As "MEMÓRIAS DO INSTITUTO BUTANTAN" têm por finalidade a apresen¬

tação de trabalhos originais que contribuam para o progresso nos campos das Ciên¬

cias Biológicas, Médicas e Químicas, elaborados por especialistas nacionais e es¬

trangeiros.

São publicadas sob a orientação da Comissão Editorial, sendo que os con¬

ceitos emitidos são de inteira responsabilidade dos autores.

The "MEMÓRIAS DO INSTITUTO BUTANTAN" are the vehicle of communi-

cation for original papers written by national and foreign specialists who contribute to

the progress of Biological, Medicai and Chemical Sciences.

They are published under the direction of the Editorial Board which assumes

no responsability for statements and opinions advanced by contributors.

Diretor do Instituto Butantan

Dr. Willy Beçak

Comissão Editorial

Presidente - Henrique Moisés Canter

Membros - Adolpho Brunner Júnior

Olga Bohomoletz Henriques

Raymond Zelnik

Sylvia Lucas

Bibliotecária - Renata Lara Paes de Barros

Indexado/lndexed: Biosis Data Base, Current Contents, Index Medicus.

Periodicidade: irregular

Permuta/Exchange: são feitas entre entidades governamentais, com publicações

congêneres, mediante consulta prévia. Exchanges with similar publications can be

settled with academic and governmental institutions through prior mutual agreement.

Endereço/Address. Instituto Butantan - Biblioteca. Av. Vital Brasil, 1.500

05504 - São Paulo, SP - Brasil

Telefone/Telephone: (011) 813-7222 - R. 129 - Telex: (011) 83325 BUTA-BR

Telefax: (011) 815-1505

SciELO

0 11 12 13 14 15 16

Governo do Estado de São Paulo

Secretaria de Estado da Saúde

Coordenação dos Institutos de Pesquisa

Instituto Butantan — São Paulo — SP — Brasil

MEMÓRIAS

INSTITUTO BUTANTAN

Volume 53, Suplemento 1, 1991

SIMPÓSIO SOBRE

"TECNOLOGIA DO DNA RECOMBINANTE

NA PRODUÇÃO DE VACINAS E NO

DIAGNÓSTICO DE DOENÇAS INFECCIOSAS"

6-7 fevereiro 1991

Instituto Butantan

São Paulo — Brasil

Publicação patrocinada pela Fundação de Amparo

à Pesquisa do Estado de São Paulo (FAPESP)

São Paulo, SP — Brasil

1991

MEMÓRIAS do INSTITUTO BUTANTAN.

Sao Paulo, SP — Brasil, 1918 -

1991,53 (1, supl. 1,

ISSN 0073-9901

MIBUAH CDD 614.07205

Solicita-se permuta/Exchange desired

SciELO

0 11 12 13 14 15 16

Mem. Inst. Butantan, v. 53, supl. 1, 1991

INSTITUTO

BUTANTAN

1901—1991

SIMPÓSIO SOBRE "TECNOLOGIA DO

DNA RECOMBINANTE NA PRODUÇÃO

DE VACINAS E NO DIAGNÓSTICO DE

DOENÇAS INFECCIOSAS"

SYMPOS1UM ON "RECOMBINANT DNA

TECHNOLOGY IN THE PRODUCTION

OF VACCINES AND THE DIAGNOSIS

OF INFECTIOUS DISEASES"

Evento comemorativo dos 90 anos de fundação do Instituto Butantan

I, | SciELO

Symposium on "Technology of recombinant-DNA in the production

of vaccines and in the diagnosis of infectious diseases".

Instituto Butantan, February 6-7,1991

PROGRAM

Febr 6,1991 -Wednesday

Morning session: Chairperson: Isaías Raw (I. Butantan, S. Paulo, Brazil)

08:30-08:45 Opening of the Symposium by the Director of the Instituto Butantan

08:45-09:45

"Genetic engineered vaccines against rabies and rabies-related viruses:

a review".

09:45-10:00

10:00-11:00

Noel Tordo (1. Pasteur, Paris, France)

Coffee Break

"Engineering bacterial toxins for the development of new vaccines".

R. Rappuoli (Lab. Sclavo, Siena, Italy)

11:00-12:00

12:00-14:00

Discussion

Lunch

Afternoon session: Chairperson: Beatriz Lieblich Fernandes (University of São Paulo,

S. Paulo, Brazil)

14:00-14:30

"Insertions of heterologous epítopes in Salmonella flagellin”.

Salete Newton (University of S. Paulo, Brazil)

14:30-15:00

"Cloning and characterization of Tripanosoma cruzi antigens and their

use in the diagnosis of Chagas disease".

Samuel Goldenberg (Fiocruz, Rio de Janeiro, Brazil)

15:00-15:20

15:20-15:50

Coffee Break

"New development in diagnosis and control of animal health"

Ingrid E. Bergman (Pan Am. Centerfor Foot-and-Mouth Disease, Rio de

Janeiro, Brazil)

15:50-16:50

Discussion

Febr 7,1991 - Thursday

Morning session: Chairperson: Wilmar Dias da Silva (I. Butantan)

08:30-09:30

"Molecular epidemiology of rabies virus throughout the world; evaluation

of the relationship between wild and vaccinal rabies strains".

Noel Tordo (1. Pasteur, Paris, France)

09:30-09:45

09:45-10:45

Coffee Break

Recombinant vaccines against whooping cough: development and

clinicai experience".

R. Rappuoli (Lab. Sclavo, Siena, Italy)

10:45-11:45

Discussion

Post-Symposium activity

Febr 8,1991 - Friday

Morning session: Coordinator: Willy Beçak (I. Butantan)

09:00-10:00

"Perspectives of anti-meningitis B vaccine production".

Cari E. Frash (Bethesda, Maryland, USA)

10:00-10:15

10:15-11:15

Coffee Break

Discussion

2 3

4 5 6 SCIELO 1Q ii 12 13 14

SciELO

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 9-10, 1991

SIMPÓSIO "TECNOLOGIA DO DNA RECOMBINANTE NA PRODUÇÃO DE

VACINAS E NO DIAGNÓSTICO DE DOENÇAS INFECCIOSAS"

O Instituto Butantan comemora este ano 90 anos de existência. Foi fundado

para atender problemas prementes de saúde pública, no caso um surto epidêmico de

peste bubônica que assolava a cidade de Santos.

No decorrer dessas nove décadas a instituição cresceu e expandiu-se.

Atualmente, as suas atividades e seus programas podem ser resumidos no desenvol¬

vimento do tripé ciência, tecnologia e educação.

A ampla gama de soros e vacinas produzidos, num total de 27 tipos diferentes,

garantem ao Instituto Butantan a posição de maior instituto de imunobiológicos da

América do Sul. No entanto, a sua filosofia não é o de uma mera fábrica de imuno¬

biológicos, mas, principalmente, o de gerar, aperfeiçoar e transferir tecnologia moder¬

na, nessa área, aos órgãos participantes dos programas de saúde pública.

Como substrato a essa tecnologia, desenvolve o Instituto Butantan ciência

básica nos vários campos relacionados a medicina e a biologia. Ênfase grande tem

sido dada a programas interdisciplinares que abrangem as novas áreas de conheci¬

mentos que utilizam as modernas técnicas de biologia molecular e de engenharia ge¬

nética.

É, portanto, não só oportuno, mas altamente significativo a inclusão no programa

de comemorações do aniversário do Instituto Butantan, do simpósio sobre "Tecnologia

do DNA Recombinante na produção de Vacinas e no Diagnóstico de Doenças Infeccio¬

sas". Os conhecimentos gerados, as descobertas, as vacinas e as proteínas purifica¬

das obtidas por engenharia genética, os tratamentos preventivos e sintomáticos de

doenças, terão nos próximos anos um impacto formidável na medicina e saúde

pública. A política de saúde projetada para o futuro implica obrigatoriamente na exis¬

tência de instituições de pesquisa e desenvolvimento que dominem as novas tecnolo¬

gias, que podem ser definidas como biotecnologia, na acepção moderna da palavra.

No Instituto Butantan, o desenvolvimento tecnológico surge, naturalmente, como

interface entre a pesquisa básica e a produção de imunobiológicos. É nesse segmento

que a tecnologia do DNA recombinante, que é o tema central desse simpósio e no

qual contamos com a participação de renomados cientistas nacionais e estrangeiros,

tem grande importância.

2 3

5 6

11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 9-10, 1991

É imprescindível para a geração dessa tecnologia em níveis e volume adequa¬

dos, a formação de uma massa crítica de pesquisadores familiarizados com esses

conhecimentos. Nesse sentido o Instituto Butantan tem investido na constante recicla¬

gem dos seus pesquisadores e técnicos e nos recém criados cursos de especialização

e curso de pós-graduação em biotecnologia. Simpósios como o que iniciamos hoje fa¬

zem parte desse programa estratégico de formação de pessoal.

Damos as boas vindas a todos os participantes e convidados desse Simpósio e

em especial aos cientistas de outras instituições do Brasil e do exterior, que estão aqui

conosco para apresentar suas contribuições e participar dos debates. É necessária

uma cooperação maior entre os cientistas de países de economia menos e mais

adiantada, para a solução dos graves problemas de saúde pública, que representam

um desafio, cuja solução é de nossa responsabilidade.

É para mim, uma honra e satisfação fazer a abertura desse importante conclave

científico, augurando sucesso.

WILLY BEÇAK

Diretor Geral

Instituto Butantan

2 3

5 6

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 11-12, 1991

OPENING REMARKS

SYMPOSIUM "RECOMBINANT DNA TECHNOLOGY IN THE PRODUCTION OF

VACCINES AND THE DIAGNOSIS OF INFECTIOUS DISEASES"

The Instituto Butantan commemorates this year 90 years of existence. It was found-

ed in order to attend urgent problems of the public health, in particular, an epidemic

outbreak of bubonic plague that devastated the city of Santos.

In the course of these nine decades, the Institution grew and developed. Its present

activities and programmes can be summarized as: development of Science, technology

and education.

The wide range of 27 different sera and vaccines produced, assures to the Instituto

Butantan the position of the largest plant of immunobiologicals of South America. How-

ever, its phylosophy is not to be a simple factory of immunobiologicals, but mainly to

generate, improve and transfer modern technology within this area to the participating

organs of the Public Health programmes.

As substrate of this technology, the Instituto Butantan develops basic Science in the

various fields related to Medicine and Biology. Emphasis is given to interdisciplinary

programs which include new areas of knowledge using modern techniques of molecu¬

lar biology and genetic engineering.

Therefore it is not only opportune, but highly significant the inclusion in the program

of commemorations of the Instituto Butantan's anniversary, the Symposium on "Recom-

binant DNA Technology in the production of vaccines and diagnosis of infectious dis-

eases". The generated knowledge, the vaccines and the purified proteins obtained by

genetic engineering, the preventive and symptomatic modern treatment of diseases will

play an important role in medicine and public health in the coming years. The future

health policy has necessarily, to imply that institutions dedicated to research and scien-

tific development should dominate the new technologies, known as Biotechnology in

the modern concept of the term.

In the Instituto Butantan, the technological development appears naturally, as an in-

terphase of basic research and production of immunobiologicals. In this segment, the

recombinant DNA, central topic of this symposium in which renowned national and for-

eign scientists will participate, is very important. To enhance the use of this technology

at adequate levei and volume we have to assemble a criticai mass of well trained sci-

SciELO

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 11-12, 1991

entists. In this sense, the Institute is investing in recycling its staff and in organizing

new courses of specialization and postgraduation mainly in biotechnology. Symposia

as the one we are opening today are part of this strategic program.

We Wellcome all participants and invited guests of this Symposium, in particular the

scientists from our country and from abroad, who accepted our invitation to present

their contributions and to participate in the discussions. A greater cooperation is neces-

sary among scientists from different countries to solve the serious problems of public

health, that are a challenge, the sollution of which is our responsability.

It is a honour and satisfaction for me to open this important scientific meeting, wish-

ing it to be a great success.

W. BEÇAK

Director

INSTITUTO BUTANTAN

SciELO

10 11 12 13

Mem. Inst. Butantan, v. 53, supl. 1, p. 13, 1991

1 - Left to right: Ingrid E. Bergmann, Pan American Center of Foot-and-Mouth Di-

sease, Rio de Janeiro, Brazil; Rino Rappuoli, Sclavo Research Center, Siena, Ita-

ly; Salete Newton, Dept. Microbiology, University of S. Paulo, Brazil; Noèl Tordo,

Dept. Virology, Institut Pasteur, Paris, France.

2 - Samuel Goldenberg, Dept. Biochemistry and Molecular Biology, Fundação Oswa-

do Cruz, Rio de Janeiro, Brazil.

3 - Cari E. Frasch, Center for Biologics Evaluation and Research, Bethesda, Mary-

land, USA.

4 - Willy Beçak, Director of the Instituto Butantan, S. Paulo, Brazil.

SciELO

Mem. Inst. Butantan, v. 53, supl. 1, p. 15-20, 1991

ENGINEERING BACTERIAL TOXINS FOR THE

DEVELOPMENT OF NEW VACCINES

M.G. Pizza, M. Domenighini, L. Nencioni, A. Podda, R. Vanni, S.

Silvestri, R. Rappuoli, Sclavo Research Centre, Siena, Italy

BACTERIAL ADP-RIBOSYLATING TOXINS

ADP-ribosylating bacterial toxins are proteins, produced by pathogenic bactéria,

which are usually released into the extracellular médium and cause disease by killing

or altering the metabolism of eukaryotic cells. The toxins are usually composed of two

functionally distinct domains: a toxic moiety and a vector which have been called do-

mains A and B, respectively. The vector (B) binds the receptors on the surface of eu¬

karyotic cells and delivers the toxic part (A) across the membrane of eukaryotic cells so

that it can reach its target proteins 1 .

While the properties and the complexity of the vector (B) differ from toxin to toxin and of-

ten also within the same family of toxins, all of the A domains have a common mechanism

of action: they are enzy mes which ADP-ribosylate eukaryotic target proteins which control

crucial circuits of eukaryotic cells, such as protein synthesis, transmembrane signaling,

oncogenesis, cytoskeleton structure. The target proteins also have a common feature and

a common structure: they are GTP-binding proteins. The only ADP-ribosylating toxins

characterized both in terms of protein and genetic structure are diphtheria toxin (DT),

Pseudomonas exotoxin A (PAETA), pertussis toxin (PT), cholera toxin (CT) and the E.coli

heat-labile toxin (LT). The main properties ofthese toxin saresummarized in Table I.

THE ACTIVE SITE OF ADP-RIBOSYLATING TOXINS HAS A COMMON STRUCTURE

The initial comparison of the aminoacid sequences of DT and PAETA did not show

sequence homology 2 . However, when by photoaffinity labeling Carroll and Collier

showed that Glu148 of diphtheria toxin is functionally equivalent to Glu553 of Pseudo-

Correspondence to: R. Rappuoli, Sclavo Research Centre, Via Fiorentina 1, 53100 Siena, Italy.

2 3

5 6

11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 15-20, 1991

TABLE

Bacterial

toxin

Acceptor

protein

Acceptor

aminoacid

Effect on Primary

eukaryotic cells

Eukaryotic cell

structure receptor

Diphtheria

Elongation

Diphtamide

Inhibition of Known

14.5 Kd protein?

toxin

factor-2

715

protein synthesis

Pseudomonas

Elongation

Diphtamide

Inhibition of pro- Known

Not known

ETA

factor-2

715

tein synthesis

Pertussis

Gi, Go, T

Cys 352

Alteration of trans- Known

160 Kd glycopro-

toxin

membrane

tein in CHO cells

signal transduction

Cholera

Gs, T (Gi

Arg 201

Ganglioside

toxin

and Go

GM1>GDb1

E. coli

Arg 201

••

Ganglioside

LT1

GM1>GDb1>GM2

monas exotoxin A 3 they realigned the aminoacid sequences using this reference point

and found a strong homoiogy between the two toxins 4 . Regions of strong homology

were also found between pertussis and cholera toxins 5 ' 6 . The two groups of toxins

(which have a different target) did not show any aminoacid homology.

In the meantime, by Chemical modification, photoaffinity labeling and site-directed

mutagenesis, a number of aminoacids were identified which are essential for the cata-

lytic activity of the different toxins 2 . Remarkably, these aminoacids were found to be

conserved in all enzymes (Fig. 1).

The availability of the three-dimensional structure of PAETA, determined by X-ray

crystallography by Allured et ai. 7 provided the template for a computer-based molecular

modeling of DT, PT and CT. Using the coordinates of ETA we were able to predict the

majority of the structure of DT and the structure of the active sites of PT and CT. As

shown in Fig. 2, the active site of the four toxins which contain all aminoacids identified

in Fig. 1 share a common structure.

SITE-DIRECTED MUTAGENESIS OF THE ACTIVE SITE AND CONSTRUCTION OF

NON TOXIC MOLECULES

The identification of the aminoacids which are in the active site of the ADP-

ribosylating toxins, provided the rational basis for the mutagenesis of their genes in or-

der to obtain non toxic molecules to be used in vaccines. A number of non toxic mole-

cules have been obtained from diphtheria toxin, pseudomonas exotoxin A and pertus¬

sis toxin 8 . In all cases, the substitution of the glutamic acid (residue 6 in Figs. 1 and 2)

or its deletion was the mutation most effective in reducing the toxicity of the molecules.

Among the many non toxic derivatives obtained, the one which has been studied in

more detail is PT-9K/129G, a non toxic mutant of pertussis toxin which is being actively

tested in clinicai trials as a new vaccine against whooping cough. This mutant contains

SciELO

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 15-20, 1991

Figure 1 - Homologies between the three regions forming the active site of Pseudomonas exotoxin

A (PAETA) and corresponding regions of diphtheria toxin (DT) pertussis toxin (PT), cholera toxin

(CT) and E. coli LT toxin (LT).

H6TFLE

H G t] K P G

H els RmT

R0L

PARS

857

N|V L

ETA

463

U R

F Y

.(7)..

Y G

W K

F Y

.(9)..

F T

U G N

CT *

124

w|y r

CT *

171

H R

WRlE

PARS

893

.. G

y p a

..(9).

.. I

. Y

..(6).

.. F

..ST

Y Y

. S T

Y Y

..(8).

ETA

549

G G R L

I\T[Í

LG H

ETA 555 [TIL G

P L A

144

S S S V

E v i

N N[W]

DT 150 [ijN N

E 0 A

PARS

983

L L Y N

e y i

V Y 0

PT 104 Y F E

V D T

125

TYOS

E Y L

A H R

CT * 122 SOI

G W Y

CT * 124 Q]Y G

YR V

Figure 2 - Structure of the active sites of PAETA, DT, CT and PT, obtained using the coordinates

of PAETA. The numbers identify the aminoacids shown in Fig. 1.

Mem. Inst. Butantan, v. 53, supl. 1, p. 15-20, 1991

two aminoacid substitutions (Arg9 —> Lys9 and Glu129 —> Gly) which make this

molecule absolutely non toxic but fully immunogenic 9 ' 10 .

GENETIC DETOXIFICA TION OF PT

The failure to induce protective immunity with recombinant subunits suggested that

the ideal vaccine should be a PT molecule whose toxicity has been eliminated by ge-

netic manipulation of the gene coding íor subunit SI. To do so, we and other investiga-

tors generated a number of recombinant SI molecules containing aminoacid substitu¬

tions and tested their enzymatic activity. Substitution of either Arg9, Aspll, Arg13,

Trp26, or Glu129 was found to reduce the enzymatic activity to undetectable leveis 8 .

Each of the above mutations was then introduced into the chromosome of B. pertussis

whose wild type genes had been deleted. These new B. pertussis strains were found to

produce molecules indistinguishable from PT in SDS-PAGE which had a toxicity that

ranged from 0.1% to 10% of wild type PT. Since even 0.1% of the toxicity is by far too

high for a molecule to be used in a vaccine, we combined some of the above mutations

and obtained PT double mutants that were at least 10 6 times less toxic than wild type

PT (Table II) 9 . Such molecules, being non toxic, were ideal candidates for new vac-

cines provided they had maintained the correct B- and T-cell epitopes and were able to

induce protective immunity in animal models. The non toxic double mutant shown in

Table II was found to have the same B- and T-cell epitopes as wild type toxin, to induce

toxin-neutralizing antibodies, and to protect mice from the intracerebral chalienge 9 ’ 10 .

GENETICALL Y DETOXIFIED MOLECULES ARE SAFER AND MORE

IMMUNOGENIC THAN CHEMICALLYDETOXIFIED TOXINS

Formaldehyde and glutaraldehyde are usually used to detoxify toxins for vaccine pur-

poses. To find whether PT-9K/129G is more immunogenic than chemically inactivated per-

TABLE II

In wVoand in vitro properties of PT-9K/129G compared with purified native PT

Property

PT-9K/129G

CHO cdl clustered growth

(ng/ml)

0.005

> 5,000

ADP ribosylation

(ug)

0.001

>20

Mitogenicity

(ug/ml)

0.1-0.3

Hemagglutination

(ug/well)

0.1

Affinity constant (anti-SI)

[Ka (L/mol)]

2.4X10 8

6.1x10 a

Affinity constant (anti- PT)

[Ka (L/mol)]

2.0x10 10

9.8x10 9

Histamine-sensitization

(ug/mouse)

0.1-0.5

> 50

Leukocytosis stimulation

(ug/mouse)

0.02

> 50

Anaphylaxis potentiation

(ug/mouse)

0.04

>7.5

Enhance isulin secretion

(ug/mouse)

>25

In vivo acute toxicity

(ug/Kg)

N.D.

> 1,500

N.D. = not determined

Mem. Inst. Butantan, v. 53, supl. 1, p. 15-20, 1991

tussistoxin molecules, we treated PT-9K/129G with 0.07% and 0.42%formaldehydefor48

hours at 37°C and then compared the in v/fro properties and the immunogenicity orthe re-

sulting molecules. As shown inTable III, formaldehydetreatment abolishedthe hemagglu-

tinating property of PT-9K/129G, decreased ils affinity for anti-PT gama globulins and

masked the epitope recognized by the protective monoclonal antibody 1B7. The immunog¬

enicity of PT-9K/129G did not decrease after formaldehyde treatment. The ELISA titer of

the sera obtained from guinea pigs immunized with natural or formalin-treated molecules

were almost identical (Table III). In rnarked contrast, the ability of the sera to neutralize PT

in the CHO assay were dramatically lower when chemically treated molecules were used

for immunization. In the intracerebral challenge assay, the formalin-treated molecules

were remarkably less potentthan natural PT-9K/129G (T able II). We have therefore shown

that Chemical treatment of toxins for vaccine production induces profound changes in the

antigenic propertiesof the molecules. These changes do not alterthe total amount of anti-

bodies induced but change dramatically the quality of the antibodies obtained. As a result,

large quantities of chemically treated molecules are required to induce a protective re¬

sponse. Under these conditions, the immune system produces mainly antibodies against

non protective epitopes (or with lower affinity for the protective epitopes) and the cellular im-

munity is overstimulated, Both conditions may favor the appearance of untoward reactions

of the Arthus type or delayed typed hypersensitivity. In conclusion, the properties of PT-9K/

129G show that molecular genetics has provided new and more eff icient tools to inactivate

toxinsfor vaccine use. These molecules have the same conformation as the native proteins

and are much betterthan chemically treated toxins in inducing protective immunity.

TABLEI

Effect of formaldehyde treatment on the properties of the genetically inactivated

pertussistoxin mutant PT-9K/129G.

Formaldehyde Hemagglutination

Affinity Constant

Immunogenicity

Vaccine Potency

(%)

(jxg/well)

Polyclonal

gamma

globulins

Monoclo¬

nal 1 B7

(guinea pigs

immuni zed with

3 pg of antiçpn)

Mice survival after

immunization with

5 pg of antigen

ELISA titer CHO titer

Intracerebral

challenge

0.0

0.5

1.15 10 9

5.510 7

3.5

1/2560

13/16

0.07

4.0

5.2610 a

6.75 10 7

3.1

1/160

8/16

0.42

>10.0

3.3

1/80

1/16

CLINICAL STUDIES

After extensive studies in animal models which have shown that PT-9K/129G is non

toxic, immunogenic and is able to protect mice from the infection with virulent B. pertus-

sis, we have tested PT-9K/129G in human adult volunteers 11 . The results of this study

showed that the vaccine was safe and induced a great increase in antibody titers

against PT both in ELISA and CHO neutralization assays. The titers of PT-specific anti¬

bodies were higher than those reported in similar studies using higher doses of chemi-

SciELO

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 15-20, 1991

cally detoxified PT. In particular, the ratio between toxin neutralizing titers and total anti

PT ELISA titers was the highest so far reported, suggesting that also in man, immuni-

zation with a molecule not chemically modified induces antibodies with higher affinity

for the native PT. After the successful phase I study in adult volunteers, the vaccine is

now being tested in 3- and 15-month old children. So far the results confirm the excel-

lent properties of the new pertussis vaccine.

REFERENCES

1. MIDDLEBROOK, J.L. & DORLAND, R.B. Bacterial toxins: cellular mechanisms of action. Mi-

crobial. Rev., 48: 199-221, 1984.

2. DOMENIGHINI, M.; MONTECUCCO, C.; RiPKA, W.C.; RAPPUOLI, R. Mol. Microbiol., 5: 23-

32, 1991.

3. CARROLL, S.F. & COLHER, R.J. NAD binding site of diphtheria toxin: identification of a resi-

due within the nicotinamide subsite by photochemical modification with NAD. Proc. Natl.

Acad. Sei. USA, 81: 3307-3311, 1984.

4. CARROLL, S.F. & COLLIER, R.J. Aminoacid sequence homology between the enzymic do-

mains of diphtheria toxin and Pseudomonas aeruginosa exotoxin A. Mol. Microbiol., 2:

293-296, 1988.

5. NICOSIA, A.; PERUGINI, M.; FRANZINI, C.; CASAGLI, M.C.; BORRI, M.G.; ANTONI, G.; AL-

MONI, M.; NERI, P.; RATTI, G.; RAPPUOLI, R. Cloning and sequencing of the pertussis

toxin genes: operon strueture and gene duplication. Proc. Natl. Acad. Sei. USA, 83: 4631-

4635, 1986.

6. LOCTH, C. & KEITH, J. Pertussis toxin gene: nucleotide sequence and genetic organization.

Science, 232: 1258-1264, 1986.

7. ALLURED, V.S.; COLLIER, R.J.; CARROLL, S.F.; MCKAY, D.B. Strueture of exotoxin A of

Pseudomonas aeruginosa at 3.0 Angstrom resolution. Proc. Natl. Acad. Sei. USA, 83:

1320-1324, 1986.

8. RAPPUOLI, R. & PIZZA, M. Strueture and evoiutionary aspects of ADN-ribosylating toxins.

In: ALOUF, J. & FREER, J., eds. Strueture, regulation and activity of bacterial toxins. (in

press), 1990.

9. PIZZA, M.; COVACCI, A.; BARTOLINI, A.; PERUGINI, M.; NENCIONI, L; DE MAGISTRIS,

M.T.; VILLA, L; NUCCI, D.; MANETTI, R.; BUGNOLI, M.; GIOVANNONI, F.; OLIVIERI, R.;

BARBIERI, J.T.; SATO, H.; RAPPUOLI, R. Mutants of pertussis toxin suitable for vaccine

development. Science, 246: 497-500, 1989.

10. NENCIONI, L; PIZZA, M.; BUGNOLI, M.; DE MAGISTRIS, M.T.; Dl TOMMASO, A.; GIOVAN-

NONI, F.; MANETTI, R.; MARSILI, I.; MATTEUCCI, G.; NUCCI, D.; OLIVIERI, R.; PILERI,

P.; PRESENTINI, R.; VILLA, L; KREEFTENBERG, H.; SILVESTRI, S.; TAGLIABUE, A.;

RAPPUOLI, R. Characterization of genetically inactivated pertussis toxin mutants: candi¬

dates for a now vaccine against whooping cough. Infect. Immun., 58: 1308-1315, 1990.

11. PODDA, A.; NENCIONI, L.; DE MAGISTRIS, M.T.; Dl TOMMASO, A.; BOSSU', P.; NUTI, S.;

PILERI, P.; PEPPOLONI, S.; BUGNOLI, M.; RUGGIERO, P.; MARSILI, I.; D'ERRIcÓ, A.;

TAGLIABUE, A.; RAPPUOLI, R. Metabolic, humoral and cellular responses in adult volun¬

teers immunized with the genetically inactivated pertussis toxin mutant PT-9K/129G. J.

Exp. Med, 772:861-868, 1990.

2 3

5 6

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 21-29, 1991

EVALUATION OF ACELLULAR DPT VACC1NES IN INFANTS

Luciano Nencioni, Audino Podda, Samuele Peppoloni, Gianfranco

Voipini, llio Marsili, ‘Bruno Contu, ‘Giuseppa Campus, ‘Maria

Antonia Cossu, Riccardo Vanni, Sérgio Silvestri, Rino Rappuoli,

Sclavo S.p.A, Siena, and ‘Health Local Unit #5, Ozieri, Italy.

ABSTRACT: Two acellular DPT vaccines containing, as pertussis components, the ge-

netically detoxified pertussis toxin mutant PT-9K/129G, either alone or combined with

FHA and 69K, were evaiuated for safety and immunogenicity in infants 8-14 months old.

Both vaccines induced very mild local reactions which were consistent with the pres-

ence of alum and the previous administration of two doses of whole-cell DPT vaccine.

A marked increase in specific antibodies to each pertussis component and in pertussis

toxin neutralizing antibodies was observed after one dose of either acellular vaccines.

All vaccinees also acquired an excellent protective immunity against diphtheria and tet-

anus, as assessed in vitro and in vivo.

INTRODUCTION

The genetically detoxified pertussis toxin (PT) mutant PT-9K/129G is naturally de-

void of the toxic properties of PT and maintains the physicochemical and the immuno-

logical properties of the wild type toxin 1 < 2 .

We have developed acellular pertussis vaccines containing the PT-9K/129G mutant

alone or together with the filamentous haemagglutinin (FHA) 3 and an outer membrane

protein named 69K or pertactin 4 . These proteins are involved in the adhesion of B. per¬

tussis to mammalian cells 5 and therefore, if included in a vaccine, are expected to pre-

vent bacterial colonization. Both FHA and 69K are purified from cultures of the recombi-

nant strain B. pertussis W28-9K/129G which produces the non toxic PT mutant PT-9K/

Correspondence to: R. Rappuoli, Sclavo Research Centre, Via Fiorentina 1, 53100 Siena, Italy

í, i SciELO

Mem. Inst. Butantan, v. 53, supl. 1, p. 21-29, 1991

129G and therefore these products cannot be contaminated with activate pertussis tox-

in. Since no Chemical treatment is necessary to inactivate the toxin, these molecules

can be safely used in pertussis vaccines without fear of potential reversion to toxicity,

loss of immunogenicity, and batch-to-batch variations 6 ' 7 .

In previous studies we have shown that acellular pertussis vaccines containing PT-

9K/129G alone 8 or combined with FHA and 69K 9 are extremely safe in adult volunteers

and that are able to induce high leveis of humoral and celiular immunity. These results

are confirmed by phase II clinicai trials in infants and children which are now in

progress. Since it is likely that one of these pertussis formulations will be used for

mass immunization and administered within the DPT vaccination schedule, we have

prepared two trivalent vaccines, in which diphtheria and tetanus toxoids have been

combined with only PT-9K/129G (DPT3/P/AH) or with PT-9K/129G, FHA and 69K

(DPT7/PFK/AH).

In this paper we report the safety and the immunogenicity in infants of these new

acellular DPT vaccines.

MATERIALS AND METHODS

Study design. Forty-five healthy infants of both sexes, 8 to 14 months of age, were

recruited at the Health Local Unit of Ozieri, Sardinia, Italy. Among these, twenty-one

and twenty-four subjects received intramuscularly one dose of DPT3/P/AH or DPT7/

PFK/AH vaccine, respectively. All infants had previously received at least two doses of

conventional whole-cell DPT vaccine.

Safety assessment and serology. Rectal temperature was monitored at 3, 6 and 24

hours after vaccination. Drowsiness, fussiness, appetite, vomiting, redness, swelling

and pain were monitored 3 and 6 hr after vaccination and then at bed-time throughout

the first week and on the 14 th evening.

For redness and swelling we have reported in table 1 only values which were > 1

cm; for fussiness and appetite only values monitored throughout the first three days af¬

ter vaccination.

Parents were instructed to record the rectal temperature of the infant and to evaluate

any local and systemic reactions. All infants were home-visited 24 hs after the vaccine ad-

TABLEI

Adverse reactions after vaccination

Reaction

DPT3/P/AH

DPT7/PFK/AH

Fever

2/21

3/24

Drowsiness

2/21

3/24

Fussiness

5/21

6/24

Appetite

4/21

4/24

Vomiting

2/21

2/24

Redness

4/21

1/24

Swelling

4/21

3/24

Pain

2/21

6/24

SciELO

10 11 12 13

Mem. Inst. Bulantan, v. 53, supl. 1, p. 21-29, 1991

ministrationby afollow-upnurse who monitored parents by phonethroughout 14days.

Venous blood samples for in vivo and in vitro evaluation of specific IgG and toxin-

neutralizing antibodies were obtained before and 4 weeks after vaccination from 7 and

11 infants receiving, respectively, the DPT3/P/AH or the DPT7/PFK/AH vaccines.

Vaccines. Both the aceliular DPT vaccines were prepared at Sclavo Laboratories

(Siena, Italy). Each 0.5 ml single dose vial of DPT3/P/AH contained 15 Lf of diphtheria

toxoid, 10 Lf of tetanus toxoid, 10 pg of PT-9K/129G, 0.05 mg of thimerosal and 2 mg

of aluminium hydroxide. Each 0.5 ml single dose vial of DPT7/PFK/AH contained 20 Lf

of diphtheria toxoid, 12.5 Lf of tetanus toxoid, 5 pg of PT-9K/129G, 5 pg of FHA, 3 pg

of 69K, 0.05 mg of thimerosal, and 1 mg of aluminium hydroxide.

Antigens. The non toxic mutant PT-9K/129G and FHA were purified from the culture

supernatants of the recombinant strain B. pertussis W28-9K/129G 1 according to a mod-

ified version of the method described by Cowell et al. 10 . The 69K protein was extracted

from cell paste of B. pertussis W28-9K/129G 1 by heat treatment and purified to the ho-

mogeneity by anionic exchange chromatography and gel filtration (Manetti & Rappuoii,

manuscript in preparation). Before use, each antigen underwent a mild stabilization

with formaldehyde.

Diphtheria and tetanus toxoids were obtained as previously described 11 .

CHO cell toxin neutralization assay. Pertussis toxin-neutralizing antibodies induced

by vaccination were using, as standard, the U.S. Reference Human Pertussis Antiser-

um (lot # 3), containing 640 neutralizing units, kindly provided by the Center for Drugs

and Biologics, Bethesda, MD. U.S.A. Briefly, sera from volunteers obtained after ad-

ministration of one or two doses of PFK/2 vaccine were diluted directly in the wells of

flat-bottomed microplates (Costar, Cambridge, MA, U.S.A.) in 25 pl of Dulbecco's Modi-

fied Eagle Médium (DMEM, Flow Laboratories, Mc Lean, VA, U.S.A.). Purified wild

type PT (120 pg) in 25 pl of Coulter médium DMEM was added to each well and the

plates were incubated for 3 h at 37°C. After the incubation period, 0.2 ml of DMEM con¬

taining 1x10 4 CHO cells, previously treated with 1 mg/ml of trypsin, were added to each

well and incubated for 48 h at 37°C in atmosphere of 5% CO z . As positive control, the

clustering effect of PT alone was titered in each plate. Neutralizing titers were ex-

pressed as the reciprocai of the highest serum dilution causing complete inhibition of

the clustering activity induced by the native toxin.

Vero cell toxin neutralization assay. Diphtheria toxin-neutralizing antibodies induced by

vaccination were tested by the VERO cell assay using as standard the NIH Reference An-

tiserum (lot #A50), containing 3,200 neutralizing units. The test was performed in the wells

of flat-bottomed microplates (Costar, Cambridge, MA, U.S.A.). Antisera from vaccinated

infants and the NIH reference serum were diluted in culture médium Ml99, containing

10% FCS, 2 mM glutamine, 25 mM Hepes, 50 pg/ml gentamicin, directly in the wells by

twofold serial dilutions. The volume of antisera added to each well is 20 pl. After that, the

diphtheria toxin (80 LF/ml diluted 1:10 5 ) was added to each well in a volume of 20 pl/ml.

The plates are then incubated for 3 h at 37°C. After the incubation time, 200 pl of culture

médium containing 10 9 VERO cells are added to each well. The microtiter plates were then

incubated at 37°C and, 48-72 h later, cells were strained with crystal violet and the stain

was solubilized with 50% solution in water (vol/vol) of ethanol. Absorbance values at 560

SciELO

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 21-29, 1991

nm of samples were measured against blank on aTitertek Mulliskan (Flow Laboratories,

Inc., McLean, VA, U.S.A.). Controls for each plate included wells with cells and diphtheria

toxin (positive control), and wells with cells but no toxin (negative control). Each serum

sample was tested in duplicate and absorbance values were averaged. The neutralizing

titer is expressed as the reciprocai of the highest dilution of serum samples giving absor¬

bance values comparable to those of the negative control (100% of protection).

Elisa

The ELISA method was performed as previously described 2 ' 8 . Wells of flat-bottomed

polystyrene microtest plates (Dynatech Laboratories, Inc., Alexandria, VA, U.S.A.) were

coated with 10Opil of PBS pH 7.4 containing 1 pg of purified PT or FHA or 69K, diphtheria

toxoid ortetanus toxoid. The coating was performed for 2 h at 37°C and overnight at 4°C in

a humidified chamber. The coating buffer was aspirated, and wells were washed with 200

(j.1 of PBS containing 0.05% Tween 20 and 0.02% sodium azide (PTA). To minimize non-

specif ic adsorption of serum proteins to the plastic, wells were coated with 200 pl of a block-

ing solution consisting of 1% bovine serum albumin (BSA) in PBS, and then incubated for 2

h at 37°C. Plates were then washed three times in PTA and 200 pi of fivefold diluted test

serum were added to the wells. The U.S. Reference Human Pertussis Antiserum (lot # 3)

containing 200 ELISA Units (EU) per ml of IgG anti-PT and 200 EU/ml anti-FHA was used

as reference standard. In the case of the 69K protein, we used an "in-house standard" im-

mune serum to which we assigned a valueof 20 EU/ml of lgGanti-69K, as well as 10 EU/ml

of IgG anti-diphtheriaand anti-tetanustoxoids. Following incubation at 37°C for 2 h, plates

were washed three times with PTA and a conjugate of anti human IgG-alkaline phospha-

tase was added. Plates were then incubated at 37°C for 2 h and washed three times with

PTA. Finally, 1 00pl of p-nitrophenyl phosphate substrate (Sigma Chemical Co., St. Louis,

MO, U.S.A.) 1 mg/ml in 1 M diethanolamine, pH 9.8, containing 1 mM MgCI 2 , was added to

each well. The enzyme-substrate reaction, which developed at room temperature, was

stopped after 30 min, and the absorbance values of the samples was measured at 405 nm

against blank (substrate in diethanolamine, pH 9.8) on aTitertek Multiskan (Flow Laborato¬

ries, Inc., McLean, VA, U.S.A.). Controls for each plate included wells with serum samples

but no antigen, and wells with antigen but no serum samples. Each serum sample was test¬

ed in duplicate and absorbance values were averaged. Total IgG antibodies to PT, FHA

and 69K in the test samples were expressed as geometric mean of EU/ml, determined

according to the parallel line bioassay procedure described by Manclark et al. 14 .

Diphtheria and tetanus antitoxin tritation in vivo. The assays were performed accord¬

ing to W.H.O. recommendations using rabbits for diphtheria and mice fortetanus antitoxin

tritation. The results, expressed in International Units (IU) per ml, were obtained accord¬

ing to the N.I.H. reference diphtheria antiserum (Lot # A50, containing 6 IU/ml) and the

N.I.H. reference tetanus antiserum (Lot # A50, containing 6 IU/ml), used as standards.

RESULTS

Safety. The follow-up performed after the administration of DPT3/P/AH (table 1)

showed that no infants had fever over 38°C with two exceptions (38.1°C and 38.2°C).

However, in both cases, the temperature was over 38°C for only one measurement

2 3

5 6

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p, 21-29, 1991

and, in no cases, the administration of anti-pyretic drugs was necessary. Unusual drow-

siness was only noticed in two infants. Five out of twenty-one infants were more irrita-

ble than usual throughout 72 hr following vaccination, but all of them maintained a nor¬

mal activity. Three infants had loss of appetite (two of them after the first three days),

while two infants had an isolate episode of vomiting on day 2 and 4, respectively. Four

out of twenty-one subjects had redness and swelling at the site of injection and finally,

two infants had local pain that in one case was very mild.

As far as vaccination with DPT7/PFK/AH is concerned, the follow-up of adverse re-

actions (table 1) showed that only one out of twenty-four infants had redness and

three had swelling. Local pain was reported in six infants. Three infants had fever over

38°C. Drowsiness in the first twenty-four hours after vaccination was noticed in three

infants. Mild fussiness occurred in six subjects and for four infants, parents reported

loss of appetite throughout 72 hr after vaccination while two infants had isolate epi-

sodes of vomiting.

Serology. The humoral response of infants to the DPT3/P/AH vaccine is reported in

figure 1 and table 2 and that one to DPT7/PFK/AFI vaccine is reported in figure 2 and

table 2.

As expected, all the infants, after two doses of conventional DPT cellular vaccine,

showed low titers of anti-PT, anti-FHA and anti-69K IgG antibodies, assayed by ELISA

(Figures 1,2), but a good immune response against diphtheria and tetanus, either evalu-

ated as passive protection in vivo (Table 2) or as specific antibodies in vitro (Figures 1,2).

(DTP3/P/AH)

10 '

10 l

E3 PT

E5S DIPHTHERIA

□ TETANUS

150

100

aio

VERO

1500

1000

500

PRE

POST

PRE POST PRE POST

Figure 1 — Specific IgG antibodies (ELISA) as well as pertussis (CHO) and diphtheria (VERO)

toxin neutralizing antibodies in infants receiving DPT3/P/AH vaccine.

SciELO

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 21-29, 1991

1(T

10 "

1 .

io u r

10 '

□

FIIA

□

69K

DIPHTHERIA

□

TETANUS

(DTP7/PFK/AH)

10 -

CHO

VERO

1000

750 W

500 K

250

POST

PRE POST PRE POST

Figure 2 — Specific IgG antibodies (ELISA) as well as pertussis (CHO) and diphtheria (VERO) tox-

in neutralizing antibodies in infants receiving DPT7/PFK/AH vaccine.

TABLE II

Diphtheria and tetanus antitoxin titration in vi vo 1

Vacine

Infant

Diphtheria

Tetanus

N a

PRE

POST

PRE

POST

DPT3/P/AH

<0.01

>0.1 < 1

> 8 < 15

> 2 < 4

>0.1 < 1

> 15

N.D. 2

> 0.1 < 1

> 1 < 2

>0.1 < 1

> 15

DPT7/PFK/AH

> 1 <2

> 4

> 15

>0.1 < 1

> 2 <4

> 1 <2

> 15

> 0.1 < 1

> 2 < 4

> 1 < 2

> 15

0.1

> 1 <2

> 15

> 0.1 < 1

> 1 <2

> 1 < 2

> 8 < 15

>0.1 < 1

> 2 < 4

> 15

N.D. 2

>0.1 < 1

> 4

> 1 < 2

> 15

> 1 < 2

> 2 < 4

> 8 < 15

> 0.1 < 1

> 1 <2

> 2 < 4

> 15

0.1

> 1 < 2

> 8 < 15

1) The test was performed in rabbits and in mice for in vivo titration of anti-diphitheria

and anti-tetanus neutralizing antibodies, respectively. Results are expressed as In¬

ternational Units (IU)/ml.

2) N.D. Not Determined.

SciELO

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 21-29, 1991

Total IgG antibodies to each vaccine components increased after the administration

of the acellular DPT vaccines. A similar rise was noted for pertussis toxin-neutralizing

antibodies, evaluated by the CHO cell assay, and for diphtheria toxin-neutralizing anti¬

bodies, assayed by the VERO cell assay (Figure 1, 2). A high serum activity against

diphtheria and tetanus toxins was also observed in animais, using the in vivo assay

recommended by WHO. In fact, sera from infants receiving one dose of either DPT

acellular vaccine showed a marked increase of passive protection against diphtheria

and even more pronounced protection against tetanus. In most cases the in vivo diph¬

theria antitoxin titers were over 4 IU/ml and the tetanus antitoxin titers were over 15 IU/

ml (Table 2).

DISCUSSION

In a previous phase I 8 ' 9 and a phase II (Podda et al., manuscript in preparation) clin¬

icai study, we had carefully tested the safety and the immunogenicity of two monoval-

ent acellular pertussis vaccines, one containing the genetically detoxified pertussis tox-

in mutant PT-9K/129G 1 and the other containing PT-9K/129G combined with FHA and

69K. Both vaccines proved to be safe and immunogenic in adults and children. Since

the final formulation of pertussis vaccine, to be introduced in the vaccination schedule

of children, is expected to contain also diphtheria and tetanus toxoids, we have pre-

pared two acellular DPT vaccines (DPT3/P/AH and DPT7/PFK/AH) which pass the test

of the American and European pharmacopea and we have tested them in infants.

Although the present study is a not controlled open evaluation carried out on a limit-

ed number of subjects, some comments about the safety and the immunogenicity of

these acellular DPT vaccines can be made.

The local reactions reported after vaccination with DPT3/P/AH were very mild and

consistent with the administration of a vaccine adsorbed with aluminium hydroxide (2

mg/dose) and with the fact that all infants had, previously, received two doses of con-

ventional whole-cell DPT vaccine. In detail, the 19% of infants had redness of 1 cm of

diameter at the site of injection, but in only one case the size of the reaction was great-

er than 1 cm. Swelling and local pain occurred in the 19% and 9% of infants, respec-

tively. The most frequent systemic reaction was an unusual irritability, reported in the

23% of subjects, which, however, did not affect the normal activity of infants. In two

cases (9.5%), moderate fever and excessive sleeping were reported.

As compared with DPT3/P/AH, the local reactions, mainly redness, were less fre¬

quent after the administration of DPT7/PFK/AH vaccine. This is probably due to the re-

duced amount of adjuvant (1 mg versus 2 mg). In detail, only one infant had redness at

the site of injection while swelling and local pain occurred in the 12.5% and 25% of in¬

fants, respectively.

Although the major aim of this small trial was the evaluation of the safety we have

also performed an immunogenicity study in a small number of infants from whom it was

possible to obtain serum samples before and one month after vaccination with the acel¬

lular DPT vaccines. All infants had a low levei of anti-PT, anti-FHA and anti-69K anti¬

bodies assessed by ELISA and a not detectable PT-neutralizing activity. Both specific

and neutralizing antibodies markedly increased after the administration of either vac¬

cines but the enhancement of the humoral immunity, probably due to the higher

amount of adjuvant, was more pronounced after vaccination with DPT3/P/AH. Also the

SciELO

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 21-29, 1991

serum antibody responses to diphtheria and tetanus toxoids, which was already good

before vaccination, augmented more than ten times aíter one dose of either acellular

DPT vaccines as evaluated in vitro in terms of ELISA and VERO titers. The enhanced

serum activity against diphtheria and tetanus was confirmed by in vivo antitoxin anti¬

body titrations. In fact, sera from all vaccinees were able to confer a high passive pro-

tective immunity to the animais reaching in most cases, after one dose of acellular DPT

vaccine, values of more than 4 IU/ml against diphtheria and even more than 15 IU/ml

against tetanus.

In conclusion, the careful follow-up of 21 infants receiving DPT3/P/AH and of 24 in-

fants receiving DPT7/PFK/AH in addition to the immune response evaluation per-

formed in some of them proved the substantial safety and the excellent immunogenicity

of these acellular DPT vaccines and encourages their further evaluation in enlarged

clinicai trials.

REFERENCES

1. PIZZA, M.G.; COVACCI, A.; BARTOLONI, A.; PERUGINI, M.; NENCIONI, L; DE MAGISTRIS,

M.T.; VILLA, L; NUCCI, D.; MANETTI, R.; BUGNOLI, M.; GIOVANNONI, F.; OLIV1ERI, R.;

BARBIERI, J.; SATO, H.; RAPPUOLI, R. Mutants of pertussis toxin suitable for vaccine de-

velopment. Science, 246: 497-500, 1989.

2. NENCIONI, L; PIZZA, M.G.; BUGNOLI, M.; DE MAGISTRIS, M.T.; Dl TOMMASO, A.; GIOVAN-

NONI, F.; MANETTI, R.; MARSILI, I.; MATTEUCCI, G.; NUCCI, D.; OLIVIERI, R.; PILERI, P.;

PRESENTINI, R.; VILLA, L; KREEFTEMBERG, J.G.; SILVESTRI, S.; TAGLIABUE, A.;

RAPPUOLI, R. Characterization of genetically inactivated pertussis toxin mutants: candi¬

dates fora new vaccine against whooping cough. Infect. Immun., 58: 1308-1315,1990.

3. RELMAN, D.A.; DOMENIGHINI, M.; TUOMANEN, E.U.; RAPPUOLI, R.; FALKOW, S. Filamen-

tous hemagglutinin of Bordetella pertussis: nucleotide sequenco and crucial role in adher-

ence. Proc. Natl. Acad. Sei. U.S.A., 86: 2637-2641, 1989.

4. ROBINSON, A.; GORRINGE, A.R.; FUNNEL, S.G.P.; FERNANDEZ, M. Specific seroconver-

sion of mice against intranasal infection with Bordetella pertussis. Vaccine, 7: 321-324,

1989.

5. PILERI, P.; PEPPOLONI, S.; NUTI, S.; TAGLIABUE, A.; RAPPUOLI, R.; NENCIONI, L. Evalu¬

ation of eukaryotic cell attachment and internalization of Bordetella. In: RAPPUOLI,

ALOUF, FALMAGNE, FEHEREMBACH, FREER, GROSS, JELJASZEWICZ, MONTECUC-

CO, TOMASI, WADSTROM, WITHOLT, eds. BactériaI Protein Toxins. Edited by Gustav

Fisher Verlag, Stuttgart, 1990, p 391-396.

6. MOXON, R. & RAPPUOLI, R. Modem vaccines: Haemophilus influenzae infections and

whooping cough. Lancet: 1324-1329, 1990.

7. NENCIONI, L; VOLPINI, G.: PEPPOLONI, S.; DE MAGISTRIS, M.T.; MARSILI, I.; RAPPUOLI,

R. Properties of the pertussis toxin mutant PT-9K/129G after formaldehyde treatment. In-

lect. Immun., 59: 625-630, 1991.

8. PODDA, A.; NENCIONI, L; DE MAGISTRIS, M.T.; Dl TOMMASO, A.; BOSSU', P.; NUTI, S.;

PILERI, P.; PEPPOLONI, S.; BUGNOLI, M.; RUGGIERO, P.; MARSILI, I.; D'ERRICO, A.;

TAGLIABUE, A.; RAPPUOLI, R. Metabolic, humoral and cellular responses in adult volun-

teers immunized with the genetically inactivated pertussis toxin mutant PT-9K/129G. J.

Exp. Med., 172: 861-868, 1990.

9. PODDA, A.; NENCIONI, L; MARSILI, I.; PEPPOLONI, S.; VOLPINI, G.; DONATI, D.; Dl TOM¬

MASO, A.; DE MAGISTRIS, M.T.; RAPPUOLI, R. Phase I clinicai trial of an acellular per¬

tussis vaccine composed of genetically detoxifiod pertussis toxin combined with FHA and

69K. Vaccine, 1990 (Submitted).

SciELO

10 11 12 13

Mem. Inst. Butantan, v. 53, supl. 1, p. 21-29, 1991

10. COWELL, J.L.; SATO, Y.; SATO, H.; AN DER LAN, B.; MANCLARK, C.R. Separation, purifi-

cation, and properties of filamentous hemagglutinin and leukocytosis promoting factor-

hemagglutinin from Bordetella pertussis. In: ROBBINS JB, HILL JC, SADOFF JC, eds.

Seminars in infectious disease. Bacterial vaccines. New York: Thieme-Stratton, 1982. v. 4.

p. 371-397.

11. RAPPUOLI, R. New and improved vaccines against diphtheria and tetanus. In: WOODROW

GC & LEVINE, MM eds. M. Dekker Inc., New generation vaccines, New York, 1990. p. 251 -

258.

12. HEWLETT, E.L.; SAUER, K.T.; MYERS, G.A.; COWELL, J.L.; GUERRANT, R.L. Induction of

a novel morphological response in Chinese Hamster Ovary Cells by pertussis toxin. Infect.

Immun., 40:1198-1203, 1983.

13. GILLENIUS, P.; JAATMAA, E.; ASKELOF, P.; GRANSTROM, M.; TIRU, M. The standardiza-

tion of an assay for pertussis toxin and antitoxin in microplate culture of Chinese Hamster

Ovary cells. J. Biol. Stand. ,13: 61-66, 1985.

14. MANCLARK, C.R.; MEADE, B.D.; BURSTYN, D.G. Serological response to Bordetella pertus¬

sis. In: ROSE NR, FRIEDMAN H, FAHEY JL, eds. Manual of clinicai laboratory immunolo-

gy. Washington, D.C.: American Society for Microbiology, 1986, p. 388-394.

2 3

5 6

11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 31-51, 1991

CONTRIBUTION OF MOLECULAR BIOLOGY TO VACCINE

DEVELOPMENT AND MOLECULAR EPIDEMIOLOGY

OF RABIES DISEASE*

Noel Tordo. Unité de la Rage, Institut Pasteur, Paris, France

I. RABIES DISEASE: AN HISTORICAL EXAMPLE OF INTERNATIONAL

COOPERATION, NOTABLY BETWEEN BRAZILAND FRANCE

In order to build an establishment devoted to the treatment of rabies disease based

on Louis Pasteur's famous "method of prevention of rabies disease after biting" 45 , a

worldwide foundation was organized. Besides numerous anonymous contributors, the

four first donators were the Tsar of Rússia, the Sultan of Turkey, Madame Boussicot

and the Emperor of Brazil, Dom Pedro II who crystalized by this action, a constant

friendship with Louis Pasteur. This first authentic example of international financial and

scientific collaboration resulted in the founding of the Pasteur Institute of Paris inaugu-

rated in 1888, and inspired numerous other projects throughout the world to fight the

rabies disease raging at the end of the 19th century. As early as 1888, Dom Pedro II

organized the building of a Pasteur Institute in Rio de Janeiro directed by Dr Ferreira

dos Santos who had studied the method of rabies prophylaxis in Paris 46 . Today, the

brazilian production of anti-rabies vaccines is dispersed in four places including the Bu¬

tantan Institute of São Paulo, which celebrates its 90th Anniversary in 1991.

II. CAUSES AND CONSEQUENCES OF RABIES DISEASE WORLDWIDE TODAY

Human deaths by rabies disease are estimated as at least 50,000 per year, although

the WHO records less alarmist statistics due to the difficulty of collecting data from some

* Summary of two conferences during the symposium.

Correspondence to: Noel Tordo, Unité de la Rage, Institut Pasteur, 25, rue du Docteur Roux,

75724, Paris, Cédex 15, France.

2 3

5 6

11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 31-51, 1991

countries (World survey of rabies 1988 no 24, WHO/Rabies/91.202). China and India are

probably the most conlaminated (25,000 cases each) followed by the Philippines, Thai-

land and Indonésia in Asia, most of África, México and Central America. In South Ameri¬

ca, besides human cases (200 to 500 in Brasil) rabies disease is also an economical

problem, responsible for considerable losses of cattle, approximately 500,000 per year 1 .

Indeed, the only continent to remain totally rabies-free is Oceania together with several

island countries such as Japan, United Kingdom, Ireland, as well as certain Antillas.

The perenniality of the disease is ensured by several wild animal species serving as

natural virus reservoirs. These reservoir species, also acting as vectors, are characte-

rized by their high susceptibility to the virus and their propensity to transmit the infec-

tion before dying. They must be distinguished from the other species (cattle, human,

etc...) that constitute a "cul de sac" for the infection.

The vector species are susceptible to considerable variation depending both on geo-

graphic and temporal criteria 1 °- 58 . Today, for example, the dog is the main reservoir in

Asia, África and Latin-America, the raccoon and the skunk in North-America, the vam-

pire bat in Central- and South-America, the red fox in Europe. Fox rabies is relatively

recent in Western-Europe where it penetrated from the far East (USSR). It first ap-

peared in France in 1968, replacing the dog or ''Street'' rabies contemporary of Louis

Pasteur which disappeared in 1920-1930 60 . Another recent european vector is the in-

sectivorous bat, initialiy described in the North of Europe and invading progressively to-

wards the South. The first french case was diagnosed late 1989 61 , others being report-

ed in Spain. Bat rabies is today a major problem because of its divergence from the

"classicar rabies viruses (see § VI. 1).

As reflected by the great variability of the host, rabies disease is caused by a group of

different agents describing the Lyssavirus genus of the Rhabdoviridae family (figure I) 4 .

RHABDOVIRIDAE FAMILY

LYSSAVIRUS GENUS

serotvpc

geoerapiiic_distrib.

animal_species

1. Rabies

world except Oceania,

U.K., Japan, islands...

carnivores

cattle, bats

human

2. Lagos bat

África: Nigéria, Zimbabwe

Ccnt. Afri. Rcp., South-Africa,

Senegal

frugivorous bats

cats

3. Mokola

África: Nigéria, Zimbabwe

Ccnt. Afri. Rep., Cameroon

shrews, rodents

cats, dogs,

human

4. Duvenhage

África: Zimbabwe

South-Africa

insectivorous bats

human

uatlassilitd

Obodhiang

África: Sudan

Mansonia

Kotonkan

África: Nigéria

Culicoids

EBL

Europe: Finland, France

insectivorous bats

(European Bat

Lyssavirus)

Figure 1 — Classification of the Lyssavirus genus.

SciELO

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl, 1, p. 31-51, 1991

These are classified on the basis of serological and antigenic relationships into four

serotypes 13 * 32 . Serotype 1 comprises the vaccinal strains and the "classical" wild ra¬

bies viruses. Serotype 2, 3 and 4 are the so called "rabies-related" viruses because of

their distant relationship with serotype 1; serotype 3 corresponds to Mokola virus; sero¬

types 2 and 4 group are bat viruses, represented by Lagos bat and Duvenhage iso-

lates, respectively. The recent european bat isolates form the European Bat Lyssavirus

(EBL) group, as yet unclassified 61 .

III. BRIEF DESCRIPTION OF THE DISEASE

Rabies is a virai infection of the nervous system affecting particularly mammals

where it causes an acute encephalitis 4 . The virus generally penetrates by effraction

(essentially biting) although some unclassical routes of infection by aerosols or licking

of mucous membranes have been described 19 . At the site of the bite, generally in the

externai tissue, a local multiplication is believed to occur but not obligatory 56 . The virus

is clearly neurotropic and tends to infect neurons of the peripheral nervous system 71 .

The putative receptor of the rabies virus is as yet unknown. On the basis of sequence

homology between the externai glycoprotein of the virus and the receptor binding site

of venom snake neorotoxins, it was postulated that the nicotinic acetylcholine receptor

could be the rabies receptor 7 ' 39 . But if this is possible in muscular cells, it seems that

at the levei of fibroblastic and above all neuronal cells, the rabies receptor(s) is(are)

more complex, also involving oligosaccharides or lipoproteinic elements such as the

sialic acids of gangliosides 71 ' 78 .

Once in the nerve, the virus replicates and ascends to the central nervous system

by retrograde axoplasmic flow. Some regions of the central nervous system are prefe-

rentially infected such as the córtex, the pons or the thalamus 71 . Late in infection, all

the central nervous system is infected as well as certain externai tissues, such as the

salivary glands, that aliow the perenity of infection. Rabies disease is characterized by

a variable incubation period (generally 1-3 months), contrasting with a short and violent

symptomatic period (less than 1 week) leading invariably to death in the absence of

any possible therapy 4 . Generally no evident histopathological perturbation accompa-

nies death, as if the virus killed the organism without killing the cell. Several electro-

physiological dysfunctions, notably at the levei of the paradoxical sleep, have been not-

ed in infected mice 29 .

IV. STRUCTURAL AND FUNCTIONAL STUDY OF THE VIRION (FIGURE 2)

1. Morphology, structure

If Pasteur was already certain that rabies disease was caused by a virus, over 80

years of technical progress was necessary, in tissue culture and electron microscopic

domains, before observing the first rabies particule in 1963, notably at the Pasteur Insti-

tute 3 ' 21 . The virion has a bullet-shaped form with a round extremity and a flat base.

The unique single-stranded RNA genome and the five viral-encoded proteins are dis-

tributed in two structural and functional units 67 : a central dense helical cylinder corre-

sponding to the ribonucleocapsid surrounded by a lipoproteinic envelope obtained from

2 3

5 6

11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 31-51, 1991

genomes

(-sense)

/TA

envelope

ENCAPSIDATION

TRANSCRIPTION

ribonucleocapsid

Figure 2 — Structure, transcription and replication mechanisms of rabies genome.

lhe host-cell surface during the budding of the virion. The ribonucleocapsid is constitut-

ed by the RNA genome tightly associated with the N nucleoprotein, and less stringently

with the Ml phosphoprotein and the L polymerase. The RNA-N protein association is

so intimate that the genome is completely insensitive to RNase. The envelope carries

the M2 membrane or matrix protein on the inner side and is traversed by the G glyco-

protein which forms spike-like glycosylated projections towards the outside. As is evi-

dent from their names, the G glycoprotein is glycosylated and the Ml phosphoprotein is

phosphorylated. The N nucleoprotein is also phosphorylated.

The rabies genome is an unsegmented RNA molecule of negative polarity. Negative

means that it is not directly translatable by the cell machinery, but forced to assure an

autonomous transcription step by producing positive stranded mRNAs, as soon as it

penetrates the cytoplasm. Unsegmented means that cistrons are juxtaposed, limited by

start and stop transcription signals and separated, by intergenic regions. This genomic

organization is shared by the Rhabdoviridae and Paramyxoviridae families which have

consistently envolved an identical multiplication strategy.

SciELO

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 31-51, 1991

2. Multiplication Strategy

Once fixed on its putative receptor, the vírus penetrates the cell by pinocytosis and

fuses its envelope with that of the lysosomal vacuole. The ribonucleocapsid is thereby

liberated into the cytoplasm to serve as a template for transcription and replication

mechanisms in this genuine form, without any decapsidation. These mechanisms have

been largely inspired from the vesicular stomatitis virus (VSV) model, far the more ex-

tensively studied unsegmented negative stranded RNA virus 5 - 26 ’ 73 ' 74 .

The transcription occurs from the 3’ to the 5' end of the genome, producing monocis-

tronic transcripts. First a small noncapped, nonpolyadenylated leader RNA, then five

capped and polyadenylated messenger RNAs (mRNAs) successively corresponding to

the N, Ml, M2, G and L proteins. This transcription is sequential and of decreasing effi-

ciency, meaning that a messenger is always transcribed after the 3' proximal one and

at lower rate. Consequently, the extent of gene expression is directly related to the ge-

nomic location. It appears that the transcription complex stops at the end of each cis-

tron, and reinitiates only partially at the beginning of the following one. This sequential

progression is dictated by the ten nucleotides long start and stop transcription signals

recognized by the running transcription complex.

It is only after the translation of the mRNAs into the corresponding proteins that the

switch to replication step occurs. This suggests that at least one of the virai protein is

involved in this switch. The N nucleoprotein ratio is currently thought to be the key

point, the replication beginning only if sufficiently large amounts of N protein are availa-

ble for encapsidation. The replication leads to the synthesis of a full length positive

stand antigenome that, in turn, will serve as a template to amplify the negative strand

genomes. These will be either encapsidated in the progeny virions or submitted to a

secondary transcription step.

Although not yet studied as completely, the rabies genome expression is coherent

with the VSV model but shows typical features:

1) the presence of very variable intergenes both in size and nucleotidic composition,

notably between G and L cistrons (423 bases) 67 - 69 ;

2) the presence of two consecutive stop signals for the G and M2 cistrons, alterna-

tively used to produce either a large or a small messenger. Because the transcription

complex is thereby released more or less far upstream from the distai start signal, alter-

native termination influences the extent of distai gene transcription. It is a typical regu-

latory mechanism since the ratio between both messengers varies during the course of

infection, and differently in fibroblastic or neuronal cells 64 - e5 - 6S .

3. Functional Role of the Virai Proteins

The ribonucleocapsid structure (N protein coated RNA genome, Ml and L proteins) is

a functional entity autonomous in transcription and replication. Typically, the N encapsi¬

dated genome is recognized as a template by the virai polymerase composed of two

functional elements. While the L protein is the actual RNA-dependent RNA polymerase

carrying most of the required activities (RNA synthesis, capping and polyadenylation),

the Ml seems more devoted to regulatory functions. It ensures notably a local decapsi¬

dation of the template upstream from the running L protein, by displacing the N proteins

to leave the template accessible for the polymerase. The affinity of the N protein mole-

cules for the Ml phosphoprotein results from the extreme electronegativity of the latter

2 3

5 6

11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 31-51, 1991

that mimics the RNA backbone. This electronegativity is due both to the richness of the

Ml protein in acidic amino-acids and to the presence of the phosphate residues. Down-

stream the polymerase complex, the N nucleoproteins immediately re-encapsidate the

genome, suggesting an exact coordination between polymerization and encapsidation.

The M2 membrane protein, occupying an intermediate position between the enve¬

lope and the ribonucleocapsid, interacts with both units. Thereby, it plays a capital role

during the maturation step which precedes the budding of the virion out of the cell. At

the ribonucleocapsid levei, it inhibits the transcription and replication mechanisms and

catalyses a strong condensation in an helical structure 40 . At the membrane levei, it des-

ignates the places where the virus will bud by facilitating local concentration of the G

glycoprotein on the cell surface. The externai spike-like glycosylated projections of the

G protein constitute the major virai antigen and are likely to mediate the binding to the

target cellular receptor although the binding site remains uncharacterized.

V. RABIES VACCINES: A CENTENARY OF CONTINUAL

TECHNICAL PROGRESSES FROM PASTEUR’S "DESSICATED SPINAL CHORDS"

TO GENETIC ENGINEERING

1. Classical Vaccines

To fight the disease, the first serious scientific approach was undertaken by Pierre-

Victor Galtier and subsequently Louis Pasteur, to find an effective vaccine. Since the

famous and successful Pasteur's injections of dessicated spinal chord of rabid rab-

bits 45 , considerable efforts were undertaken to increase the efficiency and safety of

vaccines. They were subsequently prepared from infected brains of adult animais, then

of suckling mice to avoid risks of encephalopathy, and finally from infected cell cul-

tures 17 ’ 34 ’ 59 . In parallel, performing methods to purify (zonal centrifugation) and inacti-

vate (UV, B-propriolactone) the virus were developed.

2. Subunit Vaccines without DNA Technology

The project of producing rabies subunit vaccines consists of using purified virai poly-

peptides or part of polypeptides for immunization. Biochemical techniques to purify virai

polypeptides and to restructure them in a convenient form were first assayed. The puri¬

fied glycoprotein was anchored either on an oligosaccharide bone composed of glyco-

side Quil A, or on a lipidic membrane, giving rise to "rabies iscoms" and "rabies immu-

nosomes", respectively 48 . Even though both products showed a high protective activity

in pre- or post-exposure tests, the difficulties inherent to glycoprotein purification ren-

dered impossible the development of a vaccine for commercial reasons. DNA-

recombinant technology could bypass this difficulty by producing substantial quantities

of virai polypeptides. Alternatively, a vaccinal approach by synthetic peptides corre-

sponding to B- or T-cell epitopes coupled to "carríer" molecules was investigated 17 .

3. Contribution of the DNA-recombinant Technology to Rabies Vaccine

The different steps for the general strategy to produce a rabies vaccine by DNA-

recombinant technology are (figure 3): 1) The choice of the virai antigen eliciting the

2 3

5 6

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 31-51, 1991

best immune response; 2) The choice to work either with the corresponding messenger

RNA or the total RNA genome; 3) After the reverse transcription into complementary

DNA (cDNA) and the cloning, the choice of the expression system, including the vec-

tor, the promotor and the host-cell system.

GENERAL STRATEGY FOR A

DNA-RECOMBINANT VACCINE

1) choice of the virai antigen

2) reverse transcription of the RNA gene

of the messenger RNA

3) cloning of the complementary DNA

4) choice of the expression system

(host cells - vector - promotor)

Procaryotes Eucaryotes

S. cerevisiae

E. coli

animal cells

5) purification of the recombinant protein

(unecessar when using virus vectors able

to multiply on the targeted animal)

6) protective power on animal

Figure 3 — Strategy towards a genetic engineered vaccine.

Assuming that expression is good, then the recombinant protein has to be purified,

except when using virus vectors able to multiply in the target animal itself, such as vac-

cinia vectors. Finally, the last and crutial step is to test the protective power of the re¬

combinant on animais.

2 3

5 6 SClELO 1Q 1X 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 31-51, 1991

A. Choice of the Virai Antigen

Although all virai proteins show antigenicity, they do not all play the same role in

protection 17 : the purified G protein has a protective effect against an intracerebral chal-

lenge with rabies virus while the purified ribonucleocapsid is only protective against a

less stringent peripheral challenge (intramuscular) 25 . Consistently, the glycoprotein is

the only virai protein having the capacity to elicit virus neutralising antibodies 75 . This

property is mainly dependent on the preservation of its three dimensional structure, al¬

though a linear neutralising epitope was recently characterized 16 . On the other hand, it

shares the capacity to induce a cellular immune response involving both helper and cy-

totoxic T lymphocytes with other proteins of the ribonucleocapsid, namely the N and

the Ml proteins 17 . The major epitopes for both B and T cells have been located along

the virai proteins 24 .

These studies suggest that if the glycoprotein is obviously the most important anti¬

gen for vaccination, the nucleoprotein could represent an interesting enhancer for two

principal reasons: 1) because of its capacity to substantially increase the helper T cell

immune response as is reflected by the numerous T-cell epitopes along the N protein;

2) because it is a less variable antigen, 13 ’ 67 ' 68 and could increase the spectrum of a

vaccine, notably to rabies-related viruses.

B. Working with Messenger RNAs or Total Genome

An mRNA template is easy to prime with an oligo (U), but impairs the study of the

untranscribed intergenic regions separating the cistrons, which could be important in

regulating transcription. This study is only accessible when cloning the whole genome

but the latter is not polyadenylated at the 3’ end. This obliges the need for a specific

primer deduced, for exampie, from the direct Chemical sequence of the 3' end of the

RNA genome 68 . During the last ten years of rabies molecular biology, both strategies

were adopted. The first rabies sequence to be published predicted the glycoprotein

structure of the ERA strain, and was obtained by the intermediate of its mRNA 2 . Alter-

natively, the genome of the PV strain was the first to be cloned to completion by using

three consecutive specific primers, and the sequence of its 11932 nucleotides deter-

mined 68 ’ 69 ' 70 .

Today, the structure of numerous rabies mRNAs from different strains are known 64 '

78 . The genome of the AvOI 49 and SAD B19 20 viruses, mutants of the CVS and SAD

strains, respectively, as well as the genome of the rabies-related Mokola virus 15 have

been cloned and totally or partially sequenced. All these cloned genes were then

available for expression. However, for historical reasons, almost all the expression as-

says were tempted on the ERA and CVS glycoproteins, the two first genes to be

cloned 2 ' 79 Now, the tendency is more towards diversity both at the levei of the nature

(G or N protein) and serotype (strains of rabies or rabies-related viruses) of the ex-

pressed antigen.

C. Choice of the Expression System

The key point is then to select an expression system able to perform the required

post-translational modifications to make the recombinant polypeptides identical to the

authentic protein.

2 3

5 6

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 31-51, 1991

intlg.

Rabies

slraln

Host

Vector

Promotor

Notes

Ref.

ERA

E. Coli

plasmid

lac UV5

2-5%

-inducible promotor

-unglycosylated

-unmatured (expression

without signal peptide)

-denatured (reduced)

-non protective

CVS

pBR 322

tryp

2-3%

ERA

plasmid

Ml 3 phage

25%

36,37

ERA

M13 phage

lac

B-Gal fused)

ERA

S. Cere-

visiae

2 microns

plasmid

weak

-glycosylated

31,37

high

eucaryote

cells

-CEF

-BHK

-VERO

-human

-etc...

plasmid

SV40 late

Herpes TK

weak

ERA

Bovine

papilloma

replicon

TK gene

weak

31,37

ERA

SV40

ERA

Adenovirus

CVS

Adenovirus

SV40 early

Adeno late

Adeno E3

protective power in vivo

Injection: high i

Oral : high |

ERA

Vaccinia

[Copenhagen)

7,5 KD

early-late

Injection: high j

Scarific : high I s 'd e effects

Oral : high j

CVS

Vaccinia

(N. Y. Board

of Health)

7,5 KD

early-late

11 KD

late

Injection: high j | ess

Scarific: high i side effects

Oral : weak |

CVS

Raccoon pox

7,5 KD

early-late

11 KD

late

Injection: high i host

Oral high i specificity

28,41

ERA

Fowlpox

H 6

early-late

Injection: high sp h e ° c f f ’ city

62,63

CVS

Spodoptorr

Fruglperda

Baculovirus

polyedrin

good

Protective power by

Injection: high

fVCK

Baculovirus

polyedrin

good

Protective power by

Injection: high

1 4

CVS

high

eucaryote

cells

Vaccinia

Copenhagen)

7,5 KD

early-late

CVS

Raccoon pox

7,5 KÜ

early-late

11 KD

late

high

Dose dependent

protection by injection

or tail scratching

CVS

Spodoptorc

Frugiperda

Baculovirus

polyedrin

high

Products for

diagnosis reagents

8,50,53

Figure 4 — Summary of the systems employed to express the G and N genes of Lys-

saviruses.

SciELO

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 31-51, 1991

The glycoprotein is a classic transmembrane protein with four typical domains from

the NH2 to the COOH end 67 ' 78 : 1) a 19 residues-long hydrophobic signal peptide; 2) a

hydrophilic and glycosylated externai ectodomain, approximatively 440 residues long;

3) an 22 residues-long transmembrane peptide; 4) an 85 residues long hydrophilic cy-

toplasmic domain. During the G mRNA translation, the first synthesized signal peptide

traverses the endoplasmic reticulum towards the lumen, initiating a translocation pro-

cess of the whole ectodomain. The translocation stops when the transmembrane pep¬

tide reaches the membrane where it stays anchored. A maturation process then

cleaves the signal peptide and induces the glycosylation of the ectodomain. Finally, the

mature protein appearing at the cell or virion surface is composed of the internai cyto-

plasmic domain, the transmembrane peptide and the glycosylated ectodomain protrud-

ing toward the outside.

The N mRNA translation is more simple, occuring totally in the cytoplasm, and pro-

ducing a 450 amino-acid long polypeptide which is then phosphorylated at precise

sites 23 .

Figure 4 summarizes the numerous procaryotic and eucaryotic expression systems

that were assayed.

D. Expression in Bactéria and Yeast

E. coli was extensively employed at the beginning, using several plasmid or virus

vectors and inducible promotors, but was rapidly discarded because of its inability to

glycosylate and cleave signal peptides. This required additional molecular biological tin-

kering to delete the signal sequence before expression. Although a heat inducible Ml3

phage promotor led to an important proportion of recombinant protein, approximatively

25% of the total proteins 36 ’ 37 , the resulting unglycosylated and denatured polypeptide

showed absolutely no protective power in vivo.

In yeast, the 2 microns plasmid expressed a correctly glycosylated recombinant pro¬

tein, but the production was too weak to envisage a genetic engineered vaccine 31 ■ 37 .

E. Expression in Animal Cells

In animal cells, several system were assayed, using either plasmid or viruses as

vectors (Bovine papilloma virus, SV40 virus, adenovirus) but the production of the gly¬

coprotein remained weak. Only a recent trial in adenovirus using three different promo¬

tors, allowed a substantial production of glycoprotein with a high protective power ei¬

ther by injection or by the oral route 52 .

But the real first success was obtained with the Copenhagen strain of the vaccinia

virus using the 7,5 Kd promotor 30 . The success of this system was both in expression

and protective effects obtained by injection, by scarification, or by oral route 11 ’ 37 ' 54 ' 76 '

77 Basically, a non-essential gene (here the TK gene), and a strong promotor (the 7,5

Kd) are isolated from the vaccinia virus genome. The rabies glycoprotein gene is

placed under the control of the promotor and this construction is inserted into the mid-

dle of the TK gene on a convenient plasmid vector. After cotransfection of the vector

with wild vaccinia genome in cells, double reciprocai recombination events using the

TK gene flanking sequences insert the rabies glycoprotein in the middle of the vaccinia

genome. This recombinant vaccinia virus is interesting because it is able to grow either

in cell culture, or directly in animais to be vaccinated.

2 3

5 6

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 31-51, 1991

lt was shown that laboratory animais injected intradermally or inoculated by the pa-

renteral route with the recombinant virus deveiop high titer of virus neutralising antibod-

ies and resistance to an intracerebral lethal challenge 76 ' 77 Injection with the vaccinia

wild type has no effect. Adult fox vaccinated by the oral route deveiop a neutralising an-

tibody titer still detectable between 1-1.5 years after inoculation and are consistently

protected against a lethal challenge 11,22 . Since a fox lives approximately less than two

years in the wild, this type of vaccination appeared particularly adapted. Oral vaccina-

tion campaigns by baits has been undertaken in Europe in the aim of eradicating the

wild- (essentially fox-) rabies 47 .

However, the Copenhagen strain of vaccinia virus that was initially used, was sus-

pected of inducing side effects for human and animais, as previously observed during

mass vaccination campaigns against smallpox 27 ■ 28 . This was, from the beginning, a

subject of controversy leading Copenhagen partisans to multiply the safety tests for an

incredible number of animal species 22, 54 , and Copenhagen detractors to look for new

vaccinia strains or new poxviral vectors, with limited host range. Several alternatives

were assayed. The New York Board of Health strain showing remarkably less side ef¬

fects was used with a more powerful promotor, the 11KD 27 . However, its protective ef¬

fect by the oral route appeared weak, perhaps because it is less invasive than the Co¬

penhagen strain. The racoonpox virus was also tested to improve the host-specificity

for the vaccine 28, 41 . Finally, the fowlpox virus, pathogen for poultry only, is still under

development 62, 63 . Its interest as a vector is that it is unable to make a productive in-

fection in non-permissive cells or non-permissive host. This greatly limits the possible

side effects, notably a generalized infection to the vector in immunocompromised indi¬

viduais. Despite this non-productive infection, fowlpox vector is able to express foreign

genes, and particularly the rabies glycoprotein, at the surface of non permissive cells

and thereby induces interesting leveis of neutralizing antibodies by inolucation of sever¬

al non permissive mammalian hosts.

More recently, expression of the glycoprotein gene in the baculovirus system was

performed for CVS strain rabies virus 51 as well as for the rabies-related Mokola virus 14 .

In this latter case, baculovirus appeared as a particularly convenient vector because, in

contrast to rabies virus, Mokola virus grows efficiently in insect cells and was essential¬

ly isolated from insectivores in África. It was notably responsible for human encephali-

tis, death of domestic animais vaccinated against rabies, and for a limited epizooty in

Zimbabwe 13 . The absence of cross protection by the classical rabies vaccines being

verified in the laboratory, a specific vaccine against Mokola virus is necessary to pro-

tect exposed populations. The low growth efficiency of the virus in cell culture hypothe-

sized the initial project of a classical cell culture vaccine and favours the search for an

alternative baculovirus recombinant vaccine (unpublished results).

The technique used (figure 5) was basically similar to vaccinia virus. The Mokola gly¬

coprotein gene was isolated, placed under the control of the polyhedrin promotor in a

pEV55 transfer vector. Here again, the polyhedrin is not essential for the baculovirus and

a cotransfection of the pEV55-GMok construct with the wild virus in SF9 insect cells, ex-

changes the polyhedrin gene with the Mokola glycoprotein in the recombinant virus.

When cells are infected with the recombinant virus, they synthesize a substantial

quantity of a 56 kd recombinant protein similar both by electrophoresis and immuno-

blotting techniques to the native glycoprotein. In the presence of tunicamycin, which in-

hibits the N glycosylation, the apparent molecular weight of both recombinant and na¬

tive protein are reduced to a similar extent, suggesting their effective glycosylation. By

2 3

5 6

11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 31-51, 1991

Fnud II Ssp I

(3180) B 9 1 11 (5205)

(4035)

3' | H | Ml | M2 | G | L ^ 5’

• •••••••

0 12 3 4 5 6 7 Kt>

pMD 10

pMAl O

Figure 5 — Cloning of the coding sequence of the Mokola glycoprotein in a baculovirus

vector. The cleavage sites of the restriction enzymes Bglll, Fnudll and Sspl are indicat-

ed. pMDIO and pMAIO cDNAs are ligated by their common Bglll site to form the pM7

cDNA. The Fnudll-Sspl internai fragment is excised, trimmed at the G protein NH2 side

by the Bal31 exonuclease, and placed under the control of the polyedrin promotor in a

pEV55 transfer vector.

treating the infected cells with a glycoprotein-specific monoclonal or a polyclonal anti-

body, the recombinant protein is shown to be expressed at the cell surface, as expect-

ed from its transmembrane nature. The protective power of the recombinant protein

was analyzed. Mice injected intraperitonealy with insect cells expressing the recombi¬

nant glycoprotein resist an intracerebral challenge performed three weeks later. The

protection is clearly due to the recombinant protein, since insect cells expressing the

wild baculovirus are inefficient. It is specific for Mokola virus since mice remain sensi-

tive to a challenge with the rabies CVS strain. The baculovirus recombinant Mokola gly¬

coprotein is thus protective in vivo and constitutes the first experimental vaccine

against a rabies-related virus with an excellent protective index. Additional purification

and restructuration of the recombinant protein is now necessary before proposing a

vaccine for human or veterinary use.

The expression of the IM nucieoprotein gene was also performed in vaccinia 9,41 and

baculovirus sectors 50 ' 53 .

SciELO

10 11 12 13

Mem. Inst. Butantan, v. 53, supl. 1, p. 31-51, 1991

VI. ARE THE CURRENT VACCINAL STRAINS ADAPTED

TO THE CURRENT RABIES DISEASE?

1. The Vaccinal Strains

In contrast to the continuai technical progresses in vaccine developments, very little

was attempted to concomitantly adapt the vaccinal strains to rabies virus evolution.

Consequently, the strains used for medicai or veterinary vaccines, derive from isolates

obtained between 50 and 100 years ago, following a complex history summarized in

figure 6 and described in references 18 - 38

Most of them derive from the original Pasteur's isolate collected from a rabid cow in

the suburb of Paris in 1882. They are therefore reminiscent of the dog-rabies that over-

ran Western-Europe at the end of the 19th century and ignore the recent switch to the

fox-rabies (see § II). The Pasteur's isolate was serially passaged on rabbit brains to

give rise to the first "fixed" rabies strain (L. Pasteur strain) characterized by a constant

and species-dependent incubation period. The L. Pasteur strain is still traditionally

maintained at the Pasteur Institute of Paris (2074 passages) and has been adapted to

cell culture. It was largely disseminated among the scientific and industrial communities

to generate numerous "fixed" strains serving both for molecular studies and vaccine de-

velopment. In 1940, thé L. Pasteur strain penetrated the USA where it was maintained

on rabbit brain and adapted to mouse brain by Karl Habel, resulting in the PtvTand CVS

strains, respectively. It carne back to Paris in 1965, after an obscure period in the Ce-

panzo of Buenos Aires, and gave rise to the PV strain.

Other classical strains alternatively derived from different isolates. In the USA, a rab¬

id dog originated the SADs, SAG1, ERA and Vnukono32 strains, while a human isolate

adapted to chick brain resulted in the Flury LEP and HEP strains. The Beijing31 and

Kelev strains were isolated in China and Israel, respectively.

The letigimate question arising from figure 6 is the ability of current vaccinal strains

to protect against the current rabies virus. Cross-protection studies have established

that current rabies vaccines (serotype 1): 1) are globally efficient against members of

the homologous serotype 1, although isolated cases of vaccination failures were report-

ed in África 12 ; 2) offer an imperfect protection dependent on the vaccinal strain used

against serotypes 2 and 4 35 ; 3) are ineffective against Mokola virus that constitutes the

most divergent serotype 3 33 .

This strongly plaids for the need for considerable efforts either to extend rabies vac¬

cine potency towards a polyvalent Lyssavirus vaccine, or to propose new specific vac¬

cines, particularly for viruses recognized as potentially dangerous for public health,

such as Mokola virus 13 . As a logical first step in that goal, an intensiva molecular epi-

demiological study of rabies virus appears urgent in order to appreciate the worldwide

virai evolution and the antigen divergence at the genetic levei.

2. Polymerase Chain Reaction (PCR) as an Alternative Tool for Diagnosis,

Typing and Molecular Epidemiology of Rabies Virus

We have recently developed a very simple method based on PCR amplification of

infected brain material which appears as a hopeful alternative for routine diagnosis, typ¬

ing or precise molecular epidemiological studies 55 . At the epidemiological levei, this

method is advantageous to classical approaches based on antigenic differences with

SciELO

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 31-51, 1991

SciELO

Figure 6 — Conspectus of the schematic history of the principal rabies virus strains

used for vaccine development.

Mem. Inst. Butantan, v. 53, supl. 1, p. 31-51, 1991

anti-N or -G monoclonal antibodies 24 - 43 ' 57 ■ 58 ' 72 because it does not require a previ-

ous cell culture adaptation setp, presumed to be possible source or selector of muta-

tions. Using a disposable plastic pipette, brain samples are internally collected via the

occipital foramen 6 or the retroorbital route 44 to avoid risks of contamination. Total RNA

is extracted and reverse transcribed into cDNA which is consecutively amplified. The

rabies specific primer used for reverse transcription can be one element of the couple

used for amplification or different. One can amplify either genomic (minus sense) or

messengers (plus sense) RNAs by selecting the convenient polarity for the reverse

transcription primer. The amplified fragment can be progressively processed for:

- diagnosis, by electrophoresis on ethidium bromide agarose gel,

- diagnosis or typing, by differential hybridization with internai probes of divergent

strains, either on dot or on Southern blots,

- typing, by restriction polymorphism with a limited panei of endonucleases,

- molecular epidemiology, by direct nucleotide sequencing of the fragment excised from

agarose gel (Nusieve), without any additional purification, using the dideoxy technique

(T7 Sequencing Kit, Pharmacia),

- cloning and expression in convenient vectors for fundamental or vaccinal purpose.

3. The Rabies \y Pseudogene: The Best Clock of Evolution

The genomic areas targeted for amplification are different for diagnosis or epidemio-

logical purposes. Conserved regions are more suitable for diagnosis which looks for

minute quantities of virai sequences, whatever the infecting Lyssavirus. Highly variable

regions are the more convenient for typing or molecular epidemiological studies where

the important point is to find sensitive criteria to differentiate isolates.

From a comparison between rabies and Mokola genomes 15 ' 55 ' 68 ' 69 ' 70 , which rep-

resent the two most divergent serotypes of the Lyssavirus genus, the N gene and the

G-L intergenic regions were targeted for diagnosis and epidemiological studies, respec-

tively (figure 7). The latter corresponds to a remnant protein gene, baptized \\i for

pseudogene, placing the rabies virus in an intermediate position of unsegmented nega-

tive stranded RNA virus evolution, between the Rhabdoviridae and the Paramyxoviri-

dae families 67 ' 69 . As a non protein coding region greatly susceptible to mutations, it is

more likely to represent the natural evolution of the virus outside any externai selective

pressure and therefore the most suitable target for epidemiological studies.

4. Towards a Worldwide Molecular Epidemiological Study of Lyssaviruses

Using primers located in conserved places of the flanking G and L, the gene of

worldwide fixed or wild Lyssaviruses has been successfully amplified and sequenced.

The totality of the G and N genes of several isolates were also studied. In terms of di-

vergence, the isolates rank in the same order, although at different rates, by consider-

ing either the coding G and N genes of immunological importance, or the non-coding

pseudogene, respectively. This assesses "a posteriori" the significance of our rational

approach of the virai divergence by the gene study, confirming this rapid method as

particularly adapted for rabies epidemiological studies.

Although the results are still under exploitation, several major observations summar-

ized in table I can be already shown (unpublished results):

1) The "fixed" strains used in vaccines form a dispersed group showing up to 18%

2 3

5 6

11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 31-51, 1991

1 2 3 4 5 6 Kb

Rabies genome (PV strain)

| | Protein coding region

Figure 7 — Sequence comparison of the rabies (PV strain) and Mokola genome se-

quence. Diagonal lines indicate homologous areas. Regions convenient for diagnosis

or typing-epidemiology purposes are noted.

internai divergence in the \y gene (not shown).

2) The wild isolates of geographic proximity from relatively homogeneous groups.

However, groups are substantially divergent from each other as well as from the vacci-

nal strains.

3) The West-African isolates, notably those suspected of vaccination failures, are

consistently more divergent from the vaccinal strains than the French isolates against

which the vaccines are clearly effective.

4) Bat isolates recently invading Europe or already present in Latin-America (Brasil,

Guyana, not shown) exhibit a divergence almost as important as Mokola virus against

which the rabies vaccines are clearly ineffective.

SciELO

10 11 12 13

Mem. Inst. Butantan, v. 53, supl. 1, p. 31-51, 1991

TABLE

Divergence (%) between wild Lyssavirus groups and the classical vaccinal strain PV:

France (12 isolates); West-Africa (Ivory Coast, Cameroon, Niger, Guinea, Marocco,

etc...); European bat (France, Poland, Finland).

ygene/vaccine

Ggene/vaccine

\|//group

France

14-15%

2.5%

West-Africa

25-30%

15%

Europ. bat

high

30%

Mokola

high

40-45%

VII. MAIN CHALLENGES FOR THE FUTURE

The recent progress in the molecular understanding of the rabies virus have not pro-

vided, so far, pertinont responses to the major enigmas of the disease itself. The rea-

sons for the rabies virus neurotropy are still not understood, despite numerous hypoth-

eses attempting to correlate the susceptibility of each cell type with the presence of

specific receptors. Perhaps, the secret of neurotropy will reside in the study of more

distai infection events, such as the influence of tissue-specific transcription factors on

the mechanism of rabies genome expression. Furthermore, the mystery remains

cloudy as to the nature of neuronal dysfunctions resulting in lethality, although their un¬

derstanding would be a capital step towards an effective therapy.

Despite these numerous unsolved questions, the availability of any virai gene for

studying and expression, is the most impressive contribution of the last ten years. In

the near future, examination of virai isolates from various regions of the world will per-

mit an evaluation of the spacio-temporal evolution of the virus and the influence of the

host. It will be interesting to understand the basis of the cross-protection at the se-

quence levei, in order to decipher whether the current vaccinal strains are sufficient for

animal and human health, or if additional specific vaccines are required. These could

be performed by DNA-recombinant techniques, taken into account the progress in ra¬

bies virus immunology. For example, synthetic structures carrying a recombinant glyco-

protein(s) anchored on their surface, and containing a recombinant nucleoprotein(s) (or

T-peptides) should be promising in the goal of a genetic engineering vaccine against all

(or most of) the members of the Lyssavirus genus.

ACKNOWLEDGMENTS

The unpublished results presented throughout the text were realized in the Rabies

Unit of the Pasteur Institute in Paris by: Hervé Bourhy, Responsible of the Diagnostic La-

boratory at the National and OMS Reference Centre for Rabies; Débora Sacramento,

who prepared her PhD in 1986-1991 and is presently at the Butantan Institute of Sao

Paulo (Brasil); Anne Kouznetzoff and Hassan Badrane currently preparing their PhDs.

SciELO

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 31-51, 1991

Acknowledgements are due to Pr. Pierre Sureau, head of the Rabies Unit, in whose

Laboratory this work was carried out. We thank Dr. Brian Lockhart for criticai reading of

the manuscript.

REFERENCES

1. ACHA, P.N. Buli. Ofl. Int. Epiz., 67:343 382, 1967.

2. ANILIONIS, A.; WUNNER, W.H.; CURTIS, P.J. Structure of the glycoprotein gene in rabies vi-

rus. Nature, 294: 275-278, 1981.

3. ATANASIU, P.; LÉPINE, P.; SISMAN, J.; DAUGUET, C.; WETTEN, M. Etude Morphologique du

vírus rabiquedes rues en culture de lissu. C. R. Acad. Sei. (Paris), 256: 3219-3221,1963.

4. BAER, G.M.; BELLINI, W.J.; FISHBEIIM, D.B. Rhabdoviruses. In: FIELDS, B.N. & KNIPE,

D.M., eds. Virology. New York, Ravon Press, 1990, p. 883-930.

5. BANERJEE, A.K. Transcription and replication of rhabdoviruses. Microbiol. Rov., 51: 66-87,

1987.

6. BARRAT, J. & HALEK, H. Simplified and adoquate sampling and preservation techniques for ra-

bios diagnosis in Moditerrancan countries. Comp. Immunol. Microbiol. Inf. Dis., 9:= 10,1986.

7. BECKER, Y. Putative antigenic domains in glycoprotein G of rabies virus: is the RGK se-

quenco involved in vírus adsorption receptors? Virus Ganes, 3: 277-284, 1990.

8. BELLINI, J.W.; RIED-SANDEN, F.L.; SUMNER, J.W.; SMITH, J.S. Expression of rabies nu-

cleoprotein (N) gone using a baculovirus recombinant: antigenic and immunologic proper-

ties of the N protein expressed in an insect coll line. In: Proceedings of tho 22nd Joint Virai

Diseases Panei Meeting of the Japan-U.S. Tokyo, Japan, Cooperative Medicai Science

Program,1988, p. 53.

9. BELLINI, J.W.; SUMNER, J.W.; FEKADU, M ; SCHADDOCK, J.H.; ESPISITO, J.J. Expres¬

sion of rabies nucleoprotein (N) gene by recombinant vaccinia virus. In: Proceedings of the

22nd Joint Virai Diseases Panei Meeting of the Japan-U.S. Tokyo, Japan, Cooperative

Medicai Science Program, 1988, p. 51.

10. BLANCOU, J. Epizootiology of rabies: Eurasia and África. In: CAMPBELL, J.B. & CHARLTON,

K.M., eds., Rabies. Boston, Kluwer Academic, p. 243-265, 1988.

11. BLANCOU, J.; KIENY, M.P.; LATHE, R.; LECOCQ, J.P.; PASTORET, P.P.; SOULEBOT, J.P.;

DESMETTRE, P. Oral vaccination of tho fox against rabius using a live recombinant vacci¬

nia virus. Nature, 322:373-375, 1986.

12. BOURHY, H.; LAFON, M.; BERTHONNEAU, M.C.; RENNER, Y.; ROLLIN, P.E.; SUREAU, P.

Rabies in vaccinatod dogs in Gabon. I /et. Record, 122: 361-362, 1988.

13. BOURHY, H.; SUREAU, P.; TORDO. N. From rabies to rabies-related viruses. Vet. Microbiol.,

23: 115-128, 1990.

14. BOURHY, H. & TORDO, N. Etude moléculaire des Lyssavirus apparentés à la rage. Ann.

Rech. Vót.,21: 316-319, 1990.

15. BOURHY, H.; TORDO, N.; LAFON, M.; SUREAU, P. Complete cloning and molecular organi-

zation of a rabies-related virus: Mokola virus. J. Gen. Virol., 70: 2063-2074, 1989.

16. BUNSCHOTEN, H.; GORE, M.; CLAASSEN, I.J.; UYTDEHAAG, F.G.; DIETZSCHOLD, B.;

WUNNER, W.H.; OSTERHAUS, A.D. Charactorization of a new virus-neutralizing epitope

that denotes a sequential determinant on the rabies virus glycoprotein. J. Gen. Virol., 70:

291-298, 1989.

17. CELIS, E.: RUPPRECHT, C.E.; PLOTK1N, S.A. New and improved vaccinos against rabies.

In: WOODROW, G.C. & LEVINE, M.M., eds., New genemtion vaccines, New York, Marcei

Decckcr, 1990. p. 419-439.

18. CLARK, H.F. & VIKTOR, T.J. Rabies virus. In: MAJER, M. & PLOTKIN, S.A., cds„ Strains of

Human Viruses, Basel, Karger, 1972. p. 177-182.

19. CONSTANTINE, D.G. Rabies transmission by air in bat caves. Atlanta, Public Health Service

2 3

5 6

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 31-51, 1991

Publication, 1967. 51 p.

20. CONZELMANN, K.K.;COX, J.H.;SCHNEIDER, L.G.;THIEL, H.J. Molecularcloning and complete

nucleotidesequenceof theattenuated rabies virusSADB19. Virology, 175: 485-489,1190.

21. DAVIES, M.C.; ENGLERT, M.E.; SHARPLESS, G.R.; CABASSO, V.J. The electron microsco-

py of rabies virus in cultures of chicken embryo tissues. Virology, 41: 821-832, 1963.

22. DESMETTRE, P.; LANGUET, B.; CHAPPUIS, G.; BROCHIER, B.; THOMAS, I.; LECOCQ,

J.P.; KIENY, M.P.; BLANCOU, J.; AUBERT, M.; ARTOIS, M.; PASTORET, PP. Use of vac-

cinia virus recombinant for oral vaccination of wildlife. Vet. Microbiol., 23: 227-236, 1990.

23. DIETZSCHOLD, B.; LAFON, M.; WANG, H.; OTVOS, L; CELIS, E.; WUNNER, W.H.; KO-

PROWSKI, H. Localization and immunological characterization of antigenic domains of the

rabies virus internai N and NS proteins. Virus Res., 8: 103-105, 1987.

24. DIETZSCHOLD, B.; RUPPRECHT, C.E.; TOLLIS, M.; LAFON, M.; MATTEI, J.; WIKTOR,

T.J.; KOPROWSKI, H. Antigenic diversity of the glycoprotein and the nucleocapsid proteins

of rabies and rabies-related viruses: implications for epidemiology and control of rabies.

Rev. Infect. Dis., 10: S785-S798, 1988.

25. DIETZSCHOLD, B.; WANG, H.; RUPPRECHT, C.E. Induction of protective immunity against

rabies by immunization with rabies virus ribonucleoprotein. Proc. Natl. Acad. Sei. USA, 84:

9165-9169, 1987.

26. EMERSON, S.U. Transcription of vesicular stomatitis virus. In: WAGNER, R.R., ed., The

Rhabdoviruses, New York, Plenum, 1987, p. 245-269.

27. ESPOSITO, J.; BRECHLING, K.; BAER, G.; MOSS, B. Vaccina virus recombinants express-

ing rabiesvirus glycoprotein protect against rabies. Virus genes, 1: 7-21, 1987.

28. ESPOSITO, J.J.; KNIGHT, J.C.; SHADDOCK, J.H.; NOVEMBRE, F.J.; BAER, G.M. Success-

ful oral rabies vaccination of raccoon poxvirus recombinants expressing rabies virus glyco¬

protein. Virology, ?65:313-316, 1988.

29. GOURMELON, P.; BRIET, D.; COURT, L.; TSIANG, H. Electrophysiological and sleep altera-

tions in experimental mouse rabies. Brain Res., 398: 128-140, 1986.

30. KIENY, M.P.; LATHE, R.; DRILLIEN, R.; SPEHNER, D.; STORY, S.; SCHMITT, D.; WIKTOR,

T.J.; KOPROWSKI, H.; LECOCQ, J.P. Expression of rabies virus glycoprotein from a re¬

combinant vaccinia virus. Nature, 312: 163-166, 1984.

31. KIENY, M.P.; LATHE, R.; LEMOINE, Y.; DRILLIEN, R.; WIKTOR, T.J.; KOPROWSKI, H.; LE¬

COCQ, J.P. Rage et génie génétique: développement de nouveaux vaccins. In: Vaccins et

vaccinations, Paris, Elsevier, 1985, p. 83-95.

32. KING, A. & CRICK, J. Rabies-related viruses. In: CAMPBELL, J.B. & CHARLTON, K.M., eds.,

Rabies, Boston, Kluwer Academic, p. 177-199. 1988.

33. KOPROWSKI, H.; WIKTOR, T.J.; ABERSETH, M.K. Cross-reactivity and cross-protection: ra¬

bies variants and rabies-related viruses. In: KUWERT, E.; MÉR1EUX, C.; KOPROWSKI,

H.; BÕGEL, K., eds., Rabies in the tropies, Wien, Springer-Verlag, 1985.

34. LABORATO, O.M.S. Laboratory techniques in rabies. 3 ed., Geneva, World Health Organiza-

tion, 1973, 367 p.

35. LAFON, M.; BOURHY, H.; SUREAU, P. Immunity against the European bat rabies (Duven-

hage) virus induced by rabies vaccines: an experimental study in mice. Vaccine, 6: 362-

368, 1988.

36. LATHE, R.; KIENY, M.P.; SCHMITT, D.; CURTIS, P.; LECOCQ, J.P. M13 bacteriophage vec-

tors for the expression of foreign proteins in Escherichia coli: the rabies glycoprotein. J.

Mol. Appl. Gen., 2: 231, 1984.

37. LECOCQ, J.P.; KIENY, M.P.; LEMOINE, Y.; DRILLIEN, R.; WIKTOR, T.J.; KOPROWSKI, H.;

LATHE, R. New Rabies vaccines: recombinant DNA approaches. In: KOPROWSKI, H. &

PLOTKIN, S.A., eds., World‘s debt to Pasteur, New York, Alan R. Liss, 1985, p. 259-271.

38. LECOCQ, J.P.; ZUKOWSKI, M.; LATHE, R. Cloning and expression of virai antigens in Esche¬

richia coli and other microorganisms. Methods in Virology, 8: 121, 1984.

39. LENTZ, T.L.; WILSON, P.T.; HAWROT, E.; SPEICHER, D.W. Amino acid sequence similarity

2 3

5 6

11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 31-51, 1991

between rabies glycoprotein and snake venom curaremimetic neurotoxins. Science, 226:

847-848, 1984.

40. LOCKHART, B.P. & TSIANG, H. Actin-independent maturation of rabies virus in neuronal cul-

tures. J. Gen. Virol., (in press), 1991.

41. LOPMELL, D.L.; SUMMER, J.W.; ESPOSITO, J.J.; BELLINI, W.J.; EWALT, L.C. Raccoon

poxvirus recombinants expressing the rabies virus nucleoprotein protect mice against lethal

rabies virus infection. J. Virol., 65:3400-3405, 1991.

42. MALEK, L.T.; SOOTSMEYER, G.; GARVIN, R.T.; JAMES, E. The rabies glycoprotein gene is

expressed in E. co//asadenaturedpolypeptide. In: R.A., C.R.M.a.L, eds., Modern Approach-

es to Vaccine, ColdSpring Harbour, ColdSpring Harbour Laboratories, 1983, p. 203-208.

43. MONTANO HIROSE, J.A.; BOURHY, H.; LAFON, M. A reduced panei of anti-nucleocapsid

monoclonal antibodies for bat rabies virus identification in Europe. Res. Virology, 747:571,

1990.

44. MONTANO HIROSE, J.A.; BOURHY, H.; SUREAU, P. Retro-orbital route for the collection of

brain specimens for rabies diagnosis. Vct. Record, 1991.

45. PASTEUR, L. Móthode pour prevenir la rage après morsure. C. R. Acad. Paris, Sôance, Sei:

765-774, 1885.

46. PASTEUR, L. Présentation, au nom de S.M. Dom Pedro, d'une collection de photographies et

d'une note relative à la statistique du traitement de la rage au Brésil. C. R. Acad. Paris,

Sôance, Sei: 847-848, 1888,

47. PASTORET, P.P.; BROCHIER, B.; LANGUET, B.; THOMAS, I.; PAQUOT, A.; BAUDIN, B.;

KIENY, M.P.; LECOCQ, J.P.; DE BRUYN, J.; COSTY, F.; ANTOINE, H.; DESMETTRE, P.

First field trial of fox vaccination against rabies using vaccinia-rabies recombinant virus.

Vet. Rec., 723:481-483, 1988.

48. PERRIN, P.; LAFON, M.; SUREAU, P. Rabies vaccines from Pasteur's time up to experimen¬

tal subunit vaccines today. In: MIZRAHI, A., eds., Virai Vaccines, New York, Wiley-Liss,

1990. p. 325-345.

49. POCH, O.; TORDO, N.; KEITH, G. Sequence of the 3386 3' nucleotides of the genome of the

AvOI strain rabies virus: structural similarities of the protein regions involved in transcrip-

tion. Biochimie, 70: 1019-1029, 1988.

50. PREHAUD, C.; HARRIS, R.D.; FULOP, V.; KOH, C.-L; WONG, J.; FLAMAND, A.; BISHOP,

D.H.L. Expression, characterization, and purification of a phosphorylated rabies nucleopro¬

tein synthesized in insectcells by baculovirus vectors. Virology, 776:486-497, 1990.

51. PREHAUD, C.; TAKEHARA, K.; FLAMAND, A.; BISHOP, D.H.L. Immunogenic and protective

properties of rabies virus glycoprotein expressed by baculovirus vectors. Virology, 173:

390-399, 1989.

52. PREVEC, L; CAMPBELL, J.B.; CHRISTIE, B.S.; BELBECK, L; GRAHAM, F.L. A recombi¬

nant human adenovirus vaccine against rabies. J. Infect. Dis., 767:27-30, 1990.

53. RIED-SANDEN, F.L.; SUMNER, J.W.; SMITH, J.S.; FEKADU, M.; SHADDOCK, J.H.; BELLI-

Nl, W.J. Rabies diagnosis reagents prepared from a rabies N gene recombinant expressed

in baculovirus. J. Clin. Microbiol., 28: 858-863, 1990.

54. RUPPRECHT, C.E. & KIENY, M.P. Devolopmont of a vaccinia-virus glycoprotein recombinant

virus vaccine. In: CAMPBELL, J.B. & CHARLTON, K.M. , eds., Rabies, Boston, Kluwer

Academic 1988, pp. 335-364.

55. SACRAMENTO, D.; BOURHY, H.; TORDO, N. PCR technique as an aiternative method for di¬

agnosis and molecular epidemiology of rabies virus. Mol. Cel. Probes, 5: 229-240, 1991.

56. SHANKAR, V.; DIETZSCHOLD, B.; KOPROWSKI, H. Direct entry of rabies virus into the cen¬

tral nervous System without prior local replication. J. Virol., 65: 2736-2738, 1991.

57. SMITH, J.S. Rabies virus epitopic variation: use in ecologic studies. In: MARAMOROSCH, K.,

MURPHY, F.A. & SHATKIN, A.J., eds., Advances in virus research, San Diego, Academic

Press, 1989, pp. 215-253.

58. SMITH, J.S. & BAER, G.M. Epizootiology of rabies: the Américas. In: CAMPBELL, J.B. &

2 3

5 6

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 31-51, 1991

CHARLTON, K.M., eds., Rabies, Boston, Kluwer Academic, 1988, p. 267-299.

59. SUREAU, P. Rabies vaccine production in animal cell cultures. Adv. Biomed. Engineer. Bio-

tec., 34:111-128, 1987.

60. SUREAU, P. & BOURHY, H. Des renards et des hommes ... "Histoires naturelles" sur les

risques de contamination par des animaux sauvages enragés, en France, en 1989. Ann.

Inst. Past./ActuaUtés, 3: 204-205, 1990.

61. SUREAU, P.; BOURHY, H.; LAFON, M. La rage des chauves souris en France. Buli. Epidemi-

ol. Hebdo., 40: 167, 1989.

62. TAYLOR, J. & PAOLETTI, E. Fowlpox virus as a vector in non-avian species. Vaccine, 6:466-

468, 1988.

63. TAYLOR, J.; WEINBERG, R.; LANGUET, B.; DESMETTRE, P.; PAOLETTI, E. Recombinant

fowlpox virus inducing protectiveimmunity in non-avian species. Vaccine, 6: 497-503,1988.

64. TORDO, N. Génétique molôculaire du virus de la rage, un siècle après Pasteur. Doctoral The-

sis, Paris VII. 1988.

65. TORDO, N.; BOURHY, H.; SACRAMENTO, D. Les Rhabdovirus: classification, structure,

mécanismes généraux, épidémiologie moléculaire. Ann. Rech. Vét., 21: 310-314, 1990.

66. TORDO, N. & POCH, O. Strong and weak transcription signals within the rabies genome. Vi¬

rus Res., sup.2: 30, 1988.

67. TORDO, N. & POCH, O. Structure of rabies virus. In: CAMPBELL, J.B. & CHARLTON, K.M.

eds., Rabies, Boston, Kluwer Academic, 1988, p. 25-45.

68. TORDO, N.; POCH, O; ERMINE, A.; KEITH, G. Primary stucture of leader RNA and nucleo-

protein genes of the rabies genome: segmented homology with VSV. Nucleic Acids. Res.,

14: 2671-2683,1986.

69. TORDO, N.; POCH, O.; ERMINE, A.; KEITH, G.; ROUGEON, F. Walking along the rabies ge¬

nome: is the large G-L intergenic region a remnant gene? Proc. Natl. Acad. Sei. USA, 83:

3914-3918, 1986.

70. TORDO, N.; POCH, O.; ERMINE, A.; KEITH, G.; ROUGEON, F. Completion of the rabies vi¬

rus genome sequence determination: highly conserved domains among the L (polymerase)

proteins of unsegmented negative-strand RNA viruses. Virology, 765:565-576, 1988.

71. TSIANG, H. Interactions of rabies virus and host cells. In: CAMPBELL, J.B. & CHARLTON,

K.M. eds., Rabies, Boston, Kluwer Academic Publishers, 1988. p. 67-100.

72. VINCENT, J.; BUSSEREAU, F.; SUREAU, P. Immunological relationships between rabies vi¬

rus and rabies-related viruses studied with monoclonal antibodies to Mokola virus. Ann.

Inst. Pasteur/Virol., 739:157-173, 1988.

73. WAGNER, R.R. Rhabdoviridae and their replication. In: FIELDS, B.N. & KNIPE, D.M., eds.,

Virology, New York, Raven Press, 1990, p. 867-881.

74. WERTZ, G.W.; DAVIS, N.L.; PATTON, J. The role of proteins in vesicular stomatitis virus RNA

replication. In: WAGNER, R.R., eds., The Rhabdoviruses, New York, Plenum Press,

1987, p. 271-296.

75. WIKTOR, T.J.; GYÒRGY, E.; SCHLUMBERGER, H.D.: SOKOL, F.; KOPROWSKI, H. Anti-

genic properties of rabies virus components. J. Immunol., 7 70:269-273, 1973.

76. WIKTOR, T.J.; MACFARLAN, R.I.; DIETZSCHOLD, B.: RUPPRECHT, C.E.; WUNNER, W.H.

Immunogenic properties of vaccinia recombinant virus expressing the rabies glycoprotein.

In: Vaccins et vaccinations Paris, Elsevier,1985. p. 97-103.

77. WIKTOR, T.J.; MACFARLAN, R.I.; REAGAN, K.J.; DIETZSCHOLD, B.; CURTIS, P.J.; WUN¬

NER, W.H.; KIENY, M.P.; LATHE, R.; LECOCQ, J.P.; MACKETT, M.; MOSS, B.; KO-

PROWSKI, H. Protection from rabies by a vaccinia virus recombinant containing the rabies

virus glycoprotein gene. Proc. Natl. Acad. Sei. USA, 67:7194-7198, 1984.

78. WUNNER, W.H.: LARSON, J.K.; DIETZSCHOLD, B.; SMITH, C.L. The molecular biology of

rabies viruses. Rev. Infect. Dis., 70:S771-S784, 1988.

79. YELVERTON, E.; NORTON, S.; OBIJESKI, J.F.; GEODDEL, D.V. Rabies virus glycoprotein

analogs: biosynthesis in £. coli. Science, 279:614-620, 1983.

2 3

5 6

11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 53-58, 1991

INSERTION OF HETEROLOGOUS EPITOPES IN

SALMONELLA FLAGELLIN

Newton, S.M.C.’ & Stocker, B.A.D. 2 ( 1) Department of Microbiology,

Instituto de Ciências Biomédicas, University of São Paulo, São

Paulo, Brazil; (2) Department of Microbiology and Immunology,

Stanford University, Califórnia, USA.

INTRODUCTION

Strains of Salmonella with non-reverting mutations in the common aromatic biosyn-

thesis pathway have been constructed 4 and found to be attenuated and highly effective

as live vaccine in several animal models 12 - 14 . The lack of virulence is probably asso-

ciated with their requirement for paraminobenzoic acid, not available in mammalian tis-

sues. Therefore, aromatic-dependent Salmonella multiply for only a few generations in

the host; however, they persist, as live bactéria, in the liver and spleen of mice for

weeks. Such persislence accounts for a strong immune response by the host, provid-

ing long lasting protection against challenge with homologous strains.

Aromatic-dependent Salmonella strains have been successfully used to expose het-

erologous antigens to the immune system. An aromatic-dependent S.dublin strain car-

rying a plasmid harboring the gene for the B subunit of heat-labile E. coli enterotoxin

was able to evoke serum IgG and mucosal IgA antibodies to LT-B after oral administra-

tion to Balb/c mice 2 . Similarly, an aroA strain of S. typhimurium carrying a plasmid

specifying constitutive expression of beta-galactosidase induced cellular and humoral

responses to beta-galactosidase, as measured by footpad swelling test and ELISA 1 .

The immune response to a cloned antigen may be enough to cause protection. In-

deed, an aroA strain of S.typhimuríum carrying a plasmid determining production of the

M protein of a Streptococcus pyogenes strain was able to protect against intraperitoneal

Correspondence to: Salete M.C. Newton, Department of Microbiology, Instituto de Ciências Bio¬

médicas, University of São Paulo, 05508 São Paulo, SP - Brazil.

SciELO

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p, 53-58, 1991

challenge with 100 LD 50 S. pyogenes after oral immunization of Balb/c mice 11 . Vaccina-

tion of Balb/c mice with aroA Salmonella typhimurium expressing the circumsporozoite

protein from Plasmodium berghei conferred protection against rodent malaria 13 .

Since the immune reponse directed to some antigens resides in short amino acid se-

quences (linear epitopes), it is possible to engineer recombinant bacterial proteins carry-

ing foreign antigenic determinants. When expressed by live vaccine strains of Salmonel¬

la, such hybrid proteins are likely to evoke immune responses to the inserted peptide.

Flagellin, the protein that makes up the bacterial flagellar filament has been consid-

ered an attractive target for epitope insertions, since there is great variation in its amf-

noacid composition, a fact revealed by the number of flagellar serotypes in nature. The

cloning and sequencing of four flagellin genes of different serotypes from Salmonella

showed near identity between the genes for ca. 150bp at each end of the genes, and

increasing diversity toward the central region, where no more than about 30% homolo-

gy in aminoacid sequence was detected for any pair-wise comparison 16 . A previously

identified epitope of flagellar antigen i was located with segment IV 6 . Mutants of a

cloned flagellin Hl -d gene with altered antigenicity showed a point mutation or dele-

tions within segment IV 9 . Therefore, the central, hypervariable region is thought to

specify antigenic determinants present at the flagellar filamenfs surface. Insertion of

epitope-specifying oligonucleotides in that region could result in exposition of a foreign

epitope as a flagellar antigenic determinant.

MATERIALS AND METHODS

Plasmid pLS402 is a pBR322 with a 3.8kb insert from S.muenchen harboring the

Hl -d flagellin gene (provided to us by Dr. T.M. Joys). Plasmid pLS408 was derived

from pLS402 by cloning the EcoRI insert in pUC19 and deleting a 48bp EcoRV frag-

ment, inside the hypervariable region, generating a single EcoRV site (GAT ATC) in be¬

tween two codong of the Hl -d gene. Oligonucleotides were designed to allow in-frame

insertion at the EcoRV site of plasmid pLS408. They were purchased from Operon Bio-

technologies Inc., Califórnia.

Strain CL447 is a C600-derivative with a deletion at its single flagellin gene, hag 18 . It

is non-motile but becomes motile when given plasmids pLS402 or pLS408. Strain

LB5000 is an S. typhimurium strain r-m+ used to modify plasmids from E. coli to prevent

restriction in the live vaccine strain SL5928. SL5928 is an S. dublin aroA strain with its

single flagellin gene replaced by a gene inactived by the insertion of transposon TnlO.

All molecular manipulations were carried out according to Maniatis et al 7 . Protein

analysis was made by western-blotting 15 and the immune responses followed by

ELISA 3 .

RESULTS

Epitope CTP3 of cholera toxin subunit b as insert

The CTP3 epitope comprises residues 50-64 of the B subunit of cholera toxin 5 . Oli¬

gonucleotides specifying this sequence were synthesized with a 15bp overlapping re¬

gion, leaving 5' overhangs to be repaired by Klenow fragment. The resulting double-

2 3

5 6

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 53-58, 1991

stranded segment was blunt-end ligated to EcoRV-digested pL3408 and transformed

into CL447 strain. Plasmids wilh the insert in correct orientation (as shown by sequenc-

ing) were transíerred to LB5000 strain and from this strain to SL5928. Hybrid flagellin

genes conferred motility on the flagellin-negative recipients (E. coli CL447 and S. dublin

SL5928). Western blotting of lysed bactéria or isolated flagellin showed a single band of

the expected size, binding both rabbit anti-d antigen and monoclonal anti-CTP3 anti-

body. Exposure of CTP3 at the surface of the flagella was shown by immobilization of

the bactéria by monoclonal CTP3 antibody and also by immunogold labelling. Live-

vaccine strain of S.dublin expressing the chimeric flagellin gene was given to C57BL/6

mice, at three weekly intervals, i.p. (5 x 106 bact/mouse) and their sera analysed by

ELISA. Sera from all animais reacted with CTP3 peptide and whole cholera toxin 8 .

Epitopes of hepatitis B surface antigen as inserts

Epitopes from the surface antigen (SI22-137) and from the pre-S2 region (120-145)

were inserted in flagellin essentialiy by the same procedure employed for CTP3 inser-

tion. Results are summarized in table I. Recombinant plasmids with the oligonucleo-

tides inserted in correct orientation (pLS414 for S122-137 and pLS428 for 120-145)

TABLEI

Insertion of oligonucleotides specifying hepatitis B surface epitope (residues 122-137)

in vector pLS408. Arrows outside boxes indicate direction of transcription of flagellin

gene Hl -d. Boxes represent inframe insertions, with the two possible orientations

indicated by arrows.

originated non-motile E. coli or S. dublin clones. By western-blotting, a band binding

anti-flagellar antigen d and anti-peptide antibodies was detected in both cases, but no

flagella were seen by electron microscopy. Insertion of S122-137-specifying oligonucle-

otide in reverse orientation produced functional flagella, even when 3 copies of the se-

quence were inserted (48 aminoacids). One recombinant plasmid, pLS413, showed

two copies of the oligonucleotides, one of them in correct orientation, and originated

functional flagella. Therefore, it was possible to compare the immune response to the

hepatitis epitope when present at the flagellar filamenfs surface (pLS413) or intracellu-

larly, as flagellin (pLS414). Both constructs, when expressed by the live vaccine strain

SL5928, evoked antibodies to the inserted peptide to the same levei 17 .

SciELO

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 53-58, 1991

M protein epitope as insert

The Streptococcus pyogenes type-5 M protein comprising the 16 amino terminal

residues of the mature protein was inserted in flagellin, since using the whole protein

expressed in Salmonellalo immunize mice p.o. conferred protection against challenge

with the Streptococcus strain. Immunization with the whole protein, while effective to

confer protection, has been hampered by the fact that such protein presents epitopes

that cross-react with human heart tissue. The epitope chosen for insertion in flagellin,

has been recently characterized as protective, and unable to evoke auto-immune re¬

sponses 10 .

Recombinant plasmids showing the oligonucleotides in correct orientation originated

functional flagella, similarly to what has previously been observed for the cholera toxin

epitope insertion. Exposition of the epitope at the flagellar filamenfs surface was re-

vealed by immobilization and immunogold labelling. Mice immunized i.p. with live

SL5928 expressing the chimeric flagellin gene and rabbits immunized i.m. with the

same strain, formalin-fixed, made antibodies to the M protein peptide, with opsonizing

activity (Table II).

Table III summarizes the immune response of all epitopes described herein.

TABLE II

Summary of immune responses of mice vaccinated with strain SL5928

expressing flagellin with an M5 insert or given the same strain expressing flagellin

with no heterologous epitopes.

Week

Elisa Titres (*)

Peptide S. dublin

Opsonization

(**)

Mice immunized with strain SL5928 expressing plasmid pL435 (M5 insert in flagellin)

<100

200

3,200

800

12,800

6 (***)

6,400

25,600

Mice immunized with strain SL5928 expressing plasmid pL408 (no insert in flagellin)

<100

6(“*)

<100

12,800

(*) Titres of pooled sera from 5 mice in ELISA test with synthetic M5 peptide or whole

bactéria (SL5928) as test antigens.

(**) Percent of human neutrophila with one or more associated streptococci.

(***) Two weeks after the vaccine dose, the mice were challenged by i.p. injection of

100 L.D.50 streptococci. None of the mice given the live-vaccine strain expressing

pLS408 survived. Only one of the 5 mice given the hybrid live-vaccine strain died.

SciELO

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 53-58, 1991

TABLE III

Summary of immune responses for all epitopes described in this paper.

Epitope

(residues)

Plasmid

Motility

species

doses

no. of

doses

route

ELISA

(+/tested)

Cholera toxin

pLS411

C57BLV6

5x10 6 , live

i.p.

5/5

subunit B

C57BL/6

5x10 6 , killed

i.p.

5/5

(50-64)

C57BL/6

10 9 , live

p.o.

0/5

Hepatitis B surface

pLS414

rabbit

10 9 , live

i.m.

2/2

protein

Balb/cJ

5x10 e , live

p.o.

10/10

S (122-137)

guinea pigs

10 9 , live

p.o.

3/3

Pre-S2 (120-145)

pLS428

rabbit

10 9 , live

i.m.

2/2

B10/BR

5x10 8 , live

p.o.

10/10

guinea pigs

p.o.

3/3

S (122-137)

pLS413

rabbit

10 9 , live

i.m.

2/2

Balb/oJ

5x10 8 , live

p.o.

10/10

guinea pigs

10 9 , live

p.o.

3/3

S. pyogenes

pLS435

rabbit

10 a , killed

i.m.

type-5 M protein

Balb/cJ

5x10 6 , live

i.m.

5/5

(42-57)

Balb/cJ

10 9 , live

p.o.

0/5

CONCLUSIONS

The central region of flagellin gene Hl -d seems appropriate for insertion of epitope-

specifying oligonucleotides, since humoral immune responses have been detected

against all epitopes inserted so far. The generation of immune responses to the foreign

peptide and to flagellin is not dependent upon flagellar function, a fact that greatly ex-

tends the probable usefulness of the system. Our results with the hepatitis B surface

antigen indicate that it might be possible to insert several epitopes without preventing

flagellar assembly and function. However, at the moment we do not know the features

of an aminoacid sequence that allow normal flagellar function. Immunization with live-

vaccine strains of Salmonella presents several advantages: effective immunization by

oral route, induction of cellular and humoral immune responses as well as mucosal im-

munity, and safety, due to the non-reverting aromatic mutation. Experiments are under

way to determine the maximum size of insertion compatible with flagellar function, the

possibilities of using other sites in the flagellin molecule for insertion of epitopes, and

the induction of cellular immunity to a heterologous epitope inserted in flagellin.

ACKNOWLEDGMENTS

The work at Stanford University has been supported by U.S. Public Health Service

grants AI-18872 and AI-27722 from the National Institute of Allergy and Infectious Dis-

ease, grant 000553 from the American Foundation for Aids Research, and a gift from

Praxis Biologics, Inc. S.N. held a post-doctoral fellowship from FAPESP (86/0594-9).

The authors thank Marianne Hovi for excellent technical Services.

SciELO

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 53-58, 1991

REFERENCES

1. BROWN, A.; HORMAECHE, C.E.; HORMAECHE, R.; WINTHER, M.; DOUGAN, G.; MAS-

KELL, D.J.; STOCKER, B.A.D. An attenuated aroA Salmonella typhimurium vaccine elicits

humoral and cellular immunily to cloned beta-galactosidase in mice. J.Inlect.Dis., 155: 86-

92, 1986.

2. CLEMENTS, J.D.; LYON, F.L.; LOWE, K.L.; FARRAND, A.L.; EL-MORSHIDY, S. Oral immu-

nization of mice with attenuated Salmonella enterítidis containing a recombinant plasmid

which codes for production of B-subunit of heat-lebile Escherichla coli enterotoxin. Infec.

Immun., 53: 685-688, 1986.

3. ENGVALL, E.; PERLMANN, P. Enzyme-linked immunosorbent assay (ELISA) quantitative

assay of immunoglobulin G. Immunochemistry, 8: 871-879, 1972.

4. HOISETH, S.K.; STOCKER, B.A.D. Aromatic-dependent Salmonella typhimurium are non-

virulent and effective as live-vaccine. Nature, 291: 238-239, 1981.

5. JACOB, C.O.; SELA, M.; ARNON, R. Antibodies against synthetic peptides of the B subunit

of cholera toxin: cross-reaction and neutralization of the toxin. Proc. Natl. Acad. Sei. USA,

80: 7611-7615, 1983.

6. JOYS, T.M.; MARTIN, J.F. Identification of amino acid changes in serological mutants of the i

flagellar antigen of Salmonella typhimurium. Microbios, 7: 71-73, 1983.

7. MANIATIS, T.; FRITSCH, E.F.; SAMBROOK, J. Molecular cloning. a Laboratory Manual.

New York, Cold Spring Harbor Laboratory. Cold Spring Harbor, 1982.

8. NEWTON, S.M.C.; JACOB, C.O.; STOCKER, B.A.D. Immune response to cholera toxin epi-

tope inserted in Salmonella llagellin. Science, 244: 70-72, 1989.

9. NEWTON, S.M.C.; WASLEY, R.D.; WILSON, A.; ROSENBERG, L.T.; MILLER, J.F.;

STOCKER, B.A.D. Segment IV of a Salmonella flagellin gene specifies flagellar antigen

epitopes. Molec. Microbiol., in press.

10. POIRIER, T.P.; KEHOE, M.A.; DALE, J.B.; TIMMIS, K.N.; BEACHEY, E.H. Expression of pro-

tective and cardiac tissue cross-reactive epitopes of type 5 streptococcal M protein in Es-

cherichia coli. Inlect. Immun., 48: 198-200, 1985.

11. POIRIER, T.P.; KEHOE, M.A.; BEACHEY, E.H. Protective immunity evoked by oral adminis-

tration of attenuated aroA Salmonella typhimurium expressing cloned streptococcal M pro¬

tein. J. Exp. Mod, 168: 25-32, 1988.

12. ROBERTSSON, J.A.; LINDBERG, A.A.; HOISETH, S.K.; STOCKER, B.A.D. Salmonella ty¬

phimurium infoction in calves: protection and survival of virulent challenge bactéria after im-

munization with live or inactivated vaccines. Infec. Immun., 41: 742-750, 1983.

13. SADOFF, J.C.; BALLOU, W.R.; BARON, L.S.; MAJARIAN, W.R.; BREY, R.N.; HOCKMEYER,

W.T.; YOUNG, J.F.; CRYZ, S.J.; OU, J.; LOWELL, G.H.; CHULAY, J.D. Oral Salmonella

typhimurium vaccine expressing circumsporozoite protein protects against malaria. Sci¬

ence, 240: 336-338, 1988.

14. SMITH, B.T.; REINA-GUERRA, M.; HOISETH, S.K.; STOCKER, B.A.D.; HABASHA, F.;

JOHNSON, E,; MERITT, F. Safoty and efficiency of aromatic-dependent Salmonella typhi¬

murium as live-vaccine for calves. Amer. J. Vet. Res., 45: 59-66, 1984.

15. TOWBIN, H.; STAEHELIN, T.; GORDON, J. Electrophoretic transfer of proteins from polya-

crylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad.

Sei. USA, 76: 4350-4354, 1979.

16. WEI, L.N.; JOYS, T.M. Covalent strueture of three phase-1 flagellar filament proteins of Sal¬

monella. J. Mol. Biol., 786:791-803, 1985.

17. WU, J.Y.; NEWTON, S.M.C.; JUDD, A.; STOCKER, B.A.D.; ROBINSON, W.S. Expression of

immunogenic epitope of hopatitis B surface antigen with hybrid flagellin protein by a vac¬

cine strain of Salmonella. Proc. Natl. Acad. Sei. USA. 86: 4726-4730, 1989.

18. ZIEG, J.; SIMON, M. Analysis of the nucleotide sequence of an invertible controlling element.

Proc. Natl. Acad. Sei. USA, 77: 4196-4200, 1980.

2 3

5 6

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 59-70, 1991

THE IMPACT OF RECOMBINANT DNA ON THE CONTROL

OF ANIMAL HEALTH

Ingrid E. Bergmann, Pan-American Foot-and-Mouth Disease

Center, Rio de Janeiro - RJ, Brasil

INTRODUCTION

Although there are at present a large number of different products for controlling dis-

eases and improving productivity of livestock approximately one hundred billion dollars

is lost annually to the world economy due to death losses, treatment costs or reduced

productivity. The advent of recombinant DNA techniques has the potential to develop a

variety of new diagnostic procedures as well as new vaccination strategies to help in

the campaigns against economically important diseases of livestock.

Foot-and-mouth disease virus (FMDV) will be adopted to exemplify the impact of re¬

combinant DNA of the new developments.

The causative agent of foot-and-mouth disease (FMD) is an aphthovirus belonging

to the family picornavirídae. The virion is icosahedral, without envelope, of about 25 nm

of diameter and consists of 60 copies each of 4 coat protein VP^ VP 2 , VP 3 and VP 4 .

VP, is the only structural polypeptide that, when purified and injected into cattle, is ca-

pable of inducing neutralizing antibodies 1 ' 2 ' 3 . The virai genome consists of a single-

stranded RNA of approximately 8000 nucleotides. The virai RNA is infectious and

serves as mRNA. Upon translation a polyprotein is produced which is subsequently

cleaved into a series of intermediate precursors which are further processed to give the

mature non-capsid and capsid proteins.

This system provides an excellent model to explore many of the alternative possibili-

ties of diagnosis and vaccines because: - The virus has a relatively simple structure

compared to other agents; - Foot-and-mouth disease is obviously one of the major dis-

Correspondence to: Ingrid E. Bergmann, Pan American Foot-and-Mouth Disease Center, P.O.

Box 589, 20.001 Rio de Janeiro - RJ, Brazil.

SciELO

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 59-70, 1991

eases of domestic farm animais; - lt is an acute disease characterized by vesicular

lesions which lead to significant productivity losses; - After the acute phase of the

disease, the virus established inaparent persistent infections which constitute an eco-

nomic burden to cattle exporters because of the restrictions placed on the export of

FMDV-positive animais or animal products to dísease-free areas; - Therefore the dis¬

ease causes one of the major economic impacts in livestock industry. It is estimated

that in South American countries, at least 25% of the reduction in livestock productivity

is due to this disease, causing losses and expenses of 500 million dollars annually; -

Variability of the virus during replication leads to variants often not neutralized by the

vaccines in use. Variability is criticai during persistent infections 4 ' 5 ; - The potential of

new biotechnological strategies should be evaluated in order to develop:

a) highly sensitive and specific methods for detection of carrier animais;

b) precise techniques for the characterization of FMDV to establish the similarities

between field and vaccine strains;

c) more effective vaccines: Conventional vaccines have been available since 1938

and in addition, 800 million doses are used annually, so that efficacy of new alter-

natives will have a good point of reference.

DIAGNOSTIC PROBLEM IN SUSPECTED APHTHOVIRUS DISEASE

The Identification of carrier animais infected with FMDV is usually made by the isola-

tion of FMDV of material obtained from esophageal-pharyngeal (OP) fluids, in tissue

culture 6 ' 7 ' 8 ' 9 . Moreover, screening for antibodies against the FMDV-infection associat-

ed antigen (VIAA) 1CM3 , the FMDV RNA polymerase 14 ' 15 , in animal sera is also carried

out routinely. These tests are internationally accepted for import/export testing and as

an epidemiological tool to determine the spread of FMD in animal populations. In prac-

tice, these methods do not always yield conclusive results. No clear correlations could

be established between these methods suggesting technical problems with the OP fluid

sampling and/or the tests used to detect the anti-VIAA antibodies. Virus isolation proce-

dures appear to be successful in the acute phase of the disease with extensive virus

replication, however persistent virus from OP samples is only occasionally recovered

during the whole carrier State.

The identification of serum antibodies against the VIA antigen by the immunodiffu-

sion test in agarose gels (IDAG) 10 ' 12 is not sensitive enough. Attempts to increase the

sensitivity through an enzyme-linked immunosorbent assay (ELISA) 16 raised questions

with regard to the development of VIAA antibodies in cattle vaccinated and revaccinat-

ed with vaccines produced in baby hamster kidney (BHK) cells, containing high con-

centrations of non-purified FMDV-antígens. Since the VIA antigen used for these tests

is only partially purified from BHK-infected cells, traces either of FMDV capsid polypep-

tides or BHK antigens may be recognized by sera of animais immunized with BHK-

produced vaccines, leading to false positive tests.

Another significant limitation of these methods is the requirement of a high security

laboratory unit for handling FMDV, which constitutes a problem, especially in FMD-free

areas.

INTRODUCTION OF NEW DETECTION STRATEGIES

So far we have chosen two approaches to potentially overcome the mentioned iimita-

2 3

5 6

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 59-70, 1991

tions. The first approach was based on the use of new diagnostic approaches for the de-

tection of FMDV in OP samples which included: a) molecular nucleic acid hybridization

techniques (dot blot, northern blot or in situ hybridization), using cloned virai DNA as a

diagnostic reagent for detection of the virai RNA sequences and b) amplification of spe-

cific virai genomic fragments using the polymerase chain reaction (PCR). The second

approach was to develop highly sensitive and specific methods to detect in sera of sus-

pected carrier animais the presence of antibodies against FMDV-nonstructural antigens,

including others than the VIAA traditionaliy used. Immunochemical techniques (ELISA,

immunodot and immunoblotting) using as serological probes highly purified bioengi-

neered VIAA as well as other bioengineered nonstructural antigens were attempted.

One major prerequisite for the introduction of these strategies as diagnostic tools in

suspected persistent aphthovirus infections was the molecular cloning of the aphthovi-

rus RNA genome in order to get a complementary DNA (cDNA) as a probe for the

FMDV genome, to be used in molecular hybridization tests. Besides, the cloned cDNA

is further used to obtain the bioengineered serological probes, by expressing defined

genomic regions, coding for the different nonstructural proteins, in E. coli.

Intact 35S virai RNA was prepared, copied into cDNA using oligo (dT) as a primer

and reverse transcriptase. The cDNA was then fractionated on an alkaline sucrose gra-

dient. The largest cDNA fractions were further inserted through standard techniques

into a bacterial plasmid.

Clones covering the whole coding region of the genome were obtained.

CLONED APHTHOVIRUS cDNA AS A DIAGNOSTIC REAGENT

lf one addresses nucleic acid hybridization as a diagnostic tool for the detection of

aphthovirus carriers a major prerequisite is the demonstration that the radioactively la-

beled cloned cDNA detects specifically virai nucleic acids and does not hybridize to to¬

tal cell RNA. When radioactively labeled cloned aphthovirus cDNA corresponding to

the genomic region coding for the virai polymerase was hybridized to RNA immobilized

on nitrocellulose paper, extracted from OP samples of control or experimentally persis-

tently-infected cattle, specific hybridization was found only for FMDV-infected tissue

RNA and not for total tissue RNA. The minimal detection levei under high stringency

conditions, that we observed was 1 pg of genome FMDV-RNA. Thus, the use of cloned

cDNA provides a valuable diagnostic tool for detection of carrier cattle.

A main advantage of the nucleic acid hybridization approach in suspected FMDV in¬

fections is given by the fact, that detection of different aphthovirus serotypes is possible

by using just one cloned cDNA as a probe. The molecular basis for this finding is given

by the high degree of nucleic acid sequence homology among the different serotypes

for example in the genomic region corresponding to the virai polymerase. The exact

typing of the implicated strain could also be carried out by using serotype-specific DNA

fragments corresponding to the virai genome encoding for the virai coat proteins.

Hybridization directly in fixed tissues is another interesting approach to be consid-

ered. It is a worthwhile tool especially for studies of virai pathogenicity, since it permits

the direct identification of the virus-infected target cells and an estimate of the propor-

tion of cells carrying the virai genome within a section.

Other probes for veterinary applications include: Pseudorabies, Bovine herpes virus

I, African swine fever, Bluetongue, Marek's disease, Anaplasma marginale, Babesia

2 3

5 6

11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 59-70, 1991

and Enterotoxigenic E. coli.

AMPLIFICATION OF SPECIFIC APHTHOVIRUS GENOMIC RNA SEGMENTS

A major breakthrough for diagnostic procedures is given by the polymerase chain

reaction (PCR) 17 . It consists on the capacity to amplify specific segments of DNA

through an in vitro enzymatic synthesis of a specific DNA fragment using two oligonu-

cleotide primeis that hybridize to opposite strands and flank the region of interest in the

target DNA. This synthesis is achieved through a repetitive series of cycles involving

denaturation, primer annealing and extension of annealed primers.

Since primer extension products can serve as templates in the next cycle, there is

an exponential accumulation of specific fragments whose termini are defined by the 5'

ends of the primers, so that 20 cycles can yield about one million fold amplification.

For the detection of virai RNA it is necessary to first obtain a cDNA copy by the re-

verse transcription, which is then used as target DNA.

We have applied the PCR method to the amplification of defined segments from the

genomic region coding for the structural polypeptides of partially purified FMDV-RNA of

serotype O r Primers were chosen from highly conserved regions and purified. Suc-

cessful amplification of fragments of 850 bp covering the whole VP, region and of frag¬

ments of up to 2000 bp covering the genomic region coding for the capsid polypeptides

VP 3 , VP 2 and VP v were obtained. In contrast, amplification of segments of more than

about 2000 bp normally resulted in several low molecular weight bands and only rarely

gave a specific reaction.

The specificity of the amplified material was confirmed by restriction enzyme analy-

sis of amplification products of the correct molecular weight, and is being further con¬

firmed by sequencing after subcloning into SP64.

Our future goal is to amplify FMDV-RNA segments from biological specimens such

as cells persistently-infected with FMDV and biopsies from experimentally-infected ani¬

mais. The method should be adequate for rapid diagnosis of FMD, even when minimal

quantities of virus are present.

A great advantage of this extremely rapid and sensitive method for the detection of

FMD is that nucleotide sequence information can be obtained directly from clinicai sam-

ples without the prior need for virus amplification in cell culture, thus avoiding the possi-

bility of in vitro induced selection and modification of virus populations.

DETECTION OF FMDV-ANTIBODIES IN SERA OF

PERSISTENTLY-INFECTED ANIMALS

In order to overcome the limitations mentioned above, we have attempted the use of

immunochemical techniques using bioengineered VIAA and other bioengineered aph-

thovirus nonstructural antigens.

Compared to the classical VIAA obtained from infected cell cultures, the use of bio¬

engineered antigens offers several advantages: safety, does not require a high security

laboratory for its production; yields, higher yields can be obtained in bactéria than in in¬

fected cells; and simplicity and costs, bacterial cultures require less space, time and ef-

fort than eukaryotic cell cultures and are significantly less expensive. In addition, sero-

2 3

5 6

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 59-70, 1991

logical probes can be prepared for nonstructural antigens which can not be purified

from infected cells, so that the most suited nonstructural antigens of potential diagnos-

tic significance can now be identified by studying the antibody response of the host to

the different aphthovirus nonstructural proteins during different stages of persistence.

To obtain the most adequate bioengineered VIAA for use as a serological probe, we

constructed a bacterial expression vector (carrying the inducible PL promoter of bacte-

riophage Lambda in front of a consensus Shine Dalgarno sequence) coding for the

complete VIA antigen. The strategy was desígned so that the polypeptide expressed

would only have one additional amino acid the N-terminal methionine 18 .

The recombinant polypeptide was expressed and the soluble fraction from the bac¬

terial extracts was further purified by chromatography over a series of phosphocellu-

lose column and poly(U)-Sepharose column. The identity of the purified protein was

confirmed by ímmunoblotting with sera of convalescent animais. We further analyzed

the potential of the bioengineered native VIAA to detect specifically anti-VIAA antibod-

ies in sera of experimentally persistently-infected cattle.

Analysis by Ímmunoblotting indicated that antibody binding to the bioengineered na¬

tive VIA polypeptide was constant and it was detected in sera from infected cattle as

early as at 7 dpi, giving a peak at 6 weeks postinoculation with the intensity of the bands

decreasing gradually with time after infection, and being still positive at DPI 644, time by

which all detections were negative by the classical IDAG test. In contrast to the results

obtained previously 16 , under the conditions used, no detectable reaction was obtained

with sera from control cattle (obtained from FMD-free regions) and/or vaccinated cattle.

The induction of antibodies to other nonstructural proteins was analyzed by using a

set of bioengineered antigens obtained by expressing defined regions of the genome in

E. coli, as serological probes 19 . Western blot analysis with sera of experimentally in¬

fected cattle shows that although all antigens gave a positive reaction, antigens 3A and

3B gave the highest signal/noise positive detection when tested either with sera from

convalescent or late persistently-infected cattle. Antibody induction to antigens 3A and

3B shows a peak 5 weeks later than that obtained for the VIA antigen, decreasing then

gradually with the increasing weeks postinoculation, but to a lesser extent than for the

VIA antigen. Again, none of the sera of FMD-free regions or from vaccinated cattle

gave a positive reaction.

Our ultimate goal is to develop a rapid and simple assay for the detection of carrier an¬

imais. The use of additional nonstructural antigens together with Ímmunoblotting, could

provide a method for simultaneously screening a single serum for the presence of anti¬

bodies against multiple antigens which may fluctuate during the course of the disease.

Examples of other animais diseases which have immunochemical diagnostic assays

include: Bluetongue, Salmonella, Pseudorabies, Parvovirus, etc.

MOLECULAR CHARACTERIZATION OF APHTHOVIRUS STRAINS

Control of FMDV is complicated by the occurrence of the virus in 7 serotypes. The Eu-

ropean types O, A and C, also present in South America, the South African Territorie type

Sat v Sat 2 and Sat 3 and the Asiatic type Asia,. In addition, over 60 known subtypes result-

ed from variation within each serotype, with little cross reactivity among them. Therefore,

the constant monitoring of field strains is essential to ensure that vaccines in current use

are effective, i.e., contain strains sufficiently similar to thosecirculating in the field.

2 3

5 6

11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 59-70, 1991

Several techniques for the analysis of nucleic acids and proteins are being used.

Such techniques include nucleic acid analysis through RNA fingerprinting, and se-

quencing as well as protein analysis through SDS-PAGE, isoelectrofocusing, two-

dimensional gel electrophoresis, etc. These biochemical methods together with the de-

velopment of monoclonal antibodies and their use in ELISA tests provide ideal tools for

the precise biochemical and antigenic characterization of active, evolving, vaccine and

laboratory strains.

We are intensively involved in the biochemical and serological characterization of

field and vaccine strains.

Fingerprinting is used routinely for comparing closely related strains, and so: evalu-

ate genetic stability of strains during vaccine production, establish possible vaccine ori-

gin of field outbreak and monitor origin behaviour and fate of new strains. Examples of

outbreaks caused by viruses of serotype C 3 which showed minor serological variations

from the prototype strain C 3 Resende as well as very similar fingerprinting patterns, oc-

curred in Argentina between 1982-1984 20 . The similarity of the isolated strains with that

of the prototype strain, was taken as an indication that these strains had not circulated

for long in the field and that they had been freshly introduced through escape of a la¬

boratory or an incomplete inactivated vaccine.

The significance of the evolution of viruses in the field can be exemplified by the char¬

acterization of strains isolated from an outbreak caused by strains serologically identified

as a C 3 variant which took place in Argentina during 1984 and up to 1986. This outbreak

could not be controlled by the vaccines in use containing the prototype strain C 3 Resende

20 . Representatives of this outbreak were isolated, studied and later included in the vac¬

cines in order to finally control the outbreak. Fingerprinting showed them to be significant-

ly different from the prototype strain C 3 Resende. Moreover, relevant changes in the

structural polypeptides could also be shown. Although the epidemiological origin of these

strains has not been traced, it is possible that in endemic regions viruses could be main-

tained under conditions which they are replicating such as occurs in persistently-infected

cattle. We have demonstrated that during persistence, variability is very criticai. During

this time, rapid evolution of viruses occurs. We found a clear cut relationship between the

degree of genomic variations and the number of days postinoculation, indicating a gradu¬

al and progressivo evolution of the strains during the persistent stage. In addition, we de-

scribed a decreased reactivity of FMDV persisting at 63 dpi to a set of neutralizing mono¬

clonal antibodies. Sequencing data also showed the accumulation of mutations which

represented 2x10 2 substitutions/nucleotide/year, 60% of the changes resulted in chang¬

es of amino acids. Some of the changes occurred in amino acid residues 40-50 and 135-

153, considered to be important immunological domains 5 . This fact demonstrates the

high risk for the animais and for those susceptible hosts in the surroundings.

PROSPECTIVES FOR THE DEVELOPMENT OF NEW VACCINES

Although vaccination to prevent virus infections has been practiced for over two cen-

turies, the process has changed relatively little since the time of Jenner. Most of the im-

munogens of choice used today are still killed or attenuated viruses. Recent attention

has focused upon the design and production of vaccines consisting of non-viable (non-

replicating) and non-infectious portions of the pathogenic agent that are still capable of

eliciting a protective immune response.

2 3

5 6

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 59-70, 1991

Potential vaccine designs include:

I. Subunit vaccines

a) Expression

Bactéria

Eukaryotic systems:

Yeast

Poxvirus

Baculovirus

b) Synthetic peptides

II. Attenuation by direct gene manipulation

III. Anti-idiotypes

IV. Antisense: Infectious resistant cells

V. Complementation of a virus deficient strain in cells constitutively producing the

defficient protein

VI. Anti-cell receptor: antiviral

SUBUNIT VACCINES

This approach became possible when some structural features necessary for elicit-

ing a good immune response were identified:

- the neutralizing activity is largely confined to VP,; VP 1 isolated and used as a

vaccine elicits neutralizing antibody responses and protects cattle and swine

from infections;

- the neutralizing activity was generated by fragments obtained by cyanogen bro-

mide or enzymatically, spanning the regions corresponding to amino acid resi-

dues 145-154 and 201-213.

The aphthovirus structural gene coding for VP, was expressed in E. coti. Although a

significant levei of expression was obtained, the protein produced, evoked no neutraliz¬

ing antibody response up to 30 days postrevaccination.

These results are not surprising considering that the isolated protein is weakly im-

munogenic possessing less than 0,1% of the activity of the virus particle 21 . It becomes

always more evident that VP, in the virion adopts a conformation which is highly depen-

dent from a substantial interaction with the other structural proteins 22

In the case of other infectious agents, expressed in prokaryotic systems, the first

outstanding example of a genetically engineered bacterial vaccine which became com-

mercially available was against the somatic pili of enterotoxigenic E. coli strains, the

cause of diarrheal diseases in young livestock 23 .

A recently developed baculovirus expression system utilizes the highly expressed and

regulated Autographa californica nuclear polyhedrosis virus poiyhedrin promoter modi-

fied for the insertion of foreign genes 24 . Encouraging results were obtained expressing

bluetongue virus neutralization antigens VP 2 and VP 3 25 , pseudorabies gp 50 2S , Rift Val-

ley fever virus antigens G-1 and G-2 27 and simian rotavirus SA11 capsid antigen VP 6 28 .

SYNTHETIC PEPTIDES

It is fortunate that FMDV has been one of the most studied with regard to immuniza-

2 3

5 6

11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 59-70, 1991

tion with peptide anligen and the results suggest that there could be a practical out-

come to the work.

Several approaches predicted amino acid residues 144-159 and to a lesser degree

200-213 of VP, as good candidates for eliciting a neutralizing response 29 ' 30 .

According to these predictions we synthesized a series of peptides, covering various

regions of the polypeptides, corresponding to the sequence of serotype O, Kaufbeuren,

by using the solid phase Merriefield process. The peptides were linked to keyhole lim-

pet haemocyanin and tested for immunogenicity in guinea pigs. Synthetic peptides rep-

resenting each of the two potential immunogenic regions of VP, induced high leveis of

antibodies which recognized intact virus, but only residues 140-160 were protective.

Despite the optimal results obtained in guinea pigs with the peptide corresponding to

residue 140-160, when this peptide was inoculated in cattle, very low leveis of neutral¬

izing antibodies were obtained even after revaccination 31 . The potential importance of

sequences 200-213 in enhancing the response of sequences 141-160 was recently

suggested 32 . Moreover, a very encouraging result was obtained using tandem peptide

sequences fused to bacterial proteins. Already protection of cattle and swine has been

achieved with one injection of as little as 40,ug of peptide. However, only limited ani¬

mais were used in the trial 33 .

A key issue in the field of peptide vaccines is the potency of the adjuvant. Several

new adjuvants have been tried, including the use of what has been termed immunos-

timulating complex (ISCOM), in which the virus proteins are incorporated into cagelike

structures by complexing them with saponin, a plant glycoside.

In the case of other animal infectious agents, encouraging results were obtained

with vaccines against rabies and rotavirus 34_35 .

LIVE ATTENUATED VACCINES

Live virai vaccines offer significant advantages over other types of virai vaccines,

namely, induction of more effective local immunity, longer duration of immunity and

spread of the vaccine strain among susceptible hosts within the population. Such vac¬

cines, however, are at present only of limited use due to a) the possibility that an attenuat-

ed virus strain could revert to its more pathogenicform; b) potential susceptibility of hosts

other than the one for which the vaccine is attenuated and c) an optimum equilibrium be-

tween pathogenicity and immunogenicity is not always obtained. Such limitations could

not be overcome in the late fifties when these attenuated vaccines were developed 36 .

The observation for several picornaviruses that viable virus can be rescued from

cloned cDNA was an essential breakthrough to study the genetic determinants of attenu-

ation and to construct safe attenuated vaccines. Infectious DNA can be specifically modi-

fied to obtain attenuated strains which can replicate and retain antigenic identity without

propensity for virulence. Moreover, once a stable attenuated strain is identified or gener-

ated through adequate alterations of an infectious clone, and provided that the genetic

determinants of attenuation are not located in the immunogenic regions, one can extend

the attenuated phenotype to other serotypes, by introducing through recombination via

cDNA in vitro the genes for immunogenicity from a new strain into the genome of an ideal

avirulent strain. To make such an approach feasible identification of virai genes that

specify virulence is of criticai importance. Therefore, the biological and biochemical char-

acterization of several attenuated strains of different serotypes was undertaken.

2 3

5 6

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 59-70, 1991

A remarkable feature was a common increased electrophoretic mobility of polypep-

tide 3A in the attenuated strains analyzed. Sequencing data indicates a genomic dele-

tion of approximately 60 nucleotides, depending on the attenuated strain 39 . The poten-

tial relevance of this genomic region for determining the attenuated phenotype, is being

further studied by introducing an equivalent deletion in an infectious cDNA clone, re-

cently constructed by the laboratory of Dr. Ewald Beck 40 .

With regard to other infectious agents, a modified virus strain of pseudorabies virus

was obtained by engineering a mutation into the thymidine kinase (TK) gene so that the

activity of TK is destroyed and thus the virus cannot multiply in the central nervous Sys¬

tem of pigs 41 . Similarly the TK gene has been deleted from infectious bovine rhinotra-

cheitis vaccine strains which would allow a similar development of a cattle vaccine 42 .

Further attenuation of pseudorabies was obtained through deletions in the internai and

terminal repeat region of the pseudorabies vaccine strains 43 .

One of the major disadvantages of live attenuated vaccines is that the immune re¬

sponse which is elicited cannot be easily distinguished from the one provoked by natu¬

ral infection so that many times disease control measures are complicated. One ap-

proach to overcome the diagnostic problem is through the introduction of specific gene

deletions. Such an approach was effectively used for pseudorabies. The gene coding

for the glycoprotein was removed from the live vaccine, which prevents antibodies be¬

ing evoked to this glycoprotein and so allows vaccinated pigs to be identified from pigs

naturally infected 44 .

Another way to overcome this limitation is by combining the advantages of subunit

and live attenuated vaccines through the use of vaccinia vectors.

Vaccinia virus behaves as a live-virus vaccine. Therefore it stimulates cell-

mèdiated immunity, but has the advantage that certain selected gene sequences of a

heterologous agent can be inserted and expressed. Moreover, due to its large DNA

capacity, it is possible to construct multivalent vaccines for different serotypes of the

same virus or even against entirely different pathogenic agents. The products ob¬

tained are properly glycosilated and transported to membranes and therefore mimic

the native state 45 .

The large number of animal hosts for vaccinia enables its use for immunization in

veterinary medicine. Other advantages of this system include:

- it is cheap to mass produce;

- it is possible to use poxviruses specific for each animal species;

- no animal reservoir is known;

- it is stable in lyophilized form at room temperature;

- it can be administered by the oral route.

Significant results were obtained with live vaccine of recombinant vaccinia virus

which expresses the glycoprotein of rabies virus. Racoons fed with vaccina/rabies re¬

combinant virus developed rabies neutralizing antibodies and were resistant to rabies

challenge up to 200 days after feeding 46 .

After inoculation of live recombinant vaccinia-G protein of vesicular stomatitis virus

(VSV), mice were protected against lethal encephalitis and cattle protection correlated

with the levei of neutralizing antibody produced following vaccination 47 .

More recently veterinary researches have employed vaccinia virus recombinants as

experimental vaccine against Rinderpest Sindbis, Marek's disease, fowl pox and feline

leukemia.

An interesting strategy was attempted for FMDV, combining the synthetic peptide

2 3

5 6

11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 59-70, 1991

approach with recombinant vaccinia virus. However no neutralizing antibody response

was elicited after inoculation into rabbits 48 .

The potential hazards of human and animal virus vector vaccines should be careful-

ly investigated before they are released for field use. The occurrence of generalized

vaccinia after administration of samllpox vaccine to asymptomatic carriers of human im-

munodeficiency viruses (HIVs) has been reported 49 .

Other interesting alternatives include:

Infectious resistant cells

Consists on blocking the production of virai proteins by inserting antisense genes

(DIMA fragments complementary to the virai genome) into cells. So far, attempts to use

this approach have been reported for AIDS treatment and for FMD.

Anti-idiotypes

Since the observations that anti-idiotypes can mimic foreign antigens, their potential

utilization in vaccines has been pursued 50 . In fact, many groups reported induction of

protective immunity in experimental animais upon administration of anti-idiotypes for

parasites (Schistosoma mansonii, Trypanosoma rhodesiensi), viruses (polio type I and

II, hepatites B) and bactéria (Streptococcus pneumoniae and E. coli). Their main poten¬

tial is against infectious agents which evade neutralizing by antigenic variation and for

carbohydrate antigenic determinants that cannot be genetically engineered.

CONCLUDING REMARKS

At present, several new approaches to the development of animal vaccines were

undertaken. Although their potential is unquestionable, an overall success will depend

on the identification and expression of protective epitopes, and a deeper understanding

of the molecular definitions of virulence, immunological mechanisms and molecular bi-

ology of infectious agents. Further studies are required to explain why vaccines pre-

pared with antigens obtained by genetic engineering or peptide synthesis have been of

relatively poor immunogenicity in cattle and swine when compared with antigens ob¬

tained through classical inactivated vaccines. Very encouraging is the use of new diag-

nostic test, which are of great importance for veterinary medicine diagnostic and partic-

ularly with regard to detection of carrier animais.

ACKNOWLEDGMENTS

I would like to thank Ewald Beck, Paulo Augé de Mello, Ivo Gomes, Erika Neitzert,

Beatriz H. Tiraboschi, Viviana Malirat, Maria Aparecida Affonso Boller, Pedro Jeovah

Vieira Pereira and Ronaldo de Albuquerque for their scientific contributions and Carla

Prete for typing this manuscript.

REFERENCES

1. LAPORTE, J.; GROSCLAUDE, J.; WANTYGHEM, J.; BERNARD, S.; ROUZE, P. Compte

Ftendu Hebdomaire des Seances de UAcademie des Sei. Serie, D276: 3399-3401, 1973.

SciELO

10 11 12 13

Mem. Inst. Butantan, v. 53, supl. 1, p. 59-70, 1991

2. BACHRACH, H.L.; MOORE, D.M.; MCKERCHER, P.D.; POLATNICK, J. Immunol., 115:

1636-1641, 1975.

3. CAVANAGH, D.; SANGAR, D.V.; ROWLANDS, D.J.; BROWN, F. J.Gen.Virol., 35: 149-158,

1977.

4. COSTA GIOMI, M.P.; GOMES, I.; TIRABOSCHI, B,; AUGÉ DE MELLO, P.; BERGMANN,

I.E.; SCODELLER, E.A.; LA TORRE, J.L. Virology, 162: 58-64, 1988.

5. GEBAUER, F.; DE LA TORRE, J.L.; GOMES, I.; MATEU, M.G.; BARAHONA, H.; TIRABOSCHI,

B.; BERGMANN, I.E.; AUGÉ DE MELLO, P.; DOMINGO, E. J. Virol., 62: 2041-2049,1988.

6. BURROWS, R. J. Hyg. ,64:81-90, 1966.

7. SUTMÒLLER, P. & COTTRAL, G.E. Arch. ges. Viruslorsch., 21: 170-177, 1967.

8. LECUW, P.W.; VAN BEKKUM, J.G.; TIESSINK, J.W.A. J. Hyg., 81: 415, 1978.

9. STRAVER, P.J.; BOOL, P.H.; CLAESSENS, A.M.; VAN BEKKUM, J.G. Arch. Ges. Virus¬

lorsch., 29: 113-126, 1970.

10. HEDGER, R.S. J. Hyg., 66: 27-36, 1968.

11. COWAN, K.M. & GRAVES, J.H. Virology, 30: 528-540, 1966.

12. MCVICAR, J.W. & SUTMÒLLER, P. Am J. Epidemiol., 92: 273-278, 1970.

13. ALONSO, A.; AUGÉ DE MELLO, P.; GOMES, I.; ROSENBERG, F. Boi. Centr. Panam. Fiebre

Aftosa, 17/18: 17-22, 1975.

14. POLATNICK, J. & WOOL, S.H. J. Virol., 40: 881-889, 1981.

15. POLATNICK, J. Arch. Virol., 84: 269-275, 1985.

16. ALONSO, A.; GOMES, M.P.D.; MARTINS, M.A.; SÕNDAHL, M.S. Prev. Vet. Med., 9: 233-

240, 1990.

17. MULLIS, K.; FALDONA, F.; SCHARF, S.; SAIKI, R.; HORN, G.; ERLICH, H. Cold Spring Har-

bor Symposio on Quantitative Biology, v. 51, n. 1, 267-273, 1986.

18. NEITZERT, E.; BECK, E.; AUGÉ DE MELLO, P.; GOMES, I.; BERGMANN, I.E. Virology, In

Press, 1991.

19. STREBEL, K.; BECK, E.; STROHMAIER, K.; SCHALLER, H. Journal of Virology, 57: 983-

991, 1986.

20. BERGMANN, I.E.; TIRABOSCHI, B.; MAZUCA, G.; FERNANDEZ, E.; MICHAILOFF, C.A.;

SCODELLER, E.A.; LA TORRE, J.L. Vaccine, 7, 1988.

21. BACHRACH, H.L.; MOORE, D.M.; MCKERCHER, P.D.; POLATNICK, J. In Proceedings, In¬

ternational Symposium of Foot-and-Mouth Disease (II), (C.Mackowiak R.H. Regamey,

Eds). Symposia Series in Deveiopmenls in Biological Standardization, Karger, Basel. 35:

150-160, 1976.

22. MELOEN, R.H.; BRIARE, J.; WOORTMEYER, R.J.; VAN ZAANE, D. J. Gen. Virol., 64: 1193-

1198, 1983.

23. HILDEBRAND, D. Animal Nutrition Health , 20-22, 1983.

24. VERNE, A.L. & SUMMERS, M.D. Biotechnology, 6: 47-55, 1988.

25. INAMURA, S.; GHIASI, H.; ROY, P. J. Gen. Virol., 68: 1627-1635, 1987.

26. KURODA, K.; HAUSER, C.; RUDOLF, R.; KLENK, H.D.; DOERFLER, W. Embo J. , 5: 1359-

1365, 1986.

27. TAKEHARA, K.; MIN, M.K.; BATTLES, J.K.; SUGLYEMA, K.; EMERY, V.C. Virology, 169:

452-457, 1989.

28. ESTES, M.K.; CRAWFORD, S.E., PERRARANDA, M.E.; BURNS, J.; CHAN, W.; ERICKSON,

B.; SMITH, G.E.; SUMMERS, M.D.4. Virol., 61: 1488-1494, 1987.

29. STROHMAIER. K.; FRANZE, R.; ADAM, K.H. J. Gen. Virol., 59: 295-306, 1982.

30 PFAFF, E.; MUSSGAY, M.; BOEHM, H.O.; SCHULZ, G.E.; SCHALLER, H. Embo J., 1: 869-

874, 1982.

31. BERGMANN, I.E.; AUGÉ DE MELLO, P.; SCODELLER, E.; LA TORRE, J.L. Mem. Inst. Bu¬

tantan, 50, (Supl.), 1988.

32. DIMARCH, R.; BROOKE, G.; GALE, C.; CRACKNELL, V.; DOELT, T.; MOWAT, N. Science,

232: 639-641, 1986.

SciELO

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 59-70, 1991

33. BROEKHUDSEN, M.P.; BLOM, T.; VAN RIJN, J.; POUWELS, P.H.; KLASEN, E.A.; FAS-

BENDER, M.J.; ENGER-VALK, B.E. Gene, 49: 189-197, 1986b.

34. ARNON, R. TIBS, 11: 521-524,1986.

35. BARROS-CUESTA, F.; PETIT, A.; PERY, P.; FEDON, J.; GARNIER, J.; COHEN, J. Ann.

Inst. Pasteur/Virol., 138: 437-450, 1987.

36. BERNAL, C.L.; CUNHA, R.G.; HONIGMAN, M.N.; GOMES, I. Proc. 5th Pan-Amer. Cong. Vet.

Med.Zoot., 1: 42-58, 1966.

37. ALMOND, J.W.; STANWAY, G.; CANN, A.J.; WESTROP, G.D.; EVANS, D.M.A.; FERGU-

SON, M.; MINOR, P.D.; SPITZ, M.; SCHILD, C.C. Vaccine, 2: 177-184, 1984.

38. RACANIELLO, V.R. & BALTIMORE, D. Proc. Natl. Acad. Sei., 78: 4887, 1981.

39. GIRAUDO, A.T.; BECK, E.; STREBEL, K.; AUGÉ DE MELLO, P.; LA TORRE, J.L.; SCO-

DELLER, E.A.; BERGMANN, I.E. Virology, 177: 780-783, 1990.

40. ZIBERT, A.; MAASS, G.; STREBEL, K.; FALK, M.M.; BECK, E. Journal of Virology, 64: 2467-

2473, 1990.

41. KIT, S.; KIT, M.; LAWHORN, B. et al. High Technology Route to Virus Vaccines, 82-99, 1985.

42. KIT, S.; SHEPPARD, M.; ICHIMURA, H. etal. Am. J. Vet. Res., 48: 780-793, 1987.

43. KIT, S.; KIT, M.; MCCORNELL, S. Vaccine, 14: 55-61, 1986.

44. KIT, S.; GAVI, H.; GAIRES, J.D. et al. Archives oI Virology, 86: 63-83, 1985.

45. DESMETRE, P. Biofutur, 27-31, 1986.

46. RUPPRECHT, C.E.; WIKTOR, T.J.; JOHNSTON, D.H. et al. Proc. Natl. Acad. Sei., 83: 7947-

7950, 1986.

47. MACKETT, M.; YILMA, T.; ROSE, J.K.; MOSS, B. Science, 227:433-435, 1985.

48. NEWTON, S.E.; FRANCIS, M.; BROWN, F.; APPLEYARD, G.; MACKETT, M. Vaccines, 86:

303-309, 1986. Cold Spring Harbor Laboratories.

49. REDFIELD, R.P.; WRIGHT, D.C.; JAMES, W.D.; JONES, T.S.; BROWN, C.; BURKE, D.S. N.

Engl. J. Med., 316: 673, 1987.

50. EICHMANN, K.; EMMRICH, F.; KAUFMANN, S.H.E. CRO Crit. Rev. Immunol., 1987.

SciELO

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 71-76, 1991

USE OF TRYPANOSOMA CRUZI RECOMBINANT

ANTIGENS IN THE IMMUNOLOGICAL DIAGNOSIS

OF CHAGAS' DISEASE.

Samuel Goldenberg', Marco Aurélio Kríeger 1 , Juan J. Lafaille

Elza Almeida 12 & Walter Oelemann’, Fundação Oswaldo Cruz, (1)

Dept. Bioquímica e Biologia Molecular and (2) Biomanguinhos. Rio

de Janeiro, RJ, Brasil.

INTRODUCTION

Chagas' disease is still a major endemic problem in South America where some 20

million individuais display seropositivity for Trypanosoma cruzi, causative agent of the

disease (Chagas, 1909). The life-cycle of the parasite involves two intermediary hosts

(the triatomine insect and mammals) and three well-defined developmental forms, the

infective non-replicative trypomastigotes, the replicative amastigotes (mammalian form)

and epimastigotes (insect form) (Brener, 1973). The infection occurs when trypomasti¬

gotes, released with the excreta of the triatomine, penetrate into the mammalian host

through a wound or mucosa. Alternatively, Chagas' disease can be transmitted by

transfusion with infected blood. In recent years, the transmission of Chagas' disease

by the triatomine invertebrate host has diminished in virtue of improvements in vector

control campaigns in some of the endemic countries. However, new cases of this dis¬

ease still occur due to blood transfusion with infected blood.

Transfusional Chagas' disease frequently occurs as a result of incomplete or defi-

cient diagnosis of the disease. In a few cases, there is a complete lack of blood control

and the transfusion is direct (arm-to-arm). in most cases, only one test is used for the

screening of the blood, normally agglutination, which results in many false negative re¬

sponses due to the poor sensibility of the test (Carrasco et al., 1985). However some

Correspondence Io: Samuel Goldenberg, Dept. Bioquimica e Biologia Molecular, Fundação

Oswaldo Cruz, Av. Brasil 4365, 21040 Rio de Janeiro - RJ, Brazil.

SciELO

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 71-76, 1991

of the blood is screened using at least two different tests. But even in this case, some

false results still arise due to the low sensitivity of some methods or antigens, or cross

reactivity of T. cruzi extracts used in the diagnosis with other diseases such as leishma-

niasis, syphilis, toxoplasmosis, etc. (Camargo & Takeda, 1979).

These problems might be solved by purifying parasite specific antigens from para¬

site extracts. However, these antigens are not generally available and their production

costs would be too high, in addition to provoking the risk of infection of people involved

in the manipulation of the parasite.

On the other hand, the cloning and expression of T. cruzi genes in bactéria might

provide antigens from the parasite with the required specificity and at low cost. Accord-

ingly, recent work from several laboratories has resulted in the cloning and expression

of T. cruzi genes in E. coli (Ibanez et al., 1988; Lafaille et ai., 1989; Hoft et al., 1989;

Levin et al., 1989; Cotrin et al., 1990; Paranhos et al., 1990). The results presented be-

low describe the cloning and characterization of T. cruzi specific antigens and their use

in the diagnosis of Chagas's disease.

RESULTS

Cloning of T. Cruzi Antigens

The successful attempts to clone and express T. cruzi genes in bactéria coincided with

the description of the lambda gtl 1 vector (Young & Davis, 1983). The general strategy

consisted of the screening of T. cruzi expression libraries with either chagasic sera oranti-

seraraised against the parasite. These expression libraries were either cDNA (Hoft et al.,

1989; Levin et al., 1989) orgenomic (Peterson et al., 1986; Dragon et al., 1987; Ibanez et

al., 1988; Lafaille et al., 1989; Cotrim et al., 1990; Zingales et al., 1990), in the latter case

taking advantages of the fact that the genes of the parasite are intronless.

In the work carried out in our laboratory, DNA was extracted from T. cruzi epimasti-

gotes and cloned in lambda gtl 1 after shearing to a mean size of 1 kb and subsequent

addition of EcoRI linkers. The recombinant DNA library in E. coli was immunologically

screened with a trypomastigote - specific antiserum and the positive clones were fur-

ther screened with a pool of chagasic sera (Lafaille et al., 1989). Twelve clones posi¬

tive with the two sera types were then purified to homogeneity.

In order to investigate the specificity of the recognition of these recombinant clones

by chagasic sera, we tested the respective fusion proteins by western blot analysis

against different human sera. These sera comprised both chagasic sera and sera from

patients bearing diseases which cross-react antigenically with Chagas' disease. In this

selection, two of the 12 recombinant clones were considered specific for Chagas' dis¬

ease diagnosis.

Characterization of the Clones

The selected recombinant clones were characterized in terms of their structure and

expression (Lafaille et al, 1989; Krieger et al., 1990). One of the antigens is located in

the region of the flageilum adjacent to the body of the parasite, while the other is dis-

tributed in the cytoplasm of T. cruzi. Nucleotide sequencing analysis demonstrated that

both antigens are composed of repetitive epitopes: the flagellar antigen is composed of

2 3

5 6

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 71-76, 1991

repetitions of a 68 aminoacid motif, while the repeat unit of the cytoplasmic antigen

contains 14 aminoacids. Thsse antigens were then named FRA (flagellar repetitive an¬

tigen) and CRA (Cytoplasmic repetitive antigen) (Lafaille et al., 1989). The consensus

aminoacid sequence of the repetitive epitopes is shown in Fig. 1.

Cytoplasmic Repetitive Antigen (CRA)

LYS ALA ALA GLU ALA THR LYS VAL ALA GLU ALA GLU LYS GLN

Flagellar Repetitive Antigen (FRA)

MET

GLU

GLN

GLU

ARG

GLN

LEU LEU GLU LYS ASP PRO ARG

ARG

ASN

ALA

LYS

GLU

ILE

ALA

LEU GLU GLU SER MET ASN ALA

ARG

ALA

GLN

GLU

LEU

ALA

ARG

GLU

LYS LYS LEU ALA ASP ARG ALA

PHE

LEU

ASP

GLN

LYS

PRO

GLU

ARG

VAL PRO LEU ALA ADP VAL PRO

LEU

ASP

SER

ASP

PHE

VAL

ALA

Figure 1 - Consensus aminoacid sequence of the repeat unit of CRA and FRA

lt is interesting to note that most of the genes screened from T. cruzi expression librar-

ies display various copies of repetitive motifs, indicating that repetitive epitopes are

highly antigenic. Accordingly, other groups described antigens similar to CRA (Ibanez

et al., 1988; Hoft et al., 1989; Levin et al., 1989) and to FRA (Ibanez et al., 1988; Levin

et al., 1989; Cotrim et al., 1990). Despite the ubiquitous nature of these antigens, CRA

and FRA are highly polymorphic in T. cruzi (Krieger et al., 1990). This polymorphis can

be seen at the genomic levei, where distinct restriction fingerprints are obtained for

CRA and FRA in different T. cruzi strains.

Use Of Recombinant Antigens In Chagas' Disease Diagnosis

Although polymorphic in different strains, these antigens were recognized in the

form of B-galactosidase fusion proteins by more than 95% of chagasic sera when indi-

vidually tested using a radioimmunoassay procedure. This indicated that the antigenic

determinants are well conserved. However, our first attempts to use these recombi¬

nant antigens in an ELISA test showed that some non-chagasic sera displayed a bor-

derline response, very likely due to their reactivity with the B-galactosidase portion of

the fusion proteins. In order to circumvent this problem, we adopted a strategy which

consisted of expressing these recombinant antigens in the pMSgtll vector (Scherf et

al, 1990). This vector contains a cleavage site for factor Xa in the cloning site, hence

allowing the enzymatic cleavage of the fusion protein with the subsequent release of

the B-galactosidase. We first tested CRA expressed in pMSgtl 1 and the results were

highly satisfactory: in a multi-center study carried out by the World Health Organization

and involving nine laboratories and 24 antigens, it was concluded that CRA ranked as

the best individual reagent (Moncayo & Luquetti, 1990).

_ t—1

r o

= [£l

=_o

— t—1

- t—1

_ t—1

- t—1

_ t—1

Mem. Inst. Butantan, v. 53, supl. 1, p. 71-76, 1991

However, a doubtful response was observed for some sera, as a result of reactivity

too close to the cut-off serum control. When a mixture of CRA and FRA was used in-

stead of the individual antigen, it became evident that some of the doubtful sera were

in fact positive for Chagas' disease. Consequently, in order to improve the ELISA test

we started to use a mixture (1:1) of CRA and FRA, resulting in a highly specific diagno-

sis reagent (Almeida et al., 1990; Krieger et al„ in preparation).

The CRA+FRA ELISA was then tested using sera from different endemic regions,

sera from patients bearing diseases which present cross-reactivity with Chagas' dis¬

ease, and negative sera from endemic areas and from blood banks. The results

were compared to those obtained with conventional serological tests (haemagglutina-

tion and indirect immunofluorescence), and to those obtained using an ELISA test

consisting of a cytosolic extract (CYTO) of T. cruzi as antigen source. It was obser¬

ved that the CRA+FRA ELISA, in addition of recognizing all tested chagasic sera, did

not react with sera from patients bearing other diseases. On the contrary, both the

Cyto ELISA (Table I) and the conventional serology tests (Table II) gave some false

positive responses.

TABLEI

Comparision of CRA + FRA and CYTO ELISA with different human sera

Sera

CRA+FRA

CYTO

Chagasic(n=221)

221

Negative 1 (n=193)

Negative? (=49)

Schistosomiasis (n=15)

Malaria (n=12)

Syphilis (n=14)

Leishmaniasis (n=21)

1 - Negative sera from endemic areas

2 - Negative sera from blood bank

TABLE II

Comparision of the reactivity of conventional serological methods and CRA + FRA

ELISA with sera which cross-react with Chagas' disease

Sera

IHA

IFI

CRA+FRA

Visceral Leishmaniasis (n=5)

Cutaneous leishmaniasis (n=5)

Leprosy (n=2)

Lupus (n=8)

SciELO

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 71-76, 1991

CONCLUSIONS AND PERSPECTIVES

The data discussed above indicate that the use of T. cruzi recombinant antigens in

an ELISA test provides a safe and accurate diagnosis for Chagas' disease. The main

advantage of the recombinant ELISA is the fact that it gives very few (if any) false posi¬

tive results in comparison to other reagents and methods frequently used for the diag¬

nosis of the disease. These false positive responses should be avoided in virtue of the

social problems they can cause for the patient. In addition, recombinant antigens are

cheaper to produce than antigens isolated from the parasite, and this should have a di-

rect impact on the price of the diagnostic reagent. In the particular case of the FRA

and CRA antigens, their repetitive epitope structure suggests that synthetic peptides

might be used in diagnosis in a near future. Indeed, we have recently tested a synthet¬

ic CRA peptide and the results showed that 65% of the tested chagasic sera were de-

tected in ELISA. However, further investigations are necessary in order to determine

whether this observed diminution in sensitivity was due to a technical problem related

to the binding of the peptide to the ELISA plate or, alternatively, whether the problem

was related to a poor exposition of the correct antigenic determinants.

ACKNOWLEDGEMENTS

We thank Catherine Lowndes for the criticai reading of this manuscript. Our work

received finantial support from THE UNDP/WORLD BANK/WHO Special Programme

for Research and Training in Tropical Diseases, FINEP-PADCT and Conselho Nacion¬

al de Desenvolvimento Científico e Tecnológico (CNPq).

REFERENCES

1. ALMEIDA, E.; KRIEGER, M.A.; CARVALHO, M.R.; OELEMANN, W.; GOLDENBERG, S.

Use of recombinant antigens for the diagnosis of Chagas disease and blood bank screen-

ing. Mem. Inst. Oswaldo Cruz, 85 (4): 513-517, 1990.

2. BRENER, 2. Biology of Trypanosoma cruzi. Ann. Rev. Microbiol., 27: 347-382, 1973.

3. CAMARGO, M.E. & TAKEDA, G.K.F. Diagnóstico de laboratório. In: BRENER, Z. & AN¬

DRADE, Z„ eds. Trypanosoma cruzi e Doença de Chagas. Rio de Janeiro, Guanabara

Koogan., 1979. p. 175-198.

4. CARRASCO, R.L.; BRENIERE, S.F.; POCH, O.; MIGUEZ, H.V.; SELAES, H.; ANATEZANA,

G.; DESJEUX, P.; CARLIER, Y. Chagas Serology and its Problems. Ann. Soc. Belg. Med.

Trop., 65(1): 79-84, 1985.

5. CHAGAS, C. Nova tripanosomiase humana. Estudos sobre a morfolojia e o cyclo evolutivo

do Schizotrypanum cruzi n.gen., n.sp. agente etiolojico de nova morbidade do homen

Mem. Inst. Oswaldo Cruz, 1: 159-180, 1909.

6. COTRIN, P.C.; PARANHOS, G.S.; MORTARA, R.A.; WANDERLEY, J.; RASSI, A.; CAMAR¬

GO, M.E.; FRANCO DA SILVEIRA, J, Expression in Escherichia coli of a dominant immu-

nogen of Trypanosoma cruzi rocognized by Human chagasic sera. J. Clin. Microbiol., 28:

519-524, 1990.

7. DRAGON, E.A.; SIAS, S.R.; KATO, E.A.; GABE, J.D. The genome of Trypanosoma cruzi

contains a constitutively expressed, tandemly arranged multicopy gene homologous to a

major heat shock protein. Mol. Cell. Bioi, 7: 1271-1275, 1987.

8. GONZALEZ, A.; LERNER, T.J.; HUECAS, M.; SOSA-PINEDA, B.; NOGUEIRA, N.; LIZARDI,

2 3

5 6

11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 71-76, 1991

P.M. Apparent generation of a segmented mRNA from two separate tandem gene families

in Trypanosoma cruzi. Nucleic Acids Res., 73:5789-5804, 1985.

9. HOFT, D.F.; KIN, K.S.; OTSU, K.; MOSER, D.R.; YOST, W.J.; BLUMIN, J.H.; DONELSON,

J.E.; KIRCHHOFF, L.V. Trypanosoma cruzi expresses diverse repetitive protein antigens.

Infect. Immun., 57: 1959-1967, 1989.

10. IBANEZ, C.F.; AFFRANCHINO, J.L.; MACINA, R.A.; REYES, M.B.; LEGUIZAMON, S.; CA¬

MARGO, M.E.; ASLUND, L; PETTERSON, U.; FRASCH, A.C.C. Multiple Trypanosoma

cruzi antigens containing tandemly repeated aminoacid sequence motifs. Mol. Biochem.

Parasito!., 30: 27-34, 1988.

11. KRIEGER, M.A.; SALLES, J.M.; ALMEIDA, E.; LINSS, J.; BONALDO, M.; GOLDENBERG, S.

Expression and polymorphism of a Trypanosoma cruzi gene encoding a cytoplasmic repeti¬

tive antigen. Exp. Parasito/., 70: 247-254, 1990.

12. LAFAILLE, J.J.; LINSS, J.; KRIEGER, M.A.; SOUTO-PADRON, T.; DE SOUZA, W.; GOLD¬

ENBERG, S. Structure and expression of two Trypanosoma cruzi genes encoding antigen-

ic proteins bearing repetitive epitopes. Mol. Biochem. Parasito/., 35: 127-136, 1989.

13. LEVIN, M.; MESRI, E.; BENAROUS, R.; LEVITOS, G.; SCHIJMAN, A.; LEYATI, P.; CHIALE,

P.; RUIZ, A.M.; KAHN, A.; ROSENBAUM, M.; TORRES, H.N.; SEGURA, E.L. Identifica¬

tion of major Trypanosoma cruzi antigenic determinants in chronic Chagas disease. Am. J.

Trop. Mod. Hyg., 47:530-538, 1990.

14. MONCAYO, A. & LUQUETTI, A.O. Multicentre double blind study for evaluation of Trypanoso¬

ma cruzi defined antigens as diagnostic reagents. Mem. Inst. Oswa/do Cruz, 85: 489-495,

1990.

15. PARANHOS, G.S.; COTRIN, P.C.; MORTARA, R.A.; RASSI, A.; CORRAL, R.; FREILIJ, H.L.;

GRINSTEIN, S.; WANDERLEY, J.; CAMARGO, M.E.; FRANCO DA SILVEIRA, J. Trypa¬

nosoma cruzi: cloning and expression of an antigen recognized by acute and chronic hu-

man chagasic sera. Exp. Parasito/., 77:284-293, 1990.

16. PETERSON, D.S.; WRIGHTSMAN, R.A.; MANNING, J.E. Cloning of a major surface-antigen

gene of Trypanosoma cruzi and Identification of a nonapeptide repeat. Nature., 322: 566-

399, 1986.

17. SCHERF, A.; MATTEL D.; SCHREI8ER, M. Parasite antigenic expressed in Echerichia coli.

A refined approach for epidemiological analysis. J. Immunol. Meth., 723:81-87, 1990.

18. YOUNG, R.A. & DAVIS, R.W. Efficient isolation of genes by using antibody probes. Proc.

Nat. Acad. Sei., USA, 80: 1194-1198, 1983.

19. ZINGALES, B.; GRUBER, A.; RAMALHO, C.B.; UMEZAWA, E.S.; COLLI, W. Use of two re-

combinant proteins of Trypanosoma cruzi in the serological diagnosis of Chagas' disease.

Mem. Inst. Oswa/do Cruz., 85: 519-522, 1990.

2 3

5 6

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 77-86, 1991

POST-SYMPOSIUM LECTURE

PERSPECTIVES ON PRODUCTION OF GROUP B

MENINGOCOCCAL VACCINES

Cari E. Frasch, Laboratory of Bacterial Polysaccharides, Center for

Biologics Evaluation and Research, Food and Drug Administration,

Beihesda - Maryland

ABSTRACT: Group B meningococcal disease remains a problem in many countries.

Since the group B Neisseria meningitidis polysaccharide has proven not to induce pro-

tective antibodies, existing vaccine against group B disease have been composed of

lipopolysaccharide depleted outer membranes, usuaily in the form of small vesicles.

Protection against meningococcal disease is correlated with induction of bactericidal

antibodies. The group B vaccines stimulating the highest bactericidal titers when ad-

ministered in a two dose immunization series 6 to 8 weeks appart consist of soluble

vesicles, and one of the meningococcal polysaccharides all adsorbed to aluminum hy-

droxide. Efficacy trials with such vaccines have recently been conducted in Chile,

Cuba, and Norway. The Cuban trial demonstrated 80% efficacy against disease

caused by a B:4:P1.15 strain, and was the first to clearly demonstrate that antibodies

induced to non-capsular antigens can protect against meningococcal disease.

KEY WORDS: Neisseria meningitidis, vaccine, outer membrane

INTRODUCTION

Serogroup B Neisseria meningitidis is responsible for over 80% of meningococcal

diseases in Brazil and is the predominant cause of meningococcal disease in many oth-

Correspondence to: Cari E. Frasch, Division of Bacterial Products, HFB-640, Center for Biologics

Evaluation and Research, 8800 Rockville Pike, Bethesda, Maryland 20892 USA

Fax number: 301 480-4091

2 3

5 6

11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 77-86, 1991

er countries induding the United States. A recent study of serogroup prevalence in the

United States 26 showed that approximately equivalent amounts of meningococcal dis-

ease were due to groups B and C. However, the incidence of meningococcal disease

remains approximately 1/100,000 in the US compared to 2.5/100,000 in Brazil.

GROUP B POLYSACCHARIDE

Effective capsular polysaccharide vaccines against N. meningitidis serogroups A

and C were developed in the early 1970s 18 . Later the Y and W135 polysaccharides

were added to produce a tetravalent meningococcal vaccine 2 . Although the group B

polysaccharide would be the logical choice for a group B vaccine, it is poorly immuno-

genic and antibodies induced by the polysaccharide do not appear to be protective.

The B polysaccharide is a homopolymer of alpha 2-8 linked N-acetyl neuraminic acid,

the same as on some fetal proteins. This may account for the observed poor immunog-

enicity of the B polysaccharide.

Attempts have been made to improve the immunogenicity of the polysaccharide

22,27,28 Adsorption of the polysaccharide to aluminum hydroxide appeared promising in

mice 28 , but failed to increase immunogenicity in humans 14 . Another approach was to

prepare a chemically modified polysaccharide, in which the N-acetyl groups on the pol¬

ysaccharide were replaced with N-propionyl groups 21 ' 22 . When this altered polysaccha¬

ride was chemically bound to tetanus toxoid, forming a conjugate vaccine, the N-

propionyl polysaccharide induced bactericidal antibodies. Conjugates prepared using

the native group B polysaccharide were nonimmunogenic. Studies are continuing by

Dr. Jennings to evaluate this approach.

MENINGOCOCCAL OUTER MEMBRANE ANTIGENS

Protection against meningococcal disease is correlated with the presence of bacteri¬

cidal antibodies 18 . The peak incidence of disease occurs in children under 1 year of

age; as a group, they have few or no bactericidal antibodies. In addition, the high sus-

ceptibility of individuais with a deficiency of one of the terminal complement compo-

nents (C5, C6, C7, or C8) for invasive meningococcal disease strongly implicates the

importance of bactericidal activity in host defense against these organisms. Thus, a

group B meningococcal vaccine should be based on cell surface antigens shown to in-

duce bactericidal antibodies, although opsonic antibodies probably also contribute to

protection. Studies have shown that convalescent sera from patients with group B me

ningococcal disease have bactericidal antibodies directed against surface proteins and

the lipopolysaccharide 19 ' 23 ' 32 . Most efforts to develop an effective group B vaccine

have, therefore, involved use of outer membrane proteins 8 ' 41 .

There is a large amount of antigenic diversity among group B meningococcal

strains. The outer membrane contains three to tive major proteins, comprising 5 protein

classes, class 1 through class 5 having molecular weights between 26,000 and 46,000

daltons 36 . There are approximately 20 different serotypes within group B and group C

based upon immunologic differences in the Class 2 and Class 3 major OMPs which are

between 34K and 40K 15 . These cias 2/3 proteins are the meningococcal porins. Further

antigenic diversity is seen among the approximately 46K Class 1 OMPs, which contain

2 3

5 6

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 77-86, 1991

the subtype specific antigens and may also have porin function. Antibodies to both the

serotype and subtype proteins are bactericidal.

Although sporadic causes of group B meningococcal disease may be caused by a

variety of serotypes, outbreaks and epidemics in most countries have been caused by

a small number of serotypes, 2, 4, 8 and 15 1>10 . Thus, a serotype protein vaccine need

contain membranes from a relatively small number of serotypes. In addition to our la-

boratory, laboratories in Biltoven, The Netherlands 30 ; Havana, Cuba 34 ; Oslo, Norway

16 and Washington DC, USA 41 have produced group B meningococcal vaccines, all

based on use of lipopolysaccharide depleted outer membranes.

EARLY OUTER MEMBRANE VACCINES

Other membranes may be removed from meningococci by lithium chloride-sodium

acetate extraction at 50°C. These membranes contain approximately equivalent

amounts of protein and toxic lipopolysaccharide (LPS). For vaccine use, the LPS must

be largely eliminated by selective solubilization with detergents, a number of which

have been used for this purpose including Brij-96 13 , Empigen-BB 40 , and sodium deoxy-

cholate 37 . Sodium deoxycholate may be best because it is a normal bile metabolite

present in humans, and residual detergent that may be present in the vaccine would

probably not have toxic effects.

To prepare OMV vaccines free of unknown quantities of group B meningococcal pol-

ysaccharide, we have isolated non-encapsulated variants or mutants using horse anti-

group B polysaccharide serum incorporated into a clear agar médium to detect colo-

nies not elaborating the B polysaccharide 13 .

The first outer membrane vaccines consisted of LPS depleted membranes. The vac¬

cine protein was separated from the detergent by ethanol precipitation and resuspend-

ed in 0.9% sodium chloride.

These vaccines were visibly particulate and contained aggregated outer membranes

as observed by electron microscopy. Later studies showed that the membranes were

soluble in water, but not 0.9% sodium chloride.

The early particulate vaccines as well as later vaccines contained considerable

amounts of LPS (about 5 to 10 pg/100 pg protein), yet were much less pyrogenic in

rabbits than would be expected 35 . The LPS that remained was strongly membrane as-

sociated, which probably accounted for the lower toxicity.

The particulate outer membrane vaccines were evaluated in adults then in children

using a two or three dose immunization schedule previously evaluated in animal stud¬

ies 8 . Zollinger et al. 42 found that such a vaccine failed to induce bactericidal antibodies

in five adults after three doses. A similar particulate vaccine prepared in our laboratory

induced low leveis of antibody in both adults and children as measured by ELISA, but

also failed to stimulate bactericidal antibodies. Thus, although particulate vaccines

were found safe in both adults and children, they were poorly immunogenic, a fact not

predicted by the animal studies.

Zollinger et al. were first to clinically evaluate soluble outer membrane vaccines 42 .

They found that outer membrane vaccines could be made soluble by combination with

group B meningococcal polysaccharide. Electron microscopy of similar vaccines pre¬

pared in our laboratory showed some aggregation of outer membrane vesicles (OMV)

without the polysaccharide and individual vesicles with polysaccharide 13 .

2 3

5 6

11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 77-86, 1991

Meningococci release large amounts of essentially pure outer membranes during

normal growth into the culture broth as blebs or vesicles 7 ' 13 . These membranes may

be purified from the broth and used as the starting material for preparation of a vac-

cine 13 . Since the natural orientation of the proteins in the outer membrane may be im-

portant, we have developed methods to selectively remove the LPS, leaving the mem¬

branes intact and soluble as determined by electron microscopy 13 . Soluble vaccines

were prepared with and without the group B polysaccharide and tested in animais 29 ' 37 .

The soluble OMV vaccines are colloidal suspensions rather than true Solutions. In-

creasing the ionic strength caused precipitation of the OMV, whereas addition of a neg-

atively charged polysaccharide increased OMV solubility. The polysaccharide forms a

non-covalent complex with the vesicles 43 . Recent results from our laboratory show that

the OMV-polysaccharide association is hydrophobic, because removal of the lipid tail

from the polysacharide by phospholipase blocked the interaction.

Soluble protein plus polysaccharide vaccines have been clinically evaluated 8 ' 43 .

These vaccines induced bactericidal antibodies on primary immunization with only

modest increases in antibody titers after the second immunization. We then compared

the immunogenicity of a soluble OMV vaccine with and without group B polysaccha¬

ride 8 ' 33 . Addition of the polysaccharide resulted in a significant increase in bactericidal

antibodies to a group C serotype 2a strain. The second dose 6 to 8 weeks later result¬

ed in an increase in the percent of individuais responding to the vaccine.

The target age group for a group B vaccine is young children. When the immune re¬

sponses of children were compared to those of older children and adults, by either

OMV ELISA or bactericidal assay, young children (under 6 years old) responded less

well 12 . In an effort to increase the percentage of young children developing bactericidal

antibodies, the OMV vaccine was adsorbed onto aluminum hydroxide or 14 .

Adsorption of the OMV plus polysaccharide onto aluminum hydroxide or aluminum

phosphate significantly increased the bactericidal response of mice to the outer mem¬

brane proteins 37 . These vaccines were therefore evaluated in human adults 14 . The alu¬

minum hydroxide absorbed vaccine was found safe and more immunogenic than the

same vaccine without the adjuvant. The vaccine in combination with the adjuvant in¬

duced significantly higher bactericidal titers. More recent studies in Norway suggest

that the adjuvant can be added directly to LPS depleted membranes 20 .

In summary, these data provide evidence indicating that to stimulate an optimal im¬

mune response, the surface exposed protein epitopes need to be presented to the im¬

mune system in a near-native configuration, and therefore should remain within soluble

membranes.

CURRENT VACCINES AND RECENT CLINICAL STUDIES

A study in Norway 32 using a combined serotype 2b and serotype 15 OMV vaccine

showed that bactericidal antibodies were induced to both serotypes in 70% of the

adults tested. This demonstrated the utility of combining multiple serotypes. The study

also showed a good correlation between IgG antibodies to the outer membranes and

bactericidal titers measured using human complement.

For comparative clinicai studies in Norway 20 three different vaccine formulations

were prepared containing OMV from a B:15:P1.16 strain mixed with either group C me-

ningococcal polysaccharide, or with aluminum hydroxide, or with both. These vaccine

2 3

5 6

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 77-86, 1991

formulations were evaluated in adults and students in preparation for a large scale effi-

cacy trial. Two doses of protein between 12.5 and 100 pg per dose were given at a 6

week interval. The researchers found equivalent bactericidal responses when the pro-

leins were administered adsorbed to aluminum hydroxide with or without the group C

polyssacharide, and that the 50 pg dose was optimal.

A group B vaccine has been produced in Cuba consisting of LPS depleted OMV

from a B:4:P1.15 strain, a high molecular weight protein complex, and group C menin-

gococcal polysaccharide, all adsorbed onto aluminum hydroxide 34 . Care is taken to

control the vesicle structure and size. The vaccine contains per dose 50 pg protein, 50

pg polysaccharide, and 2 mg aluminum hydroxide. It is administered as a two dose im-

munization schedule with a 6 to 8 week interval. The dosage and interval were arrived

at after evaluation of different immunization schedules in adults and children. Eighty-

eight percent of school children responded with 2-fold or greater increases in outer

membrane antibodies as measured by ELISA.

GROUP B EFFICACY TRIALS

A number of efficacy trials have now been conducted with varying results using

group B meningococcal vaccines (Table 1). A number of important observations can be

drawn from these trials, but the foremost is that antibodies to non-capsular surface anti-

gens can prevent group B meningococcal disease 34 . The first trial was carried out in

Cape Town, South África in 1981 9 against a B:2b:P1.2 epidemic (peak incidence; 150/

100,000) using a 2a:P1.2 outer membrane vaccine combined with group B polysaccha¬

ride, but no adjuvant. Although insufficient cases occurred to estimate efficacy, no sero-

type 2 disease occurred in the vaccinated children, but equivalent amounts of disease

due to other group B serotypes occurred in vaccinated and control children (received

meningococcal AC vaccines). Thus the vaccine failed to protect against nonserotype 2

disease.

An efficacy trial was performed in Iquique, Chile in 1988-1990 using an outer mem¬

brane protein vaccine from a B:15:P1.3 strain combined with group C meningococcal

TABLEI

Field trials of group B meningococcal outer membrane vaccines

Years

Vaccine formulation

Location

Est.

Efficacy

Ref.

1981-82

2a:P1.2 + B polysacch

Cape Town,

South África

Too few cases

1987-89

4:P1.15 + C polysacch

+ AI(OH)3

Cuba

80%

1988-90

15:P1.3 + C polysacch

+ AI(OH)3

Iquique,

Chile

51%

1989-91

15:P1.16 +AI(OH)3

Norway

In progress

_ 1—1

r o

ü CiH

= _ O

— y—1

- y—1

_ y—1

- y—1

_ y—1

Mem. Inst. Butantan, v. 53, supl. 1, p. 77-86, 1991

polysaccharide and aluminum hydroxide 5 . During the trial over 90% of the group B me-

ningococcal disease was due to the B:15:P1.3 clone. In this double-blind trial 40,800

volunteers, ages 1-21 years, received two doses of either the B vaccine or ACYW135

polysaccharide vaccine given 6 weeks appart. The estimated efficacy was 51%. The

vaccine differed from other vaccines in that efforts were taken to reduce the LPS con-

tent to very low leveis, which probably disrupted the membrane structure. The antibody

responses of the children as measured by ELISA were good, but the numbers respond-

ing with bactericidal titers were low. This illustrates the need for measurement of bacte-

ricidal antibodies.

A randomized double-blinded efficacy trial was carried out in Cuba between 1987

and 1989 using the B:4:P1.15 vaccine described above 34 . The trial was conducted in

197 boarding schools, randomized by school, where there were 106,000 students be¬

tween 10 and 14 years of age, half of which received the serotype 4 vaccine. During

the trial 95% of group B disease was due to B:4:P1.15 and 3% to B:15:P1.15. Thus,

this vaccine, like the vaccine used in Chile, was evaluated against a single group B

clone. The estimated efficacy was 80%, clearly demonstrating that a protein vaccine

can prevent B:4:P 1.15 disease. However, since the epidemic was caused by one

clone, the trial was not able to provide evidence for the degree of protection that could

be expected against other group B strains.

IMPORTANT VACCINE CHARACTERISTICS AND PROBLEMS

WITH CURRENT VACCINES

Immunogenicity and efficacy studies conducted with a different outer membrane

vaccines have demonstrated a number of physical characteristics that are important for

optimal immunogenicity of these vaccines (Table 2). A number of studies have shown

that isolated outer membrane proteins induce few antibodies reactive against surface

exposed epitopes. These proteins have loop structures Crossing the membrane several

times 25 , that are not conserved upon removal from a membrane environment. We have

therefore sought to maintain the vesicle structure, following detergent treatment to re¬

move the LPS, and this is monitored by electron microscopy. Additional reasons to

work with the intact membranes are that antibodies to no one protein are likely to pro-

TABLE II

Important characteristics of a meningococcal outer membrane protein vaccine

1. Must be soluble - Solubility improved by addition of polysaccharide

2. Native conformation of proteins should be maintained by:

Retaining outer membrane structure

OR Insertion into liposomes

3. Maintain approximately 50 pg of LPS per mg protein to retain near native outer

membrane conformation

4. Outer membrane proteins normally expressed during infection should be included:

Iron regulated proteins

Stress proteins (heat shock)

2 3

5 6

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 77-86, 1991

vide broad protection against group B meningococcal disease, and we do not yet know

to which proteins the criticai protective antibodies are directed.

A minimum amount of LPS is required to maintain the outer membrane conforma-

tion, without which the membranes disintegrate. Isolated outer membranes contain ap-

proximately equal quantities of protein and LPS. Deoxycholate treatment removes

about 95% of the LPS without disrupting the membranes, while stronger detergent

treatment to remove additional LPS generally disrupts the membranes.

There are problems with all of the outer membrane protein vaccines that have re-

ceived clinicai evaluation to date. First, the existing vaccines are rather serotype specif-

ic. The antigenic composition of the vaccines need to be changed to provide for induc-

tion of broadly protective antibodies. In this regard, it now appears that antibodies to no

single protein will provide broad protection. Most all of the outer membrane proteins ap-

pear to have antigenic variants. Second, the vaccines fail to include a number of impor-

tant cell surface proteins that are expressed during infection. These in vivo proteins in¬

clude the iron regulated outer membrane proteins 3 - 4 , and probably heat-shock

proteins 39 . Third, the vaccines do not appear to induce a clear booster response as

would be expected of a protein antigen when the second immunization is given 6 weeks

afterthe first. In addition, no data have been presented demonstrating whether vaccina-

tion primes for a booster response if the child is reimmunized 6 months or a year after

the primary two doses immunization series. Most other protein vaccines are given as a

multiple immunization series. Lastly, it may be necessary to use genetic engineering to

remove the class 4 protein from the vaccine strains. These proteins are equivalent to

gonococcal Protein III, and Protein III has been shown to induce blocking antibodies 31 .

The class 4 protein has been removed without effecting growth properties 24 .

Although vaccines consisting of LPS depleted outer membranes offer the best ime-

diate approach, there are alternatives. An ideal meningococcal vaccine would be immu-

nogenic in all age groups and protect against all group B strains in addition to the other

disease associated serogroups, A, C, Y and W135. Development of such a vaccine will

likely require a better understanding of the basic mechanisms by which only some me¬

ningococcal strains are able to gain enterance into the host and cause disease. Wetz-

ler et al. 38 have successfully used purified outer membrane proteins inserted into lipo-

somal membranes to induce high leveis of bactericidal antibodies. Now we only need

to know which proteins should be included. Another very promising finding is that alka-

line detoxified meningococcal LPS remained immunogenic and induced bactericidal an¬

tibodies in mice 6 .

RESUMO: A doença meningococcica continua sendo um problema em muitos países.

Uma vez que o polissacáride da Neissería meningitidis B mostrou-se incapaz de indu¬

zir a formação de anticorpos protetores, as vacinas existentes contra a doença menin-

gocócica pelo grupo B tem apresentado em sua composição, membrana externa com

quantidade reduzida de lipopolissacáride.sob a forma de vesículas. A proteção contra

a doença meningococica está correlacionada com a indução de anticorpos bacterici-

das. A vacina contra o grupo B estimula a produção de altos títulos de anticorpos bac-

tericidas quando administrada em duas doses com intervalo de 6 a 8 semanas e con¬

siste de vesículas solúveis e um polissacaride de meningococo, todos absorvidos ao

hidróxido de alumínio. Testes de eficácia com tais vacinas foram recentemente feitos

no Chile, Cuba e Noruega. A vacina Cubana demonstrou 80% de eficácia contra a

SciELO

10 11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p, 77-86, 1991

doença meningococcica causada por cepas B:4:P1.15, e foi a primeira a demonstrar

claramente que anticorpos induzidos por antígenos não capsulares podem proteger

contra a doença meningococcica.

PALAVRAS CHAVES: Neisseria meningitidis, vacina, membrana externa

REFERENCES

1. ABDILLAHI, H. & POOLMAN J.T. Neisseria meningitidis group B serotyping using monoclonal

antibodies in whole-cell ELISA. MicrobialPath., 4: 27-32, 1988.

2. ARMAND, J.; ARMINJON, F.; MYNARD, M.C.; LAFAIX, C. Tetravalent meningococcal po-

lysaccharide vaccine groups A, C, Y, W135: clinicai and serological evaluation. J. Biol.

Stand, 10: 335-340, 1982.

3. BHATNAGAR, N.B. & FRASCH, C.E. Expression of Neisseria meningitidis iron-regulated out-

er membrane proteins, including a 70-Kilodalton transferrin receptor, and their potential for

use as vaccines. Infect. Immun., 58: 2875-2881, 1990.

4. BLACK, J.R.; DYER, D.W.; THOMPSON, M.K.; SPARLING, P.F. Human immune response

to iron-repressible outer membrane proteins of Neisseria meningitidis. Infect. Immun., 54:

710-713, 1986.

5. BOSLEGO, J.; ZOLLINGER, W.; GARCIA, J; CRUZ, C.; RUIZ, S.; BRANDT, B.; MARTINEZ,

M.; ARTHUR, J.; UNDERWOOD, P.; HANKINS, W.; MAYS, J.; GILLY, J.; THE CHILEAN

NATIONAL COMMITTEE FOR MENINGOCOCCAL DISEASE. Efficacy trials of a menin¬

gococcal group B (15:P 1.3) outer membrane protein vaccine in Iquique, Chile. In: INTER¬

NATIONAL PATHOGENIC NEISSERIA CONFERENCE, 7, 1991. Abstracts p.23.

6. DELVIG, A.; KRASNOPROSHINA, L; KUVAKINA, V.; DMITRIEV, B. Protective activity of de-

toxified lipopolysaccharides of Neisseria meningitidis serogroups A and B in mice model. In:

INTERNATIONAL PATHOGENIC NEISSERIA CONFERENCE, 7,1991. Abstracts p.119.

7. DEVOE, I.W. & GILCHR1ST. Release of endotoxin in the form of cell wall blebs during in vitro

growth of Neisseria meningitidis. J. Exp. Med., 138: 1156-1167, 1973.

8. FRASCH, C.E. Immunization against Neisseria meningitidis. In: EASMON, C.S.F. &

JELJASEWICZ, J. eds. Medicai microbiology. New York, Academic Press, 1983. v. 2,

pi 15-144.

9. FRASCH, C.E. Status of a group B Neisseria meningitidis vaccine. Eur. J. Clin. Microbiol., 4:

533-536, 1985.

10. FRASCH, C.E. Development of meningococcal serotyping. /n;VEDROS, N.A., ed. Evolution

of meningococcal disease. Boca Raton, CRC Press, 1987. p.39-54.

11. FRASCH, C.E. Vaccines for prevention of meningococcal disease. Clin. Microbiol., Rev., 2:

S134-S-138, 1989.

12. FRASCH, C.E.; COETZEE, G.J.; WU, L; WANG, L-Y.; ROSENQVIST E. Immune response

of adults and children to group B Neisseria meningitidis outer membrane protein vaccines.

/n;ROBBINS, J.B., ed. Bactéria! vaccines. New York, Praeger, 1987. p.262-272.

13. FRASCH, C.E. & PEPPLER, M.S. Protection against group B Neisseria meningitidis disease:

preparation of soluble protein and protein-polysaccharide immunogens. Infect. Immun., 37:

271-280.

14. FRASCH, C.E.; ZAHRADNIK, J.M.; WANG, L-Y; MOCCA, L.F.; TSAI, C.M. Antibody re¬

sponse of adults to an aluminum hydroxide adsorbed Neisseria meningitidis serotype 2b

protein group B polysaccharide vaccine. J. Infect. Dis., 158: 700-708, 1988.

15. FRASCH, C.E.; ZOLLINGER, W.D.; POOLMAN, J.T. Serotype antigens of Neisseria menin¬

gitidis and a proposed scheme for designation of serotypes. Rev. Infect. Dis., 7:504-510,

1985.

16. FROHOLM, L.O. Recent meningococcal epidemiology in Norway and eight years of experi-

ence with serotyping for strain characterization. Rational for vaccine development. In: IN-

SciELO

10 11 12 13

Mem. Inst. Butantan, v. 53, supl. 1, p. 77-86, 1991

TERNATIONAL PATHOGENIC NEISSERIA CONFERENCE, 7, 1991. Abstracts p.24.

17. GOLDSCHNEIDER, I.; GOTSCHLICH, E.C.; ARTENSTEIN, M.S. Human immunity to the me-

ningococcus. I. tho role of humoral antibiodes. J. Exp. Med., 129: 1307-1326, 1969.

18. GOTSCHLICH, E.C.; GOLDSCHNEIDER, I.; ARTENSTEIN, M.S. Human immunity to the

meningococcus. IV. immunogenicity of group A and group C polysaccharides in human vol-

unteers. J. Exp. Med., 129: 1367-1384, 1969.

19. GRIFFISS, J.M.; BRANDT, B.L.; BROUND, D.D.; GORROFF, D.K.; BAKER C.J. Immune re¬

sponse of infants and children to disseminated infections with Neisseria meningitidis. J. In-

fec. Dis., 150: 71-79, 1984.

20. HOIBY, E.A.; ROSENQVIST, E.; BJUNE, G.; CLOSS, O.; FROHOLM, L.O. Serological re¬

sponses in adult human volunteers to a meningococcal 15:P1.16 outer membrane complex

vaccine in Norway (phase II studies). In: INTERNATIONAL PATHOGENIC NEISSERIA

CONFERENCE, 7. 1991. Abstracts p.98.

21. JENNINGS, H.J.; GAM1AN, A.; ASHTON, F.E. N-propionylated group B menincococcal poly-

saccharide mimics a unique epitope on group B Neisseria meningitidis. J. Exp. Med., 165:

1207-1211, 1987.

22. JENNINGS, H.J.; ROY, R.; GAMIAN A. Induction of meningococcal group B polysaccharide-

specific IgG antibodies in mice using a N-propionylated B polysacharide-tetanus toxoid con-

jugato vaccine. J. Immunol., 137: 1708-1713, 1986.

23. JONES, D.M. & ELDRIDGE, J. Development of antibodies to meningococcal protein and lip-

opolysaccharide serotype antigens in healthy carriers. J. Med. Microbiol., 12: 107-111,

1979.

24. KLUGMAN, K.P.; GOTSCHLICH, E.C.; BLAKE, M.S. Sequence of tho structural gene (rmpM)

for the classe 4 outer membrane protein of Neisseria meningitidis, homology of the protein

to gonococcal protein III and Escherichia coli OmpA, and construction of meningococcal

strains that lack class 4 protein. Infect. Immun., 57: 2066-2071,1989.

25. MCGUINNESS, B.; BARLOW, A.K.; CLARKE, I.N.; FARLEY, J.E.; ANILIONIS, A.; POOL-

MAN, J.T.; HECKLES, J.E. Deduced amino acid sequence of class 1 protein (PorA) from

three strains of Neisseria meningitidis. Synthetic peptides define the epitopes responsible

for serosubtype specificity. J. Exp. Med., 171: 1871-1882, 1990.

26. MOCCA, L.F.; FRASCH, C.E.; WEAVER, R.E.; GELLIN, B.; PINNER, R.W.; THE MENINGO-

COCCAL DISEASE STUDY GROUP. Serotypes of Group B and Group C Meningococcal

Isolates from a 1986-87 Surveillance Study in the United States. J. infect. Dis., (in press).

27. MORENO, C.; LIFELY, M.R.; ESDAILE, J. Immunity and protection of mice against Neisseria

meningitidis group B by vaccination using polysaccharide complexod with outer membrane

proteins: a comparison with purified B polysaccharide. Infect. Immun., 47: 527-533, 1985.

28. MORENO, C.; LIFELY, M.R.; ESDAILE, J. Effect of aluminum ions on Chemical and immuno-

logical properties of meningococcal group B polysaccharide. Infect. Immun., 49: 587-592,

1985.

29. PEPPLER, M.S. & FRASCH, C.E. Protection against group B Neisseria meningitidis disease:

effect of serogroup B polysaccharide and polymyxin B on immunogenicity of serotype pro¬

tein preparations. Infect. Immun., 37: 264-270, 1982.

30. POOLMAN, J.T. Polysaccharides and membrane vaccines. Adv. Biotechnoi, Processes, 13:

57-86, 1990.

31. RICE, P.A.; VAYO, H.E.; TAM, M.R.; BLAKE, M.S. Immunoglobulin G antibodies directed

against protein III block killing of serum-resistant Neisseria gonorrhoeae by immune serum.

J. Exp. Med., 164: 1735-1748, 1986.

32. ROSENQVIST, E.; HARTHUG, S.; FROHOLM, L.O.; HOIBY, E.A.; BOVRE, K.; ZOLLINGER,

W.D. Antibody response to serogroup B meningococcal outer membrane antigens after

vaccination and infection. J. Clin. Microbiol., 26: 1543-1548, 1988.

33. ROSENQVIST, E.; TJADE, T.; FROHOLM, L.O.; FRASCH, C.E. An ELISA study of the anti¬

body response after vaccination with a combined meningococcal group B polysaccharide

2 3

5 6

11 12 13 14 15

Mem. Inst. Butantan, v. 53, supl. 1, p. 77-86, 1991

and serotype 2 outer membrane protein vaccine. NIPH Ann., 6: 139-149, 1983.

34. SIERRA, G.; CAMPA, H.C.; VALCARCEL, N.M.; GARCIA, I.L.; SOTOLONGO, P.F., 12-

QUIERDO, P.L.; CASANUEVA, G.V.; BARO, S.M.; LEGUEN, C.F.; RODRIGUEZ, C.R.;

TERRY, H.M. Efficacy evaluation of the Cuban vaccine VA-MENGOC-BC against group B

Neisseria meningitidis caused disease. In: ACHTMAN, M., cd., Proceedings Seventh Inter¬

national Pathogenic Neisseria Conference, 1991 (In press).

35. TSAI, C.M.; FRASCH, C.E.; RIVERA, E.; HOCHSTEIN, H.D. Measurements of lipopolysac-

charide (endotoxin) in meningococcal protein and polysaccharide preparations for vaccine

usage. J. Biol. Stand., 17: 249-258, 1989.

36. TSAI, C.M.; FRASCH, C.E.; MOOCA, L.F. Five structural classes of major outer membrane

proteins in Neisseria meningitidis. J. Bacteríol., 146: 69-78, 1981.

37. WANG, L-Y & FRASCH, C.E. Development of a Neisseria meningitidis group B serotype 2b

protein vaccine and evaluation in a mouse model. Infec. Immun., 46: 408-414, 1984.

38. WETZLER, L.M.; BLAKE, M.S.; GOTSCHLICH, E.C. Characterization and specificity of anti-

bodies to protein I of Neisseria gonorrhoeae produced by injection with various protein I-

adjuvant preparations. J. Exp. Med., 168: 1883-1897, 1988.

39. WOODS, M.L.II; BONFIGLIOLI, R.; MCGEE, Z.A.; GEORGOPOULOS, C. Synthesis of select

group of proteins by Neisseria gonorrhoeae in response to themal stress. Infect. Immun.,

58: 719-725, 1990.

40. ZOLLINGER, W.D.; BOSLEGO, J.; BRANDT, B.L.; MORAN, E.E.; RAY, J. Safety and antige-

nicity studies of a polyvalent meningococcal protein-polysaccharide vaccine. Antonie van

Leewenhoek, 52: 225-228, 1986.

41. ZOLLINGER, W.D.; BOSLEGO, J.; FROHOLM, L.O.; RAY, L.S.; MORAN, E.E.; BRANDT,

B.L. Human bactericidal antibody response to meningococcal outer membrane protein vac-

cines. in: POOLMAN, J.T., ed. Gonococci and Meningococci. Dordrecht, Kluwer Academic,

1990. p.209-217.

42. ZOLLINGER, W.D.; MANDRELL, R.E.; ALTIERI, P.; BERMAN, S.; LOWENTHAL, J.; ARTEN-

STEIN, M.S. Safety and immunogenicity of a meningococcal type 2 protein vaccine in ani¬

mais and human volunteers. J. Infect. Dis., 137: 728-739, 1978.

43. ZOLLINGER, W.D.; MANDRELL, R.E.; GRIFFISS, J.M.; ALTIERI, P.; BERMAN, S. Complex

of meningococcal group B polysaccharide and type 2 outer membrane protein immunogen-

ic in man. J. Ctin. Invest., 63: 836-848, 1979.

SciELO

10 11 12 13

AUTHORS' INDEX

ÍNDICE DE AUTOR

ALMEIDA, E. 71

BERGMANN, I. E. 59

CAMPUS, G. 21

CONTU, B. 21

COSSU, M. A. 21

DOMENIGHINI, M. 15

FRASCH, C. E. 77

GOLDENBERG, S. 71

KRIEGER, M. A. 71

LAFAILLE, J. J.71

MARSILI, I. 21

NENCIONI, L 15, 21

NEWTON, S. M. C. 53

OELEMANN, W. 71

PEPPOLONI, S. 21

PIZZA, M. G. 15

PODDA, A. 15, 21

RAPPUOLI, R. 15, 21

SILVESTRI, S. 15, 21

STOCKER, B. A. D. 53

TORDO, N. 31

VANNI, R. 11, 21

VOLPINI, G. 21

INSTRUÇÕES AOS AUTORES

1. Somente serão aceitos trabalhos inéditos e que se destinem cxdusivamentc à revista. É proibida a reprodução

com fins lucrativos. Os artigos de revisão serão publicados a convite da Comissão Editorial.

2. Os trabalhos deverão ser redigidos cm português, inglês ou francês, datilografados preferencialmente cm máquina

elétrica, em espaço duplo cm 3 (três) vias, em papel formato ofício e numerados no angulo superior direito.

3. No preparo do original será observada, sempre que possível, a seguinte estrutura: Página de rosto: título do arti¬

go, nomefa) do(s) autor(es) e filiação científica. Texto: introdução, material e métodos, resultados, discussão, con¬

clusões, agradecimentos e referência bibliográfica. Material de referência: resumos (em português e inglês); uni-

termos (palavras ou expressões que identificam o conteúdo do artigo; devem ser incluídas até um limite máximo

de três, cm português e inglês).

4. As referências bibliográficas deverão ser ordenadas alfabeticamente e numeradas.

Exemplos:

Para livros: autor, título, edição, local de publicação, editor, ano, páginas.

7. BIER, O. Microbiologia e imunologia. 24.ed. São Paulo: Melhoramentos, 1985.1234p.

Para artigos: autor, título do artigo, título do periódico, volume, página inicial e final, ano.

8. MACIIADO, J.C. & SILVEIRA P, J.F. Obtenção experimental da pancreatite hemorrágica aguda no cão por ve¬

neno escorpiônico. Mcm. Inst. Buiantan,v, 40/41, p. 1-9, 1976/77.

As citações no texto devem ser por números-índices correspondentes às respectivas referências bibliográficas.

Exemplos:

... método derivado de simplificação de armadilha de Disney 1

... segundo vários autores 13 - 1

5. As ilustrações (fotos, tabelas, gráficos etc.) deverão ser originais e acompanhadas de legendas explicativas. As le¬

gendas serão numeradas e reunidas em folha à parte. Os desenhos deverão ser a nanquim e as fotografias bem

nítidas, trazendo no verso o nome do autor o a indicação numérica da ordem a ser obedecida no texto. As ilus¬

trações deverão ser organizadas de modo a permitir sua reprodução dentro da mancha da revista (22 x 12,5cm).

6. Os artigos deverão conter no máximo 6 (seis) ilustrações (branco e preto). De cada trabalho serão impressas 50

(cinquenta) separatas, sendo 10 para a Biblioteca do Instituto e 40 para os autores.

7. Os textos originais não serão devolvidos c os originais das ilustrações estarão à disposição dos autores.

1NSTRUCTIONS TO AUTHORS

1. Manuscripts submitted to the Editorial Board should be unpublished texts and should not be under consideration

for publication elsewhere. Reproduction for commercial purposos is not allowcd. The Editorial Board vvill plan

the publication of revlsion artlcles.

2. The original and tvvo copies of papers should be typevvritten in Portuguese, English or Fronch, double spaced, on

typing paper (31 x 21cm). Pages should be numbered consecutively at the upper right comer,

3. The following strueture should be considered in the preparation of the manuscript: Title page: with articlc title,

nume of author(s), professional address. Text: with introduction, material and methods, rcsults, discussion, con-

clusions, acknoH-ledgments, references, abstraets (in Portuguese and English), and keywords. A maximal number

of 03 keywords should be includcd in Portuguese and English.

4. References in alphabetical order should be numbered consecutively.

Examples:

Books

7. BIER, O. Microbiologia e imunologia. 24.cd. São Paulo: Melhoramentos, 1985. 1234p.

Artides

8. MACI IADO, J.C. & SILVEIRA F°, J.F. Obtenção experimental da pancreatite hemorrágica aguda no cão por ve¬

neno escorpiônico. Mcm. Inst. Butantan, v. 40/41, p. 1-9, 1976/77.

Citations in lhe text should be identificd by the reference number.

Examples:

... método derivado de simplificação de armadilha de Disney 1

... segundo vários autores 13 - 1

5. lllustrations (photographs, tablcs, figures etc.) should be the originais and lcgends should be submitted typevvrit¬

ten on a separate sheet. Line-drawings should be with China ink and photographs must be of top quality. On the

back of each figuro of photograph the name of the author(s) should be lightly written and the number indicating

the sequence in the text. lllustrations should (it in a page measuring 22 x 12,5cm.

6. No more than 6 illustrations will be accepted and photographs should be black and white. Fifty reprints of each

article are provided without chargc, and 10 will be kept at the library.

7. Submitted manuscripts will not be returned to the author(s) but the original illustrations are available to author(s)

by request.

SciELO

11 12 13 14 15 16 17