Experimental neuroinvasiveness of wild and laboratory junin virus strains

(~) INSTITUT PASTEUR/ELSEVIER Paris 1992

Res. Virol. 1992, 143, 259-267

Experimental neuroinvasiveness of wild and laboratory Junin virus strains S.I. Medeot, M.S. Contigiani, M.S. Sabattini and G. Diaz Instituto de Viroiogia "'Dr. J.M. Vanella", Facuitad de Ciencias Medicas, Universidad Nacional de Cordoba, Ciudad Universitaria, Estafeta 32, 5016 Cordoba (Argentina)

SUMMARY

The neuroinvasiveness of Candid 1 and XJCL3 laboratory strains and CbalV4454 and CbaFHA5069 wild strains of Junin virus was studied in albino mice, guinea pigs, and a South American wild rodent, Calomys musculinus (Cm), of different ages inoculated by a non-neural route. Infectivity in brain, blood and organs, as well as lethality, were determined. The results with the 3 hosts indicate that Junin virus neuroinvasiveness is virus-straindependent, host species- and age-dependent, with the Candid 1 strain proving to be the least neuroinvasive of the strains studied. The lethal efficiency index (log PFU/LDso) in 2-day old albino mice and the neuroinvasiveness index (Log PFU/NDso) in 6 + 1 dayold Cm of the various strains using the intraperitoneal (ip) route could therefore be useful markers of Junin virus neuroinvasiveness. Moreover, different patterns of infection were established using the results of the presence of infectious virus in brain and viraemia in the 3 hosts. In nearly all cases, virus neuroinvasion was present without detectable viraemia (virus in plasma).'Current evidence leads to the assumption that virus might reach the brain associated with the white cells in blood (undetectable by conventional isolation methods) or by another possible mechanism of neuroinvasion which is not haematogenous.

Key-words: Junin virus, Central nervous system; Neuroinvasiveness, Experimental infection.

INTRODUCTION Junin virus (Arenaviridae) is the aetiological agent of Argentine haemorragic fever (AHF), an acute and severe disease, with moderate morbidity in some areas of the country and with a casefatality rate of between 15 and 30 % in untreated cases (Maiztegui et al., 1979). AHF exhibits varying degrees of central nervous system (CNS) involvement in the acute stage. A late neurological

Submitted March 11, 1992, acceptedJune 29, 1992.

syndrome without fatal consequences associated with antiserum treatment has been also described (Maiztegui et al., 1979). Despite progress in understanding the aetiology and diagnosis of AHF, the CNS pathophysiology in humans is not well understood, although immunopathogenic phenomena cannot be ruled out (Cossio et al., 1977). Several studies have been conducted in animals and a varying degree of CNS involve-

260

S.L M E D E O T E T AL.

ment has been also reported. Suckling mice and rats have been found to show T-lymphocytemediated fatal encephalitis (Besuschio et al., 1973 ; Giovanniello et al., 1973 ; Taratuto et al., 1973; Nota et al., 1977; Weissenbacher et al., 1987), but the mechanisms o f CNS damage have not been elucidated for monkeys, guinea pigs and C a l o m y s musculinus (Cm), the main Junin virus reservoir. Recently, a differential neurovirulence of Candid 1 vaccine strain of Junin virus in different hosts at various ages was described (Medeot et al., 1990). This appears to be important in order to investigate the neuroinvasiveness of the attenuated candidate vaccine strains, as it has led to field trials in human volunteers to evaluate the innocuity and protective efficacy of the Candid I strain (Dr. J.I. Maiztegui, personal communication). Therefore, we initiated a systematic and comparative study of laboratory and wild strains of Junin virus in 3 animal hosts under defined experimental conditions. The proposed vaccine strain, Candid 1, has been included in this study. This reports describes the lethality and viral growth curves in the brain in albino mice, guinea pigs and Cm inoculated by an extraneural route. The viraemia and viral growth curves in organs were also examined, since experimental studies and observations in man have indicated that most viral CNS infections are acquired from the blood (Johnson, 1982).

MATERIALS

AND METHODS

Viral strains The following strains of Junin virus were used : XJCL3 obtained from a highly passaged XJ (2 guinea pigs and 14 mice passages), cloned in rabbit MA-I 11 cells (Guerrero et al., 1969) and used in this study

Argentine haemorragic fever.

AHF Cm CNS

= = =

central nervous system.

ic

=

intracerebral(ly).

im ip

= =

intramuscular(ly). intraperitoneal(ly).

Calomys musculinus.

with 9 suckling mice passages after cloning. Candid 1 strain (2 guinea pigs, 44 mice, and 19 foetal Rhesus lung cells passages), developed by Dr. J. Barrera Oro and kindly provided by Dr. J.I. Maiztegui, National Institute of Viral haemorragic Disease (INEVH), Pergamino, Argentina. The CbaFHA5069 and CbalV4454 wild strains recovered from a fatal human case and a mild AHF respectively, were taken through 5 passages in suckling mice. Neurovirulence properties of these strains for mice, guinea pigs and Cm have been previously reported (Medeot et ai., 1990).

Animals The 3 host species used in these experiments were outbred in our laboratory as previously described (Medeot et al., 1990). The age and the sex of the animals were selected in accord with the results of previous works (Contigiani and Sabattini, 1984; Medeot et ai., 1990) including the animal systems that have exhibited virus replication in brain and a different behaviour between the distinct virus strains when inoculated by the intracerebral (ic) route. Mice of both sexes, 2-, 11- and 14-days old, were used. I 1-day old guinea pigs of both sexes and adult and 6 + 1 dayold Cm were also used.

Experimental design Junin virus neuroinvasiveness was studied by determining the growth in brain tissue and the lethality of each viral strain after a non-neural inoculation in each host. Viraemia and viral growth curves in visceral and lymphatic organs were also determined. Lethality was determined by assessing the efficiency of lethal infection in mice after intraperitoneal inoculation (ip) of serial decimal dilutions of viral stocks in groups of 8 mice to obtain the 50 °70 lethal dose (LDs0), in parallel with the determination of the plaque-forming units (PFU) titre in C176 Vero cell cultures. The efficiency of lethal infection index was expressed as the log PFU/LDs0 (Contigiani and Sabattini, 1977) for each viral strain and animal age. Instead, in Cm, lethality rates were obtained from the daily number of deaths occurring over a 30-day period in animals inoculated ip with a single dose (3.1 to 4.6 Log PFU) of each viral stock.

LDso MEM NDso pi PFU

= = = = =

50 % lethal dose. minimal essential medium. 50 ~0 neuroinvasion dose. post-inoculation. plaque-forming unit.

J U N I N VIRUS N E U R O I N V A S I V E N E S S For viral growth, 3 to 5 animals of each host and age were inoculated with approximately 3,0 log PFU of each strain and sacrificed at different times postinoculation (pi). Samples of brain, lymph nodes, spleen, kidney, lung, liver, salivary gland and blood, diluted l:10 in minimal essential medium (MEM) with 5 mg 07o of gentamicin sulphate, were kept frozen at -80°C for virus assay.

6

261

Log PFU/0.1 ml

dose: 3.5 Log PFU Log PFUILD60:1.96

Virus assay C176 Vero cell cultures were employed for virus assay (Earley et al., 1967). Total blood samples and organ pieces were homogenized 1:10 (w/v) in MEM with 5 °7o of foetal bovine serum and 5 mg % gentamicin sulphate and then centrifuged at 11,400 g and 4°C for 30 min. Supernatants were then titrated by plaquing 10-fold serial dilutions in cell culture monolayers under agarase as previously described (Contigiani and Sabattini, 1977). By employing this conventional viral isolation technique, we detected only plasma viraemia.

0

5

I

i

I

I

10

15

20

2,5

pi days RESULTS Albino mice In this host, virus was recoved from brain and blood only with the CbaFHA5069 strain in 2-day old mice (fig. 1). Viraemia was detected, as in brain, from day 7 pi. Moreover, the lethality index obtained indicates that approximately 2 log PFU produced 1 LD50 (fig. 1). No virus in brain or blood and no lethality were detected in 11- and 14-day old mice with CbaFHA5069, and in 2-, 11- and 14-day old mice with XJCL3 and Candid 1 strains (data not shown). These results indicate that, although the presence of virus has been reported in the brain of these animals inoculated by the ic route (Medeot et al., 1990), the virus did not reach the CNS when inoculated by an extraneural route.

•

Brain

t

Blood

Fig. 1. Brain and blood virus titres in 2-dayold albino mice inoculated by the ip route with the CbaFHA5069 strain of Junin virus. Each point representsthe mean value of 2 to 4 animals. Each graph indicates the virus dose inoculated.

tween these two strains (Contigiani and Sabattini, 1984). Viral growth curves in the brain and blood are shown in figure 2. Both strains were recovered from brain with similar virus titres, but no viraemia was detected in the two cases. Moreover, no virus was found in lymph nodes, lung, spleen, liver and kidney at any of the pi time points shown in figure 2 with both strains (data not shown).

Guinea pigs

C a l o m y s musculinus Eleven-day old animals were inoculated intramuscularly (im) with the CbaIV4454 and XJCL3 strains in accord with previous results that suggested a different neuroinvasiveness be-

As stated above, we selected 6 + 1-day old Cm inoculated by the ip route with CbaFHA5069, XJCL3 and Candid 1 strains, and adult Cm

262

2.5

S.l. M E D E O T E T A L .

CbalV4454

XJCL3

STRAIN

STRAIN Log PFU/0.1nl 2.5

Log PFU/0.1ml

2.0

2.0

1.5

1.5

1.0

1.0

dollg:, ' ~1.....

0.5

0.5 0,0 |

0

doee:

I I I I 5 10

I 15

I 20

I 25

l

30

pi days --

Brsin --t-- BIooa

0.0'

I= 1 '

0

5

I I I 10 15

1 20

25

30

clays

• Brain -4-- Blood

Fig. 2. Brain and blood virus titres in 11-dayold guinea pigs inoculated im with two strains of Junin virus. Each point represents the mean value of 2 to 3 animals. Each graph indicates the virus dose inoculated.

inoculated by the ip route with CbaFHAS069 and XJCL3 strains. In all these animal-strain systems, we had previously demonstrated virus replication in brain when inoculated by the ic route (Medeot et al., 1990). Viral growth curves in the brain and blood, and lethality expressed as the rate between dead and total inoculated animals are shown in figure 3. The growth curves of organ-derived virus from the same animals are shown in figure 4. The CbaFHAS069 strain was uniformly recovered from the brain and blood from day 11 pi, but there were no deaths during the observation period. Variable viral curves from organs depending on the organ and age of the animal were obtained with this strain (figure 4). The XJCL3 strain was present in the brain of 6 + 1 day-old Cm until 30 days pi, but virae-

mia was irregularly detected between days 11 and 20 pi. This strain was lethal for these animals (4/14) between days 12 and 21 pi. In contrast, this strain was not recovered from the brain or blood of adult Cm. Moreover, no virus was detected in lymph nodes, liver, spleen or salivary glands of Cm of either age (data not shown). Finally, the Candid 1 strain was recovered from brain of 6 + 1 day-old Cm, but no virus was detected in blood or organs at any of the days pi tested. This strain proved to be somewhat lethal for these animals (3/14) between days 10 and 28 pi. The similar viral growth curves in the brain o f 6 + 1-day old C m obtained with the 3 strains led us to compare brain virus titres and the day of virus appearance in these animals inoculated ip with different doses of the CbaFHAS069 and Candid 1 strains, in order to detect any possi-

J U N I N VIR US N E U R O I N V A S I V E N E S S

263

N3tJkTS

6.*1 DAYS OLD 8 Log PFUIO. 1 ml

8 Log PFUIO, I ml

1.

L, 0/24 cloee: 8.2 Log PFU

0119 !

.

8

4

cboFHA 5069

0

/

2

.°

1

0

0 10

20

30

:

: :

0

40

:' 10

I

I

gO

8O

40

eO

40

p, Oay~ Lo~ H:u/o.1 *w

t..o9 PFU/0.1 ml

L= 0114 dole: 8.1 Log PFU

elb 4114 Idoel= 4.5 Log PFU

XJCL3

1

0

i

0

10

O0

2O

.

0

"'

.

.

.

. . . . 10

.

.

.

! gO

{

claye

Log PFU/0.1 n./

doel~. 4.3 Log PFU

4

Cand id 1

. . . . . .

Fig. 3. Brain and blood virus titres in C. muscuiinus of different ages inoculated by the ip route with 3 strains of Junin virus.

8

Each point represents the mean value of 2 to 3 animals L=animals dead/total inoculated. Each graph indicates the virus dose inoculated.

2

1

0

0

10

~

gO

Bnlin

80

.--4- ~

40

264

S.I. M E D E O T E T A L .

ADtJkTS

6 .+ 1 DAYS OLD Log PFU/0.1 rnl

Log PFU/0,1 ml

5

5

~-

4.1t Log PFU 4

4

3

1

0

fH L 10

0

2

I

i

20

3o

4o

o

1

I

I

~o

2o

3o

days Lt~,Ir

-4-- I _ n o f l e s

40

~ days SDleet~

-Q-

S.glan~l

~

Ll~r

--l.-- L . n o ~ 8

---1(-- ~ a l ~

~

S,gla~8

Fig. 4. Organ virus titres in C. musculinus of different ages inoculated by the ip route with the CbaFHA5069 strain of Junin virus. Each point is the mean value from the same animals as in figure 3.

ble difference in the virus neuroinvasiveness between the strains. The results are shown in table I. Cumulative brains with infectious virus were determined at the 14th day pi; then the 50 °70 dose of neuroinvasion (NDso) was calculated for each strain according to Reed and Muench (1938), while simustaneously each virus strain was titrated by plaque assay in Vero cells. The neuroinvasion index for each strain was expressed as virus titre in log(PFU/0.1 ml)/virus titre in log (NDs0/0.1 ml), meaning the log PFU of each viral strain is necessary to produce 1 ND50 in this host (table I). The CbaFHA5069 strain was always recovered from brain, regardless of the viral dose employed, since day 9 pi and with similar virus titres. Only 2.2 or less log PFU of this strain

produced 1 NDS0. On the other hand, the Candid I strain was recovered only from day 9 pi from the brain of Cm inoculated with the highest dose; 4.2 log PFU of this strain were necessary to produce 1 ND50. Brains from days 2, 5, 7 and 8 were negative with both strains (data not shown).

DISCUSSION

The results described in this work clearly show that the neuroinvasiveness of Junin virus, as previously established for its neurovirulence (Medeot et al., 1990), depends on the viral strain, animal species and age. Three different patterns of infection were established using the results of the viral growth in

J U N I N VIRUS N E U R O I N V A S I V E N E S S

265

Table I. Brain virus titres and neuroinvasiveness index of CbaFHA5069 and Candid 1 strains of Junin virus inoculated ip into 6 + 1-day old Calomys musculinus. Viral strain CbaFHA5069

Pi day 9 11 14

Candid 1

9 11 14

Dose (') 4.70 3.18 2.70 4.70 3.18 2.70 4.70 3.18 2.70 4.62 3.20 2.15 4.62 3.20 2.15 4.62 3.20 2.15

N ('')

(10- J ~ (10 - 2 3" (lo -3 J (10 -] ~ (10 -2t (lo -3 (lO - t (10 -2"~ (10 -3"t (nd) (lO-') (lO-') (nd) (10- J) (10 -2 ) (nd) (10- l) (10 -2 ) •

•

Brain titre Mean ('") Range

3/3 3/3 3/3 3/3 3/3 3/3 3/3 3/3 3/3 3/3 0/3

4.25 4.38 3.37 4.50 3.30 2.90 4.10 3.98 3.75 3.40 0 (.... )

0/3

0

2/2 0/2 0/2 3/3 0/3 0/3

4.52 0 0 3.95 0 0

3.70-4.85 4.00-4.75 3.00-3.77 4.20-4.95 3.00-3.70 2.20-3.25 3.90-4.55 3.15-4.28 3.40-3.95 3.00-3.70

Log PFU/NDs0 ~< 2.2

4.2

--

4.25-4.85 3.25-4.32

(*) Log PFU (stockdilution). (**)Brainswithvirus/total testedper dilution. (***)Log PFU/0.1 mi (meanvalue). (****)Negative in 10 ~/obrain suspension, nd= not diluted.

the brain and blood in the 3 hosts (table II), as follows. 1) Lack of neuroinvasiveness without evidence of viraemia, as shown by the Candid I strain in 2-day old mice; the XJCL3 strain in 2- and 11-day old mice and adult Cm; and the CbaFHA5069 strain in 11 and 14 days-old mice. 2) Neuroinvasiveness with the presence of viraemia, as seen with the CbaFHA5069 strain in 2-day old mice and 6 _+ 1-day old and adult Cm, and with the XJCL3 strain in 6 + 1 day-old Cm. 3) Neuroinvasiveness without detectable viraemia, as found with the CbaIV4454 and XJCL3 strains in 11-day old guinea pigs, and the Candid 1 strain in 6 + 1-day old Cm. The high viraemia levels developed by the CbaFHA5069 strain concomitant with brain virus appearing on approximately day 10 in most systems analysed suggests that plasma viraemia did not initiate viral replication in the brain. It

may represent a secondary vira,emia caused by release of virus from brain tissue into the bloodstream. Moreover, Junin virus neuroinvasion without detectable viraemia was repeatedly observed in the 3 hosts studied. In order to assess the role o f viraemia in the spread of virus to the brain, the demonstration that viraemia preceded the invasion o f the CNS appears to be important (Johnson, 1982). Nevertheless, in many infections, blood-borne virus is cell-associated and sometimes undetectable by conventional methods. The orthomyxoviruses and paramyxoviruses as well as some togaviruses and papovaviruses adsorb to red blood cells (Johnson, 1982). Other viruses can be isolated from human leukocytes during viraemia (Chantler et al., 1981 ; Scott et al., 1980). Transit within white cells not only protects the virus from phagocytosis by the reticuloendothelial system, but also shields it from neutralization by circulating antibody or inactivation

266

S.I. MEDEOT E T AL.

by non-specific serum inhibitors. Therefore, virus could easily gain access to CNS.

strain and the laboratory strains (Candid 1 vaccine) of Junin virus.

Recently, Junin virus was isolated from peripheral blood mononuclear cells during an acute period of A H F (Ambrosio et al., 1986). Thus, current evidence suggests that cellular viraemia, not detected by the isolation method employed, might be related to the virus neuroinvasion without detectable viraemia demonstrated here (table II).

The presence of the XJCL3 virus strain in the brain of I 1-day old guinea pigs that were inoculated with 3.29 log PFU did not correlate with previous results (Contigiani and Sabattini, 1984) that have shown that more than 5.18 log PFU of this strain were necessary to produce 1 LDs0 in these animals inoculated by the im route, suggesting that it was not neuroinvasive. However, the apparent differing results might be related to heterogeneity in the virulence of different viral subpopulations in the viral stock of the XJCL3 strain, as reported in a previous work (Contigiani and Sabattini, 1987). Moreover, an irregular lethal response was also observed with the lowest stock dilutions of this strain in 11-day old guinea pigs (Contigiani and Sabattini, 1984). A remarkable result of this work is that the Candid 1 vaccine strain was the least neuroinvasive, favouring its use as a human vaccine.

As shown in table II, the presence of virus in brain could therefore be a useful marker of Junin neuroinvasiveness in the 3 hosts studied, although the lethality and the presence of virus in blood are not reliable markers of virus neuroinvasion. A differing neuroinvasion ability between the distinct strains was demonstrated in albino mice and Cm. The lethal efficiency index (log PFU/LDs0) in 2-day old mice and the neuroinvasion index (Log PFU/NDs0) in 6 + 1-day old Cm of the various strains using the ip route could therefore be useful markers of Junin virus neuroinvasiveness. These markers differentiated the neuroinvasion capacity of the wild CbaFHA5069

Table

II. Different patterns of infection of Junin virus.

Pattern

Infectious virus Brain Blood -

II

+

+

III

+

-

Systems CbaFHA5069 11- and 14-day old mice; XJCL3 2- and l 1-day old mice; and adult Cm Candid 1-2-day old mice CbaFHA5069 2-day old mice and 6 + l-day old and adult Cm; XJCL3 6:t: l-day old Cm CbalV4454 11-day old guinea pigs; XJCL3 l l - d a y old guinea pigs; Candid 1 6 + 1-day old Cm

Nevertheless, the demonstrated neuroinvasiveness of the Candid 1 strain in suckling Cm and of the XJCL3 laboratory strain in adult Cm and 11-day old guinea pigs points to its possible dangers as a vaccine for children. More detailed studies are needed with these animal models to elucidate the possible mechanisms of Junin virus neuroinvasion by attenuated strains. At the present time, experiments are being carryied out in our laboratory to assess the role of infected blood ceils in haematogenous Ju= nin virus spread to the CNS. Moreover, the "in vivo" Junin virus neuroinvasiveness markers defined in this work are important in order to detect an increase in neurovirulence or neuroinvasiveness with the attenuated strains, candidates for human vaccines, since a neurotropism shift of a given strain of Jurtin virus has been reported after multiple passages in guinea pigs (Oubifia et aL, 1985).

Acknowledgements

We thank Dr. J.I. Malztegui for providing the Candid 1 strain of Junin virus and V.R. Vega for technical help. One of the authors (S.M.) was the recipient of a fellowship from CONICET.

JUNIN VIRUS NEUROINVASIVENESS This study was supported by grants from the Secretaria de Ciencia y Tecnica de la Nacion and the Consejo Nacional de Investigaciones Cientificas y Tecnicas (CONICET).

Invasion neurale exp~rimentale par des souches sauvages et att~nu~es du virus Junin L'invasion neurale par des souches att6nu6es Candid 1 et XJCL3 et par des souches sauvages CbaIV4454 et CbaFHAS069 du virus Junin, a 6t6 analys6e chez des souris albinos suisses, chez des cobayes et chez Calomys musculinus (Cm) d'ages divers, infect6s par une voie autre que la voie neurule. Le pouvoir l~tal du virus et le pouvoir infectant dans le cerveau, le sang et les organes, ont 6t~ d6termin6s. Les r6sultats obtenus indiquent que l'invasion neurale par le virus Junin est variable: elle d6pend de la souche virale, de l'esp6ce h6te et de l'fige de l'animal. La souche Candid 1 a 6t6 la souche la moins envahissante de toutes. Le taux d'infection 16tal (Log PFU/DLs0) chez des souris fig6es de 2 jours, et le taux d'invasion neurale (Log PFU/NDs0 ) chez Cm ag6s de 6 + 1 jours infect6s par voie p6riton6ale, deviennent des rep6res utiles de l'invasion neurale par le virus Junin. Trois diff6rents modules d'infection ont ~t~ 6tablis, d'apr6s la pr6sence du virus infectieux duns le cerveau et dans le sang. Etant donn6 que duns certains cas la pr6sence du virus dans le cerveau n'est pus en relation avec celle du virus duns le plasma, cela nous sugg6re que le virus pourrait arriver duns le cerveau de ces animaux associ6 aux cellules blanches sanguines, sans toutefois nous faire exclure la possibilit6 de la pr6sence d ' u n autre m~canisme d'invasion neurale non h6matologique. Mots-clds: Virus Junin, Syst6me nerveux central; Infection exp~rimentale, M6canisme invasif.

References Ambrosio, A.M., Enria, D.A. & Malztegui, J.I. (1986), Junin virus isolation from lympho-mononuclear cells of patients with Argentine Hemorrhagic Fever. Intervirology, 25, 97-102. Besuschio, S.C., Weissenbacher, M.C. & Schmufiiz, G.A. (1973), Different histopathological response to Arenavirus infection in thymectomized mice. Arch. ges. Virusforsch., 40, 21-28.

267

Contigiani, M.S. & Sabattini, M.S. (1977), Virulencia diferencial de cepas de virus Junin pot marcadores biologicos en ratones y cobayos. Medicina (Bs.As.), 37, 244-251. Contigiani, M.S. & Sabattini, M.S. (1984), El cobayo lactame como indicador diferencial de la virulencia de las cepas atenuadas de virus Junin. Medicina (Bs. As.), 44, 376-382. Contigiani, M.S. & Sabattini, M.S. (1987), Heterogeneidad en la virulencia de subpoblaciones virales derivadas de una cepa atenuada de virus Junin. Rev. lat.-amer. Microbioi., 29, 345-352. Cossio, P.M., Laguens, R.P., Malztegui, J.I., Rabinovich, A., Gonzalez, P.H. & Arana, R.M. (1977), Preseacia de antigenos de virus Junin e inmunogiobulinas autologas depositadas in vivo en Sistema Nervioso Central (SNC) de 4 casos fatales de Fiebre Hemorragica Argentina. Medicina (Bs.As.), 37, 503-504. Chantler, J., Ford, D. & Tingle, A. (1981), Rubellaassociated arthritis: rescue of rubella virus from peripheral blood lymphocytes two years post vaccination. Infect. Immun., 32, 1274-1280. Earley, E., Peralta, P.H. & Johnson, K.M. (1967), A plaque neutralization method for arboviruses (32194). Proc. Soc. exp. Biol. (N.Y.), 125,741. Giovanniello, O.A. & Boxaca, M.C. (1973), Effect of cyclophosphamide on Junin virus infection of mice. Medicina (Bs.As.), 33, 368-376. De Guerrero, L.B., Weissenbacher, M.C. & Parodi, A.C. (1969), Inmunizacion contra la FHA con una cepa atenuada del virus Junin. q I. Estudio de una cepa modificada del virus Junin. Inmunizacion de cobayos. Medicina (Bs.As.), 29, 1-10. Johnson, R.T. (1982), Pathogenesis of CNS infections, in "Viral infections of the Nervous System" (R.T. Johnson) (.pp. 37-60). Raven Press, New York. Malztegui, J.I., Fernandez, N.J. & Da~lano, A.J. (1979), Efficacy of immune plasma in treatment of AHF and association between treatment and a late neurological syndrome. Lancet, 8, 1216-1217. Medeot, S.I., Contigiani, M.S., Brandan, E.R. & Sabattini, M.S. (1990), Neurovirulence of wild and laboratory Junin virus strains in animal host. J. Med. Virol., 32, 171-182. Nota, N.R., Nejamkis, M.R. & GiovannieUo, O.A. (1977), Patogenia de la encefalitis del raton infectado por virus Junin. Medicina (Bs.As.), 37 (supl. 3), 114-120. Oubifia, J.R. & Carballal, G. (1985), Neurotropism of a high-passage XJ strain of Junin virus. J. Med. Virol., 15, 157-161. Reed, L.H. & Muench, H. (1938), A simple method of estimating 50 per cent end points. Amer. J. Hyg., 27, 493-497. Scott, R., Nisalak, A., Cheamudon, U., Seridhornakul, S. & Nimmannitya, S. (1980), Isolation of dengue viruses from peripheral leukocytes of patients with hemorrhagic fever. 3". infect. Dis., 141, 1-6. Taratuto, A.L., Nota, N.R., Nejamkis, M.R. & Giovanniello, O.A. (1973), Junin virus encephalitis in mice: its inhibition by antithymocyte serum. Arch. ges. Virusforsch., 43, 173-183. Weissenbacher, M.C., Laguens, R.P. & Coto, C.E. (1987), Argentine Hemorrhagic Fever. Curr. Top. Microbiol. Immunol., 134, 79-116.