Pathogenesis of Rift Valley fever virus (RVFV) in inbred rats

Microbial Pathogenesis 1987, 2 : 283-29

Pathogenesis of Rift Valley fever virus (RVFV) in inbred rats * G . W .Anderson, Jr ."fl[.W . Slone, Jr . z and C . J . PnterS 3 'Department of Immunology anal Infectious Disease, 71e,Johns Hopkins University School of Hygiene and Public Health, Baltimore, Maryland 21205, 8~A .;iEL DuPnnrC», Hascall Laboratory, Nemm4, Delaware /97/y//S .4 . ; 3Disease Assessment Division, United States Army Research Institute of Infectious Diseases, Frederick, Maryland 21701-5011 U.S-A

(Received September 17, 1 986; accepted in revised form December 12, 1986)

Anderson, G . W . (Dept . of Immunology and Infectious Disease, The Johns Hopkins University Sohnol of Hygiene and Public Health, Baltimore, Maryland 21205, U .S .A.) T . W. Slone and C . J . Peters . Pathogenesis of Rift Valley fever virus (RVFV) in inbred rats . Microbial Pathogenesis 1987 ;2 :288-293 .

The pathogenesis of Rift Valley fever in adult rats from 3 inbred strains (LBW. MAXX, WF) was investigated . VVIF rats all died by day 2 postinoculation with viral tissue titers reaching 9 log,, PFU/0 . LEW and MAXX rats were resistant to liver disease, but fatal necrotising encephalitis developed in 16 and 44% of the rats, respectively . Detection of serum neutralising antibody on day 3 coincided with clearance of virus from serum and liver, although infectious virus was detected in spleen homogenates as late as day 1 9 postinfection . Viral titers in LEW and MAXX rats did not exceed 4 .5 iog,v PFU/g . Cyclophosphamide immunosuppression of LEW rats led to death 5-9 days postinfection ; early patterns of viral replication were not affected, but continued growth in the liver resulted in fatal hepatitis . These animals could be protected by passive antibody therapy administered on days 2-5 postinfection to mimic theserum neutralising antibody pattern seen in unmanipulated infected LEW rats . Thus, RVF virus replication and spread is rapid in the VVF rats tissues, whereas in LEVVand MAXX rats viral growth is less due to an intrinsic mechanism which allows sufficient time for an immune response to terminate infection . A slightly diminished immune response may lead to the development of encephalitis more frequently in yWAXXthun LEVVeto . These rat strains should be useful in duciclating those mechanisms of resistance which limit RVFV-induced hepatitis and encephalitis . Key words : Rift Valley fever virus ; rats ; cyclokde ; pathogenesk ; Phlebovirus ; encepha!Nc

Introduction

Rift Valley fever (RVF) is an arthropod-borne disease of sub-Saharan Africa . Because it causes high mortality and abortion rates among domestic ruminants, particularly of Europeanurigin'RVFiuofuonxidomb!oooonomioanda0riou!tunm!inopununuo .Dospite large numbers of human infections in nature, as well as laboratory infections, only a ^/nou^uownomonmoarcxumm,iueuinmivmpo*moin,ooinutomxuhomutomn"suiu^/o,camand Use of Laboratory Animals", as promulgated by the Committee on Care and Use of Laboratory Animals of the Institute of Laboratory Animal Resources, National Research Council . The facilities are fully accredited by the American Association for Accreditation of Laboratory Animal Care . The views of the authors do not purport to reflect the position of the Department of the Army or the Department of Defense . , Disease Assessment Division, Fort Detrick, Frederick, 0882-40 10/87/040283+11 $03 .00/0

xxo2)7o1sn11 .u .s.A. @ 1987 Academic Press Inc . (London) Ltd



Pathogenesis of Rift Valley fever virus

285

I

100a~ ~a~ o.,o 1(

90 80 L t 70 >>_ 60 U)

50

a '&

`

a

~~FTf R

i L

k

(87/104)

r

i T I 1

-

~'T FTTTTTY'F•

(66/117)

0 40 -r 30

1 I 1

20 i

10 1(0/24) 0 _i I I I I I I I 25 5 10 15 20 30 Days

Fig . 1 . Comparative survival of 3 adult rat strains inoculated subcutaneously with 5 .7 log 10 PFU RVFV . Symbols : ∎, WF ; V, MAXX ; A, LEW .

90- Serum

m

0 U a-0

7 .0 5.0 3 .0 1 .0

n f

- Spleen

T. Ir -T -T - I

0 .7

- Kidney

Brain

0 9.0 J

Liver

7.0 5 .0 ?r '

3 .0 1 .0

0 .7

8

56 104 240

-

456

8

56 104 240 456

8

56 104 240 456

Hours Fig . 2 . Comparative viral tissue titers of 3 adult rat strains inoculated subcutaneously with 5 .7 log 10 PFU RVFV . Symbols : /, WF; V, MAXX ; A, LEW. Each point is a geometric mean based on the results from 3

rats.

late as day 19 in the spleen of some rats . A maximal viral titer of 4 .3 log 10 PFU/g occurred at 80 hours in the spleen . Virus was detected in the brain of all MAXX rats demonstrating CNS signs ; the titers exceeded 5 .7 log10 PFU/g in moribund rats . The LEW rats had a peak viremia of approximately 3 .0 log,, PFU/ml at 40 hours postinfection . Infectious virus was not detected in sera after 40 hours although virus was still detectable in the liver and spleen . The maximum viral spleen titer of 4 .0 log 10 PFU/g at 80 hours postinfection was similar to that of the MAXX rats . The spleen was the major site of viral replication in both the MAXX and LEW rats, though the liver appeared to represent the first site of significant replication . Pathology of RVF

Sequential histological examination revealed different patterns of pathology in adult rat strains . Small foci of hepatocellular necrosis were evident as early as 8 hours postinfection in the WF rats . Liver necrosis progressed until the majority of hepatocytes were necrotic or degenerating, except for a single row of normal cells around some



286

G . W . Anderson et al.

portal areas and central veins . A scattered neutrophilic infiltration accompanied the necrosis by 16 hours and was present thereafter . Periarteritis, characterised by edema and a neutrophilic infiltrate in the adventitia of large and small pulmonary arteries, occurred in the lung of 1 WF rat at 16 hours and 2 at 24 hours . Mild inteeoitial pneumonia, comprised of an infiltrate of neutrophils, which were often necrotic, was present in all WIF rats at 32 and 40 hours . Moderate numbers of intact and necrotic neutrophils were located in red pulp of the spleen from 24 to 40 hours . Mild lymphoid depletion of the periarteriolar lymphatic sheath occurred at 32 and 40 hours in the WF rats . Occasionally, oma!l amounts of cellular debris and a few neutrophils were located in glomerular tufts of the kidney at 24 and 32 how& The number of glomeruli affected increased at 40 hours, with essentially no change in the number of neutrophils . Additionally, hyaline droplet degeneration and protein casts occurred in proximal Wbules at 40 hours . Occasional neutrophils, often necrotic, and individual necrotic epithelial cells were located in the cortex of the adrenal gland at 32 and 40 hours . In the MAXX rats, a few necrotic and degenerative hepatocytes were seen in the liverfrom 24 to 80 hours postinfection . Mild lymphoid depletion of splenic periarteriolar lymphatic sheaths occurred at 80 and 104 hours . On day 14 then was adid to moderate neubtising encephalitis, characterised by neuronal necrosis, with an associated infiltrate of neutrophils, many of which were karyorrhectic in 1 of 3 rats . Perivascular accumulations of lymphocytes occurred in enna!l numbers in brain and meninges . Similar, but more severe, encephalitic lesions occurred in two moribund rats, 1 examined on day 11 and 1 at day 14 . On day 19, there were only small numbers of perivascular lymphocytes in the brain and meninges of 1 out of3 K8AXXra1s . The findings in the LEW rats mimicked the MAXX rats, though additional rats with signs of CNS disease had to be added to the study group to doe, the encephalitic lesions due to the low occurrence of CNS disease and path6ogy . Tissue samples from oon1rol animals for a!l three strains were normal histologically . Detection of viral antigen by immunofluorescence Due to the high viral content of the Wood, viml tissue titers were not considered indicative of sites of viral replication . Therefore, in determining these sites, viral antigen accumulation was used as the criterion . The liver and the adrenal cortex of the WF rat strain exhibited faintly stained single cells by 8 hours postinfection . This progressed to intensely staining cells 24 to 40 hours postinfection, with most of the liver cells positive for RVFviruF antigen . 8y24 hours postinfection, scattered fluorescent granules were detected in most areas of the spleen (except the periarteriolar sheath), renal glomeruli, cerebrum, and littoral macrophages of the lymph node . The ma/ginul zone of the spleen had a large accumulation of viral antigen 40 hours postinfection . Single cell foci appeared 24 hours postinfection in the liver and cortex of the adrenals in LEW and MAXX rats . These foci increased in size to 2-4 cells by 80 hours postinfection, which correlates well with histological observations . Figure 3 (a) and (b) compares the difference in viral antigen distribution in livers taken from LEW and IN raw 24 hours pootinfeotiow The cerebrum of moribund MAXX and LEW rats with CNS signs contained large foci of vkal antigen "which correlates with the areas of necrotding encephalitis seen histologically . Detection of neutralising antibody Tub!o1oompernothoneutno!ioingantibudyre*ponseofLEVVandK8AXXrntsinonu}ated subcutaneously with 53 /og,o PFU of infectious RVFV . Eight rats per strain were bled

287

Pathogenesis of Rift Valley fever virus

no . 3 . Specific immunofluorescence of liver cells 24 hours postinoculation from rats inoculated subcutaneously with o .7!ou,,ppunvpv . (a) WF . (u) Law, Table 1 Neutralising antibody response (PRN80) of rats to the ZH501 strain of Rift Valley fever virus' Days posinfecdon nmotain 3

7

14^

MAXX

33' (10-8n^

891 676 (320-2560) (320-1280)

LEW

47' (z0-1m8

1349 1047 (640-1280) (1340-2560

'Eight rats were inoculated with 5 .7 mg," PFU RVFV . 'P < 0 .02, t-test, comparing difference in antibody response on day 14 . 'Reciprocal of geometric mean titer, pnw m,w = 8 . All 8 were positive at 3, 7, 14 days . 'Range.

serially, and neutralising antibody was detected in all sera by day 3 when viremias had been cleared . The neutralising antibody responses were equivalent on days 3 and 7 postinfection, but were two-fold higher in LEW rats on day 14 (P < 0 .02) . To determine if the WF rat strain could respond to RVF viral antigens, a comparison of the neutralising antibody response to an inactivated RVF vaccine administered subcutaneously was conducted among the 3 rat strains (Table 2) . On day 3 postinoculation, LEW rats had significant increases in antibody titer compared to WF and MAXX rats . After day 3, there was no significant difference between WF and LEW rat geometric mean titers, but on day 4, the LEW and WF rats had 10/10reopooduo '



288

G . W. Anderson et al.

Table 2 Neutralising antibody response to Rift Valley fever virus vaccine in rats' Days post vaccination Rat strain

1

2

3

4

5

14

WF

<55

6 .5(2)` (10-20)°

9 .3(3) (10-40)

19 .1 (10) (10-40)

43 .7(10) (20-80)

69 .2(10) (20-160)

MAXX

<5

6 .3(2) (20)

10 .0(5) (10-30)

11 .2(5) (10-60)

24 .5(10) (20-60)

39 .8(10) (20-160)

LEW

<5

6 .3(2) (20)

15 .1(6) (20-80)

30.2(10) (20-80)

64.4(10) (30-160)

91 .2(10) (20-320)

a Rats were inoculated subcutaneously with 0.4 ml of the NDBR 103 formalininactivated RVF viral vaccine . e Reciprocal of geometric mean titer, PRN 80 . Positive responders out of 10 rats . 'Range of positive values .

6 .0 5 .0 E 4 .0 u-

a.

0s 3 .0 J -J

2 .0

1 .0

2 .0

1 .0 Minutes Log

)o

Fig . 4. Clearance of RVFV from the plasma of rats inoculated with 7 .0 log lo PFU RVFV via the intracardiac route . Each point is a geometric mean (± standard deviation) based on data from 7 rats . Zero time point calculated by assuming 0 .17 kg rats with a blood volume of 10 .9 ml . Symbols :', WF; A, LEW.

compared to 5/10 for the MAXX rats . The neutralising antibody titers of LEW rats were significantly increased on days 5, 7, and 14, compared to MAXX rats . Viral clearance Since large differences in clearance have been associated with virulence in alphaviruses, we measured the ability of WF and LEW rat strains to clear 7 .0 log o PFU of RVFV injected intracardially . The ability to clear infectious virus was equivalent over a 1 hour period (Fig . 4) . Immunosuppression and serum therapy LEW rats were immunosuppressed with a subcutaneous sublethal dose of cyclophosphamide (50 mg/kg) on day -1, and inoculated subcutaneously with 5 .7 log o PFU of RVFV on day 0 . The mortality in this group was 100% (30/30, MTD 6 .3 ±1 .4



28 9

Pathogenesis of Rift Valley fever virus

uu n .n No Vo 2 .0

9

4 .0 Vo 2 .0 /'»
5

7

/

e»

5

7

/

2

3

5

D"s Fig . 5 . Comparative RVFV tissue titers of normal and cyclophosphamide-immunosuppressed LEW rats after subcutaneous inoculation of 5.7 log,, PFU of RVFV . Points represent geometric mean titers of 3 rats . Symbols : A . Normal LEW; ', LEW treated with cyclophosphamide (50 mg/kg) given 1 day before infection .

days) . The immunosuppressed, infected rats died between days 4 and 11 with liver necrosis . To deteremine if immunosuppression affected early or late events, virological and histopathological parameters were followed sequentially . Immunosuppression had little effect on viral tissue titers occurring during the first 24 hours, but thereafter, the immunosuppressed rats could not limit viral replication undl after day 5 (Fig . 5) . By day 7 ' when viral titers were declining, in addidon to severe liver necrosis, some rats shooed evidence of hepatocellular regeneration and Kupffer cell hyperplasia . Two moribund rats examined on days 8 and 9 had cleared their viremias and had neutralising titers of 1 :640 and 1 :1280, The effect of immunosuppression on the antibody response was determined by inoculating suppressed LEW rats with the formalin-inactivated RVFV vaccine . Appearance of neutralising antibodies xvaa delayed by about 2 days, compared to the nonsuppressed controls (data not shown) . We reconstituted the delayed antibody response of immunosuppressed, infected rats with virus-specific immune serum (Fig . 6) . Administration of serum on days 2-5 to mimic the !evel of neutralising antibodies developing in unmanipulated infected LEW rats protected a!! 10 animals from fatal disease . If half the dosage of serum was given, 44% (4/9) of We animals died . FiNtopathological examination of moribund rats on days 10 through 13 showed necrotising encephalitis as the fatal lesion with only mild biliary hyperplasia in the liver . Discussion and conclusions

RVFV infection of inbred rats can result in fulminant disease with extensive liver necrosis, fatal encephalitis, or an immunising infection with no obvious signs of



290

G . W . Anderson at al.

100 (10/10)

90 80 70 0

o-o o-o

I O I

60

(}oOOO-)-o-OOOO .O°OO • (5/9)

50 40

I

30 t 20 10

'V y y y §\ (0/30) 0 1 )1?--U-O-9 I I 1 1 5 10 15 20 25 30 Days

Fig . 6 . Effect of RVFV specific serum therapy on survival of cyclophosphamide (50 mg/kg intraperitoneally on day -1)-immunosuppressed LEW -rats infected subcutaneously with 5 .7 log,, PFU RVFV . Symbols : 0, Cyclophosphamide, no therapy ; 0, Low-dose serum therapy ; A, High-dose serum therapy; 1, Arrows indicate administration of serum therapy .

illness . $ In the initial report these outcomes, which may be analogues of the human syndromes of hemorrhagic fever, encephalitis, and benign RVF, appeared to be determined by the rat genotype . This study, employing larger numbers of animals and serial measurements of virological events, demonstrated that the major distinction is between rat strains such as WF, which suffer an early predictable death with extensive virus replication, and other strains (MAXX, LEW) that control virus growth much more effectively . LEW rats, initially thought to be resistant to disease, developed late encephalitis in 16% of animals and the MAXX rats in 44% . In all 3 rat strains the major site of viral replication during the first 24 hours of infection appeared to be the liver . In the WF rat, viral titers were higher than those of LEW or MAXX rats in every sample from the earliest timepoints . Fluorescent antibody studies sharply contrasted the spreading infection in the highly permissive WF rat with the pattern seen in more restrictive genotypes . The in vitro studies of Rosebrock & Peters" demonstrated a pattern that may resemble the in vivo infection . LEW resident peritoneal macrophages were refractory to productive RVF infection at a multiplicity of inoculum that destroyed WF macrophages . However, it is not clear which cell type within the liver is the primary site of viral replication, although antigen-positive cells identified early after infection often resembled hepatocytes, as suggested by Mims for the highly susceptible mouse . 12 Classical genetic studies with WF and LEW rats have shown that the resistant phenotype is controlled by a single dominant locus or a closely linked gene complex, 10 a finding which has led to the derivation of a congenic resistant rat strain to provide cells for ongoing in vivo and in vitro studies to elucidate the relative importance of hepatocytes and Kupffer cells in resistance to RVFV . The importance of intrinsic resistance to RVFV replication and spread in determining the outcome of infection was further emphasised by experiments relating to the immune response. The appearance of serum antibody in LEW and MAXX rats as early as day 3, its close temporal relation to the termination of viremia, and the efficient high-titered in vitro activity suggest that this may be the major immunological defense mechanism operative in recovery from RVF . When formalin-inactivated RVFV was



Pathogenesis of Rift Valley fever virus

291

used as an immunogen to compare the immune response to non-replicating RVF in different rat strains, WF and LEW rats had similar serum neutralising antibody responses, while MAXX rats were significantly lower . To further study the role of the primary immune response in susceptibility to fatal RVF, we immunosuppressed LEW rats with cyclophosphamide . RVFV specific antibody was delayed, and fatal liver necrosis ensued . However, virus replication during the first few days after inoculation was still curtailed and death did not occur until 5-7 days, as compared to susceptible WF rats dying at 1-2 days postinoculation . Moribund cyclophosphamide-treated rats had much higher virus titers than the maximum achieved in unmanipulated animals, but were still 100- to 1000-fold less than naturally susceptible WF rats . The importance of antibody in recovery of the LEW rat was further demonstrated by passive transfer experiments . All cyclophosphamide immunosuppressed LEW rats injected with convalescent serum to mimic the neutralising antibody titers found in unmanipulated infected animals survived . Thus, RVFV infection in WF rats proceeds rapidly with extensive target organ damage before an effective immune response is possible . LEW and MAXX rats mount a timely antibody response which effectively controls the infection due to the intrinsically restricted rate of viral replication and spread . The occurrence of encephalitis typically in the second and third weeks of infection initially led us to speculate that an immunopathological mechanism might be involved . However, high brain virus titers and the multifocal necrotising lesions suggested a direct viral cytopathic effect . This posed an apparent paradox, since viremia was only detected during the first 40 hours of infection and serum neutralising antibody was readily measured from day 3 onward . Direct intracranial inoculation of small quantities of RVFV into LEW or MAXX rats leads to death from necrotising encephalitis within only 3-4 days . $ Since lymphoid tissues from LEW and MAXX rats had high viral titers and yielded infectious virus as late as 19 days postinfection, we speculate that infectious virus may enter the CNS in circulating lymphocytes or macrophages after the cessation of viremia, leading to a multifocal necrotising viral encephalitis . Further studies are needed on the ability of the immune response to protect against or modulate the development of encephalitis . For example, it is known that passively transferred antibody can protect the mouse against fatal liver necrosis, but permits the encephalitic process to proceed to death . 1314 Indeed, the immune response may determine which individual rats develop CNS disease . MAXX rats have twice the incidence of encephalitis compared to LEW rats, and their neutralising antibody response to inactivated virus or to infection is significantly lower . The efficiency of viral clearance from the circulation has been suggested to be an important determinant of the outcome of infection . For example, rapid removal of alphaviruses by hepatic reticuloendothelial cells predicts survival in different hostvirus interactions ." In the mouse, a host that succumbs to hepatitis, RVFV has been shown to circulate with a very long half life ." We found that RVFV was cleared fairly rapidly in both resistant and susceptible rats . The rat model has several advantages over the classical mouse model . The rat is the only species for which inbred strains exist which are either resistant or susceptible to RVFV-induced hepatocellular necrosis . The mechanism which limits early viral replication in the LEW rat is under the control of a single Mendelian dominant gene, therefore it has been possible to breed this gene into the genetic backgrond of the WF rat to produce a congenic rat strain (G . W . Anderson and C . J . Peters, manuscript in preparation) . The WF rat strain can also distinguish Egyptian from sub-Saharan strains of RVFV . 10 Finally, this model provides a framework to develop testable hypotheses concerning susceptibility to encephalitis in the same species .



292

G . W . Anderson et al.

Materials and methods Animals . Three inbred strains of Rattus norvegicus were used at 10 weeks of age . Female LEW/fMai, WF/fMai, and MAXX rats were obtained from Microbiological Associates, Inc ., Walkersville, MD, NIH rat and mouse ration (Ralston Purina Co ., St . Louis, MO) and water were provided ad libitum . All rats were inoculated subcutaneously with 5 .7 log o PFU RVFV,

unless noted otherwise in the text . Rats were observed daily for 30 days . Virus . All experiments were conducted with the ZH501 strain of RVFV originally isolated by Dr James Meegan (Naval Medical Research Unit No . 3, Cairo, Egypt) from a fatal human hemorrhagic fever case in Zagazig, Egypt, in 1977 . The virus was reisolated from human serum in DBS-103 fetal rhesus lung cells (FRhL) and passed a second time in these cells before use . The virus was stored in aliquots at -70°C in Eagles's minimal essential medium (EMEM) supplemented with 10% heat-inactivated (56°C for 30 minutes) fetal bovine serum, 200 units/ml penicillin, and 50 mg/ml streptomycin . Dilutions were prepared in Hank's balanced salt solution (HBSS) buffered to pH 7 .4 with HEPES containing 2% heat-inactivated fetal bovine serum plus antibiotics . All procedures were conducted by immunised personnel in a P3 containment laboratory . The RVFV, formalin-inactivated, cell-culture vaccine (NDBR 103) was made from the Entebbe strain of RVFV as previously described ." Plaque-reduction neutralisation assay . Serum-neutralising antibody titers were determined by using a modification of the method of Earley et al." Briefly, 4-fold dilutions of serum were made in 96-well plates in HESS with HEPES, 2% heat-inactivated bovine serum, and antibiotics . The dilutions were incubated for 1 hour at 37°C with 50 to 100 PFU of RVFV . Residual virus was quantitated by inoculating 50,ul of serial 4-fold dilutions onto duplicate 16-mm monolayer cultures of Vero cells in 24-well plastic plates . After 1 hour of adsorption, 0 .6 ml of an overlay consisting of 0 .5% agarose (SeaKem ME, Marine Colloids Div., Rockland, ME), Basal medium (Eagle) in Earle's balanced salt solution with 4% heat-inactivated fetal bovine serum and antibiotics was added to each well . The plates were held at 37°C in a 5% CO 2 humidified

incubator . A second overlay prepared as above, but containing 0 .1 mg/ml neutral red (Grand Island Biological Company, Grand Island, NY) was added on day 3 . Plaques were enumerated the following day . The plaque-reduction neutralisation titer (PRN,,) was determined as the highest dilution of serum reducing 80% of the input virus . Infectious virus assay. Ten-fold serial dilutions were assayed by inoculating duplicate 16mm Vero cell monolayers with 50 yl of sample diluted in HBSS with HEPES and containing 2% heat-inactivated bovine serum . The plaques were enumerated as in the plaque-reduction neutralisation assay. Immunofluorescence method . Tissues were embedded in Tissue-Tek II O .C .T. compound (Lab-Tek, Naperville, IL) and frozen at -70°C . Sections were cut 4-6 lcm thick (Tissue-Tek II cryostat) and fixed in acetone at -70°C overnight . Fluorescein-conjugated sheep antiserum to RVFV was used to detect viral antigen by the direct immunofluorescence antibody technique . Slides were incubated in a humidified chamber at 23°C for 30 minutes . The sections were washed 3 times in phosphate buffered saline, pH 7 .4 . Sections were examined and photographed with a Zeiss fluorescence microscope with epi-illumination, high-pressure xenon lamp, and the FITC blue interference combination 455-490 filter . Cyclophosphamide treatment . Cyclophosphamide (Mead Johnson, Evansville, IN), was administered intraperitoneally, 50 mg/kg, 1 day prior to subcutaneous inoculation of RVFV . Immune sera . LEW rats were inoculated subcutaneously with 5 .7 log o PFU of RVFV (ZH501

strain) . Thirty days postinfection, these rats were rechallenged with the same dose of RVFV . Seven days later the rats were exsanguinated and the sera pooled . This material was used in the serum therapy studies . Necropsy and specimen collection . Rats were killed with CO 2, and gross pathological changes

were noted . Tissues (liver, brain, spleen, kidney, adrenal, lung) were taken and divided for detection of infectious virus, viral antigen, and histological changes . Samples for infectious virus and viral antigen were immediately frozen and held at -70°C until assayed . Tissues for



Pathogenesis of Rift Valley fever virus

293

histological examination were fixed in 10% neutral buffered formalin . Paraffin sections were prepared and stained with hematoxylin and eosin (H&E) . Statistical analysis . The Student's t-test and Chi-square test were used to analyse the differences among groups . Differences that yielded probability values <0 .05 were considered statistically significant .

We thank Robert C . Bishoff for expert technical assistance .

References 1 . Peters CJ, Meegan J . Rift Valley fever . In : Beran GW ed . CRC Handbook series in zoonoses, Sect . B, Viral zoonoses, Vol . 1 . Boca Raton : CRC Press, 1981 . 2 . Van Velden DJJ, Meyer JD, Olivier J, Gear JHS, McIntosh B . Rift Valley fever affecting humans in South Africa : a clinicopathological study . S Afr Med J 1977 ; 51 : 867-71 . 3 . Imam IZE, Darwish MA . A preliminary report on an epidemic of Rift Valley fever (RVF) in Egypt . J Egypt Pub Health Assoc 1977 ; 52 : 417-8 . 4 . Laughlin LW, Meegan JM, Strausbaugh LJ, Morens DM, Watten RH . Epidemic Rift Valley fever in Egypt : Observations of the spectrum of human illness . Trans R Soc Trop Med Hyg 1979 ; 73 : 630-3 . 5 . Maar SA, Swanepoel, R, Gelfand M . Rift Valley fever encephalitis . A description of a case . Cent Afr J Med 1979; 25 : 8-11 . 6 . Swanepoel R, Manning B, Watt JA . Fatal Rift Valley fever of man in Rhodesia . Cent Afr J Med 1979 ; 25 : 1-8 . 7 . Shope RE, Peters CJ, Davies FG . The spread of Rift Valley fever and approaches to its control . Bull WHO 1982 ; 60 :299-304 . 8 . Peters CJ, Slone TW . Inbred rat strains mimic the desperate human response to Rift Valley fever virus infection . J Med Virol 1982 ; 10 : 45-54. 9 . Bang FB . Genetics of resistance of animals to viruses : I . Introduction and studies in mice . Adv Virus Res 1978; 23 : 270-348. 10 . Peters CJ, Anderson Jr GW . Pathogenesis of Rift Valley fever. In : Swartz TA, Goldblum N eds . Contributions to Epidemiology and Biostatistics, New York : Karger, 1981 ; 121-41 . 11 . Rosebrock JA, Peters CJ . Cellular resistance to Rift Valley fever virus (RVFV) infection in cultured macrophages and fibroblasts from genetically resistant and susceptible rats . In vitro 1982 : 18 : 308 . 12 . Mims CA . Rift Valley fever virus in mice . I . General features of the infections . Br J Exp Path 1956 ; 37 : 99-109 . 13 . Bennett DG, G,lock RD, Gerone PJ . Protection of mice and lambs against pantropic Rift Valley fever virus using immune serum . Am J Vet Res 1965 ; 26 : 57-62 . 14 . Peters CJ, Reynolds JA, Slone TW, Jones DE, Stephen EL . Prophylaxis of Rift Valley fever with antiviral drugs, immune serum, an interferon inducer, and a macrophage activator . Antiviral Res 1986 ; 6 : 28597 . 15 . Jahrling PB, Gorelkin L . Selective clearance of a benign clone of Venezuelan Equine Encephalitis virus from hamster plasma by hepatic reticuloendothelial cells . J Infect Dis 1975 ; 132 : 667-76 . 16 . Mims CA . Rift Valley fever virus in mice . II . Adsorption and multiplication of virus . Brit J Exp Path 1956; 37 : 110-119, 17 . Randall R, Binn LN, Harrison VR . Immunization' against Rift Valley fever virus . Studies on the immunogenicity of lyophilized formalin-inactivated vaccine . J Immunol 1964 ; 93 : 292-9 . 18 . Earley E, Peralta PH, Johnson KM . A plaque neutralization method for arboviruses . Proc Soc Exp Biol Med 1967 ; 125 : 741-7 .