Hantavirus infections in humans and commensal rodents in Singapore

248 TRANSA~~ONS OPTHEROYALb

OPTROPICAL. MEDICINE ANDHYGIENE (1989)

Hantavirus

in humans

infections

and commensal

83, 248-251

rodents

in Singapore

Tze Wai Wonf, Yow Cheong Char?, Yong Gyu Joo3, Ho Wang Lee’, Pyung Woo Lee3 and Richard Yanagihara4 Department of Community, Occupational and Family Medicine, National University of Singa@re; ZDepartment of Microbiology, National University of Singapore, Singapore; 3WH0 Collaborating Laboratory for Virus Refeence and Research(HFRS), Korea University Medical College, Korea; 4Laboratory of Central Nervous System Studies, National Institute of Neurological and Communicative Disorders and Stroke, National Institutes of Health, Bethesda, Maryland, USA Abstract To determine the extent of hantavirus infection in Singapore, serological studies using the indirect immunofluorescent antibody (IFA) test were conducted on commensalrodents and on human patients in 4 diagnostic groups. Virus isolation using a Vero E6 cell line was performed on hantaviral antigen-positive rodent lung iissue. Of 142 rodents and 3 msectivores studied. 37 (26%) were serooositive for IFA. Rattus ncwegi& was thepredomin&t speciescaptured, with the highest species-specific seropositive rate of 32% (36 of 113). A hantavirus strain, R36, was isolated from one R. ncrvegicus.Seropositive rates for human patients were: 8% respectively for dengue haemorrhagic fever suspects &id for -non-A non-B hepatitis natients. 3% for lentosnirosis susnects and 2% for acute nephritis patients.* 2 patients had marked liver dysfunction but mild renal involvement. This hepatitis-like manifestation appearsto be a clinical variant of hantavirus infection. hroduction Haemorrhagic fever with renal syndrome (HFRS) is an endemic disease in China, Korea, Japan, the Soviet Union, Scandinavia and eastern Europe (WHO, 1983). The aetiologic agents are Hantaan and related viruses, namely Seoul, Puumala and Prospect Hill viruses. Hantaan virus, first isolated by LEE et al. (1978) from the lung tissue of the striped field mouse (Apodemusagrarius), is the prototype of the newly defined genus Hantavirus (SCHMALJOHN et al., 1985). Other reservoirs for hantaviruses are the bank vole (Clethrionomysglareolus), the meadow vole (Microtus pennsylvanicus) (YANAGIHARAet al., 1984) and the commensal rat (Rattus species) (LEE et al., 1982). Serological evidence of hantavirus infection in humans in the absenceof clinical diseasehas been found in various parts of the world, including the United States, Canada, South America and Africa (LE DUCet al., 1985, 1986; LEE et al., 1984; GONZALEZet al., 1984). Smce commensal rodents are ubiquitous m distribution, sero-epixootiological studies would yield useful information on the distribution of hantavirus infection in nature. Singapore is a cosmopolitan city situated on the southern tip of the Malayan peninsula. A major seaport in south-eastAsia, Singapore has a population Correspondenceto Dr Tze Wai Wang, Departmentof Community, Occupationaland Family Medicine, National University of Singapore,Lower Kent Ridge Road, Singapore 0511.

of 2.6 million people living in an admixture of old and new residential areas which is conducive to a large rodent populations. The first case of haemorrhagic fever with renal syndrome was reported in 1985 (WONGet al., 1985). However, the extent of unreported hantaviral infection in the community has not been established. To determine the presence and extent of hantavirus infection in Singapore, we initiated a sero-epidemiological survey of humans and commensal rodents. Materials and Metbods Serological survey of commensalrodents Commensal rodents were live-trapped in 7 selected areas in Singapore representing different ecological environments, namely old residential areas (Bugis Street and Chinatown), residential and commercial areas along waterways (Singapore River and Rochor Canal), an industrial town, a public housing estate, and the harbour. From August 1985to February 1986 142 rodents and 3 insectivores were caught. The species, sex and trap site were recorded. Blood was drawn by cardiac puncture and allowed to clot at room temperature. Sera, obtained by centrifugation in a Beckman microfuge at 6000 rpm for 1 minute, were stored at -20” for antibody studies. Lung tissues of all rodents were removed aseptically and stored at -70” until tested. Serosutvey in human patients Four groups of patients were studied. From January 1985 to June 1986 824 unpaired sera were collected from the Department of Pathology, the central laboratory serving all government hospitals and clinics in Singapore which also provides services to private hospitals and clinics. These included 312 serafrom patients with acute nephritis, 301 serafrom patients with suspected leptospirosis and 217 sera from patients w&non-A non-B-hepatitis. In addition 21 sera from 12 natients (9 natients had &red serum samples) with suspected’d&ue haemo-rrhagicfever were collected from the Department of Microbiology, National Universitv of Sinaawre. Patients with acute nephritis had a negative- kti-streptolysin 0 titre (ASOT). Lentosnirosis susnects were those oatients &th chnicalZfeatures suggestive of leptospirosis but negative ‘sensitized erythrocyte lysis’ tests for leptos~ira. Henatitis non-A non-B natients were diaanosed by exchkion; that is, they presented clinical& with henatitis but their sera lacked IeM antibodies anainst hepatitis A virus and hepatitis B surface antigeL and also antibody. Dengue haemorrhagic fever suspects

249 Rodents with high IFA titres (reciprocal titre 2 256) were found in only 3 areas, namely Bugis Street, Singapore River and Rochor Canal. Rodents from the ha&our were seronegative. Of the 37 seropositive rats only one had hantaviral antigen in the lung tissue demonstrable by IFA technique. A hantavirus strain, designated R36, was isolated from the lung tissue of this rat.

were those with fever and haemorrhagic manifestation whose sera were negative for haemagglutinationinhibition antibodies against the endemic type 1 and 2 dengue virus. All set-awere stored at -20°C until testing. methods Rodent and human sera were screened by the indirect immunofluorescent antibody (IFA) test, using Vero E6 cells infected with prototype Hantaan virus, strain 76118, beginning at a dilution of 1:16 for rodent sera and 1:32 for human sera. Fluorescein isothiocyanate-conjugated goat anti-rat and anti-human IgG were used in the rodent and human serosurveys respectively. Positive sera were titrated. The plaque-reduction neutralization (PRN) test, using Hantaan virus, strain 76118, as described by TAKENAKA et al. (1985), was done on all positive human sera. Frozen sections of lung tissues of seropositive rodents were examined for hantaviral antigens by the IFA method using positive serum from an infected rat and FITC-labeled goat anti-rat NJ.

Laboratory

Serosuwg of patients Among the 4 groups of patients, dengue haemorrhagic fever (DHF) suspectshad the highest seropositive rate of 8*3%, followed closely by hepatitis non-A non-B patients with a seropositive rate of 8.1% (Table 2). The leptospirosis suspects and acute nephritis patients had seropositive rates of 266% and 1.92% respectively. Based on their antibody titre, 2 groups of seropositive patients could be discerned: 29 patients had reciprocal titres s 128 and 3 had titres > 512. Of the three subjects with a high reciprocal IFA titre (2 512), one (Table 3, patient A) presented with fever and renal failure but no generalized haemorrhagic manifestation except microscopic haematuria. The other 2 patients (B and C) had liver dysfunction but only mild renal involvement. Patient B had a severe illness with shock, generalized bleeding tendency, a prolonged prothrombin time and evidence of disseminated intravascular coagulation. Serum transminases, alkaline phosphatase and lactic dehydrogenase were grossly elevated. The only evidence of renal involvement was mild proteinuria and elevation of serum urea levels. This patient worked in Malaysia and contracted the diseasebefore coming to Singapore for treatment. Patient C likewise had a hepatitis-like presentation but no generalized bleeding tendency. All 3 patients recoveredwith supportive treatment. A retrospective study of paired seraof patient B showed a more than fourfold rise in IFA. The 3 patients’ sera were also tested for the presence of neutralizing antibodies to Hantaan virus by the PRN test. Only patient B had neutral&g antibodies against Hantaan virus, the reciprocal PRN titre being 2000.

Results

Serosuwg on comtnensalrodents 142 rodents (113 Rattus twwegicus, 21 R. rams, 5 R. exulans and 3 Mus musculus)and 3 insectivores (Suncu.rmurinus) were trapped. The reciprocal immunofluorescent antibody (IFA) titres to Hantaan virus according to species and sex are shown in Table 1. The overall seropositive rate was 255% (37 of 145), using a reciprocal titre 5 16 as positive. The species-specificseropositive rates for R . twwegicus and R. rams were 31.9% (36 of 113) and 4.8% (1 of 21), respectively. None of the other specieswas seropositive. Of the 37 seropositive rodents, 21 had reciprocal titres 3 256, and 9 rodents (all R. twwegicus) had titres > 1024. The reciprocal geometric mean titre was 100. Seropositive rodents were found in 5 of the 7 areas surveyed. Of the 37 seropositive rodents, 24 (64.9%) were trapped from Bugis Street (an old residential area) while 8 (21.6%) were from Singapore River. Table

1. Reciprocal

Reciprocal IFA titre

indiict

fluorescent

antibody

titre to Hantaan

virus by species of host

R.

ninvegicus

R. rattus

R. exulans

M. musculus

S. murinus

All species

<16 :2”

77 12

20 -1

5 -

3 -

3 -

108 13

1;;

:

-

-

-

-

:

256 512 1024 2048

--i

-

-

-

-

:

-

-

-

37

3

145

-

25.5

i 3

Total no. of seropositive rodents’ Total no. of rodents tested

36

1

113

21

Seropositive rate (%)

31.9

4.8

’ A reciprocal IFA titre of 216 was considered positive.

5 -

3 -

:, 3

250 Table 2. Seropositive rates in four groups of patients by indirect fluorescent Patient group

antibody

test

No. of sera

No. of subjects

No. of positive sera’

Acute nephritis

312

312

6

32-

512

1.92

Leptospirosis suspect

301

301

8

32 - 1024

2.66

12

1

64

8.33

211

17

64-

Dengue baemorrhagic fever suspect

21b

Hepatitis non-A non-B

211

Reciprocal IFA titre (range)

Seropositive rate (Oh)

512

8.06

a A reciprocal IFA titre of 3 32 was considered positive. b 9 patients had paired sera.

Table 3. Reciprocal neutralization (PRN)

antibody titres to Hantaan virus by indirect tests in three patients with hi IFA titres

fluorescent

antibody

(IFA)

and plaque-reduction

Reciprocal antibody titres fo Hantaan virus (76/118) IFA

A

F/36

Diagnostic group Acute nephritis

B

M/39

Leptospirosis suspect

Patients

Sex/age

PRN

Acute

Convalescent

Convalescent

NA’

512

Negative

1024

2000

x3-a

128b

sera

sera

512 Hepatitis non-A non-B NA Negative MM0 C ’ NA = not available. b Paired sera of patient B were obtained retrospectively by tracing the unused serafor other diagnostic tests in the Department of Patbology.

Discussion

Since the isolation of Hantaan virus by LEE et al. in 1976 and the recognition of its aetiologic role in HFRS (LEE et al., 1978), there has been rapid progress in the understanding of the epidemiology of this diseaseas well as the virology of the hantaviruses. Outside the traditional endemic areas, hantaviral infections in rodents and humans have been detected in the Americas, western and eastern Europe, southeastAsia, parts of Africa and Australia (LE DUCet al., 1986). We have demonstrated that Singapore does harbour infected rodents, R. norvegizus being the predominant reservoir. This finding is similar to those in many other cities, such as Hong Kong (SHORTRIDGEet al., 1987), Penang (Malaysia) (LIM et al., 1985>,Belkm (Brazil) (LE Due et al., 1985), Buenos Aires (Argentina) (LE DUC et al., 1986) and Baltimore, New Orleans and several other cities in the United States(TSAI er al., 1985). The infection rate in commensal rats was 25*5%, which is similar to those

found in other Asian cities-31% in Hong Kong, 32% in Manila, 22% in Rangoon and 22% in Taiwan. In Burma, Bandicota bengalen& was found to be an important reservoir, whereas in Thailand, R. indica was the predominant species infected with hantavirus.

S. murintu, the musk shrew, and Felis cati, the domestic cat, have been shown to harbour hantavirus in China (TANG et al., 1985). Although their role in causing haman diseasehas not been clearly defined, the possibility that domestic cats are a secondary reservoir cannot be ignored.

In the present study, the infection rate for R. nomegicuswas 31.9% and that for R. rattus was 4.8%. There was no evidence of infection among the other species (R. exulans, M. musculus and S. murinus). However, their numbers were too small for us to conclude that they play no part in the epidemiology of hantaviral infection in the local habitat. In our serosurvey among patients, we have not attempted to study the history of exposure to rodents. It is reasonableto assumethat local peridomestic rodents are the reservoirs of human diseasebecauseof their close association with the human habitat. Moreover, the other known reservoirs for HFRS, namely, Apodemussp. and Clethrionomyssp., are absent from Singapore. Epidemiological and experimental studies elsewhere have shown that the mode of transmission from rodents to human is via infectious aerosols (LEE et al., 1981; YANAGIHARA et al., 1985). Infection among local rodents could have been

introduced via the seaport from endemic countries or on land from Malaysia. Wherever the original source of infection may be, it appears that hantaviral infections have been enzootic in Singapore for quite some time? as evidenced by the widespread geographical &striburion of seropositive rodents in the various areassurveyed. The successful isolation of a

hantavirus strain (R36) from the lung tissue of R. norvegicusgives further evidence that the latter plays an important role asa reservoir of hantaviruses. LEE et al. classified hantaviruses into 4 serotypes according to the rodent reservoir (LEE et al., 1985). The

251 Singapore isolate falls within the Rorrus-derived serotype. Its morphological and serological characteristics are being studied using monoclonal antibodies. The serosurvey among patients for evidence of hantaviral infection showed that it accounted for only a small proportion of acute nephritis and suspected leptospirosis in Singapore, seropositive rates for the two groups being 2.0% and 2.7% respectively. The

corresponding rates for dengue haemorrhagic fever (DHF) susnects and heuatitis non-A non-B patients were &out~l%. However, the diagnostic sigdifcance of low IFA titres (< 128) is doubtful, becauseof the

lack of specificity of the test. If such sera are excluded, the seropositive rates would be much lower in all the patient groups. In a separate, community

based serosurvev. 200 healthv individuals aaed 12 to 65 years were-randomly selected and t&ted for hantaviral antibodies by the IFA test. All were seronegative at 1 in 32 (Wong et al., unpublished

observations). Two of the 3 patients with high IFA titres had evidence of gross liver dysfunction and mild renal involvement. This hepatitis-like picture has not been reported previously in patients in China and Korea. We consider this as a clinical variant from the classical HFIU of eastern Asia and nephropathia epidemica of Scandinavia. Only one of the 3 patients (who contracted the disease in Malaysia) had neutralizing antibodies to Hantaan virus and a more than fourfold rise in IFA. A full account of his clinical illness has been published elsewhere (CHANet al., 1987). For the other 2 patients, diagnosis was based on a suggestive clinical picture, a high IFA titre, and the absenceof evidence of infection by other agents which might have accounted for their illnesses. It is possible that these 2 patients had been infected by a local heterologous strain, antibodies to which do not cross-react with Hantaan virus by the PRN test. In summary, we have demonstrated that hantavirus infection is endemic in commensal rodents in Singapore. We have also demonstrated a low infection rate in human patients and a variant, hepatitis-like syndrome with only mild renal involvement. With the isolation of a local hantavirus strain from R . mvegicus further development of diagnostic tests for seroepidemiological and virological studies among patients becomes possible, leading eventually to a better understanding of this disease. References

Chan, Y.C., Wong, T. W., Tan, H.C., Lee, H. W., Chu, Y. K. & Lee, P. W. (1987). Haemorrhagic fever with renal syndrome affecting the liver. Medical Journal of Auszralia, 147, 248-249. Gonzalez, J. P., McCormick, J. B., Baudon, J. B., Gautun, J. P., Meunier, D. Y., Doumon, E. & Georges, A. J. (1984). Serological evidence of Hantaan-related virus in Africa. Lance?, ii, 1036-1037. Le Due, J. W., Smith, G. A., Pinheiro, F. P., Vasconcelos, P. F. C., Rosa, E. S. T. & Maixtegui, J. I. (1985). Isolation of a Hantaan-related virus from Brazilian rats and serologic evidence of its widespread distribution in South America. Americun 3ournal of Tropical Medicine and Hygiene, 34, 810-815. Le Due, J. W., Smith, G. A., Childs, J. E., Pinheiro, F. P., Maiztegui, J. I., Niklasson, B., Antomades, A., Robinson, D. M., Khin, M., Shortridge, K. F., Wooster, M. T., Elwell, M. R., Ilbery, P. L. T., Koech, D., Rosa, E.

S. T. & Rosen, L. (1986). Global survey of antibody to Hantaan-related viruses among peridomestic rodents. Bulletin of the WorU Health Organization, 64, 139-144. Lee, H. W., Lee, P. W. & Johnson, K. M. (1978). Isolation of the etiologic agent of Korean haemorrhagic fever. Jourual of Znfectious Diseases, 137, 298-308. Lee, H. W., Lee, P. W., Baek, L. Jr, Song, C. K., Seong, I. W. (1981). Intraspecific transtmssion of Hantaan virus., fever, in the the etiologic agent of Korean haemorrhagic rodent Apodemus agrarius. American Joumal of Tropical Medicine and Hygiene, 30, 1106-1112. Lee, H. W., Baek, L. J. & Johnson, K. M. (1982). Isolation of Hantaan virus, the etiologic agent of Korean haemorrhagic fever, from wild urban rats. Journal of Znfectious Diseases,

146, 638-644.

Lee, H. W., Seong, I. W., Baek, L. J., McLeod, D. A., Seo, J. S., Kang, C. Y. (1984). Positive serological evidence that Hantaan virus, the etiologic agent of haemorrhagic fever with renal syndrome, is endemic in Canada. Canadian Journal of Microbiology, 30, 11371140. Lee, P. W., Gibbs, C. J., Gajdusek, D. C. & Yanagihara, R. (1985). Serotypic classification of hantaviruses by indirect immunoihtorescent antibody and plaque reduction neutralixation tests. Journal of Clinical Microbiology, 22, 940-u Lim, T. W., Ambu, S., Baek, L. J., Ju, Y. K., Lee, H. W. 8t Ng, C. S. (1985). Investigation on the presence of Hantaan and/or related virus, etiologic agent of haemorrhagic fever with renal syndrome (HFRS) among rodents in the seaport of Penangand Perlis. Tropical Biomedicine, 2, 73-79.

Schmaljohn, C. S., Hasty, S. E., Dalrymple, J. M., LeDuc, J. W., Lee? H. W., Bomsdorff, C. M., BrummerKorvenkonuo, M., Vaheri, A., Tsai, T. F., Regnery, H. L., Goldgaber, D. & Lee, P. W. (1985). Antigenic and genetic properties of viruses linked to baemorrbagic fever with renal syndrome. Science, 227, 1041-1044. Shortridge, K. F., Lee, H. W., LeDuc, J. W., Wong, T. W., Chau, G. W. & Rosen, L. (1987). Serological evidence of Hantaan-related viruses in Homr Konn. Transactious of the Royal Societyof Tropical Medicine and Hygiene, 81, 4oo-402. Take&a, A., Gibbs, C. J. & Gajdusek, D. C. (1985). Antiviral neutralixina antibody to Hantaan virus as determined by plaque reduction technique. Archives of Virolow. 84. 197-206. Tang, T.V., I&, Z. Y., & Tsai, T. F. (1985). Isolation of HFRS virus from Suncusnmrinus,an insectivore. Lancer,

i, 513-514.

Tsai, T. F., Bauer, S. P., Sasso,D. R., Whitfield, S. G., McCormick, J. B., Caraway, T. C., McFarland, L., Bradford, H. & Kurata, T. (1985). Serological and virological evidence of a Hantaan virus-related enxootic in the United States.Journal of Infectious Diseases, 152, 126136. WHO (1983). Haemorrhagicfever with renal syndrome: memorandum from a WHO meeting. Bulletin of rhe World Health Organization, 61, 269-275. Wong, T. W., Chan, Y. C. & Lee, H. W. (1985). Haemorrhagic fever with renal syndrome in Singapore. Southeast Asian 3ournal of Clinical Medicine and Public Health, 16, 525-527.

Yanagihara, R., Svedrnyr, A., Amyx, H. L., Lee, P. W., Goldgaber, D., Gajdusek, D. C., Gibbs, C. J. & Nystrom, K. (1984). Isolation and propagation of nephropathia epidemica virus in bank voles. Scandinavian 3oumal of Znfectious Diseases, 16, 225-228.

Yanagihara, R., Amyx, H. L. & Gajdusek, D. C. (1985;. Experimental infection with Puumala virus, the etiologic agent of nephropathia epidemica, in bank voles (Clethrionomys ghzreolus). Journal of Virology, 55, 34-38. Received accepted

27 May 1988; revised 23 August publication 25 August 1988

for

1988;