Tula Virus in Populations of Small Terrestrial Mammals in a Rural Landscape

Abstract

Over 5 years (2000–2004), populations of small mammals from a rural landscape in southern Moravia (Czech Republic) were investigated for the presence of Tula virus (TULV) antigen using the ELISA set Hantagnost. In total, 1566 individuals from 10 species were examined. The prevalence in the common vole (Microtus arvalis Pallas 1778), the main reservoir of TULV, was 10% (n = 871). The prevalence of TULV antigen increases with its population numbers. The highest number of TULV antigen-positive common voles was found in set-aside plots and winter crops, such as rape and winter wheat. All these habitats are important for common vole overwintering. Older and heavier individuals were more often hantavirus antigen positive. From the other small mammal species, 186 pygmy field mice (Apodemus uralensis Pallas, 1811) were examined, of which 3 were positive, which represents the first hantavirus antigen positive record for this species, and of 195 wood mice (Apodemus sylvaticus Linnaeus, 1758) only 1 was positive. The remaining five rodent species (Apodemus flavicollis Melchior, 1834, Mus musculus Linnaeus, 1758, Micromys minutus Pallas, 1771, Myodes glareolus Schreber, 1780, Microtus subterraneus de Sélys-Longchamps, 1836) and two Soricomorpha (Sorex araneus Linnaeus, 1758, Sorex minutus Linnaeus, 1766) were hantavirus antigen negative.

Introduction

H antavirus infections are important “emerging” infections in which rodents are the main host organisms. Hantavirus serotypes are known to be rodent specific and can spread to humans in a dried and/or aerosolized form from urine, feces, and saliva by inhalation or even through bites (Mills and Childs 2001, Zeier et al. 2005). Although usually inducing nephropatia or influenza-like syndromes in humans in Europe, infection occurs in rodents without any obvious syndromes (Pejčoch et al. 2003b). Earlier surveys in the Czech Republic have suggested that among rodents it is mainly the common vole (Microtus arvalis) that is susceptible to the western type of hantavirus, Tula virus (TULV), and is the main reservoir known for transmission of this hantavirus species (Daneš et al. 1986, 1991, Pejčoch et al. 1992, 2003a, Zeier et al. 2005). The genospecies TULV was described by Plyusin et al. (1995) in voles Mi. arvalis and Microtus levis caught in 1987 in Tula. The first successful isolation of the virus in central Europe came from a population in southern Moravia where the same antigen was also screened in a human (Plyusnin et al. 1995, Vapalahti et al. 1996). However, the presence of common vole TULV antigen was later also documented in other rodent species, such as the yellow-necked mouse (Apodemus flavicollis), wood mouse (Apodemus sylvaticus), bank vole (Myodes glareolus), and European pine vole (Microtus subterraneus) (Pejčoch et al. 1992, 2003a). Although the basic host–virus relationships have already been described (Vapalahti et al. 1996), detailed information on the prevalence of TULV in rodent populations in relation to their demography and ecology is scarce.

Here we present the results from the monitoring of the TULV antigen prevalence in small rodent populations on a long-term scale, with a particular emphasis on common vole populations as the main reservoir of the virus. In this species we evaluate hantavirus prevalence in relation to its demographic structure (body mass, the sex), population dynamics (in years and seasons), and habitat type. We underline the importance of agroecosystems, where set-aside plots and some particular crop provide optimal habitat for the common vole. Hence, detailed information on TULV antigen prevalence in small mammals relative to season and crop is of great epidemiological importance to farmers.

Material and Methods

Analysis of rodents

The research was carried out in an agricultural landscape in southern Moravia, Czech Republic, between March 2000 and December 2004. The rodents were sampled by snap-trapping in lines in fields of various crops and in small woods. Sampling was done monthly on the same plots in set-aside fields (3 plots in each of 2 fields) and also regularly in crop fields (122 plots) following its phenological stage. In woodland (3 woods, 15 plots), trapping was done five times a year. In total, 4620 small mammals were caught during 33,100 trapnights. Traps were inspected in the morning [for details on trapping methods, see Suchomel and Heroldova (2004) and Heroldová et al. (2005)]. The trapped mammals were processed on the same day in a laboratory to determine species, sex, reproductive condition, and body dimensions. All individuals of the common vole were examined for TULV antigen except 429 voles whose carcasses were damaged by cannibalism or partially consumed by other animals. The other species were examined selectively by picking up a random subset of individuals from the sample. In total, 695 mostly adult animals out of 3320 individuals were screened for TULV antigen. For scientific names we followed Wilson and Reeder (2005) classification.

We measured abundance of common vole populations twice a year using a population index based on the counts of active (food, fresh feces near burrow entrances, vegetation eaten around the burrow, burrowing activity, smooth burrow opening) burrow entrances per hectare. This index is routinely used by the State Phytosanitary Administration in the Czech Republic to monitor vole population numbers as part of management program for voles (for detailed description of the method, see Lisická et al. 2007).

TULV detection

The samples of lung collected at dissection were immediately frozen and stored at −20°C for further analysis. Testing was done monthly. In few cases, where the interval was longer, the samples were stored at −70°C. Each sample representing the whole left lung was subsequently separately homogenized in a friction dish with 0.005 L of physiological solution. TULV antigen was detected by ELISA Hantagnost sets (Poliomyelitis Institute, Moscow, Russia) as described by Tkachenko et al. (1981). These detection sets contained a peroxidase conjugate based on IgG antibody against Puumala virus from convalescent human serum (Groen et al. 1991). The results were analyzed on the basis of optical densities by comparing the sample of lung homogenate with a specific antibody and human serum without antibody as a negative control. The sample was considered positive if the optical density at a wavelength of 492 nm was higher than 2.1 (e.g., Escutenaire et al. 2000). The test was repeated in each positive sample. The positive tests were verified in three samples from southern Moravia common voles by sequencing of L segment, 355 bp (done by L. Ragosa of Comenius University, Bratislava, Slovakia, and B. Klempa, Institute of Virology, SAS, Bratislava, Slovakia).

Data analysis

The presence of positive TULV antigen in the common vole was analyzed by fitting generalized linear models in relation to the following factors: body mass, sex, reproductive condition, year, season, type of crop (annual, perennial, forest), and relative population density as measured by vole index. We further compared the prevalence of TULV antigen between woods and open habitats.

Results

A total of 1566 specimens of 10 mammal species were screened for the presence of the TULV antigen during the period 2000–2004 (Table 1). Of these, 89 were found to be positive for TULV antigen: 85 were common voles, 1 was a wood mouse (a gravid female), and 3 were pygmy field mice (all adult males). The positivity of the latter represents the first record of the TULV antigen ever documented for the pygmy field mouse. The most numerous species in the entire sample was the common vole (871 individuals). The prevalences of TULV antigen in males and females were 36 (12%) and 49 (9%) individuals, respectively, with no significant differences between the sexes (χ ² = 1.83, p = 0.17). Of the positive females, 86% were adults, of which 35% were pregnant and 14% subadults. Among the positive males, 61% were adults, 36% subadults, and 3% juveniles. Reproductive condition was not a significant factor in TULV hantavirus positivity (χ ² = 0.61, p = 0.43). Heavier individuals were more often positive (χ ² = 32.15, p < 0.01; Table 2). There was a significant effect of season (χ ² = 11.08, p < 0.01), as TULV antigen-positive individuals were more frequent in early spring (overwintering individuals; Figs. 1 and 2). From a total sample of 1185 individuals collected in open habitats, 86 (7.3%) were positive for TULV antigen, whereas it was only 3 (0.8%) out of 381 individuals of mammals collected in woodland that showed a positive response (χ ² = 31.5, p < 0.001; Table 1). Moreover, these three positive individuals in woods were the common voles. Significantly more TULV-positive individuals were found in annual and biannual crops compared with perennial crops and forest habitats (χ ² = 9.87, p < 0.01; Table 2). However, there were no differences among annual crops (rape, wheat, barley; χ ² = 1.59, p > 0.1). Set-aside plots in the crop-field landscape seemed to be important for TULV prevalence. During winter the common vole tends to concentrate in fallow areas and on plots of overwintering crops, such as rape and wheat (Fig. 1). TULV prevalence increases with the increasing population numbers (χ ² = 4.54, p = 0.033; Fig. 3).

FIG. 1.

Influence of crops and month of the year on TULV positive prevalence in the common vole (in percent of TULV antigen-positive voles). Prevalence shown for each month represents the cumulative totals during the 5-year period. TULV, Tula virus.

FIG. 2.

The year chronology (in months) of the TULV positive prevalence in common vole (in percent of TULV antigen-positive voles). Prevalence shown for each month represents the cumulative totals during the 5-year period.

FIG. 3.

Relation between TULV prevalence in common vole in the years 2000–2004 (in percent of TULV antigen-positive voles) compared with common vole densities expressed by index (burrow entrances/ha).

Table 1.

Summary of Small Terrestrial Mammal Materials Examined for Tula Virus Antigen Positivity

	Woods		Fields
Species	Examined (n)	Positive (n)/%	Examined (n)	Positive (n)/%
Apodemus flavicollis	189	0	18	0
Apodemus sylvaticus	32	0	163	1/0.6
Apodemus uralensis	2	0	184	3/1.6
Mus musculus	0	0	3	0
Micromys minutus	0	0	1	0
Myodes glareolus	94	0	0	0
Microtus arvalis	63	3/5	808	82/10
Microtus subterraneus	1	0	0	0
Sorex araneus	0	0	5	0
Sorex minutus	0	0	3	0

Table 2.

Prevalence of Tula Virus Antigen in Common Voles in Relation to Sex and Weight Categories (133 Pregnant Females Were Excluded), and the Type of Habitat/Crop

		Examined, n (% of positive) based on weight categories
Sex	Type of habitat	To 10 g	10.5–20 g	20.5–30 g	Over 30 g
Female	Annual	3 (0)	13 (0)	27 (28.6)	9 (22.2)
	Perennial	5 (0)	154 (2.6)	132 (9.8)	62 (9.6)
	Forest	0	18 (5.6)	7 (14.3)	1 (0)
Male	Annual	1 (0)	15 (13.3)	31 (25.8)	21 (61.9)
	Perennial	9 (0)	141 (2.1)	37 (10.8)	17 (35.3)
	Forest	0	18 (0)	16 (6.3)	1 (0)

Discussion

Although hantavirus-positive rodents pose an important epidemiological risk to farmers, detailed information on their prevalence relative to rodent demography, season, and crop is scarce. In this study, three dominant species of rodents in central European agroecosystems were shown to be TULV antigen positive.

The positivity in common voles and wood mice has already been demonstrated (Pejčoch et al. 2003a), but that in the pygmy field mouse represents the first record of the TULV antigen in this species. However, these findings still need to be confirmed genetically. Pygmy field mice often use the burrows of other rodents, particularly the common vole (personal observations), and together with the vole, they were found at high densities in set-aside. Close contact between species may facilitate the spread of the virus. One individual of A. sylvaticus was also TULV antigen positive. Prevalence of TULV in Apodemus sp. was not realized and confirmed genetically in our material. Research on hantaviruses continue in this area, with stress on genetic confirmation of TULV prevalence in other species than common vole.

Other species exhibiting TULV antigen positivity in the Czech Republic include Mi. subterraneus and My. glareolus (Pejčoch et al. 2003a), both of which were negative in our study. Small mammals from woodland (with the exception of three common voles) were largely TULV antigen negative. This striking difference with results for the agricultural landscape is primarily caused by the composition of the small mammal community in the two habitats. We captured only low number of common voles in woodland and the most common forest species in our samples (wood mice and bank vole) have only rarely been found to be TULV antigen positive (Pejčoch et al. 2003a).

Among TULV antigen-positive voles, no significant difference was found between the sexes as to which was likely to be infected. Adult common vole males have three times larger home range than adult females (Reichstein 1960). Their contact with other individals are, therefore, expected to be more frequent and the risk of infection is more likely. Aggressive behavior is often recorded among rodents at higher densities. Bite wounds inflicted during fighting appear to be a major mode of virus transmission among rodents (Mills and Childs 2001). Adult females, on the other hand, have intimate and frequent contacts with their offspring and other relatives (Mackin-Rogalska 1979), even though their home ranges are smaller.

In California, research on the presence of hantavirus antigen (Bunyaviridae) was conducted on Peromyscus maniculatus and Reithrodontomys megalotis, which demonstrated its correlation with age (Bennett et al. 1999). Body weight and reproductive stage are indicators of age and our results that mature adults and the heaviest individuals were more likely to be TULVantigen positive accords with this result. High TULV prevalence in winter crops and set-aside in early spring can thus be a consequence of vole age composition. All TULV-antigen voles were adult and overwintering individuals and were from 6 to 9 months old (Pelikán 1986, Zejda and Nesvadbová 2000). Moreover, these individuals also often spend the winter months in common nests of three to seven individuals (Zejda et al. 2002), promoting the transmission of hantavirus among individuals. The overlapping space use by rodents might be an important factor affecting the local transmission of several hantaviruses (Root et al. 2005).

The incidence of hantaviruses is directly linked to the population and environmental conditions of their natural hosts (Pejčoch et al. 2003a). One of the factors we investigated was the influence of habitat (crop type). Unar et al. (1996) indicated that mesophilous grasslands were the main natural foci of the TULV. Some set-aside plots were grasslands, which can be important for vole overwintering (Heroldová et al. 2005). Annual crops (spring barley, winter wheat, winter rape) also provide suitable cover and food for the common vole (Heroldová et al. 2007). Also, the long growing period of winter rape and winter wheat (from September/October to the harvest in July the following year) is also an important factor that increases the possibility of the transmission of TULV. It is known that small mammal communities undergo permanent changes during annual crop management over the course of a year and populations can be under stress, which may weaken individual immunity (Heroldová et al. 2007).

The dynamics of hantavirus prevalence correlates with human infections. This effect was demonstrated through research on bank voles, the main host of Puumala virus. Three years of epidemic cycles in humans were found to correspond with the highest rates of prevalence in rodents (Escutenaire et al. 2000, Olsson et al. 2003). Here we show that the variability of TULV prevalence correlates with common vole population dynamics. But there are no data on dynamic of TULV human infection. Only mean seroprevalence rate of antihantavirus antibodies in Czech population is known to range about 0.8%. Much higher rates have been reported for our study area (southern Moravia) where seropositivity rates reached 9.9% and 29.4% in two groups of elderly farmers (Pejčoch and Kříž 2003). However, human infection with TULV shows limited symptoms (fever and exanthema) (Schultze et al. 2002). But in one man from Germany, serious symptoms (renal syndrome, pneumonia, and fever) were proved to be caused by TULV (Klempa et al. 2003). More studies are needed to understand this TULV incidence in humans, which is often unnoticed or misdiagnosed.

A highly agricultural landscape with only small wooded islands provides the most suitable environment for the common vole, the main reservoir of TULV. These environmental conditions may provide natural foci for TULV prevalence likely as in other countries (Klempa et al. 2003).

Footnotes

Acknowledgments

This research was supported by MSM 6215648902 and grants NR 9420-3/2007 of the Grant Agency of the Ministry of Health of the Czech Republic and NAZV QH 72075 Grant Agency of MAF. The authors are grateful to coworkers from the National Institute of Public Health Prague and Brno, especially to E. Pěkná, S. Plosová, and P. Klapušová, who helped us with laboratory and field work. For the genetic detection of TULV, the authors are grateful to L. Radosa, Department of Microbiology and Virology, Comenius University, Bratislava, Slovakia, and B. Klempa from Institute of Virology, SAS, Bratislava, Slovakia. The authors express gratitude to Dr. Carl Smith for language correction.

Disclosure Statement

No competing financial interests exist.

References

Bennett

, Webb

Jr. , Madon

, Childs

et al. Hantavirus (Bunyaviridae) infections from Orange and San Diego Counties, California. Am J Trop Med Hyg, 1999; 60:75–84.

Daneš

, Pejčoch

, Hubálek

, Halouzka

et al. Hantaviruses in small wild living mammals in Czechoslovakia. Results of a 1983–1989 study. J Hyg Epidemiol Microbiol Immunol, 1991; 35:281–288.

Daneš

, Tkachenko

, Ivanov

, Lím

et al. Hemorrhagic fever with renal syndrome in Czechoslovakia: detection of antigen in small terrestrial mammals and specific serum antibodies in man. J Hyg Epidemiol Microbiol Immunol, 1986; 30:79–85.

Escutenaire

, Chalon

, Verhagen

, Heyman

et al. Spatial and temporal dynamics of Puumala hantavirus infection in red bank vole (Clethrionomys glareolus) populations in Belgium. Virus Res, 2000; 67:91–107.

Groen

, Jordans

HGM

, Clement

et al. Identification of hantavirus serotypes by testing of post-infection sera in immunoflorescence and enzyme–linked immunosorbent assays. J Med Virol, 1991; 33:26–32.

Heroldová

, Bryja

, Zejda

, Tkadlec

. Structure and diversity of small mammal communities in agriculture landscape. Agric Ecol Environ, 2007; 120:206–210.

Heroldová

, Jánová

, Bryja

, Tkadlec

. Set-aside plots–source of small mammal pests? Folia Zool, 2005; 54:337–350.

Klempa

, Meisel

, Rath

, Bartel

et al. Occurrence of renal and pulmonary syndrome in a region of northeast Germany where Tula hantavirus circulates. J Clin Microbiol, 2003; 41:4894–4897.

Lisická

, Losík

, Zejda

, Heroldová

et al. Measurement error in burrow index to monitor relative population size in the common vole. Folia Zool, 2007; 56:169–176.

10.

Mackin-Rogalska

. Elements of the spatial organization of a common vole population. Acta Theriol, 1979; 24:171–199.

11.

Mills

, Childs

. Rodent-born hemorrhagic fever viruses. Williams

, Barker

. Infectious Diseases of Wild Mammals, 3rd. Ames: Iowa State University Press, 2001; 254–270.

12.

Olsson

, Ahim

, Elgh

, Verlemyr

et al. Hantavirus antibody occurrence in bank voles (Clethrionomys glareolus) during a vole population cycle. J Wildl Dis, 2003; 39:299–305.

13.

Pejčoch

. Hantaviry a nákazy jimi vyvolané. Review article. Klin Mikrobiol Infek Lékařství, 2003b. 9:4–11(In Czech with English abstract.)

14.

Pejčoch

, Daneš

, Unar

, Zahrádková

, Pěkná

. Prevalence nákaz drobných savců hantaviry v Jihomoravském kraji. Epidemiol Mikrobiol Imunol, 1992; 41:92–100(In Czech with English abstract.)

15.

Pejčoch

, Heroldová

, Zejda

, Treml

et al. Nálezy hantavirového antigenu u hlodavců v České republice. Epidemiol Mikrobiol Imunol, 2003a. 52:18–24(In Czech with English abstract.)

16.

Pejčoch

, Kříž

. Prevalence of antibodies against hantaviruses among the adult population of the Czech Republic. Cent Eur J Public Health, 2003; 11:169–172.

17.

Pelikán

. Small mammals in windbreaks and adjacent fields. Acta Sci Nat Brno, 1986; 20:1–38.

18.

Plyusnin

, Cheng

, Vapalahti

, Pejcoch

et al. Genetic variation in Tula hantaviruses: sequence analysis of the S and M segment of strains from central Europe. Virus Res, 1995; 39:237–250.

19.

Reichstein

. Untersuchungen zum Aktionsraum und zum Revierverhalten der Feldmaus Microtus arvalis (Pall.) Markierungsversuche Z Säugetierkunde, 1960; 25:150–169.

20.

Root

, Wilson

, Calisher

, Wagner

et al. Spatial clustering of Murid rodents infected with hantavirusis: implication from meta-analyses. Ecol Appl, 2005; 15:565–574.

21.

Schultze

, Lundquist

, Blauenstein

, Heyman

. Tula virus infection associated with fever and exanthema after rodent bite. Eur J Clin Microbiol Infect Dis, 2002; 21:304–306.

22.

Suchomel

, Heroldová

. Small terrestrial mammals in two types of forest complexes in intensively managed landscape of south Moravia (Czech Republic) Ekológia (Bratislava), 2004; 23:377–384.

23.

Tkachenko

, Ivanov

, Dzagurova

, Donets

et al. Immunosorbent assays for diagnosis of haemorrhagic fever with renal syndrome. Lancet, 1981; 318:257–258.

24.

Unar

, Daneš

, Pejčoch

. The basic ecology and vegetation characteristic of the natural foci of hantaviruses in South Moravia. Zprávy Vlastivědného muzea v Olomouci, 1996; 273:9–55.

25.

Vapalahti

, Lundkvist

, Kukkonen

, Cheng

et al. Isolation and characterization of Tula virus, a distinct serotype in the genus Hantavirus, family Bunyaviridae. J Gen Virol, 1996; 77:3063–3067.

26.

Wilson

, Reeder

. Mammal Species of the World. A Taxonomic and Geographic References. Baltimore: The Johns Hopkins University Press, 2005.

27.

Zeier

, Handermann

, Bahr

, Rensch

et al. Ecological aspects of hantavirus infection: a change of a paradigm and a challenge of prevention. Rev Virus Genes, 2005; 30:157–180.

28.

Zejda

, Nesvadbová

. Abundance and reproduction of the common vole, Microtus arvalis in crop rows and associated agricultural habitats. Folia Zool, 2000; 49:261–268.

29.

Zejda

, Zapletal

, Pikula

, Obdržálková

et al. Hlodavci v zemědělské a lesnické praxi. Agrospoj Praha, 2002(In Czech with English abstract.)