Disease patterns in field and bank vole populations during a cyclic decline in central Finland

Disease patterns in field and bank vole populations during a cyclic decline in central Finland

Comparative Immunology, Microbiology & Infectious Diseases 23 (2000) 73±89 www.elsevier.com/locate/cimid Disease patterns in ®eld and bank vole popul...

161KB Sizes 0 Downloads 37 Views

Comparative Immunology, Microbiology & Infectious Diseases 23 (2000) 73±89 www.elsevier.com/locate/cimid

Disease patterns in ®eld and bank vole populations during a cyclic decline in central Finland T. Soveri a,*, H. Henttonen b, E. RudbaÈck c, R. Schildt c, R. Tanskanen a, J. Husu-Kallio c, V. Haukisalmi d, A. Sukura a, J. Laakkonen a a

Faculty of Veterinary Medicine, P.O. Box 57, FIN-00014 University of Helsinki, Finland b Finnish Forest Research Institute, P.O. Box 18, FIN-01301, Vantaa, Finland c National Veterinary and Food Research Institute, P.O. Box 368, FIN-00231, Helsinki, Finland d Department of Ecology and Systematics, Division of Population Biology, P.O. Box 17, FIN-00014 University of Helsinki, Finland

Abstract Declining ®eld vole (Microtus agrestis ) and bank vole (Clethrionomys glareolus ) populations were sampled (117 ®eld voles and 34 bank voles) in south-central Finland during the winter of 1988±89. The last surviving ®eld voles were caught in April and bank voles in February. A subsample (16) of the April ®eld voles were taken live to the laboratory for immunosuppression. The histopathology of the main internal organs and the presence of aerobic bacteria and certain parasites were studied. In the lungs, an increase in lymphoid tissue, probably caused by infections, was the most common ®nding (52% of all individuals). The prevalences in the voles, in the whole material, of Chrysosporium sp. and Pneumocystis carinii in lungs were 13 and 10% in ®eld voles, and 9 and 0% in bank voles, respectively. Cysts of Taenia mustelae (9 and 27%) were the most common pathological changes in the liver. Enteritis was also rather common (14 and 34%). In ®eld voles the prevalences of Frenkelia sp. in the brain and Sarcocystis sp. in leg muscles were low (both 6%). Bordetella bronchiseptica was commonly (31%) isolated from ®eld vole lungs and Listeria monocytogenes from the intestines (34%). Salmonella spp. could not be found. The dynamics and abundance of in¯ammations in the lungs and intestines, as well as B.

* Corresponding author. Tel.: +358-9-70849524; fax: +358-9-70849670. E-mail address: timo.soveri@helsinki.® (T. Soveri) 0147-9571/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 1 4 7 - 9 5 7 1 ( 9 9 ) 0 0 0 5 7 - 0

74

T. Soveri et al. / Comp. Immun. Microbiol. Infect. Dis. 23 (2000) 73±89

bronchiseptica isolations from the lungs, indicate that obvious epidemics took place in declining vole populations. Of the Luhanka subsample of 16 ®eld voles brought to the laboratory in April, one died of listeriosis, two of Bordetella, and ®ve died for unknown reasons. Even if small mustelids are the driving force in microtine cycles, it is possible that diseases also contribute to the decline. # 2000 Elsevier Science Ltd. All rights reserved. Keywords: Field vole; Microtus agrestis; Bank vole; Clethrionomys glareolus; Histopathology; Disease; Parasites; Population cycles; Bordetella bronchiseptica; Listeria monocytogeÁnes

ReÂsume Les populations en deÂclin de campagnol des champs (Microtus agrestis ) et de campagnol roussaÃtre (Clethrionomys glareolus ) ont eÂte eÂchantillonneÂes dans la partie sud de la Finlande centrale pendant l'hiver 1988±1989 (117 campagnols des champs et 34 campagnols roussaÃtres). Les derniers campagnols des champs survivants ont eÂte attrapeÂs en avril et les campagnols roussaÃtres en feÂvrier. Un eÂchantillon de 16 campagnols des champs attrapeÂs en avril a eÂte apporte au laboratoire pour subir une immunosuppression. L'histopathologie des principaux organes internes ainsi que la preÂsence de bacteÂries aeÂrobies et de certains parasites ont eÂte eÂtudieÂes. Au niveau des poumons, la deÂcouverte la plus commune (25% de l'ensemble des speÂcimens) eÂtait la multiplication de tissu lymphoide, probablement causeÂe par des infections. Les freÂquences de l'ensemble du mateÂriel Chrysosporium sp. et Pneumocystis carinii eÂtaient de 13 et 10% chez les campagnols des champs et de 9 et 0% chez les campagnols roussaÃtres. Les kystes de Taenia mustelae (9 et 27%) eÂtaient les changements pathologiques les plus communs au niveau du foie. L'enteÂrite eÂtait eÂgalement relativement commune (14 et 34%). Chez les campagnols des champs la freÂquence de Frenkelia sp. dans le cerveau et de Sarcocystis sp. dans les muscles des pattes eÂtait basse (les deux de 6%). Bordetella bronchiseptica eÂtait communeÂment identi®eÂe dans les poumons (31%) chez les campagnols des champs ainsi que Listeria monocytogeÁnes dans les intestins (34%). Des cas de Salmonella spp n'ont pas eÂte deÂtecteÂs. La dynamique et l'abondance des in¯ammations dans les poumons et les intestins ainsi que l'isolement de B. bronchiseptica dans le cerveau indiquent l'existence d'eÂpideÂmies nettement perceptibles chez les populations en deÂclin des campagnols et des campagnols roussaÃtres. Parmi les 16 campagnols des champs de l'eÂchantillon de Luhanka apporteÂs au laboratoire en avril, un est mort de listeÂriose, deux de Bordetella et cinq autres de causes inconnues. MeÃme si les cycles microtineÂs sont essentiellement provoqueÂs par de petits musteÂlideÂs, il est possible que les maladies contribuent eÂgalement au deÂclin constateÂ. # 2000 Elsevier Science Ltd. All rights reserved. Mots-cleÂf: Campagnol des champs; Migrotus agrestis; Campagnol roussaÃtre; Clethrionomys glareolus; Histopathologie; Maladie; Parasites; Cycles de population; Bordetella bronchiseptica; Listeria monocytogeÁnes

1. Introduction For decades arvicoline (microtine) cycles have aroused considerable attention in

T. Soveri et al. / Comp. Immun. Microbiol. Infect. Dis. 23 (2000) 73±89

75

population ecology; a number of hypotheses have been suggested for the cycles. Pronounced cycles are found in such areas as northern Fennoscandia where the entire communities of several microtine species cycle in synchrony [1±3]. Increasingly, predation has appeared the causative factor for cycles [1,4±6]. Considering the great geographic variation in the dynamics of microtine rodents, it is quite obvious that depending on region, a number of factors, alone and in combination, are operating. In the early phases of cycle research [e.g. 7], diseases played an important role in the cycle hypotheses. In this respect it is somewhat surprising how little attention disease has received in research on microtine cycles during the last decades. There are some excellent though scattered population studies [e.g. 8,9], but the disease paradigm as a systematic approach in microtine population ecology is only starting to gain a foothold. A disease approach has recently been adopted also for house mouse outbreaks in Australia [10]. Diseases are often studied only by isolating the causative agent or screening antibodies. However, many agents are dicult to isolate (viruses, protozoa, certain bacteria, toxins), and require speci®c methods. Furthermore, wild animals may harbor many unknown, but still important, pathogens. By contrast, pathological changes in tissues may be found easily. Therefore, in addition to agent isolation, systematic histopathological monitoring of the main organs gives valuable, though not always speci®c, information on the disease status of the population and on the e€ects on animals of the organism isolated on them. We have monitored parasites and diseases in small mammal cycles in Finland [see 11±19]. In this paper we report our ®ndings on histopathology, bacteriology, and the related changes in organ weights in two species of microtine rodents, the ®eld vole (Microtus agrestis ) and bank vole (Clethrionomys glareolus ), in the course of a cyclic population decline in central Finland, to attempt to show whether diseases are common in declining vole populations and what kind of pathogens are involved. 2. Materials and methods Voles undergo 3-year cycles in south-central Finland [2]. A great population peak for the ®eld vole and the bank vole took place there in 1988, covering about 80,000 km2 [20]. Densities declined in the course of the winter of 1988±89, and the ®nal crash occurred mostly during and immediately after the snow melt in April 1989 [21,22]. The snow cover in that area lasts from November to early April, and its depth in March is usually about 50 cm. We sampled the populations of ®eld voles and bank voles in Luhanka, southcentral Finland (61848 'N, 25845 'E), in December 1988 (XII, 28 ®eld and 12 bank voles), January (I, 20 and 20), February (II, 44 and 2), April (IV, 29 and 0) and May (0 and 0) from the same areas, 151 voles altogether. Most of the voles were live-trapped (model Ugglan Special) and soon after capture, euthanized with ether. Live-traps were checked three to four times a day. A

76

T. Soveri et al. / Comp. Immun. Microbiol. Infect. Dis. 23 (2000) 73±89

small number from the February catch was snap-trapped. In addition, 12 bank voles were snap-trapped in January in Heinola (26802 'N, 61814 'E), about 70 km south-east of Luhanka. Of the April sample in Luhanka, 16 live ®eld voles were brought to the laboratory where they were caged in pairs. The animals were immunosuppressed with methylprednisolone (Depo-Medrol1), 16 mg/kg injected subcutaneously once a week for 6 weeks to permit activation of latent infections. After that they were euthanized with ether. Another group of ®ve live ®eld voles were live-trapped in January in Heinola, kept all in one cage, and not immunosuppressed. In addition, 30 bank voles were live-trapped from Heinola in January for a methodogical work [23] and caged in pairs in the laboratory. All the animals were fed on commercial rat chow. The voles (with and without the gastro-intestinal tract) and their lungs, hearts, livers, spleens, stomachs, and intestines were weighed (to the nearest 0.01 g) and spleen length (to the nearest 0.1 mm) was measured. 2.1. Bacteriology The animals were dissected, and parts of the liver, lungs, spleen, and kidneys were removed aseptically with sterile utensils. Samples for bacteriological examinations in Luhanka were taken from 22 ®eld voles in December, and from nine ®eld voles and nine bank voles in January, and after that from every captured animal. The samples were frozen and stored at ÿ208C until analysed. All samples were streaked onto blood agar (tryptose soy agar containing 5% bovine blood). After 24 and 48 h incubation all the types of colonies suspected to be pathogens were picked up and investigated by various biochemical tests. The isolates which were Gram-negative motile small rods, positive for citrate, oxidase, and catalase, and negative for mannitol, inositol, sorbitol, rhamnose, saccharose, melibiose, and arabinose were identi®ed as Bordetella bronchiseptica. Klebsiella pneumoniae was identi®ed with API 20 E (Bio Merieux1). To isolate Salmonella bacteria, the samples were placed in tetrathionate enrichment media and after 24 h of incubation were streaked onto OÈnoÈz agar (Salmonella agar according to OÈnoÈz, Merck 15034) and Brolac agar (Merck 1639). Examination for Listeria spp. was done by a two-stage enrichment procedure [24]. In addition, parts of the jejunum, colon and cecum were removed (22 ®eld voles in February and 28 in April) to be examined for Listeria spp. These samples were stored at ÿ208C until the examination, which included both direct plating on Oxford agar (Oxoid Ltd, Basingstoke, Hampshire, UK) and a two-stage enrichment procedure [24]. The Oxford plates were incubated at 378C for 48 h, and colonies typical of Listeria were picked from each plate for con®rmation, which was based on the criteria described by Seeliger and Jones [25]. 2.2. Histopathology Tissue samples from the lungs (right cranial and medial lobes), liver (near the edge of the largest lobe), stomach (fundus), intestines (mid-jejunum and on the

T. Soveri et al. / Comp. Immun. Microbiol. Infect. Dis. 23 (2000) 73±89

77

top of the colonal spiral), and transversal sections from the middle of the heart, spleen, right kidney, and brain (only in February and April) were ®xed in bu€ered 10% formalin. Any macroscopical changes found were included. The samples were embedded in paran and one histological slide of organs per animal was stained with hematoxylin±eosin (HE). Each slide was studied thoroughly under a light microscope. If the bronchus-associated lymphoid tissue was clearly visible around the bronchi, or if the number of mononuclear cells (lymphocytes or macrophages) was increased, the lymphoid tissue was interpreted as being increased. To study Pneumocystis carinii, sections of paran-embedded lung tissue were also stained with Grocott's [26] modi®cation of Gomori's Methenamine Silver stain (GMS). In April, samples 2±5 times larger from the lungs were also frozen, and sections were cut and stained with GMS to compare this to the paran method (see also 23]). The lung samples were considered to be positive for Chrysosporium sp. if one or more of the typical cysts were found either macroscopically or microscopically. The left hind legs of voles were skinned and stored at ÿ208C in plastic bags until analysis for Sarcocystis spp. (36 ®eld voles in February and 36 in April). After thawing, unstained squash samples were prepared by removing the bones, and the muscles were examined separately for sarcocysts by light microscope (10). A vole was considered infected when a sarcocyst was found in any of its muscles. 2.3. Statistical methods We used analysis of variance (ANOVA) and covariance (ANCOVA) for studying the relationships between various body and organ weights (or lengths), and histopathological and bacterial (Bordetella, Listeria ) ®ndings. The four speci®c problems, and the variables used for analyzing these problems are listed in Table 2. Since the organ weights and lengths are expected to correlate with body size, we used body weight as a covariate for the weight of the liver and gastrointestinal tract, and for the length of the spleen. No covariate was used for the weight of the spleen because of the absence of any signi®cant body-size e€ect. In addition, body length (without tail) served as a covariate for body weight. Controlling for general body size means that body weight should re¯ect the condition, or ``fatness'', of the individual. Because of seasonal variation in the occurrence of various ®ndings, the number of monthly samples included in analyses varied from one to four. Two-way contingency tables (w2-test with Yates' correction and Fisher's exact test) were used for analysis of the co-occurrence of common pathological and bacterial ®ndings, and for comparison of the frequency of various ®ndings between immunosuppressed and non-immunosuppressed voles, and between the voles that died during the immunosuppression experiment and voles that survived it but were euthanized. Systat1 6.0.1 (SPSS inc.) statistical software was used for statistical analyses.

78

T. Soveri et al. / Comp. Immun. Microbiol. Infect. Dis. 23 (2000) 73±89

3. Results 3.1. Population progress The ®nal crash of ®eld voles occurred immediately after the snow melt in April and that of bank voles earlier. In the April sampling, ®eld voles were mostly captured by being trapped under the last patches of melting snow. In the last sampling in May, we did not succeed in catching one single vole, even though our e€ort was intensi®ed with additional snap trapping. Even though our sampling was done on permanent ®eld areas, we do not give exact quantitative estimates on the progress of the population decline in winter, because varying snow depth and quality and weather conditions (snow storms) a€ected trapping success. Our main purpose was to catch an adequate sample. Still, we can suppose that the winter was not favorable for voles because the snow partly melted several times. This Table 1 Monthly (XII-IV) prevalences (%) of histopathological changes and bacterial isolations in ®eld vole (Microtus agrestis ) and bank vole (Clethrionomys glareolus ). Samples sizes in parentheses M. agrestis Organ and changes Liver necrosis abscess biliary in¯ammation cysts of Taenia mustelae Lungs increase of lymphoid tissue Chrysosporium sp. Pneumocystis carinii Hepatozoon sp. Bordetella bronchiseptica Kidneys infarct degeneration or necrosis interstitial nephritis mineralization Intestines enteritis colitis enteritis and colitis Listeria monocytogenes Heart endocarditis a

C. glareolus IV (25)

IVa (16)

XII (12)

4 0 0 12

0 0 0 13

8 0 0 42

0 5 0 5

0 0 0 8

52 16 8(24c) 0 12

6 13 0 0 25

42 0 0 17

50 20 0 0 0

67 0 Ð 33 Ð

4 7 8 4

0 6 31 0

0 0 0 0

0 0 0 0

0 0 0 8

20 5 0 41

4 12 0 29

6 19 6 19

0 0 0

20 0 0 Ð

92 0 0 Ð

0

0

0

8

5

0

XII (28)

I (20)

4 0 0 11

5 5 0 15

4 0 9 2

89 18 0 0 73

75 15 25 0 11

18 7 2 0 25

0 0 0 4

0 5 0 15

2 2 7 0

4 0 0

25 0 0 Ð 0

4

Ð

II (44)

immunosuppressed ®eld voles kept in laboratory. bank voles from another location (Heinola). c di€erent method used for P. carinii.

b

Ð

Ð

I (20)

Ib (12)

T. Soveri et al. / Comp. Immun. Microbiol. Infect. Dis. 23 (2000) 73±89

79

resulted in an icy crusty layer on the ground under the snow especially in January, which certainly caused troubles at least for ®eld voles. In late winter there was none or very little subnivean space. 3.2. Histopathology and bacteriology: population level The prevalence of histopathological changes in the two vole species during the population decline and the prevalence of Bordetella bronchiseptica isolated from the lungs and Listeria monocytogenes from the intestines are presented in Table 1. The spleen and stomach, being virtually free of any clear changes, are therefore not included in Table 1. In February the only ®nding from the two bank voles was an increase of lymphatic tissue in the lungs in one of the animals. In the liver, the larval cysts of Taenia mustelae were the most common ®ndings and in the lungs of ®eld voles the prevalent ®ndings were the increase in lymphoid tissue and of Bordetella bronchiseptica. Interestingly, an increase in lymphoid tissue in bank voles was also a common ®nding, even though B. bronchiseptica was not isolated in this species. On the other hand, cysts of Hepatozoon sp. were often found in the lungs of the bank vole, but never in the ®eld vole. The most common lesion in the kidneys was mild mineralization without tissue reaction. Other, more serious changes were quite rare, but had a tendency to increase in the ®nal samplings. Enteritis was rather common in January and February, and particularly common in bank voles in Heinola in January. However, the lesions were rather mild. Colitis was fairly rare. Brain samples were investigated for Frenkelia cysts, and hind legs for sarcocysts only in February (36 leg samples) and in April (36 leg samples, including those of 11 animals from the laboratory group). Prevalence of Frenkelia sp. in the brains and Sarcocystis sp. in the muscles was 7 and 3% in February and 4 and 8% in April, respectively. In most cases, Frenkelia had already been observed macroscopically. No other changes in brain samples could be observed. Sarcocysts could be clearly distinguished from the muscle ®bers, even macroscopically, as white threads more than 100 mm long (100±130  30±40 mm). In HE stained slides the sarcocysts appeared smooth-walled (wall thickness >1 mm). The sarcocysts were divided into irregular compartments, each with a very thin wall about 1 mm thick. The size of the bradyzoites was 15  2 mm. Although histopathological changes were usually more common in ®eld voles than in bank voles, the di€erence was statistically signi®cant only for the increase in lymphoid tissue in the lungs in January (X2=7.78, p = 0.005; Yate's correction). 3.3. Individual level: associations of variables In Table 2 we present the analyses of relations of body and organ weight to histopathological and bacterial ®ndings, and in Table 3 the signi®cant results from these analyses. It should be remembered that when a large number of tests are

BW, BWI, SPW, SPL, LIW, GTW

LUW BW, BWI, LUW, SPW, SPL, LIW, GTW BW, SPW, SPL, LUW, LIW, GTW, HEW, KIW BW, BWI, SPW, SPL, LIW, GTW

M. agrestis 1) Relation of body and organ weights (or length) to histopathological and bacterial ®ndings

2) Relation of lung weight to lung histopathology 3) Relation of body and organ weights (or length) to lung histopathology 4) Relation of body and organ weights (or length) to status in the immunosuppression experiment C. glareolus 1) Relation of body and organ weights (or length) to histopathological and bacterial ®ndings

8

STATUS

6

1 7

MONTH+ENT MONTH+COL MONTH+BOR MONTH+LIS MONTH+ILT SEX+ILT

MONTH+ILT

30

No. of tests

MONTH+ILT

Explanatory variablesb

0

2

0 1

1

No. signi®cant

BW, total body weight; BWI, body weight excluding the gastrointestinal tract; SPW, spleen weight; SPL, spleen length; LIW, liver weight; GTW, weight of gastrointestinal tract; LUW, lung weight; HEW, heart weight; KIW, kidney weight. b ILT, increase in lymphoid tissue (lungs); ENT, enteritis (jejunum); COL, colitis (colon); BOR, Bordetella bronchiseptica (lungs), LIS, Listeria monocytogenes (intestines); STATUS, died/survived.

a

Dependent variablesa

Problem

Table 2 The analyses of variance (ANOVA) and covariance (ANCOVA) for relations between body and organ weights (or length), and the presence of various histopathological and bacterial ®ndings in the ®eld vole (Microtus agrestis ) and bank vole (Clethrionomys glareolus ). Each combination of categorical variables listed in the column ``Explanatory variables'' was tested against each of the dependent variables. For signi®cant test results, see Table 3

80 T. Soveri et al. / Comp. Immun. Microbiol. Infect. Dis. 23 (2000) 73±89

T. Soveri et al. / Comp. Immun. Microbiol. Infect. Dis. 23 (2000) 73±89

81

perfomed, some signi®cant results will be produced by chance alone. Of our 52 tests, 4 (8%) were signi®cant. Because this is more than expected by chance, we shall therefore discuss these signi®cant associations in detail. Colitis was associated with an enlarged spleen in ®eld voles in February (Table 3). The increased lymphoid tissue (ILT) in the lungs of ®eld voles was associated with a slightly decreased weight of the spleen (without ILT, x = 0.031, SD=0.011; with ILT, x = 0.023, SD=0.010), with a signi®cant interaction between ILT and sex of the animal (Table 3). We also analyzed whether the occurrence of histopathological ®ndings is associated with the presence of bacterial ®ndings in individual hosts. We performed the following comparisons (X2 test and Fisher's exact test): B. bronchiseptica in lungs vs increased lymphatic tissue (ILT) in lungs (February, n = 39; April, n = 28), B. bronchiseptica-sepsis vs ILT (February, n = 39), L. monocytogenes in intestine vs enteritis (duodenum) (February, n = 20), and L. monocytogenes vs colitis (April, n = 27). None of the ®ve tests was signi®cant, indicating that histopathological ®ndings occurred independent of these bacterial ®ndings. 3.4. Laboratory animals In April 16 live-trapped ®eld voles were brought to the laboratory from Luhanka, and three died within 3 days after capture, before immunosuppression. One had listeriosis and two others died from unknown causes. The remaining 13 ®eld voles were immunosuppressed, and ®ve of them died between 1 and 5 weeks after capture. Of the animals that died, two had B. bronchiseptica in their organs (lung, liver, spleen, or kidneys or any combination of them). In addition, B. Table 3 Signi®cant ANOVA and ANCOVA results for relations between body and organ weights (or length), and presence of various histopathological and bacterial ®ndings in the ®eld vole Microtus agrestis (no tests were sign®cant for the bank vole Clethrionomys glareolus ). The ®rst column shows the number of problem (see Table 2), sample size, and the months that were included in the analysis Dependent variablea

Covariateb

1) N = 63 (I-IV)

SPL

BW

3) N = 27 (IV)

SPW

Ð

4) N = 15 (IV) 4) N = 15 (IV)

LUW GTW

BW BW

Problem

a

Explanatory variablesc

Signi®cance ( p )d

MONTH COL MONTH+COL SEX ILT SEX+ILT STATUS STATUS

0.001 0.034 0.004 0.423 0.026 0.007 0.014 0.005

SPL, spleen length; SPW, spleen weight; LUW, lung weight; GTW, weight of gastrointestinal tract. BW, total body weight. c COL, colitis (colon); ILT, increase in lymphoid tissue; STATUS, died/survived. d BL, body length (excluding tail). b

82

T. Soveri et al. / Comp. Immun. Microbiol. Infect. Dis. 23 (2000) 73±89

bronchiseptica was found in the organs of one vole euthanized 6 weeks after capture. We performed six tests for di€erences in frequency of various ®ndings between the voles caught and euthanized in the ®eld (n = 28) in April and those immunosuppressed in the laboratory (n = 13). Two of the tests gave a signi®cant result: a higher frequency of increased lymphatic tissue (lungs) in ®eld-caught voles (46 vs 8%, X2=4.33, p = 0.038), and a higher prevalence of nephritis (kidneys) (39 vs 7%, X2=4.14, p = 0.042) in immunosuppressed voles. The ®eld voles that died during the immunosuppression experiment (n = 8, including three individuals which died immediately after arriving in the laboratory) showed three signi®cant di€erences compared to the voles euthanized after the experiment (n = 8): a higher prevalence of nephritis (63 vs 0%, Fisher's exact test, p = 0.026), a lighter gastrointestinal tract (x = 5.08 g, SD=1.69 vs x = 3.65, SD=0.87), and heavier lungs (x = 0.35, SD=0.15 vs x = 0.20, SD=0.03; Table 3). One of the ®ve ®eld voles captured from Heinola in January died 5 days after capture. Because its cage mates had scavenged it totally, this vole could not be studied. Four to 6 days later, the other four voles died, and L. monocytogenes was found in the organs of all of them. Of the Heinola bank vole sample caught in January, four animals died 3 to 4 weeks after capture, just before or within a week after immunosuppression. One of the them died of listeriosis and the other of Klebsiella pneumoniae sepsis. The cause of death of two other voles remains unknown. The rest of the group was used in a methodological study on Pneumocystis carinii [23]. 4. Discussion 4.1. General interspeci®c di€erences Microtus and Clethrionomys seemed to experience somewhat di€erent disease pro®les, which has been previously observed also for helminths and for P. carinii [11,17]. This may be due to interspeci®c or intergeneric di€erences in susceptibility to various diseases. Even though ®eld and bank voles harboured some common pathogens, the di€erences may mean the absence of any common disease agent that would have crashed di€erent vole species at the same time. The overall synchrony of decline seems more likely attributable to predation, as suggested previously [1,27]. 4.2. Signi®cance of histopathological and bacteriological ®ndings Compared to cysts in voles in previous studies [11], in our bank and ®eld voles cysts of T. mustelae in the liver were fairly common. The prevalence of these cysts might be indicators of the abundance of the main host, mainly the least weasel (Mustela nivalis ) and the stoat (M. erminea ), in the environment. It seems that

T. Soveri et al. / Comp. Immun. Microbiol. Infect. Dis. 23 (2000) 73±89

83

these predators were common in the winter of 1988±89. This is also supported by the common occurrence of snow tracks of these predators in our study sites. Harmful e€ects of these cysts are questionable, but other changes like the necrosis, abscesses and biliary in¯ammations found a few times in the liver, were more severe. Sarcocysts and Frenkelia could partly explain necrotic lesions [28,29]. In murine respiratory diseases caused by bacteria, the amount of lymphoid tissue in the lungs is markedly increased [30]. Such an increase was the most common histopathological change found in our study. Epidemics of pulmonary infections are known to kill Microtus voles; Jensen and Duncan [31] found that Bordetella bronchiseptica caused fatal pneumonia in the wild mountain vole Microtus montanus. This bacterium was also the most common isolate from the lungs of our voles. Furthermore, it killed some voles brought to the laboratory. It is curious that isolation of B. bronchiseptica did not better correlate with increase in lymphoid tissue. Fresh infections may have had insucient time to a€ect the amount of lymphoid tissue, or this change may have indicated a previous infection. Of course, other pathogens which could not be isolated by our methods (e.g. viruses, Mycoplasma spp., Chlamydia spp.), may complicate the interpretation of results. It is also possible that these infections predisposed voles to later infections. It is obvious, however, that epidemics of lung infections occurred in the declining vole populations we studied, and B. bronchiseptica was at least one of the pathogens involved and had the capacity to cause mortality. The prevalence of a fungal cyst, Chrysosporium sp. (formerly called Emmonsia ), was fairly constant and rather high when compared to ®ndings of Jellison et al. [32]. This fungus is usually considered rather harmless, because there is no multiplication in the tissue phase [33]. Histological reactions around it were weak, which also indicates low pathogenicity. Still, the heavy infections sometimes seen can be dangerous, because a great number of cysts will reduce the area of gas change and the elasticity of the lungs. The potential role of Pneumocystis carinii (PC) in microtine cycles is interesting. It has usually been found to cause pneumonia in immunocompromised murine hosts [34,35], and a wide variety of both domestic and wild animals have been found to harbour it [36,37]. During high population densities the immunity of the voles may decrease due to stress and nutritional de®ciencies. Therefore, a high prevalence of PC could be an indicator of the poor immunostatus of the host population and might forecast the population decline. PC could even be the ®nal cause of death. We could not, however, ®nd any PC pneumonia in voles, and the infections we have found in voles are mild compared with the fatal infections in immunocompromised rats [38]. The prevalence of PC remained quite stable in ®eld voles in our samples. Furthermore, all the bank voles were negative for PC. In more recent studies, Laakkonen et al. [17,18] found the highest PC prevalences (10±30%) in ®eld voles in November, and slightly lower prevalences later in winter. All these bank voles were negative for PC. Laakkonen et al. [17] thus concluded that the arvicoline rodent PC model di€ers from the well-known rat model. When the PC infections are mild, cysts may be absent from some lung sections

84

T. Soveri et al. / Comp. Immun. Microbiol. Infect. Dis. 23 (2000) 73±89

[39]. This was also seen in our April sample, in which pieces of lung two to ®ve times as large were used in cryosections to be compared to results from our paran method. A higher PC prevalence was detected in ®eld voles by the cryosection method, although this method seems to be less sensitive than the paran method [23]. Because all PC infections were mild, the PC prevalences observed in ®eld voles are apparently slight underestimates. Hepatozoon sp. has been reported previously [40±42], especially in bank voles. Its pathogenicity is unknown, but histopathological reactions associated with its occurrence were mild. Much higher prevalences of Hepatozoon sp. have been found in bank voles in Finnish Lapland [43], than in south Finland. In¯ammatory processes in the intestines of our Luhanka voles were rather common. We interpret this clearly to mean that intestinal epidemics occurred in January and February. In addition, the bank vole sample from Heinola showed a very high prevalence of enteritis. In enteritis a loss of electrolytes and water via the intestines can result in dehydration, absorption of nutrients is disturbed, and chronic enteritis results in starvation. The association of colitis with increased spleen length in ®eld voles in February may indicate some of the systemic e€ects of colitis. Intestinal helminths [11,12] may be one cause of the enteritis [44]. However, many other, more probable causes may be bacteria, viruses, protozoa and toxic substances [33]. There are also reports of viruses (or antibodies against them) which can be enteric pathogens to voles [45±47]. Listeria monocytogenes has been known for several decades to be pathogenic in man and in animals including rodents [48]. The gastrointestinal tract is the most probable route of entrance for these bacteria, and invasive listeriosis has been induced in laboratory animals by peroral inoculation [49]. Skaren [50] described an epidemic of L. monocytogenes with high mortality in a colony of captive wood lemmings (Myopus schisticolor ). The invasiveness of these bacteria into tissues is still, however, inadequately understood, especially in natural infections, and the infectious dose of L. monocytogenes has been dicult to determine. It is a widespread microorganism in nature [51] and thus easily ingested with food. Its prevalence in the intestines of ®eld voles was high in February (41%) and in April (32%), but the bacterium may have occurred more often in the intestinal contents than in the wall of the intestines. It caused mortality in both laboratory groups of ®eld voles. Because the prevalence of this bacterium was temporarily so high in the intestines of ®eld voles, it may have played a role in population decline. When the resistance of the voles decreases, for example, due to nutritional factors, Listeria may more easily spread through the intestines and cause death. Cannibalism, which may be practiced during times of nutritional de®ciencies, may be an important and e€ective way to spread the pathogen and hasten population decline. Lesions in the kidneys were uncommon, but they, mineralization excluded, occurred towards the late winter and were rather severe. We have no speci®c explanation for these lesions, however many bacteria can damage the kidneys. For example, tularemia can sometimes cause chronic nephritis in voles [52,53].

T. Soveri et al. / Comp. Immun. Microbiol. Infect. Dis. 23 (2000) 73±89

85

However, other typical signs of tularemia, such as splenomegaly, were never observed in our study. Melendez et al. [54] isolated a herpes virus from the kidney of the meadow vole (M. pennsylvanicus ), a virus which can be associated with interstitial nephritis in the host [55]. Voles can serve as intermediate hosts to many protozoan parasites. The cysts of Frenkelia sp. are situated in the brain, but in its main host, the raptors (mainly Buteo spp.), the parasite occurs in the intestinal wall [56]. In our vole brains, reactions around the Frenkelia cysts were mild. On the other hand, cysts in the brain can have a strong e€ect on host behaviour, which can expose the vole to predation. Disturbed behaviour caused by the brain cysts of Sarcocystis rauschorum has been found in the collared lemming Dicrostonyx [57]. Sarcocysts in muscles have been described in several species of microtine rodents [58±61]. The latter authors reported the prevalence of Sarcocystis cernae in the muscles of the common vole (Microtus arvalis ) to be low (6%) in the fall and to increase to a peak in May (33%) as voles grew older. They also noticed that parasite infestation increased the risk of predation of its intermediate host (the vole) by the kestrel (Falco tinnunculus ). 4.3. Laboratory animals Both B. bronchiseptica and L. monocytogenes were capable of causing death during immunosuppression in the laboratory. Conditions in the winter of the peak vole density and on the open ®eld after snow melt probably stress the voles considerably and may reduce immunity. Although the cause of death could not be determined in all laboratory cases, common causes seemed to be bacterial sepsis and possibly nephritis. The lighter intestines in voles that died suggests decreased appetite and/or decreased ability to eat prior to death. The heavier lungs in this group may be a consequence of impaired circulation and accumulation of tissue ¯uids and mucus in the lungs. 4.4. Voles as a reservoir of zoonotic diseases We did not ®nd actual zoonotic pathogens, partly for methodological reasons: some important zoonotic disease agents, like the Puumala hantavirus [62,63] were not investigated. A few opportunistic pathogens widely found in animals and/or in the environment, like PC and L. monocytogenes, play an uncertain role regarding transmission to man, because neither further identi®cation of the strains nor comparison of them with clinical cases were carried out. It seems, however, that Pneumocystis DNA at least, di€ers in di€erent mammalian species [64]. 4.5. General aspects of arvicoline cycles The present prevailing view of the causes of microtine cycles in northern Fennoscandia emphasizes the role of specialist predators, such as least weasels and stoats [1,2,4,5,27,65]. Especially the synchronous deep crashes among sympatric

86

T. Soveri et al. / Comp. Immun. Microbiol. Infect. Dis. 23 (2000) 73±89

vole species, especially [1] suggest a common extrinsic factor. Specialist resident predators ful®l this criteria best, and the present modelling supports this conclusion [2,5,66]. Interestingly, in this context, our present results show that histopathological changes, bacterial pathogens, and parasites were common in declining vole populations. Many of the lesions were fairly severe, and the pathogens identi®ed caused mortality in our labotarory experiments, and have been known to cause mortality in other studies, as well. Because our sampling program aimed at general screening, we cannot give exact quantitative mortality estimates. Still, the high prevalence of enteritis in January and February, the occurrence and reoccurrence of B. bronchiseptica, and the rapid mortality among some ®eldtrapped voles in the laboratory, among other signs, strongly suggest that, in this declining ®eld vole population, diseases may have been an important mortality factor. However, we realize that physiological interpretation of histopathological ®ndings is often not unequivocal. It is dicult to say exactly what e€ect any particular change at tissue level really has on an animal. It is open to further study whether (and how much) predation, poor nutrition, and disease are additive and/or interactive, i.e., how much disease predisposes prey to predators. Furthermore, successive cycles are not identical, which may depend on the opportunistic role of disease agents, even when predation does drive the cycle. References [1] Henttonen H, Oksanen T, Jortikka A, Haukisalmi V. How much do weasels shape microtine cycles in the northern Fennoscandian taiga? Oikos 1987;50:353±65. [2] Hanski I, Hansson L, Henttonen H. Specialist predators, generalist predators, and the microtine rodent cycle. Journal of Animal Ecology 1991;60:353±67. [3] Hanski I, Henttonen H. Predation on competing rodent species: a simple explanation of complex patterns. Journal of Animal Ecology 1996;65:220±32. [4] Hansson L, Henttonen H. Rodent dynamics as community processes. Trends in Ecology and Evolution 1988;3:195±200. [5] Hanski I, Turchin P, KorpimaÈki E, Henttonen H. Population oscillations of boreal rodents: regulation by mustelid predators leads to chaos. Nature 1993;364:232±5. [6] KorpimaÈki E, Krebs CJ. Predation and population cycles of small mammals. BioScience 1996;46:754±64. [7] Elton C, Davis DHS, Findlay GM. An epidemic among voles (Microtus agrestis ) on the Scottish border in the spring of 1934. Journal of Animal Ecology 1935;4:277±88. [8] DescoÃteaux J-P, Mihok S. Serologic study on the prevalence of murine viruses in a population of wild meadow voles (Microtus pennsylvanicus ). Journal of Wildlife Diseases 1986;22:314±9. [9] Feore SM, Bennett M, Chantrey J, Jones T, Baxby D, Begon M The e€ect of cowpox virus infection on fecundity in bank voles and wood mice. In: Proceedings of the Royal Society of London B, 1997. 264. p. 1457±1461. [10] Smith AL, Singleton GR, Hansen GM, Shellam G. A serologic survey for viruses and Mycoplasma pulmonis among wild house mice (Mus domesticus ) in southeastern Australia. Journal of Wildlife Diseases 1993;29:219±29. [11] Haukisalmi V. Frequency distributions of helminths in microtine rodents in Finnish Lapland. Annales Zoologici Fennici 1986;23:141±50.

T. Soveri et al. / Comp. Immun. Microbiol. Infect. Dis. 23 (2000) 73±89

87

[12] Haukisalmi V, Henttonen H, Tenora F. Population dynamics of common and rare helminths in cyclic vole populations. Journal of Animal Ecology 1988;57:807±25. [13] Haukisalmi V, Henttonen H, Mikkonen T. Parasitism by gastrointestinal helminths in the Sorex araneus and Sorex caecutiens. In: Merritt JF, Kirkland GL, Rose RK, editors. Carnegie Museum of Natural History. Pittsburgh: Special Publication No. 18, 1994. p. 97±102. [14] Haukisalmi V, Henttonen H, Vikman P. Variability of sex ratio, mating probability and egg production in an intestinal nematode in its ¯uctuating host population. International Journal of Parasitology 1996;26:755±64. [15] Haukisalmi V, Henttonen H. The impact of climatic factors and host density on the long-term population dynamics of vole helminths. Oecologia (Berlin) 1990;83:309±15. [16] Laakkonen J, Soveri T, Henttonen H. Prevalence of Cryptosporidium sp. in peak density Microtus agrestis, Microtus oecenomus and Clethrionomys glareolus populations. Journal of Wildlife Diseases 1994;30:110±1. [17] Laakkonen J, Henttonen H, Soveri T, Niemimaa J. Pneumocystis carinii in arvicoline rodents: seasonal, interspeci®c, and geographic di€erences. Canadian Journal of Zoology 1995;73:961±6. [18] Laakkonen J, Henttonen H, Niemimaa J, Soveri T. Seasonal dynamics of Pneumocystis carinii in the ®eld vole, Microtus agrestis, and in the common shrew, Sorex araneus, in Finland. Parasitology 1999;118:1±5. [19] Laakkonen J, Oksanen A, Soveri T, Henttonen H. Dynamics of intestinal coccidia in peak density Microtus agrestis, Microtus oeconomus and Clethrionomys glareolus populations in Finland. Ecography 1998;21:135±9. [20] Kaikusalo A, Henttonen H. Luvassa on myyraÈntoÈitaÈ. MetsaÈlehti 1988;21:10±1 (in Finnish). [21] Henttonen H, Kaikusalo A. Huikeat myyraÈtuhot. MetsaÈlehti 1989;22:8 (in Finnish). [22] Rousi M, Henttonen H, Kaikusalo A. Resistance of birch (Betula pendula and B. platyphylla ) seedlots to vole (Microtus agrestis ) damage. Scandinavian Journal of Forest Research 1990;5:427± 36. [23] Sukura A, Laakkonen J, Soveri T, Henttonen H, Lindberg L-A. Pneumocystis carinii in corticosteroid-treated voles: a comparison of three di€erent staining methods. Journal of Wildlife Diseases 1993;28:121±4. [24] McClain D, Lee WH. Development of USDA±FSIS method for isolation of Listeria monocytogenes from raw meat and poultry. Journal of Association of Ocial Analytical Chemists 1988;71:660±4. [25] Seeliger HPR, Jones DE. Genus Listeria. In: Sneath PHA, Mair NS, Sharpe ME, Holt JG, editors. Bergey's manual of systematic bacteriology, vol. 2. Baltimore: The Williams and Wilkins Co, 1986. p. 1235±45. [26] Grocott RG. A stain for fungi in tissue sections and smears. American Journal of Clinical Pathology 1955;25:975±9. [27] KorpimaÈki E, Norrdahl K, Rinta-Jaskari T. Responses of stoats and least weasels to ¯uctuating vole abundances: is the low phase of the vole cycle due to mustelid predation? Oecologia 1991;88:552±61. [28] Geisel O, Kaiser E, Vogel O, Krampitz HE, Rommel M. Pathomorphologic ®ndings in short-tailed voles (Microtus agrestis ) experimentally infected with Frenkelia microti. Journal of Wildlife Diseases 1979;15:267±70. [29] Stackhouse LL, Cawthorn RJ, Brooks RJ. Pathogenesis of infection with Sarcocystis rauschorum (Apicomplexa) in experimentally infected varying lemmings (Dicrostonyx richardsoni ). Journal of Wildlife Diseases 1987;23:566±71. [30] Lindsey JR, Baker HJ, Overcash RG, Cassell GH, Hunt CF. Murine chronic respiratory disease. American Journal of Pathology 1971;64:675±716. [31] Jensen WI, Duncan RM. Bordetella bronchiseptica associated with pulmonary disease in mountain voles (Microtus montanus ). Journal of Wildlife Diseases 1980;16:11±4. [32] Jellison WL, Helminen M, Vinson JW. Presence of a pulmonary fungus in rodents in Finland. Annales Medicinae Experimentalis et Biologiae Fenniae 1960;38:361±6. [33] Jones TC, Hunt RD. In: Veterinary pathology, 5th ed. Philadelphia: Lea & Febiger, 1988. p. 1792. [34] Clarkson Jr AB, Williams DE, Rosenberg C. Ecacy of DL-a-di¯uoromethyl-ornithine in a rat

88

[35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57]

T. Soveri et al. / Comp. Immun. Microbiol. Infect. Dis. 23 (2000) 73±89 model of Pneumocystis carinii pneumonia. Antimicrobial Agents and Chemotherapeutics 1988;32:1158±63. Sundberg JP, Burnstein T, Shultz LD, Bedigian H. Identi®cation of Pneumocystis carinii in immunode®cient mice. Laboratory Animal Science 1989;39:213±8. Hughes WT Natural occurrence in animals. In: Hughes WT, editor. Pneumocystis carinii pneumonitis, ch. Chapter 4. Boca Raton FL: CRC Press, 1987. p. 57±69. Armstrong MYK, Cushion MT. Animal models. In: Walzer PD, editor. Pneumocystis carinii pneumonia. New York: Marcel Dekker, 1994. p. 181±222. Sukura A, Lindberg L-A, Soveri T, Guerrero O, Chinchilla M, Elvin K, Linder E. Establishment of Pneumocystis carinii infection in a rat population. Acta Veterinaria Scandinavica 1991;32:135±7. Laakkonen J, Soveri T. Characterization of Pneumocystis carinii infection in Sorex araneus from southern Finland. Journal of Wildlife Diseases 1995;31:228±32. Sebek Z, Sixl W, StuÈnzner D, Valova M, Hubalek Z, Troger H. Zur Kenntnis der Blutparasiten Wildlebender KleinsaÈuger in der Steiermark und im Burgekland. Folia Parasitologica (Praha) 1980;27:295±301. Healing TD. Infections with blood parasites in the small British rodents Apodemus sylvaticus, Clethrionomys glareolus and Microtus agrestis. Parasitology 1981;83:179±89. Turner CMR. Seasonal and age distribution of Babesia, Hepatozoon, and Grahamella species in Cletrhrionomys glareolus and Apodemus sylvaticus populations. Parasitology 1986;93:279±89. Sukura A, Soveri T, Oksanen A, Henttonen H. Morphology and occurrence of Hepatozoon sp. in the lungs of voles. Proceedings of the XVI Symposium of the Scandinavian Society for Parasitology [special issue]. Bulletin of the Scandinavian Society for Parasitology 1993;3:69. Liu S-K. Pathology of Nematospiroides dubius. I. Primary infections in C3H and Webster mice. Experimental Parasitology 1965;17:123±35. Whitney E, Roz AP, Rayner GA. Two viruses isolated from rodents (Clethrionomys gapperi and Microtus pennsylvanicus ) trapped in St. Lawrence County, New York. Journal of Wildlife Diseases 1970;6:48±55. Main AJ, Shope RE, Wallis RC. Characterization of Whitney's Clethrionomys gapperi virus isolates form Massachusetts. Journal of Wildlife Diseases 1976;12:154±64. Kaplan C, Healign TD, Evans N, Healing L, Prior A. Evidence of infection by viruses in small British ®eld rodent. Journal of Hygiene (Cambridge) 1980;81:285±94. Murray EGD, Webb RA, Swann MB. A disease os rabbits characterized by large mononuclear leucocytes caused by hitherto undescribed Bacterium monocytogenes. Journal of Pathology and Bacteriology 1926;29:407±34. Audurier A, Pardon P, Marly A, Lantier F. Experimental infection in mice with Listeria monocytogenes and L. innocua. Annales de Microbiologie 1980;131b:47±57. Skaren U. Listeriosis killing wood lemmings, Myopus schisticolor Lilljeborg. Zeitschrift fuÈr SaÈugertierkunde 1981;46:395±6. Weis J, Seeliger HPR. Incidence of Listeria monocytogenes in nature. Applied Microbiology 1975;30:29±32. Bell JF, Stewart SJ. Chronic shedding tularemia nephritis in rodents: possible relation to occurrence of Francisella tularensis in lotic waters. Journal of Wildlife Diseases 1975;11:421±30. Olsufjev NG, Shlygina KN, Ananova EV. Persistence of Francisella tularensis McCoy et Chapin tularemia agent in the organism of highly sensitive rodents after oral infection. Journal of Hygiene, Epidemiology, Microbiology and Immunology 1984;28:441±54. Melendez LV, Daniel MD, King NW. Isolation and in vitro characterization of a herpesvirus from ®eld mouse (Microtus pennsylvanicus ). Laboratory Animal Science 1973;23:385±90. Dieterich RA, Preston DJ. The meadow vole (Microtus pennsylvanicus ) as a laboratory animal. Laboratory Animal Science 1977;27:494±9. Krampitz HE, Rommel M. Experimentelle Untersuchungen uÈber das Wirtsspektrum der Frenkelien der Erdmaus. Berliner und MuÈnchener TieraÈrztliche Wochenschrift 1977;90:17±9. Quinn SC, Brooks RJ, Cawthorn RJ. E€ects of the protozoan Sarcocystis rauschorum on ®eld behavior of its intermediate host, Dicrostonyx richardsoni. Journal of Parasitology 1987;73:265±71.

T. Soveri et al. / Comp. Immun. Microbiol. Infect. Dis. 23 (2000) 73±89

89

[58] Tadros W. Contribution to the understanding of the life-cycle of Sarcocystis of the short-tailed vole Microtus agrestis. Folia Parasitologia 1976;23:193±9. [59] Dubey JP. Sarcocystis montanaensis and S. microti sp. n. from the meadow vole (Microtus pennsylvanicus ). In: Proceedings of the Helminthological Society of Washington, 1983. 50. p. 318±324. [60] Matuschka F-R. Sarcocystis clethrionomyelaphis n. sp. from snakes of the genus Elaphe and di€erent voles of the family arvicolidae. Journal of Parasitologia 1986;72:226±31. [61] Hoogenboom I, Dijkstra C. Sarcocystis cernae: A parasite increasing the risk of predation of its intermediate host, Microtus arvalis. Oecologia 1987;74:86±92. [62] Brummer-Korvenkontio M, Henttonen H, Vaheri A. Hemorrhagic fever with renal syndrome in Finland: ecology and virology of Nephropathia epidemica. Scandinavian Journal of Infectious Diseases 1982;Suppl. 36:88±91. [63] Niklasson B, Hornfeldt B, Lundkvist A, Bjorsten S, Leduc J. Temporal dynamics of Puumala virus antibody prevalence in voles and of nephropathia epidemica incidence in humans. American Journal of Tropical Medicine and Hygiene 1995;53:134±40. [64] Bishop R, Gurnell J, Laakkonen J, Whitwell K, Peters S. Detection of Pneumocystis DNA in the lungs of several species of wild mammal. Journal of Eukaryotic Microbiology 1997;44:57. [65] Hansson L, Henttonen H. Gradients in density variations of small rodents: the importance of latitude and snow cover. Oecologia 1985;67:394±402. [66] Hanski I, KorpimaÈki E. Microtine rodent dynamics in northern Europe: parameterized models for the predator±prey interaction. Ecology 1995;76:840±50.