- Email: [email protected]
Environmental determinants of the spatial distribution of Trichinella britovi and Trichinella spiralis in Hungary
Accepted Manuscript Title: Environmental determinants of the spatial distribution of Trichinella britovi and T. spiralis in Hungary Author: Z. Tolnai Z. Sz´ell G. Marucci E. Pozio T. Sr´eter PII: DOI: Reference:
S0304-4017(14)00233-7 http://dx.doi.org/doi:10.1016/j.vetpar.2014.04.024 VETPAR 7231
To appear in:
Veterinary Parasitology
Received date: Revised date: Accepted date:
15-11-2013 2-4-2014 21-4-2014
Please cite this article as: Tolnai, Z., Sz´ell, Z., Marucci, G., Pozio, E., Sr´eter, T.,Environmental determinants of the spatial distribution of Trichinella britovi and T. spiralis in Hungary, Veterinary Parasitology (2014), http://dx.doi.org/10.1016/j.vetpar.2014.04.024 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
SHORT COMMUNICATION
2 3
ip t
4
cr
5
Environmental determinants of the spatial distribution of Trichinella britovi
7
and T. spiralis in Hungary
us
6
an
8
Z. Tolnaia, Z. Szélla, G. Maruccib, E. Poziob, T. Srétera,*
9
11
a
M
10
Laboratories of Parasitology, Fish and Bee Diseases, Veterinary Diagnostic Directorate, National Food Chain Safety Office, Tábornok utca 2, H−1143 Budapest, Hungary
b
te
13
d
12
Department of Infectious, Parasitic and Immunomediated Diseases, Istituto Superiore di
14
16 17 18 19
Ac ce p
15
Sanitá, viale Regina Elena 299, 00161 Rome, Italy
20 21 22
* Corresponding author: E-mail: [email protected]; Fax +36 1 252 5177; Tel. +36 1 460
23
6322
24 1 Page 1 of 13
Abstract
25
Trichinella spiralis and T. britovi are the two most common species of the genus Trichinella
26
persisting in the European wildlife. To investigate the spatial distribution of these Trichinella
27
spp. and the factors influencing their circulation in Hungary, 3,304 red foxes (Vulpes vulpes)
28
and 0.29 million wild boars (Sus scrofa) were tested for Trichinella sp. infection in Hungary
29
from 2006 to 2013. Trichinella spp. larvae from 68 (2.0%) foxes and 44 (0.015%) wild boars
30
were identified by a multiplex PCR as T. britovi or T. spiralis. The locality of origin of foxes
31
and wild boars were recorded in a geographic information system database. There was no
32
correlation between environmental parameters in the home range of foxes and wild boars and
33
the T. spiralis larval counts, but there was a positive correlation between the boundary zone of
34
Hungary and T. spiralis infection (P < 0.0001; odds ratio: 24.1). These results indicate that the
35
distribution of T. spiralis in the Hungarian wildlife is determined by the transborder
36
transmission of the parasite from the surrounding endemic countries. Multiple regression
37
analysis was performed with environmental parameter values and T. britovi larval counts.
38
Based on the statistical analysis, non-agricultural areas (forests, scrubs, herbaceous vegetation
39
and pastures) and the mean annual temperature (P < 0.0001; odds ratios: 9.53 and 0.61) were
40
the major determinants of the spatial distribution of T. britovi in Hungary. The positive
41
relationship with non-agricultural areas can be explained by the generalist feeding behaviour
42
including scavenging of foxes in these areas. The negative relationship with the mean annual
43
temperature can be attributed to the slower decomposition of wildlife carcasses favouring a
44
longer survival of T. britovi larvae in the host carrion and to the increase of scavenging of
45
foxes.
Ac ce p
te
d
M
an
us
cr
ip t
24
46
2 Page 2 of 13
46
1.
47
Introduction Nematode parasites of the genus Trichinella are among the most widespread zoonotic
pathogens in the world (Pozio and Murrel, 2006). In Hungary, T. britovi is the most common
49
species circulating among carnivore and omnivore wild animals (Széll et al., 2008, 2012). To
50
assess the consumer risk for T. britovi and T. spiralis infections, it is important to know the
51
spatial distribution pattern of these parasites and environmental factors (e.g., temperature, land
52
cover) influencing this pattern. Geographical information systems (GIS) represent new tools
53
for studying the spatial distribution of parasites (Rinaldi et al., 2006). To investigate
54
environmental factors which can play a role in Trichinella spp. transmission in wildlife, red
55
foxes and wild boars, the most important indicator species of these parasites in the European
56
wildlife, were sampled and examined for Trichinella spp. infection in Hungary. The pattern of
57
infection in foxes and wild boars, and the relationship of these patterns with landscape and
58
climate was analysed by GIS.
62 63
te
61
2.
Materials and methods
Ac ce p
60
d
59
M
an
us
cr
ip t
48
2.1. Sample collection, parasite isolation and identification From November 2008 to February 2013, carcasses of red foxes killed by hunters in
64
Hungary were sent in individual plastic bags to the National Food Chain Safety Office of
65
Budapest. Carcasses were labelled by the hunters with an identification number reporting the
66
information on the nearest place to killing on the topographic map. Red fox carcasses,
67
representing more than 4% of the total fox population of each county, were randomly selected
68
out of all the foxes from 19 counties and from the Budapest municipality (Fig. 1). Muscle
69
samples were collected and examined as previously described (Széll et al., 2008).
70 71
Between June 2006 and September 2013, a total of 0.29 million hunted wild boars were tested at official control laboratories for Trichinella sp. larvae during routine meat inspection. 3 Page 3 of 13
When a Trichinella spp. positive animal was detected, a 100 g sample was collected from the
73
predilection muscles and sent to the National Food Chain Safety Office, Budapest, Hungary,
74
for the confirmatory diagnosis, evaluation of the worm burden (number of larvae/g, LPG) and
75
parasite isolation (Széll et al., 2012). As the place of killing of uninfected wild boars is not
76
recorded in Hungary, these data could not be included in the analysis.
Trichinella spp. larvae were washed, counted, and forwarded to the Istituto Superiore di
cr
77
ip t
72
Sanità, Rome, Italy, for the species identification as previously described (Széll et al. 2008,
79
2012).
us
78
81 82
an
80
2.2. Geographic information system database, spatial and statistical analysis The locality of origin of foxes and wild boars and the number of larvae/g were marked on a point layer by the Quantum GIS 1.8.0 softwareQGIS Team, 2012). The vector layers of
84
altitude, land cover, permanent water bodies, protected areas according to the Hungarian Law
85
No. LII/1996 (national parks, landscape protection areas, nature conservation areas), soil water
86
retention and soil permeability, were obtained from VÁTI Hungarian Nonprofit Ltd. for
87
Regional Development and Planning (Budapest, Hungary). The vector layers of the mean
88
annual temperature and annual precipitation were created and vector-based analysis was
89
carried out by the Quantum GIS 1.8.0 software on the basis of the georeferenced digital map of
90
the Hungarian Meteorological Service. The spatial resolution of the vector layers varied from
91
10 m (land cover, country border) to 50–100 m (other layers). The radius around the locality of
92
animal origin was restricted to 2.5 and 3.0 km, which was assumed to represent the average
93
home range of foxes and wild boars, respectively (Staubach et al., 2001; Calenge et al., 2002).
94
Along permanent water bodies, a 100 meter wide buffer zone was created, where the
95
probability of the presence of reservoir hosts was high. Along country border, a 10 km wide
96
buffer zone was created to evaluate the transborder transmission of these parasites from
97
neighbouring countries. The digitized home range and the vector data were used to calculate
Ac ce p
te
d
M
83
4 Page 4 of 13
the altitude, mean annual temperature, annual precipitation, areas of land cover types, and the
99
buffer zones of permanent water bodies and country borders. The associations between
100
environmental factors and the larval burden of T. britovi and T. spiralis infection were
101
analysed by multiple linear regression analysis and logistic regression analysis as previously
102
described (Tolnai et al., 2013).
ip t
98
104
3.
cr
103 Results
T. spiralis larvae were detected in 8 foxes and 12 wild boars (Table 1). T. britovi larvae
106
were identified in 60 foxes and 32 wild boars (Table 1). The spatial distribution of T. britovi
107
and T. spiralis was highly clumped in Hungary (Fig. 1).
an
108
us
105
There was no correlation between environmental parameters and the T. spiralis LPG. The multiple regression analysis and the logistic regression analysis show a positive correlation
110
(95% CI of odds ratios: 3.81–108.3) between the buffer zone of the Hungarian border and T.
111
spiralis LPG in foxes (P < 0.0001). When the data of T. spiralis infected wild boars were
112
included in the analyses, the results were very similar (95% CI of odds ratios: 7.9–73.4).
te
d
M
109
There was no correlation between protected areas, permanent water bodies, soil water
114
retention or soil permeability in the home range of foxes and wild boars and the T. britovi LPG.
115
A positive correlation between the annual precipitation and T. britovi LPG in foxes (P <
Ac ce p
113
116
0.05) or in foxes and wild boars (P < 0.005) was observed; however, this relationship was not
117
confirmed by the multiple regression analysis and the logistic regression analysis.
118
A positive correlation between the average altitude of the home range of foxes (P <
119
0.0001) or of foxes and wild boars (P < 0.0001) and the T. britovi LPG was observed;
120
however, the multiple regression analysis revealed a multicollinearity with the temperature
121
(redundant information). Therefore, the altitude was removed from the regression models.
122 123
A negative correlation between agricultural areas (permanent crops, arable land and heterogeneous agricultural areas) of the home range of foxes (P < 0.0001) or of foxes and wild 5 Page 5 of 13
124
boars (P < 0.0001) and T. britovi LPG was observed. As the multiple regression analysis
125
revealed a multicollinearity with non-agricultural areas, agricultural areas was removed from
126
the regression models. No other relationship with land cover types could be observed. A positive correlation between non-agricultural areas (forests, scrubs, herbaceous vegetation
ip t
127
and pastures) of the fox home range and the T. britovi LPG (P < 0.0001) and a negative
129
correlation between the mean annual temperature of the fox home range and T. britovi LPG (P <
130
0.0001) were observed (i.e., the lower the mean temperature, the higher the T. britovi prevalence
131
in foxes). The percentage of non-agricultural areas and the mean annual temperature of the home
132
range of infected foxes were 61 ± 3% and 9.1 ± 0.1°C, respectively. The multiple regression
133
analysis (P < 0.005; P < 0.001) and the logistic regression analysis (P < 0.0001; P < 0.005)
134
confirmed these relationships . Although the best fit model was the logistic regression model
135
(classified correctly with P = 0.5: 98.2%; Hosmer-Lemeshow test: P = 0.81; area under ROC
136
curve: 0.76 ± 0.03), the overall fit of both models was satisfactory (P < 0.0001). Based on the
137
logistic regression analysis, the percentage of non-agricultural areas (95% CI of odds ratios:
138
2.219–19.516) and the mean annual temperature (95% CI of odds ratios: 0.477–0.832) were the
139
major determinants of the spatial distribution of T. britovi in foxes of Hungary. If the data of T.
140
britovi infected wild boars were included in the analyses, the results were very similar (P <
141
0.0001; classified correctly with P = 0.5: 97.2%; Hosmer-Lemeshow test: P = 0.65; area under
142
ROC curve: 0.79 ± 0.02; 95% CI of odds ratios: 3.879–23.407 and 0.488–0.772).
143 144 145
4.
Ac ce p
te
d
M
an
us
cr
128
Discussion
GIS was used only in one study focusing on the environmental determinants of the spatial
146
distribution of Trichinella spp. in Europe (Pozio et al., 2009). As the geographic coordinates of
147
Trichinella-free animals were not available in that study, only the spatial distribution and
148
environmental determinants of T. spiralis and T. britovi could be compared. In the present study,
149
the geographic coordinates of the Trichinella spp. negative foxes were also available; therefore, 6 Page 6 of 13
150
the environmental determinants of the spatial distribution of T. spiralis and T. britovi could be
151
analysed.
152
The prevalence of T. spiralis in wildlife increases when this parasite species is also circulating in the domestic habitat (e.g., Croatia, Romania, Serbia) (Pozio and Murrell, 2006).
154
In those countries (e.g., Austria, Hungary, Slovenia), where T. spiralis has been almost
155
completely eradicated from the synathropic cycle (Széll et al., 2012), it can be detected only
156
sporadically in wildlife (ITRC, www.iss.it/site/Trichinella). The statistical analysis confirms (P
157
< 0.0001; odds ratio: 24.1) that the distribution of T. spiralis in the Hungarian wildlife is
158
determined by the transborder transmission of the parasite from the neighbouring endemic
159
countries (i.e., Croatia, Romania, Serbia) by wild mammals (Széll et al., 2013) and not by
160
environmental factors. Hungary is surrounded by several border rivers (e.g., Dráva, Ipoly,
161
Danube, Tisza) coming from other countries. These rivers serve as ecological corridors for
162
wildlife and may play a significant role in the long distance spread of T. spiralis (Széll et al.,
163
2013). Moreover, these river valleys with dense vegetation, cool and humid microclimate and
164
high population density of reservoir hosts may also favour the long term persistence of T.
165
spiralis on both sides of these rivers (e.g., in the border region of Slovakia). The transborder
166
transmission and the possible long term persistence of the parasite in the environment of border
167
rivers is a significant risk for the reintroduction of T. spiralis to the domestic habitat in
168
Hungary.
cr
us
an
M
d
te
Ac ce p
169
ip t
153
The positive relationship between the prevalence of T. britovi in foxes and non-agricultural
170
areas observed in the present study can be explained by the different food resources of foxes in
171
the agricultural and non-agricultural areas (Pozio, 1998). In agricultural areas, foxes become
172
specialist predators due to prey profitability and focus predation on small rodents (Ferrari, 1995).
173
Moreover, foxes can feed on anthropogenic food resources in these areas. In non-agricultural
174
areas, foxes are generalist predators due to the shortage of food resources, which increases
175
scavenging and cannibalism (Kidawa and Kowalczyk, 2011), favouring Trichinella spp. 7 Page 7 of 13
transmission (Pozio, 1998). The relationship between low environmental temperature and T.
177
britovi infection can be attributed to the slower decomposition of animal carcasses, the increase
178
of scavenging of foxes (i.e., the lower the temperature, the higher the scavenging activity) and
179
the longer survival of T. britovi larvae at lower temperatures (Von Köller et al., 2001; Selva et
180
al., 2005; Sharanowski et al., 2008; Pozio et al., 2009). The present results explain the spatial
181
distribution of T. britovi in Europe. In wildlife, T. britovi is more widespread in the colder
182
northern regions with lower agricultural activities (e.g., Scandinavian and Baltic countries) than
183
in the central and southern regions of Europe (Pozio, 1998; Pozio and Murrell, 2006Pozio et al.,
184
2009). In the latter regions, Trichinella sp. infections are more prevalent among foxes living in
185
non-agricultural areas (e.g., mountains, protected areas, forested regions) (Pozio, 1998; Pozio et
186
al., 2009). Our results also confirm the findings of a North American GIS-based study
187
identifying temperature and land cover as the major determinants of the distribution of other
188
Trichinella spp. in wildlife (Masouka et al., 2009).
M
an
us
cr
ip t
176
te
d
189 References
191
Calenge, C., Maillard, D., Vassant, J., Brandt, S., 2002. Summer and hunting season home
192 193 194 195 196
Ac ce p
190
ranges of wild boars (Sus scrofa) in two habitats in France. Game Wildl. Sci. 19, 181–301.
Ferrari, N., 1995. Influence of the abundance of food resources on the feeding habits of the red fox, Vulpes vulpes, in western Switzerland. J. Zool. 236, 117–129.
Kidawa, D., Kowalczyk, R., 2011. The effects of sex, age, season and habitat on diet of the red fox, Vulpes vulpes, in northeastern Poland. Acta Theriol. 56, 209–218.
197
Masuoka, P.M., Burke, R., Colaccico, M., Razuri, H., Hill, D., Murrell, K.D., 2009. Predicted
198
geographic ranges for North American sylvatic Trichinella species. J. Parasitol. 95, 829–
199
937.
200
Pozio, E., 1998. Trichinellosis in the European Union: epidemiology, ecology and economic
8 Page 8 of 13
202 203 204
impact. Parasitol. Today 14, 35–38. Pozio, E., Murrell, K.D., 2006. Systematics and epidemiology of Trichinella. Adv. Parasitol. 63, 367–439. Pozio, E., Rinaldi, L., Marucci, G., Musella, V., Galati, F., Cringoli, G., Boireau, P., La Rosa,
ip t
201
G., 2009. Hosts and habitats of Trichinella spiralis and Trichinella britovi in Europe. Int. J.
206
Parasitol. 39, 71–79.
cr
205
QGIS Team, 2012. Quantum GIS. Accessed at: http://www.qgis.org.
208
Rinaldi, L., Musella, V., Biggeri, A., Cringoli, G., 2006. New insights into the application of
us
207
geographical information systems and remote sensing in veterinary parasitology. Geospat.
210
Health 1, 33−47.
213
M
212
Selva, N., Jedrzejewska, B., Jedrzejewski, W., Wajrak, A., 2005. Factors affecting carcass use by a guild of scavengers in European temperature woodland. Can. J. Zool. 83, 1591−1601. Sharanowski, B.J., Walker, E.G., Anderson, G.S., 2008. Insect succession and decomposition
d
211
an
209
patterns on shaded and sunlit carrion in Saskatchewan in three different seasons. Forensic
215
Sci. Int. 179, 219–240.
te
214
Staubach, C., Thulke, H.H., Tackmann, K., Hugh-Jones, M., Conraths, F.J., 2001. Geographic
217
information system-aided analysis of factors associated with the spatial distribution of
218 219 220 221 222
Ac ce p
216
Echinococcus multilocularis infections of foxes. Am. J. Trop. Med. Hyg. 65, 943−948.
Széll, Z., Marucci, G., Bajmóczy, E., Cséplő, A., Pozio, E., Sréter, T., 2008. Spatial distribution of Trichinella britovi, T. pseudospiralis and T. spiralis in red foxes (Vulpes vulpes) in Hungary. Vet. Parasitol. 156, 210−215.
Széll, Z., Marucci, G., Ludovisi, A., Gómez-Morales, M.A., Sréter, T., Pozio, E., 2012. Spatial
223
distribution of Trichinella britovi, T. spiralis and T. pseudospiralis of domestic pigs and
224
wild boars (Sus scrofa) in Hungary. Vet. Parasitol. 183, 393−396.
225
Széll, Z., Marucci, G., Pozio, E., Sréter, T., 2013. Echinococcus multilocularis and Trichinella 9 Page 9 of 13
226 227 228
spiralis in golden jackals (Canis aureus) of Hungary. Vet. Parasitol. 197, 393−396. Tolnai, Z., Széll, Z., Sréter, T., 2013. Environmental determinants of the spatial distribution of Echinococcus multilocularis in Hungary. Vet. Parasitol. 198, 292−297. Von Köller, J., Kapel, C.M.O., Enemark, H.L., Hindsbo, O., 2001. Infectivity of Trichinella
230
spp. recovered from decaying mouse and fox muscle tissue. Parasite 8, S209−S212.
ip t
229
Ac ce p
te
d
M
an
us
cr
231
10 Page 10 of 13
231
Table 1. Prevalence and intensity (larvae per gram) of Trichinella spiralis and T. britovi
232
infections in red foxes (Vulpes vulpes) (n = 3,304) and wild boars (Sus scrofa) (n = 0.29
233
million) collected in Hungary from 2006 to 2013. T. britovi
ip t
T. spiralis Prevalence (%)
Intensity
Prevalence (%)
(95% CI)
(± SE)
(95% CI)
Red foxes
0.24 (0.16−0.32)
3.2 ± 1.0
1.82 (1.59−2.05)
15.9 ± 2.5
Wild boars
0.004 (0.003−0.005)
33.2 ± 25.8
0.01 (0.008−0.012)
16.6 ± 2.7
us
cr
(± SE)
Ac ce p
te
d
M
an
234
Intensity
11 Page 11 of 13
235
Legend to figure
236 Figure 1. Map of Hungary showing animals sampled from 2006 to 2013. Panel A. Uninfected
238
and Trichinella spiralis infected red foxes (Vulpes vulpes) (circles) and wild boars (Sus scrofa)
239
(triangles). Panel B. Uninfected and Trichinella britovi infected red foxes (circles) and wild
240
boars (triangles).
cr
ip t
237
Ac ce p
te
d
M
an
us
241
12 Page 12 of 13
an
us
cr
ip t
A
M
241
Ac ce p
te
d
B
242 243
13 Page 13 of 13
S0304-4017(14)00233-7 http://dx.doi.org/doi:10.1016/j.vetpar.2014.04.024 VETPAR 7231
To appear in:
Veterinary Parasitology
Received date: Revised date: Accepted date:
15-11-2013 2-4-2014 21-4-2014
Please cite this article as: Tolnai, Z., Sz´ell, Z., Marucci, G., Pozio, E., Sr´eter, T.,Environmental determinants of the spatial distribution of Trichinella britovi and T. spiralis in Hungary, Veterinary Parasitology (2014), http://dx.doi.org/10.1016/j.vetpar.2014.04.024 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
SHORT COMMUNICATION
2 3
ip t
4
cr
5
Environmental determinants of the spatial distribution of Trichinella britovi
7
and T. spiralis in Hungary
us
6
an
8
Z. Tolnaia, Z. Szélla, G. Maruccib, E. Poziob, T. Srétera,*
9
11
a
M
10
Laboratories of Parasitology, Fish and Bee Diseases, Veterinary Diagnostic Directorate, National Food Chain Safety Office, Tábornok utca 2, H−1143 Budapest, Hungary
b
te
13
d
12
Department of Infectious, Parasitic and Immunomediated Diseases, Istituto Superiore di
14
16 17 18 19
Ac ce p
15
Sanitá, viale Regina Elena 299, 00161 Rome, Italy
20 21 22
* Corresponding author: E-mail: [email protected]; Fax +36 1 252 5177; Tel. +36 1 460
23
6322
24 1 Page 1 of 13
Abstract
25
Trichinella spiralis and T. britovi are the two most common species of the genus Trichinella
26
persisting in the European wildlife. To investigate the spatial distribution of these Trichinella
27
spp. and the factors influencing their circulation in Hungary, 3,304 red foxes (Vulpes vulpes)
28
and 0.29 million wild boars (Sus scrofa) were tested for Trichinella sp. infection in Hungary
29
from 2006 to 2013. Trichinella spp. larvae from 68 (2.0%) foxes and 44 (0.015%) wild boars
30
were identified by a multiplex PCR as T. britovi or T. spiralis. The locality of origin of foxes
31
and wild boars were recorded in a geographic information system database. There was no
32
correlation between environmental parameters in the home range of foxes and wild boars and
33
the T. spiralis larval counts, but there was a positive correlation between the boundary zone of
34
Hungary and T. spiralis infection (P < 0.0001; odds ratio: 24.1). These results indicate that the
35
distribution of T. spiralis in the Hungarian wildlife is determined by the transborder
36
transmission of the parasite from the surrounding endemic countries. Multiple regression
37
analysis was performed with environmental parameter values and T. britovi larval counts.
38
Based on the statistical analysis, non-agricultural areas (forests, scrubs, herbaceous vegetation
39
and pastures) and the mean annual temperature (P < 0.0001; odds ratios: 9.53 and 0.61) were
40
the major determinants of the spatial distribution of T. britovi in Hungary. The positive
41
relationship with non-agricultural areas can be explained by the generalist feeding behaviour
42
including scavenging of foxes in these areas. The negative relationship with the mean annual
43
temperature can be attributed to the slower decomposition of wildlife carcasses favouring a
44
longer survival of T. britovi larvae in the host carrion and to the increase of scavenging of
45
foxes.
Ac ce p
te
d
M
an
us
cr
ip t
24
46
2 Page 2 of 13
46
1.
47
Introduction Nematode parasites of the genus Trichinella are among the most widespread zoonotic
pathogens in the world (Pozio and Murrel, 2006). In Hungary, T. britovi is the most common
49
species circulating among carnivore and omnivore wild animals (Széll et al., 2008, 2012). To
50
assess the consumer risk for T. britovi and T. spiralis infections, it is important to know the
51
spatial distribution pattern of these parasites and environmental factors (e.g., temperature, land
52
cover) influencing this pattern. Geographical information systems (GIS) represent new tools
53
for studying the spatial distribution of parasites (Rinaldi et al., 2006). To investigate
54
environmental factors which can play a role in Trichinella spp. transmission in wildlife, red
55
foxes and wild boars, the most important indicator species of these parasites in the European
56
wildlife, were sampled and examined for Trichinella spp. infection in Hungary. The pattern of
57
infection in foxes and wild boars, and the relationship of these patterns with landscape and
58
climate was analysed by GIS.
62 63
te
61
2.
Materials and methods
Ac ce p
60
d
59
M
an
us
cr
ip t
48
2.1. Sample collection, parasite isolation and identification From November 2008 to February 2013, carcasses of red foxes killed by hunters in
64
Hungary were sent in individual plastic bags to the National Food Chain Safety Office of
65
Budapest. Carcasses were labelled by the hunters with an identification number reporting the
66
information on the nearest place to killing on the topographic map. Red fox carcasses,
67
representing more than 4% of the total fox population of each county, were randomly selected
68
out of all the foxes from 19 counties and from the Budapest municipality (Fig. 1). Muscle
69
samples were collected and examined as previously described (Széll et al., 2008).
70 71
Between June 2006 and September 2013, a total of 0.29 million hunted wild boars were tested at official control laboratories for Trichinella sp. larvae during routine meat inspection. 3 Page 3 of 13
When a Trichinella spp. positive animal was detected, a 100 g sample was collected from the
73
predilection muscles and sent to the National Food Chain Safety Office, Budapest, Hungary,
74
for the confirmatory diagnosis, evaluation of the worm burden (number of larvae/g, LPG) and
75
parasite isolation (Széll et al., 2012). As the place of killing of uninfected wild boars is not
76
recorded in Hungary, these data could not be included in the analysis.
Trichinella spp. larvae were washed, counted, and forwarded to the Istituto Superiore di
cr
77
ip t
72
Sanità, Rome, Italy, for the species identification as previously described (Széll et al. 2008,
79
2012).
us
78
81 82
an
80
2.2. Geographic information system database, spatial and statistical analysis The locality of origin of foxes and wild boars and the number of larvae/g were marked on a point layer by the Quantum GIS 1.8.0 softwareQGIS Team, 2012). The vector layers of
84
altitude, land cover, permanent water bodies, protected areas according to the Hungarian Law
85
No. LII/1996 (national parks, landscape protection areas, nature conservation areas), soil water
86
retention and soil permeability, were obtained from VÁTI Hungarian Nonprofit Ltd. for
87
Regional Development and Planning (Budapest, Hungary). The vector layers of the mean
88
annual temperature and annual precipitation were created and vector-based analysis was
89
carried out by the Quantum GIS 1.8.0 software on the basis of the georeferenced digital map of
90
the Hungarian Meteorological Service. The spatial resolution of the vector layers varied from
91
10 m (land cover, country border) to 50–100 m (other layers). The radius around the locality of
92
animal origin was restricted to 2.5 and 3.0 km, which was assumed to represent the average
93
home range of foxes and wild boars, respectively (Staubach et al., 2001; Calenge et al., 2002).
94
Along permanent water bodies, a 100 meter wide buffer zone was created, where the
95
probability of the presence of reservoir hosts was high. Along country border, a 10 km wide
96
buffer zone was created to evaluate the transborder transmission of these parasites from
97
neighbouring countries. The digitized home range and the vector data were used to calculate
Ac ce p
te
d
M
83
4 Page 4 of 13
the altitude, mean annual temperature, annual precipitation, areas of land cover types, and the
99
buffer zones of permanent water bodies and country borders. The associations between
100
environmental factors and the larval burden of T. britovi and T. spiralis infection were
101
analysed by multiple linear regression analysis and logistic regression analysis as previously
102
described (Tolnai et al., 2013).
ip t
98
104
3.
cr
103 Results
T. spiralis larvae were detected in 8 foxes and 12 wild boars (Table 1). T. britovi larvae
106
were identified in 60 foxes and 32 wild boars (Table 1). The spatial distribution of T. britovi
107
and T. spiralis was highly clumped in Hungary (Fig. 1).
an
108
us
105
There was no correlation between environmental parameters and the T. spiralis LPG. The multiple regression analysis and the logistic regression analysis show a positive correlation
110
(95% CI of odds ratios: 3.81–108.3) between the buffer zone of the Hungarian border and T.
111
spiralis LPG in foxes (P < 0.0001). When the data of T. spiralis infected wild boars were
112
included in the analyses, the results were very similar (95% CI of odds ratios: 7.9–73.4).
te
d
M
109
There was no correlation between protected areas, permanent water bodies, soil water
114
retention or soil permeability in the home range of foxes and wild boars and the T. britovi LPG.
115
A positive correlation between the annual precipitation and T. britovi LPG in foxes (P <
Ac ce p
113
116
0.05) or in foxes and wild boars (P < 0.005) was observed; however, this relationship was not
117
confirmed by the multiple regression analysis and the logistic regression analysis.
118
A positive correlation between the average altitude of the home range of foxes (P <
119
0.0001) or of foxes and wild boars (P < 0.0001) and the T. britovi LPG was observed;
120
however, the multiple regression analysis revealed a multicollinearity with the temperature
121
(redundant information). Therefore, the altitude was removed from the regression models.
122 123
A negative correlation between agricultural areas (permanent crops, arable land and heterogeneous agricultural areas) of the home range of foxes (P < 0.0001) or of foxes and wild 5 Page 5 of 13
124
boars (P < 0.0001) and T. britovi LPG was observed. As the multiple regression analysis
125
revealed a multicollinearity with non-agricultural areas, agricultural areas was removed from
126
the regression models. No other relationship with land cover types could be observed. A positive correlation between non-agricultural areas (forests, scrubs, herbaceous vegetation
ip t
127
and pastures) of the fox home range and the T. britovi LPG (P < 0.0001) and a negative
129
correlation between the mean annual temperature of the fox home range and T. britovi LPG (P <
130
0.0001) were observed (i.e., the lower the mean temperature, the higher the T. britovi prevalence
131
in foxes). The percentage of non-agricultural areas and the mean annual temperature of the home
132
range of infected foxes were 61 ± 3% and 9.1 ± 0.1°C, respectively. The multiple regression
133
analysis (P < 0.005; P < 0.001) and the logistic regression analysis (P < 0.0001; P < 0.005)
134
confirmed these relationships . Although the best fit model was the logistic regression model
135
(classified correctly with P = 0.5: 98.2%; Hosmer-Lemeshow test: P = 0.81; area under ROC
136
curve: 0.76 ± 0.03), the overall fit of both models was satisfactory (P < 0.0001). Based on the
137
logistic regression analysis, the percentage of non-agricultural areas (95% CI of odds ratios:
138
2.219–19.516) and the mean annual temperature (95% CI of odds ratios: 0.477–0.832) were the
139
major determinants of the spatial distribution of T. britovi in foxes of Hungary. If the data of T.
140
britovi infected wild boars were included in the analyses, the results were very similar (P <
141
0.0001; classified correctly with P = 0.5: 97.2%; Hosmer-Lemeshow test: P = 0.65; area under
142
ROC curve: 0.79 ± 0.02; 95% CI of odds ratios: 3.879–23.407 and 0.488–0.772).
143 144 145
4.
Ac ce p
te
d
M
an
us
cr
128
Discussion
GIS was used only in one study focusing on the environmental determinants of the spatial
146
distribution of Trichinella spp. in Europe (Pozio et al., 2009). As the geographic coordinates of
147
Trichinella-free animals were not available in that study, only the spatial distribution and
148
environmental determinants of T. spiralis and T. britovi could be compared. In the present study,
149
the geographic coordinates of the Trichinella spp. negative foxes were also available; therefore, 6 Page 6 of 13
150
the environmental determinants of the spatial distribution of T. spiralis and T. britovi could be
151
analysed.
152
The prevalence of T. spiralis in wildlife increases when this parasite species is also circulating in the domestic habitat (e.g., Croatia, Romania, Serbia) (Pozio and Murrell, 2006).
154
In those countries (e.g., Austria, Hungary, Slovenia), where T. spiralis has been almost
155
completely eradicated from the synathropic cycle (Széll et al., 2012), it can be detected only
156
sporadically in wildlife (ITRC, www.iss.it/site/Trichinella). The statistical analysis confirms (P
157
< 0.0001; odds ratio: 24.1) that the distribution of T. spiralis in the Hungarian wildlife is
158
determined by the transborder transmission of the parasite from the neighbouring endemic
159
countries (i.e., Croatia, Romania, Serbia) by wild mammals (Széll et al., 2013) and not by
160
environmental factors. Hungary is surrounded by several border rivers (e.g., Dráva, Ipoly,
161
Danube, Tisza) coming from other countries. These rivers serve as ecological corridors for
162
wildlife and may play a significant role in the long distance spread of T. spiralis (Széll et al.,
163
2013). Moreover, these river valleys with dense vegetation, cool and humid microclimate and
164
high population density of reservoir hosts may also favour the long term persistence of T.
165
spiralis on both sides of these rivers (e.g., in the border region of Slovakia). The transborder
166
transmission and the possible long term persistence of the parasite in the environment of border
167
rivers is a significant risk for the reintroduction of T. spiralis to the domestic habitat in
168
Hungary.
cr
us
an
M
d
te
Ac ce p
169
ip t
153
The positive relationship between the prevalence of T. britovi in foxes and non-agricultural
170
areas observed in the present study can be explained by the different food resources of foxes in
171
the agricultural and non-agricultural areas (Pozio, 1998). In agricultural areas, foxes become
172
specialist predators due to prey profitability and focus predation on small rodents (Ferrari, 1995).
173
Moreover, foxes can feed on anthropogenic food resources in these areas. In non-agricultural
174
areas, foxes are generalist predators due to the shortage of food resources, which increases
175
scavenging and cannibalism (Kidawa and Kowalczyk, 2011), favouring Trichinella spp. 7 Page 7 of 13
transmission (Pozio, 1998). The relationship between low environmental temperature and T.
177
britovi infection can be attributed to the slower decomposition of animal carcasses, the increase
178
of scavenging of foxes (i.e., the lower the temperature, the higher the scavenging activity) and
179
the longer survival of T. britovi larvae at lower temperatures (Von Köller et al., 2001; Selva et
180
al., 2005; Sharanowski et al., 2008; Pozio et al., 2009). The present results explain the spatial
181
distribution of T. britovi in Europe. In wildlife, T. britovi is more widespread in the colder
182
northern regions with lower agricultural activities (e.g., Scandinavian and Baltic countries) than
183
in the central and southern regions of Europe (Pozio, 1998; Pozio and Murrell, 2006Pozio et al.,
184
2009). In the latter regions, Trichinella sp. infections are more prevalent among foxes living in
185
non-agricultural areas (e.g., mountains, protected areas, forested regions) (Pozio, 1998; Pozio et
186
al., 2009). Our results also confirm the findings of a North American GIS-based study
187
identifying temperature and land cover as the major determinants of the distribution of other
188
Trichinella spp. in wildlife (Masouka et al., 2009).
M
an
us
cr
ip t
176
te
d
189 References
191
Calenge, C., Maillard, D., Vassant, J., Brandt, S., 2002. Summer and hunting season home
192 193 194 195 196
Ac ce p
190
ranges of wild boars (Sus scrofa) in two habitats in France. Game Wildl. Sci. 19, 181–301.
Ferrari, N., 1995. Influence of the abundance of food resources on the feeding habits of the red fox, Vulpes vulpes, in western Switzerland. J. Zool. 236, 117–129.
Kidawa, D., Kowalczyk, R., 2011. The effects of sex, age, season and habitat on diet of the red fox, Vulpes vulpes, in northeastern Poland. Acta Theriol. 56, 209–218.
197
Masuoka, P.M., Burke, R., Colaccico, M., Razuri, H., Hill, D., Murrell, K.D., 2009. Predicted
198
geographic ranges for North American sylvatic Trichinella species. J. Parasitol. 95, 829–
199
937.
200
Pozio, E., 1998. Trichinellosis in the European Union: epidemiology, ecology and economic
8 Page 8 of 13
202 203 204
impact. Parasitol. Today 14, 35–38. Pozio, E., Murrell, K.D., 2006. Systematics and epidemiology of Trichinella. Adv. Parasitol. 63, 367–439. Pozio, E., Rinaldi, L., Marucci, G., Musella, V., Galati, F., Cringoli, G., Boireau, P., La Rosa,
ip t
201
G., 2009. Hosts and habitats of Trichinella spiralis and Trichinella britovi in Europe. Int. J.
206
Parasitol. 39, 71–79.
cr
205
QGIS Team, 2012. Quantum GIS. Accessed at: http://www.qgis.org.
208
Rinaldi, L., Musella, V., Biggeri, A., Cringoli, G., 2006. New insights into the application of
us
207
geographical information systems and remote sensing in veterinary parasitology. Geospat.
210
Health 1, 33−47.
213
M
212
Selva, N., Jedrzejewska, B., Jedrzejewski, W., Wajrak, A., 2005. Factors affecting carcass use by a guild of scavengers in European temperature woodland. Can. J. Zool. 83, 1591−1601. Sharanowski, B.J., Walker, E.G., Anderson, G.S., 2008. Insect succession and decomposition
d
211
an
209
patterns on shaded and sunlit carrion in Saskatchewan in three different seasons. Forensic
215
Sci. Int. 179, 219–240.
te
214
Staubach, C., Thulke, H.H., Tackmann, K., Hugh-Jones, M., Conraths, F.J., 2001. Geographic
217
information system-aided analysis of factors associated with the spatial distribution of
218 219 220 221 222
Ac ce p
216
Echinococcus multilocularis infections of foxes. Am. J. Trop. Med. Hyg. 65, 943−948.
Széll, Z., Marucci, G., Bajmóczy, E., Cséplő, A., Pozio, E., Sréter, T., 2008. Spatial distribution of Trichinella britovi, T. pseudospiralis and T. spiralis in red foxes (Vulpes vulpes) in Hungary. Vet. Parasitol. 156, 210−215.
Széll, Z., Marucci, G., Ludovisi, A., Gómez-Morales, M.A., Sréter, T., Pozio, E., 2012. Spatial
223
distribution of Trichinella britovi, T. spiralis and T. pseudospiralis of domestic pigs and
224
wild boars (Sus scrofa) in Hungary. Vet. Parasitol. 183, 393−396.
225
Széll, Z., Marucci, G., Pozio, E., Sréter, T., 2013. Echinococcus multilocularis and Trichinella 9 Page 9 of 13
226 227 228
spiralis in golden jackals (Canis aureus) of Hungary. Vet. Parasitol. 197, 393−396. Tolnai, Z., Széll, Z., Sréter, T., 2013. Environmental determinants of the spatial distribution of Echinococcus multilocularis in Hungary. Vet. Parasitol. 198, 292−297. Von Köller, J., Kapel, C.M.O., Enemark, H.L., Hindsbo, O., 2001. Infectivity of Trichinella
230
spp. recovered from decaying mouse and fox muscle tissue. Parasite 8, S209−S212.
ip t
229
Ac ce p
te
d
M
an
us
cr
231
10 Page 10 of 13
231
Table 1. Prevalence and intensity (larvae per gram) of Trichinella spiralis and T. britovi
232
infections in red foxes (Vulpes vulpes) (n = 3,304) and wild boars (Sus scrofa) (n = 0.29
233
million) collected in Hungary from 2006 to 2013. T. britovi
ip t
T. spiralis Prevalence (%)
Intensity
Prevalence (%)
(95% CI)
(± SE)
(95% CI)
Red foxes
0.24 (0.16−0.32)
3.2 ± 1.0
1.82 (1.59−2.05)
15.9 ± 2.5
Wild boars
0.004 (0.003−0.005)
33.2 ± 25.8
0.01 (0.008−0.012)
16.6 ± 2.7
us
cr
(± SE)
Ac ce p
te
d
M
an
234
Intensity
11 Page 11 of 13
235
Legend to figure
236 Figure 1. Map of Hungary showing animals sampled from 2006 to 2013. Panel A. Uninfected
238
and Trichinella spiralis infected red foxes (Vulpes vulpes) (circles) and wild boars (Sus scrofa)
239
(triangles). Panel B. Uninfected and Trichinella britovi infected red foxes (circles) and wild
240
boars (triangles).
cr
ip t
237
Ac ce p
te
d
M
an
us
241
12 Page 12 of 13
an
us
cr
ip t
A
M
241
Ac ce p
te
d
B
242 243
13 Page 13 of 13