Introduction
Angiostrongylus vasorum, commonly known as the ‘French heartworm’, is a metastrongylid nematode with an indirect life cycle. A wide range of gastropods serve as obligatory intermediate hosts for this parasitic nematode, and dogs and foxes are the typical definitive hosts (Rosen et al., Reference Rosen, Ash and Wallace1970; Guilhon & Cens, Reference Guilhon and Cens1973; Bolt et al., Reference Bolt, Monrad, Henriksen, Dietz, Koch, Bindseil and Jensen1992). The gastropod hosts become infected by foraging on the carnivore faeces containing first-stage larvae (L1). Such larvae-infested gastropods may transmit the infection when ingested by a carnivore, where the adult worm resides in the pulmonary artery and the heart (Rosen et al., Reference Rosen, Ash and Wallace1970; Guilhon & Cens, Reference Guilhon and Cens1973). In Denmark, recurrent epidemiological studies on fox and dog populations suggest that both the spatial distribution of A. vasorum and its prevalence are increasing (Willingham et al., Reference Willingham, Ockens, Kapel and Monrad1996; Saeed et al., Reference Saeed, Maddox-Hyttel, Monrad and Kapel2006; Koch & Willesen, Reference Koch and Willesen2007). However, information on the diversity of slug intermediate hosts and their distribution is lacking. Such information is central for the prevention of parasite transmission. The present study was therefore performed to investigate the spatial distribution of A. vasorum infected slugs in urban and peri-urban areas of Copenhagen.
Materials and methods
Study areas
Terrestrial slugs were collected from forests and parks in a 30 km radius of Copenhagen, Denmark, during the period from September to October 2007. Slugs were collected from six different localities: Frederiksberg Botanical Garden, Frederiksberg Park, Assistens Cemetery Park, Jaegersborg Forest and Deer Park, West Amager Forest and Rude Forest.
Among these study areas Frederiksberg Botanical Garden, Frederiksberg Park and Assistens Cemetery Park are regarded as public grassland parks with varieties of trees, herbs and ornamental vegetation. People use these areas for recreational purposes and to exercise their dogs and it is also possible that urban foxes visit these places at night in search for food. The peri-urban areas Jaegersborg Forest and Deer Park, West Amager Forest and Rude Forest are maintained as natural forest areas with open public access, and with a variety of trees and vegetation and a natural fauna. Jaegersborg Forest and Deer Park is populated by approximately 2000 semi-domesticated deer. All study areas have natural populations of foxes.
Collection and maintenance of slugs in the laboratory
Slugs collected by hand from the soil surface and vegetation were kept in plastic boxes according to locality, species and size. Species determination of slugs was done after Bondesen (Reference Bondesen1981), and sizes of the slugs were categorized, according to weight, into three groups (large, >10 g; medium, 5–10 g; and small, < 5 g). Boxes were lined with moistened tissue paper and supplemented with cucumber and lettuce as a food source. Slugs were kept at 15°C until the parasitological examination took place.
Parasitological procedures
Isolation of A. vasorum larvae from slugs was conducted by tissue digestion. Individual slugs were cut into small pieces (1–2 mm) and incubated at 37°C in artificial digestion fluid (12 ml HCl (30%) and 30 ml liquid Pepsin (1:30,000 IU) in 1 litre 42°C tap water) under constant magnetic stirring at 300 rpm for 35 min, strained through a 180 μm sieve and subsequently allowed to sediment for 75 min. The sediment was washed twice with water until the suspension became clear and recovered larvae were counted in a stereomicroscope. The larvae were fixed on an object glass by a drop of Lugol's solution and examined under the microscope at 20 × and 40 × magnification to determine the larval stage, as described by Ash (Reference Ash1970) and Rosen et al. (Reference Rosen, Ash and Wallace1970). Larvae that could not be identified as A. vasorum or could not be identified to a specific larval stage were excluded from the count.
Statistical analysis
Descriptive analysis of the explanatory variables (weight and species of the slugs and locality) and outcome variable (presence or absence of A. vasorum larvae) was performed by Fisher's exact test. Statistical analysis to evaluate the effect of weight, locality and species on the outcome variable was tested by using the logistic analysis procedure for binomial distribution. In logistic analysis, taking all explanatory variables into account, the most significant variable was selected by excluding the non-significant and/or least significant variable in a backward elimination strategy. P values of < 0.05 were considered to be statistically significant. All the statistical analyses were performed using SAS 9.1 package (SAS Institute Inc., Cary, North Carolina, USA) and Microsoft Excel 2000.
Results
From the six different study areas 48–50 slugs of up to four different species were examined for the presence of natural A. vasorum infection (table 1). Arion lusitanicus was the most common species in both park and forest areas, except in West Amager Forest. Arion ater and A. ater rufus were found exclusively in forest areas, and low numbers of Limax maximus were found at all localities.
Table 1 Angiostrongylus vasorum infection in naturally infected slugs in the Copenhagen area of Denmark.
Out of 298 examined slugs, 28 (9.40%) were infected with A. vasorum. Among infected localities the level of infection varied from 4% in Frederiksberg Park to 26% in Rude Forest. No naturally infected slugs were found in Frederiksberg Botanical Garden and Assistens Cemetery Park areas (table 1). Angiostrongylus vasorum was recovered from all of the four slug species but infection varied spatially. The highest overall prevalence was found in Rude Forest where all of the examined species, e.g. A. ater, A. ater rufus, A. lusitanicus and L. maximus were found to be positive for infection. In contrast only A. lusitanicus harboured the infection in Frederiksberg Park. In Jaegersborg Forest and Deer Park A. ater and A. ater rufus had comparable prevalence of infection. In West Amager Forest, two out of three L. maximus were infected (table 1). Only third-stage larvae (L3) were recovered from the slugs (ranged from 1 to 392 per slug). Highest larval burdens were found in A. lusitanicus and A. ater slugs; about 82% of the total infected slugs harboured < 10 larvae and only 14% harboured >100 larvae (table 1, fig. 1). No significant differences in larval burden could be demonstrated according to slug weight, although the highest burdens were found in larger specimens (figs 1 and 2). Logistic analysis for binomial distribution showed a significant association between the presence of natural A. vasorum infection and the locality from which slugs were collected (P = 0.04); however, the difference was not significant between species and size of the slug and presence of infection (P = 0.11 for species and P = 0.17 for size of the slug).
Fig. 1 Distribution of Angiostrongylus vasorum larvae in different slug species according to locality and body weight: (A) Arion lusitanicus, (B) A. ater, (C) A. ater rufus, (D) Limax maximus species. Frederiksberg Park (●), Rude Forest (○), Jaegersborg Forest and Deer Park (△), West Amager Forest (⋄).
Fig. 2 Distribution of Angiostrongylus vasorum larvae in different localities according to body weight of the slugs. Frederiksberg Park (●), Rude Forest (○), Jaegersborg Forest and Deer Park (△), West Amager Forest (⋄).
Discussion
Previous investigations have shown that A. vasorum can utilize a wide range of gastropods as intermediate hosts (Rosen et al., Reference Rosen, Ash and Wallace1970; Guilhon & Cens, Reference Guilhon and Cens1973; Sauerlander & Eckert, Reference Sauerlander and Eckert1974; Simpson & Neal, Reference Simpson and Neal1982; Barcante et al., Reference Barcante, Barcante, Costa Dias and Lima2003). Accordingly, the present study demonstrated the occurrence of natural infection of A. vasorum in four slug species, i.e. A. lusitanicus, A. ater, A. ater rufus and L. maximus, which supports observations by Guilhon & Cens (Reference Guilhon and Cens1973) and Tonsberg (Reference Tonsberg2006). Such lack of specificity in intermediate hosts has also been documented for other Angiostrongylus species, such as A. cantonensis (Wallace & Rosen, Reference Wallace and Rosen1969; Yousif & Lammer, Reference Yousif and Lammer1975b) and A. costaricensis (Rambo et al., 1997; Laitano et al., Reference Laitano, Genro, Fontoura, Branco, Siqueira, Maurer, Graeff-Teixeira, Chiaradia and Thomé2001).
The local variations in occurrence of slug species most likely rely on the diversity of the basic habitat, whereas the spatial differences in parasite prevalence in the slugs appears more complex and may depend on susceptibility, density and behaviour of the slug population, as well as on the quantity and infectivity of the larval population excreted by the local carnivore definitive hosts (Yousif & Lammer, Reference Yousif and Lammer1975a; Bolt et al., Reference Bolt, Monrad, Koch and Jensen1994; Saeed et al., Reference Saeed, Maddox-Hyttel, Monrad and Kapel2006; Weidema, Reference Weidema2006; Kozlowski, 2007). The most likely reasons for the absence of parasite-infected slugs in Frederiksberg Botanical Garden and in Assistens Cemetery Park are that the two areas are central urban parks with little faunal diversity and the low number of slugs examined. In spite of the observed local differences in larval burden among the slug species (table 1) a general pattern is not obvious due to the low numbers of specimens collected for some slug species, e.g. L. maximus. The few observations of a very high larval burden for L. maximus may reflect its long lifespan (up to 3 years, as compared to 1 year for Arion spp.) (Rollo, Reference Rollo1983) and the higher risk of exposure to larvae excreted in carnivore faeces. Although no significant association was found between slug weight and prevalence, the infection level tended to be higher in larger specimens than in the small slugs. A comparable association was also reported by Yousif & Lammer (Reference Yousif and Lammer1975a) for A. cantonensis in the snail Biomphalaria glabrata.
Overdispersion of parasites in their host populations, as described by Anderson & Gordon (Reference Anderson and Gordon1982), is evident from the present study. The majority of the slugs (270 out of 298) were not harbouring the infection, five slugs (14% of the infected slugs) had a high larval burden (>100 larvae), whereas 82% of the infected slugs (23 specimens) harboured < 10 larvae. Similar findings were reported in a field survey in Brazil on A. costaricensis (Laitano et al., Reference Laitano, Genro, Fontoura, Branco, Siqueira, Maurer, Graeff-Teixeira, Chiaradia and Thomé2001). Such overdispersion in the intermediate slug host may indicate that the definitive carnivore populations only rarely eat slugs with a high larval burden and thus many cases of dog angiostrongylosis may remain subclinical.
The recovery of exclusively third-stage A. vasorum larvae in the slugs supports previous observations (Tonsberg, Reference Tonsberg2006) where slugs collected over the summer (June–September) had an increasing proportion of A. vasorum L3 larvae (up to 100%). The collection period in the present study (September–October) is an extension of this period and it is therefore not surprising that all larvae recovered from the slugs had developed to the final, third stage. Overall, the present study illustrates that A. vasorum is widely distributed in several slug species in the forest and park areas of Copenhagen and that transmission of the infection is occurring via gastropods.
Acknowledgement
Hanne Rawat is appreciated for her competent technical work.
