Gastropods as intermediate hosts of Angiostrongylus spp. in the Americas: bioecological characteristics and geographical distribution

BACKGROUND Intermediate hosts are key organisms in maintaining parasite life cycles, because they can act as amplifiers in the transmission from natural reservoirs to humans. One of the most important groups of intermediate hosts for zoonotic nematode infections are gastropods,slugs and snails. These are essential organisms in the larval development of Angiostrongylus species. OBJECTIVES The objective of this paper is to review reports of Angiostrongylus spp. in naturally infected gastropods from the Americas, taking into account the diagnostic methods used in their identification, to be able to provide more accurate list of their intermediate hosts. We also discuss the factors that aid the dispersion of Angiostrongylus spp. in the Americas. METHODS This study reviews scientific publications and book sections on Angiostrongylus spp. in the Americas, including original works assessing larvae of Angiostrongylus in intermediate hosts. The eligible reports were classified accordingly to their geographical location, year of first record, and the larvae identification methodologies used. Digital repositories were used for the search. The bioecological characteristics of the main intermediate hosts are summarised. FINDINGS A total of 29 gastropod species that are naturally infected with Angiostrongylus spp. have been reported as intermediate hosts, 16 of which are land snails, two are freshwater snails, and 11 land slugs. MAIN CONCLUSIONS This study highlights the importance of integrative studies, analysing both the etiological agent and its transmission dynamic in the environment, the biological and ecological characteristics of the hosts, and the impact on host populations. It is necessary to increase interdisciplinary studies to determine the potential epidemiological health risk of angiostrongyliasis in the Americas, and thus be able to establish prevention, monitoring and contingency strategies in the region.

Angiostrongyliasis, caused by both Angiostrongylus cantonensis (Chen 1935) and Angiostrongylus costaricensis Morera and Céspedes 1971 is a zoonotic disease. Adults of A. cantonensis inhabit the pulmonary artery or right ventricle of the heart, while adults of A. costaricensis are found in the mesenteric arteries of the caecum of their definitive host (DH) (e.g., mustelids, procyonids, felids, canids, and mainly rodents). (7) DHs release first-stage larvae in faeces, and these utilise slugs and/ or snails as IHs. The gastropods are infected by ingestion or penetration of first-stage larvae; while DHs are infected by eating infected gastropods or by contact with their slime. Transmission can also involve ingestion of paratenic hosts, such as planarias, frogs, freshwater shrimps and crabs. (8,9,10) Species of Angiostrongylus have been recorded in a wide range of gastropods, indicating low host specificity for their His. (6,11,12) In the Americas, gastropod species have been reported harbouring Angiostrongylus spp., (13) but many reports are of experimentally infected gastropods that would be unlikely to act as IHs in nature. (12,14) In other cases, identification was performed using only morphological characteristics, but it is known that the larvae features in this genus are insufficient for identification at species level. It was therefore important to review the list of IHs and highlight the identification methods used for larvae determination.
In this study, we review reports of Angiostrongylus spp. in naturally infected gastropods from the Americas, taking into account the diagnostic methods used in their identification, to be able to provide more accurate Zoonoses are diseases transmitted from wild/domestic animals to humans. (1) Transmission to humans is via preparation or ingestion of infected food or through close contact with infected animals. (2) This dynamic is maintained a permanent focus of pathogen circulation in a specific geographical environment. (1,3) Although direct contact is not frequent between humans and natural reservoirs in wildlife, exposure to a pathogenic agent can occur through any intermediate host (IH) (e.g., invertebrates, fishes) used as bait or in consumption. (4) Intermediate hosts act as a bridge in the transmission of zoonotic pathogens, because they are amplifiers that maintain the transmission from natural reservoirs to humans. In consequence, monitoring possible IHs as pathogen transmitters is important in epidemiology, particularly for the prevention and control of emerging zoonoses. (5) Among the invertebrates, gastropods constitute one of the most important IHs groups for zoonotic nematode infections. In the case of Angiostrongylidae (Metastrongyloidea) slugs and snails are essential organisms in which the larvae develop. (6) list of their intermediate hosts. Also, we discuss the factors that aid the dispersion of Angiostrongylus spp. in the continent, mainly considering the bioecological characteristics, origin and geographical distribution of each gastropod species acting as an IH.

MATERIALS AND METHODS
The literature reviewed in this study includes scientific publications and book sections about Angiostrongylus spp. in the Americas. Original papers assessing larvae of Angiostrongylus in IHs have been included. The reports found, up to March 2020, were in Spanish, Portuguese and English using the descriptor words: Angiostrongylus sp, intermediate host, angiostrongyliasis. An exhaustive search was performed using the digital repositories Scielo, Redalyc, Scopus, Dialnet, Pubmed, and Google Academic. Personal communications at congresses and conference reports were not included.
The eligible reports were classified according to their geographical location, first year of record at this site, hosts recorded at the site, and the larvae identification methodologies used. To provide a complete characterisation, previously known data on DHs compiled in Valente et al. (15) were added to each Angiostrongylus species.
To understand nematode dispersion and to detect patterns, information on diet, microhabitat, environment and origin was summarised, among other bioecological characteristics of the main intermediate hosts.

RESULTS
Diagnosis of the larvae -Angiostrongylus larvae were recorded in different IHs and sites throughout the Americas. Several reports of Angiostrongylus L3 in gastropods were based only on molecular techniques (Tables I-II). Other reports resulted from the implementation of two methods of identification, some combining morphological and experimental methods, and others using both morphological and molecular techniques. However, many records were based only on morphological characterisation of Angiostrongylus L3.
Most reports of A. cantonensis were based on experimental and/or molecular methods, while most of the records of the other Angiostrongylus spp. larvae have only been based on morphological analyses.
In addition, on several occasions the susceptibility to infection of different gastropod species has been evaluated with positive results, as in the case of Biomphalaria spp. and Pomacea canaliculata (Lamarck 1822). (12,14,16) However, this study includes a list only of naturally infected IHs. Table III shows the bioecological characteristics of the 29 gastropod species reported as IHs of Angiostrongylus spp. and the ecological and environmental features associated with these.

Intermediate hosts and their bioecological characteristics -
Bradybaena similaris (Férrusac 1822), Achatina fulica Bowdich 1822, Sarasinula linguaeformis (Semper 1885) and Subulina octona (Bruguiére 1789) are the most frequent land gastropods reported as IHs of A. cantonensis in the Americas. Some of these species are considered crop pests due to their trophic habits and A.
fulica is known to also have coprophagic habits. (17) These gastropods are found in residential areas, constituting a link between the synanthropic definitive hosts and humans, indicating a beneficial scenario for the development of A. cantonensis. (17,31) Among the IHs, A. fulica has a wide geographical distribution in the Americas (see Table III). (9,18,19,20,21,22,23) This invasive gastropod has high reproductive rates and periods of estivation/hibernation, and as the most invasive species, its particular biological characteristics give it adaptive plasticity, and favour its role as the main IH for A. cantonensis. (5,24) Among freshwater IHs of A. cantonensis (Table III), only Pomacea lineata (Spix 1827) and Pomacea maculata (= P. insularum) Perry 1810 were reported. (25,26,27) The lack of records of Pomacea canaliculata (Lamarck 1822) as an IH is striking, because its distribution in Brazil overlaps that of P. maculata. P. canaliculata is found in Argentina, Brazil, Chile, Colombia, Ecuador, Paraguay, Uruguay, (25,27) and is the main freshwater IH of A. cantonensis in Asia. (28) In the case of A. costaricensis, the reports of naturally infected IHs are mainly associated with land slugs:  (29) ). These gastropods are considered as pests due their herbivorous habits. During estivation periods, some of them produce mucus around their bodies that isolate them from unfavourable conditions. (30,31) Several reports show a wide variety of IHs experimentally infected with A. costaricensis such as, B. glabrata (Say 1818), B. straminea (Dunker 1848), B. tenagophila (d'Orbigny 1835), Phyllocaulis boraceiensis Thomé 1972 and Megalobulimus abbreviatus (Bequaert 1948), and all these gastropod species are susceptible to develop infective larvae. (14,32,33) For the remaining Angiostrongylus species only two gastropods, B. glabrata and Polygra septemvolva (Say 1818), have been reported as intermediate hosts of A. schmidti Kinsella 1971, (11) and B.glabrata and P. canalicuta as IHs of A. vasorum (Bailliet 1866). (12,16,34) However, all these reports come from experimental gastropod infestations.
Most studies have focused on the gastropod species that are already known as IHs for Angiostrongylus. Therefore, it is possible that the number of IH species in the Americas has been underestimated.
Geographical distribution -Angiostrongylus larvae were recorded in 11 American countries, with Brazil being the country where most studies were performed. A. cantonensis is widely distributed in Central and South America (Table I), but there is no evidence that it was introduced from the western hemisphere by its DH (e.g., Rattus spp.) or its IH (e.g., the invasive gastropod A. fulica). (4,35,36) In contrast, A. costaricensis is

A. fulica
Morphological (60) not widely spread in the Americas, even though it is an endemic species (Table II (10,15) Their DHs are wild animals, and there are no reports of these Angiostrongylus species in synanthropic hosts from urban/agricultural areas. Furthermore, there are no records of their IHs, except for A. vasorum and A. schmidti from experimental research (Table II). Their zoonotic potential cannot be evaluated due to the lack of reports of their definitive and intermediate hosts. (10) Native and exotic gastropods -Exotic species can influence ecosystem health by introducing parasites from their native area, or by amplifying parasites already existing in the introduced range. (37) Some invasive species can impact human health by carrying zoonotic parasites. When an exotic parasite reaches a native host species, the phenomenon is called "spillover". When a native parasite infects an invasive host, leading to increased opportunities to infect autochthonous species, the phenomenon is called "spillback". (37) One of the causes that favours these interactions is the low host specificity of the parasite. (38) Examples of low host specificity in this work are A. cantonensis and A. costaricensis, since both exotic and native species are involved as IHs. In the first case, only four of the 18 IH species reported are exotic: A. fulica, B. similaris, Subulina octona and Paropeas achatinaceum (Pfeiffer 1846) (Table III). Achatina fulica is native to Africa, being observed in the Americas for the first time in Hawaii in 1936, and later it was reported in the rest of the continent, except Chile and Uruguay. (39,40,41) Bradybaena similaris is native to Asia, being observed for first time in the Americas in Brazil in 1835, and later it spread to almost all South America (Table III). (23,42) S. octona originated in Europe, being recorded in South America in 1914, (23) while whereas P. achatinaceum is native to Southeast Asia. (43) Angiostrongylus costaricensis is endemic to the Americas and almost the same number of exotic and of endemic gastropod species have been reported as IHs (6 vs 5) (Table III).
Thus, the dispersion of Angiostrongylus spp. in the Americas is favoured by low host specificity and the high number of naturally infected gastropod species.

DISCUSSION
The present study lists all the records and IHs identified as reservoirs of Angiostrongylus spp. larvae in the Americas. Reports were analysed, taking into account the diagnostic methods. Additionally, the bioecological characteristics, origin and geographical distribution of the IHs recorded in the literature were also summarised.
Currently, only two of the nine Angiostrongylus species distributed in the Americas have been confirmed as zoonotic agents (i.e., A. cantonensis and A. costaricensis). Nevertheless, the potential health risk of the rest of the Angiostrongylus species remains unknown. PCR-RFLP: polymerase chain reaction-restriction fragment length polymorphism; qPCR: quantitative polymerase chain reaction.
Identification of larvae of Angiostrongylus spp. by morphological methods is very difficult. The main morphological characteristics used by the authors are shape and body length, oesophagus length, excretory pore to the anterior end distance, genital primordium to posterior end distance, tail length, and tip of tail. (44) Unfortunately, the small size (400-600 µm) and the absence of diagnostic morphological characters developed of the infective larvae (L3) does not allow a good identification.
Fortunately, in recent years the diagnostics of parasites has been supported by experimental and molecular tools. (45,46,47) There are two types of experimental methods: in one case, captive mammals (e.g., Rattus sp.) are fed with a pool of live larvae (L3) obtained from naturally infected IH (e.g., A. fulica). After 28 days post infection, immature (L4-L5) and adult worms are found in the DH, of which accurate morphological identification is possible (e.g., Andersen et al. (46) ). In other cases, firststage larvae are isolated from the faeces of the DH. Gastropods are fed with a pool of isolated larvae (L1). After 30 days post infection, L3 that emerge from the IH are identified morphologically. (11,12,16) In the case of the molecular methods, gastropods are artificially digested according to the Wallace & Rosen (48) and Baermann-Moraes techniques. (15) After sedimentation, the material is analysed under a stereomicroscope and the larvae recovered from each collection pool are identified following DNA extraction and sequencing methods.
From 1977 to 2007, all the reports of A. cantonensis larvae were based on morphological and experimental methods, and they were recorded only in Central America. From 2007 to the present, its larvae were identified by molecular techniques, and were focused in South America. In contrast, all identifications of A. costaricensis larvae were carried out only by morphological methods. From 1999 until now, no new gastropod species have been added as IHs for Angiostrongylus spp., but probably the range of IHs has been underestimated. In the cases of A. vasorum and A. schmidti in the Americas, the larvae were obtained by experimental methods, presenting a similar eco-epidemiological scenario to that of A. costaricensis.
Host-parasite interactions are affected by genetic, biological and ecological aspects of both members of the relationship. Land gastropod species, such as A. fulica, B. similaris, Phyllocaulis spp., Sarasinula spp. and Belocaulus angustipes, show some bioecological characteristics that would favour Angiostrongylus spp. dispersion, such as polyphagia, peridomiciliary restriction, nocturnal habits, and estivation/ hibernation periods. (17) These IHs act as a link between synanthropic definitive hosts and humans. Additionally, unlike other species of gastropods, A. fulica is eaten raw by humans in many regions (e.g., Ecuador, Jamaica), increasing the risk of disease. (41) Environment plays an important role in the interaction between Angiostrongylus spp. and their IHs. Any disturbances in the structure and function of the ecosystem could indirectly influence zoonoses transmission. (49) The conversion of natural habitats to agricultural land, pastures and plantations due to the expanding human population has caused the displacement of wildlife and the wild/urban interface, causing more contact between humans and wild animals. (5,50) In this context, it is very important to continuously monitor different environments, to detect those changes that could favour the installation of a new focus of infection.    Introduction of exotic species could affect the equilibrium of an ecosystem by the invasion of new parasite species in native hosts, or amplify the parasites that already exist in the area. (37) The former seems to be the case in the introduction of A. cantonensis in the Americas, which has been able to invade at least 14 native gastropod species by "spillover". On the other hand, A. costaricensis, which parasitises native species, was able to invade exotic gastropods, also favouring its dispersion by "spillback". Apparently, both phenomena added to the low host specificity that these Angiostrongylus species seem to have, favouring the dispersion of these zoonoses in the Americas. The current paper demonstrates that the spread of A. cantonensis and A. costaricensis through gastropod IHs in the Americas is not limited to exotic or native origins. As has been stated, the success in obtaining larvae of Angiostrongylus spp. by infecting different freshwater gastropods denotes their low host specificity, (43,51) which constitutes a risk since it favours the dispersion of the species.
The information in the present review about the IHs of Angiostrongylus spp. complements the previous study of Valente et al. (15) on DHs, and sumarises all the available information in the Americas. In addition, this study highlights the importance of integrative studies, analysing both the etiological agent and the dynamic of its transmission in the environment, the biological and ecological characteristics of its hosts, and the impact on host populations. This is in agreement with the concept of "One Health" that states that human and animal health are interdependent and are linked to the environment in which they coexist. In this way, the etiological agent, the host and the environment constitute a triad that is influenced by continuous variations caused by environmental and social changes. (52) It is necessary incrase interdisciplinary studies to determine the potential epidemiological health risk of angiostrogiliasis in the Americas, and thus be able to establish prevention, monitoring and contingency strategies in the region.