Native and Nonindigenous Anurans and Lizards Are Intermediate Hosts for the Invasive Pentastome Parasite Raillietiella orientalis in Florida

PENTASTOMES are a parasitic group of crustaceans (Pare, 2008) that have been introduced during nonindigenous reptile and amphibian invasions and have become established in many geographic areas far from their native range. Invasive pentastomes include Raillietiella indica (also known as R. frenatus) and R. orientalis in both Australia (Barton, 2007; Kelehear et al., 2011, 2014) and North America (Pence and Selcer, 1988; Miller et al., 2018). The extensive rapid spread of raillietiellid pentastomes and their health impacts on reptile hosts (Pence and Selcer, 1988; Farrell et al., 2019; Bogan et al., 2022) makes them an emerging conservation concern.

Like most adult pentastomes, R. orientalis is a hematophagous endoparasite that resides in reptile lungs when mature (Kelehear et al., 2014; Miller et al., 2018). The definitive host is typically a snake (Christoffersen and De Assis, 2013), though it also has been found in large lizards in Florida (Fieldsend et al., 2021). Pentastome eggs are released in the lung of definitive host, swallowed, and then passed into the environment via host feces. Pentastomes typically exhibit heteroxenous life cycles with larvae infecting one or two intermediate hosts before the adult stage (Pare, 2008). Unfortunately, little is known about which species are used as intermediate hosts by R. orientalis. Laboratory infection studies indicate that roaches (Blaberus discoidalis) can serve as a first intermediate host, and anurans (Anaxyrus terrestris, Rana sphenocephala) and lizards (Anolis sagrei) can serve as a second intermediate host for R. orientalis (Palmisano et al., 2022). The consumption of R. orientalis-infected roaches results in visceral pentastomiasis (small pentastome larvae located in the viscera) in anurans and lizards that appears to have little impact on intermediate host health (Palmisano et al., 2022). Observations of R. orientalis in natural intermediate hosts are rare. Miller et al. (2018) observed a single larval R. orientalis in a shrew from Africa, and the only published observation of larvae in Australia or North America was in several Southern Toads (Anaxyrus terrestris) in Florida (Palmisano et al., 2022). Based on the diets of the Australia (Kelehear et al., 2014) and Florida (Miller et al., 2020) snake species that are most frequently infected by R. orientalis, anurans have been suggested to be important intermediate hosts. Developing an understanding of the intermediate hosts utilized by R. orientalis in invaded regions is critical to predicting the definitive host species likely to be threatened by parasitic infection and may also help predict the extent of geographic range expansion.

Chalkowski et al. (2018) discussed the impact of invasive host species on parasite–host relationships, emphasizing their role as disease facilitators: host species that enhance parasite abundance and distribution. For instance, Cane Toads (Rhinella marina) in Australia appear to act as strong disease facilitators for R. indica, as the invasion of Cane Toads in Australia is associated with rapid increases in the geographic range of R. indica (Kelehear et al., 2013). Similarly, in Florida the spread of R. orientalis may involve invasive species acting as disease facilitators. Florida is a major site for imports of live herpetofauna and is a hotspot for biological invasions, with at least 64 established nonindigenous herpetofauna species (Krysko et al., 2019). Many of these nonindigenous species are widespread and abundant and likely play an important role in the life cycle of invasive parasites in Florida, including R. orientalis. Recent studies indicate that two species of nonindigenous lizards, the Argentinian Black and White Tegu (Salvator merianae; Goetz et al., 2021) and the Tokay Gecko (Gekko gecko; Fieldsend et al., 2021), can serve as definitive hosts for R. orientalis, but the impact of the nonindigenous herpetofauna on other life stages of R. orientalis and their potential to serve as disease facilitators is unknown.

Snake dissections have documented the rapid spread of R. orientalis up to 650 km northwest of the initial discovery locations in Florida (Miller et al., 2020; Palmisano et al., 2023, 2025a) indicating that R. orientalis has suitable intermediate hosts throughout Florida. We studied frog and lizard species in central Florida to assess how widespread R. orientalis is in food webs and to improve our understanding of its life cycle. We sampled two study sites to determine the prevalence of infection with R. orientalis in both native intermediate host species and two extremely abundant nonindigenous species, the Cuban Treefrog (Osteopilus septentrionalis) and the Cuban Brown Anole (Anolis sagrei).

MATERIALS AND METHODS

Field collection of animals.—

We hand collected anurans and lizards from two sites where we had previously collected snakes infected by R. orientalis in Volusia County, Florida between November 2020 and August 2024. The Brandywine site (Site BW; 29°04′51.80″N, 81°19′18.07″W) was an urbanized site situated in the center of a residential area with homes and a few small (1–3 hectare) woodlots. The collection area was centered on a retention ditch that was a breeding site for several anuran species including large numbers of Southern Toads (Anaxyrus terrestris) and O. septentrionalis. From this site, we collected 263 individuals of six species of potential intermediate host species for R. orientalis including A. sagrei, A. terrestris, Green Treefrogs (Hyla cinerea), O. septentrionalis, Southern Leopard Frogs (Rana sphenocephala), and Eastern Spadefoot Toads (Scaphiopus holbrookii; sample sizes in Table 1).

Table 1.Observed prevalence (with 95% confidence intervals) and the mean and median infection intensity of R. orientalis in anuran and lizard species sampled at an Site BW. NC indicates a value was not calculated for a species.

This table shows the prevalence (column 2), the upper and lower 95% confidence intervals for prevalence (columns 3 and 4), the mean infection intensity and standard error (column 5), and the median infection intensity (column 5) for R. orientalis at study site BW for five host species. “NC” in the table indicates a value was not calculated for that species.

View Fullscreen

The second collection site, Lake Woodruff National Wildlife Refuge (Site LW; 29°06′24.63″N, 81°22′22.55″W), contained large expanses of freshwater wetlands and mesic forest and was 5.4 km northwest of Site BW. The habitats and diverse herpetofauna found at this site are described in more detail in Farrell et al. (2011). Prior research at Site LW (Farrell et al., 2019) indicated that by 2017 infections with R. orientalis were frequently observed in Pygmy Rattlesnakes (Sistrurus miliarius). We collected 185 frogs and lizards from this site representing members of eight abundant species including Green Anoles (Anolis carolinensis), A. sagrei, A. terrestris, H. cinerea, Southeastern Five-lined Skinks (Plestiodon inexpectatus), R. sphenocephala, and Little Brown Skinks (Scincella lateralis; sample sizes in Table 2).

Table 2.Observed prevalence (with 95% confidence intervals) and the mean and median infection intensity of R. orientalis in anuran and lizard species sampled at Site LW. NC indicates the value was not calculated for that species.

This table shows the prevalence (column 2), the upper and lower 95% confidence intervals for prevalence (columns 3 and 4), the mean infection intensity and standard error (column 5), and the median infection intensity (column 5) for R. orientalis at study site LW for six host species. “NC” in the table indicates a value was not calculated for that species.

View Fullscreen

Dissection techniques.—

We euthanized all anurans and lizards on the day of their capture and then either immediately dissected them or froze them until dissection. During dissections, we made incisions to expose the body cavity from throat to pelvis and examined the surface of internal organs, membranes, and the body wall using a dissecting microscope (8–20X magnification). We removed observed parasites, wet mounted them on a slide, and examined them under a compound microscope (50–400X magnification) to determine if the parasites were pentastome larvae. For each anuran and lizard, we recorded infection status (infected or not infected with R. orientalis) and infection intensity (the total number of larvae of R. orientalis observed in an infected individual).

Identification of pentastomes.—

We examined the extracted pentastome larvae microscopically (100 to 400X) and found their morphology to be consistent with Raillietiella (e.g., Ali and Riley, 1983; Bosch, 1986) and dissimilar to other pentastome genera found in North America. We used molecular techniques to confirm the identification of 14 larvae (nine from Site BW and five from Site LW) that were found in six host species (A. carolinensis, A. sagrei, A. terrestris, P. inexpectatus, R. sphenocephala, S. holbrookii). For three of these larvae, we used PCR and Sanger DNA sequencing of the 18S rRNA gene, as in Farrell et al. (2019), to determine if the larvae were R. orientalis. For 11 larvae, we used the COX1 gene target specific to Raillietiella orientalis with primers and PCR protocols from a recently developed conventional PCR assay (Palmisano et al., 2025b).

Statistical analysis.—

We calculated prevalence of R. orientalis (# infected individuals/total # individuals) for each host species at each of the study sites. For species with eight or more dissected individuals, we calculated the 95% confidence interval for prevalence using the Wilson method with a continuity correction (Newcombe, 1998). We used logistic regression (JMP v17, https://www.jmp.com/en_us/home.html) to determine if host infection status (infected or uninfected by R. orientalis) was influenced by host species, study site, or a host species by study site interaction effect. In this analysis, we restricted our focus to four species (A. sagrei, A. terrestris, H. cinerea, and R. sphenocephala) that were sampled at both sites. To determine if host species influenced intensity of infection with R. orientalis, we focused on the BW site where three species (A. sagrei, A. terrestris, and O. septentrionalis) had eight or more infected individuals. As the infection intensity data displayed a right-skewed distribution, we used regression analysis with a negative binomial distribution model error term and a logarithmic link function as suggested by Alexander (2012) to determine if infection intensity was associated with host species.

RESULTS

DISCUSSION

Our results indicate that R. orientalis is using at least five species of anurans and three species of lizards as intermediate hosts in central Florida. The appearance of larvae of R. orientalis in dissected hosts was identical to that observed in A. terrestris and A. sagrei from laboratory infection studies (see figure 2 in Palmisano et al., 2022). The eight species we observed carrying larvae of R. orientalis included members of four anuran families (Bufonidae, Hylidae, Ranidae, Scaphiopodidae) and two lizard families (Dactyloidae, Scincidae). The broad phylogenetic niche breadth of R. orientalis in terms of intermediate host use is similar to this parasite’s extremely broad phylogenetic range of definitive hosts, which includes at least five families of snakes, three families of lizards, and a species of tortoise (reviewed by Palmisano et al., 2025a). The ability of R. orientalis to utilize many definitive and intermediate host taxa is likely why it is a highly successful invasive species in terms of geographic range expansion and abundance.

Anurans, particularly A. terrestris and O. septentrionalis, often had high infection prevalence and infection intensity, indicating they are important intermediate hosts in the life cycle of R. orientalis in Florida. This finding is consistent with prior research that inferred frogs and toads were key intermediate hosts given the anuran-rich diets of the snake species most frequently infected by R. orientalis in both Australia (Kelehear et al., 2014) and Florida (Miller et al., 2020). The high prevalence of larvae of R. orientalis in lizards, including A. sagrei and P. inexpectatus, indicate this parasite has important intermediate hosts other than anurans and may be the primary reason why snakes with lizard-rich diets in more xeric habitats in Florida, including Black Racers (Coluber constrictor) and Coachwhips (Masticophis flagellum), have high prevalence of R. orientalis (Miller et al., 2020; Palmisano et al., 2023). Our survey results are also consistent with a prior laboratory infection experiment that indicated A. sagrei was easily infected with R. orientalis, wherein over 90% of Cuban Brown Anoles were infected after consuming part of a single pentastome-infected roach (Palmisano et al., 2022). Interestingly, R. sphenocephala was also easily infected with R. orientalis, but less than 3% of the individuals of this species that we field collected were infected (Palmisano et al., 2022).

The prevalence of R. orientalis was significantly influenced by both study site and host species. Site BW had higher prevalence than we observed at Site LW, but it is unclear what causes this difference in prevalence. We observed high prevalence in three anuran species (A. terrestris, O. septentrionalis, S. holbrookii) and two lizard species (A. sagrei, P. inexpectatus). Anaxyrus terrestris exhibited high prevalence at both sites, but A. sagrei had much lower prevalence at Site LW than Site BW, leading to significant site by species interaction. As R. orientalis is a trophically transmitted parasite, the variation in prevalence among sites and species is likely to be strongly influenced by local food web structure. Understanding patterns of prevalence in the herpetofauna that are the second intermediate hosts will require a detailed, site-specific understanding of both the diets of herpetofauna and the infection prevalence in the coprophagous insects that are probably the key first intermediate hosts in the life cycle of R. orientalis (Palmisano et al., 2022).

The infection intensity of R. orientalis was highly variable within the anuran and lizard species we sampled. There were no significant differences in mean intensity among A. sagrei, A. terrestris, and O. septentrionalis, the three most frequently infected species. Small larvae of R. orientalis (approximately 1 mm in length) are unlikely to pose a major health risk to the infected anurans and lizards. Prior experimental laboratory studies indicate that infections by larval R. orientalis have no observable impact on the growth rate and survival of A. sagrei and R. sphenocephala, even though these experimentally impacted animals had higher mean infection intensity than those we observed in the field collected animals (Palmisano et al., 2022). The impact of pentastomiasis appears to be much greater in the snakes that serve as the definitive hosts of R. orientalis where the large adult parasites (females 40–80 mm in length) apparently can cause morbidity and mortality (Farrell et al., 2019; Walden et al., 2020; Bogan et al., 2022).

Several recent studies (Farrell et al., 2019; Miller et al., 2020; Walden et al., 2020; Palmisano et al., 2023, 2025a) document the rapid spread of R. orientalis from south Florida locations, where Miller et al. (2018) initially documented this parasite in North America, to counties bordering Alabama and Georgia. Most recently, R. orientalis was found in the panhandle of Florida over 600 km from its presumptive area of establishment in Florida (Palmisano et al., 2025a). The high prevalence of R. orientalis in A. sagrei and O. septentrionalis, two nonindigenous synanthropic species that are extremely abundant throughout peninsular Florida (Krysko et al., 2019), may help explain this parasite’s rapid geographic range expansion. Cuban Treefrogs and Cuban Brown Anoles frequently undergo long-distance translocations via the horticulture industry (Glorioso et al., 2018), vehicular rafting (Campbell, 1996), and the pet trade and are likely to carry parasites with them. Both A. sagrei (Goldberg and Bursey, 2000) and O. septentrionalis (Ortega et al., 2015; Galt et al., 2021) appear to have aided long-distance invasions of a variety of pathogens and parasites; our data suggest they may be important dispersal hosts for R. orientalis as well.

The observed patterns of prevalence and infection intensity of R. orientalis have important implications for the definitive hosts, which are primarily snakes and larger lizards (reviewed by Palmisano et al., 2025a). Parasites with heteroxenous life cycles often have high host specificity for intermediate hosts and, therefore, are limited in abundance and geographic range by the availability of suitable intermediate hosts (Stein et al., 2010). This seems to be unlikely for R. orientalis in North America, as six of the seven genera that we found infected with R. orientalis, including Anaxyrus, Anolis, Hyla, Plestiodon, Rana, and Scaphiopus, are abundant, speciose, and have broad geographic ranges that span North America. The other infected genus, Osteopilus, is widespread in the Caribbean and rapidly expanding its geographic range in North America (Glorioso et al., 2018). At least one species of the intermediate host genera we identified as being infected in our study is sympatric with every species of North American snake found north of Mexico.

The intermediate host diversity for R. orientalis that we document in this study is likely a conservative estimate of all the intermediate host species that exist in Florida. Additionally, more hosts may continue to become available as the diversity of herpetofauna in Florida is rapidly growing with nonindigenous species establishing and expanding their initial ranges (Krysko et al., 2019). We sampled nine species of anurans and lizards, only 9.5% of the 95 species of frogs and lizards found in Florida. Furthermore, we do not know if mammals, birds, or a variety of other taxa can serve as intermediate hosts. The observation of an African shrew infected by a larval R. orientalis (Miller et al., 2018) and the observation that Eastern Diamondback Rattlesnakes (Crotalus adamanteus), a species that eats primarily mammals, are frequently infected (Metcalf et al., 2019; Andrew Durso, pers. comm.) are both indications that some mammals may also be important intermediate hosts. Our study did not investigate the insects that are probably the initial intermediate hosts for R. orientalis; the only information on insect infection comes from laboratory infection studies (Palmisano et al., 2023). A detailed understanding of the life cycle of R. orientalis will require further research involving a great diversity of potential host species.

Invasive species often act as disease facilitators (Chalkowski et al., 2018), and for infections with R. orientalis in snakes, this is likely to be the case for A. sagrei and O. septentrionalis. Both species are often eaten by many species of snakes in Florida (Meshaka, 2001; Meshaka et al., 2004) and, given the high prevalence and infection intensity of R. orientalis we observed in these species, they are likely to cause severe pentastomiasis in the snake species that consume them. The low intermediate host specificity and high prevalence and infection intensity we observed in R. orientalis indicates that this invasive parasite is currently an important conservation concern in the southeastern United States and has substantial potential to become a problem in other regions.

DATA ACCESSIBILITY

Unless an alternative copyright or statement noting that a figure is reprinted from a previous source is noted in a figure caption, the published images and illustrations in this article are licensed by the American Society of Ichthyologists and Herpetologists for use if the use includes a citation to the original source (American Society of Ichthyologists and Herpetologists, the DOI of the Ichthyology & Herpetology article, and any individual image credits listed in the figure caption) in accordance with the Creative Commons Attribution CC BY License.

AI STATEMENT

The authors declare that no AI-assisted technologies were used in the design and generation of this article.