In 1946, following a mass mortality event (MME) of oyster stocks in Louisiana (Gulf of Mexico, United States), Mackin et al. (1950) identified a protist present in host tissue sample as the causative agent. First affiliated to a fungal lineage and named Dermocystidium marinum (Mackin et al., 1950), electronic microscopy observations then showed morphological characteristics, such as the presence of a subpellicular membrane, micropores, and a conoid-like structure (zoospore stage) suggesting a re-affiliation of this protist into the Apicomplexa phylum (Alveolata) (Perkins and Menzel, 1967; Perkins, 1976). While doubts remained about the accuracy of this affiliation (Siddall et al., 1997), molecular phylogenies based on ribosomal and actin sequences revealed that Perkinsus species were more closely related to dinoflagellates than apicomplexans (Reece et al., 1997). In 2003, multiple protein phylogeny indicated that Perkinsus is an early branch in dinoflagellate lineage (see schematic representation in Figure 2A; Saldarriaga et al., 2003).

Nowadays, seven species are described within the genus Perkinsus: P. marinus, P. olseni, P. qugwadi, P. chesapeaki, P. mediterraneus, P. honshuensis, and P. beihaiensis (see Figures 1B, 2B for the phylogenetic classification of these parasites). However, only P. marinus and P. olseni, the etiological agents of “Dermo” disease and Perkinsosis, respectively, have significant negative impacts on mollusk populations worldwide (Siddall et al., 1997). These two infectious agents are today within the list of notifiable diseases in the World Organization for Animal Health (OIE) (Oie-Listed Diseases, 2021: OIE—World Organisation for Animal Health).

Some Perkinsus species have a putative BHR among mollusk species as, for example P. olseni infecting various clams (e.g., R. decussatus, R. philippinarum, Austrovenus stutchburyi, Tridacna maxima, T. crocea, and Pitar rostrata), oysters (Crassostrea rhizophorae, Saccostrea sp.), pearl oyster (e.g., Pinctada imbricata and P. fucata), and abalone (e.g., Haliotis rubra and H. laevigata) (Lester and Davis, 1981; Cremonte et al., 2005; Dungan et al., 2007; Sheppard and Phillips, 2008; Sanil et al., 2010; Ramilo et al., 2015; Pagenkopp Lohan et al., 2018), or P. chesapeaki, which can infect clams (R. decussatus, R. philippinarum) and oysters (C. rhizophorae, C. virginica) (Coss et al., 2001; Arzul et al., 2012; Dantas Neto et al., 2016). Conversely other Perkinsus species show an intermediate host range as, for example, P. marinus infecting mostly oysters (e.g., C. agar, C. rhizophorae, C. virginica, C. agar, and Saccostrea palmula) (e.g., Enríquez-Espinoza et al., 2010; Cáceres-Martínez et al., 2012; da Silva et al., 2013, 2014).

Until now, most of the host species described above are of commercial interest but other bivalve and gastropod species may also be susceptible to infection, especially in the described geographical range. However, the Perkinsus putative wide host range established solely using microscopic and molecular methodology may also be overestimated. Indeed, these results do not allow the identification of dead-end hosts that prevent the parasite transmission to the definite hosts. Furthermore, despite their BHR character, host susceptibility needs to be considered to prevent and protect most sensible bivalve stocks. For example, the two genetically related oysters C. gigas and C. virginica respond with different susceptibility when challenged by P. marinus. Although infection is established in both hosts, only infection in C. virginica is lethal (Barber and Mann, 1994). Among Perkinsus species, different levels of prevalence depending on hosts species are recorded across the world (Figure 3). In the Chesapeake Bay, P. marinus and P. chesapeaki infect different species of oysters and clams living in sympatry. Among these hosts, P. marinus rarely infects clams compared with P. chesapeaki and exhibits oyster preference (Reece et al., 2008). Variability in pathogenicity across the BHR of Perkinsus spp. raises the question about the existence of specific strains. For example, such specific strain was recently highlighted for Perkinsus marinus with the emergence between 1983 and 1990 of a new hypervirulent phenotypic strain that includes a shortened life cycle and a trophism shift from deeper connective tissues to digestive epithelia (Carnegie et al., 2021). Authors hypothesized that the development of this new strain may be related to reduced oyster abundance and the rapid establishment of the exotic parasite H. nelsonii in 1959. Hence, given the increase in epizootic diseases (= disease event in a non-human animal population analogous to an epidemic in humans) worldwide, consideration of the host range is absolutely essential because i) they may be the starting point for a host shift, ii) they represent vector of transmission with no apparent signs of disease, and iii) new native parasitic strains can emerge in response to biotic or abiotic stressors.

Perkinsus infection appears to occur directly by filtration without an intermediate host, with gills and labial palps playing a crucial role as entrance portal for the parasite (Chintala et al., 2002; Allam et al., 2013; Wang et al., 2018). Four different life stages (Figure 4A) are described today occurring inside or outside the host with an exception for P. qugwadi where all stages can be observed in host tissues (Blackbourn et al., 1998). All of them can induce new infections into a healthy host. The trophozoite life stage proliferates by vegetative multiplication (palintomy) in the host tissue. Disruption of the cell wall allows the release of immature spherical trophozoites that will gradually enlarge becoming mature vegetative cells (Perkins, 1996). In case of heavy infection, trophozoites can invade totally host tissues, with occasional production of cutaneous white nodules (inflammatory reaction), and induce a global decrease in host fitness (e.g., lower filtration activity, retard in growth and reproduction) (Dittman et al., 2001; Lee et al., 2001; Choi and Park, 2010). When the host becomes moribund, the parasite enlarges and develops a thick cell wall becoming a hypnospore life stage (Valiulis and Mackin, 1969; Perkins, 1996). When transferred to fresh seawater under favorable conditions of temperature and salinity (Auzoux-Bordenave et al., 1996), hypnospores undergo multiple divisions leading to the formation of free-living biflagellate life stages, the zoospores, into a cellular structure called zoosporangium. At maturity, the zoosporangium releases zoospores in the medium allowing the infection of new hosts.

Perkinsus species are one of the most famous examples of successful invasive pests in marine environments (Figure 3A). Three different scenarios are described through three different Perkinsus species: (1) Expansion: P. marinus has shown a significant increase in its geographic area in the United States along the East coast by more than 500 km in 2 years (1990–1992). This expansion, which is now permanent, has been correlated with an increase in winter surface temperature due to milder winters (Ford, 1996; Ford and Smolowitz, 2007). (2) Introduction: P. chesapeaki appears to have been accidentally introduced in Europe via its vectors Mya arenaria, or the hard clam, Mercenaria mercenaria from the United States (Arzul et al., 2012). The sporadic detection of P. chesapeaki does not reflect an expansion dynamic. To date, no mortality event affiliated to P. chesapeaki has been recorded. (3) Introduction and expansion: P. olseni has been co-introduced in Europe with its host the Manila clam, Ruditapes philippinarum, in 1972 for the development of clam aquaculture (Ruano et al., 2015). After its first detection in the late 1980s, the first mortalities of clam stocks were attributed to this parasite at Ria de Faro in Portugal (Ruano and Cachola, 1986). Since then, P. olseni has produced mortalities in Spain, Portugal, and Italy, infecting both the exotic Manila clam, Ruditapes philippinarum, and the native European clam, Ruditapes decussatus (Azevedo, 1989; Figueras et al., 1992; Pretto et al., 2014). Following these first mortalities, P. olseni was detected along the French Atlantic coast with infection prevalence up to 100% (Lassalle et al., 2007), in contrast to the non-detection of Perkinsus spp. in most of (these) previously studied sites (Goggin, 1992). Similarly, histological studies conducted by Vilela (1951) on Minchinia tapetis (Haplosporidia) infection in the Formosa lagoon in Portugal, where the prevalence of P. olseni infection is now very high, ranging from 60 to 90% depending of the season (Leite et al., 2004), did not mention any infection with Perkinsus spp. Finally, in 2011, using microsatellite genetic diversity analysis, Vilas et al. described low genetic diversity of the P. olseni parasite in Spain and Portugal compared with samples collected in Japan and New Zealand (Vilas et al., 2011), which might result in a founder effect testifying an introduction event of this parasite. These examples of Perkinsus spp. dissemination demonstrates the ubiquitous nature (exploitation of an important range of abiotic and biotic niches) of Perkinsus species, which is a threat for shellfish farming economy and for ecosystem equilibrium. Bivalves are known to provide important ecosystem services through their status as habitat engineers, as filter feeders, and as a connecting compartment between primary producers and predators (Jones et al., 1996; Coen and Grizzle, 2007; Gallardi, 2014). By cascade effect, parasitic infection can have deleterious effects on the different functions attributed to these populations: (i) by altering their bioturbatory activity and, thus, by modifying biogeochemical cycles at the water–sediment interface (Dairain et al., 2019), (ii) by impacting their filtration activity participating in the nitrogen cycle (= production of NH₄⁺ leading to an increase in primary production) and helping to reduce turbidity (due to the increase in light penetration) (Gallardi, 2014), and (iii) by altering their physiological traits (e.g., growth, reproduction, and survival), which will modify the structure of habitat, biotic interaction, and the associated species richness (e.g., Thomas and Poulin, 1998). In addition to this infectious threat, benthic biodiversity is threatened by human activities (McCauley et al., 2015). Indeed, anthropogenic eutrophication and temperature increase are all factors that can weaken the immune system of the benthic organisms and lead to a decrease in their resistance to parasitic protists as Perkinsus spp. (Morley, 2010). Furthermore, the host range of these parasitic organisms might be wider than described in the scientific literature with the existence of other “non-commercial” potential hosts belonging to suborders infected by Perkinsus spp. Thus, even though many arguments allow classifying the Perkinsus species as a BHR parasite, we cannot determine the real consequences on stocks of “non-optimal” hosts especially when it is related to sporadic infections with low prevalence. Due to research focus on economically valuable species of mollusks like C. virginica or R. philippinarum, consequences of these infections on the global host range of Perkinsus spp. are still mainly undiscovered.

Norén et al. (1999) identified small round structures abnormally present within the toxic dinoflagellate Dinophysis spp. on the west coast of Sweden. Molecular and microscopic analyses revealed that this parasitic organism was affiliated to the Perkinsids group and erected a new phylum, the Perkinsea (Norén et al., 1999). So far, five genera have been described and cultivated (Jeon et al., 2018). The first genus described was Parvilucifera genus encompassing P. infectans (Norén et al., 1999), P. multicavata (Jeon et al., 2018), P. sinerae (Figueroa et al., 2008), P. rostrata (Lepelletier et al., 2014b), P. corolla (Reñé et al., 2017b), P. catillosa, and Parvilucifera sp. (Alacid et al., 2020). Recently, three new genera were described, the genus Snorkelia containing P. prorocentri (Leander and Hoppenrath, 2008) renamed Snorkelia prorocentri (Reñé et al., 2017a), the genus Dinovorax including D. pyriformis (Reñé et al., 2017a), Tuberlatum encompassing T. coatsi (Jeon et al., 2018), and finally Maranthos including M. nigrum (Reñé et al., 2021a). All these genera seem to have a basal position within the Parviluciferaceae group, with the exception of the Maranthos species, whose phylogenetic position is not yet clear (Jeon et al., 2018; Reñé et al., 2021a; Figures 1B, 2).

Members of the Parviluciferaceae family share many common traits including development and characteristic of their life cycle, and their infection strategy (Reñé et al., 2017a). The infection begins with the entry of the zoospore into the dinoflagellate host cell (Figure 4B). Once inside, the parasite develops in the cytoplasm forming a spherical cell, the trophont (= trophocyte), which consumes its host for its own growth. At the end of the feeding stage, a free-living non-motile cell, the sporocyte (= immature sporangium), occupying the whole intracellular space, is released by breaking the theca of the host. Next, the cell undergoes morphological transformations into a spherical multinucleated dark structure called the sporangium (= mature sporocyte). The sporangium remains in dormancy until the detection of a host chemical signal (Garcés et al., 2013b). After a short maturation (∼48 h), a sporangium with open aperture(s) releases several hundred new infectious cells (zoospores) (Alacid et al., 2017). To date, all described species are parasites of dinoflagellates including toxic species and possess a BHR (here belonging to a “true” generalist class) (Figueroa et al., 2008; Garcés et al., 2013a; Jeon et al., 2018; Rodríguez and Figueroa, 2020; Reñé et al., 2021b) except S. prorocentri, isolated from Boundary Bay (Canada), which was known to only infect the marine benthic dinoflagellate Prorocentrum fukuyoi (Leander and Hoppenrath, 2008; Figure 1B). However, in 2017, Reñé et al. isolated a new species of Snorkelia from a different location (Catalan Coast, NW Mediterranean) infecting the dinoflagellate Levanderina fissa (Reñé et al., 2017a). The host range of Snorkelia species, therefore, remains enigmatic. For the “true” generalists, P. sinerae, P. corolla, and P. rostrata, the host range was determined mainly by in vitro cross-infection. In vitro experiments could result in artificial linkage between a host species and a parasite, which is not representative of the ecological reality (Råberg et al., 2014; Alacid et al., 2016; Reñé et al., 2021b). Infection of non-preferred hosts could be based on chemical and physiological processes shared by several related hosts (e.g., Garcés et al., 2013b). Furthermore, when a parasite infects a host but is not able to produce viable parasitic cells, such “host” cannot be considered as a “true” host for the parasite, e.g., infection of a chlorophyte strain of Pyramimonas by P. corolla (Rodríguez and Figueroa, 2020). The infection process is probably the result of plesiomorphic mechanisms among P. corolla’s hosts, but the parasite cannot achieve its whole life cycle. In this review, we considered Parvilucifera species as generalists when they realize a viable life cycle in a large repertoire of host species. A BHR gives them the full potential of host shifting in new environments, although there are evidences of host preference in nature (Alacid et al., 2016). However, much remains to be done to evaluate the contribution of the plesiomorphic and convergent traits of a parasite to the success of generalists.

Despite a “hot-spot” of detection in Europe, the distribution of these parasites of dinoflagellates is worldwide in marine environments (Figure 3B). As major contributors, with diatoms, to the fixation of inorganic carbon through photosynthesis (Falkowski et al., 1998), dinoflagellates are one of the most important components of marine phytoplankton as primary producers and grazers. Parvilucifera parasites could participate in regulating dinoflagellate populations in the blooming period, as shown in Spain where 5 to 18% of the population of the noxious dinoflagellates, Alexandrium minutum, were killed by Parvilucifera sp. during natural bloom (Alacid et al., 2017). For methodological limitation reasons, most studies describing Parvilucifera spp. have been carried out when prevalence is the highest on blooming host species, thus, increasing the chances to detect and describe new parasitic species. Unfortunately, this bias hides a possible important part of host and parasite diversity (Reñé et al., 2021b). Indeed, in some studies, the host range of these parasites appears to be much wider, including non-harmful species (e.g., Figueroa et al., 2008; Garcés et al., 2013a; Rodríguez and Figueroa, 2020), even if the in vitro specificity tests do not take into account the complexity of planktonic interaction network and the in situ parasitic host preferences. This lack of knowledge on Parvilucifera distribution, diversity, and trophic niche might become problematic, given its BHR nature and the global trades. Indeed, ballast waters, which are one of the most important vectors of introduction in marine systems (Carlton, 1985; Barry et al., 2008), could contribute to the dispersal of Parvilucifera trophonts or dormant sporangia (e.g., Darling et al., 2018), thus, increasing the risk of settlement in new areas. Hence, introductions of these BHR parasites could act positively as a top–down control of the toxic dinoflagellate blooming species and, conversely, could play a key role in deregulating the planktonic compartment given the importance of dinoflagellates as primary producers inducing significant shifts in marine food webs.

In 1969, Brooks et al. of two flatfish species, the black-tipped flounder, Psettichthys melanostictus, and the Pacific plaice, Platichthys stellatus (Brooks et al., 1969). Thirty-five years later, a protozoan named X-cell was identified as responsible for this pathology inducing also multiple lesions and swelling of the gill filaments (Miwa et al., 2004). Using small and large concatenated subunit (SSU and LSU) rDNA phylogeny reconstruction, Freeman et al. (2017) investigated the phylogenetic position of this enigmatic X-cell protist and showed that it formed two highly distinct clades, one corresponding to the pseudobranchial parasites of Gadiformes, and the other to gill and epidermal X-cells from Perciforms and Pleuronectiformes called Gadixcellia and Xcellia, respectively. More recently, Karlsbakk et al. (2021) described a new genus named Salmoxcellia, a sister group to Gadixcellia. Despite an unusually high genetic divergence between the three genera, with a similarity of 74.9% for the SSU rDNA sequences, these two close sister clades form a new family called Xcellidae (or Xcellins) branching within the Perkinsea lineage (Freeman et al., 2017; Karlsbakk et al., 2021) (Figure 2B).

Histological examination and scanning electron microscopy of xenomas of infected fish revealed a large number of clustered round parasitic cells surrounded by host connective tissue. Currently, only one developmental life stage has been described up to today; hence, the life cycle of this parasite still remains unknown (Freeman et al., 2017). However, Freeman et al. (2017) suggested that infection occurs via contact between fish and the benthos. Indeed, X-cell infection occurs in fish species with at least one benthic stage during their life cycle (Fahay, 1983; Lough et al., 1989). An experimental infection in situ in tanks revealed that only fish exposed, even transiently, to the benthos were infected and showed symptoms of disease (Eydal et al., 2010). The infected fish samples are mainly young and adult individuals. Indeed, the first pathogenic symptoms (pseudobranchial xenomas) appear macroscopically in young wild cod (around 6 months old, 6- to 13-cm length) from Icelandic waters with a prevalence peak of 23% at the age of 22 months, whereas older or larger fish (40- to 76-cm length) have a lower prevalence around 7 and 1%, respectively (Eydal et al., 2010). Similar trends had been observed in Atlantic cod (Morrison et al., 1982) and the Pacific cod (Stich et al., 1976). However, the question remains open whether the mortality induced by the X-cell infectious agent on juveniles could remove an age class from the field sampling and bias the observed prevalence. Although mortality induced by these infectious agents still needs to be demonstrated, the transfer of wild infected individuals of Gadus morhua in tanks has shown an important mortality (Eydal et al., 2010).

Parasites belonging to these three genera have been detected in more than 20 fish species belonging to five orders of teleost: Pleuronectiformes (flatfishes) (Freeman, 2009; Freeman et al., 2011), Perciformes (perch-like fish) (Katsura et al., 1984), Gadiformes (cods) (Freeman et al., 2011), Siluriformes (catfishes) (Diamant et al., 1994), and Salmoniformes (salmonids) (Dyková et al., 1993; Karlsbakk et al., 2021). Until now, five species have been described (Figure 1B). The parasite Xcellia pleuronecti infection has been confirmed in Hippoglossoides dubius and Pseudopleuronectes obscurus but also suspected in many other species from the same area: Cleisthenes herzensteini, C. pinetorum, Glyptocephalus stelleri, Kareius bicoloratus, Hippoglossoides elassodon, Liopsetta pinnifasciata, Platichthys stellatus, Parophrys vetulus, Pseudopleuronectes schrenki, and Verasper moseri. Equally, Xcellia lamelliphila infection is detected in Limanda limanda, Lycodes spp., Macruronus novaezelandiae, Merluccius gayi gayi, and Trematomus spp. (Freeman et al., 2017). Finally, the newly described parasite Salmoxcellia vastator is detected in Oncorhynchus mykiss and in Salmo salar (Karlsbakk et al., 2021). These three parasites are, therefore, BHR parasites (Figure 1B). Conversely, Xcellia gobii and Gadixcellia gadi infecting, respectively, Acanthogobius flavimanus and Gadus morhua are NHR parasites and classified as specialists (Freeman et al., 2017; Figure 1B). This classification will certainly evolve as new knowledge on Xcellidae group becomes available.

In addition, Xcellidae protists have been detected in a restricted geographical area (Northern Europe and Japan) (Figure 3C), but the monitoring of these X-cell infections is hard to carry out and stay largely incomplete because of migratory fish hosts in the worldwide ocean. Their host range is mainly unknown, but we can hypothesize that it may be broader than that currently identified. An extended host range would not be surprising considering the Perkinsea phylum constituted of a majority of BHR parasites closely related phylogenetically (e.g., the Perkinsidae) (Figure 2B). Overall, the consequences of the X-cell infection on fish populations are not well known. However, as it affects important halieutic resources as described for the salmonid parasite, Salmoxcellia vastator, it makes farmed salmonid fillets unsuitable for sale (Karlsbakk et al., 2021). Moreover, these parasites could infect vulnerable species, such as Atlantic cod fish G. morhua, extensively farmed, but mainly in the north of Europe (see FAO, 2021 Programme—Gadus morhua). Commercially important fish, a concern of the global trades, which could act as reservoirs allowing the dissemination in farmed and wild fish populations.

With up to 50% of all threatened species, amphibian populations are emblematic representatives of the sixth mass extinction event to which highly virulent wildlife diseases contribute (Stuart et al., 2004; Gewin, 2008; Chambouvet et al., 2020). Recent works have mainly identified two pathogens as drivers of this decline, the fungus Batrachochytrium dendrobatidis and the Ranavirus (e.g., Bosch et al., 2001; Fisher et al., 2009; Gray et al., 2009; Olson et al., 2013).

However, in April 2006, in a pond in northeast Georgia (United States), Davis et al. (2007) observed a massive mortality event of southern leopard tadpoles, Rana sphenocephala (syn. Lithobates sphenocephalus), attributed to an unknown protist affiliated by phylogenetic analysis to the Perkinsea lineage. Using phylogenetic analysis, it has been shown that this organism belongs to a discrete clade named pathogenic Perkinsea clade (PPC) (Isidoro-Ayza et al., 2017) within the wider monophylic cluster of the Novel Alveolate Group 01 (NAG01) (Chambouvet et al., 2015) obtained from disparate freshwater environments or internal organs (e.g., liver tissues) of a wide variety of Neobatrachia suborder (Chambouvet et al., 2015; Isidoro-Ayza et al., 2017; Figures 1B, 2B). Infectious agents of the PPC clade have been identified as responsible for the die-offs of tadpole throughout the United States (Isidoro-Ayza et al., 2017). Infected tadpoles with pathological symptoms, named “severe Perkinsea infection” (SPI), showed lethargic swimming with enlarged and histopathologic lesions of the liver, mesonephros, spleen, pancreas, gills, gastrointestinal tract, skeletal muscle, dermis, and peritoneum (Green et al., 2002; Davis et al., 2007; Jones et al., 2012; Isidoro-Ayza et al., 2017). Histological examination of infected liver tissues revealed a massive number of round infective cells replacing the normal hepatic parenchyma. Two distinct putative life stages were identified invading the hepatocyte cells corresponding to hypnospore-like and trophozoite-like life stages (Isidoro-Ayza et al., 2017, 2019). SPI symptoms have been reported from summer to early autumn in boreal and temperate regions, and from late winter to early spring in subtropical areas (Isidoro-Ayza et al., 2017; Figure 3D). Although this parasite is mainly described as infecting tadpole life stage, one report highlights infection of the adult populations with granulomatous lesion in the legs (Jones et al., 2012). These results suggest that either most of the tadpoles die before the metamorphosis or that mature immune systems acquired after the metamorphosis may drive back the infection (Isidoro-Ayza et al., 2017). Although the relationship between the infectious agent and the disease is not yet well established, symptoms of the disease could result from co-infection between the Perkinsea parasite and others infectious agents, such as the FV3-like virus, responsible for frog population declines, which infects both larval and adult amphibians (Gray et al., 2009; Lesbarrères et al., 2012), and/or alteration of tadpole immune systems (Isidoro-Ayza et al., 2017).

During monitored SPI outbreaks, mortality rate can reach up to 95% of the population leading to the loss of an entire age class or, in case of chronic infection, to a reduced recruitment (Green et al., 2002; Isidoro-Ayza et al., 2017, 2019). Now recognized as the third most common infectious disease of anuran species, SPI infectious agent is described as an emerging pathogen. It is therefore now fundamental to understand the relationship between this infectious agent and the outcome of the disease and how other pathogen communities and/or environmental factors could affect disease susceptibility. Furthermore, as the globalization of trade where amphibian species, generally involved in meat or pet trade, could spread the parasitic invaders via reservoir species into native and/or naive amphibian populations, the study of this pathogen is crucial.

Molecular methodologies have recently highlighted that our vision of Perkinsea diversity, mainly assessed by culture-based methods, is still only the tip of the iceberg. In both marine and freshwater environments, analysis of genetic diversity based on SSU rDNA sequences of the smallest sizes of plankton (<5 μm), revealed a previously unknown diversity of these organisms (Moon-van der Staay et al., 2001; López-García et al., 2001, 2003; Lefranc et al., 2005; Mangot et al., 2009).

A recent overview of freshwater Perkinsea genetic diversity, especially in lakes, revealed that this group appears to be diverse and abundant, with an observation of 2,927 operational taxonomic units (OTUs) defined at 95% of sequence similarity, which represented 3% of the eukaryotic diversity analyzed in Debroas et al. (2017), suggesting an important role in the trophic web (e.g., Lepère et al., 2008, 2011; Jobard et al., 2020). Recent studies have highlighted that freshwater clades could be a parasitic protist of the colonial green algae Sphaerocystis sp. (Jobard et al., 2020). These results are consistent with the description of Rastrimonas gen. nov. (Brugerolle, 2003), previously described as Cryptophagus subtilis, infecting the free-living Cryptophyte Chilomonas paramaecium (Brugerolle, 2002). However, apart from these two descriptive publications, no molecular analysis has yet been carried out to definitively affiliate Rastrimonas sp. within the Perkinsea lineage.

In marine environments, molecular signatures of Perkinsea were also found in extreme environments, such as hydrothermal vents or anoxic fjords (López-García et al., 2003; Zuendorf et al., 2006), but most of environmental genomics studies target water column samples [e.g., surface or deep chlorophyll maximum (DCM) depths] (e.g., Moon-van der Staay et al., 2001; López-García et al., 2003; De Vargas et al., 2015). This lack of genetic signatures in marine environmental surveys represent a real paradox, since the two main cultivable groups described, Perkinsus spp. and Parvilucifera spp., are marine. In 2014, targeting the V4 hypervariable region of the rDNA and rRNA templates using 454 sequencing technology, Chambouvet et al. (2014) evaluated the genetic diversity of the eukaryotic microbial community in two sampling depths (surface and DCM) and in the sediment across four different locations across Europe (Oslo, Norway; Naples, Italy; Barcelona, Spain; Roscoff and France). The analysis revealed an unexpected genetic diversity of ribosomally active organisms belonging to Perkinsea other than than Parviluciferaceae and Perkinsus clusters mainly detected in the sediment with a total of 265 sequences clustered in 150 OTUs defined at 99% similarity. The ribosomal RNA sequences present in these clades belong to metabolically active organisms, which certainly play an active role in the ecosystem functioning. This cryptic diversity, only detected by their genetic signatures, raises new scientific questions about the putative existence of non-parasitic heterotrophic organisms, the host range of parasitic Perkinsea, and their role and impact on the aquatic food web. Finally, these results suggested that sediments can act as parasite reservoirs, as suggested for Amoebophrya parasitoids (Chambouvet et al., 2011) and for marine Perkinsea (Chambouvet et al., 2014; Freeman et al., 2017; Reñé et al., 2021b). However, studies on diversity and distribution are scarce and restricted to specific environments (e.g., coastal water or few freshwater environments). A wider study of the Perkinsea distribution could give a more complete picture of environmental distribution unveiling potential reservoirs of pathogens. A phylogenetic analysis of Perkinsea illustrates two main evolutionary divisions between clades associated to Perkinsidae and Xcellidae, and Parviluciferaceae prior to marine/freshwater specialization (Jobard et al., 2020). This also suggests that transitions between marine and freshwater were few in the life history of Perkinsea, as differences between these environments constitute a barrier for cross-colonization for most organisms (Logares et al., 2009; Bråte et al., 2010). However, a deep analysis of environmental DNA may reveal new information on the evolution and environmental colonization of Perkinsea.