Pudicinae (Nematoda) coparasitic in Proechimys roberti (Rodentia: Echimyidae) from the Brazilian Amazonia: description of a new species of Pudica, redescription of Pudica evandroi (Travassos) and updated key to the species of the genus.

The re-examination of nematodes collected decades ago from a spiny rat, Proechimys roberti (Echimyidae) from Pará State, revealed the presence of Acanthostrongylus acanthostrongylus Travassos, 1937, and two coparasitic species of Pudica (Heligmonellidae: Pudicinae), one of them new to science. Pudica minima n. sp. mostly resembles Pudica tenua Durette-Desset, 1970, described from Proechimys semispinosus in Colombia. Nevertheless, it differs from P. tenua by having a synlophe with 12 ridges with poorly developed careen, bursal pattern of type 2-2-1 on the right lobe, bursal rays 2 and 3 similar in length, and spicules longer with spoon-shaped tips. Pudica evandroi (Travassos, 1937) was already described from Pr. roberti but its synlophe remained undescribed and the original description lacked some details. A detailed study of the synlophe is provided and the range of morphometric data for males and females is enlarged. The study of the synlophe allowed ratifying the placement of this species within Pudica, as early suggested by previous authors. The species list and identification key to species of Pudica are updated with the inclusion of three species. The status of Heligmosomum alpha Travassos, 1918, which was transferred to Pudica in 1990, is emended and the species is considered a Nippostrongylinae incertae sedis.


INTRODUCTION
The genus Proechimys Allen, 1899 is the most speciose and geographically most widely distributed of the echimyidae rodents. Known commonly as "spiny rats", all species are essentially limited to lowland rainforest habitats in Central America and Amazonian South America, where they are important components of the terrestrial forest community, serving as seed predators and dispersal agents (Patton and leite 2015). Species of Proechimys are of public health interest as they act as reservoirs of parasitic zoonoses such as leishmaniasis and trypanosomiasis (Cordeiro et al. 2015a, Patton andleite 2015). The genus comprises at present 22 species distributed in 10 species groups (Patton and leite 2015). Among them, Proechimys roberti Thomas, 1901 occurs throughout the rainforest of Amazonian Brazil south of the Amazonas River and extends south to the Cerrado biome in east-central Brazil (Weksler et al. 2001). There have been relatively few studies examining the helminth fauna of this species (Cordeiro et al. 2015a). To date, it has been cited as host of seven species of nematodes: four out of them are Heligmonellidae: Acanthostrongylus acanthostrongylus Travassos, 1937, Pudica evandroi (Travassos, 1937), Heligmostrongylus squamastrongylus (Travassos, 1937), and Squamasnema amazonica Cordeiro, Melo, Furtado, giese, Maldonado Jr. and dos Santos, 2015. The other species are Molinema dessetae (Bain, 1973) (Onchocercidae), Spirura carajaensis Cordeiro, Melo, Maldonado and Santos, 2015 (Spiruridae) and one species of Gongylonema Molin, 1857 (Travassos 1937, Bain 1973, Vicente et al. 1997, Cordeiro et al. 2015a, b, Durette-Desset et al. 2017). Up to now, for all these species except for Gongylonema sp., Pr. roberti is the type and unique host. The old parasitological literature (Travassos 1937) refer the host of A. acanthostrongylus, P. evandroi and H. squamastrongylus to as Proechimys oris Thomas, 1904. However, more recent studies indicate that Pr. oris should be considered a junior synonym of Pr. roberti (Weksler et al. 2001, Patton andleite 2015).
In 1971, several nematodes collected from one Pr. roberti from Pará State were deposited at the Muséum national d'Histoire naturelle (MNHN) (Paris, France). These nematodes are studied herein. Among this material, three coparasitic taxa were identified: A. acanthostrongylus, P. evandroi, and other specimens belonging to a new species of Pudica Travassos and Darriba,1929 (Heligmonellidae: Pudicinae). In this paper, the new species is described and illustrated, a redescription of P. evandroi (whose synlophe remained unknown) is provided, and an updated key to the species of Pudica is included. The ZooBank Life Science Identifier (LSID) of this publication is: urn:lsid:zoobank.org:pub:DBeDBe0D-2396-4e4e-BD41-8C46C75F333e.

MATERIALS AND METHODS
The material was deposited at the Muséum national d'Histoire naturelle (MNHN) (Paris, France) under the number 136 Rl. It consisted of nematodes collected from a Pr. roberti from Itaituba (Pará State, Brazil) in 1971, in the frame of a project involving the survey for sylvatic reservoirs of T. cruzi-like trypanosomes under the auspices of the Wellcome Trust (london, UK) and the Instituto evandro Chagas (Belém, Brazil) (see lainson et al. 1979).
The nematodes were recovered in the field and identified preliminary by one of us (MCDD) as P. evandroi at the time of collection. Their study started at the MNHN shortly after recovery but it was arrested for several years, until it was restarted at the Division Zoología Invertebrados of the Museo de la Plata (la Plata, Argentina).
Nematodes were studied in temporary mounts in Amman's lactophenol under a leica DM 2500 microscope provided with a drawing attachment. The synlophe was studied following Durette-Desset (1985) and the nomenclature referring to the axis of orientation of the ridges follows Durette-Desset and Digiani (2005). Ridges are numbered with respect to the oblique axis of orientation: from left to right, 1 to n for right-dorsal ridges and 1' to n' for leftventral ridges. The nomenclature used for the study of the bursa (pattern of lateral lobes and symmetry) follows Durette-Desset and Digiani (2012). Figures  8, 11b and 26 are original drawings made in 1971, whose scale bars are missing. Measurements are provided in micrometres (except body length, in mm.), as follows: in the new species description, holotype/allotype followed by range and mean of paratypes in brackets; in the redescription of P. evandroi, range followed by mean in parenthesis. Spl/Bl and Utl/Bl mean the proportion of the spicule length to the body length and of the uterus length to the body length, respectively. The type material of the new species described was deposited at the Helminthological Collection of the Instituto Oswaldo Cruz (CHIOC), Rio de Janeiro.
General. Very small nematodes, irregularly coiled, with coiling mainly affecting the anterior part of the body. excretory pore situated anterior to oesophago-intestinal junction in both sexes (43-82% of oesophagus length in males and 56-89% in females). Deirids situated asymmetrically with respect to the apex, with left deirid at level of the excretory pore and right deirid some 20 mm more anteriorly (Figure 1).
Synlophe (studied in 2 males and 2 females). In both sexes cuticle bearing longitudinal, uninterrupted ridges, most of them appearing posterior to cephalic vesicle. Twelve ridges (careen, 5 right-dorsal and 5 left-ventral) in both sexes, at oesophago-intestinal junction, at mid-body and posterior third of body. From oesophago-intestinal junction to mid-body, ridges forming careen largest but not hypertrophied (Figures 3-6). Remaining ridges medium to small. gradient in ridge size absent. Dorsal ridges similar in size. Male: at midbody, ridges 2'and 3' larger than 4'-6' (Figure 3).
Female: at anterior third of body length, ridges 2'-4' larger than 5' and 6', upper part of right-ventral quadrant free of ridges ( Figure 5). From mid-body, ridges 2' and 5' diminishing in size and ridges forming careen less salient (Figure 6). At level of uterus, all ridges small and similar in size and careen not differentiated. Some struts not visible (Figure 7). Axis of orientation of ridges single, inclined at about 45° to sagittal axis in both sexes.
Males Bursa subsymmetrical ( Figure 8). Prebursal papillae not observed. Right lobe with pattern of type 2-2-1. left lobe with pattern of type 2-2-1 tending to 1-3-1. Rays 2 and 3 parallel and similar in length. Rays 4 and 5 diverging slightly at distal part of their length. Rays 6 long and straight, with similar diameter through their length. Dorsal lobe medium-sized. Dorsal ray strongly thickened, especially dorso-ventrally, divided at about distal third into two branches (Figures 9, 10). each branch divided into two sub-branches, rays 9 (external) and rays 10 (internal). Rays 8 of similar length, not reaching level of extremity of dorsal ray, arising from the ventral surface of the dorsal ray at the angle of the dorsal ray and the lateral common trunk of rays 2 to 6 ( Figure 9). Origin of rays 8 not clearly visible in dorsal view due to the thickening of dorsal ray and both lateral trunks at their base ( Figure 10 Figure 12).

ReMARKS
The presence of a cephalic vesicle and a synlophe with an axis of orientation not bilaterally symmetrical places these specimens into the superfamily Heligmosomoidea, as redefined by Durette-Desset (1983) and Durette-Desset and Chabaud (1993), which is composed of six families. Two of these, the Heligmonellidae and the Viannaiidae are well represented in caviomorph rodents. The presence of a careen and spicules with conspicuous alae, together with the absence of gubernaculum in the present specimens, allow to assign them to the Heligmonellidae: Pudicinae. Among these, and according to Durette-Desset and Justine (1991) and Durette-Desset et al. (2017), the following characters enable us to assign this species to the genus Pudica: synlophe with careen made up of two continuous ridges, three to five continuous dorsal ridges, three to five continuous ventral ridges, at least two ventral comaretes present, together with a bursa of type 2-2-1 and 9th rays slightly longer than 10th. Within the genus, these specimens strongly resemble Pudica tenua Durette-Desset, 1970, parasitic in Proechimys semispinosus Tomes, 1860 from Colombia, by the tiny body size, the dissymmetrical position of the deirids (only recorded in these two species), the bursal pattern and general arrangement of the bursal rays, and the similar number of ridges of the synlophe (11-12). However, these specimens are differentiated from P. tenua by a bursal pattern of type 2-2-1 on right lobe (vs. 2-2-1 t 1-3-1), rays 2 and 3 similar in length (vs. rays 2 shorter), rays 8 not reaching level of extremity of dorsal ray, branches of dorsal ray bifid (vs. trifid, with phasmids visible), origin of rays 8 from the angle of dorsal ray and lateral trunks (vs. from proximal third of dorsal ray), longer spicules (370-460 mm vs. 182 mm) with spoon-shaped tips (vs. sharp tips) and greater SpL/BL (21-31% vs. 14%). Differences concerning the synlophe at mid-body are: careen less developed, 12 ridges (vs. 11 in the figure of P. tenua), and ridges 3' and 4' well developed (vs. 2', 3' and 4'). These differences allow us to consider these specimens as a new species, for which the name Pudica minima n. sp. is proposed. Etymology. The specific name refers to the extremely small size of the specimens.
Pudica evandroi (Travassos, 1937) Evandroia evandroi Travassos, 1937: 352. Pudica evandroi ˗ Durette-Desset, 1983: 60. with cuticular struts; remaining three, on rightventral quadrant, smaller and perpendicular to body surface, without cuticular struts at this level (Figures 13, 14). These three smaller ridges acquire cuticular struts but remain poorly developed and without orientation, disappearing at mid-body in females and at 60% of total length in males, leaving right-ventral quadrant free of ridges (Figures 15-20). At mid-body: 13 ridges in male, 10 in female (Figures 18, 19). Size of ridges: at the anterior half of spicules in males (80-85% of Bl) and at level of uterus in females (ca. 75% of Bl): 10 in both sexes, all ridges small and similar in size; right-ventral quadrant still free of ridges (Figures 21, 22). From ovejector level, ridges of synlophe widened to form prevulvar alae: two latero-ventral and two laterodorsal. In one of the two observed females, one medio-dorsal ala also present ( Figure 23). Axis of orientation of ridges single, inclined at about 55° to sagittal axis in both sexes. Disappearance of ridges within the 200 mm proximal to bursa in males, just proximal to vulva in females.
General. Small nematodes, irregularly coiled, with coiling mostly affecting the anterior half of the body. excretory pore situated anterior to oesophago-intestinal junction in both sexes (55-86% of oesophagus length in males and 77-80% in females). Deirids situated symmetrically with respect to the apex (Figure 24).
Head. Cephalic vesicle present. In apical view, rounded buccal opening surrounded by very thick ring. Papillae not observed ( Figure 25).

ReMARKS
The specimens described herein have morphological and morphometric characters that match those of P. evandroi. Evandroia evandroi Travassos, 1937 was the type and sole species of Evandroia Travassos, 1937but Durette-Desset (1983 synonymized Evandroia with Pudica, arguing that the synlophes of the type species of both genera were identical. However, the synlophe of P. evandroi has never been formally described peculiar form and thickness of the dorsal ray and the asymmetrical arising of its branches. On the other hand, the range of morphometric data is enlarged, including measurements of uterus, infundibulum and vestibule in females. The classification of the Pudicinae into genera is based mainly on characters of the synlophe (Durette-Desset and Justine 1991, Durette-Desset et al. 2017). even though the synlophe of the specimens studied herein is not exactly identical to that of the type species of Pudica, as argued by Durette-Desset (1983), the different characters observed in this structure correspond largely better to those observed in species of Pudica than to those present in any other genus of Pudicinae. These characters include a careen made up of two ridges similar in size, and the remaining ridges continuous. Such characters, together with a bursa mainly of type 2-2-1 and 9th rays slightly longer than 10th, allow us to ratify the placement of this species within the genus Pudica, as early suggested by Durette-Desset (1983) and assumed by Durette-Desset and Justine (1991) and Durette-Desset et al. (2017).

DISCUSSION
With the description of Pudica minima n. sp. the number of species of Pudicinae described from Pr. roberti throughout its distribution increases to five. Proechimys roberti is the type and unique host for all these parasite species. Pudica minima n. sp. and P. evandroi were coparasitic in the same individual host. This phenomenon of coparasitism of related species is frequent among heligmonellids of rodents, and particularly in spiny rats and agoutis (Dasyproctidae), where a unique host may harbor two, three and up to five species of Pudicinae (see Durette-Desset et al. 2017). Although we can strongly suspect coparasitism among the species described by Travassos (1937) from Pr. roberti (= Pr. oris), all from the same locality and time, this phenomenon was not documented. In fact, in most taxonomic works, the coparasitism is overlooked or simply not recorded (Durette-Desset et al. 2017). Durette-Desset et al. (2017) accounted for 13 species of Pudica, for which they also provided an identification key. Shortly after the publication of such revision, two new species were added to the genus: Pudica cattani Digiani, landaeta-Aqueveque, Serrano andNotarnicola, 2017 andPudica degusi (Babero andCattan, 1975), both parasitic in Chilean caviomorphs (Octodontidae and Abrocomidae) .
The findings of Digiani et al. (2017) together with the present description of Pudica minima n. sp., increase in three the number of known species of Pudica in comparison to those listed in Durette-Desset et al. (2017). However, just after the publication of such list, the writers realized that Pudica alpha (Travassos, 1918) should be excluded from Pudica, for the reasons explained below. Heligmosomum alpha Travassos, 1918 was described from a wild mouse "camondongo selvagem" (sic) from Angra dos Reis, State of Rio de Janeiro. The original description of H. alpha did not include illustrations but the species was redescribed and illustrated by Travassos (1921). The host was not identified at specific level but the author specified that it was a Muridae. The combination Longistriata alpha was then proposed by Travassos and Darriba (1929). Two years later, Henry (1931) reported the finding of L. alpha in France from a nutria or coypu, Myocastor coypus (Molina, 1782) introduced from South America. The record of Henry (1931) was a misidentification, as demonstrated by Travassos (1937), who stated that the parasite found by Henry (1931) corresponded in fact to Longistriata maldonadoi Artigas and Pacheco, 1933 (now Pudica maldonadoi), described from the same host in Brazil. Travassos (1937)  sensu Henry (1931) nec Travassos (1918) as a synonym of L. maldonadoi. Several years later, Durette-Desset (1990), probably based on the report of Henry (1931), assumed erroneously that L. alpha was a parasite of the coypu, and transferred L. alpha to Pudica, creating the new combination Pudica alpha. The two reasons argued for this transference were, on one hand, the similarity of the bursa to that of P. maldonadoi, and on the other hand, the (erroneous) assumption that the host was a caviomorph. This mistake is partially overcome in Durette-Desset et al. (2017), where P. alpha sensu Travassos appears correctly having a Muridae as type host, and P. alpha sensu Henry is treated as a synonym of P. maldonadoi. However, there were no longer reasons to keep the species of Travassos in Pudica, nor even in the Pudicinae. First, due to the fact that the synlophe remains unknown, second, because the host is not a caviomorph but a Muridae, and, finally, the fact that the specimens possess a gubernaculum, a typical character of the Nippostrongylinae, whereas the Pudicinae are characterized by the absence of this structure (Durette-Desset et al. 2017). We consider that these arguments are enough to dismiss the combination Pudica alpha and to declare Heligmosomum alpha as a Nippostrongylinae incertae sedis.
The number of species of Pudica is then modified to 15, and the key to the species of the genus can be rewritten as follows: Identification key to the species of Pudica