A putatively extinct higher taxon of Spirotrichea (Ciliophora) from the Lower Cretaceous of Brazil

Fossil microeukaryotes are key elements for understanding ancient ecosystems at microscopic level and improving the knowledge on the diversification of microbial life as a whole. We describe Palaeohypothrix bahiensis gen. et sp. nov., an exceptionally well-preserved Lower Cretaceous (Berriasian–Barremian; 145–125 Mya) amber-entrapped microeukaryote, identified as a spirotrich ciliate. The preservation of structures interpreted as the nuclear apparatus and remains of the ciliature revealed a novel ground plan, not found in modern Spirotrichea, thus representing a putatively extinct higher taxon lineage, viz. the Palaeohypotricha nov. tax. Based on cladistic analysis, the new taxon is hypothesized as phylogenetically related to the Protohypotrichia.

Fronto-ventral-transverse ciliature rather particular, because formed by ciliary units (indeterminable whether ordinary cilia or thin cirri) emerging from minute pits, roughly paired, arranged in tight rows resembling brushes (apparently anlagen-like; see Supplementary Discussion); exact number of ciliary units per row not determined (Figs. 1b, 2a, 3a,c-f). Frontal ciliature formed by apparently four of said brushes. Leftmost brushes (possibly three) hardly separable, appearing as a continuation of distal end of AZM lapel, possibly also connected posteriorly with paroral. Right frontal brush (i.e., fourth one) terminates at about the level of leftwards preceding frontal brush (Figs. 1b, 2i, 3c-f). Ventral region with two ciliary brushes: Leftmost one extending almost to body equator, wider anteriad, becomes thinner and composed of more spaced ciliary units towards distal end; right ventral brush running adjacent to left marginal row, possibly terminating slightly above the level of contractile vacuole (Figs. 2a, 3a,g,h, 4a).
At least four (five?) strong, transverse cirri-like structures emerging from conspicuous elongate, oblique pits longitudinally aligned below equatorial region of body, slightly displaced rightwards (Figs. 1b, 3g,i,j, 4b). Some undetermined surface structures present in the space between transverse cirri-like row and left marginal brush set (Figs. 1b, 4a). Left marginal ciliature with about ten ciliary brushes of decreasing length outwards, inserted in shallow furrows; outermost ones slightly oblique (Figs. 1b, 2a,f, 4e). One right marginal row, likely formed by cirri (Figs. 1b, 2a, 4b-d). Two short ciliary brushes (perhaps ventrally located caudal cirri-like structures) between transverse cirri-like structures and left marginal ciliature (Figs. 1b, 4a,d). Dorsal side with at least seven narrow cortical ribs slightly pronounced at anterior region of body, disappearing towards equatorial region of body (Figs. 1c, 4d). Dorsal ciliature not preserved. A tentative reconstruction of the live specimen is provided as Fig. 1d,e.
Identification as a spirotrich ciliate. At first glance, the specimen roughly resembles a palynomorph because of its outline and size. However, this is readily refuted as one perceives the structures herein interpreted as remains of the nuclear apparatus (of which the anterior macronuclear nodule is the most evident) and of the AZM; both appearing concordant with "polyhymenophore" ciliates in shape and topology.
We identified the studied specimen as a spirotrich ciliate because it has both macro-and micronuclei (i.e., nuclear dualism), with morphological vestiges of discoid DNA replication bands in the macronuclear nodules. www.nature.com/scientificreports/ The former is a time-honored synapomorphy of the Ciliophora, and the latter of the Spirotrichea 1,13 . Moreover, the dorsoventrally flat body with putative remains of specialized ciliature concentrated on the ventral side, all indicate the specimen belongs to the hypotrichs s. l. 2 . Among such structures, the nuclear apparatus is sometimes found in amber fossilized ciliates [14][15][16] , thus its preservation is not surprising. The DNA replication bands appear as conspicuous morphological features of the macronucleus in most spirotrichs, and are often easily visible under the light microscope even without the need of stain techniques (e.g., see Fig. 2 in Paiva et al. 17 ). Hence, replication bands vestiges in fossilized macronuclear nodules are foreseeable, even though preservation of actual ancient DNA content is unlikely. Remarkably, recent findings, accompanied by experimental taphonomy results, suggest that fossilization of microeukaryote nuclei is more common than previously thought to be possible 18,19 . On the other hand, preservation of ciliary structures seems rare, as they tend degrade during resin embedding 14 , even though structures interpreted as flagella were found in inclusions identified as fossilized flagellates 16,20 . Comparison with related taxa. When compared to the main higher taxa of hypotrichs s. l. (viz., Discocephalida, Euplota, Hypotricha and Protohypotrichia), the absence of a prominent cephalization and the rightwards vertically aligned transverse cirri-like structures (vs. horizontally) likely exclude placement among the discocephalids 21,22 . Considering the remaining taxa, macronuclear DNA replication bands were not reported in protohypotrichs 2 , but regardless, the studied specimen presents notable protohypotrich features, as discussed further below. The nuclear apparatus composed of two ellipsoid nodules, associated with at least one small micronucleus is a recurrent feature among hypotrichs, and notably frequent in the "oxytrichids", now assigned to the Postoralida 12 . On the other hand, the oral apparatus in the studied specimen is mostly reminiscent of the euplotids because of its large peristome to body length ratio and the presence of a multi rowed paroral membrane, which only barely resembles an euplotid polystichomonad pattern. In regard to this, the paroral configuration is worthy of attention for its uniqueness among the Spirotrichea. It represents a derived form due to its high complexity in relation not only to modern spirotrichs 21,[23][24][25] , but also to the SAL supergroup, to where they belong 26 . Another possibility which cannot be disregarded is that the multiple oblique fragments interpreted as part of the paroral do not belong to such structure, but could actually be homologous to the buccal cirri of modern hypotrichs 3,[27][28][29] . The absence of the endoral membrane, although not confirmed, is another common feature in euplotids, except for the diophryines 3,30 . However, the right portion of the complex paroral structure could be an atypical endoral. Unfortunately, shedding light on such matters would require investigation of ontogenetic processes.
Many euplotids and some protohypotrichs typically have the contractile vacuole positioned in the right side of body, below the equatorial region, as in the present specimen 31,32 . Conversely, in hypotrichs such vacuole(s) is(are) usually positioned along the left side of body 3,27-29 . The contractile vacuole in the present specimen was interpreted as such due to its position matching with those of euplotids, including the ventral overture. It cannot be excluded, however, that said structure is a cytoproct preserved during excretion. Anyhow, a fecal pellet was not found anywhere near the specimen, thus such interpretation is less favored.
The ventral ciliary pattern of the studied specimen has a unique feature among the Spirotrichea: Its basic organization displaying brushes of ciliary units at specific regions. Such pattern, and the connection of the paroral to the frontal brushes, are all reminiscent of the proter fronto-ventral-transverse (FVT) anlagen set of some modern hypotrichs, as developing during middle stage morphogenesis [33][34][35] . However, overall evidence seems to favor the interpretation of the specimen as a morphostatic cell: While the macronuclear replication bands indicate at least a very early stage of divisional morphogenesis, the absence of opisthe stomatogenesis (which normally precedes visible FVT anlagen differentiation) 27,36 demonstrates the ciliature was still morphostatic when the ciliate was entrapped by the resin. It is worth to mention that we cannot completely exclude the possibility of the structures interpreted as the left marginal brushes being the oral anlage of the opisthe. However, the ciliary brushes and their insertion furrows can be distinguished, at least partially, either by direct observation under bright field or after reconstructions, and they appear different from typical euplotid (subsurface) or hypotrich (surface) opisthe oral anlage.
Physiological reorganization could also explain proter-like FVT anlagen, but it is overruled because macronuclear replication bands are uncommon during that process 37 . Therefore, a suitable explanation for the curious ventral ciliature pattern of the studied specimen could be neoteny, because it exhibits anlage-like organization of most structures in a morphostatic, mature cell. Said pattern is, perhaps, comparable to that of the protohypotrich Kiitricha marina, in which anlagen segregation into cirri is belated, occurring after cytokinesis 2 . However, contrarily to that species, the juvenile pattern of the ciliature in P. bahiensis was retained during interphase, to the point it was still present when the macronuclear replication bands appeared-hence the neoteny. Lastly regarding the FVT ciliature, the longitudinal alignment of the transverse cirri-like structures is also reminiscent of protohypotrichs 2,31,38-41 . Noteworthy, we observed some undetermined structures on the ventral surface, in the space between the transverse cirri-like structures and the left marginal brushes. Although suggestive of remains of scattered ciliary elements or perhaps extrusomes, they did not conform with the criteria used to discern actual anatomic features from artifacts (see "Methods"). Therefore, were not included in the live specimen reconstruction presented in Fig. 1d,e. The presence of ciliary rows on both right and left body margins is consistent with all hypotrich s. l. groups, except euplotids, which normally lack the right marginal and display reduced number of left marginal cirri 1,32 . In protohypotrichs, possible right marginal cirri are not distinguishable from ventral ones, and the left marginal cirri are not ontogenetically homologous to those of the other groups 2 . Nevertheless, the arrangement of the left marginal ciliature of the studied specimen as brushes is also unique among hypotrichs s. l. www.nature.com/scientificreports/ Dorsally, the specimen has cortical ribs, indicating it had a rigid pellicle. Cortical ribs are common in various euplotids 42,43 , but not in hypotrichs, albeit exceptions occur, such as in Hemiholosticha kahli 44 . It is worthy of note that the posterior shortening of dorsal ribs could be an artifact of preservation in the studied specimen, therefore, should not be over interpreted.
Although comparable fossil hypotrichs s. l. are presently unknown, a last common ancestor of the Hypotricha was hypothesized by Berger 3 as resembling a typical non-dorsomarginalian 18-FVT cirri Hypotricha, thus being remarkably different from P. bahiensis.

Phylogenetic position and systematic implications to the Spirotrichea. A cladistic analysis of combined morphological and molecular characters (Supplementary Materials and Supplementary Data 2)
resulted in one optimal cladogram ( Fig. 5; Supplementary Fig. S1) in which the inclusion/exclusion of P. bahiensis from the analysis did not alter the resulting topology. The phylogenetic pattern is largely in agreement with previous studies, considering that the position and sometimes the monophyly of the main taxa vary slightly 4,5,45 . The placement of P. bahiensis (and consequently of the Palaeohypotricha) as sister of the marine Protohypotrichia is sustained by two morphological synapomorphies, namely: (i) the rightwards inclination of transverse cirri, tending to be verticalized-a pattern which also evolved independently in some euplotids, such as Aspidisca 32   www.nature.com/scientificreports/ maturate 2,39 . However, in P. bahiensis the fronto-ventral and the whole left marginal ciliature were retained as anlagen-like structures during interphase, because they occur concomitantly with the macronuclear replication bands. Hence, such ciliary structures are possibly neotenic. Traits such as (i) the complex arrangement of the paroral and its connection with the left most elements of the frontal ciliature; (ii) the continuity of the adoral membranelles of the lapel with the frontal ciliature; and (iii) the permanence of fronto-ventral and left marginal ciliary elements in anlage-like brushes during interphase are thus considered autapomorphies of P. bahiensis, and possible synapomorphies of Palaeohypotricha nov. tax. These are clearly ground plan defining traits not found in the more than 1300 known species of hypotrichs s. l. 46,47 , hence warranting the taxonomic acts herein established and the reasonable assumption of Palaeohypotricha as a putatively extinct ciliate higher taxon lineage. The paleobotanical affinity of this amber based on the record of fossil plants and pollen is difficult, as hitherto there are no well documented studies on paleobotany on the Caruaçu Member. Gas chromatography-mass spectrometry (GC/MS) analysis from the extracts of this amber identified alkylbenzenes, alkylnaphtalenes, alkylhydronaphtalenes, parafins, phenols, carboxylic acids and terpenoids 52 . Based on biomarkers as fenchone, camphor, 16, 17, 19-trisnorabieta-8, 11, 13-triene and methyl 16, 17-bisnordehydroabietate, the botanical origin should be related to Araucariaceae 52 . Therefore, later studies on the chemical composition of this amber 50,53 indicated that there are phyllocladanes and kauranes, also chemosystematic markers of other plant families besides Araucariaceae. The original resins could also be produced by plants from the Cupressaceae or Podocarpaceae families, but excludes Pinaceae. As triterpenoid compounds were not found, most angiosperms are not to be considered as the source of the resin.

Methods
Preservation quality of microscopic eukaryotes in amber may vary depending on the botanical origin of resin 14 , but other factors, in special temperature when embedded, appear to be determinant in structural preservation 54 . The resin in which the studied specimen was embedded stained the cellular structures we interpreted as the remains of nuclear apparatus and ciliature elements in the fashion of a faint protargol preparation-a rare event, herein reported for the first time in a ciliate amber inclusion.
Morphology and taxonomy. Amber fragments of roughly 1 cm 3 were cut in slices of about the thickness of a typical microscope glass slide and polished on both sides. Four slices were then searched for inclusions suggestive of microeukaryotes. To do so, the slides were covered with a water droplet to reduce the mirror effect caused by microscopic irregularities and cracks, and a coverslip for observation under the light microscope. We measured and photographed the studied specimen under bright field, 1.000 × (oil immersion), on various focal planes, based on which volume reconstructions were performed with Volume Viewer 2.01 plugin in Fiji 55 , using z-aspect of 2.0, varying alpha offsets, with and without oblique illumination. When applying the oblique illumination effect, presumed vestiges of ciliary structures appeared as cloudy masses at different focal planes. To discern from potential artifacts, we considered valid only those structures which (i) could be detected from different angles and were traceable through focal planes; (ii) were consistent with observations made under bright field; and (iii) homology with modern ciliates could be determined. The photographs used to make volume reconstructions are available as Supplementary Data 1. An interpretative reconstruction of the live specimen was made based on our observations of the fossil structures and comparison with related modern organisms.
Terminology of spirotrich ciliate morphology is basically according to Berger 3,28 and systematics above subclass level follows Lynn 1 . Given the potentially misleading effect of excessive scale bars when fixation shrinkage and other preparation artifacts alter the actual size of cell structures in modern ciliates 56 , we limited the use of scale bars to avoid informing false precision due to size distortions after volume reconstruction, preferring to provide measurement values of taxonomically relevant structures in the main text.
Cladistic analysis. Taxon sample contained species representing the Spirotrichea higher taxa, plus two species of Blepharisma (Heterotrichea) as outgroup, totalizing 29 terminals with P. bahiensis included (Supplementary Materials, Supplementary Table S1). We analysed a matrix of 26 morphological characters combined with the 18S rDNA (of modern species). Morphological characters were proposed as statements 57 , of which primary homologies rely on the interpretation of the studied specimen as a spirotrich ciliate. Characters were coded as either binary or multistate, treated as unordered. Issues of non-applicable character states were resolved via contingent coding 58,59 . The 18S matrix was aligned through the MUSCLE algorithm implemented in MEGA X 60 , and refined manually to minimize changes among nucleotides; leading and trailing gaps coded as missing data.
A cladistic analysis was performed in TNT 1.5 61 with parsimony uninformative characters disabled and gaps treated as a fifth base. We analysed the combined matrix under extended implied weights regime 62  www.nature.com/scientificreports/ optimal cladograms, we employed a combination of parsimony-ratchet, tree-drifting, tree-fusing and sectorial search routines 64,65 . Clade support was evaluated via 1000 symmetric resamples, expressed as GC frequencies (a metric which ranges from − 100 to 100), for which values < 0 were considered unreliable 66,67 , and the traditional Bremer index 68,69 , given in fit units to ensure consistency with analysis optimality criterion. Moreover, the ensemble consistency (CI) and retention (RI) indexes were calculated to measure character logical consistency and synapomorphy retention onto the tree, respectively 58 . Root placement was made a posteriori 70 . The analysed matrix is provided as a TNT formatted file (Supplementary Data 2).

Data availability
The analysed data used in this study are available as part of the Article and Supplementary Materials. The new genus and species in publication have been registered at ZooBank (LSID: D98CA5A2-B5B4-4264-B703-1EDE1EFF22BF).