Physiological reorganization in the hypotrich ciliate Apoamphisiella vernalis ( Protista , Ciliophora , Hypotricha )

Physiological reorganization is described for the fi rst time for the hypotrich ciliate genus Apoamphisiella Foissner, 1997, based on a population of A. vernalis (Stokes, 1887) Berger, 2006 collected from an eutrophic pond in the state of Minas Gerais, Brazil. Stomatogenesis is epiapokinetal, and the cirral pattern reorganizes from six typical fronto-ventral-transverse primordia, plus at least one short primordium located between IV and V. Primordia I and II originate from disaggregating undulating membranes and buccal cirrus, respectively; primordium III originates from left frontoventral cirrus plus a streak of basal bodies extending from the anterior end of oral primordium; primordium IV and the short extra primordia are formed from basal bodies perhaps associated to the posterior end of III plus the disaggregation of anteriormost cirri of the left ventral row. Primordia V and VI arise within the right ventral row as a single streak, splitting to form the two primordia. Marginal primordia develop initially within the pre-existent marginal rows, continuing to reorganize outside, pushing the old marginal cirri leftwards. Two dorsomarginal kineties develop associated to the right marginal row primordium. Dorsal ciliature also reorganizes from within the pre-existent dorsal kineties. The two macronuclear nodules approach each in middle reorganizers, briefl y touching each other. At least one micronucleus undergo division. The process of physiological reorganization in A. vernalis resembles that of the North American Paraurostyla weissei complex representative, and along with features of the interphase morphology, indicate that Apoamphisiella belongs or is related to the Cyrtohymena-Paraurostyla group, within the Dorsomarginalia.

In the present study, we describe, for the fi rst time, stages of physiological reorganization in the hypotrich genus Apoamphisiella Foissner, 1997, based on a strain of Apoamphisiella vernalis (Stokes, 1887) Berger, 2006, from an eutrophic pond in the Atlantic Forest in the state of Minas Gerais, Brazil.The origins of the various elements of the ciliature are traced from ontogeny during physiological reorganization, and results are discussed in comparison with the literature on supposedly related species.

MATERIAL AND METHODS
Specimens of Apoamphisiella vernalis were obtained from limnetic cultures enriched with crushed rice grains, made using water with sediments collected from a shallow eutrophic pond in the state of Minas Gerais, Brazil, in November of 2002 (approximate geographic coordinates: 21°59'36.4"S,44°57'17.7"W).Specimens were identifi ed using standard light and electron microscopy techniques commonly used for ciliate taxonomy, which included bright-field, differential interference contrast microscopy, scanning electron microscopy and protargol-impregnation (Dragesco & Dragesco-Kernéis, 1985;Dieckmann, 1995;Da Silva-Neto et al., 2012).The interphase (non-dividing) cells of this strain were studied in detail by Paiva et al. (2014), thus the reader is referred to that work and the original description by Stokes (1887) for the general morphology of A. vernalis.

RESULTS
Physiological reorganization in Apoamphisiella vernalis begins with the formation of an elongated, likely epiapoknetal, oral primordium, which originates de novo in the space between the left marginal and left ventral cirral rows.Such primordium is slender and slightly curved rightwards anteriorly.At its posterior region, there are small patches of basal bodies, suggesting the oral primordium arises from multiple points (Fig. 1).The oral primordium then becomes progressively elongated, forming a short anterior projection and growing slender posteriad (Figs 2, 3).Next, it becomes shorter and clavate, and then commences to produce new adoral membranelles, which will replace approximately half of the old lapel (Figs 4,5).At the anterior region of body, the paroral and endoral membranes simultaneously dissociate.Starting distally, the disaggregation of basal bodies moves posteriad until both membranes form primordium I.The fronto-ventral-transverse (FVT) primordium II originates from the disaggregation of the buccal cirrus, and primordium III from a long streak that proliferates anteriad from the anterior projection of oral primordium, fusing to the dissociating left frontoventral cirrus.Such long streak possibly contributes also to primordium II.At this stage, a streak of basal bodies which precedes primordia V and VI, grows within the right ventral row (Fig. 4).Primordia IV and IV ?arise likely associated with basal bodies near the posterior region of III, with participation of one or two anteriormost cirri from the left marginal row.As primordium II grows posteriad, the five streaks form a fan-shaped structure.The right marginal primordium is already visible at this stage, growing initially within the original structure.Meanwhile, the streak of basal bodies within the right ventral row splits into primordia V and VI (Fig. 5).
In middle-reorganizers, the postperistomal cirrus have been resorbed, and possibly incorporated in the differentiating oral primordium.The right frontoventral cirrus may be resorbed in middle reorganizers or retained until late stages.The left marginal primordium originates at this stage, similarly to the primordium for the right marginal row.Both right and left marginal primordia elongate posteriad, outside and to right of the old structures, pushing them leftwards.Simultaneously, there appear some short basal body streaks continuous with the posterior region of the fan-shaped cluster of FVT primordia, located between IV and V, and forming a ladderized pattern.At this stage, the left frontal procirrus begins to differentiate rightwards from the distal end of the undulating membranes primordium, and the basal bodies of the oral primordium organize into new adoral membranelles (Figs 6, 7).As reorganization proceeds, the FVT cirral streaks begin to differentiate, with cirri maturating posteriad (Fig. 7).The dorsal primordia arise within the old structures (Fig. 8), and fragmentation of dorsal kinety 3 [which happens in divisional morphogenesis (De Castro et al., 2016) and is noticeable in interphase (Paiva et al., 2014)] possibly occurs in very late reorganizers.
In middle-to-late reorganizers, the short streaks of basal bodies that originate between FVT primordia IV and V, viz.primordia IV ?, form each a single transverse cirrus (Figs 9, 10).During the process, they produce vestigial procirri, which are resorbed (Fig. 9).However, in some specimens (Fig. 10), one of such procirri in the leftmost IV ?primordium can maturate, becoming an extra postperistomal cirrus.Dorsomarginal kineties are formed right of right marginal row primordium, near its anterior end.The outer dorsomarginal kinety starts its development contiguously with the juvenile marginal cirri, then separates and moves rightwards.The paroral and endoral membranes appear differentiated, but still disposed parallel to each other, hence, the cyrtohymenid curvature and subsequent deepening of the oral cavity must occur in late reorganizers, which were not observed among the studied specimens (Fig. 10).During physiological reorganization, the distance between the macronuclear nodules is conspicuously reduced, the nodules briefly touching each other (Fig. 11).Fusion of the macronuclear nodules was not observed.At least one or two micronucleus per cell becomes ellipsoid and divides .
After reorganization, the remaining old ciliary structures are fully resorbed, except by the collar and approximately half of the lapel adoral membranelles, which are maintained after physiological reorganization.The origins of each FVT primordia and their cirral products are summarized in Tab.I.

DISCUSSION
Until now, the process of physiological reorganization in Apoamphisiella was never described.When compared to species of supposedly related genera, the reorganization in Apoamphisiella vernalis resembles that described by Jerka-Dziadosz & Frankel (1969) for a North American population of the Paraurostyla weissei complex, viz.Paraurostyla polymicronucleata Merriman, 1937, according to Berger (1999).In both cases, the oral primordium is curved rightwards and originates de novo [vs.originates adjacently (parakinetal?) to postperistomal cirri row in the European P. weissei (Wirnsberger et al., 1985)]; the undulating membranes start disaggregation before fusing; about half of the membranelles of the lapel are replaced; and the two rightmost FVT primordia originate from a primordium growing within the rightmost ventral row, which then splits in two.
Another important shared feature is the participation of basal bodies originating from the oral primordium to the formation of the FVT primordia.In A. vernalis, one long streak of basal bodies extends from the anterior end of the oral primordium and merges with the disaggregating left frontoventral cirrus, resulting in an small anarchic field of basal bodies near the posterior end of III, from where primordia IV and then IV ?arise.Such long streak possibly also contributes to primordium II (Fig. 4).On the other hand, in the North American P. weissei and in Paraurostyla coronata Arora et al., 1999, there are two or three of such streaks, which are more-or-less parallel and apparently contribute to primordia II, III andIV [see Fig. 15 in Jerka-Dziadosz &Frankel (1969) andFigs 23, 24 in Arora et al. (1999)].Jerka-Dziadosz & Frankel (1969) also reported the formation of a "bare patch" of basal bodies in the space between the anterior region of the oral primordium and the proximal termini of the three leftmost FVT primordia of the North American P. weissei.Such structure is likely associated to agglomeration of basal bodies of the above mentioned streaks originating from the oral primordium.In A. vernalis, as only one of such streak develops, a proper patch is not formed.The short FVT primordia which arise between IV and V, herein named "IV ?" are possibly related to IV, and may represent slightly modified replicas of it (De Castro et al., 2016).Evidence of such relationship comes from the primordia behavior, which is similar to IV, viz.forming transverse cirri, but not pretransverse; and sometimes forming a postperistomal cirrus (Fig. 10).Moreover, serial repetitions of primordium IV are common in hypotrichs.They are characteristic of the urostyloids, and probably evolved independently in pattersoniellids, neokeronopsids, and perhaps also in uroleptids and rigidothrichids (Foissner et al., 2004;Foissner & Stoeck, 2006, 2008;Paiva et al., 2009), forming the so-called midventral complex (Berger, 2006).The pattern in which primordia IV ?appear in early-to-middle reorganizers, arranged in a ladderized sequence of short basal body streaks (Figs 7, 9), recall the corresponding developmental stage of a short pattersoniellid or Territricha-Apoterritricha-like midventral complex (Foissner, 1987;Berger & Foissner, 1988;Kim et al., 2014).Such structure, in A. vernalis, does not complete its development because the procirri which could maturate into midventral pairs are resorbed before differentiation.Reorganization of the marginal cirral rows and dorsal kineties had no peculiarities, occurring as in most dorsomarginalians (e.g.Grimes & Adler, 1978;Jerka-Dziadosz & Frankel, 1969;Kamra & Sapra, 1991;Arora et al., 1999;Blatterer & Foissner, 2003).
During physiological reorganization, the macronucelar nodules of A. vernalis approximate in middle reorganization, briefly touching each other and then to beginning to separate in middle-to-late reorganizers.Thus, complete fusion of the nodules was not observed.However, macronuclear fusion is reported in other dorsomarginalians (e.g.Arora et al.,1999;Zou & Ng ,1991;Blatterer & Foissner, 2003), including the North American P. weissei, as described in Jerka- Dziadosz & Frankel (1969).Hence, we cannot exclude the possibility of a curt fusion being overlooked because the corresponding stage could be absent among the investigated specimens.In A. vernalis, not all micronuclei seemed to divide during physiological reorganization, since the studied reorganizers exhibited only one or two dividing micronuclei (among on average five of such structures).Conversely, in the North American P. weissei, the micronuclei divide to produce a total of eight units, indicating most or all micronuclei divide, even though subsequently from reorganization, this number is reduced to about four (Jerka-Dziadosz & Frankel, 1969).
The data on ontogeny during physiological reorganization confirmed the presence of dorsomarginal kineties, corroborating the classification of genus Apoamphisiella within the Dorsomarginalia Berger, 2006, as mentioned in Paiva et al. (2014).Moreover, the ontogenetic similarities with the North American population of P. weissei, plus other morphological features, such as the cyrtohymenid pattern of the undulating membranes and the morphology and arrangement of cortical granules (Grimes & L'Hernault, 1978;Foissner, 1997;Paiva & Silva-Neto, 2004;Küppers & Claps, 2013), support the phylogenetic position of Apoamphisiella in the Cyrtohymena-Paraurostyla (Berger, 1999;Foissner et al., 2004)

Figs 4
Figs 4, 5. Ventral side of early-to-middle reorganizers, after protargol impregnation.Fig. 4. Early development of primordia I-III.Black arrowhead shows the streak of basal bodies which will form primordia IV and V; white arrowhead shows the postperistomal cirrus.Notice a streak of basal bodies originating from anterior end of oral primordium (asterisk), which contributes to the formation of primordium III.Fig. 5. Early development of primordia IV-VI and right marginal primordium.The arrows point to the right frontoventral cirrus.Fronto-ventral-transverse primordia numbered in romans.(IV ?, short primordia possibly related to IV; AZM, adoral zone of membranelles; E, endoral; LMR, left marginal cirral row; LVR, left ventral cirral row; OP, oral primordium; P, paroral; RMP, right marginal primordium; RMR, right marginal cirral row; RVR, right ventral cirral row).Scale bar = 20µm.
Figs 9-14.Reorganizers after protargol-impregnation.Figs 9, 10.Ventral side of middle-to-late reorganizers.Notice the formation of surplus transverse cirri (black arrowheads) by primordia IV ?, the leftmost forming also a second postperistomal cirrus (double black arrowhead).White asterisk shows vestigial procirri originating from primordia IV ?, which are later resorbed; black asterisk shows late differentiation of adoral membranelles; black arrow shows differentiating undulating membranes; white arrow indicates residual right frontoventral cirrus; double white arrowhead shows outer dorsomarginal kinety originating from anterior end of right marginal primordium; and white arrowheads indicate newly formed dorsomarginal kineties.Figs 11-14.Nuclear apparatus of different specimens.Notice the dividing (white arrowheads) and non-diving micronuclei (black arrowhead).Fronto-ventral-transverse primordia numbered in romans.Scale bar = 20 µm.
group of dorsomarginalian hypotrichs, which was recently hypothesized by De Castro et al. (2016).