Interspecific micronuclear transplantation in Paramecium: nucleogenesis and stomatogenesis in asexual and sexual reproduction

The micronucleus and the macronucleus of ciliated protozoa are traditionally thought to be demarcated sharply in terms of their functions. The micronucleus is obviously indispensable in nuclear reorganization during sexual reproduction, while the macronucleus governs somatic functions and asexual propagation. This traditional view has been critically reexamined, and the sharp functional demarcation between the micronucleus and the macronucleus is no longer tenable in view of compelling evidence demonstrating the somatic function of the micronucleus (reviewed by Ng, 1986). In particular, in Paramecium tetraurelia, it has been shown that the micronucleus plays an important role in the development of the oral apparatus, in both asexual and sexual reproduction (Ng & Mikami, 1981; Ng & Newman, 1984b; Tam & Ng, 1986; Chau & Ng, 1988a,b,c). These studies have made use of (1) amicronucleate cell lines generated by removal of the micronucleus with a microinjection needle, (2) defective-micronucleus cell lines generated by creating lesions in the micronucleus with laser microbeam irradiation of the micronucleus, or treatment of the cells with cis-dichlorodiammineplatinum II and also (3) haploids.

An amicronucleate cell line can be obtained following the removal of the two micronuclei from the cell. Such cell lines enter a growth depression period manifested by the development of defective, but mostly functional oral apparatuses, characterized by abnormal oral ciliature (‘membranelles’) and reduction of oral length. These cell lines can gradually recover and the oral apparatus returns to a near normal condition. On the other hand, amicronucleate cells are doomed when induced to go through the sexual cycle (autogamy or conjugation), during which they resorb the preexisting oral apparatus, as usual, but fail completely to develop a new one. Stomatogenesis in the sexual cycle is arrested characteristically at an early stage of alignment of the basal bodies of the oral anarchic field into parallel rows, in preparation for the formation of the oral membranelles (‘initiation’); no oral structures can subsequently develop, resulting in astomy and death. Of particular interest, this specific stomatogenic function relies on the activity of micronuclear postmeiotic divisional derivatives, some of which reside in a ‘paroral cone area’ that is to the immediate right of the oral anarchic field, at the time when oral membranelle assembly is initiated (see Ng & Newman, 1984a). Cells possessing defective micronuclei are able to initiate the assembly of oral membranelles in the sexual cycle, but the oral apparatus subsequently developed is frequently defective. We have concluded from these observations that the micronucleus plays an important, but replaceable, role in the development of a normal oral membraneliar pattern and length during asexual propagation, but its function in the initiation of oral membranelle assembly in the sexual cycle is indispensable. The studies of defective-micronucleus cell lines also suggest that the micronucleus plays a role in determining normal oral membraneliar pattern and length in the sexual cycle.

Heterospecific micronuclear transplantation, the transplantation of a micronucleus of one species into an amicronucleate of another species, offers an additional approach to the analysis of the stomatogenic function of the micronucleus. Micronuclear anomalies may develop in a foreign cytoplasm during asexual and sexual reproduction, if incompatibility between the two species exists, and the stomatogenic consequences in such heterospecific transplants can be studied. Furthermore, if the two species employed differ somewhat in their oral apparatus, the control of the micronucleus over particular stomatogenic steps can be dissected in the heterospecific transplants.

In the present study, heterospecific micronucleus transplantation is performed using P. jenningsi amicronucleates as recipients and P. tetraurelia as micronucleus donors. There is enough similarity between the two species to warrant, a priori, the successful maintenance and propagation of the micronucleus in the transplants. Both types of micronuclei belong to the ‘aurelia’ group (see Wichterman, 1986). They both undergo autogamy naturally upon starvation and their micronuclear cycles in autogamy are largely similar (for P. jenningsi, see Diller & Earl, 1958; Mitchell, 1963). Furthermore, P. jenningsi resembles P. tetraurelia in the control of the micronucleus over stomatogenesis, in both asexual and sexual reproduction (M.F.C. & S.F.N., in preparation; cf. Ng & Mikami, 1981; Ng & Newman, 1984b). The oral apparatuses of the two species are similar in the pattern of the oral membranelles, but differ substantially in length (P. tetraurelia, 28 μm; P. jenningsi, 36 μ m) (Figs 1, 2). There are thus both similarities and differences in the micronucleus-stomatogenesis system in the two species that make them suitable for the heterospecific analysis.

Homospecific transplantations, involving amicronucleate recipients and micronucleus donors of the same species (P. tetraurelia), could generate genetically normal cell lines (Ng, 1981) and rescue the cells from stomatogenic abnormalities and failures in asexual and sexual reproduction (Ng & Tam, 1987). Furthermore, in the Paramecium caudatum complex, heterospecific micronuclear transplantation between two sibling species (syngens) has been shown to be able to promote recovery from growth depression (Fujishima & Watanabe, 1981). Heterospecific micronuclear transplantations are expected to entail different stomatogenic consequences, unless the micronuclear stomatogenic signals in the two species concerned have much in common.

Two species of Paramecium were used (i) Paramecium tetraurelia stock d4-lll, a derivative of stock 51 VII, is homozygous for the nd3^b gene and cannot discharge trichocysts when stimulated by saturated picric acid (for details, see Sonneborn, 1974). (ii) Paramecium jenningsi, obtained from K. Hiwatashi, Japan, originally from the Sonneborn collection, USA. For culture and handling of cells, the methods of Sonneborn (1950, 1970) were followed. Phos-phate-buffered cerophyl medium (2·5 g 1^-1, pH 7), inoculated with Enterobacter aerogenes and supplemented with 5mg 1”‘stigmasterol, was used. Experiments were performed at 27°C, while back cultures were stored at 13-15°C.

Enucleation and renucleation were performed according to the method described in Ng (1981), except one needle was used. Four P. jenningsi amicronucleate cell lines (amic JI, amie J2, amic J3a, amic J3b) were derived independently from three bimicronucleate clones no. 1–3 (amic J3a and J3b being from the same clone, no. 3), by simultaneously removing the two micronuclei from a cell at the clonal ages of 18, 7, 30 and 30 fissions, respectively. At the same time, two unimicronucleate cell lines (mic JI, mic J2) were also generated from bimicronucleate clones no. 1 and no. 2, respectively, by emicronucleation; a unimicronucleate clone, mic J3, was derived by isolation from clone no. 3 exhibiting mis-segregation of micronuclei during asexual propagation.

Two schemes of renucleation experiments were performed: (i) Homospecific transplantation, with P.jenningsi as micronucleus donors. Three renucleated cell lines, RJ2, RJ3a and RJ3b, were generated by renucleating amicro-nucleate cell lines, amie J2, J3a and J3b, at the clonal ages of 104, 45 and 50 fissions, respectively, (ii) Heterospecific transplantation, with P. tetraurelia stock d4-lll as micronucleus donors. Eight renucleated cell lines were generated from four amicronucleate cell lines, amic JI, J2, J3a and J3b (RT1 from amic JI, at 80 fissions; RT2.1 and RT2.2 from amie J2, at 116 fissions; RT3a.l and RT3a.2 from amic J3a, at 48 fissions; RT3b.l, RT3b.2 and RT3b.3 from amic J3b, at 48, 54 and 54 fissions, respectively). The homospecific transplants were collectively designated as RJ2–RJ3b, and the heterospecific transplants as RTl–RT3b.3 in the text. The clonal relationship of these cell lines was apparent from their listing in the tables.

To reveal oral and nuclear structures, cells in vegetative propagation, and also postautogamous cultures in each cell line were sampled for silver impregnation (Chatton & Lwoff, 1936; Corliss, 1953) at different times (number of fissions) after renucleation (for details, see Tables 1–4). In the sampling of postautogamous cultures, astomatous cells of abnormal shapes were mostly not included. Samples of the progenitor cell lines, including P. jenningsi amicronucleate recipients, and P. jenningsi and P. tetraurelia micronucleate donors, were obtained at similar clonal ages for comparison. The length of the oral apparatus was defined by the anteroposterior span of the quadrulus in silver-impregnated cells and measured with an ocular micrometer under ×1 000 phase-contrast optics (see Figs 1C, 2B). For observing nuclei, aceto-orcein staining was also applied (Beale & Jurand, 1966; without osmium fixation).

One-tailed t-test for comparison of mean length of the oral apparatuses with formula to cater for inequality of variances and 2×2 G-test of independence for comparison of two percentages were performed as described by Sokal & Rohlf (1981).

The maintenance of the micronucleus in asexual propagation, nuclear reorganization in the sexual cycle, and the pattern and length of the oral apparatuses in eleven renucleated cell lines (RJ2–RJ3b with P. jenningsi micronuclei, RTl-RT3b.3 with P. tetraurelia micronuclei) in the asexual and sexual cycles are reported in the following sections.

The micronuclei in the renucleated cell lines did not exhibit any structural defects (Figs 3,4). However, abnormality in the maintenance of the micronucleus was evident in all renucleated cell lines, since they gave rise to amicronucleates. The percentage of cells possessing micronuclei in the renucleated cell lines varied from 25 % to 81 % (RJ2, 35 %; RJ3a, 53 %; RJ3b, 67%; RT1, 25%; RT2.1, 34%; RT2.2, 62%; RT3a.l, 81%; RT3a.2, 65%; RT3b.l, 57%; RT3b.2, 59%; RT3b.3, 78%; at 10-15 fissions after renucleation). This revealed the subnormal behaviour of the transplanted P. jenningsi and P. tetraurelia micronuclei during asexual propagation. Nevertheless, the micronucleus of both types could be maintained in the transplants for a long period: in most of the renucleated cell lines, the percentage of cells possessing micronuclei ranged from 55 to 72 % when assessed 35–48 fissions after transplantation; the three exceptions were RJ1 (0 %, 35 fissions), RT1 (6%, 35 fissions) and RT2.1 (0%, 38 fissions).

Amicronucleates P. jenningsi experienced a depression period shortly after emicronucleation, during which the growth rate was reduced and the oral apparatuses were grossly abnormal in membranellar pattern and in length (Figs 5–7). However, they gradually recovered in growth rate, and the oral membraneliar pattern and oral length returned to near normal in 30-60 fissions (to be detailed in a separate communication).

In the present study, cell lines RJ2, RT1, RT2.1 and RT2.2 were derived by renucleation of amicronucleates JI and J2 that had already recovered (at 62–116 fissions after enucleation, when only 2 % of the cells of amic J2 possessed slightly abnormal oral apparatuses, Table 1). These renucleated cell lines, as expected, also formed normal oral apparatuses after renucleation (Table 1). However, all of the other transplants were derived by renucleation of amicronucleates J3a and J3b that were still in the depression period (at 15–24 fissions after enucleation). Interestingly, micronucleus reimplantation in these seven cell lines brought about a full recovery in the pattern of oral membranelles in as early as 10–15 fissions after renucleation, in contrast to their amicronucleate progenitors which still exhibited substantial percentages of abnormal oral apparatuses at this time (Table 1, amic J3a, 23 %; amic J3b, 28 %). These amicronucleates had taken 15–30 fissions more to approach full recovery in oral membranellar pattern (Table 1). This shows that micronucleus reimplantation has brought about a faster recovery in oral membranellar pattern. Significantly, the heterospecific micronucleus (from P. tetraurelia) was as competent as the micronucleus from the same species (P. jenningsi) in promoting this recovery.

Apart from the rapid recovery in membranellar pattern, renucleated cell lines from both homo- and heterospecific transplantations also exhibited a prominent increase in oral length compared to amicronucleates. In ten renucleated cell lines (apart from RJ2), the mean oral lengths of cells possessing micronucleus in 26 vegetative samples (out of 28, at different numbers of fissions after renucleation) were significantly longer than those of their amicronucleate progenitors at similar ages (Table 2). Moreover, the highest value of mean oral lengths exhibited by the renucleated cell lines was also greater than the highest value attained by the amicronucleate counterparts. These observations demonstrate that the micronucleus participates in determining oral length, since its presence has allowed the development of longer oral apparatuses approaching that in normal P. jenningsi (mean of mean oral length of the various renucleate cell lines = 34·4 μm, range = 30·0—37·3μm, Table 2; c.f normal oral length, Table 2, footnotes). The converse of the above observations was exhibited by one renucleated cell line, RJ2; this could be due to abnormal functioning of the transplanted micronucleus in this case.

As with the recovery of normal membranellar pattern, the recovery of oral lengths following micronucleus reimplantation was rapid. This is best shown in the renucleated cell lines RJ3a, RJ3b and RT3a.l-RT3b.3 as these were renucleated with amicronucleates (J3a and J3b) within the depression period. At 10–15 fissions after renucleation, these renucleated cell lines already attained the mean oral lengths of 32·2–35·9 μm, while their amicronucleate counterparts were only of 26·3–26·9 μm. Their oral lengths were even longer than those attained by the amicronucleates at 30 fissions later (Table 2, amic J3a, 31·5 μm-, amic J3b, 26·5μm).

Again, in both homo- and heterospecific transplants the increase in oral length was substantial. The effectiveness of the two types of micronucleus reimplantation was analysed by comparing their mean oral lengths (one-tailed r-test). To avoid interclonal variations of oral length and also intraclonal variations arising from different clonal ages, renucleated cell lines belonging to the same clone (i.e. RJ3a vs RT3a.l and RT3a.2; RJ3b vs RT3b.l, RT3b.2 and RT3b.3) and at similar ages were compared (Table 2, see footnote c). In 4 of the 15 cases analysed, the mean oral length of the cell line renucleated with micronucleus from the same species was significantly greater than that in the heterospecific transplant; in 5 other cases there was no significant difference; the remaining 6 cases showed the converse. This ratio, 4:5:6, is essentially 1:1:1. There is thus no indication that the heterospecific micronucleus was different from the micronucleus of the same species in their stomatogenic function.

Because of postautogamous death in the homospecific transplants (see below: section 2 of ‘Sexual Reproduction’), the possibility exists that the transplanted micronuclei were damaged during the operation, probably due to their large size. This raises the concern as to whether the stomatogenic functions of these micronuclei were impaired, and hence the oral lengths attainable by homospecific transplantation may be underestimated in the present study. While this may well account for the situation in RJ2, we believe that in RJ3a and RJ3b the micronuclear stomatogenic functions were largely normal, for the following reason. In a previous study, homospecific micronuclear transplants of P. tetraurelia maintaining normal micronuclei (as demonstrated by good post-autogamous survival) nevertheless possessed oral apparatuses only of near-normal length (Ng & Tam, 1987). It was concluded that nonmicronuclear factors had contributed to the subnormal oral length in the transplants and that the micronuclear stomatogenic functions in such transplants were largely normal. In the present study, near-normal oral lengths were attained in both homo- and heterospecific transplants (apart from RJ2). It thus seems to be the case that the micronuclei in these transplants were able to exercise normal stomatogenic functions. In any case, it is remarkable that a P. tetraurelia micronucleus can promote near-normal recovery of oral length characteristic of P. jenningsi, in view of the large difference in oral length of the two species (P. tetraurelia, 28 μm; P. jenningsi, 36 μm).

All of the renucleated -cell lines exhibited micronuclear abnormalities during autogamy. This was characterized by the failure of some of the postautogamous cells to possess either micronuclei or macronuclear anlagen, with the latter situation being more frequent, and others possessing none (Table 3, +,— and–, –classes). Even in cells possessing both micronuclei and macronuclear anlagen, some did not contain the normal complement of two micronuclei and two macronuclear anlagen (2,2). Instead, they exhibited 1,1; 3,3; 2,1; 3,1; 1,3 and rarely 4,4 nuclear constitutions.

In the majority of cases analysed (10 out of 13), the percentages of postautogamous cells exhibiting abnormal nuclear constitutions in the heterospecific transplants were significantly greater compared to the homospecific transplants (Table 3; see footnote d for details). This shows that the heterospecific micronuclei (from P. tetraurelia) behaved more abnormally in nucleogenesis compared to the homospecific ones (from P. jenningsi) in the cytoplasm of P. jenningsi.

The macronuclear anlagen formed in the homo- and heterospecific transplants were morphologically different and conformed to their specific origins. In the former, the anlagen were characterized by the presence of ‘chromatinic centres’ resembling bull eyes (Fig. 8), typical of P. jenningsi (Diller & Earl, 1958; Mitchell, 1963). The latter had homogeneous anlagen resembling that of the micronucleus donor, P. tetraurelia (Fig. 9). Both types of anlagen in the transplants failed to support somatic functions resulting in 100 % postautogamous death.

We have shown previously in P. tetraurelia that, during sexual reproduction, the micronucleus mediates an early stomatogenic step of oral membranelle assembly (‘initiation’); in amicronucleates, stomatogenesis becomes arrested at this crucial step resulting in astomy and death (see Introduction). This unique stomatogenic function of the micronucleus was also demonstrated by amicronucleate P. jenningsi cell lines in autogamy (to be detailed separately in another report). The importance of the presence of the micronucleus in the renucleated cell lines was further demonstrated by isolating, from renucleated cell line RT1, five vegetative cells that had lost the renucleated micronuclei and expanding these into cell lines; they also failed to initiate oral membranelle assembly and became astomatous after autogamy (total 227 cells sampled). In contrast, in sublines of both homo- and heterospecific micronuclear transplants maintaining micronuclei the development of a new oral apparatus was possible in some, though not all of the autogamous cells, so that the postautogamous culture contained feeders, besides the thin, deformed astomatous nonfeeders. This shows that the presence of the micronucleus, even of another species, can mediate the initiation of oral membranelle assembly and the subsequent development of the oral apparatus. Clearly, the micronuclear signal for this crucial stomatogenic step is not speciesspecific.

The astomatous postautogamous cells of the transplants had originated from two sources. As noted above, the renucleated cell lines generated some amicronucleates during asexual propagation. Obviously, some of the postautogamous astomatous cells were derived from such amicronucleates present in thepreautogamous culture. Others were generated as a result of abnormal nucleogenesis in micronucleates during autogamy, in particular when the micronuclear cycle had become arrested after meiosis so that none of the postmeiotic micronuclear derivatives were retained for further nuclear development (see Tam & Ng, 1986; Chau & Ng, 1988a). Because of the heterogeneous origin of the astomatous postautogamous cells, the exact percentage of the latter type was not assessed in the present study. Inspection of the postautogamous cultures of the heterospecific transplants revealed that up to 50 % of the cells were feeders in some cases. This indicates that in a significant proportion of the autogamous cells of the heterospecific transplants, the micronucleus was competent in mediating the initiation of oral membranelle assembly. This ability, however, was not compared in the homo- and heterospecific transplants.

The postautogamous feeders were selectively harvested for silver impregnation, to allow a comparison between the homo- and heterospecific transplants in their oral apparatus and in nucleogenesis (see sections 3 and 4 below).

Some postautogamous cells possessing oral apparatus nevertheless did not bear any micronuclei or macronuclear anlagen (Table 3, —, —cases). It is likely that, in these, the micronucleus had exercised its stomatogenic function (in the initiation of oral membranelle assembly) before disappearing later in the nucleogenic cycle. This aspect has been explored and discussed in some detail in another study employing P. tetraurelia cell lines bearing defective micronuclei (Chau & Ng, 19886, and unpublished data).

The new oral apparatuses formed in the autogamous renucleated cell lines were not always normal (Table 3). The two types of transplants gave rise to similar types of oral abnormalities, including fragmentation of oral membranelles (involving the quad-rulus, but more often also dorsal and ventral peniculi; Fig. 10), spreading out of the anterior basal body rows of the dorsal peniculus (Fig. 11), and reduction of the length of oral membranelles and buccal cavity. These oral defects were similar to those exhibited by cell lines possessing defective micronuclei in autogamy (Tam & Ng, 1986; Chau & Ng, 1988a,b, and unpublished data), suggesting that the oral abnormalities in the renucleated cell lines had also stemmed from abnormal micronuclear functioning in the sexual cycle.

Unlike nuclear reorganization, postautogamous cells from the homo- and heterospecific transplants exhibited similar degrees of oral membranellar abnormalities. When the percentages of postautogamous cells possessing defective oral apparatuses in the two types of transplants from the same clone were compared, no significant difference was found in ten cases (Table 3, see footnote c for details). In two other cases, the heterospecific transplants showed significantly higher percentages of oral abnormalities, while the remaining one case showed the converse. In addition, there was probably only a slight indication, if any, of more oral abnormalities in the heterospecific transplants, in which the highest percentage of oral abnormalities was greater than that’observed in the homospecific transplants (RT3a.2, 24% vs RJ3b, 14%). This analysis therefore shows that whenever an oral apparatus was formed in the transplants in autogamy, the probability of producing an abnormal oral membranellar pattern and the types of abnormalities produced were about the same in the two types of transplants. This suggests that the micronuclear stomatogenic signals in the patterning of the oral membranelles during sexual reproduction are largely similar in the two species.

The oral apparatuses formed in autogamy in the renucleated cell lines were mostly significantly longer than that of P. tetraurelia (14 out of 15 cases of the heterospecific transplants), but shorter compared to the micronucleate P. jenningsi controls (13 out of 21 cases of homo- and heterospecific transplants, Table 4). As in asexual reproduction, the homo- and heterospecific transplants produced oral apparatuses of similar lengths during autogamy (Table 4). Out of 13 cases compared, in 4 cases the oral lengths attained by the homospecific transplants was significantly greater. In another 4 cases, there was no significant difference between the two. The remaining 5 cases demonstrated the converse. This ratio, 4:4:5, is obviously not different from 1:1:1. Thus, the heterospecific micronucleus was as effective as the homospecific one in the provision for the development of normal oral length. It is of interest to emphasize that the oral length exhibited by the heterospecific transplants were mostly significantly longer than that of the P. tetraurelia micronucleus donors, and approaching that of the recipient species, P. jenningsi (Table 4, mean of mean oral length of the cell lines from heterospecific transplantation, 32·0μm, range, 28·5–34·8 μm).

In the present study, homo- and heterospecific micro-nuclear transplantations entailed similar stomatogenic consequences in asexual as well as sexual reproduction. During vegetative propagation, the heterospecific transplants were as competent as the homospecific transplants in effecting recovery of the membranellar pattern to normal and in allowing the development of longer oral apparatuses with length approaching that of micronucleate P. jenningsi. Moreover, the recovery in both types of transplants was rapid, taking only 10–15 fissions.

During autogamy, micronuclei from P. tetraurelia could trigger the initiation of oral membranelle assembly and consequently the formation of functional oral apparatuses in the heterospecific transplants. As in asexual reproduction, the oral lengths attained by the heterospecific micronuclear transplants were not significantly different from those reimplanted with micronuclei from the same species and, remarkably, they approached the oral length of the cytoplasmic recipient (P. jenningsi) instead of the micronucleate donor (P. tetraurelia). In addition, similar percentages and types of oral membranellar abnormalities in postautogamous cells were demonstrated by the homo- and heterospecific transplants. These observations allowed the following conclusions on micronuclear stomatogenic functions to be made.

The development of an oral apparatus during vegetative propagation and autogamy can be subdivided into two stages: the critical step of initiation of oral membranelle assembly (“initiation”) and subsequent stomatogenic developments. In previous studies, the micronucleus was shown to exercise control over the initiation of oral membranelle assembly during sexual but not asexual reproduction, and it also affected the patterning of the oral apparatus during both cycles (see Introduction).

The ability of a heterospecific micronucleus to trigger oral membranelle assembly during sexual reproduction in the cytoplasm of a different species is particularly revealing. This clearly indicates that the micronuclear stomatogenic controls over the initiation step in the two Paramecium species have much in common. The similarity of this micronuclear stomatogenic function in the two species suggests that this micronuclear function is fundamental to the genus.

The other micronuclear stomatogenic functions, in the control of oral membranellar pattern during cell division, are also conserved. Both homo- and heterospecific micronuclear transplantations conferred equally rapid recovery of membranellar pattern to near-normal condition, within 10–15 fissions of propagation of the transplants. Additionally, micronuclear control over the patterning of the oral apparatus during the sexual cycle can also be interpreted as conserved, since the percentages and types of oral membranellar abnormalities exhibited by postautogamous cells from homo- and heterospecific transplants were similar. Understandably, this last observation is open to other explanations, since similarity in oral abnormality in the sexual cycle in the two types of transplants does not necessarily imply similar modes of micronuclear controls in them. The comparison of the nature of micronuclear controls on oral patterning during autogamy in the two species therefore awaits further investigation.

The micronucleus also plays a role in determining the length of the oral apparatus during asexual and sexual reproduction (see Introduction). Heterospecific micronuclear transplantation is particularly illuminating in this regard. The micronucleus donor, P. tetraurelia, has a much shorter oral apparatus, compared to that of the recipient, P. jenningsi (28 vs 36μm). The heterospecific transplants thus furnished a test of the efficacy of micronuclear control over oral length.

Both homo- and heterospecific transplants developed oral apparatuses comparable in length to that of the cytoplasmic recipient P. jenningsi, in asexual and sexual cycles. Hence, the function of the micronucleus in the determination of oral length is not species-specific. Whereas the micronucleus provides the necessary conditions for the development of normal oral length, the oral length characteristic of a species is determined by nonmicronuclear factors. A macronucleus-acting gene, shl, has been shown to be involved in the determination of oral length in P. tetraurelia (Tam & Ng, 1987).

In the studies of Ng (1981) and Tam & Ng (1987) employing P. tetraurelia, about two thirds of cell lines generated by homospecific micronuclear transplantation were normal in the propagation of the micronucleus during cell division. In addition, nuclear reorganization during the sexual cycle was also mostly normal, as shown by the high survival rate and normal expression of specific genes derived from the implanted micronuclei. However, in the present study, all of the homo- and heterospecific renucleated cell lines were abnormal in micronuclear propagation during vegetative division and in nuclear reorganization during sexual reproduction. All postautogamous cells of these transplants died, except when there was regeneration of fragments of the recipients’ presexual macronuclei. Could these micronuclear abnormalities be the result of damage of the micronuclei during transplantation? While this could well be the reason in the homospecific transplants, because of the large size of the P. jenningsi micronucleus, we do not think this accounts for the situation in the heterospecific transplants, as the size of the P. tetraurelia micronucleus is much smaller. Moreover, the heterospecific transplants exhibited a higher percentage of abnormalities in nuclear reorganization in the sexual cycle, suggesting interspecific incompatibility between the micronucleus and the recipient. It has been well documented that the micronucleus comes under the influence of the cytoplasm during nuclear reorganization in different phases of the sexual cycle: the progress through premeiotic S phase (Fujishima & Hiwatashi, 1978); the survival of one postmeiotic product in the paroral cone (Sonneborn, 1954; Yanagi & Hiwatashi, 1985); the development of the zygotic nucleus (Harumoto & Hiwatashi, 1982) and the differentiation of macronuclear anlagen (Mikami, 1980; Grandchamp & Beisson, 1981; Mikami & Ng, 1983). Moreover, incompatibility has also been shown in interspecific conjugation between sibling species of the Paramecium aurelia complex (see Sonneborn, 1974), and also in interspecific nuclear transplantations in Amoeba and in amphibians (reviewed by Jeon & Lorch, 1979; Gallien, 1979).

The present finding is analogous to heterospecific (e.g. chick-duck; chick-mouse) dermal-epidermal inductive interactions in the development of feathers and hairs (reviewed by Sengel, 1971; Deuchar, 1975). The mesodermal inductive stimulus on the differentiation of epidermal feather or hair was shown to be common to both species, in addition to the involvement of other species-specific factors. Similarly, the Paramecium micronucleus exerts an ‘intracellular inductive stimulus’ common to the two species on the oral anarchic field to promote oral development, but the final length of the oral apparatus is determined by species-specific factors residing outside the micronucleus. This analogy is particularly interesting in view of the close spatial relationship between the oral anarchic field and the micronuclear postmeiotic divisional derivatives at the time when the latter are supposed to execute their stomatogenic functions (see Introduction).