Karyotype analysis of the plant-parasitic nematode Heterodera glycines by electron microscopy: II the tetraploid and an aneuploid hybrid

Cytogenetic studies have revealed different patterns of reproduction in the plant-parasitic nematodes (Triantaphyllou, 1971, 1979)-Some forms undergo meiosis during gametogenesis and reproduce by cross-fertilization (amphimixis) or, in the absence of males, by meiotic parthenogenesis. Other forms do not undergo meiosis and reproduce exclusively by mitotic parthenogenesis. Analyses of the synaptonemal complexes (SC) of pachytene nuclei have revealed certain structural modifications of the SC; e.g. decondensed chromatin regions, SCs without distinct central elements, etc., which appear to be peculiar to nematodes. These modifications may represent steps in the evolution of the ameiotic type of maturation of gametocytes and the establishment of mitotic parthenogenesis (Goldstein & Triantaphyllou, 1978 a, b).

To characterize further the SC complement of plant-parasitic nematodes, a diploid amphimictic species, Heterodera glycines was analysed in the first part of this study and its 9 SCs were reconstructed by electron microscopy of serial sections (Goldstein & Triantaphyllou, 1979). We selected H. glycines because in addition to the prevalent diploid amphimictic form, a tetrapioid (18 bivalents) form was identified recently (Triantaphyllou & Riggs, 1979). Crosses between the diploid and the tetrapioid forms have yielded aneuploid hybrids (14 bivalents) which are viable and have been propagated for many generations (Triantaphyllou, unpublished). The study of all these forms could provide useful information about the behaviour of the SC in different states of ploidy, a situation often encountered in plant-parasitic nematodes. Chromosome pairing in diploid, triploid, and autotetraploid forms of Bombyx mori has also been analysed by electron microscopy and revealed 2 distinct phases of chromosome pairing (Rasmussen, 1976, 1977; Rasmussen & Holm, 1979).

In the present study, we have illustrated the pachytene karyotype of the tetrapioid and aneuploid forms of H. glycines following serial sectioning and 3-D reconstruction. We have also attempted to clarify the mechanism of formation of tetrapioid and hybrid karyotypes.

The tetrapioid population of Heterodera glycines used in this study is the one isolated by Triantaphyllou & Riggs (1979). It has been propagated on soybean seedlings in the greenhouse since its discovery in 1973.

The aneuploid hybrid was produced by crossing females of the diploid population used in the first part of this study (Goldstein & Triantaphyllou, 1979) with males of the tetrapioid population. A cytological study by light microscopy established that the tetrapioid had 18 bivalents, whereas, the hybrid isolate of the present study had 14 bivalent chromosomes at metaphase of the first maturation division of the oocytes. Both tetrapioid and aneuploid forms were viable and reproduced exclusively by cross-fertilization.

The procedures and methods employed for electron microscopy were those reported in the study of the diploid form (Goldstein & Triantaphyllou, 1979). In addition to the nuclei completely reconstructed, numerous nuclei were also examined under the electron microscope to check for consistency in number of modified synaptonemal complexes present. Thus, the data in Tables 1–3 are considered to be typical and representative of the tetrapioid and aneuploid hybrid forms.

The general morphology of the gonad and the formation and behaviour of the synaptonemal complexes in pachytene nuclei in the tetrapioid and hybrid forms are similar to those of the diploid form, as described in the previous article (Goldstein & Triantaphyllou, 1979).

Three pachytene nuclei were completely reconstructed from serial sections: nuclei no. 1 and no. 1 a were adjacent nuclei in the same ovary; nucleus no. 2 was from a different female (Table 1). There are 18 SCs in each pachytene nucleus (compared to 9 in the diploid). Pairing of the chromosomes appears to be complete. The chromatin along the SC is distinctly more condensed than in the diploid (Fig. 2). One end of each SC is attached to the inner nuclear membrane and the other end is free in the nucleoplasm (Fig. 3). Attachment of the SCs appears to be random and no bouquet formation is observed. The relative length of the SCs of the 3 nuclei studied varies considerably more than in the diploid (Table 1) (Goldstein & Triantaphyllou, 1979) since in the diploid the individual SC lengths were similar. The total karyotype length is on the average 268 μm, i.e. about the same as in the diploid.

There are 2 modified SC regions (MSC) in each tetrapioid nucleus. These are identical in structure to those observed in the diploid, as they consist of a heterochromatic ball within which the SC appears disorganized. However, the position of MSC no. 2 in relation to the attachment of the SC on the nuclear envelope is different from that of the diploid (12–22% as compared to 53-55%) (Table 3, p. 232). In each nucleus, one of the MSCs is located on the longest SC, whereas the other is located on a mid-sized SC (Table 1). Both MSCs appear to be of similar sizes.

A large aggregate of chromatin is always displaced to one side of the pachytene nucleus (Fig. 9A-F, p. 235). Within this chromatic mass, there are two separate SCs which vary in length from 7-4 to 9·0 μm (Fig. 9F, Table 1).

The nuclear morphology appears slightly irregular as the nuclear envelope is convoluted and the mitochondria are not normally shaped because of irregular membrane structure (Fig. 2). The nuclear volume is on the average 134 μm³ (Table 1) and this is about one-half the volume of the nucleus of the diploid form.

There are 14 normal SCs, i.e. are attached to the nuclear envelope, in each of the 2 pachytene nuclei of different females studied and the lengths of the SCs range from 3-7 to 94-4 μm (Table 2, Fig. 6). There is no apparent bouquet arrangement of SC ends (Fig. 4). The relative lengths of the SCs in the 2 nuclei are not the same (Table 2). The average total karyotype length is 434 μm. One end of each SC is attached to the inner nuclear membrane and the other end is free in the nucleoplasm (Fig. 4). The location of the nucleolar organizer region (NOR) on the SC suggests that either end of the SC is able to attach to the nuclear envelope (thus, in nucleus no. 1 the NOR is located 84% from the attached end of the SC while in nucleus no. 2 the NOR is located 16% from the attached end of the SC, or 84% from the free end; Table 3). There are segment(s) of SCs that are not attached to the nuclear envelope and are free in the nucleoplasm (Fig. 4). In one nucleus, there is 1 piece, 24 · 9 μ m long and in the second nucleus there are 2 pieces, 10 · 1 and 28 – 4 μ m in length (Table 2). Such unattached SCs were not observed in the diploid or tetrapioid.

SC formation is not normal throughout the entire length of the chromosomes (Fig. 1). Only 108 μ m, i.e. 25%, of the total karyotype length of 434 μ m consists of SCs of normal structure (Table 2). Along the rest of the bivalents, the SC material is present, but poorly organized (Fig. 1). We recognize that oblique views of normal SCs may occasionally appear ‘disorganized’, but this is not the case here.

Seven distinct heterochromatic masses (HM) were observed in each nucleus with a total volume of approximately 1 · 9 μ m³. In nucleus no. 1, 4 HMs are on different SCs and 3 are located on the longest SC (94–4 μm), on which 2 MSCs and the NOR (Nucleolar Organizer Region) are also located. In nucleus no. 2, 6 HMs are located on 6 different SCs and one on the unattached segment. The HMs are always interstitially located on the SCs and never at the ends.

Four modified SC regions are present in each hybrid pachytene nucleus (Table 2), similar in structure to those observed in the diploid and tetrapioid. The relative NOR are located. The other 2 MSCs are on different SCs (Table 2).

Polyploidy is rare among cross-fertilizing animals. The tetrapioid studied here probably represents an autotetraploid of recent origin (Triantaphyllou & Riggs, 1979). Irregularities in chromosome pairing are observed at metaphase I (20% of nuclei observed) when quadrivalents, trivalents and univalent chromosomes are seen in addition to bivalents. Still, telophase I is usually normal and metaphase II and anaphase II are always normal (Triantaphyllou & Riggs, 1979). In spite of the presence of multivalents at metaphase I, 18 normal SCs were observed in the 3 pachytene nuclei studied suggesting that formation of SCs and therefore, pairing of homologous chromosomes, is regular. No multivalent associations were observed at pachytene. Such associations may occur secondarily, at a later stage (e.g. diakinesis), due to attraction of homologous or homeologous chromosomes, and do not represent true pairing. In autotetraploid Bombyx females, multivalent associations are absent at mid-late pachytene and only bivalents are formed (Rasmussen & Holm, 1979).

The tetrapioid has twice the number of SCs as the diploid but the total karyotype length is similar in the tetrapioid and the diploid. This suggests that a possible way of derivation of the form with 18 bivalent chromosomes (18 SCs) may be through fragmentation of the chromosomes of the diploid (9 bivalents). However, the tetraploid form has more than 1-5 times the amount of DNA per nucleus than the diploid, and has juveniles and adults of significantly larger body measurements, as would be expected from a polyploid (Triantaphyllou & Riggs, 1979). It seems that comparisons of SC lengths of the tetrapioid and diploid may be misleading since they may have been measured at different stages of pachytene. Drastic shortening of SC lengths at late pachytene have indeed been reported in Drosophila (Carpenter, 1975); Zea (Gillies, 1973; Maguire, 1978a) and Chinese hamster (Moses, Slatton, Gambling & Starmer, 1977). Furthermore, chromatin is much more condensed in the tetrapioid than in the diploid. It is very likely that chromatin distribution is related to the organization of the SC, and consequently, to the shortening of the total SC length in the tetrapioid.

In the diploid, the NOR and the MSCs were located on different SCs (Goldstein & Triantaphyllou, 1979). In the tetrapioid, the NOR and 1 MSC are located on the same SC. This observation strongly suggests that a chromosomal translocation has occurred in the tetrapioid in addition to the doubling of the chromosome number.

If the tetrapioid has arisen from the diploid via doubling of the chromosomes, 4 MSCs (twice the number present in the diploid) would be expected to be found in the tetrapioid, rather than the two observed. However, a large chromatic association, that is displaced to one side of the nucleus and has 2 SCs within it, has been observed in the tetrapioid nucleus, but not in the diploid or hybrid nuclei. It is possible that the 2 SCs associated with this chromatic mass are actually the 2 missing MSCs, expressed in a different form. This structural change may indicate a change in function, possibly inactivation of the chromatin associated with them. Furthermore, it is possible that the MSCs, in general, are the segments of the karyotype where the sex-determining chromatin is located. The MSC is a unique SC structure and alteration of SC structure usually implies a function. Thus, the chromatic mass, with its 2 SCs would then be the extra sex chromatin in an inactive state. In many other organisms, extra sex chromosomes are not advantageous and may be selectively inactivated or eliminated (White, 1973). The preservation of the normal (diploid) complement of sex chromatin would aid in the viability of the tetrapioid and would allow normal sex expression (males and females, but no intersexes) and amphimictic reproduction. This is especially true in the aneuploid hybrid, which otherwise would have been unbalanced concerning sex factors.

The cross between the diploid and tetrapioid yielded a viable aneuploid hybrid with 14 SCs in the pachytene nucleus. Approximately 25% of the total karyotype of the hybrid consists of regions where the SC has normal structure. Along the rest of the chromosomes (75%), the SCs are poorly organized (Figs. 7, 8). To our knowledge, this situation has not been encountered in any other organism. However, it is known that the process of pairing of homologous chromosomes and the formation of SCs during the first meiotic prophase is influenced by many factors including DNA-to-DNA binding, nuclear envelope-mediated chromosome movement and the unique timing of synaptic competence (Moens, 1973). The condensation of chromatin during prophase I is specific in its rate and extent of coiling and varies from species to species. It may be that the entire segmental association of chromatin with the SC would conceivably be in a state of flux (Maguire, 1978a) with some localized areas (areas where chiasmata occur) that might remain constant. Due to the extensive condensation of the chromatin in the tetrapioid at prophase I, compared to the diploid, the pairing of homologous chromosomes and formation of SCs may not be synchronous in the 2 forms. This possible lack of synchrony of condensation of chromosomes may result in hybrid bivalents with poorly formed SCs since binding sites would only, by chance, be in register. However, short stretches of SC at cross-over sites could serve the demands for homologous pairing (Holliday, 1977) and recombination (Maguire, 1978b).

An interesting feature of the karyotype of the hybrid is that 1 of its SCs has 2 MSCs and the NOR located on it. This SC apparently has resulted from the association of 2 SCs, one from one parent (with one MSC and the NOR) and another from the other parent (with a MSC). Therefore, some chromosomal rearrangements appear to have occurred during the establishment of the aneuploid hybrid. This then would generate non-homologous pairing of chromosomes. The variable location of the MSCs in the hybrid (Table 3) suggests non-homologous pairing.

Unattached pieces of SCs are not present in the diploid or tetrapioid. However, such pieces were observed in nucleus no. 1 of the hybrid (Fig. 9), 1 segment of 24 · 9 μ m; and in nucleus no. 2, 2 segments of 10 · 1 and 28 · 4 μ m. These segments are of considerable length and would appear as individual chromosomes when viewed with the light microscope, thus creating haploid chromosome counts of n = 14, 15 and 16. The extent of normal SC formation is 25%, similar to the rest of the karyotype. None of the 4 MSCs are located on the unattached segment. An explanation of the possible mechanism of formation of these segments is presented in Fig. 8. Pairing between 2 chromosomes of different lengths can be assumed. One of the axes would extend past the other and pair with another unpaired axis from a different chromosome, to which it may be homologous. The exact association between such chromosomes is not defined clearly, since in this nematode unpaired axial cores are not visible. Therefore, the paired segments would appear ‘unattached’ to the rest of the chromosomes.

We thank Mr Eugene McCabe for valuable technical assistance. Part of this work was done in the laboratory of Dr M. J. Moses, Department of Anatomy, Duke University, and we thank him for his cooperation, discussions and review of this manuscript. Financial assistance was provided by the National Science Foundation Grant DEB 76-20968 A02 to A. C. Triantaphyllou and by the International Meliodogyne Project, Contract no. AID/ta8-C-i234.

Paper no. 6171 of the Journal Series of the North Carolina Agricultural Research Service, Raleigh, N.C.