In telocentric trisomics (telotrisomics) of organisms in which the chromosomes normally have two distinct arms, a single chromosome arm with a centromere is present in addition to a complete diploid set of chromosomes. It is the simplest form of polysomy and suitable for analyzing meiotic pairing and recombination patterns in situations where chromosomes compete for pairing. When no suitable meiotic chromosome markers are available, four metaphase I configurations can be distinguished. Their relative frequencies are indicative of the pairing and recombination patterns. In short arm (1RS) telotrisomics of chromosome 1R of rye (Secale cereale) we observed great differences in pairing and recombination patterns among spikes from different tillers and clones of the same plants. Anthers within spikes were only very rarely different. We analyzed a large number of genotypes, including inbreds as well as hybrids. The effects of genetic and environmental conditions on heterogeneity, if any, were limited. Considering that the reproductive tissue of a spike is derived from one primordial cell, it seems that at the start of sexual differentiation there was variation among cells in chromosomal control, which at meiosis determines pairing and crossing-over competence. We suggest that it is an epigenetic system that rigidly maintains this pattern through generative differentiation. In competitive situations the combination most competent for pairing will pair preferentially, forming specific meiotic configurations with different frequencies for different spikes of the same plant. This would explain the heterogeneity between spikes and the homogeneity within spikes. The epigenetic system could involve chromatin conformation or DNA methylation. There were no signs of heterochromatinization.

---

IN organisms with chromosomes that normally have two arms, telocentric trisomics (telotrisomics) have one chromosome arm with a terminal centromere in addition to a complete complement. They are the simplest form of polysomy and particularly suitable for the analysis of pairing and recombination patterns in competitive situations. They are used in gene localization, in gene transfer, and in breaking close linkages (SYBENGA 1992). In a previous article (SYBENGA et al. 2007), we reported mathematical models for estimating pairing and chiasma parameters from metaphase I observations in telotrisomic plants. The models were designed for organisms where insufficient chromosome markers are available or where the morphology of the chromosomes prevents the recognition of specific critical chromosome segments at meiotic metaphase I. We applied the models to two types of telotrisomics involving the short arm of the satellite chromosome (1RS) of rye (Secale cereale), which forms readily recognizable meiotic metaphase I configurations. When the observations started (late 1960s), no useful C-band or other marker polymorphisms were available. In a preliminary analysis of a small part of the same material (SYBENGA 1976), the behavior of the polysomic 1RS appeared to be chaotic, with an unexpected large variation among different spikes of the same plant. This was confirmed in the SYBENGA et al. (2007) analysis using models for estimating pairing preference and chiasma formation parameters as mentioned above, but the pattern of variation was not studied. This analysis is presented here. We now include the comparison of anthers within a spike, where, in contrast with the great differences between tillers of the same plants, no or very little variation was found. In the study of BENAVENTE and ORELLANA (1985) on C-banded metaphase I preparations of marked telotrisomics of rye chromosome 1RS using limited material, few within-plant replications were reported. QI et al. (1988) analyzed synaptonemal complex and metaphase I configurations of C-banded telotrisomics of chromosome 1RS of rye, where one chromosome 1R was replaced by the corresponding two telocentrics (telocentric substitution). Their material was limited in scope and no analysis of heterogeneity was reported. Studies of metaphase I association patterns of unbanded telocentric trisomics in rye and other material have been published (SYBENGA 1975; ALONSO and KIMBER 1983; KIMBER and ALONSO 1984) but not on the same scale, nor with especially prepared material and without paying attention to heterogeneity. In an early study of tetrasomic chromosome arm 1RS of rye, with one isochromosome and two telocentrics, we observed large and systematic within-plant heterogeneity (SYBENGA and VERHAAR 1975). Heterogeneity for different configurations of a translocation heterozygote of barley (Hordeum vulgare L.) (KÜNZEL 1963) and for chiasma frequencies in rye (REES and NAYLOR 1960) were reported. In none of these cases was heterogeneity between anthers of the same spike studied in comparison with heterogeneity between spikes of the same plants or clones. The number of cells and genotypes analyzed was never as large as in this study.

In earlier publications on this and similar material (SYBENGA and VERHAAR 1975; SYBENGA 1976, 1999; BENAVENTE and SYBENGA 2004; SYBENGA et al. 2007), we ascribed the large heterogeneity between spikes of the same plant to some form of epigenetic control of meiotic pairing competence. The increased interest in epigenetic control of genetic processes prompted us to analyze the pattern of heterogeneity in more detail and to try to formulate a hypothesis explaining how this would fit in with molecular models of chromosome pairing and recombination. Two types of telotrisomics were studied:

1. Simple telotrisomics with one telocentric arm 1RS in addition to the two normal chromosomes 1R (Figure 1a).
   
2. Telotrisomics with telocentric substitution, where, in addition to telotrisomy, one chromosome 1R was replaced by the two corresponding telocentrics 1RL and 1RS (telocentric substitution, Figure 1b; compare QI et al. 1988).
   

View larger version (4K): In this window In a new window Download PPT slide   FIGURE 1.— The somatic chromosomes of telotrisomic 1RS (a) and telotrisomic (1RS) with telocentric substitution 1R (b).

 
SYBENGA et al. (2007) defined meiotic "pairing" as a process of mutual approach of homologous chromosomes, potentially leading to a crossover, which is then seen in metaphase I preparations as an attachment ("chiasma") of two or more homologous chromosomes. Since there are three pairing partners in telotrisomics (codes 1, 2, and 3 in Figure 1), with only two ultimately associating to form a chiasma, there are three possibilities of association at any locus: 1 with 2, 3 not paired; 1 with 3, 2 not paired; and 2 with 3, 1 not paired. The observation of partner switch at metaphase I, with a chiasma on each side of the switch, shows that there are at least two segments (A and B in Figure 1) in 1RS where pairing can result in a chiasma and that between them a pairing partner switch can occur. These segments cannot be further defined.

In unmarked material, only four metaphase I configurations could be distinguished (marked l, m, n, and o) in Figure 2. When no chiasma is formed in the long arm, the configurations are slightly different, but can readily be classified in terms of 1RS pairing. The different configurations and their origin were discussed extensively by SYBENGA et al. (2007).

View larger version (8K): In this window In a new window Download PPT slide   FIGURE 2.— Metaphase I configurations l, m, n, and o. (Top) Simple telotrisomic. (Bottom) Telotrisomic with telocentric substitution. Compare with Figure 1. X, chiasma; C,centromere

 
In configuration l, there has been a pairing partner switch between segments A and B (Figure 1) with a chiasma on both sides. All three copies of 1RS are associated by chiasmata to form a branched configuration, but the resulting configurations are different in the two chromosomal types.

In configuration m of the simple telotrisomic, the two short arms of the morphologically identical chromosomes (the complete chromosomes of Figure 1a) have paired and formed a chiasma. In the telocentric substitution type, these are the two telocentrics (Figure 1b). One copy of 1RS is not associated by a chiasma with any other chromosome: the telocentric as a univalent in the simple telotrisomic (Figure 1a) and the short arm of the normal chromosome in the substitution (Figure 1b).

In configuration n, the telocentric is associated with the short arm of a normal chromosome 1RS in the simple trisomic, and the 1RS arm of the other complete chromosome is free. In the substitution trisomic, one telocentric is free, and the other is associated with the complete chromosome 1R. Finally, in configuration o, all 1RS arms lack chiasmata and consequently are not associated with any homolog. Photographs of the prophase and metaphase I configurations of the substitution telotrisomic are presented by QI et al. (1988).

The relative frequencies of the four configurations are determined by the pairing relations between the three arms and the frequency of chiasma formation in the two segments A and B. In turn, with the proper conversion models, pairing and chiasma parameters can be derived from these frequencies (SYBENGA et al. 2007).

We used the same materials and methods as in SYBENGA et al. (2007) and summarize them here. The telocentrics 1RS and 1RL (Figure 1) were derived as simple telocentric trisomics from spontaneous centromere breakage of primary trisomics of 1R, in turn derived from one spontaneous triploid plant, originating from spring-type Petkus rye. The telocentric substitution was formed by combining the long arm and short arm telocentrics by crossing the long arm telotrisomic as the female parent with the short arm telotrisomic as the male parent, because the transmission of the long arm through pollen is very limited. The telotrisomic with telocentric substitution was obtained by crossing the 1RS telotrisomic as the female with the substitution as the male parent.

To be able to study the meiotic behavior of the telotrisomics in different genetic backgrounds, the simple telotrisomics of 1RS and the telocentric substitution were backcrossed for at least six generations with each of four different, highly inbred lines with accession nos. 001, 015, 029, and 200. These lines were originally derived from different accessions of open-pollinated winter-type rye.

Backcrossing telocentric trisomics presents difficulties: the telocentric must recombine with the short arm of the recurrent 1R, but may pair and recombine with the nonrecurrent homolog and then fail to receive the intended segment, or even loose it. Thus, after six generations, it is not certain that the extra chromosome has indeed the desired genetic composition (SYBENGA et al. 2007).

The resulting trisomics were subsequently hybridized with each one of the inbred lines or with the backcrossed substitution lines to obtain inbred and hybrid trisomics of both chromosomal types.

Plants were grown from seeds, and the desired chromosomal types were selected at the seedling stage. To increase the material of each genotype and each plant, the plants were split at the tillering stage, and separate tillers, clones of the original seedlings, were grown in pots, where they continued tillering. The cloned progeny were grown in the greenhouse or, for some genotypes, in growth chambers at 20° or 15° to study possible effects of environmental conditions. Spring-type plants were selected.

Plants in this study were grown and fixations were made in different years, indicated by the first two digits of the plant number and the fixation number.

The stage of meiosis was checked in one of the three anthers of a floret, and if correct, the remaining two were fixed in acetic alcohol 3:1. Anthers from more than one floret of the same spike were included in the same vial with fixative. Each vial was given an individual fixation number, which is the "fixation" referred to in text and tables. Thus, the anthers of one fixation were derived from one spike. Rarely, anthers from a different spike of a tiller from the same pot and flowering at the same time were included.

Usually 100 cells were scored from a single anther, the next 100 from another, etc. Originally, a separate list of scores contained the data of the 100 cells of a single anther, and when these lists were kept, different anthers from the same spike could be compared. This is the case for three genotypes of the telotrisomic with telocentric substitution. However, in many cases the original lists could not be recovered, and only the totals of all anthers studied from a fixation were available, and thus comparisons could be made only between fixations. The genotypic combinations analyzed are shown in Tables 1 and 2 (compare with Figure 1).

View this table: In this window In a new window   TABLE 1 Genotypes of simple telotrisomic 1RS (Figure 1a)

 

View this table: In this window In a new window   TABLE 2 Genotypes of telotrisomic 1RS with telocentric substitution (Figure 1b)

 
Heterogeneity between anthers of the same spike, between different tillers of the same plant, between plants, between environmental conditions (temperatures), and between years (1976 and 1977, when relevant) was measured by ² analyses. Large heterogeneity between fixations within plants may produce spurious significance of differences at the levels of plants, environment, and genotype. This was checked by an analysis of variance (ANOVA).

Simple telotrisomic (Figure 1a, Table 1)

Data for individual fixations are available, but are not presented here. Tables 3–7 give the number of cells for each configuration and their relative frequencies and tests of heterogeneity (over all fixations, between fixations within plants, between plants, between temperatures, and between years where relevant). When relevant, an analysis of variance (ANOVA) was performed to assign the source of heterogeneity to the proper factor.

View this table: In this window In a new window   TABLE 3 Inbred 001 x 001: simple telotrisomic

 

View this table: In this window In a new window   TABLE 4 Inbred 015 x 015: simple telotrisomic

 

View this table: In this window In a new window   TABLE 5 Hybrid 029 x 001: simple telotrisomic

 

View this table: In this window In a new window   TABLE 6 Hybrid 029 x 015

 

View this table: In this window In a new window   TABLE 7 Hybrid 029 x 200

 

Inbreds:

The inbreds were grown in the greenhouse.

Inbred 001 (Table 3):

Two plants of inbred 001 were studied, with four and two fixations, respectively, and 500 cells per fixation. Table 3 shows the total number of cells for each configuration (l, m, n, and o), their relative frequencies (fraction), the overall heterogeneity between all fixations for each configuration and for their sum, and the difference between the two plants and the heterogeneity between fixations within plants.

The overall heterogeneity was large and almost entirely due to large heterogeneity between fixations, representing different clones of the same plants. The difference between plants was significant only for configuration l, caused primarily by a difference in chiasma formation in the interstitial segment B, and almost so for o, the value of which is determined by the overall chiasma frequency.

Inbred 015 (Table 4):

There were five plants of inbred 015 with two fixations each and 500 cells per fixation. Overall heterogeneity over all fixations was large. The heterogeneity between fixations within plants was large for the sum of the configurations and for configuration o. Heterogeneity between plants was considerable, especially for configurations m and n. This made it interesting to check the between-plants variation by an ANOVA. The F-value showed that for only these two configurations (m and n) the difference between plants was significant. The significant between-fixations heterogeneity of o may well have been the reason for the insignificant difference between plants in the ANOVA. A similar but less striking phenomenon is seen for l. The plants were apparently homogeneous for chiasma formation.

Inbreds 029 and 200:

These inbreds will be discussed only briefly. Together with the other genotypes they are reviewed in Table 8.

View this table: In this window In a new window   TABLE 8 Summary of configuration frequencies and heterogeneity for simple telotrisomic 1RS

 
In inbred 029, the heterogeneity between plants was very large for all configurations. The same was true for fixations within plants, but slightly less. In contrast with inbred 015, the analysis of variance showed that the differences between plants were statistically significant for l, n, and o, but not for m, probably because it had the largest heterogeneity between fixations.

With inbred 200, two closely related populations of plants were studied. The overall heterogeneity was large for all configurations. Heterogeneity between fixations within plants was especially large for configurations o and (less so) for m, but not for l and n. Heterogeneity between plants and between the two populations was large, but the analysis of variance showed that the difference between the two populations was significant only for l. Apparently, the low between-fixations heterogeneity of l and the large heterogeneity of the other configurations played a role. The overall variation between plants was not a result of the difference between the populations as such, but of variation between the plants of the second population.

Hybrids:

Of the hybrid plants, separate clones were grown under different temperature regimes (15° and 20°). More fixations were studied than in the inbreds. There were three hybrids, all involving 029 as the mother, which contributed the extra telocentric 1RS.

Hybrid 029 x 001 (Table 5):

The overall heterogeneity was large and variable, as was that between fixations. In the ANOVA, the difference between temperatures was significant for branched trivalents (l), chain trivalents (n), and univalents (o), but not for rings with univalents (m). There was no significant difference between plants, although the heterogeneity ² would suggest so. This must have been a consequence of the large between-fixations heterogeneity.

Hybrid 029 x 015 (Table 6):

There were two populations, which differed by origin. Plant 79369-4 had a translocation in its ancestry, which was lost during backcrossing; plants 79372-4 and 79372-8 did not have a translocation in their ancestry. Clones were grown at 15° and 20°.

The overall heterogeneity was very large. The difference between temperatures might seem significant for o on the basis of the heterogeneity ², but appeared to be due entirely to the heterogeneity between fixations and plants, as is clear from the variance ratios (F-values).

The differences between plants were significant for the branched trivalents (l) and trivalents (n), and the interaction between plants and temperatures for l. As appeared from the number of cells for each configuration, this was due mainly to a difference between the sister plants 79372-4 and 79372-8, not to a difference between the populations.

Hybrid 029 x 200 (Table 7):

The heterogeneity was large, except for configuration l, and was mainly between fixations within the single plant and within temperatures. The difference between temperatures was large only for configuration m, but the ANOVA variance ratio F shows that even here it was not due to intrinsic temperature effects.

Telotrisomics with telocentric substitution (Figure 1b, Table 2)

Inbreds:

Inbred 001 (Tables 9 and 10):

Observations of three individual anthers within fixations were recovered. As referred to above, these had been derived from the same spike or, rarely, from a spike developmentally close to the first. Of each of two anthers, 100 cells were studied; of the third anther, only 50 cells were studied. The number of several configurations was critically small in the sample of 50 cells, so these anthers were not included in the comparisons. In Table 9, the total number of cells for each configuration is given, but the ² analysis refers to the difference between the two anthers with 100 cells each, again for each configuration separately, as well as their sum. Configuration l was excluded as the number of cells with l was always low. As the frequency of configuration l varied, the totals of the remaining configurations were not identical for the different fixations. This had to be taken account of in the calculations. Excluding l is not expected to affect the conclusions.

View this table: In this window In a new window   TABLE 9 Inbred 001 x 001: telotrisomic with telocentric substitution

 

View this table: In this window In a new window   TABLE 10 Inbred 001 x 001: telotrisomic with substitution

 
The two anthers of fixation 76212-1 (plant 76355-1) were very different. The two anthers within all other 15 fixations were strikingly similar, in contrast to the differences between the fixations (Table 10).

Inbred 015 (Tables 11 and 12):

Data for separate anthers were again available. As in inbred 001, anthers with 50 cells and configuration l were omitted from Table 11 and the calculations.

View this table: In this window In a new window   TABLE 11 Inbred 015 x 015: telotrisomic with substitution

 

View this table: In this window In a new window   TABLE 12 Inbred 015x015: telotrisomic with substitution

 
The anthers were very similar in all 12 comparisons, including configuration o, which was very heterogeneous between fixations (Table 11). The ² values were very small, except two that deviated slightly, with P-values of 0.060 and 0.055, in both cases due to only a slight difference in m. For the 12 comparisons this may mean that, if there was a difference between the anthers of the same spike, it was very rare.

There was considerable heterogeneity between fixations (data for anthers pooled) for configurations l (branched) and o, which are determined mainly by the chiasma frequency, but not for m and n, which are determined mainly by the pairing pattern. This implies that the pairing pattern was constant, but that the chiasma frequencies varied. There was hardly any difference between the two plants, the ² being significant only for l, but the ANOVA shows that this was entirely due to between-fixations heterogeneity.

Inbred 029:

One plant (78389-9) was studied. Six fixations were made from clones grown in the growth chamber at 15° and another six fixations were grown at 20° for a total of 3000 cells. No separate data for anthers were available. No details are shown but a summary is given in Table 16.

View this table: In this window In a new window   TABLE 16 Summary of configuration frequencies and heterogeneity of all genotypes of the telotrisomic with telocentric substitution studied

 
The heterogeneity was again great for l and o, less so for m, and absent for n. It was almost entirely due to variation between fixations within growth conditions (temperatures) within clones of the single plant. At both temperatures one very deviating fixation caused much of the heterogeneity. The ² for the effect of temperatures was significant only for m, which was apparently due to the fact that for this configuration the fixations were so homogeneous that even small differences became significant. However, even for m the F-value of the ANOVA turned out to be insignificant.

Hybrids:

Hybrid 001 x 015 (Tables 13 and 14):

Data for separate anthers were available. In Table 13, the number of cells for each configuration for each hybrid is given, in addition to the ² of the differences between two anthers (with 100 cells each and excluding l) for the population grown in 1976 only.

View this table: In this window In a new window   TABLE 13 Hybrid 001x015: telotrisomic with substitution

 

View this table: In this window In a new window   TABLE 14 Hybrid 001x015: telotrisomic with substitution

 
No differences among anthers were observed, including fixations that deviated considerably from the mean. The smallest P-value was 0.182.

Table 14 shows the overall data for the same population (76356) grown in 1976 and for an additional population (77332) grown in 1977, without scores for individual anthers. The parents had the same genetic background. The plants were grown in the greenhouse.

Within the two hybrid populations, the fixations and individual plants were quite homogeneous when one fixation (76058-1) was omitted. This fixation was very different from the remainder. The data of plants within years were pooled for the calculation of the t-value of the ANOVA. The differences between the two hybrid populations (in years) were real and great, in spite of the fact that they were genetically very similar.

Hybrid 015 x 001 (Table 15):

The reciprocal hybrid (015 x 001, Table 15) was grown as two populations in the same two years (1967 and 1977) as hybrid 001 x 015. In population 76357 (five plants, a total of 11 fixations), there was considerable total heterogeneity for the configurations l and o (determinants of the chiasma parameters), not for m and n (primarily determining the pairing parameters). Apparently, the difference between plants (where m was also heterogeneous) was intrinsic and not due to the heterogeneity between fixations, which was limited. In population 77333 (three plants, 16 fixations), configurations l and o were also heterogeneous, in addition to m, which was the only configuration significantly different among the plants. In contrast to 76357, all configurations except n showed significant heterogeneity between fixations. The between-populations (in years) heterogeneity ² was significant for l and m, but in the ANOVA the significance was maintained only for l. In almost all respects, the two reciprocal hybrids between 001 and 015 were different.

View this table: In this window In a new window   TABLE 15 Hybrid 015x001: telotrisomic with substitution

 

Hybrids 015 x 029 and 029 x 015:

In hybrid 015 x 029, the overall heterogeneity was limited to configuration o, primarily due to heterogeneity between plants, where n was slightly heterogeneous. This was confirmed in the ANOVA (F-test), which reflects between-plant heterogeneity rather well because of the limited heterogeneity between fixations.

The reciprocal hybrid was different, with considerable heterogeneity between fixations and plants for several configurations. The remaining hybrids (see Table 1) are reviewed in Table 16.

The most striking phenomenon is the great overall heterogeneity observed. Heterogeneity between fixations within plants and environmental conditions was often the main contributor. Analyses of variance usually showed that differences between plants and between temperatures were not due to intrinsic effects, but to the great variability within these factors. Some between-plant heterogeneity was observed, but significant differences between temperatures (15° and 20°) were rare. Total heterogeneity and heterogeneity between fixations were not evenly distributed over the four configurations, and some genotypes showed more and different distributed heterogeneity than others.

In three genotypes (two inbreds and one hybrid) of the telotrisomic with telocentric substitution, it appeared that, with very rare but striking exceptions, there was complete correspondence between the anthers, in contrast to the usually large differences between fixations. The way fixations were made precludes a complete reconstruction of the anatomical relations between the tissues from which the anthers had developed. The two anthers recorded separately were usually derived at least from the same spike and often from the same floret, but occasionally from a neighboring spike fixed at the same time at the same developmental stage. It was clear that two anthers of the same fixation had a much more recent common developmental (anatomical) ancestry than anthers of different fixations. It is tempting to suggest that in the few cases where the two anthers analyzed were significantly different, they were derived from different spikes.

There was a difference between the two chromosomal types, both in the relative frequencies of the four configurations and in respect to heterogeneity. SYBENGA et al. (2007) found a similar large difference between the two chromosomal types in respect to pairing and chiasma parameters. In the substitution telotrisomic, pairing between morphologically identical, but genetically different, chromosomes was less frequent than random, but in the simple telotrisomic it was more frequent than random. The cause of the apparent relation between pairing preference and heterogeneity in pairing is not at all clear. In the simple telotrisomic, the two morphologically identical chromosomes were two complete chromosomes; in the substitution type, they were two telocentrics. This would imply an effect of the long arm, contradicted by the fact that one of the two 1RS telocentrics in the substitution telotrisomic paired preferentially with a complete chromosome rather than with the other telocentric.

Although rarely published, infrequent configurations and specific patterns of chromosome behavior often occur in clusters (MAGUIRE 1976). The heterogeneity described here may have a similar basis, but involves configurations that are definitely not rare, and the "clusters" are large, corresponding with entire spikes. Because heterogeneity between spikes was large, and within spikes very limited, any mechanism causing differences between spikes must have been initiated at an early stage of spike development, possibly at the time the differentiation of reproductive tissue starts. Previous publications (SYBENGA 1999; BENAVENTE and SYBENGA 2004; SYBENGA et al. 2007) proposed that the cause of these differences would be variable patterns of epigenetic control regulating meiotic pairing and genetic exchange, which were already present in the primordial cells from which the generative tissue of the spike develops. This could be the first stage of the process that, at premeiotic replication, removes blocks that normally prevent somatic pairing and crossing over. In the meiotic cell, the final switch involves all chromosomes, but, at the level of the primordial reproductive cell, the switch is apparently not homogenous.

If we conclude that variation in pairing and chiasma patterns may have an epigenetic component, possible mechanisms are variation in chromosome conformation (caused by post-translational histone modification or otherwise) or methylation of critical transcribed DNA segments. WOLF (1963) could relate the frequency of crossing over in Phryne to heterochromatinization. In our material, heterochromatinization was not observed. WU and LICHTEN (1995) found chromatin structure to be important for the formation of double-strand breaks and the further course of recombination, not directly at the site of the breaks, but at some distance. MIKHAILOVA et al. (1998) showed that the effect of the mutant ph1 on homeolog pairing in wheat (Triticum aestivum) was accompanied by and probably resulted from alteration of chromatin structure. In their material, in addition to changes in chromatin structure in meiotic cells, similar changes were found in tapetal cells, which are somatic, but related to meiotic cells and have the same anatomical origin. The importance of changes in chromosome and chromatin structure during meiosis was further shown by KLECKNER et al. (2004) and ZICKLER and KLECKNER (1999).

Epigenetic control is usually considered to involve regulation of transcription by methylation. There are indications that a relationship exists between chromosome pairing and transcription with pairing sites at or near the promoter (COOK 1997; HABER 1997; SYBENGA 1999; BENAVENTE and SYBENGA 2004). With the increasing interest in and knowledge of small unique noncoding RNAs (WILLINGHAM and GINGERAS 2006) and short DNA segments with regulating properties, short RNAs are an interesting candidate for serving as an intermediate between pairing chromosomes. This fits in with the suggestion that there is a relation between pairing and transcription.

In situations where more than two homologous chromosome segments compete for pairing (trisomics and tetrasomics), the decision of which two ultimately synapse depends on the pairing competitiveness of the segments involved. The choice determines which configuration results at metaphase I. Slight variation in competitiveness between homologous segments can result in considerable differences in metaphase configuration frequencies between different sectors of the same plant. A critical but not homogeneous, epigenetically controlled change in chromosome conformation at the moment of differentiation from the vegetative to the reproductive developmental stage, affecting pairing competence, would explain the heterogeneity between fixations representing different sectors of the same plant.

Important suggestions and comments by C. Heyting, J.H. de Jong, P. Stam, and two anonymous referees are gratefully acknowledged. This work was financed in part by Zuiver Wetenschappelijk Onderzoek (ZWO).

ALONSO, L. C., and G. KIMBER, 1983 A study of genome relationships in wheat based on telocentric chromosome pairing II. Z. Pflanzenz. 90: 273–284.

BENAVENTE, E., and J. ORELLANA, 1985 Evidence for preferential pairing in telotrisomic plants of rye. Heredity 55: 181–186.[CrossRef]

BENAVENTE, E., and J. SYBENGA, 2004 The relation between pairing preference and chiasma frequency in tetrasomics of rye. Genome 47: 122–133.[Medline]

COOK, P. R., 1997 The transcriptional basis of chromosome pairing. J.Cell Sci. 110: 1033–1040.[Abstract]

HABER, J. E., 1997 A super new twist on the initiation of meiotic recombination. Cell 89: 163–166.[CrossRef][Medline]

KIMBER, G., and L. C. ALONSO, 1984 The theory of the pairing of telocentric chromosomes in triploids and trisomics. Can. J. Genet. Cytol. 26: 231–236.[Medline]

KLECKNER, N., D. ZICKLER, G. H. JONES, J. DEKKER, R. PADMORE et al., 2004 A mechanical basis for chromosome function. Proc. Natl. Acad. Sci. USA 101: 12592–12597.[Abstract/Free Full Text]

KÜNZEL, G., 1963 Differenziertes Bindungsverhalten der Meiose-Chromosomen innerhalb der Ähre einer röntgeninduzierten Gerstenmutante mit reziproker translokation in heterozygotem Zustand. Kulturpflanze 2: 517–534.

MAGUIRE, M., 1976 Clustering of specific crossovers in maize microsporocytes. Genetics 82: 19–24.[Abstract/Free Full Text]

MIKHAILOVA, E. I., T. NARANJO, K. SHEPHERD, J. WENNEKES-VAN-EDEN, C. HEYTING et al., 1998 The effect of the wheat Ph1 locus on chromatin organisation and meiotic chromosome pairing analysed by genome painting. Chromosoma 107: 339–350.[CrossRef][Medline]

REES, H., and B. NAYLOR, 1960 Developmental variation in chromosome behaviour. Heredity 15: 17–27.[CrossRef]

QI, L.-L., J. SYBENGA and J. H. DE JONG, 1988 Synaptonemal complex pairing and metaphase I association in a telo-substituted telotrisomic of rye. Chromosoma 97: 88–95.[CrossRef]

SYBENGA, J., 1975 Meiotic Configurations. Springer-Verlag, Berlin/Heidelberg, Germany/New York.

SYBENGA, J., 1976 Quantitative variation in meiotic chromosome pairing affinities within a species: Secale cereale L., pp. 143–150 in Current Chromosome Research, edited by K. JONES and P. BRANDHAM, Elsevier/North-Holland Medical Press, Amsterdam.

SYBENGA, J., 1992 Cytogenetics in Plant Breeding. Springer-Verlag, Berlin.

SYBENGA, J., 1999 What makes homologous chromosomes find each other in meiosis? A review and hypothesis. Chromosoma 108: 209–219.[CrossRef][Medline]

SYBENGA, J., and H. M. VERHAAR, 1975 Meiotic behaviour of single-arm tetrasomic combinations of isochromosomes, telocentrics and normal chromosomes in rye (Secale cereale L.). Chromosoma 53: 295–320.[CrossRef]

SYBENGA, J., H. VERHAAR and D. G. A. BOTJE, 2007 Estimating meiotic chromosome pairing and recombination parameters in telocentric trisomics. Genome 50: 1014–1028.[Medline]

WILLINGHAM, A. T., and T. R. GINGERAS, 2006 TUF love for "junk" DNA. Cell 125: 1215–1220.[CrossRef][Medline]

WOLF, B. E., 1963 Kontrolle des Crossing over von temperaturbedingten Allozykliegrad und von -heterochromatin des X-Chromosoms bei Phryne cincta. Chromosoma 13: 647–701.

WU, T.-C., and M. LICHTEN, 1995 Factors that affect the location and frequency of meiosis-induced double-strand breaks in Saccharomyces cerevisiae. Science 263: 515–518.[CrossRef]

ZICKLER, D., and N. KLECKNER, 1999 Meiotic chromosomes: integrating structure and function. Annu. Rev. Genet. 33: 603–754.[CrossRef][Medline]

Communicating editor: J. A. BIRCHLER

International Access Link

Online ISS 1091-6490
Copyright 2008 by the Genetics Society of America
phone: 412-268-1812   fax: 412-268-1813   email: genetics-gsa{at}andrew.cmu.edu