233
Genome Downloaded from www.nrcresearchpress.com by DIRECTORATE OF COLDWATER FISHERIES RES on 10/15/15 For personal use only.
ARTICLE Incomplete pole orientation of kinetochores in complex meiotic metaphase I configurations delays metaphase–anaphase transition in Secale J. Sybenga
Abstract: To prevent unbalanced chromosome segregation, meiotic metaphase I – anaphase I transition is carefully regulated by delaying anaphase until all kinetochores are well oriented (anaphase checkpoint) in mammals and insects. In plants this has not yet been established. In heterozygotes of two reciprocal translocations of Secale cereale, with one chromosome replaced by its two telocentric arms, anaphase delay was correlated with the orientation of the kinetochores of the complex of five chromosomes. The terminal kinetochores of the half chromosomes were readily elongated and pole oriented. Chains of five chromosomes with all five kinetochores orienting on alternate poles where the first to start anaphase. Kinetochores of two adjacent chromosomes when oriented on the same pole were partly shielded and less well pole directed. Anaphase was delayed. Cells with this configuration accumulated during anther development. Kinetochores in metacentric chromosomes lacking chiasmata in one arm (in trivalents and bivalents) were slightly better pole oriented and delayed anaphase less. Release of chromatid cohesion as triggered by kinetochore stretch is apparently delayed by inadequate exposition and pole orientation of the kinetochores. It is a mild form of an anaphase checkpoint, in normal material synchronizing bivalent segregation. Key words: meiosis, multivalent orientation, kinetochore activity, anaphase checkpoint, Secale. Résumé : Afin de prévenir une ségrégation chromosomique inégale au cours de la méiose, la transition métaphase I – anaphase I est étroitement régulée en retardant l’anaphase I jusqu’a` ce que tous les kinétochores soient bien orientés (checkpoint de l’anaphase) chez les animaux et les insectes. Chez les plantes, cela n’a pas encore été démontré. Chez des plantes de seigle (Secale cereale) hétérozygotes pour deux translocations réciproques, où l’un des chromosomes était remplacé par deux bras télocentriques, le délai a` l’entrée en anaphase était corrélé avec l’orientation des kinétochores du complexe formé de cinq chromosomes. Les kinétochores terminaux des demi-chromosomes étaient facilement étirés et orientés vers les pôles. Les chaînes formées de cinq chromosomes avec les cinq kinétochores orientés en direction de pôles alternes ont été les premières a` marquer le début de l’anaphase. Les kinétochores de deux chromosomes adjacents, lorsqu’ils étaient orientés vers le même pôle, étaient partiellement voilés et moins bien orientés vers le pôle. L’anaphase était retardée dans ces cas. Les cellules présentant cette configuration s’accumulaient au cours du développement des anthères. Les kinétochores situés sur des chromosomes métacentriques auxquels il manquait des chiasmas sur un bras (au sein de trivalents ou de bivalents) étaient légèrement mieux orientés vers les pôles et entraînaient un moindre délai a` l’entrée en anaphase. Le relâchement de la cohésion des chromatides, déclenché par l’étirement des kinétochores, était apparemment retardé par une exposition et une orientation polaire inadéquates des kinétochores. Il s’agit d’une forme faible de checkpoint de l’anaphase, lequel permet de synchroniser la ségrégation des bivalents chez un matériel normal. [Traduit par la Rédaction] Mots-clés : méiose, orientation des multivalents, activité des kinétochores, checkpoint de l’anaphase, Secale.
Introduction In meiosis the homologous chromosomes pair to form bivalents. Each chromosome consists of two chromatids. Pairing is completed with the formation of the synaptonemal complex where recombinational chromatid exchange is completed. After the protein scaffold of the synaptonemal complex has been shed, the chromosomes become diffuse for a short period, then condense with the start of diplotene. The chromosomes tend to separate, but Rec8 cohesin (Stoop-Myer and Amon 1999) keeps the chromatids together over their entire length. Where chromatid exchange has occurred, chiasmata appear between the chromosomes. As long as the chromatid cohesion is not released, the chiasmata cannot move and the paired chromosomes are kept together. The cell forms two poles outside the nucleus and orga-
nizes the spindle. When the nuclear membrane is dissolved the spindle attaches to the kinetochores of the centromeric region of the chromosomes. The kinetochores of the two chromatids remain as closely associated as the remainder of the chromosome and orient as a unit to the same pole. This “precocious” activity of the cell relative to the chromosomes (“precocity theory”, Darlington 1965), compared with mitosis, makes a functional meiosis possible. The chiasmata keep the paired homologues together in a bivalent, which subsequently lines up in the equator with one chromosome oriented towards each pole. At anaphase I the cohesin between the chromatids along the length of the chromosome is released but not at the centromeric region which does not normally split before the second division.
Received 17 April 2014. Accepted 9 June 2014. J. Sybenga.* Laboratory of Genetics, Wageningen Uiniversity, Wageningen, the Netherlands. E-mail for correspondence: [email protected]. *Present address: Fazantenlaan 4, 4351 RV Veere, the Netherlands. Genome 57: 233–238 (2014) dx.doi.org/10.1139/gen-2014-0067
Published at www.nrcresearchpress.com/gen on 11 June 2014.
Genome Downloaded from www.nrcresearchpress.com by DIRECTORATE OF COLDWATER FISHERIES RES on 10/15/15 For personal use only.
234
Functional anaphase requires synchronization, i.e., early bivalents must wait until the kinetochores of delayed bivalents are well oriented and stretched. In yeast and mammals a clear anaphase checkpoint can be distinguished (Wijnker and Schnittger 2013), and in mammals incomplete activation of a single kinetochore can block anaphase long enough for the cell to fail to complete the cycle and die (De Boer et al. 1986). Kinetochore stretch is associated with dephosphorylation in insects (Li and Nicklas 1997). When all kinetochores are stretched and ready to move, dephosphorylation is completed and the anaphase checkpoint is triggered to start the release of cohesin between the sister chromatids (except in the centromeric region) to allow the homologues to segregate. Bizzari and Marston (2011) reporting on spindle assembly in meiosis in budding yeast emphasize the role of phosphatases again, now in the release of chromatid cohesion after the checkpoint has been triggered. In plants the situation is unclear (Wijnker and Schnittger 2013). Yu et al. (2000) report on proteins in plant kinetochores similar to those that play a role in anaphase checkpoints in mammals and insects. When the chiasmata are resolved, the chromatids stay together and the unsplit chromosomes move to their poles: their number is halved. The present report deals with the question if, when the kinetochores are not optimally oriented and stretched at meiotic metaphase, the transition from metaphase to anaphase is delayed or even blocked in plants. This is best studied in complex configurations in which more than two chromosomes are associated by chiasmata and more than two kinetochores are involved. Some or even all chromosomes then have more than one partner to orient on instead of one, which can result in complications in the orientation of chromosomes in the middle of the complex. Developmental progress of cells with all kinetochores optimally oriented and stretched can then be compared with that of cells where this is not the case. Examples of chromosomal rearrangements with complex configurations are reciprocal translocations and Robertson fusions or Robertson splits or combinations, as in the present report. Two heterozygous translocations were analyzed, with one metacentric chromosome replaced by the two separate arms (Robertson split), each with its own functional centromere: T662W (between chromosomes 3R and 6R, with chromosome 3R split) and T248W (between 1R and 6R and translocated chromosome 1R/6R split). In T662W the long arm of chromosome 6R is involved, in T248W the short arm. The chromosome diagram of T662W with centric split is shown in Fig. 1A, that of T248W in Fig. 1C. In reciprocal translocations non-homologous chromosomes have exchanged segments. When, in heterozygotes, homologous segments pair and form chiasmata, more than two chromosomes become associated into a multivalent (Figs. 1B and 1D). Whereas in bivalents the two centromeres readily establish stable orientation in metaphase with one kinetochore to each pole, in multivalents more than two must find a more complex way to establish stable orientation. Balanced segregation of the participant chromosomes depends on which centromeres eventually move to which pole. Unbalanced segregation products include duplications, deficiencies, and aneuploidy and may result in unbalanced progeny and reproductive impairment. In the present material five chromosomes are involved. When chiasmata are formed between all segments (indicated by X in Figs. 1B and 1D), a chain of five chromosomes is formed with the two telocentric half chromosomes at the ends (Fig. 2). This is the most frequent configuration, but in a considerable number of cases a chiasma lacks in the short translocated segment opposite the centric split, and then a trivalent and a bivalent are formed (Fig. 2D). Rarely, other configurations, especially those lacking another chiasma, or with an extra chiasma are observed. These are not considered. In the way the kinetochores of these configurations establish their orientations, the telocentric chromosomes at the end play
Genome Vol. 57, 2014
Fig. 1. Translocations T662W (3R/6R) with centric split 3R, and T248W (1R/6R) with centric split translocation chromosome 1R/6R. (A) The five chromosomes of T662W. (B) The pairing configuration of (A). (C) The five chromosomes of T248W. (D) The pairing configuration of (C).
an important role. They either orient towards the same pole or towards opposite poles, in both cases apparently well stretched. Considering the chain of five chromosomes, when both telocentrics orient towards the same pole (Fig. 2A), their direct neighbors have a strong tendency to orient towards the other pole and the middle chromosome tends to orient towards the same pole as the end chromosomes, except for a very few cases where this chromosome stays between its neighbors in a variable, probably unstable position. Since alternating chromosomes orient towards the same pole, this configuration is named “alternate”. When in the chain of five, the two end kinetochores orient towards different poles, they are also well stretched, and their orientation is again apparently stable (Fig. 2B). Again the neighboring chromosomes orient towards the opposite pole, but the middle chromosome tends to orient towards the same pole as one of its neighbors. As at least two adjacent chromosomes in the multivalent orient towards the same pole, for simplicity this configuration is named adjacent. With C-banding almost exclusively two types of adjacent can be distinguished (Fig. 2B) due to the different size of the terminal C-bands (Sybenga et al. 1989). A third possibility is the stretching of one chromosome between its two neighbors, which orient towards different poles (linear, Fig. 2C). The kinetochore seems to be inactive. This is relatively rare. Published by NRC Research Press
Genome Downloaded from www.nrcresearchpress.com by DIRECTORATE OF COLDWATER FISHERIES RES on 10/15/15 For personal use only.
Sybenga
Fig. 2. The metaphase I configurations of T662W with centric split 3R. The C-bands near the distal chiasmata are shown as large circles, the distal chiasmata without C-bands as small circles. (A) Alternate orientation of the chain of five chromosomes (Table 1: V alt). The terminal kinetochores are directed towards the same pole. There is an infrequent variant where the kinetochore of chromosome 6 is not well oriented towards the pole. (B) Two “adjacent” chains of five chromosomes, distinguished by C-banding. The terminal kinetochores are directed towards different poles (Table 1: V adj). (C) Two linear variants of adjacent orientation (Table 1: V linear). (D) Two combinations of a bivalent and a trivalent. Alternate with telocentric kinetochores pointed to the same pole (Table 1: II + III alt) and adjacent with telocentric kinetochores pointed at different poles (Table 1: II + III adj). There are rare variants with a linear trivalent.
235
The stage of development of the cell populations on which the observations are made can relatively simply be determined for male meiosis in plants, where large numbers of partly synchronized cell populations are found in the anthers (e.g., Koduru 1984; Rees and Naylor 1960; Rickards 1983; Sybenga 1968). The numbers of cells at metaphase I, where the observations are made, and at stages later and earlier than metaphase can be counted and from their ratios the stage of development of the anther can be derived. When the frequencies of different orientation types are different between different developmental stages of the cell populations, it may be assumed that the accumulating orientation type has delayed cellular development. The possibility of reorientation can present a problem in the interpretation of changes in the frequency of different orientation types during anther development. The phenomenon is real as has been demonstrated in living cells (Bauer et al. 1961, Sybenga and Rickards 1987) when kinetochores approach the poles too close without sufficient counterforce. Indirect suggestions of reorientation have been presented for instance by Sybenga (1968) and Koduru (1984). The published reports suggest that reorientation is limited to early metaphase I.
Materials and methods Translocations T662W and T248W were induced by pollen irradiation in self-compatible spring type rye (Secale cereale) and selected for easy mitotic classification. The centric splits were derived from a primary and a tertiary trisomic, respectively, which in turn were derived from a spontaneous triploid and a translocation heterozygote. The final karyotypes were produced by hybridization. To preserve the original metaphase I orientation as completely as possible, squashing was mild. Of T662W aceto orcein squashes, of T248W aceto carmine squashes were made. This does not permit the distinction between chromosomes apart from the two telocentric arms. To distinguish between the two predominant types of adjacent orientation of the multivalent, a limited number of Giemsa C-banding preparations were made of T662W. The raw data of T662W were published earlier (Sybenga and de Vries 1987; Sybenga et al. 1989); the analyses and the interpretation are new. All data from T248W and the analysis are new. The developmental stage of the anthers studies was determined exclusively by the fractions of cells at stages later than metaphase I. This implies that the developmental stage of anthers with no or very few anaphase cells is not accurately determined.
Results and discussion
For the bivalent–trivalent combination two orientation types are distinguished, in analogy with the multivalent, for simplicity named alternate and adjacent (Fig. 2D). Very rarely the trivalent shows a linear orientation with the central kinetochore apparently inactive. This is not further considered. In living material delay in reaching anaphase can be studied directly. In fixed material the stage of development of the cells in which the observations are made must be established indirectly. Cells that are delayed are still in metaphase when cells in the same cell population, which are not delayed, have already reached anaphase. Delayed cells then appear to accumulate at late metaphase I. If this accumulation is associated with specific orientation characteristics, these can be correlated with meiotic developmental delay.
Translocation 662 with centric split 3R The observations on T662W are shown in Table 1. The anthers are ordered 1 to 8 according to the stage of development, as measured by the frequency of meiocytes later than metaphase I (cells > MI). In the first column the different configurations are given, in the following columns the numbers or frequencies. Three types of chains of five chromosomes (Figs. 2A, 2B) were found. 1. Alternate (V alt) with the end kinetochores and the central kinetochore oriented towards the same pole (Fig. 2A). In a very small minority the central kinetochore did not clearly point towards a pole. 2. Adjacent (V adj) with the end kinetochores oriented towards opposite poles and two adjacent kinetochores towards the same pole (Fig. 2B). The two types, not distinguished without C-banding, are pooled. 3. Linear (V linear) with the end kinetochores directed to opposite poles and one chromosome next to the end chromosome stretched between its two neighbors (Fig. 2C). The two possible types cannot be distinguished without C-banding. Published by NRC Research Press
236
Genome Vol. 57, 2014
Genome Downloaded from www.nrcresearchpress.com by DIRECTORATE OF COLDWATER FISHERIES RES on 10/15/15 For personal use only.
Table 1. Translocation 662 (3R/6R) with centric split 3R: observations. Anther
1
2
3
4
5
6
7
8
Sum
Configuration V alt V adj sum V fract alt V V linear II+III alt II+III adj sum II+III fract alt II+III sum V+II+III fract V fr alt V fr alt II+III Other Fraction cells > MI cells anal fract > MI
96 54 150 0.64 14 68 61 129 0.527 279 0.537 0.344 0.244 14 0.014 0 319 0
167 127 294 0.568 30 69 49 118 0.584 412 0.713 0.405 0.167 22 0.051 0 511 0
74 22 96 0.770 5 45 50 95 0.473 191 0.502 0.387 0.236 24 0.112 7 232 0.030
16 22 38 0.421 3 33 30 63 0.523 101 0.376 0.158 0.327 8 0.073 115 273 0.421
19 15 34 0.558 2 36 27 63 0.571 97 0.350 0.195 0.371 6 0.058 149 270 0.551
4 38 42 0.095 0 19 32 51 0.372 93 0.451 0.043 0.204 7 0.07 550 734 0.749
10 19 29 0.344 8 22 28 50 0.44 79 0.367 0.126 0.278 2 0.025 351 444 0.790
0 22 22 0 1 11 9 20 0.55 42 0.523 0 0.262 2 0.045 591 657 0899
386 319 705 3.398 63 303 286 589 0.514 1294 0.544 1.660 2.089 85 0.061 1763 3440 3.442
Note: Configurations: V, chain of five chromosomes; II+III, heteromorphic bivalent with trivalent; alt, alternate; adj, adjacent; linear, one chromosome stretched between neighbors; other, cells not belonging to main categories; cells > MI, cells later than metaphase I.
There were two types of the combination of a heteromorphic bivalent and a trivalent. 1. Alternate (II + III alt) with the kinetochores of the telocentrics pointed to the same pole. 2. Adjacent (II + III adj) with the kinetochores of the telocentics pointed to different poles, in analogy with the chain of five, although these terms do not apply exactly to the situation. The very few cells with a linear trivalent are not considered separately. There were a few metaphase I cells with configurations that did not fit the descriptions above (other), usually including a univalent or rare interstitial chiasmata. These are not considered in the analysis. The total number of cells analyzed for determining the stage of development of the anther is slightly larger than the sum of sum V + II + III and other as it includes cells earlier than metaphase I and disrupted cells at metaphase I. Cell populations (anthers) without anaphases may include very early cells, in the tables clustered at 0 in later than MI. The fraction of alternate orientation among the chains of five chromosomes is closely and negatively correlated with the stage of development of the anther (r = –0.863, df = 7, p = 0.001). The regression equation is y = –0.611x + 0.688 (Fig. 3). An analysis of variance (ANOVA) of the fractions of altV shows significant (biologically meaningful) variation in addition to that caused by the regression: p = 0.006. Accounting for the great variation in number of observations per point only slightly increases the significance. An ANOVA does not show the contribution of the individual observations to the variation. Therefore, 2 tests were carried out with the corresponding points on the regression line as reference (expected values). The 2 of altV was 24.470, df = 7, p = 0.003. The individual 2 of anther 2 was by far the greatest, in part as the result of the great weight it carries due to the large number of observations. When this anther is left out, the 2 value is 15.320, df = 6, p = 0.018, still significant. If this anther and other early anthers could have been classified to the left of x = 0, the regression might be slightly different and possibly slightly exponential. The two types of “adjacent” of the chain of five cannot be distinguished in orcein preparations. Sybenga et al. (1989) report on Giemsa C-banded pollen mother cells in the same material. In two early anthers with no cells later than metaphase I, a total of 104 cells with alternate and 76 with adjacent chains of five were observed. In six anthers with a majority of cells later than metaphase
Fig. 3. Regression lines for translocation T662W (open circles), with centric split 3R (662), and translocation T248W (closed circles), with translocation chromosome 1R/3R split (248), for fraction of alternate chains of five chromosomes (abscissa) in the cells with a chain of five, to the fraction of cells later than metaphase I (ordinate, compare Tables 1, 2, and 3).
I, there were 54 alternate and 81 adjacent chains of five. This fits the regression of the orcein material quite well. All chromosome arms have a terminal C-band, except the long arm of 6R. Therefore, the chiasma between 6R and the translocated chromosome 3R/6R is between arms with a C-band, and the chiasma between 6R and the translocated chromosome 6R/3R is not. Thus, when there is a C-band between the adjacent orienting chromosomes, 6R and 3R/6R orient towards the same pole (adjacent a, Fig. 2B left). When there is no C-band between the adjacent orienting chromosomes, 6R and 6R/3R orient towards the same pole (adjacent b, Fig. 2B right). Of the 76 adjacent cells of the early anthers, 57 had adjacent a orientation and 19 adjacent b. Of the 81 adjacent cells of late anthers, 59 cells had adjacent a orientation and 22 adjacent b. As the adjacent a and adjacent b ratios are the same, both types of adjacent apparently react the same to anther development. The two types of adjacent, however, have different frequencies. Published by NRC Research Press
Sybenga
237
Genome Downloaded from www.nrcresearchpress.com by DIRECTORATE OF COLDWATER FISHERIES RES on 10/15/15 For personal use only.
Table 2. Translocation 662 (3R/6R) with centric split 3R. Configuration
Correlation
Intercept
Slope
2
p
Regression
alt V adj + linear V II+III Sum
−0.963 0.600 0.561
0.378 0.176 0.445 1
−0.397 0.218 0.179 0
11 283 30 106 59 954
0.127 MI) in the same anthers. Correlation between the configuration frequencies and frequencies of cells > MI. Regression data (intercept, slope, and regression formula).
Sybenga et al. (1989) ascribe this difference to the difference in distance between the centromeres of 6R and 6R/3R (long arms connected by distal chiasmata) and the distance between 6R and 3R/6R (short arms connected). The shorter the distance, the more the orientations tends to be alternate and stable (Vosselman 1981). The frequency of linear orientation of the chain of five chromosomes is low in most anthers (Table 1), and a statistical analysis of a possible correlation with anther development has limited relevance. The two possible types of linear orientation cannot be distinguished without C-banding. The correlation between the fraction of the linear orientation among the total metaphase I cells analyzed and anther development is negative but negligible (r = –0.157). As a fraction of the chains of five, “linear” is positively correlated with anther development but again not significantly: r = 0.103. Linear orientation may cause some delay but does clearly not accumulate significantly. Possibly, a completely inactivated kinetochore is not a major cause of anaphase delay, which can be attributed primarily to the two incompletely stable kinetochores of the adjacent chains of five. The change in the ratio between alternate and adjacent in the bivalent–trivalent combination (fract alt II + III) with developing anther is negative but not significant: correlation r = –0.357, p = 0.1, regression: y = –0.068x + 0.535. There is no additional heterogeneity: 2 = 3.649, p = 0.819. This is in contrast with the very significant heterogeneity in the frequency of the bivalent–trivalent combination as a whole with anther development (Table 2). This heterogeneity is a reflection of the great variation in chiasma formation in the translocated segments of chromosomes 6R and 3R/6R (Fig. 1). There is no apparent correlation between heterogeneity in chiasma formation and any other parameter in this report. As the differences are negligible the two adjacent and the linear chains of five may be pooled, and similarly, the alternate and adjacent bivalent with trivalent may be pooled. Then, three major categories remain: alternate chain of five, adjacent and linear chain of five, and the combination of a bivalent with a trivalent. The analysis is shown in Table 2.
Translocation T248W with centric split 1R/6R Several anthers at different stages, each with large numbers of cells, have been analyzed. The original data were lost and only the averages of three groups of anthers at different levels of anther development (Table 3) were recovered. In Fig. 3, the regression line and the three individual fractions of T248W (closed circles) are given in addition to the data of the alternate chain of five of translocation T662W (open circles). In T662W the initial frequency of alternate V among the chains of five chromosomes is higher than that of adjacent V (Table 1) and even more so in T248W (Table 3 and Fig. 3). This means that at the start of metaphase I orientation, the kinetochores of the five chromosomes have a combination of distances that favors alternate orientation. The difference between T662W and T248W remains during anther development. With anther development progressing, the frequency of alternate chains of five chromosomes decreases strongly, which is accompanied by an increase of multivalents in adjacent orientation and an, albeit smaller, increase of trivalent–bivalent combinations. No effect of linear orientation of the chain of five could be detected, but the small numbers may not be considered representative. In alternate chains, the three kinetochores of the central metacentrics experience pull from two neighbors and are well oriented and stretched as are the terminal kinetochores of the telocentrics at the ends. Such chains apparently complete metaphase rapidly. In all configurations, the terminal kinetochores of the two telocentric chromosomes were always well stretched with a stable pole orientation, either to the same pole or to different poles. A kinetochore in a metacentric chromosome is in a different position. The orientation of the centromeric regions, and consequently that of its kinetochore, is affected by the angle between the two arms and their directions. When in a chain of five chromosomes both neighbors of a chromosome orient towards opposite poles (Fig. 2A), the angle between the arms is narrow, the kinetochore is exposed and directed straight towards its pole, and a stable situation arises. This is the case for all three metacentric chromosomes in the alternate chain of five. Published by NRC Research Press
Genome Downloaded from www.nrcresearchpress.com by DIRECTORATE OF COLDWATER FISHERIES RES on 10/15/15 For personal use only.
238
When the kinetochore of a chromosome orients towards the same pole as one of its neighbors and the other neighbor orients on the opposite pole (adjacent orientations of the chain of five, Fig. 2B), the angle between the two arms is wide and the kinetochore is “shielded” by the pericentromeric chromatin (Sybenga and Rickards 1987). Its orientation is then bent slightly away for its pole. This makes it difficult to establish a stable interaction with the spindle, in principle creating a situation potentially leading to temporary kinetochore inactivation or even reorientation (Sybenga and Rickards 1987). This in turn delays the necessary conditions for progressing into anaphase. When a chromosome is stretched between two neighbors in a linear configuration and the kinetochore was inactivated, delay is less pronounced. Although the data on which any conclusion can be based are very limited, it is probable that completely inactive kinetochores cause little or no anaphase delay. In the bivalent–trivalent combination both the bivalent and the trivalent have a metacentric chromosome at one end. One arm is connected to another chromosome and directed towards the opposite pole. The other arm is free. The kinetochore is mildly shielded and slightly bent towards one side, but less so than with an adjacent chain of five. Spindle assembly may be slightly disturbed causing the limited anaphase delay observed. The conclusion is justified that when in S. cereale all meiotic metaphase I kinetochores are fully active and stable, anaphase I will be triggered without delay. When some kinetochores are inactive or not stably oriented, anaphase I will be delayed but not prevented. There is no strict checkpoint requiring all kinetochores to be fully active before anaphase starts. Ultimately all cells will pass through anaphase I since the total number of metaphase cells found diminishes with anther development. As adjacent orientation leading to unbalanced segregation does not block cell development, there is no selection against formation of unbalanced progeny as may occur in some mammals (de Boer et al. 1986). In humans, translocation t(10;21) which causes unbalanced progeny with the Down phenotype (Williams et al. 1975), the unbalanced segregation must be the consequence of adjacent orientation, apparently without blocking anaphase. In plants selection against unbalanced male gametes takes place later, especially during pollen growth (Janse 1987), not as a result of anaphase block. A limited delay synchronizes anaphase when submetacentric or subacrocentric chromosomes regularly fail to have a chiasma in one arm and the kinetochore is not fully exposed. There are clear indications that a mild form of anaphase I block occurs in plants.
Genome Vol. 57, 2014
Acknowledgements I thank Erik Wijnker (Strasbourg University, France) and two anonymous referees for many important suggestions for the manuscript, and Hans Jansen (Wageningen University, the Netherlands) for the analysis of variance.
References Bauer, H., Dietz, R., and Röbbelen, C. 1961. Die Spermatocytenteilungen der Tipuliden. III Mitteilung. Das Bewegungsverhalten der Chromosomes in Translokationsheterozygoten von Tipula oleracea. Chromosomen, 12: 116– 189. Bizzari, F., and Marston, A.L. 2011. Cdc55 coordinates spindle assembly and chromosome disjunction during meiosis. J. Cell Biol. 193: 1218–1228. doi:10. 1083/jcb.201103076. PMID:21690308. Darlington, C.D. 1965. Cytology. J. and A. Churchill Ltd., London. De Boer, P., Searle, A.G., van der Hoeven, F.A., De Rooij, D.G., and Beechey, C.V. 1986. Male pachytene pairing in single and double translocation heterozygotes and spermatogenic impairment in the mouse. Chromosoma, 93: 326– 336. doi:10.1007/BF00327591. PMID:3698747. Janse, J. 1987. Certation between euploid and aneuploid pollen grains from a tertiary trisomic of rye, Secale cereale L. Genome, 29(2): 353–356. doi:10.1139/ g87-058. Koduru, P.R.K. 1984. Centromere orientation, reorientation, and segregation in an interchange quadrivalent during anther development in pearl millet. Chromosoma, 90: 94–102. doi:10.1007/BF00292445. Li, X., and Nicklas, R.B. 1997. Tension-sensitive kinetochore phosphorylation and the chromosome distribution checkpoint in praying mantid spermatocytes. J. Cell Sci. 110: 537–545. PMID:9092936. Rees, H., and Naylor, B. 1960. Developmental variation in chromosome behaviour. Heredity, 15: 17–27. doi:10.1038/hdy.1960.53. Rickards, G.K. 1983. Orientation behavior of chromosome multiples of interchange (reciprocal translocation) heterozygotes. Annu. Rev. Genet. 14: 443– 498. doi:10.1146/annurev.ge.17.120183.002303. PMID:6364963. Stoop-Myer, C., and Amon, A. 1999. Meiosis: REC8 is the reason for cohesion. Nat. Cell Biol. 1: E125–E127. doi:10.1038/12956. PMID:10559953. Sybenga, J. 1968. Orientation of interchange multiples in Secale cereale. Heredity, 23: 73–79. doi:10.1038/hdy.1968.6. Sybenga, J., and De Vries, J.M. 1987. Variation in orientation of a multivalent formed by a reciprocal and a Robertsonian translocation in rye. Biol. Zent. bl. 106: 409–415. Sybenga, J., and Rickards, G.K. 1987. The orientation of multivalents at meiotic metaphase I: a workshop report. Genome, 29(4): 612–620. doi:10.1139/g87-102. Sybenga, J., Hsiao, C., and De Vries, J.M. 1989. Meiotic multivalent orientation and cell developmental delay. In Mechanisms of chromosome distribution and aneuploidy. Edited by M.A. Resnick and B.K. Vig. Alan R. Liss, New York. pp. 325–334. Vosselman, L , 1981. Meiotic segregation of five different reciprocal translocations in the onion fly, Hylemya antiqua (Meigen).Chromosoma, 81: 727–738. doi:10.1007/BF00329583. PMID:7472022. Wijnker, E., and Schnittger, A. 2013. Control of the meiotic cell division program in plants. Plant Reprod. 26(3): 143–158. doi:10.1007/s00497-013-0223-x. PMID: 23852379. Williams, J.D., Summitt, R.L., Martens, P.R., and Kimbrell, R.A. 1975. Familial Down syndrome due to t(10;21) translocation: evidence that the Down phenotype is related to trisomy of a specific segment of chromosome 21. Am. J. Hum. Genet. 27: 478–485. PMID:125542. Yu, H.-G., Hiatt, E.N., and Dawe, R.K. 2000. The plant kinetochore. Trends Plant Sci. 5: 543–547. doi:10.1016/S1360-1385(00)01789-1. PMID:11120477.
Published by NRC Research Press
Genome Downloaded from www.nrcresearchpress.com by DIRECTORATE OF COLDWATER FISHERIES RES on 10/15/15 For personal use only.
ARTICLE Incomplete pole orientation of kinetochores in complex meiotic metaphase I configurations delays metaphase–anaphase transition in Secale J. Sybenga
Abstract: To prevent unbalanced chromosome segregation, meiotic metaphase I – anaphase I transition is carefully regulated by delaying anaphase until all kinetochores are well oriented (anaphase checkpoint) in mammals and insects. In plants this has not yet been established. In heterozygotes of two reciprocal translocations of Secale cereale, with one chromosome replaced by its two telocentric arms, anaphase delay was correlated with the orientation of the kinetochores of the complex of five chromosomes. The terminal kinetochores of the half chromosomes were readily elongated and pole oriented. Chains of five chromosomes with all five kinetochores orienting on alternate poles where the first to start anaphase. Kinetochores of two adjacent chromosomes when oriented on the same pole were partly shielded and less well pole directed. Anaphase was delayed. Cells with this configuration accumulated during anther development. Kinetochores in metacentric chromosomes lacking chiasmata in one arm (in trivalents and bivalents) were slightly better pole oriented and delayed anaphase less. Release of chromatid cohesion as triggered by kinetochore stretch is apparently delayed by inadequate exposition and pole orientation of the kinetochores. It is a mild form of an anaphase checkpoint, in normal material synchronizing bivalent segregation. Key words: meiosis, multivalent orientation, kinetochore activity, anaphase checkpoint, Secale. Résumé : Afin de prévenir une ségrégation chromosomique inégale au cours de la méiose, la transition métaphase I – anaphase I est étroitement régulée en retardant l’anaphase I jusqu’a` ce que tous les kinétochores soient bien orientés (checkpoint de l’anaphase) chez les animaux et les insectes. Chez les plantes, cela n’a pas encore été démontré. Chez des plantes de seigle (Secale cereale) hétérozygotes pour deux translocations réciproques, où l’un des chromosomes était remplacé par deux bras télocentriques, le délai a` l’entrée en anaphase était corrélé avec l’orientation des kinétochores du complexe formé de cinq chromosomes. Les kinétochores terminaux des demi-chromosomes étaient facilement étirés et orientés vers les pôles. Les chaînes formées de cinq chromosomes avec les cinq kinétochores orientés en direction de pôles alternes ont été les premières a` marquer le début de l’anaphase. Les kinétochores de deux chromosomes adjacents, lorsqu’ils étaient orientés vers le même pôle, étaient partiellement voilés et moins bien orientés vers le pôle. L’anaphase était retardée dans ces cas. Les cellules présentant cette configuration s’accumulaient au cours du développement des anthères. Les kinétochores situés sur des chromosomes métacentriques auxquels il manquait des chiasmas sur un bras (au sein de trivalents ou de bivalents) étaient légèrement mieux orientés vers les pôles et entraînaient un moindre délai a` l’entrée en anaphase. Le relâchement de la cohésion des chromatides, déclenché par l’étirement des kinétochores, était apparemment retardé par une exposition et une orientation polaire inadéquates des kinétochores. Il s’agit d’une forme faible de checkpoint de l’anaphase, lequel permet de synchroniser la ségrégation des bivalents chez un matériel normal. [Traduit par la Rédaction] Mots-clés : méiose, orientation des multivalents, activité des kinétochores, checkpoint de l’anaphase, Secale.
Introduction In meiosis the homologous chromosomes pair to form bivalents. Each chromosome consists of two chromatids. Pairing is completed with the formation of the synaptonemal complex where recombinational chromatid exchange is completed. After the protein scaffold of the synaptonemal complex has been shed, the chromosomes become diffuse for a short period, then condense with the start of diplotene. The chromosomes tend to separate, but Rec8 cohesin (Stoop-Myer and Amon 1999) keeps the chromatids together over their entire length. Where chromatid exchange has occurred, chiasmata appear between the chromosomes. As long as the chromatid cohesion is not released, the chiasmata cannot move and the paired chromosomes are kept together. The cell forms two poles outside the nucleus and orga-
nizes the spindle. When the nuclear membrane is dissolved the spindle attaches to the kinetochores of the centromeric region of the chromosomes. The kinetochores of the two chromatids remain as closely associated as the remainder of the chromosome and orient as a unit to the same pole. This “precocious” activity of the cell relative to the chromosomes (“precocity theory”, Darlington 1965), compared with mitosis, makes a functional meiosis possible. The chiasmata keep the paired homologues together in a bivalent, which subsequently lines up in the equator with one chromosome oriented towards each pole. At anaphase I the cohesin between the chromatids along the length of the chromosome is released but not at the centromeric region which does not normally split before the second division.
Received 17 April 2014. Accepted 9 June 2014. J. Sybenga.* Laboratory of Genetics, Wageningen Uiniversity, Wageningen, the Netherlands. E-mail for correspondence: [email protected]. *Present address: Fazantenlaan 4, 4351 RV Veere, the Netherlands. Genome 57: 233–238 (2014) dx.doi.org/10.1139/gen-2014-0067
Published at www.nrcresearchpress.com/gen on 11 June 2014.
Genome Downloaded from www.nrcresearchpress.com by DIRECTORATE OF COLDWATER FISHERIES RES on 10/15/15 For personal use only.
234
Functional anaphase requires synchronization, i.e., early bivalents must wait until the kinetochores of delayed bivalents are well oriented and stretched. In yeast and mammals a clear anaphase checkpoint can be distinguished (Wijnker and Schnittger 2013), and in mammals incomplete activation of a single kinetochore can block anaphase long enough for the cell to fail to complete the cycle and die (De Boer et al. 1986). Kinetochore stretch is associated with dephosphorylation in insects (Li and Nicklas 1997). When all kinetochores are stretched and ready to move, dephosphorylation is completed and the anaphase checkpoint is triggered to start the release of cohesin between the sister chromatids (except in the centromeric region) to allow the homologues to segregate. Bizzari and Marston (2011) reporting on spindle assembly in meiosis in budding yeast emphasize the role of phosphatases again, now in the release of chromatid cohesion after the checkpoint has been triggered. In plants the situation is unclear (Wijnker and Schnittger 2013). Yu et al. (2000) report on proteins in plant kinetochores similar to those that play a role in anaphase checkpoints in mammals and insects. When the chiasmata are resolved, the chromatids stay together and the unsplit chromosomes move to their poles: their number is halved. The present report deals with the question if, when the kinetochores are not optimally oriented and stretched at meiotic metaphase, the transition from metaphase to anaphase is delayed or even blocked in plants. This is best studied in complex configurations in which more than two chromosomes are associated by chiasmata and more than two kinetochores are involved. Some or even all chromosomes then have more than one partner to orient on instead of one, which can result in complications in the orientation of chromosomes in the middle of the complex. Developmental progress of cells with all kinetochores optimally oriented and stretched can then be compared with that of cells where this is not the case. Examples of chromosomal rearrangements with complex configurations are reciprocal translocations and Robertson fusions or Robertson splits or combinations, as in the present report. Two heterozygous translocations were analyzed, with one metacentric chromosome replaced by the two separate arms (Robertson split), each with its own functional centromere: T662W (between chromosomes 3R and 6R, with chromosome 3R split) and T248W (between 1R and 6R and translocated chromosome 1R/6R split). In T662W the long arm of chromosome 6R is involved, in T248W the short arm. The chromosome diagram of T662W with centric split is shown in Fig. 1A, that of T248W in Fig. 1C. In reciprocal translocations non-homologous chromosomes have exchanged segments. When, in heterozygotes, homologous segments pair and form chiasmata, more than two chromosomes become associated into a multivalent (Figs. 1B and 1D). Whereas in bivalents the two centromeres readily establish stable orientation in metaphase with one kinetochore to each pole, in multivalents more than two must find a more complex way to establish stable orientation. Balanced segregation of the participant chromosomes depends on which centromeres eventually move to which pole. Unbalanced segregation products include duplications, deficiencies, and aneuploidy and may result in unbalanced progeny and reproductive impairment. In the present material five chromosomes are involved. When chiasmata are formed between all segments (indicated by X in Figs. 1B and 1D), a chain of five chromosomes is formed with the two telocentric half chromosomes at the ends (Fig. 2). This is the most frequent configuration, but in a considerable number of cases a chiasma lacks in the short translocated segment opposite the centric split, and then a trivalent and a bivalent are formed (Fig. 2D). Rarely, other configurations, especially those lacking another chiasma, or with an extra chiasma are observed. These are not considered. In the way the kinetochores of these configurations establish their orientations, the telocentric chromosomes at the end play
Genome Vol. 57, 2014
Fig. 1. Translocations T662W (3R/6R) with centric split 3R, and T248W (1R/6R) with centric split translocation chromosome 1R/6R. (A) The five chromosomes of T662W. (B) The pairing configuration of (A). (C) The five chromosomes of T248W. (D) The pairing configuration of (C).
an important role. They either orient towards the same pole or towards opposite poles, in both cases apparently well stretched. Considering the chain of five chromosomes, when both telocentrics orient towards the same pole (Fig. 2A), their direct neighbors have a strong tendency to orient towards the other pole and the middle chromosome tends to orient towards the same pole as the end chromosomes, except for a very few cases where this chromosome stays between its neighbors in a variable, probably unstable position. Since alternating chromosomes orient towards the same pole, this configuration is named “alternate”. When in the chain of five, the two end kinetochores orient towards different poles, they are also well stretched, and their orientation is again apparently stable (Fig. 2B). Again the neighboring chromosomes orient towards the opposite pole, but the middle chromosome tends to orient towards the same pole as one of its neighbors. As at least two adjacent chromosomes in the multivalent orient towards the same pole, for simplicity this configuration is named adjacent. With C-banding almost exclusively two types of adjacent can be distinguished (Fig. 2B) due to the different size of the terminal C-bands (Sybenga et al. 1989). A third possibility is the stretching of one chromosome between its two neighbors, which orient towards different poles (linear, Fig. 2C). The kinetochore seems to be inactive. This is relatively rare. Published by NRC Research Press
Genome Downloaded from www.nrcresearchpress.com by DIRECTORATE OF COLDWATER FISHERIES RES on 10/15/15 For personal use only.
Sybenga
Fig. 2. The metaphase I configurations of T662W with centric split 3R. The C-bands near the distal chiasmata are shown as large circles, the distal chiasmata without C-bands as small circles. (A) Alternate orientation of the chain of five chromosomes (Table 1: V alt). The terminal kinetochores are directed towards the same pole. There is an infrequent variant where the kinetochore of chromosome 6 is not well oriented towards the pole. (B) Two “adjacent” chains of five chromosomes, distinguished by C-banding. The terminal kinetochores are directed towards different poles (Table 1: V adj). (C) Two linear variants of adjacent orientation (Table 1: V linear). (D) Two combinations of a bivalent and a trivalent. Alternate with telocentric kinetochores pointed to the same pole (Table 1: II + III alt) and adjacent with telocentric kinetochores pointed at different poles (Table 1: II + III adj). There are rare variants with a linear trivalent.
235
The stage of development of the cell populations on which the observations are made can relatively simply be determined for male meiosis in plants, where large numbers of partly synchronized cell populations are found in the anthers (e.g., Koduru 1984; Rees and Naylor 1960; Rickards 1983; Sybenga 1968). The numbers of cells at metaphase I, where the observations are made, and at stages later and earlier than metaphase can be counted and from their ratios the stage of development of the anther can be derived. When the frequencies of different orientation types are different between different developmental stages of the cell populations, it may be assumed that the accumulating orientation type has delayed cellular development. The possibility of reorientation can present a problem in the interpretation of changes in the frequency of different orientation types during anther development. The phenomenon is real as has been demonstrated in living cells (Bauer et al. 1961, Sybenga and Rickards 1987) when kinetochores approach the poles too close without sufficient counterforce. Indirect suggestions of reorientation have been presented for instance by Sybenga (1968) and Koduru (1984). The published reports suggest that reorientation is limited to early metaphase I.
Materials and methods Translocations T662W and T248W were induced by pollen irradiation in self-compatible spring type rye (Secale cereale) and selected for easy mitotic classification. The centric splits were derived from a primary and a tertiary trisomic, respectively, which in turn were derived from a spontaneous triploid and a translocation heterozygote. The final karyotypes were produced by hybridization. To preserve the original metaphase I orientation as completely as possible, squashing was mild. Of T662W aceto orcein squashes, of T248W aceto carmine squashes were made. This does not permit the distinction between chromosomes apart from the two telocentric arms. To distinguish between the two predominant types of adjacent orientation of the multivalent, a limited number of Giemsa C-banding preparations were made of T662W. The raw data of T662W were published earlier (Sybenga and de Vries 1987; Sybenga et al. 1989); the analyses and the interpretation are new. All data from T248W and the analysis are new. The developmental stage of the anthers studies was determined exclusively by the fractions of cells at stages later than metaphase I. This implies that the developmental stage of anthers with no or very few anaphase cells is not accurately determined.
Results and discussion
For the bivalent–trivalent combination two orientation types are distinguished, in analogy with the multivalent, for simplicity named alternate and adjacent (Fig. 2D). Very rarely the trivalent shows a linear orientation with the central kinetochore apparently inactive. This is not further considered. In living material delay in reaching anaphase can be studied directly. In fixed material the stage of development of the cells in which the observations are made must be established indirectly. Cells that are delayed are still in metaphase when cells in the same cell population, which are not delayed, have already reached anaphase. Delayed cells then appear to accumulate at late metaphase I. If this accumulation is associated with specific orientation characteristics, these can be correlated with meiotic developmental delay.
Translocation 662 with centric split 3R The observations on T662W are shown in Table 1. The anthers are ordered 1 to 8 according to the stage of development, as measured by the frequency of meiocytes later than metaphase I (cells > MI). In the first column the different configurations are given, in the following columns the numbers or frequencies. Three types of chains of five chromosomes (Figs. 2A, 2B) were found. 1. Alternate (V alt) with the end kinetochores and the central kinetochore oriented towards the same pole (Fig. 2A). In a very small minority the central kinetochore did not clearly point towards a pole. 2. Adjacent (V adj) with the end kinetochores oriented towards opposite poles and two adjacent kinetochores towards the same pole (Fig. 2B). The two types, not distinguished without C-banding, are pooled. 3. Linear (V linear) with the end kinetochores directed to opposite poles and one chromosome next to the end chromosome stretched between its two neighbors (Fig. 2C). The two possible types cannot be distinguished without C-banding. Published by NRC Research Press
236
Genome Vol. 57, 2014
Genome Downloaded from www.nrcresearchpress.com by DIRECTORATE OF COLDWATER FISHERIES RES on 10/15/15 For personal use only.
Table 1. Translocation 662 (3R/6R) with centric split 3R: observations. Anther
1
2
3
4
5
6
7
8
Sum
Configuration V alt V adj sum V fract alt V V linear II+III alt II+III adj sum II+III fract alt II+III sum V+II+III fract V fr alt V fr alt II+III Other Fraction cells > MI cells anal fract > MI
96 54 150 0.64 14 68 61 129 0.527 279 0.537 0.344 0.244 14 0.014 0 319 0
167 127 294 0.568 30 69 49 118 0.584 412 0.713 0.405 0.167 22 0.051 0 511 0
74 22 96 0.770 5 45 50 95 0.473 191 0.502 0.387 0.236 24 0.112 7 232 0.030
16 22 38 0.421 3 33 30 63 0.523 101 0.376 0.158 0.327 8 0.073 115 273 0.421
19 15 34 0.558 2 36 27 63 0.571 97 0.350 0.195 0.371 6 0.058 149 270 0.551
4 38 42 0.095 0 19 32 51 0.372 93 0.451 0.043 0.204 7 0.07 550 734 0.749
10 19 29 0.344 8 22 28 50 0.44 79 0.367 0.126 0.278 2 0.025 351 444 0.790
0 22 22 0 1 11 9 20 0.55 42 0.523 0 0.262 2 0.045 591 657 0899
386 319 705 3.398 63 303 286 589 0.514 1294 0.544 1.660 2.089 85 0.061 1763 3440 3.442
Note: Configurations: V, chain of five chromosomes; II+III, heteromorphic bivalent with trivalent; alt, alternate; adj, adjacent; linear, one chromosome stretched between neighbors; other, cells not belonging to main categories; cells > MI, cells later than metaphase I.
There were two types of the combination of a heteromorphic bivalent and a trivalent. 1. Alternate (II + III alt) with the kinetochores of the telocentrics pointed to the same pole. 2. Adjacent (II + III adj) with the kinetochores of the telocentics pointed to different poles, in analogy with the chain of five, although these terms do not apply exactly to the situation. The very few cells with a linear trivalent are not considered separately. There were a few metaphase I cells with configurations that did not fit the descriptions above (other), usually including a univalent or rare interstitial chiasmata. These are not considered in the analysis. The total number of cells analyzed for determining the stage of development of the anther is slightly larger than the sum of sum V + II + III and other as it includes cells earlier than metaphase I and disrupted cells at metaphase I. Cell populations (anthers) without anaphases may include very early cells, in the tables clustered at 0 in later than MI. The fraction of alternate orientation among the chains of five chromosomes is closely and negatively correlated with the stage of development of the anther (r = –0.863, df = 7, p = 0.001). The regression equation is y = –0.611x + 0.688 (Fig. 3). An analysis of variance (ANOVA) of the fractions of altV shows significant (biologically meaningful) variation in addition to that caused by the regression: p = 0.006. Accounting for the great variation in number of observations per point only slightly increases the significance. An ANOVA does not show the contribution of the individual observations to the variation. Therefore, 2 tests were carried out with the corresponding points on the regression line as reference (expected values). The 2 of altV was 24.470, df = 7, p = 0.003. The individual 2 of anther 2 was by far the greatest, in part as the result of the great weight it carries due to the large number of observations. When this anther is left out, the 2 value is 15.320, df = 6, p = 0.018, still significant. If this anther and other early anthers could have been classified to the left of x = 0, the regression might be slightly different and possibly slightly exponential. The two types of “adjacent” of the chain of five cannot be distinguished in orcein preparations. Sybenga et al. (1989) report on Giemsa C-banded pollen mother cells in the same material. In two early anthers with no cells later than metaphase I, a total of 104 cells with alternate and 76 with adjacent chains of five were observed. In six anthers with a majority of cells later than metaphase
Fig. 3. Regression lines for translocation T662W (open circles), with centric split 3R (662), and translocation T248W (closed circles), with translocation chromosome 1R/3R split (248), for fraction of alternate chains of five chromosomes (abscissa) in the cells with a chain of five, to the fraction of cells later than metaphase I (ordinate, compare Tables 1, 2, and 3).
I, there were 54 alternate and 81 adjacent chains of five. This fits the regression of the orcein material quite well. All chromosome arms have a terminal C-band, except the long arm of 6R. Therefore, the chiasma between 6R and the translocated chromosome 3R/6R is between arms with a C-band, and the chiasma between 6R and the translocated chromosome 6R/3R is not. Thus, when there is a C-band between the adjacent orienting chromosomes, 6R and 3R/6R orient towards the same pole (adjacent a, Fig. 2B left). When there is no C-band between the adjacent orienting chromosomes, 6R and 6R/3R orient towards the same pole (adjacent b, Fig. 2B right). Of the 76 adjacent cells of the early anthers, 57 had adjacent a orientation and 19 adjacent b. Of the 81 adjacent cells of late anthers, 59 cells had adjacent a orientation and 22 adjacent b. As the adjacent a and adjacent b ratios are the same, both types of adjacent apparently react the same to anther development. The two types of adjacent, however, have different frequencies. Published by NRC Research Press
Sybenga
237
Genome Downloaded from www.nrcresearchpress.com by DIRECTORATE OF COLDWATER FISHERIES RES on 10/15/15 For personal use only.
Table 2. Translocation 662 (3R/6R) with centric split 3R. Configuration
Correlation
Intercept
Slope
2
p
Regression
alt V adj + linear V II+III Sum
−0.963 0.600 0.561
0.378 0.176 0.445 1
−0.397 0.218 0.179 0
11 283 30 106 59 954
0.127 MI) in the same anthers. Correlation between the configuration frequencies and frequencies of cells > MI. Regression data (intercept, slope, and regression formula).
Sybenga et al. (1989) ascribe this difference to the difference in distance between the centromeres of 6R and 6R/3R (long arms connected by distal chiasmata) and the distance between 6R and 3R/6R (short arms connected). The shorter the distance, the more the orientations tends to be alternate and stable (Vosselman 1981). The frequency of linear orientation of the chain of five chromosomes is low in most anthers (Table 1), and a statistical analysis of a possible correlation with anther development has limited relevance. The two possible types of linear orientation cannot be distinguished without C-banding. The correlation between the fraction of the linear orientation among the total metaphase I cells analyzed and anther development is negative but negligible (r = –0.157). As a fraction of the chains of five, “linear” is positively correlated with anther development but again not significantly: r = 0.103. Linear orientation may cause some delay but does clearly not accumulate significantly. Possibly, a completely inactivated kinetochore is not a major cause of anaphase delay, which can be attributed primarily to the two incompletely stable kinetochores of the adjacent chains of five. The change in the ratio between alternate and adjacent in the bivalent–trivalent combination (fract alt II + III) with developing anther is negative but not significant: correlation r = –0.357, p = 0.1, regression: y = –0.068x + 0.535. There is no additional heterogeneity: 2 = 3.649, p = 0.819. This is in contrast with the very significant heterogeneity in the frequency of the bivalent–trivalent combination as a whole with anther development (Table 2). This heterogeneity is a reflection of the great variation in chiasma formation in the translocated segments of chromosomes 6R and 3R/6R (Fig. 1). There is no apparent correlation between heterogeneity in chiasma formation and any other parameter in this report. As the differences are negligible the two adjacent and the linear chains of five may be pooled, and similarly, the alternate and adjacent bivalent with trivalent may be pooled. Then, three major categories remain: alternate chain of five, adjacent and linear chain of five, and the combination of a bivalent with a trivalent. The analysis is shown in Table 2.
Translocation T248W with centric split 1R/6R Several anthers at different stages, each with large numbers of cells, have been analyzed. The original data were lost and only the averages of three groups of anthers at different levels of anther development (Table 3) were recovered. In Fig. 3, the regression line and the three individual fractions of T248W (closed circles) are given in addition to the data of the alternate chain of five of translocation T662W (open circles). In T662W the initial frequency of alternate V among the chains of five chromosomes is higher than that of adjacent V (Table 1) and even more so in T248W (Table 3 and Fig. 3). This means that at the start of metaphase I orientation, the kinetochores of the five chromosomes have a combination of distances that favors alternate orientation. The difference between T662W and T248W remains during anther development. With anther development progressing, the frequency of alternate chains of five chromosomes decreases strongly, which is accompanied by an increase of multivalents in adjacent orientation and an, albeit smaller, increase of trivalent–bivalent combinations. No effect of linear orientation of the chain of five could be detected, but the small numbers may not be considered representative. In alternate chains, the three kinetochores of the central metacentrics experience pull from two neighbors and are well oriented and stretched as are the terminal kinetochores of the telocentrics at the ends. Such chains apparently complete metaphase rapidly. In all configurations, the terminal kinetochores of the two telocentric chromosomes were always well stretched with a stable pole orientation, either to the same pole or to different poles. A kinetochore in a metacentric chromosome is in a different position. The orientation of the centromeric regions, and consequently that of its kinetochore, is affected by the angle between the two arms and their directions. When in a chain of five chromosomes both neighbors of a chromosome orient towards opposite poles (Fig. 2A), the angle between the arms is narrow, the kinetochore is exposed and directed straight towards its pole, and a stable situation arises. This is the case for all three metacentric chromosomes in the alternate chain of five. Published by NRC Research Press
Genome Downloaded from www.nrcresearchpress.com by DIRECTORATE OF COLDWATER FISHERIES RES on 10/15/15 For personal use only.
238
When the kinetochore of a chromosome orients towards the same pole as one of its neighbors and the other neighbor orients on the opposite pole (adjacent orientations of the chain of five, Fig. 2B), the angle between the two arms is wide and the kinetochore is “shielded” by the pericentromeric chromatin (Sybenga and Rickards 1987). Its orientation is then bent slightly away for its pole. This makes it difficult to establish a stable interaction with the spindle, in principle creating a situation potentially leading to temporary kinetochore inactivation or even reorientation (Sybenga and Rickards 1987). This in turn delays the necessary conditions for progressing into anaphase. When a chromosome is stretched between two neighbors in a linear configuration and the kinetochore was inactivated, delay is less pronounced. Although the data on which any conclusion can be based are very limited, it is probable that completely inactive kinetochores cause little or no anaphase delay. In the bivalent–trivalent combination both the bivalent and the trivalent have a metacentric chromosome at one end. One arm is connected to another chromosome and directed towards the opposite pole. The other arm is free. The kinetochore is mildly shielded and slightly bent towards one side, but less so than with an adjacent chain of five. Spindle assembly may be slightly disturbed causing the limited anaphase delay observed. The conclusion is justified that when in S. cereale all meiotic metaphase I kinetochores are fully active and stable, anaphase I will be triggered without delay. When some kinetochores are inactive or not stably oriented, anaphase I will be delayed but not prevented. There is no strict checkpoint requiring all kinetochores to be fully active before anaphase starts. Ultimately all cells will pass through anaphase I since the total number of metaphase cells found diminishes with anther development. As adjacent orientation leading to unbalanced segregation does not block cell development, there is no selection against formation of unbalanced progeny as may occur in some mammals (de Boer et al. 1986). In humans, translocation t(10;21) which causes unbalanced progeny with the Down phenotype (Williams et al. 1975), the unbalanced segregation must be the consequence of adjacent orientation, apparently without blocking anaphase. In plants selection against unbalanced male gametes takes place later, especially during pollen growth (Janse 1987), not as a result of anaphase block. A limited delay synchronizes anaphase when submetacentric or subacrocentric chromosomes regularly fail to have a chiasma in one arm and the kinetochore is not fully exposed. There are clear indications that a mild form of anaphase I block occurs in plants.
Genome Vol. 57, 2014
Acknowledgements I thank Erik Wijnker (Strasbourg University, France) and two anonymous referees for many important suggestions for the manuscript, and Hans Jansen (Wageningen University, the Netherlands) for the analysis of variance.
References Bauer, H., Dietz, R., and Röbbelen, C. 1961. Die Spermatocytenteilungen der Tipuliden. III Mitteilung. Das Bewegungsverhalten der Chromosomes in Translokationsheterozygoten von Tipula oleracea. Chromosomen, 12: 116– 189. Bizzari, F., and Marston, A.L. 2011. Cdc55 coordinates spindle assembly and chromosome disjunction during meiosis. J. Cell Biol. 193: 1218–1228. doi:10. 1083/jcb.201103076. PMID:21690308. Darlington, C.D. 1965. Cytology. J. and A. Churchill Ltd., London. De Boer, P., Searle, A.G., van der Hoeven, F.A., De Rooij, D.G., and Beechey, C.V. 1986. Male pachytene pairing in single and double translocation heterozygotes and spermatogenic impairment in the mouse. Chromosoma, 93: 326– 336. doi:10.1007/BF00327591. PMID:3698747. Janse, J. 1987. Certation between euploid and aneuploid pollen grains from a tertiary trisomic of rye, Secale cereale L. Genome, 29(2): 353–356. doi:10.1139/ g87-058. Koduru, P.R.K. 1984. Centromere orientation, reorientation, and segregation in an interchange quadrivalent during anther development in pearl millet. Chromosoma, 90: 94–102. doi:10.1007/BF00292445. Li, X., and Nicklas, R.B. 1997. Tension-sensitive kinetochore phosphorylation and the chromosome distribution checkpoint in praying mantid spermatocytes. J. Cell Sci. 110: 537–545. PMID:9092936. Rees, H., and Naylor, B. 1960. Developmental variation in chromosome behaviour. Heredity, 15: 17–27. doi:10.1038/hdy.1960.53. Rickards, G.K. 1983. Orientation behavior of chromosome multiples of interchange (reciprocal translocation) heterozygotes. Annu. Rev. Genet. 14: 443– 498. doi:10.1146/annurev.ge.17.120183.002303. PMID:6364963. Stoop-Myer, C., and Amon, A. 1999. Meiosis: REC8 is the reason for cohesion. Nat. Cell Biol. 1: E125–E127. doi:10.1038/12956. PMID:10559953. Sybenga, J. 1968. Orientation of interchange multiples in Secale cereale. Heredity, 23: 73–79. doi:10.1038/hdy.1968.6. Sybenga, J., and De Vries, J.M. 1987. Variation in orientation of a multivalent formed by a reciprocal and a Robertsonian translocation in rye. Biol. Zent. bl. 106: 409–415. Sybenga, J., and Rickards, G.K. 1987. The orientation of multivalents at meiotic metaphase I: a workshop report. Genome, 29(4): 612–620. doi:10.1139/g87-102. Sybenga, J., Hsiao, C., and De Vries, J.M. 1989. Meiotic multivalent orientation and cell developmental delay. In Mechanisms of chromosome distribution and aneuploidy. Edited by M.A. Resnick and B.K. Vig. Alan R. Liss, New York. pp. 325–334. Vosselman, L , 1981. Meiotic segregation of five different reciprocal translocations in the onion fly, Hylemya antiqua (Meigen).Chromosoma, 81: 727–738. doi:10.1007/BF00329583. PMID:7472022. Wijnker, E., and Schnittger, A. 2013. Control of the meiotic cell division program in plants. Plant Reprod. 26(3): 143–158. doi:10.1007/s00497-013-0223-x. PMID: 23852379. Williams, J.D., Summitt, R.L., Martens, P.R., and Kimbrell, R.A. 1975. Familial Down syndrome due to t(10;21) translocation: evidence that the Down phenotype is related to trisomy of a specific segment of chromosome 21. Am. J. Hum. Genet. 27: 478–485. PMID:125542. Yu, H.-G., Hiatt, E.N., and Dawe, R.K. 2000. The plant kinetochore. Trends Plant Sci. 5: 543–547. doi:10.1016/S1360-1385(00)01789-1. PMID:11120477.
Published by NRC Research Press
