Protoplasma

, Volume 251, Issue 1, pp 127–143

Meiosis-I in Mesostoma ehrenbergii spermatocytes includes distance segregation and inter-polar movements of univalents, and vigorous oscillations of bivalents

Authors

  • Jessica Ferraro-Gideon
    • Department of BiologyYork University
  • Carina Hoang
    • Department of BiologyYork University
    • Department of BiologyYork University
Original Article

DOI: 10.1007/s00709-013-0532-9

Cite this article as:
Ferraro-Gideon, J., Hoang, C. & Forer, A. Protoplasma (2014) 251: 127. doi:10.1007/s00709-013-0532-9

Abstract

In this article, we describe meiosis-I in spermatocytes of the free-living freshwater flatworm Mesostoma ehrenbergii. The original observations of Oakley (1983, 1985) and Fuge (Eur J Cell Biol 44:294–298, 1987, Cell Motil Cytoskeleton 13:212–220, 1989, Protoplasma 160:39–48, 1991), the first to describe these cells, challenge our understanding of cell division, and we have expanded on these descriptions with the aim of laying the framework for further experimental work. These cells contain three bivalents and four univalent chromosomes (two pairs). Bivalent kinetochores oscillate vigorously and regularly throughout prometaphase, for up to several hours, until anaphase. Anaphase onset usually begins in the middle of the kinetochore oscillation cycle. Precocious cleavage furrows form at the start of prometaphase, ingress and then remain arrested until the end of anaphase. The four univalents do not pair, yet by anaphase there is one of each kind at each pole, an example of “distance segregation” (Hughes-Schrader in Chromosoma 27:109–129, 1969). Until proper segregation is achieved, univalents move between spindle poles up to seven times in an individual cell; they move with velocities averaging 9 μm/min, which is faster than the oscillatory motions of the bivalent kinetochores (5–6 μm/min), and much faster than the anaphase movements of the segregating half-bivalents (1 μm/min). Bipolar bivalents periodically reorient, most often resulting in the partner kinetochores exchanging poles. We suggest that the large numbers of inter-polar movements of univalents, and the reorientations of bivalents that lead to partners exchanging poles, might be because there is non-random segregation of chromosomes, as in some other cell types.

Keywords

Chromosome oscillationsNon-random segregationDistance segregationPrecocious cleavage furrow

Introduction

We have studied meiosis-I in spermatocytes of the freshwater flatworm Mesostoma ehrenbergii (a relative of Planaria) because several unique aspects of this division raise important questions that challenge our understanding of cell division. One of these aspects is distance segregation, the proper segregation of chromosomes that are not conjoined. The usual picture of meiosis is straightforward: chromosome partners (homologues) are attached at chiasmata, the chromosomes are bipolarly oriented, attached to the two spindle poles, and sister kinetochores (on each partner) are syntelically oriented such that once the partners disjoin the spindle fibres attached to each pole cause the partners to move to opposite poles. In many cell types, however, segregating chromosomes are not attached to each other (e.g. Camenzind and Nicklas 1968; Hughes-Schrader 1969; Forer and Koch 1973; Oakley 1983; Oakley 1985) yet their segregation nonetheless is accurate. In crane-fly spermatocytes, for example, the X and Y univalent chromosomes are not attached, they move separately on the spindle during prometaphase, and when they reach the equator at metaphase both univalent chromosomes have spindle fibre attachments to both poles (amphitelic orientation). As the two sex chromosomes move to opposite poles in anaphase (after the completion of autosomal anaphase) the spindle fibres to both poles persist: as each one moves poleward during anaphase, one of its attached spindle fibres shortens and the other elongates (e.g. Forer, Ferraro-Gideon and Berns 2013). The major puzzle of distance segregation in these cells is how each sex chromosome segregates to its respective spindle pole despite them not having been paired. It is unclear what mechanism is utilized by the sex chromosomes to ensure that movements are to opposite spindle poles, nor is it clear how the two sex chromosomes influence each other’s movements throughout anaphase (e.g. Forer and Koch 1973 and Forer et al. 2013). This example, one of many, raises the issue of how movements of not-attached chromosomes are coordinated to achieve distance segregation and what “communication” takes place between chromosomes that are not attached. An equally challenging univalent segregation pattern occurs in Mesostoma spermatocytes, as previously described by Oakley (1983, 1985), whose conclusions we now summarise.

Mesostoma spermatocytes have three autosomal bivalents and four univalents. The univalents appear as two pairs, two of each kind, distinguishable by shape and position of their kinetochores (Husted and Ruebush 1940). At the start of spindle formation, the univalents move to the two spindle poles as the bivalents become bipolarly oriented. The univalents remain at the spindle poles throughout prometaphase and metaphase but sometimes they move between the poles. At anaphase each pole is associated with one of each kind of univalent but early in division there often is “mis-segregation” of univalents: three at one pole and one at the other, or two of the same kind at each pole. The movements between poles thus seem to correct the mis-segregation and allow proper segregation of the univalents. Oakley speculated that not only do the univalents assort one of each kind at each pole, but they do so non-randomly (e.g. female derived univalents at one pole and male derived univalents at the other), the kind of non-random segregation that is well documented in other cells [e.g. Gryllotalpa hexadactyla (mole crickets) (Camenzind and Nicklas 1968); Sciara (Gerbi 1986) and mealy-bugs (Nur 1982; Schrader 1921)]. Oakley based this suggestion on two observations: (1) that univalents seem to move between poles more often than necessary in order to achieve one of each kind at each pole, and (2) that in some cells univalents of the same kind changed poles after proper segregation had been achieved. In this article we describe inter-polar movements of univalents throughout prometaphase as they move from one spindle pole to the other, prior to the onset of anaphase. We describe the frequency and velocities of univalent movements, and whether they affect bivalent kinetochore oscillations.

A second remarkable feature of Mesostoma spermatocytes is the presence of a precocious cleavage furrow. The cleavage furrow begins ingression shortly after the bivalents become bipolarly oriented in prometaphase and the furrowing then becomes arrested shortly thereafter. The furrow can gradually become slightly more constricted as the cell gets closer to anaphase but the furrow continues its ingression and cleaves the cell only after anaphase (Forer and Pickett-Heaps 2010). This seems remarkable in itself since most textbooks and articles assume that the stimulus for cleavage furrow ingression is released after anaphase onset. Almost more remarkable is that the position of the furrow along the length of the cell changes when a univalent moves from one pole to the other (Forer and Pickett-Heaps 2010), when a bivalent re-orients, or when a component of the spindle is altered by UV microbeam irradiation (unpublished data). We do not deal with this phenomenon herein, except in passing.

A third noteworthy feature of Mesostoma spermatocytes is the vigorous and persistent oscillations of bivalent kinetochores, as described by Fuge (1987, 1989). Oscillations that occur in prometaphase and/or metaphase cells in other cell types generally are of the entire chromosomes and are irregular, for short periods prior to anaphase (up to 10–25 min), for short distances (at most 1–3 μm), perhaps tapering off as chromosomes approach the metaphase plate, and at moderate speeds of the order of 1–3 μm/min (Skibbens et al. 1993; Ke et al. 2009; Jaqaman et al. 2010), though sometimes oscillations are more regular (Wan et al. 2012; Civelekoglu-Scholey et al. 2013). Speeds and distances can be larger under unusual experimental conditions such as in monopolar spindles (Bajer 1982; Ault et al. 1991; Skibbens et al. 1993; Cassimeris et al. 1994) or in chromosomes with only one kinetochore (e.g. Skibbens et al. 1995; Khodjakov and Rieder 1996; Khodjakov et al. 1997), and can be larger for oscillations of entire nuclei (e.g. Aist and Bayles 1988) or spindle poles (Riche et al. 2013). Compared with this usual description of entire chromosomes that oscillate, oscillations in Mesostoma spermatocytes as described by Fuge (1987, 1989), are regular, cover longer distances, take place throughout most of prometaphase and are best described as kinetochore oscillations since kinetochores on the same chromosome do not always act the same. According to Fuge, each kinetochore oscillates regularly to and from its spindle pole during the maximum observation period (up to an hour) with one kinetochore oscillation cycle to and from the pole taking about 100 s, encompassing distances of 5–7 μm or more with velocities from 8–10 μm/min. The oscillations last throughout prometaphase and there is no real metaphase plate: individual bivalents centre at the metaphase plate only when the two kinetochores move toward their respective poles at the same time. This sometimes happens but often does not, so when one kinetochore moves toward its pole and the other moves away from its pole, the bivalent shifts off the equator. The oscillation distances are so large that the chromosomes often are greatly stretched: during oscillations interkinetochore distances of the bivalents vary from 27 to 40 μm (Fuge 1987, 1989). Fuge (1989) noted that kinetochores sometimes remain more-or-less motionless for varying periods, but that the oscillations generally are regular. Since he did not follow the cells into anaphase, there is no information about when or whether kinetochore oscillations stop prior to anaphase onset. Fuge (1989) also suggested (based on chromosome behaviour in a few cells) that there is some coordination between movements to the same pole: of the kinetochores of the three bivalents in the cell, he suggested, two pairs move in phase and the third moves in opposite phase. In this article we present extensive data on oscillatory movements of kinetochores in Mesostoma spermatocytes, including description of chromosome movements during anaphase.

Materials and methods

Living cell preparations

We reared laboratory stocks of M. ehrenbergii and obtained spermatocytes from Mesostoma testes according to the protocol described by Hoang et al. 2013. Briefly, in our Mesostoma stocks we kept approximately five to 15 animals in plastic jars filled with dechlorinated water and fed them daily with laboratory-reared live brine shrimp (Brine Shrimp Direct, Ogden, Utah, USA). The jars were stored at 25 °C in incubators with a 16 h light:8 h dark cycle. We removed the testes from 3- to 4-week-old animals by inserting through the body wall of the animal a glass needle, pulled from 10 μl micropipettes (Fisher), and sucking up the testes using Tygon tubing (Fisher) attached to the needle. We expelled the testes from the needle onto a glass coverslip, added an equal-sized drop of Mesostoma Ringer’s solution (61 mM NaCl, 2.3 mM KCl, 0.5 mM CaCl2 and 2.3 mM phosphate buffer, pH 6.9) that contained 0.2 mg/mL Fibrinogen (Calbiochem), as previously described (Forer and Pickett-Heaps 2005), spread the fibrogen/testes mixture on the coverslip, and added a drop of thrombin (Sigma) to create a fibrin clot. The coverslip was then inverted onto a perfusion chamber (Forer and Pickett-Heaps 2005) was sealed with wax, and perfused with 2–3 mL of Mesostoma Ringer’s solution.

Data analysis

Phase-contrast images of living cells were obtained using a Nikon Diaphot microscope with ×100 (NA1.3) phase-contrast objective, were recorded on DVDs, then time-lapsed using Virtual Dub (http://www.virtualdub.org), a freeware program, and analysed as previously described (e.g. Sheykhani et al. 2013). Briefly, we used an in-house program (Wong and Forer 2003) in which we marked the position of a fixed point (the spindle pole) and then marked each kinetochore. The measured distances between pole and kinetochores were then imported into a commercially available program (SlideWrite) for analysis.

Statistical analysis

Student's t tests were conducted to determine statistical significance of velocities, amplitudes and periods of oscillations. When the values in question were calculated as percentages (e.g. Fig. 5), the percentages were first converted to proportions, and the proportions were transformed by taking their square root and then the arcsin (sin−1) of the square roots; Student's t tests were performed on the transformed values.

Results

Overview of Mesostoma spermatocytes

Mesostoma spermatocytes have five pairs of chromosomes, three bivalents with bipolar orientation and four unpaired univalents (Husted and Ruebush 1940), as illustrated in Fig. 1. The spermatocytes we study are different from those previously described by Oakley and Fuge in both karyotype and size. Both cells have four univalents, two of each kind, but the cells studied by Oakley and Fuge have two acrocentric bivalents (bivalents with no visible arms) plus one metacentric bivalent, whereas the cells we study have three metacentric bivalents. There is only one chiasma per bivalent so each of the half-bivalents in the cells we studied had a free arm whereas only one bivalent in the cells studied by Oakley and Fuge had visible free arms. Additionally, their cells and chromosomes are considerably larger than ours; their cells have pole-to-pole distances of 40 μm or greater, whereas in the cells we studied pole-to-pole distances were 30 μm or less; in their cells, chromosome lengths were 27–40 μm (Oakley 1985; Fuge 1987) whereas in the cells we studied they were 20–25 μm. These morphological differences may exist because Oakley and Fuge studied cells from Mesostoma populations that were originally derived from animals given to them by Professor Heitkamp in Germany whereas we studied cells from Mesostoma populations that were originally derived from animals from Lake Rondeau in Ontario, Canada, given to us by Dr. Paul Hebert (Hebert and Beaton 1990). The chromosomal differences that exist between the European and North American subspecies of Mesostoma are well known in the literature (Husted et al. 1939; Husted and Ruebush 1940; Hebert and Beaton 1990).
https://static-content.springer.com/image/art%3A10.1007%2Fs00709-013-0532-9/MediaObjects/709_2013_532_Fig1_HTML.gif
Fig 1

Illustrates a fixed and sectioned Mesostoma spermatocyte taken from Husted and Ruebush (1940) that has been modified using arrows to show the three bivalents (open arrow) and four univalents (arrowheads). The arrow labelled K points to the kinetochore of a bivalent and the arrow labelled C points to a chiasma

Kinetochores of the three bivalents oscillate to and from the spindle poles throughout prometaphase (Fuge 1987; Fuge 1989; Fuge 1991). The general impression from viewing time lapsed movies of these oscillations is that there is continuous and rapid movement of all kinetochores toward and away from the spindle poles throughout prometaphase and until the very start of anaphase, which occurs as long as 2 h after the start of prometaphase. The oscillatory movements are so fast that one can easily see them in living cells when viewed on a TV screen. Univalents, on the other hand, do not oscillate. They remain stationary at the spindle poles (Oakley 1985), and only move from one pole to the other, presumably to achieve the proper distribution of two univalents at each pole, one of each kind, before the onset of anaphase (Oakley 1985). Univalent movements between spindle poles can easily be seen in time lapsed movies and are more rapid than the oscillatory movements of the bivalents, as we describe in detail below.

Most Mesostoma spermatocytes have a dumbbell-shaped appearance resulting from a precocious, "pre-anaphase", cleavage furrow that begins ingression once the bivalents have established bipolar orientation; ingression of the furrow then ceases until the onset of anaphase (Forer and Pickett-Heaps 2010).

We now describe these phenomena in more detail, namely kinetochore oscillations, anaphase chromosome movement, bivalent reorientations and univalent movements between spindle poles in Mesostoma spermatocytes.

Detailed descriptions of Mesostoma spermatocytes

Oscillations in Mesostoma spermatocytes generally appear regular for any given kinetochore but sometimes there is no movement of one kinetochore, no movement of two or more kinetochores to the same pole or no movement of partner kinetochores. From time-lapse movies of 152 kinetochores in 40 cells we determined that 42 kinetochores in 28 cells either did not move or “jiggled”, by which we mean that kinetochore movement is less than 1.0 μm to or from a pole and is not regular. In those cells in which kinetochores did not move, most often only one kinetochore did not move (16/28 cells) or two kinetochores associated with the same pole did not move (8/28 cells) (Table 1). Of those 42 kinetochores from 28 cells that did not oscillate at the start of the observation period, only five kinetochores resumed movement. We now give detailed descriptions of those kinetochores that oscillated regularly.
Table 1

Summary of the number of kinetochores that oscillate throughout prometaphase compared to the number of kinetochores that remain stationary and do not move throughout prometaphase

Total number of KTs (n = number of cells)

Number of moving KTs

Number of KTs NOT moving

Number of cells with kinetochores not moving

One KT

Two KTs to the same pole

Partner KTs

KTs to different poles

152 (n = 40)

110

42 (n = 28)

16 (57 %)

8 (29 %)

3 (11 %)

1 (3 %)

Oscillations

Average velocities, amplitudes and periods

Regular kinetochore oscillations in Mesostoma spermatocytes resemble a sawtooth, wave-like pattern when movement is plotted on a distance versus time graph (Fig. 2a). We determined the average velocity for each kinetochore by measuring the slope of five sawtooth waves as the kinetochore moved to the spindle pole and five as they moved away from the spindle pole. From a total of 176 kinetochores (in 88 cells), the average velocity of kinetochore movement to the pole was 6.2 ± 1.8 (SD) μm/min (range 0.9 to 12.5 μm/min) and the average velocity of kinetochore movement away from the pole was 5.2 ± 2.2 μm/min (range 0.9 to 10.8 μm/min). The distribution is shown graphically in Fig. 3a. The average velocity of kinetochore movement to the pole is statistically significantly faster than that of kinetochore movement away from the pole (Table 2).
https://static-content.springer.com/image/art%3A10.1007%2Fs00709-013-0532-9/MediaObjects/709_2013_532_Fig2_HTML.gif
Fig 2

a Distance of the kinetochores of partner half-bivalents from the edge of the cell (top pole) in micrometers versus time in minutes in a Mesostoma spermatoctye. b Higher temporal resolution from the graph of one kinetochore in Figure 2a. The dark black arrows point to the troughs of the sawtooth waves and the open arrows point to the peaks of the sawtooth waves. The distance from the trough to the peak represents the amplitude of the sawtooth wave and the distance between troughs or peaks represents the period of the sawtooth wave. The segment highlighted by the box illustrates that the peaks of the sawtooth waves are rounded and not pointed

https://static-content.springer.com/image/art%3A10.1007%2Fs00709-013-0532-9/MediaObjects/709_2013_532_Fig3_HTML.gif
Fig 3

a Range of average velocities (in μm/min) to the pole and away from the pole. b Range of average amplitudes (in μm). c Range of average periods (in minutes)

Table 2

Summary of the velocity, amplitude, and period of kinetochore movement to the pole and away from the pole in Mesostoma primary spermatocytes and secondary spermatocyte

Kinetochore movement

Number of kinetochore measurements (n = number of cells)

Length of time series analyzed

Range of velocities (μm/min)

Average velocities (μm/min)

Average amplitude (μm)

Average period (s)

Mesostoma primary spermatocytes

Away from polea

880

2,491.5 min (41.5 h)

0.89–10.84

5.23 ± 1.82

4.0 ± 1.38 (1.0–9.0)

92.5 ± 20 (50–180)

To the polea

880

0.89–12.45

6.24 ± 2.23

Combined

1,760 (n = 88)

0.89–12.45

5.08 ± 1.75

Mesostoma secondary spermatocyte

Away from pole

20 (n = 1)

15 min

 

5.3

4.3

83

To the pole

20 (n = 1)

6.5

n number of cells, ± refers to standard deviation

aSignificant difference between to the pole and away from the pole kinetochore velocities, p = 0.01

Using the same 176 kinetochores, we determined the amplitudes of kinetochore movement by measuring the distance from the trough to the crest of each sawtooth wave, and we determined the periods by measuring the time interval between troughs of multiple waves. The average distance the kinetochores moved away from the pole was 4.0 μm (Table 2, Fig. 3b). The average time it took a kinetochore to move away from the pole and then back to the pole was 92.5 s (Table 2, Fig. 3c). At low time-resolution, kinetochores seem to reverse their directions immediately as the tips of the sawtooth waves appear pointed (Fig. 2a), but at a higher time-resolution the tips of the sawtooth waves appear rounded (Fig. 2b). The lag time (i.e. the ‘rounded’ period between the two linear parts of the curve) for kinetochores at the pole was 11.4 ± 11.4 (SD) seconds (n = 495) and for those away from the pole was 10.6 ± 10 s (n = 509). These are not statistically different.

We compared various oscillation parameters for kinetochores in the individual cells (Fig. 4, Table 3). We recorded differences in the parameters in three groupings, 0–10 % differences, 10–20 % differences or differences greater than 20 %; we consider that differences of 0–10 % in any of the categories are not significant. Overall, neither velocities nor amplitudes nor periods in a cell are constant within 10 % (Table 3): differences of 10 % or less occur less than half the time for partner kinetochores (with one exception), for kinetochores moving to the same pole and for kinetochores moving to opposite poles. Even if we extend to 20 % the differences that are not significant, more than 20 % of the kinetochores were different for all parameters (Fig. 4, Table 3). Thus, it appears that the movement parameters of each kinetochore’s oscillation are individual, i.e. not necessarily the same as those of other kinetochores in the same cell.
https://static-content.springer.com/image/art%3A10.1007%2Fs00709-013-0532-9/MediaObjects/709_2013_532_Fig4_HTML.gif
Fig 4

a–d Difference in the a velocity of kinetochore movement away from the pole, b velocity of kinetochore movement to the pole, c amplitude of kinetochore movement and d period, for partner kinetochores, kinetochores to the same pole and kinetochores to different poles

Table 3

Difference in the velocity of kinetochore movement away from the pole, velocity of kinetochore movement to the pole, amplitude of kinetochore movement to and away from the pole and period of kinetochore oscillations for partner kinetochores, for kinetochores moving to the same pole and for non-partner kinetochores moving to opposite poles in the same cell

Type of kinetochore

Total number of cells

 
  

Difference in velocity away from pole

Difference in velocity to the pole

  

0–10 %

10–20 %

Greater than 20 %

0–10 %

10–20 %

Greater than 20 %

Partner KT

36

22 (61 %)

5 (13 %)

9 (25 %)

12 (33 %)

10 (27 %)

14 (36 %)

KTs to the same pole

34

17 (50 %)

8 (24 %)

9 (26 %)

14 (42 %)

9 (27 %)

11 (32 %)

Non-partner KTs moving to opposite poles

32

17 (54 %)

9 (28 %)

6 (19 %)

12 (38 %)

7 (22 %)

13 (41 %)

  

Difference in amplitude

Difference in period

0–10 %

10–20 %

Greater than 20 %

0–10 %

10–20 %

Greater than 20 %

Partner KT

36

14 (39 %)

11 (30.5 %)

11 (30.5 %)

22 (61 %)

6 (17 %)

8 (22 %)

KTs to the same pole

34

12 (35 %)

10 (29 %)

12 (35 %)

14 (41 %)

7 (21 %)

13 (38 %)

Non-partner KTs moving to opposite poles

32

12 (36 %)

7 (22 %)

13 (41 %)

18 (56 %)

7 (22 %)

7 (22 %)

We looked at the same parameters at different times during division, to see if those in early prometaphase were different from hours later, just before anaphase. Because many of our cells were observed for shorter time periods prior to experimentation of one kind or another, we had a smaller sample size for comparisons at different stages. We compared individual kinetochores in three different time intervals for each cell, usually 15 min intervals, though sometimes 20 or 35 min. The time intervals were different in different cells but were consistent within each cell. To determine if amplitude, period and velocity of kinetochore oscillations either increase or decrease as prometaphase progresses, we normalized the data by comparing the second and third time intervals to the first time interval. We compared the first and second time intervals in 20 cells and from those 20 cells we compared the first and third time intervals for nine cells. As seen in Fig. 5, there is a small but gradual, statistically significant (p < 0.01) decrease in the amplitude, period and velocity of kinetochore movement to the pole and away from the pole as prometaphase progresses for each time interval measured, with the largest decrease observed in the amplitude of kinetochore oscillations. The decrease in velocity of kinetochore movement away from the pole, however, was not statistically significant between the second and third time intervals. Each of these parameters seem to be ‘labile’, decreasing from prometaphase toward anaphase, as also seen in prometaphase chromosome behaviours in other cell types (e.g. Dietz 1956).
https://static-content.springer.com/image/art%3A10.1007%2Fs00709-013-0532-9/MediaObjects/709_2013_532_Fig5_HTML.gif
Fig 5

Difference between three time intervals of kinetochore movement throughout prometaphase (in percentage) for amplitude, period, velocity to the pole and velocity away from the pole. The difference between the three time intervals was statistically significant (p < 0.01) for each parameter except the velocity of kinetochore movement away from the pole between the second and third time intervals

Throughout prometaphase, kinetochores oscillate either in-phase or out-of-phase. Following the original definitions by Fuge (1987), we consider partner kinetochores as moving in-phase when both kinetochores move to their respective pole at the same time and away from the pole at the same time. Partner kinetochores are considered as moving out-of-phase when one kinetochore moves to its pole and the other kinetochore moves away from its pole. In the cells we studied, for any individual bivalent the two kinetochores sometimes moved in phase and sometimes moved out of phase. Kinetochores of one bivalent can oscillate in-phase while at the same time kinetochores of a different bivalent in the same cell can oscillate out-of-phase. Bivalents can shift between phases multiple times; they always shifted phase toward the termini of their oscillations, either at the pole or away from the pole. We illustrate on graphs of distance versus time two phase shifts when the kinetochore was moving away from the pole (Fig. 6a) and one when the kinetochore was at the pole (Fig. 6b). In graphs encompassing 41.5 h of kinetochore oscillations in 91 cells, partner kinetochores were in-phase 74 % of the time (515/694 min). There were shifts of phase between partner kinetochores in 45 cells, and multiple phase shifts in 22/45 cells (Table 4). Of the 84 phase shifts observed, most (66/84) took place when the kinetochore was at the pole terminus of the oscillation (Fig. 6b). We do not know why kinetochores oscillate in-phase versus out-of-phase, why bivalent kinetochores oscillate in-phase more often than out-of-phase, if different mechanisms are required for each type of movement, or why kinetochores change phases throughout prometaphase.
https://static-content.springer.com/image/art%3A10.1007%2Fs00709-013-0532-9/MediaObjects/709_2013_532_Fig6_HTML.gif
Fig 6

ab Distance of the kinetochore of a half-bivalent from the edge of the cell (pole) in micrometers versus time in minutes. The segments highlighted by the boxes illustrate shifts in phase a as the kinetochore moved away from the pole and b when the kinetochore was at the pole

Table 4

Summary of the number of phase shifts that occur throughout prometaphase at the pole versus away from the pole and a summary of the length of time single bivalents oscillate in-phase versus out-of-phase

 

Phase shifts

In-phase vs out-of-phase

Length of time series analyzed (min)

When KTs were at the pole (# of phase shifts)

When KTs were away from the pole (# of phase shifts)

Length of time series analyzed (min)

Length of time IN phase (min)

Length of time OUT of phase (min)

Total

2,491.5

66

18

690

508.5

181.5

Percentage (%)

 

79

21

 

73.7

26.3

Mean

 

1 in ∼38 min

1 in ∼138 min

   
Fuge (1987) suggested that tension in the bivalent was a driver of the kinetochore oscillations. When bivalent kinetochores oscillate in-phase both kinetochores move to the pole at the same time and away from the pole at the same time and therefore tension is built up as the chromosome stretches and is dissipated as the chromosome shortens. One might imagine that absence of tension induces force toward the pole and when the tension in the chromosome is maximum the force releases and the tension pulls the two kinetochores toward each other. But when bivalent kinetochores oscillate out-of-phase, one kinetochore moves to the pole while the other kinetochore moves away from the pole and therefore tension is not built up. Rather, the chromosome remains at a more-or-less constant length, as seen in plots of interkinetochore distance versus time (Fig. 7a). Since partner kinetochores move out of phase 26 % of the time (181.5/690 min; Table 4), changing tension in the chromosome would not seem to be the factor causing the oscillations. However, though bivalents maintain constant length when they move out of phase, as seen in boxed region in Fig. 7 that constant length often is not the minimum length of the chromosomes seen on the same graphs, so there still can be some tension across the bivalent that is important for the oscillations to occur.
https://static-content.springer.com/image/art%3A10.1007%2Fs00709-013-0532-9/MediaObjects/709_2013_532_Fig7_HTML.gif
Fig 7

a Interkinetochore distance of partner half-bivalents in μm versus time in minutes. The segment highlighted by the box illustrates out-of-phase kinetochore movement where one half-bivalent kinetochore moves to the pole and its partner kinetochore moves away, maintaining a constant bivalent length. The segments outside of the box illustrate in-phase kinetochore movement where both half-bivalent kinetochores move to the pole and away from the pole at the same time, changing bivalent length. b Distance of the kinetochore of partner half-bivalents illustrated in Fig. 7a from the edge of the cell (pole) in micrometers versus time in minutes. The segment highlighted by the box illustrates the same time series highlighted in (a) to demonstrate that partners continue to oscillate even when they are oscillating out-of-phase and the bivalent length remains constant

Arm movements

When kinetochores oscillate to and away from the pole, the arms of the bivalents usually move with the kinetochores, maintaining a constant angle from the kinetochore as the kinetochore moves, as if the same forces act on both the kinetochore and the arm. In some bivalents, in some cells, however, the tips of the arms remain stationary and thus the arms change angles as the kinetochore oscillates (Fig. 8, Supplementary Movie). In time lapsed sequences of 130 Mesostoma spermatocytes, the arms moved with the kinetochores 87 % (329/377 kinetochores) of the time. In the cells in which arms did not move with the kinetochore (48/377 kinetochores), the tips of the arms remained stationary. In some cells, partner kinetochores had one arm move with the kinetochore and the other arm remain stationary. We do not know why some arms remain stationary and others move with their kinetochore.
https://static-content.springer.com/image/art%3A10.1007%2Fs00709-013-0532-9/MediaObjects/709_2013_532_Fig8_HTML.gif
Fig 8

a–b Montage of phase contrast images of two Mesostoma spermatocytes whose bivalent arms do not move with their associated kinetochore and change angles as their kinetochore moves to the pole and away from the pole. The white arrows point to the bivalent arms that do not move with the kinetochore

Granules in the spindle

Spindles do not contain granules in most cells (Nicklas 1972). In Mesostoma spermatocytes, there are granules along the edges of either side of the cell and also in the centre of the spindle, intermixed with the bivalents. Granules in the spindle move short distances before reversing their direction, moving back and forth, in an undirected fashion, and are not transported to the pole or out of the spindle, as happens in other spindles (Nicklas 1972). ‘Granules’ seen in electron microscopy images of spindles in Mesostoma spermatocytes appear to be mitochondria (Fig. 9).
https://static-content.springer.com/image/art%3A10.1007%2Fs00709-013-0532-9/MediaObjects/709_2013_532_Fig9_HTML.gif
Fig 9

a–d Electron microscopy images at ab low magnification, c medium magnification and d high magnification of a Mesostoma spermatocyte, illustrating granules in the spindle that we identified as mitochondria. The arrowheads point to different mitochondria

Phase dense fibres between kinetochores and poles

We have often seen phase dense “fibres” extending between kinetochores and pole (Fig. 10). They are not always seen, and even when seen they often disappear as they change planes of focus. When seen, they extend between kinetochores and poles, and they seem to elongate as the kinetochores move away from the pole and shorten when the kinetochores move towards the pole. They may represent the chromosomal (kinetochore) spindle fibres.
https://static-content.springer.com/image/art%3A10.1007%2Fs00709-013-0532-9/MediaObjects/709_2013_532_Fig10_HTML.gif
Fig 10

ab Phase contrast images of two Mesostoma spermatocytes with visible phase dense fibres that could represent spindle fibres. The white arrows point to the possible phase dense fibres

Anaphase

Bivalents do not align at a metaphase plate prior to the onset of anaphase. Rather, in the apparent middle of an oscillation the half bivalents separate and move poleward (Fig. 11). We do not know the exact length of time from prometaphase to anaphase, because we have not followed cells from nuclear membrane breakdown, and we generally experiment on cells before anaphase occurs, but several cells have been followed for over one and a half hours of continuous oscillations before they entered anaphase (Forer and Pickett-Heaps 2010). The cells elongate an average of 1.3 μm (range 0 to 3 μm) from the onset of anaphase to the completion of anaphase and once the half-bivalents reach the poles, the cleavage furrow ingresses in the exact position of the arrested precocious cleavage furrow. In the cells in which we did see anaphase, half-bivalents disjoined at the same time and moved an average of 1.9 ± 1.2 (SD) μm (range 1 to 5 μm) toward their respective poles for approximately 165 s with constant velocities to the pole averaging 1.2 ± 0.9 μm/min (n = 37 half-bivalents; Table 5). Anaphase velocities in Mesostoma spermatocytes are within the range of anaphase chromosome velocities in other cell types (Carlson 1977) but are considerably slower than oscillation velocities in the same cells (Table 5).
https://static-content.springer.com/image/art%3A10.1007%2Fs00709-013-0532-9/MediaObjects/709_2013_532_Fig11_HTML.gif
Fig 11

a–b Distance of the kinetochores of partner half-bivalents from the edge of the cell (pole) in micrometers versus time in minutes in a Mesostoma spermatocyte, illustrating anaphase. The thick black line represents that onset of anaphase. a Anaphase occurred after approximately 104 min of prometaphase bivalent oscillations. Bivalents separated into two half-bivalents as the kinetochores moved to the pole. b In this cell both kinetochores moved approximately 2 μm away from the pole before the bivalent disjoined and each kinetochore moved to its respective pole. The black arrows point to segments in the graph where the kinetochores reversed direction when they disjoined

Table 5

Summary of the average velocities of kinetochore movement for bivalent kinetochore oscillations, anaphase chromosome movement, univalent movement and bivalent reorientations in control Mesostoma spermatocytes

Mesostoma spermatocyte control cells

Kinetochore movement

Number of cells

Number of kinetochores analyzed

Range of velocities (μm/min)

Average velocities (μm/min)

Oscillations

88

176*

0.89–12.5

5.1 ± 1.75

Anaphase

16

37

0.1–3.4

1.2 ± 0.9

Univalent

12

22

2.6–21.1

9.4 ± 4.3

Bivalent reorientations

12

16

4.9–14.1

8.2 ± 2.7

± refers to standard deviation

* see Table 2 for details on the number of measurements

We analyzed distance versus time graphs to see if we could better understand when anaphase onset takes place based on possible irregularities in the sawtooth waves prior to anaphase. In the five sawtooth waves immediately preceding the final sawtooth wave prior to onset of anaphase there was no change in the amplitude or period and there was little to no change (less than 15 % difference) in the velocity of kinetochore movement to the pole; however, there was a significant decrease (approximately 30–50 %) in the velocity of kinetochore movement away from the pole, with the most drastic decrease (50 %) seen in the two sawtooth waves immediately preceding anaphase onset. In 16 of the 20 cells we studied, anaphase occurred as half-bivalent kinetochores started to move away from pole: the half-bivalents disjoined and moved towards their respective spindle poles before reaching their furthest away-from-the-pole position (Fig. 11b). In most (10/16) of these cells, bivalents disjoined after half-bivalent kinetochores moved approximately 2 μm away from the pole; bivalents also disjoined after half-bivalent kinetochores moved only 1 μm away from the pole (4/20), less than 1 μm away from the pole (1/20) or greater than 2 μm away from the pole (1/20). In the other four of the 20 cells we studied, anaphase occurred as half-bivalent kinetochores moved to the pole, the bivalents actually disjoining once the kinetochores reached their respective spindle poles (Fig. 11a). This information does not help us predict when anaphase will occur when we are watching a live cell, but it does show that bivalents usually separate and enter into anaphase when half-bivalent kinetochores are moving away from the pole.

Second division

We have observed meiosis II in one spermatocyte that we followed from the completion of meiosis I. The primary spermatoctye we originally followed (Fig. 12a) from prometaphase of meiosis I entered anaphase after approximately 1 h and 20 min. When we returned to the same cell 40 min later, a secondary spermatocyte was present (Fig. 12b) and meiosis II was underway. The secondary spermatocyte then entered into anaphase approximately 15 min later (Fig. 12c). Kinetochore oscillations were similar to those in meiosis I: average velocity of kinetochore movement to the pole was 6.5 μm/min and the average velocity of kinetochore movement away from the pole was 5.3 μm/min, the average distance the kinetochore moved away from the pole was 4.3 μm and the average period was 83 s (Table 2). Though we have data from only one secondary spermatocyte, these oscillation parameters in second division were similar to those in primary spermatocytes, as seen in Table 2.
https://static-content.springer.com/image/art%3A10.1007%2Fs00709-013-0532-9/MediaObjects/709_2013_532_Fig12_HTML.gif
Fig 12

a Phase contrast image of a primary Mesostoma spermatocyte. b Phase contrast image of a secondary spermatocyte that was created following the completion of meiosis I of the primary spermatocyte in (a). c Distance of the kinetochores of partner half-bivalents from the edge of the cell (pole) in micrometers versus time in minutes for the secondary Mesostoma spermatocyte in (b). The vertical black line indicates when the chromosomes disjoined at the start of anaphase

We now describe reorientations of bivalents, and movements of univalents in primary Mesostoma spermatocytes.

Bivalent reorientations

We have observed 49 bivalent reorientations in time lapsed movies of 25 Mesostoma spermatocytes corresponding to 2,492 min (41.5 h) of filming; one example is seen in Supplementary Movie 1. We saw two kinds of bivalent reorientations: (1) bivalents that were mono-oriented when we first starting filming that subsequently became bipolarly oriented; (2) bipolarly oriented bivalents, with normal oscillations, that became mono-oriented after a half-bivalent kinetochore detached and re-attached to the opposite pole (i.e. the same pole as its partner). After the bipolar univalents became monopolar, bipolarity was re-established when one of the two monopolarly oriented kinetochores detached and re-oriented to the opposite pole; the bipolarity was re-established either when the previously detached kinetochore returned to its original pole or when the two kinetochores exchanged places. Reorientations occurred after stable bipolar orientation of the bivalents in question, not just after initial, perhaps incomplete, attachments, because we have filmed up to 75 min of normal oscillation behaviour before some detachments and reorientations.

In 8/49 cells with reorientations, mono-oriented bivalents that were mono-oriented when we started filming became bipolarly oriented during our filming sequence. In seven of the eight cells, no further reorientations took place once the mono-oriented bivalent became bipolarly oriented.

In 41/49 cells with reorientations, a kinetochore of a half-bivalent of a bipolarly oriented bivalent detached from its pole, moved to the opposite pole and became mono-oriented. Most (37/41) of these returned to bipolarity within seconds to 15 min: one of the two kinetochores returned to the original pole. We were able to identify which kinetochore moved back to the original pole of detachment in only 17/41 reorientations. In 13/17 of these the two kinetochores changed places (Fig. 13a), whereas in 4/17 the originally detached kinetochore returned to its original pole (Fig. 13d).
https://static-content.springer.com/image/art%3A10.1007%2Fs00709-013-0532-9/MediaObjects/709_2013_532_Fig13_HTML.gif
Fig 13

ad Distances of the kinetochores of partner half-bivalents from the edge of the cell (pole) in micrometers versus time in minutes illustrating bivalents reorienting throughout prometaphase. a The lower half-bivalent kinetochore (red circles) detaches from lower pole, moves to the upper pole and attaches to it. Its partner half-bivalent kinetochore (blue circles) then detaches from the upper pole and moves to the lower pole where it attaches. b Distance versus time from 15 to 25 min from (a): the solid arrows point to the dampened oscillations of the kinetochore as it moves polewards. c Distance versus time from 7 to 14 min from (a): the dashed box highlights that the upper half-bivalent (blue circles) oscillates with a dampened amplitude as its partner half-bivalent (red circles) moves to the opposite pole with few or no oscillations. d In a different cell the upper half-bivalent kinetochore (plus sign) moves to the lower pole and then reorients and moves back to its original pole. The dashed box highlights that the lower half-bivalent (blue circles) oscillates with a dampened amplitude as its partner half-bivalent (plus sign) detaches and moves to the opposite pole

In 4/41 kinetochore detachments and movement to the opposite pole, the bivalent remained mono-oriented for as long as we followed the cell (up to 30 min), with both kinetochores either oscillating toward the same pole or remaining stationary.

From distance versus time graphs, we determined that some reorienting kinetochores continued to oscillate (albeit with dampened amplitude) as they moved away from the pole, with linear movement between oscillations (e.g. Fig. 13b, d). Other reorienting kinetochores moved to the opposite pole with constant velocity (Fig. 13c, d). When a kinetochore initially detached and moved to the opposite pole, the oscillations of its partner kinetochore had reduced amplitudes and velocities (with unchanged periods), e.g. Fig. 13c, d, and then it stopped oscillating completely after the detached kinetochore reached the pole.

Reorienting bivalent kinetochores moved between spindle poles with velocities that on average were 3 μm/min faster than bivalent kinetochore oscillations, 8.2 ± 2.7 (SD) μm/min (range 4.9 to 14.1 μm/min, n = 16 kinetochores), compared to 5 μm/min (Table 5). In the cells in which the bivalents were monopolarly oriented when we first started filming, when the four monopolar bivalents became bipolarly oriented the kinetochore reorientation velocities were the same as movements of detached bipolarly oriented bivalents, 8.3 ± 2.0 μm/min (range 6.3 to 10.3 μm/min).

Univalents

We analyzed univalent inter-polar movements to determine if these excursions affect bivalent kinetochore oscillations. Univalents usually moved from one pole to the other very rapidly with constant velocity, as seen in Fig. 14a. In some cells however, the univalents paused briefly at the equator, for between10 and 60 s, before continuing to move to the opposite spindle pole. We have seen 56 univalent excursions from pole to pole in approximately 1,200 min of filming 28 cells; there were multiple univalent excursions in 12 of those cells. From plots of 22 univalent excursions, the average velocity of univalent movement was 9.4 ± 4.3 μm/min (range 2.6 to 21 μm/min; Table 5). In cells in which multiple univalent excursions took place, the time between excursions varied from 2.5 to 26 min. In most of the cells we observed, there was only one univalent excursion during prometaphase but there were up to seven univalents excursions in single cells (Fig. 14b). Univalent movements did not affect the oscillation movements of the bivalents in the same cell, except for two excursions in 2/11 cells, one of which is illustrated in Fig. 14c. No other oscillation parameters were affected, so we conclude that univalent segregation does not affect or interfere with bivalent kinetochore oscillations.
https://static-content.springer.com/image/art%3A10.1007%2Fs00709-013-0532-9/MediaObjects/709_2013_532_Fig14_HTML.gif
Fig 14

a Distance of the kinetochores of partner half-bivalents and univalents from the edge of the cell (pole) in micrometers versus time in minutes, illustrating three univalent excursions in a single cell. In the first and second univalent excursions, the univalents move between spindle poles in a step-like fashion, whereas, in the third univalent excursion, the univalent moves to the lower spindle pole in a linear fashion. b Numbers of univalent excursions observed in single Mesostoma spermatocytes. c Distance of the kinetochores of partner half-bivalents and univalents from the edge of the cell (pole) in micrometers versus time in minutes illustrating two univalent excursions that decreased the amplitude of the kinetochore oscillations of the upper half-bivalent (blue circles)

Discussion

We have described cell division in meiosis-I spermatocytes of M. ehrenbergii with the aim of laying the framework for experimental work on this system, with its several unique attributes. We described the regular oscillations of each bivalent that occur throughout prometaphase and until anaphase, and that there is no metaphase configuration recognisable as such; that the oscillations to the pole are faster than those away from the pole; that oscillating kinetochores periodically shift phase; that the free arms of the bivalents usually move with the kinetochores, but sometimes do not; and that while the three bivalents appear to oscillate together, the oscillation parameters of each kinetochore and of the kinetochore in its partner half-bivalent in the same bivalent differ by more than 20 % in 20–40 % of the cells, so that the kinetochore oscillations in a bivalent (and between bivalents) seem independent. Anaphase most often occurs in the middle of an oscillation cycle, and anaphase movements are much slower than oscillation velocities (∼1 μm/min rather than 6 μm/min). We described how bipolar bivalents periodically reorient, most often resulting in the partner kinetochores exchanging poles. We described the pole-to-pole movements of univalent chromosomes, and how these generally do not affect the oscillations of the bivalents.

The oscillations we observed in Mesostoma spermatocytes are similar to those described by Fuge, who was the first to describe and characterise the regular oscillations of the bivalents, despite being limited in his equipment and grant support (Fuge 1987). Our observations agree with his in general that there are continuous oscillations with high velocities, and periodic changes in phase of kinetochore oscillations, but they also differ in some details (Table 6). Some of these differences can be attributed to the different species we studied. We studied a North American species of M. ehrenbergii, whereas Fuge studied a European species of M. ehrenbergii. But we also differ in some interpretations. Fuge (1987) thought that the chromosomes in Mesostoma spermatocytes must be “elastic bodies under tension” as previously reported in grasshopper spermatocytes by Nicklas and Staehly (1967); therefore, as kinetochores move to the pole tension is created and when maximum tension is achieved, the elasticity in the chromosomes matches the poleward forces which results in kinetochore movement away from the pole (Fuge 1987). Based on our data, oscillations cannot be solely due to tension in the bivalent: oscillations continued even when chromosomes oscillated out-of-phase and when interkinetochore distances remained constant (Table 4, Fig. 7). Fuge (1987) also thought that the oscillation parameters (velocity, period and amplitude) were the same for all bivalents and thus that there must be interdependency between kinetochores, especially since he observed a progressive amplification in the amplitude of all three bivalents throughout prometaphase. In our experiments, however, all parameters of the oscillations decreased during prometaphase, and kinetochores seemed to oscillate independently of each other since oscillation parameters differed between kinetochores by more than 20 % (Table 3, Fig. 4). While some of our results and interpretations differ from his, Fuge (1987, 1989, 1991) laid the groundwork in describing Mesostoma spermatocytes, upon which we were able to build.
Table 6

Comparison of Mesostoma spermatocyte control cells from Fuge (1987, 1989) and our data

 

Mesostoma spermatocyte control cells

 

Fuge (1987, 1989) European M. ehrenbergii

Our data North American M ehrenbergii

Average velocity (range including to the pole and away from pole)

8 to 10 μm/min

5 to 6 μm/min

Average amplitude

5 to 7 μm

4 μm

Average period

100 s

92.5 s

Amount of time in-phase: amount of time out-of-phase

1:4

3:1

Anaphase chromosome movements started in the middle of oscillation cycles, most commonly when kinetochores moved away from the pole, interrupting the oscillatory movement away from pole. Anaphase chromosome velocities to the pole are much slower than oscillations, 1 μm/min compared to 6 μm/min; at first glance this might seem puzzling since microtubule depolymerisation acts as the rate-limiting step for movement velocity (Forer et al. 2003, Forer et al. 2008, Pickett-Heaps and Forer 2009) and since both prometaphase and anaphase require depolymerisation of kinetochore microtubules. We think that the explanation for this is something related to kinetochore microtubules changes at or immediately prior to anaphase. It might be, for example, that during prometaphase oscillations kinetochore microtubules depolymerise at the kinetochore and the kinetochores "chew" their way to the pole as described by the PacMan model (Rieder and Salmon 1994) but that during anaphase kinetochore microtubules depolymerise at the pole, as in the classic traction fibre theory or the "flux model" (Cameron et al. 2006). In general, one could explain the slower speeds during anaphase by the microtubule depolymerising enzymes at pole and kinetochore (Sharp and Ross 2012) changing their depolymerisation rates at anaphase; if the spindle matrix (Forer et al. 2008; Pickett-Heaps and Forer 2009; Johansen et al. 2011) propels microtubules and chromosomes poleward, the rate of movement would change depending on the rates of microtubule depolymerisation at the two ends of the microtubules.

Univalents periodically moved between spindle poles, sometimes up to seven times in the same cell. Many cells have a 2:2 distribution of univalents at the poles from early prometaphase, but some have 3:1 or even 4:0 distributions. If pole-to-pole movements of univalents were required only to achieve distance segregation, then the cell would need at most one or two movements to have one X univalent and one Y univalent at each pole. In 6/28 cells, however, there were three or more excursions (Fig. 14). Oakley attributed the more-than-needed number of excursions to the need to not only balance one X univalent and one Y univalent at each pole (distance segregation) but because there is non-random segregation of the univalents, requiring X1Y1 at one pole and X2Y2 at the other. Our data confirm that there are more-than-needed univalent movements in Mesostoma spermatocytes, which is consistent with Oakley’s interpretation that there may be non-random segregation of the univalents. Perhaps relevant to this are the bivalent reorientations that occur in these cells.

Bivalent reorientations occur with reasonable frequency in the cells we studied: 49 reorientation in 2,490 min of filming, or one reorientation every 50 min on average. This seems to us to be quite high, since we know of no reports on any other not-treated cells that describe any reorientation of bipolarly oriented bivalents, though in acentrosomal mouse oocytes there seems to be considerable chromosome instability prior to bipolar orientation of bivalents (Katajima et al. 2011). The high frequency is not because this is correction of faulty initial attachment: we expect the attachments to be stable because they occurred after lengthy periods of normal oscillations and hence seemingly normal attachments, for periods of up to 75 min. That there are so many bivalent reorientations, and that most reorientations (13/17) resulted in the sister kinetochores switching poles, may be related to the speculation by Oakley (1983, 1985) that univalent movements between the poles is required both for distance segregation and to achieve non-random segregation (for example that male-derived and female-derived univalents must be at different poles). It may be that bipolar attachment of bivalents is necessary but not sufficient, that if directed, non-random segregation is required as well, then the reorientation of bivalents may be the mechanism used so that kinetochores of some bivalents can switch the poles to which they are oriented.

Bivalents oscillate regularly and continuously until anaphase in Mesostoma spermatocytes. In other cells described in the literature bivalents oscillate irregularly and only briefly throughout prometaphase before stabilizing at the metaphase plate (Pickett-Heaps et al. 1979; Pickett-Heap and Tippit 1980; Skibbens et al. 1993; Civelekoglu-Scholey et al. 2006; Jaqaman et al. 2010), though in some cells regular oscillations for some chromosomes continue through metaphase (Wan et al. 2012). While kinetochores in Mesostoma spermatocytes oscillate until anaphase, several lines of evidence show that oscillations are not required for anaphase onset. Mesostoma spermatocytes sometimes enter into anaphase when one or more kinetochores are not moving, as described above; they also enter anaphase when kinetochore oscillations are stopped with an optical trap (Ferraro-Gideon et al. 2013). Therefore, the lengthy prometaphase oscillations observed in Mesostoma spermatocytes are not actually required for anaphase onset.

During prometaphase oscillations in Mesostoma spermatocytes, kinetochore movements to the pole are statistically faster than movements away from the pole. This is not necessarily the case in other cells. In more usual cells, the velocity of kinetochore movement to and away from the pole are the same, though different mechanisms have been thought to be required to produce these movements (Ault et al. 1991; Skibbens et al. 1993; Skibbens et al. 1995; Khodjakov and Rieder 1996; Campas and Sens 2006; Ke et al. 2009). In one model, to the pole and away from pole movement are both thought to be generated by tension in the kinetochore (Skibbens et al. 1993; Skibbens et al. 1995); in a different model away from pole movement is thought to be generated by polar ejection forces acting on the chromosome arms (Ault et al. 1991; Khodjakov and Rieder 1996; Liu et al. 2007; Campas and Sens 2006; Ke et al. 2009). Our data indicate that tension alone is not responsible for producing kinetochore oscillations, as discussed above, but a combination of these models may explain why kinetochore movement to the pole is significantly faster than kinetochore movement away from the pole in Mesostoma spermatocytes.

In summary, we have described cell division in meiosis-I spermatocytes of M. ehrenbergii and we have provided detailed description of the regular bivalent oscillations that occur throughout prometaphase; the periodic reorientations of bipolar bivalents and the pole-to-pole movements of univalent chromosomes. We hope our descriptions help lay a foundation for further experimental work on these unique oscillations, on trying to understand the distance segregation of univalents, on the possible non-random segregation of univalents and half-bivalents, on the normally occurring reorientation of bipolarly oriented bivalents and on mechanisms that control their precocious cleavage furrows.

Acknowledgements

This work was supported by grants from the Canadian Natural Sciences and Engineering Council to A.F.

Conflict of interest

None of the authors have any conflict of interest.

Supplementary material

View video
ESM 1

(AVI 19626 kb)

View video
ESM 2

(AVI 35390 kb)

View video
ESM 3

(AVI 4326 kb)

Copyright information

© Springer-Verlag Wien 2013