pubmed.ncbi.nlm.nih.gov

Homologous chromosome pairing in Drosophila melanogaster proceeds through multiple independent initiations - PubMed

  • ️Thu Jan 01 1998

Homologous chromosome pairing in Drosophila melanogaster proceeds through multiple independent initiations

J C Fung et al. J Cell Biol. 1998.

Abstract

The dynamics by which homologous chromosomes pair is currently unknown. Here, we use fluorescence in situ hybridization in combination with three-dimensional optical microscopy to show that homologous pairing of the somatic chromosome arm 2L in Drosophila occurs by independent initiation of pairing at discrete loci rather than by a processive zippering of sites along the length of chromosome. By evaluating the pairing frequencies of 11 loci on chromosome arm 2L over several timepoints during Drosophila embryonic development, we show that all 11 loci are paired very early in Drosophila development, within 13 h after egg deposition. To elucidate whether such pairing occurs by directed or undirected motion, we analyzed the pairing kinetics of histone loci during nuclear cycle 14. By measuring changes of nuclear length and correlating these changes with progression of time during cycle 14, we were able to express the pairing frequency and distance between homologous loci as a function of time. Comparing the experimentally determined dynamics of pairing to simulations based on previously proposed models of pairing motion, we show that the observed pairing kinetics are most consistent with a constrained random walk model and not consistent with a directed motion model. Thus, we conclude that simple random contacts through diffusion could suffice to allow pairing of homologous sites.

PubMed Disclaimer

Figures

Figure 1
Figure 1

Chromosome orientation and coordinate system in cycle 13 and cycle 14 embryos. Relative orientation of the nuclei to the surface of the embryo is shown for both cycle 13 and cycle 14 embryos. Direction of Rabl orientation (centromere to telomere polarity) is also diagrammed. The bottom schematic shows the coordinate system used throughout our analysis.

Figure 2
Figure 2

Probe positions and a typical 3-D optical data set of nuclei from a cycle 14 embryo. (A) Names and positions of probes on chromosome 2. Locations refer to cytological map positions determined by hybridization to polytene or metaphase spread chromosomes. P1 indicates that probes were derived from P1 clones. Rsp refers to the Responder locus. Probes were either labeled directly with rhodamine-4-dUTP or indirectly with digoxigenin-dUTP and detected using rhodamine, fluorescein, or Cy 5-labeled anti-digoxigenin antibodies. (B) Representative multi-wavelength 3-D data stack from a cycle 14 embryo. Each wavelength shows a subset of nine sections from a 3-D data stack of nuclei. Each section is separated by 0.5-μm focal steps starting from the top left corner and ending at the bottom right corner. FISH data from the P1 probe 09-93 (far left), and from the histone probe (middle left) together with lamin immunofluorescence (middle right) and DAPI chromatin staining (far right) were simultaneously visualized for the same set of nuclei. For most nuclei, one or two FISH signals for each probe were identified representing, respectively, the paired or unpaired states of the locus.

Figure 3
Figure 3

Site-specific pattern of homologous pairing for chromosome arm 2L. A surface mesh plot was fit to the measured pairing frequencies (yellow O) to better depict general trends in pairing. The height and color of the surface serve to indicate the level of pairing. High pairing frequencies are coded by increasingly darker shades of pink whereas low frequencies are represented by deepening colors of blue. Embryonic age and probe positions were spaced at equal intervals along their respective axes. Below the position axis, a probe's relative location is indicated by dashed lines to a representation of chromosome 2. Similarly, the relation between different embryonic ages is mapped to a time line drawn below the age axis.

Figure 4
Figure 4

Nuclear elongation as a gauge for elapsed time in cycle 14 interphase. (A) Procedure for live analysis of nuclear volume using fluorescent-labeled dextrans. Fluorescent-labeled dextran (40,000 mol wt) is injected into the lumen of the embryo. During mitosis, dextran is able to enter into the nuclear region due to NE breakdown but at interphase, dextran is excluded from the nucleus. (B) Time series recording morphological changes in the embryonic nucleus. Single optical sections from 3-D data sets of nuclei as imaged by exclusion of the FITC-labeled dextran. Both the cell cycle stage and elapsed time in each stage is marked. Entry into mitosis is observed by the appearance of an even background of fluorescence (see cycle 13, t = 17.5 min) brought upon by the NE breakdown. The start of cycle 14 (cycle 14, t = 0.0) is set at the first reappearance of excluded nuclear volumes. (C) Representative 3-D data set of nuclei at cycle 14, t = 5.5 min showing that nuclear length can be easily determined from the excluded volume images. (D) Plot of nuclear length against elapsed time in cycle 14 interphase showing a monotonic increase in average nuclear length with increasing time. The quadratic equation was determined by a least squares fit to the plotted data where t = time and ht = nuclear ht. Such an equation allows the direct conversion from measurements of nuclear height from FISH prepared embryos into elapsed time.

Figure 5
Figure 5

Time profiles of histone pairing. (Top) Pairing frequency was measured from FISH signals to the histone locus in cycle 14 embryos and plotted as a function of elapsed time in interphase as measured by length of the nucleus. A substantial amount of pairing, starting from ∼20% and rising to ∼80%, is completed within the first 20 min of cycle 14 interphase. (Bottom) Average inter-homologue distance between unpaired histone loci was plotted as a function of elapsed time in interphase. The average distance between unpaired histone loci at the beginning of interphase is ∼1.2 μm apart and gradually increases to >2 μm as interphase progresses.

Figure 6
Figure 6

Evaluation of two potential models for homologue pairing. (A) In the random walk model, the movement of a locus can be separated into unit steps that are independent from the previous step, random in direction, and unrelated to the movements of the homologous locus. The diffusion coefficient, D, reflects how fast the random walk is occurring. In the nucleus, chromatin domains and the nuclear boundary would restrict movement from being completely random, so a constrained rather than pure random walk would be expected. In contrast, in the directed motion model, each pair of homologous loci take a unit step in a direction pointed towards the other locus. Here, a velocity constant is used to present how fast motion is occurring. (B) Random walk motion: plots of the pairing frequency as a function of elapsed time in cycle 14 were generated from simulations based on the random walk model of pairing for several values of the diffusion constant and are represented by different line patterns as follows: D = 5 × 10−11 cm2/s (top-most line), D = 2 × 10−11 cm2/s (second line from top), D = 1 × 10−11 cm2/s (third line from top), D = 0.5 × 10−11 cm2/s (fourth line from top), D = 0.1 × 10−11 cm2/s (bottom line). Data marked by circles are the actual experimental values for the pairing frequency of the histone locus replotted from Fig. 7 (top). Plots of the inter-homologue distances are generated from the simulation using the same D values as above. Circles represent the experimentally measured distances. (C) Directed motion: same types of plots as in A except this time they were generated using the directed motion model for several values of the velocity as follows: v = 10 × 10−7 μm/s (top-most line), v = 5 × 10−7 μm/s (second line from top), v = 1 × 10−7 μm/s (third line from top), v = 0.5 × 10−7 μm/s (fourth line from top), v = 0.1 × 10−7 μm/s (bottom line). Again, the appropriate experimental data was plotted (O) for comparison.

Figure 7
Figure 7

Characterization of the temporal changes in the distribution of histone loci in cycle 14 interphase nuclei for use in obtaining simulation boundaries. The top two panels report the average radial position (r) and its variance (σr) for the histone locus as a means to characterize histone's radial distribution in cycle 14 interphase. Note that the average radial position and the variance remain relatively constant throughout the observed portion of cycle 14 interphase. The bottom two panels report the average vertical position (z) and its variance (σz) for the histone locus. Here, the average vertical position increases but the variance remains relatively constant. The increasing average vertical position reflects the fact that the nuclei are lengthening. However, since the variance remains relatively constant, this indicates that volume over which the loci occupy remain relatively the same and that no significant spreading effect is caused by the upward movement of the locus. In the simulation, the normal distributions are generated based on variances measured at the earliest timepoints. Volume for the simulations is calculated from these variances as described in Materials and Methods.

Figure 8
Figure 8

Effects of chromosome rearrangement on the establishment of pairing. (A) Diagram illustrating chromosome arm 2L in a homozygous wild-type strain and in a homozygous lt×13 strain where 2L is translocated to the end of 3R. The dark filled areas indicate heterochromatic regions. The cross-hatched areas indicate approximate location of the histone locus. (B) Plot of the simulated pairing frequency for the histone locus in a lt×13 strain as a function of elapsed time in cycle 14. The experimentally determined pairing frequency for the histone locus in wild-type strain from Fig. 5 is plotted as empty circles. Plots were generated using actual experimental data for the initial distribution of the histone locus. D = 1.0 × 10−11 cm2/s was used for the diffusion coefficient. Note that the pairing frequency is predicted to be much lower in the lt×13 strain than in the wild type (compare with Fig. 6 B, D = 1.0 × 10−11 cm2/s).

Figure 9
Figure 9

Perturbation of pairing occurs at anaphase. (A) A volume-rendered image of metaphase chromosomes in several nuclei from a cycle 13 embryo overlaid with FISH signals from the histone probe. A 3-D data stack was projected down the optical (z) axis to create the volume rendered image. Over half the nuclei show a single FISH signal indicating that pairing of the histone locus is maintained. (B) A volume-rendered image of anaphase figures in several nuclei from a cycle 13 embryo overlaid with FISH signals from the histone probe. Here most anaphase figures display four FISH signals showing that pairing has been disrupted during this stage.

Similar articles

Cited by

References

    1. Agard DA, Hiraoka Y, Shaw P, Sedat JW. Fluorescence microscopy in three dimensions. Methods Cell Biol. 1989;30:353–377. - PubMed
    1. Aramayo R, Metzenberg RL. Meiotic transvection in fungi. Cell. 1996;86:103–113. - PubMed
    1. Arnoldus EP, Peters AC, Bots GT, Raap AK, van der Ploeg M. Somatic pairing of chromosome 1 centromeres in interphase nuclei of human cerebellum. Hum Genet. 1989;83:231–234. - PubMed
    1. Ashburner, M. 1989. Drosophila. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. 380 pp.
    1. Bennett MD. The timing and duration of meiosis. Phil Trans R Soc Lond Ser B. 1977;277:201–226. - PubMed

MeSH terms

Substances