Tuatara: Volume 13, Issue 1, April 1965
The Cell Nucleus
The Cell Nucleus
In Part I of this article (Tuatara Vol. 12/2) I attempted to review in some detail the structure of the cell nucleus and to briefly indicate the importance of the nucleus to the cell as a working, integrated unit. This second part of the article describes the cell nucleus and the cell as a whole in division. I have tried to picture cell division as a dynamic process with many facets that interact to produce the whole. Thus I have given some detailed attention to a causal analysis of what appear to be the most important phenomena of cell division, rather than confining the article to a redescription of the well known (though often poorly understood) elements of the process. It is unfortunate that nearly all text books treat the subject as if it were completely understood, or emphasize certain hypotheses to the exclusion of others and factual evidence, so as to present the picture that these hypotheses are virtually factual. For this reason I have tried to differentiate clearly between fact and hypothesis; what is understood and what is not; and to create in the reader an attitude of enquiry. I have also tried to indicate the intimate relationship between mitosis, meiosis, fertilization and heredity.
Two distinct phases of cell division are distinguishable, karyokinesis (nuclear division) and cytokinesis (cytoplasmic division), and when cells divide these two processes generally go hand in hand in nearly all tissues, with cytokinesis geared in time and place to the closing phases of karyokinesis. Karyokinesis is referred to as mitosis. The term mitosis comes from the Greek mitos meaning a thread, and -osis indicating a process, and mitosis literally means a process (of division) involving thread like bodies. These thread like bodies we call chromosomes, discrete rod shaped structures that represent the state of the nuclear chromatin during the visible phases of nuclear division. The drama of mitosis does in fact centre round the chromosomes but we shall see that it involves much more than just the chromosomes, or even just the nucleus; the cytoplasm also undergoes important changes during mitosis, changes without which nuclear division could not possibly be achieved. As with all aspects of the workings of cells, mitosis involves the cell as a unit.
Two types of mitosis are associable with cell division, somatic and meiotic. Somatic mitosis (usually referred to simply as mitosis), coupled with cytokinesis, gives rise to new body (soma) page 44 cells; it is a process of duplication/reduction and so it maintains a constant number and type of chromosome in each cell, and endows an identical set of genes to all its products. Meiotic mitosis or meiosis (from the Greek meio, to lessen) is a special type of nuclear division associated with the formation of gametes in male and female sexual reproductive organs; it is, in essence, a reduction division, giving the haploid number and set of chromosomes to each cell produced and so counteracts the doubling of the chromosomes at fertilization. Meiosis also permits the reshuffling of genes between homologous chromosomes and so paves the way for character variation in different individuals of a species.
Mitosis is a continuous dynamic process but for descriptive purposes it is convenient to divide it into five phases; prophase, prometaphase, metaphase, anaphase and telophase. Prophase begins from an interphase-metabolic nucleus that has been undergoing preparations for division; telophase transforms each of the two daughter nuclei produced into a new interphase-metabolic state. Figs. 1, a-l represent these stages of mitosis as seen in the root tip meristem of the onion plant Allium cepa; Figs. 2, a-d are similar stages of the first cleavage mitosis in developing eggs of the horse round worm Ascaris megalocephala.
Prophase (Figs. 1, a-c)
Prophase represents the stage of division in which all the mitotic components are mobilized prior to their organization into a characteristic metaphase configuration. One of the most important of events that takes place during this phase is the transformation of the chromatin ‘network’ of the interphase-metabolic nucleus into morphologically distinct units, the chromosomes, the structure of which was discussed in Part I. This transformation is brought about mainly by the imposition of a series of coils upon the chromosome thread (see Fig. 6, Part I), and as prophase progresses the chromosomes, at first long and thin, become much shorter and thicker as the coils gradually increase in size and compactness and become reduced in number. How this spirilization phenomenon is brought about is not very well understood; it has been suggested as being brought about by changes in a chromosome's DNA molecules. Obviously, however, it is very important to mitosis for it enables the chromosomes to move and separate with little risk of entanglement and subsequent disruption of movement.
As well, two other noticeable changes take place during prophase, the breakdown of the nuclear membrane and the loss of the nucleolus. The nuclear membrane is broken down into small page 45 particles very similar to fragments of endoplasmic reticulum, and it appears that cytoplasmic organelles as mitochondria and lysosomes* play a significant role in the process, perhaps as sources of lytic enzymes. The significance of the breakdown of the nuclear membrane probably lies in its allowance of a more complete interaction between nucleus and cytoplasm, and the incorporation of cytoplasmic components into the nucleus, components that otherwise might be inaccessible.
The reason for the disappearance of the nucleolus is not understood. One concept suggests that an exchange of material occurs between the nucleolus and the chromosomes, or perhaps an exchange involving the spindle apparatus, but no conclusive evidence is available concerning this. It is known for certain, however, that the nucleolus is important to the preparatory events of mitotic prophase, for point irradiation of the nucleolus prior to prophase inhibits mitosis, yet once the nucleus has entered prophase no such inhibition can be brought about.
If present, as in many animal cells, the duplicated centriole, lying just outside the nuclear membrane before prophase, becomes visibly active during this phase, its duplicates separating at the start of prophase and gradually migrating round the nuclear membrane to take up positions opposite each other, so fixing the poles of the cell (Fig. 2a). Though we know little of these movements barring the probability of the activity of phenomena similar to those found at anaphase, and involving spindle elements, one must mention that these centrioles are important guiding elements to which anaphase chromosomes migrate. They appear to have an important bearing on all the organizations that follow prophase. In plant cells and many animal cells where centrioles are not distinguishable the poles, no doubt, are also fixed during prophase.
* See Sampson, Tuatara Vol. 11/3, 1963.
Fig. 1, a-l: Stage of mitosis from sections of a root tip meristem of Allium cepa, stained in haematoxylin. a-c, prophase; d, metaphase; e-j, anaphase; k, telophase; l, interphase-metabolic nuclei. The darkly stained regions in l, a and b are the nucleoli; note their absence in c-j and their early stages of reformation in k. f is a single chromosome at early anaphase from a squash preparation showing the centromeres (appearing as breaks in the chromosome arms) beginning to peel the sister chromatids apart. In k the beginnings of the cell plate formation can be seen.
Prometaphase-metaphase (Figs. 1d and 2a)
After the completion of prophase transformations the chromosomes become arranged along the equator mid-way between the two poles to form the so called ‘metaphase plate’. This orientation phase, involving marked movements of the chromosomes, is referred to as prometaphase and its result as metaphase.
The start of prometaphase coincides with the completion of the breakdown of the nuclear membrane towards the end of prophase. During prometaphase the chromosomes move, the centromeres leading these movements, dragging as it were the chromosome arms behind. At its completion the two chromatid centromeres of each chromosome lie one above and one below the equator, one directed to one pole, the other to the opposite pole; they have become co-oriented.
Not only are the chromosomes oriented during prometaphase; the spindle apparatus is as well. Spindle material, manufactured during preparations for division and often seen in living and fixed cells as a clear zone surrounding the outside of the prophase nuclear membrane, is organized into an array of ‘fibres’ extending from pole to pole and pole to centromeres (See Fig. 12, Part I). Indeed these two discriptively distinct orientations are, as we shall discuss later, intimately associated with each other.
Metaphase is the climax of mitosis for here the chromosomes are poised ready for their actual feat of division. At this stage we can see most of the structures that interact to accomplish this division; the two chromatids lying alongside each other; the two centromeres of each chromosome in a co-oriented position; the poles, marked in many cases by centrioles; the chromosome spindle fibres connecting the centromeres to their appropriate poles, and the continuous fibres from pole to pole. Metaphase passes into anaphase as this machinery is set into activity to bring about an orderly and exact mitosis.
Anaphase (Figs. 1, e-j, and 2, b and c)
Anyone who has followed a living cell through mitosis under the microscope or watched a cine-micrograph of mitosis could not have failed to have been impressed at the unique movements that the chromosomes, literally huge macromolecules, show during anaphase. Even examining a series of static preparations as shown here, one is suitably impressed. Indeed these movements have been the cause of not just a few cytologists pondering in almost complete bewilderment; and today much has still to be learned about anaphase mechanics. This aspect of anaphase is discussed later.
Fig. 2, a-d: Metaphase-telophase stages of the first cleavage mitosis of the horse round worm Ascaris megalocephala from sections stained in haematoxylin. The conspicuous centrioles mark the positions of the poles in each stage. a, metaphase; b-c, anaphase; d, telophase. In d, the cleavage furrow is almost complete.
Telophase (Figs. 1k and 2d)
The events that take place during telophase are essentially the reverse of those that are found in prophase. The nucleolus is gradually reformed through the organization of material at a locus on the nucleolus organizing chromosome; a nuclear membrane is reassembled, here one round each daughter nucleus; and the chromosomes gradually lose their coils and a fine ‘network’ of chromatin is formed. The nuclei are transformed into a typical interphase-metabolic state.
The division of the cytoplasm whereby complete daughter cells are produced is most commonly geared to occur in the late anaphase-telophase of mitosis and between the two separated groups of chromosomes so that each cell produced from a complete cell division is in a uninucleate condition. The process of cytokinesis typically associated with cell division is achieved in one of two rather distinct ways, by cell plate formation or by furrowing, the former characteristic of plant cells, the latter mostly of animal cells. In animal cells the cell membrane, a somewhat elastic structure, becomes pinched inwards all around along the position of the cell equator (Fig. 2d). This furrow so produced gradually extends inwards and eventually cleaves the parent cell into two.
Plant cells are surrounded by a comparatively rigid cell wall and so cytokinesis by furrowing is apparently impossible. Instead, cell wall material is laid down at the centre of the equator, and as cytokinesis proceeds, increasing amounts of such material forms a so called cell plate that grows outwards towards the parent cell walls until it cuts the cytoplasm in two (Fig. 1k). Differentiation along each side of the cell plate forms a new partition between the two daughter cells.
The plane of division as indicated corresponds exactly to the equatorial plane as fixed by the position of the poles in the previous prophase. This relationship becomes clear if at metaphase the poles and spindle apparatus is not centrally placed in the cell.page 51
An unequal placement of the mitotic apparatus takes place under normal conditions during some patterns of tissue development following the movement of the nucleus to one end or side of the cell prior to division, and displacement of the spindle has been induced artificially with a microneedle. As a result, unequal cells are produced.
It might be asked further whether or not the spindle apparatus itself is important in bringing about cytokinesis. Present evidence through experiments on the removal or disruption of the spindle elements at various stages of mitosis suggests that the relationship between spindle and cytokinesis is indirect rather than direct; the fixing of the poles during prophase determines on one hand the spindle and its equator, and on the other hand the plane of cytokinesis. The correspondence between the spindle equator and plane of cytokinesis alone, though, suggests a possible functioning of spindle elements in cytokinesis. Some hypotheses in this respect have been suggested.
Mitosis accomplishes more than simply making two cells out of one. It is a very exact process whereby the daughter cells formed have the same number and complement of chromosomes in their nuclei. By the very early stages of division each chromosome has duplicated itself into two identical halves (chromatids), and prometaphase orientation and anaphase separation ensure that these halves pass to opposite poles. This reproduction/co-orientation/separation occurs in every chromosome of the parent nucleus and so the daughter nuclei produced after division have come to possess identical chromosome complements and, therefore, identical gene complements. They are genetically identical.
The importance of division in the cytoplasm is also more profound than a glance might merely suggest; we understand it as being very important to cell differentiation. Prior to embarking on cell division it is understood at present that a cell may become polarised within its cytoplasm, i.e. localized parts of the cytoplasm may take on different forms (e.g. in structure, enzyme content, chemical composition, metabolic activity). After nuclear division cytokinesis may accentuate this polarity by segregating these differences into different cells, so that, in contrast to their nuclei, daughter cells come to have quite distinct cytoplasms. This is believed to be the basis of how two genetically identical cells can diverge in their paths of differentiation to give rise to distinct structural and functional tissues.
Mitosis is nature's plan for growth and differentiation: it also forms the basis for the accomplishment of the aims of meiosis.
The distinctiveness of the process of meiosis can be appreciated from a study of the products of this division as well as from a study of the cytological features of the process itself. Two principal differences single out meiotic products from those produced through somatic mitosis. Firstly, they possess the haploid rather than the diploid complement of chromosomes; and secondly, depending on a number of factors we will consider below, each nuclear product is unique in its gene make-up. These two differences reflect on a number of very distinct cytological features that can be recognized from a study of the process; and two quite unique phenomena distinguish meiosis from somatic mitosis — synapsis and crossing-over. Meiosis hinges on these two features.
The meiotic products are four in number and meiosis consists of two nuclear divisions (meiosis I and meiosis II) accompanied by only one division of the chromosomes. These two divisions then bring about a reduction in the chromosome number and at the same time, as we shall see, they reshuffle the genes that homologous chromosomes bear and thus stamp individuality on the genetic make-up of their products. Figs. 3, a-r represent successive stages of a meiotic division as found in pollen mother cells of the onion weed Allium triquetrum.
The First Division of Meiosis — Meiosis I
Fig. 3, a-r: Stages of meiosis from squash preparations of pollen mother cells of Allium triquetrum, stained in Feulgens. a, leptotone; b, pachytene; c, diplotene; d, diakinesis of prophase I. Notice the characteristic “beaded” appearance of the synapsed chromosomes in b; the desynapsing chromosomes of c with chiasmata; the 9 compact bivalents in d. In e, metaphase I, the chromatin adjacent to the centromeres (arrows) is markedly drawn out from the chiasmata. f-h, anaphase I. Notice the different forms cf bivalents which can be interpreted according to the number and positions of the chiasmata in each. In g, the chiasmata have been lost (see Fig. 7) except that of the right hand bivalent (upper arrow): also note the widely separated chromatids of each chromosome (lower arrow). i, telephase I. In j, interphase, the new cell wall separating the nuclei is clear; also note the original parent cell wall. k, prophase II; l, metaphase II; m, one diad at metaphase II, the other anaphase II; n-p, anaphase II; q, telophase II with beginnings of cell plate formation; r, tetrad of pollen grains before liberation from the parent cell wall. a-g have been squashed during preparation to show the relevant features more clearly. Nucleoli are not visible as they do not stain with Feulgens.
Prophase I (Figs. 3, a-d, and 5)
The key to meiosis is found in prophase I, indeed, in the two phenomena referred to earlier as synapsis and crossing over. These two events involve prophase I in a complex series of phenomena and the phase is consequently divided into five subphases, leptotene, zygotene, pachytene, diplotene and diakinesis. As prometaphase follows prophase of somatic mitosis so prometaphase I follows diakinesis of prophase I in meiosis.
During prophase I the nuclear chromatin becomes transformed into chromosomes as in somatic mitosis, the nucleolus likewise gradually disappears, and the nuclear membrane breaks down. This transformation, basically one of coiling, is much more marked than that of mitosis (probably due largely to the prolongation of the prophase stages) and the chromosomes at diakinesis are thus much more compact than their corresponding somatic chromosomes (compare Figs. 3d and 4b). Chromosome contraction begins at leptotene, and continues through to diakinesis. Zygotene is the stage of division during which synapsis occurs (Fig. 5). Recall from Part I of this article that a diploid organism carries two homologous sets of chromosomes in its body cells, one derived from its male parent, the other from its female parent; the two chromosomes of each pair are homologous in morphology and basic gene make-up. Synapsis is the pairing of these homologous chromosomes, not at random but arm with arm, centromere with centromere, chromomere with chromomere and probably (though it cannot be certain) gene locus with gene locus. The intimacy of synapsis is clearly demonstrated in the band to band pairing of the salivary gland chromosomes of Drosophila discussed in the first part of this article.
At pachytene synapsis is complete so that each pachytene thread is in fact made up of two chromosomes closely paired throughout their lengths so as to appear as one. A feature of these pachytene chromosomes is their logitudinal ‘beaded’ nature (Fig. 3b) which in some cases (e.g. in maize chromosomes) may show quite distinct patterns and permit identification of particular chromosomes in the complement.
It is not until the following phase, diplotene, when the homologous chromosomes begin to fall apart or desynapse that we can consider further what has taken place prior to or at pachytene. The homologous chromosomes desynapse at diplotene along their lengths except at certain regions termed chiasmata (chiasma, singular). These chiasmata (Fig. 6) are considered to be responsible for maintaining the association of homologous chromosomes until they separate at anaphase I; and it will be seen also that the orientation of the chromosome pairs prior to anaphase is similarly dependent on the chiasmata. Chiasmata govern a regular meiosis. But what are these chiasmata?page 57
Fig. 4, a and b: Feulgen squash preparations of Allium triquetrum chromosomes, a, diploid complement (2n = 18, two of each type of chromosome) from a root tip meristem cell in mitosis. The arrows point to one pair of homologous chromosomes. The constrictions in the chromosomes in median, sub-median, and sub-terminal positions are the centromeres, made conspicuous through the use of colchicine during preparation. b, Haploid complement (n = 9, one of each type of chromosome) from a maturation division of a pollen grain produced from a previous meiosis. The arrow points to the single chromosome, equivalent to the two in a. The chromosomes have been squashed from the pollen cell wall (right).
Diplotene chromosomes take on a characteristic ‘shaggy’ appearance (Figs. 3c and 6) which is not clearly understood, though it is probably related to a particular metabolic state within the diplotene nucleus and is presumably indicative of intense activity between chromosomes and cytoplasm, activity which is almost certainly concerned with past or future events in the division cycle.
The chiasmata hold together pairs of homologous chromosomes, or bivalents as they are usually called. At diakinesis, when the chromosomes reach their maximum contraction, the bivalents can usually be readily counted; their number corresponds to the haploid number of chromosomes of a particular species for bivalents consist of an associated pair of chromosomes. There are nine bivalents in Allium triquetrum (Fig. 3d).
By diakinesis the loss of the nucleolus and breakdown of the nuclear membrane are complete and the bivalents, usually distributed at random, pass into their phase of orientation.
Prometaphase — Metaphase I (Fig. 3e)
As in somatic mitosis the chromosomes, as pairs or bivalents here, move during prometaphase I into co-oriented positions along the equator, midway between the two poles. This orientation, however, does not involve chromatid centromeres as in mitosis. Sister chromatid centromeres in a bivalent act as one and we can say that bivalent orientation involves centromeres of homologous chromosomes. As a result of this, metaphase I takes on a rather distinct appearance when compared with that of mitosis (c.f. Figs. 3e and 1d), though basically they can be considered the same. In mitosis chromatid centromeres are oriented to opposite poles, while the chromatids remain together associated at regions adjacent to the centromeres and along the chromosome arms. In prophase I of meiosis the bivalent centromeres are oriented to opposite poles, and the chromosomes remain associated by their shared chiasmata. The variable position of the chiasma nearest the pair of centromeres in a bivalent determines the distance these centromeres become separated at metaphase I and thus the appearance of the oriented bivalent as a whole (Figs. 3e and f).
At metaphase I the bivalents are poised ready for anaphase separation for at this stage the spindle apparatus has likewise become organized. A careful thought will indicate what, compared with mitosis, will segregate at anaphase (see Fig. 7). This is important.
Anaphase I (Figs. 3 f-h)
As with metaphase I., anaphase I of meiosis takes on distinctive characters, though essentially the process is the same as in mitosis. During anaphase I the co-oriented centromeres of each bivalent move apart to opposite poles, at the same time forcing the chiasmata along towards the ends of the bivalents as cross-over portions of chromatids peels off from their sister portions (Fig. 7). Each centromere carries with it two chromatids and by mid-anaphase the chiasmata have been lost and the chromosomes freed (Fig. 3g). By the end of anaphase I, when movements cease, the homologous chromosomes have become widely separated, and Telophase I phenomena (Fig. 3i), essentially the same as those of telophase in mitosis, then transform the two chromosome groups into two interphase nuclei (Fig. 3j). Note that each transforming group of chromosomes consists, not of a diploid number of single chromosomes (cf. mitosis) but a haploid number of double (chromatid) chromosomes.
Fig. 5: Zygotene in pollen mother cell meiosis of Lilium regale. The arrows point to regions clearly showing synapsis of homologous chromosomes. (Courtesy J. McLeish, John Innes Inst., England.)
The meiotic interphase must be recognized as being quite distinct from that which follows a mitotic division or precedes a meiotic one. At this phase there is no chromosome duplication for the chromatids that are to separate during meiosis II have already been formed. They were formed at the very early phases of meiosis. Here then we see clearly the essence of meiosis as a reduction division — two divisions with only one duplication.
Perhaps, since duplication has already occurred and duplication appears to require the chromatin to be in an interphase state (see later), the prolonged interphase, as found in Allium triquetrum, is a vestige and of no use. However, it must be remembered that in some species additional metabolic activity, dependent also on an interphase state of the chromatin, might have to be undertaken before meiosis II can begin; and these species would then possess a meiotic interphase of some completeness and duration.
Also during telophase I and interphase cytokinesis usually takes place to form two complete daughter cells from the original parental cell. Following this each of these two daughter cells will ‘go it alone’ and pass (though often synchronized) into the second stage of meiosis in order to complete the reduction in chromosome number that was initiated in the first division.
The Second Division of Meiosis — Meiosis II
During meiosis II each sister cell produced from meiosis I divides into two following the separation of the two chromatids of each chromosome into alternate nuclei. The characteristic tetrad of products thus results from a complete meiotic division.
Meiosis II can be divided into prophase, metaphase, anaphase and telophase II and these stages are very similar to those found in somatic mitosis. Some important differences should be noted, however. Firstly, the chromatids of prophase and metaphase II are those already formed by the early stages of prophase I, and secondly, they are widely separated from each other except at their centromere regions. Thirdly, individuals of a pair of chromatids are not genetically identical. This inequality has been brought about through crossing over and will be discussed again later.
Mechanically, meiosis II is essentially the same as somatic mitosis. The chromosomes contract during prophase II (Fig. 3k); they become oriented along the cell equator during prometaphase page 61 II, their chromatid centromeres in a co-oriented position (Fig. 3l and m), and the spindle is organized; the chromatids of each chromosome then pass to opposite poles, headed by the centromeres and mediated through spindle activities during anaphase II (Fig. 3, m-p) and telophase II transforms the groups of chromosomes into interphase-metabolic nuclei (Fig. 3q). Cytokinesis is again geared to telophase and through it the formation of four meiotic products is completed (Fig. 3r).
Meiosis II completes the reduction of the chromosome number that was initiated during meiosis I: without either division reduction is not achieved. The reason why we cannot consider that the chromosome number has been halved at the end of meiosis I is because the chromosomes at this stage are double for, disregarding crossing over, meiosis I has separated the products of synapsis, not the products of duplication. The products of duplication separate at anaphase II. Meiosis I and meiosis II are complementary divisions, and for this reason the latter must never be considered as simply a somatic mitosis following on from a division regarded as the reductional sequence.
The essential characteristics of meiosis will be seen from Fig. 7. both in respect of its reduction from a diploid to haploid complement of chromosomes (cf. Figs 4a and b), and its reshuffling of chromosome segments.
Preparations for Mitosis
One of the most important of the preparations undertaken for mitosis is clearly the duplication of the chromosomes, for without this mitosis could not possibly proceed in any regular manner. As a result of chromosome reproduction the two chromatid units involved in anaphase separation are formed; and each anaphase chromosome, if ultimately passing through a following cell division, will duplicate itself to form another two chromatids for that division.
Chromosome reproduction can be studied with the use of radioactive materials that are incorporated into either the protein or DNA components of chromosome structure; the radioactivity permits their identification, localization, distribution and time of incorporation. Many such studies in recent years have provided very valuable information regarding the duplication phase of mitosis. One of the most interesting of the results has shown that chromosome reproduction as measured by DNA and protein synthesis takes place during the interphase-metabolic state of the nucleus, and the chromatids of mitosis are thus formed well in advance of the initiation of division. Recently it has been suggested that the interphase metabolic nucleus can be divided page 62 into three phases, termed G1, S, and G2, and in root tip meristems of the broad bean Vicia jaba e.g., these three phases are of about equal eight hour periods. S is the phase of chromosome duplication; G1 is a ‘gap’ between the end of a previous cell division and the start of chromosome duplication; and G2 is a second ‘gap’ between the finish of chromosome duplication and the time when the nucleus enters prophase of mitosis. As might be expected there is no uniformity in different species in the time of the beginning or ending of the S period in relation to G1 and G2, but it is clear that the major part of chromosome duplication always takes place during interphase.
Does the same situation hold in meiosis? A few cytological observations suggesting the doubleness of leptotene chromosomes have been followed up in recent years with the finding that DNA synthesis for chromosome duplication does occur largely in the premeiotic interphase nucleus, and chromosomes have already duplicated by the very early stages of prophase I. Crossing over in meiosis involves the chromatids of each chromosome so it is clear that this phenomenon occurs after, or at least at, the time of chromosome duplication. It seems too that synapsis at zygotene occurs when each chromosome consists of two chromatids. These two aspects related to chromosome reproduction will be discussed again later.
What can be said of the G1 and G2 periods in regard to preparations for mitosis? The relative metabolic inertness of the chromosomes during mitosis suggests that during the interphase-metabolic nucleus, parallel paths of metabolism operate to build up all the necessary components involved in mitosis prior to prophase initiation. One such pathway in chromosome reproduction has already been mentioned. Another concerns the spindle apparatus. Experiments indicate clearly that molecules which are built up to form the metaphase spindle are manufactured as precursor molecules during interphase and are held in readiness for their prometaphase orientation as distinct structures. We have indicated this already for spindle material becomes apparent as a clear zone surrounding the prophase nucleus prior to the breakdown of the nuclear membrane.
Though only fragments of information are available, inhibition and other experiments strongly point to the concept that mitosis page 63 proceeds at the expense of energy that has been manufactured prior to mitosis and stored in readiness for future activities. What these pools of energy are one cannot be certain at present, and high energy compounds such as ATP or surf-hydryl compounds that are manufactured and stored specifically for mitosis have not as yet been unequivocably found. But an interesting possibility suggested by Mazia is that the energy requirements might actually be built into the spindle apparatus, whose activity then demands no further external energy sources. Research is on the fringe of understanding these aspects of cell division and they represent big challenges for the future.
A final word on preparation for division should be given concerning the centrioles. When present the centriole of each nucleus duplicates very early in the cycle of cell division. Indeed they begin to duplicate at the closing telophase stages of a mitotic division, and thus represent the earliest cytological indication of a forthcoming mitosis.
All these and undoubtedly other preparations for cell division set the stage for the initiation of mitosis. This initiation is considered in a little detail below.
The Initiation of Mitosis
Fig. 6: Diplotene bivalent from a spermatocyte of the salamander Oedipina poelzi. The cross configuration in the centre is a chiasma, a cross-over between two chromatids, one from each of the two synapsed homologous chromosomes (cf. Fig. 7). The dark portions are the centromeric regions. (Courtesy J. Kezer, Univ. of Oregon.)
How does a nucleus precipitate into mitosis after its preparatory period? And what initiates subsequent periods of preparation? Unfortunately, particularly from the point of view of uncontrolled cell division in malignancy, the answer is not clear. A relationship must clearly exist between growth (cell size) and division; hormones will stimulate many dormant and mature tissues to become meristematic; but it seems, however, that no mitotic ‘trigger’ alone sets cell division into motion, but rather a number of different metabolic pathways lead a cell to mitosis. In other words, successive interactions between the nuclear genes and the cytoplasm create intracellular environments that lead to activities related to cell division. Thus one event leads to another and so on to set the environment that will eventually lead to mitotic activity.
The other problem to mention is the transition from somatic mitosis to meiosis. Here, however, the basic questions remain largely unanswered too. The unique events of meiotic prophase I presumably reflect on characteristic physiological conditions in the cytoplasm that bring about a meiotic rather than a mitotic division, though numerous attempts to induce meiosis or inhibit it have not been helpful in solving the problem.
Prophase I of meiosis is often described as being precocious compared with somatic mitosis in that it is early in its initiation and its spindle formation in relation to its chromosome production. This precocity hypothesis of Darlington's is still much debated among cytologists. It is based on the assumptions that leptotene chromosomes are not duplicated and that homologous chromosomes pair in order to satisfy a ‘need’ for prophase chromosomes to be associated in pairs. In mitosis the association of sister chromatids satisfies this need. In meiosis synapsis brings satisfaction. Present evidence from chromosome duplication does not entirely fit this hypothesis, though in the absence of challenging suggestions and with certain modifications it still holds wide acceptance. A little more will be said concerning this hypothesis under the section on synapsis.
The process of pairing of homologous chromosomes is the most marked cytological phenomenon that distinguishes meiosis from page 65 somatic mitosis, and yet it is probably the one that is most puzzling and least understood. What is the driving force behind synapsis? Why does it take place during meiosis and not mitosis? Before attempting to mention a little of present day indications to the answers of these questions, however, a few basic features of synapsis should be emphasized.
Fig. 7: Diagram to illustrate the essential features of meiosis and show how a reduction in chromosome number and a reshuffling of segments of homologous chromosomes is achieved. Two homologous chromosomes have been shown differentiated (solid and outline) for clarity, a, pachytene, showing the synapsed homologous chromosomes, each of the chromatids: the break in two non-sister chromatids represents the region of one cross-over, b, diplotene showing one chiasma as an exchange between non-sister chromatids. c-e, stages of anaphase I showing progressive terminalization of the chiasma. f, anaphase II, showing one chromosome at each pole: two chromosomes are parental, two are recombinants, the latter having arisen through the exchange of segments at crossing-over.
From the brief consideration above on meiosis in polyploids, it will be apparent that the units of synapsis are not homologous chromosomes, but rather homologous segments of chromosomes. Indeed, observations in organisms with sufficiently large and clear pachytene chromosomes show that chromomere pairs with chromomere very specifically, and in Drosophila salivary gland chromosomes, a specific chromatic band pairs with its homologous band. What the lower size limit is to the unit of synapsis is of course debateable; it is often considered that chromosomes possess a pairing face composed of many molecular sized synaptic units.
Synapsis is initiated some time after the start of meiotic prophase and the chromosome thread is already in a partly coiled condition when it sets in. This fact complicates any model of synapsing chromosomes one tries to imagine and must be considered in any elaborate thoughts on the nature of the forces of synapsis. At initiation the homologous chromosomes are first brought into apposition at one or a few places, and pairing then continues from these ‘contact’ points in a zipper-like fashion. Contact points are usually at regions of the centromere and chromosome ends. In the majority of organisms there appears to be no time limit for zygotene and hence synapsis is complete, though cases of localized unpaired segments are known. As would be expected, in abnormal haploid cells undergoing meiosis unpaired chromosomes regularly appear during the division as only one of a particular type of chromosome is present. The very rare occurrence of interlocking bivalents at diplotene and diakinesis presents somewhat of a problem when analysing the process of synapsis.
The problem of the nature of the forces of synapsis will best be appreciated from a consideration of whether long range forces are operational over considerable distances, or only short range forces after chance association at contact points.page 67
It is difficult to ascertain the spatial distribution of chromosomes preparing to undergo synapsis so that little is known of the distances that might separate members of a pair of homologous chromosomes; regular synapsis in a cell undergoing meiosis immediately following fertilization* suggests that perhaps these distances are of several microns. In diploid organisms it has been suggested that perhaps during premeiotic anaphase the homologous chromosomes are attracted to each other and thus become located close to each other in readiness for synapsis in the ensuing meiosis. And the so called bouquet stage in some animal cells where chromosome ends become oriented to one side of the nucleus prior to zygotene, and the ‘knot’ stage in some plant cells in which the chromosomes become clumped to one side of the nucleus, are thought perhaps to be phenomena associated with and aiding synapsis.
Fig. 8: Illustration of the genetic significance of a crossing over. Two pairs of gene units (arbitrarily lettered) are shown on the homologous chromosomes, one of the chromosomes with the genetic units as T and W, the other with homologous, though allelic, units ° and w. At a, the chromosomes are duplicated along their lengths and, therefore, at their genetic loci, ° and w or T and W. A single cross-over between the two genes is shown. b, anaphase I. c, anaphase II. Each of the four chromosomes at anaphass II has a unique genetic make-up; two are parental (Tw and °w); two are recombinants (Tw and °W).
* Some simple animals and plants have haploid somatic cells, produce gametes by mitosis, and meiosis immediately follows fertilization.
Finally, from a physical point of view there has been some thought on what kind of forces (e.g. Van der Waal's) might be necessary to operate synapsis over long and short distances, but these must be considered largely speculative at present. The difficulty here is that generalized attraction between chromosomes is not sufficient to account for the specificity of synapsis; a great many different forces must be operative, each for and characteristic of one particular pairing locus. Whatever forces or models of synapsis are hypothesized they must as well take into account many other characteristics of the synaptic phenomenon, some of which are mentioned above.
Crossing-over and Chiasma Formation
Crossing-over, both from a cytological and a genetical point of view, holds a unique and most important place in the process of meiosis; through it the two aims of meiosis that have already been stressed are realized. As with the discussion on synapsis above it is advisable here to clearly point out a number of basic features of crossing-over before giving a few indications of the mechanics involved in the process.
From a theoretical point of view alone it will be clear that crossing-over between homologous chromosomes must produce products (chromatids) that are exactly reciprocal to each other; otherwise the integrity and homologous nature of a pair of chromosomes could not possibly be maintained from generation to generation, for just one unequal crossing-over would produce one chromatid duplicated for certain segments while the other would be deficient for these segments. Genetic studies have shown quite definitely that exact reciprocal products are indeed normally formed. This feature of crossing-over suggests a close relationship between it and the high specificity already noted in synapsis; what this relationship is will be considered below.
The second point to consider is that the chromatid, not the chromosome, is the unit of crossing-over (Fig. 6). At any one point along the chromosome pair only two of the chromatids are involved in recombination and each of the four chromatids that ultimately segregate in the two divisions of meiosis conies to have its own unique chromosome make-up (Fig. 7). But although at any one point only two chromatids cross-over, all four may become involved if more than one cross-over is formed. If we label the four chromatids A, A' (sisters) and a, a' (sisters) then four cross-overs might involve chromatids A and a' at one point, A and a at another, then A' and a', and finally A' and a. How many cross-overs are possible and do arise between a chromosome pair, then?page 69
Fig. 9: Illustration of the type of evidence that shows that a chiasma represents an exchange of segments between non-sister chromatids of synapsed homologous chromosomes, a, a homologous pair of somatic chromosomes, one with a terminal deficiency, b, synapsis between such a pair of chromosomes at meiosis showing a cross-over and chiasma (as hypothesized) between the centromeres and abnormality, c, metaphase bivalent with one chiasma. d, anaphase I showing unequal chromatid arms of each chromosomes as a result of the cross-over (see text).
* A cross configuration as in Fig. 6 would result if during diplotene desynapsis pairs of sister chromatids fell apart on one side of the X locus, and pairs of non-sister chromatids on the other side. This of course would not represent a genetic cross-over and would not produce anaphase configurations as in Fig. 9. There are some cytological indications that such or other types of ‘pseudo-chiasmata’ may arise in certain cases.
A priori, the number of chiasmata in a bivalent should be an indication of the number of cross-overs that have arisen. However, in a large number of organisms a gradual reduction in chiasma number occurs between diplotene and metaphase I. This reduction is brought about by a process known as terminalization, caused through the movement of a chiasma to a more distal position, so causing its neighbours to fuse or to be lost. The process of terminalization is essentially the same as that occurring in anaphase I as illustrated in Fig. 7. though in the former ‘repelling’ forces of desynapsing chromosomes are thought to be the main cause, whereas polar centromere movements bring it about at anaphase.
Another feature concerning crossing-over and chiasmata is the phenomenon known as interference — the suppression by one cross-over of the occurrence of others within a short adjacent segment. There is much cytological evidence from chiasma studies and genetical evidence from crossing-over data for the expression of such a phenomenon. It must clearly influence the number of chiasmata that can arise within a bivalent. A peculiar feature is that suppression does not appear to cross the centromere, i.e. a cross-over in one arm appears to have no suppressing effect on another in the other arm.
From the above comments it will be evident that a number of factors affect crossing-over and chiasma frequency. Large chromosomes often have three, four or more chiasmata per bivalent, while other equally large ones may have only one or two. Smaller chromosomes generally have fewer chiasmata. Except in a few very specialized cases, one chiasma must arise in each bivalent, for upon this chiasma hinges the continued association of the two chromosomes and their subsequent orientation and segregation.
Studying bivalents at diplotene (Fig. 6) one gets the strong impression that crossing-over and chiasma formation is brought about by a process of breakage and reunion so that reciprocal portions of sister chromatids become united (Fig. 7). Chromosome breakage and reunion are known to occur spontaneously (and can be induced) outside meiosis, but whether or not such a process occurs in crossing-over and chiasma formation has not been entirely proven. Darlington has suggested that at chromatid formation during pachytene, tensions in the coiled chromosomes cause localized breakages and reunions, but there is little positive evidence for such an interpretation. Another hypothesis suggests that ‘errors’ in chromosome duplication at pachytene lead to crossing-over, i.e. chromatids are formed of material reproduced from parts of both, rather than from one of the two synapsed chromosomes. This hypothesis requires chromosomes duplication to occur at pachytene and thus does not entirely fit available data (see page 61.).page 71
The exact reciprocal nature of crossings-over and the high specificity of synapsis suggests that the former occurs at the completion of the latter, or in other words, crossing-over occurs at pachytene. This is visualized in the two hypotheses mentioned above. But though crossing-over involves the chromatids of a chromosome it may occur during chromosome reproduction prior to pachytene synapsis. and not necessarily after duplication. This clearly presents other problems as some sort of association of homologous chromosomes is a pre-requisite for crossing-over, but recent hypotheses suggest that contact occurs between homologous chromosomes during the interphase of meiosis and at a time when the chromosomes are duplicating. ‘Errors’ in duplication then give rise to crossing-over. These hypotheses fit the available evidence more satisfactorily, but much uncertainty still remains. For a new concept see Moens (1964).
While little is positively known about the mechanics of crossing-over, a great deal has been learned of its effect. Crossing-over undoubtedly results in an exchange of segments of homologous chromosomes as illustrated in Fig. 7 and the result is clear; with only one chiasma in one bivalent the four products of meiosis come to differ in their chromosome make-up, and thus in their genetic make-up. For a fuller appreciation of the reason for this the reader is referred to any textbook on first principles of genetics.* Chromosomes carry many genes, in pairs on homologous chromosomes. One chromosome pair at a particular locus may carry a pair of genes which, though basically the same (e.g. both carry a pair of genes which, though basically the same (e.g. both control height) may differ in their influence (one for tallness, the other dwarfness): and homogues may differ to greater or lesser degrees according to this principle of gene alleles. Crossing-over reshuffles these differences and so produces different products (Fig. 8). These different products of meiosis are manifest in a diploid organism after fertilization. Consider the 23 pairs of chromosomes in Homo sapiens, each with many genes, each with a minimum of one chiasma at meiosis; then witness the variation between individuals of the progeny of such a species!
Mechanics of Prometaphase
* e.g. GENETICS, by R. P. Levine Modern Biology Series, 1962.
The very important studies in recent years of mitosis and meiosis in living plant and animal cells have provided a great deal of information regarding prometaphase movements. They show conclusively that the centromere is the organ of movement of the chromosome. And it is evident from irradiation and chemical inhibition studies that these movements are basically the same as those found at anaphase, being brought about through interactions between centromere and spindle. A very clear illustration of this comes from the use of drugs (e.g. colchicine) that inhibit spindle activities. They inhibit anaphase separation; they also inhibit metaphase orientation, and the cells are blocked at the end of prophase. The nature of the interactions between centromeres and spindle that are necessary for orientation will be considered in more detail in the next section. In essence they bring about an orientation of spindle substances from a diffuse state round the prophase nucleus to a highly organized series of ‘fibres’ extending from centromeres to pole; and they bring the chromosomes into a position to form a regular metaphase plate. The interactions appear to start with a contraction stage where the scattered chromosomes suddenly form a tighter mass before the more specific movements begin. This start seems to coincide with the loss of the nuclear membrane, a probable indication of the initial incorporation of spindle substances into the nucleus.
What is most interesting concerning the mechanics of prometaphase is that a chromosome or bivalent may not move directly onto the equator in a co-oriented position. It may do so, but it may first move to one or the other pole, then perhaps to the other pole before becoming stabilized along the equator.
These movements suggest that each centromere (of bivalent or chromosome) is capable at one time of an orientation and movement to one pole, and that from this basic property the two types of movement of an associated pair of centromeres indicated above page 73 are bought about. If the two associated centromeres orient to opposite poles, the chromosome pair will move directly onto the equator, its centromeres in a co-oriented position. If, on the other hand, the two centromeres initially orient to the same pole, the chromosome pair will move to that pole. Hence the two types of movement that appear to characterize prometaphase.
With a chromosome or bivalent oriented and at one pole, a phenomenon of re-orientation appears to operate to achieve its subsequent co-orientation. If one of the two centromeres re-orientates to the opposite pole the bivalent or chromosome will move to the equator in a co-oriented position. Re-orientation of both centromeres would cause the bivalent or chromosome to move to the opposite pole and further re-orientation would have to be undertaken before achieving co-orientation.
Whether or not these ideas are in fact the basis of prometaphase mechanics has yet to be ascertained, and a lot of aspects remain to be understood.
One aspect that should be apparent is that for co-oriented centromeres to remain in the equator and for polar chromosomes or bivalents to be brought into the equator following re-orientation, two features are necessary. Firstly, the chromosomes of a bivalent, and chromatids of a mitotic chromosome must be linked together. This linkage is provided by the chiasmata in the former, and by some little understood substance or structure in the latter; co-orientation depends on pairs of chromosomes or chromatids so linked. Secondly, the magnitude of the ‘force’ acting between a centromere and pole must very directly with the distance separating the two; the greater the distance, the greater the force. Metaphase centromeres must be in an equilibrial position. There is a lot of evidence available for this assumption.
There are other aspects of prometaphase mechanics that are not well understood. Why, for instance, do chromosomes not clump into the centre of the cell at metaphase but remain regularly spaced out? Repulsion forces are perhaps important. And why do some chromosomes show preferences for particular positions along the equator? These and other questions remain largely unanswered, and the reader is referred to Schrader's book ‘Mitosis’ for a comprehensive analysis of these and related problems.
Mechanics of Anaphase
The movement of the chromosomes from the equator to the poles during anaphase is probably one of the most intriguing yet most puzzling of biological movements. Since mitosis was first described its theme has been constantly discussed and many theories have been put forward in explanation. Space, however, will not permit mention of most of these and it must be admitted that this important phenomenon still eludes complete understanding. page 74 In this discussion a few important facts can be made clear before attempting some understanding of the problem involved.
Firstly, the importance to chromosome movements of the centromere and its associated spindle fibres is today clearly recognized. A chromosome without a centromere or whose centromere has been deactivated by point irradition fails to move, and similar maltreatments of the spindle by X-rays or poisons likewise inhibit anaphase. Secondly, the chromosome spindle fibres shorten as anaphase progresses and it is clear that centromeres and spindle interact to accomplish chromosome movements; and since each chromosome moves independently of the others, it is individual centromeres and their associated spindle elements that interact for this purpose.
The concept that a chromosome is passively dragged in movement by the shortening of fibres attached to its centromere and based at a pole can no longer be considered as a possible mechanism of chromosome movement in the light of present day evidence. Birefringence studies have indicated that a wave of activity precedes the centromere as it moves to the pole, and other evidence also points to the conclusion that the centromere is a very active organ in bringing about chromosome movements.
What can be said of the chromosome apart from its centromere?
Studies in living material have shown that in mitosis the whole chromosome is active during the initiation of anaphase. The two sister chromatids, that up to anaphase are closely associated along their lengths, appear to relax their attraction for each other before the centromeres actually begin to peel the chromosome arms away from each other to opposite poles. Since all the chromosomes in a cell are simultaneously affected in this way it seems very likely that this is an indication of a change in the chromosome environment that triggers anaphase and permits further centromere/spindle activity for polar movements. Similar activities can be thought of as being present at the initiation of anaphase in meiosis as well, though here not affecting the chromosome regions adjacent to the centromeres until anaphase II. However, it should be remembered that the fact that the chromosome segments between centromere and chiasmata are often stretched during anaphase I (Fig. 3f) indicates that the chiasmata afford considerable resistance to chromosome separation, a resistance that would not be expected if all attraction between chromatids were lost at initiation. Chromosomes in anaphase of mitosis show no such stretching. Yet by mid-anaphase I of meiosis the chromatids of each chromosome are usually widely separated from each other except at their centromeres (Fig. 3g).
In meiosis II, anaphase initiation is centred round regions of the centromeres where the chromatids previously maintained association.page 75
Two systems of movement are at present considered to operate after the initiation of anaphase. The first has already been indicated; that observable cytologically as a shortening of the spindle fibres and a movement of the chromosomes towards the poles. The second is an increase in the distance between the two poles which, while not moving the chromosomes nearer to the poles, increases the distance between the two separating groups of chromosomes and thus contributes significantly to anaphase: this phenomenon is readily observable cytologically if centrioles are present, but otherwise as an increase in the lengths of the pole to pole fibres of the spindle. The relative importance of these two systems varies with the organism; the latter appears characteristic more of animal rather than plant cells, though in both the former is considered to be the most important.
During anaphase a physical and/or chemical change appears cytologically to take place in the interzonal region between the two separating chromosome groups. Whether this is a third phenomenon associated with anaphase movements is not clear. It may be related to a ‘pushing’ action acting between the chromosome groups; or it may be more related to cytokinesis rather than to anaphase.
To close this section a few comments should be made on the nature of the important forces of anaphase; those associated with centromere/spindle activities. It is no longer acceptable today to consider the spindle fibres as series of protein threads that become folded during anaphase and thus draw the chromosomes nearer the poles. Rather, the spindle fibres are considered at present to represent regions of general spindle molecules that are in a highly oriented state (thus appearing as fibres); and that perhaps a change in this oriented state, brought about by the centromere, results in chromosome movements. Before more elaborate explanations can be given, however, a great deal must yet be learned of the physical and chemical changes that precede and accompany anaphase, a field that is only just being pioneered. Elusive though these explanations are, there is no call for dispair for little more is known of the perhaps equally or more interesting action of muscle tissue or other forms of biological movements. Some further indication of the problems concerned and the type of research they stimulate will be had from reference to Schrader's book on ‘Mitosis’ and the more recent articles by Mazia in ‘The Cell’ and those from the Second Conference on Cell Division.
Duration of Mitosis
Considerable data have become available in recent years on the duration of mitosis, of its various phases, and of the associated intermitotic time, and while generalizations tend to be misleading they give some indication of upper and lower limits. Rarely does page 76 mitosis extend beyond a duration of three hours, while in only a relatively few known cases is it completed within 30 minutes under optimal temperature conditions, though a mitotic time of 5 minutes has been recorded. As far as the various phases of mitosis are concerned it can be stated that those involving chromosome movements (prometaphase and anaphase) are quite dramatic, rarely exceeding 10 minutes, while prophase and telephase are usually of considerably greater duration. Metaphase is often of some duration also, giving the impression that the cell is waiting for something to happen: but though nothing can be seen to occur at metaphase this is not an indication of inactivity.
The prophase of meiosis is of much greater duration than the prophase of a corresponding mitosis. This fact may be related to the suggested precocious nature of prophase I already discussed.
As far as division in its completeness is concerned, a cell spends a large proportion of its time in preparations between successive mitoses. This is particularly so in meristem tissues where the cells maintain a near constant volume through successive division cycles (Fig. 1l). The length of the intermitotic time is clearly of importance to cell growth: in the first number of cleavages of a developing egg cell, mitosis follows mitosis very rapidly, intermitotic times are short, and successive daughter cells become increasingly smaller in size.
The time course of the various stages of mitosis and meiosis can be well appreciated from the numerous movie films on cell divisions, films that also vividly dramatize the swift and decisive movements of the chromosomes.
It has been impossible in this article to cover all aspects and problems associated with cell division and to mention all evidence, for and against, concerning those features I have discussed. But it is hoped that the article has indicated something of the level of present-day understanding of this important biological process, and will stimulate thought for further enquiry. For this latter purpose a brief list of appropriate books is given below.
Inoué, S., and Bajer, A. (1961). Ciné-micrographs of endosperm mitosis. Chromosoma (Berl.) Vol. 12, and subsequent papers in Chromosoma and Experimental Cell Research.
Mazia, D. (1961). Mitosis and the physiology of Cell Division. The Cell Vol. III. Brachet & Mirsky (Editors), Academic Press, N.Y.
Rhoades, M. M. (1961). Meiosis. The Cell Vol. III. (as above).
Schrader, F. (1953). Mitosis. The movement of chromosomes in cell division. 2nd edition. Columbia Univ. Press, N.Y.
Swanson, C. P. (1960). Cytology and Cytogenetics. 2nd Edition. Macmillan & Co. Ltd., London.
Second Conference on the mechanism of cell division. N.Y. Acad. of Sciences. Vol. 90(2), 1960. p. 345-613.
Moens, P. B. (1964). A new interpretation of meiotic prophase in Lycopersicon esculentum (tomato). Chromosoma 12: 48-63.