Why do gametes have haploid numbers

The problem of reduction: How the chromosome distribution is regulated in meiosis

Research Report 2008 - Max Planck Institute for Molecular Cell Biology and Genetics

Zachariae: Control mechanisms of cell division through proteolysis (Dr. Wolfgang Zachariae)
MPI for Molecular Cell Biology and Genetics, Dresden

Meiosis precedes the formation of egg and sperm cells. In this special form of cell division, maternal and paternal chromosomes are separated so that genetically unequal gametes develop. This requires several processes that do not occur in normal cell division, mitosis. Work on yeast cells in Wolfgang Zachariae's group shows that these processes are triggered by a single enzyme, the Dbf4-dependent Cdc7 protein kinase. The inactivation of this kinase transforms the meiosis into a mitosis-like cell division in which two genetically identical gametes are formed.
The research group of Wolfgang Zachariae shows that in yeast, the Dbf4-dependent Cdc7 kinase (DDK) provides a link between premeiotic S phase and the segregation of homologous chromosomes in meiosis I. Independently from its established role in initiating DNA replication, DDK promotes double -strand break formation, the first step of recombination, and the recruitment of the monopolin complex to kinetochores, which is essential for monopolar attachment of sister kinetochores. Thus, activation of DDK both initiates DNA replication and commits meiotic cells to reductional chromosome segregation in meiosis I.

Mitosis and meiosis

The reproduction of almost all animals, fungi and plants is based on two different forms of cell division, mitosis and meiosis. In mitotic cell division, the DNA is first doubled, with two identical sister chromatids being created from each chromosome. These are then distributed evenly to the two daughter cells with the help of the spindle apparatus (Fig. 1). Many such divisions ultimately turn the fertilized egg cell, the zygote, into a person with more than 10 trillion cells. The zygote contains a maternal and a paternal copy of each chromosome, making it diploid. So that the number of chromosomes does not double in each generation, egg and sperm cells (gametes) must emerge from a special cell division that reduces the double set of chromosomes to a single (haploid) set. Meiosis also begins with the replication of the DNA, but this is followed by two nuclear divisions (Fig. 2). In meiosis I (reduction division) homologous chromosomes of maternal and paternal origin are separated and the number of chromosomes is halved. In meiosis II (equatorial division), like in mitosis, sister chromatids are separated so that ultimately four haploid gametes are formed.

From peas and yeast

The special type of chromosome distribution in meiosis is the reason why the transmission of “hereditary factors” (genes) follows certain principles that were discovered more than 140 years ago by Gregor Mendel. From crossings of pea plants, he concluded that “normal” cells contain a maternal and a paternal copy (an allele) of each genetic factor. However, gametes carry either one or the other copy, but never both together. The "mechanics" of chromosome segregation was first described in organisms (e.g. sea urchins and tapeworms) with large germ cells and easily visible chromosomes. However, unicellular yeasts play a decisive role in the elucidation of the molecular mechanisms, since their genome can be manipulated with high accuracy and relatively little effort. For research into meiosis, it is particularly useful that all cells of a yeast culture can be “switched” from mitotic cell division to meiosis. The resulting haploid cells (spores) can also be isolated and multiplied individually so that all products of a meiosis can be analyzed.

The mechanics of meiosis

Correct segregation or splitting of the chromosomes depends on the correct connection with the fibers (microtubules) of the spindle apparatus. Microtubules bind to chromosomes via large protein complexes, the kinetochores, which form on a specific chromosome segment, the centromere. In mitosis (and meiosis II) the kinetochores of the sister chromatids have to bind to microtubules of opposite spindle poles (bipolar attachment). Binding to microtubules of the same pole (monopolar attachment) or attachment of only one kinetochore would lead to errors in the chromosome distribution. But how do mitotic cells recognize correct, bipolar attachment? During DNA replication, the sister chromatids are connected over their entire length by circular protein complexes (cohesin) (Fig. 1, replication). Tensile stress is only created at the kinetochores with bipolar attachment, as the spindle forces now work against this connection between the sister chromatids (Fig. 1, metaphase). Kinetochores without tension generate a signal that blocks the progress of mitosis. Only when all kinetochores are under tension is the cohesin destroyed and the spindle can pull the sister chromatids apart (Fig. 1, anaphase).

In meiosis, both homologous chromosomes and sister chromatids must first be linked and then separated again (Fig. 2). This requires three processes that differ from mitosis: (1) After DNA replication, homologous chromosomes (called homologs for short) are attached to one another and are then linked by recombination. To do this, the DNA strands are cut at several points. A strand break is repaired with the help of a homologous strand, but not with the sister strand. This creates a structure called the chiasma, which holds the two homologues together (Fig. 2, prophase I). A by-product of meiotic recombination is the mixing of maternal and paternal alleles. (2) By linking the homologues, tensile stress can now be generated by pulling the maternal and paternal centromeres apart. To do this, however, the two sister kinetochores of a chromosome must bind to microtubules from the same spindle pole (Fig. 2, metaphase I). Although monopolar attachment is the biochemical basis for homolog separation and thus for Mendel's rules, this process is one of the least understood aspects of meiosis. (3) The separation of the homologues is initiated by the dissolution of the chiasmata. To do this, the cohesin on the arms of the chromosomes is destroyed (Fig. 2, anaphase I). The cohesin around the centromeres is retained, however, as it is required for the bipolar attachment of the sister kinetochores in meiosis II (Fig. 2, metaphase II). The destruction of the centromer cohesin ultimately leads to the separation of sister chromatids and thus to the formation of haploid gametes (Fig. 2, anaphase II).

At the molecular level, we understand mitosis much better than meiosis. In mitotic cell division, it is known at least in principle how the individual processes take place and how they are coordinated with one another. For some processes of meiosis, the important proteins have yet to be discovered and it is unclear how the various processes are arranged in a sequence that creates haploid gametes.

The Dbf4-dependent Cdc7 kinase: more than just a replication enzyme

In order to find new regulators of the reduction division, the group of Wolfgang Zachariae analyzed proteins that bind in meiosis I to a protein kinase known as Polo kinase. Protein kinases are enzymes that transfer phosphate groups to other proteins and thereby regulate their activity. Polo kinase is produced in yeast and multicellular organisms during mitotic and meiotic nuclear division. Surprisingly, the most prominent binding partner of Polo kinase was identified as another conserved protein kinase, the Dbf4-dependent Cdc7 kinase (DDK for short) [1]. This consists of a catalytic (Cdc7) and an activating subunit (Dbf4) and is required for the initiation of DNA replication. The binding of DDK to Polo kinase in meiosis I indicated that this kinase also has other functions. To test this possibility, DDK had to be inactivated without blocking DNA replication and thus the progress of meiosis. This was achieved using two different tricks. On the one hand, the replication enzyme, which is normally switched on by DDK, was activated by a mutation. On the other hand, the DDK activity was reduced to such an extent that it was just sufficient for normal DNA replication. Both manipulations have the same effect on meiosis: No longer four haploid, but two diploid, viable spores are formed! This cell division resembles mitosis, since sister chromatids are separated and genetically identical (clonal) spores are formed [1]. In meiosis, DDK has other functions in addition to replication that are needed for the segregation of homologous chromosomes and thus for halving the number of chromosomes.

A master of reduction

In order to understand the role of DDK in the reduction division, the two meiosis-specific processes - recombination and monopolar attachment to the spindle - were examined in more detail. The recombination is triggered by DNA strand breaks, to which recombination enzymes then attach. However, the DNA from DDK mutants was intact and the recombination enzymes did not form complexes [1]. The lack of recombination normally leads to an extremely faulty distribution of chromosomes in meiosis I and thus to spores that are not viable. In DDK mutants, there must be another defect that prevents the segregation of unlinked chromosomes in meiosis I without, however, preventing the formation of spores. Previous work has shown that bipolar attachment of the sister kinetochores to the spindle blocks any chromosome segregation in meiosis I. In this case the spindle forces work against the cohesin, which holds the sister centromeres together and is only destroyed in meiosis II. Due to the lack of nuclear division in meiosis I, the sister chromatids are then distributed over only two spores (Fig. 3).

To investigate this possibility, cellular structures were marked with fluorescent proteins so that they can be observed in living cells by video microscopy (Fig. 4). In DDK mutants, labeled cohesin disappears as planned from the arms of the chromosomes, but without triggering nuclear division. Labeling of sister centromeres showed that these are under tension in DDK mutants in meiosis I, which is an indication of their bipolar attachment. In yeast cells, monopolar attachment depends on the binding of the protein complex monopolin to the kinetochore in meiosis I. It turned out that the monopoline complex is formed in DDK mutants, but cannot bind to the kinetochore [1]. This defect is probably due to the lack of phosphorylation of a monopoline subunit. Meiosis I without recombination or with bipolar attachment results in non-viable spores. The combination of these two defects in DDK mutants results in two living spores that are genetic clones of the parent cell (Fig. 3).