D2.1 Cell and nuclear division

All new cells come from existing cells. In cell division a parent cell (often called the mother cell) divides to form two daughter cells. In eukaryotes this happens in two stages: first the nucleus divides (by mitosis or meiosis), then the cytoplasm divides. Division of the cytoplasm is cytokinesis.

Animal and plant cells divide differently

• Animal cells: a ring of actin filaments and myosin forms just inside the plasma membrane at the equator. Myosin pulls the actin filaments past each other, so the ring contracts like a drawstring. The membrane is pulled inwards as a cleavage furrow, which deepens until it meets in the middle and the cell is pinched in two.
• Plant cells: a rigid cell wall means the cell cannot be pinched. Instead, vesicles from the Golgi apparatus, carrying cell-wall materials, are guided by microtubules to the equator. They fuse to form a cell plate, which grows outwards until it joins the existing wall. The vesicle membranes become the plasma membranes of the two daughter cells, and their contents become the new cell wall between them.

Equal and unequal cytokinesis

Cytokinesis usually divides the cytoplasm equally, but not always. However unevenly the cytoplasm is shared, neither daughter cell can be left without mitochondria (nor, in plants, without plastids such as chloroplasts). A cell cannot build these organelles from nothing; new ones form only when existing ones grow and divide.

• Oogenesis in humans: both divisions of meiosis are unequal. Nearly all the cytoplasm stays in one large cell, which becomes the egg, and the other nucleus is pushed out with very little cytoplasm in a tiny polar body that degenerates. The egg keeps the nutrients and organelles that the early embryo will need.
• Budding in yeast: a yeast cell forms a small outgrowth (a bud). After mitosis one nucleus moves into the bud, which also receives mitochondria, and the bud pinches off while still much smaller than the parent cell.

Why the nucleus must divide first

A eukaryotic cell divides its nucleus before its cytoplasm, so that each daughter cell gets a nucleus. A cell without a nucleus (an anucleate cell, such as a mature mammalian red blood cell) cannot make new proteins and cannot divide. There are two kinds of nuclear division:

DNA replication comes first

Before either kind of division, the DNA of every chromosome is replicated (in the S phase of interphase). Afterwards each chromosome consists of two chromatids: two long, identical DNA molecules called sister chromatids, held together along their length and especially at the centromere. They stay together until anaphase. When they separate, each chromatid becomes a chromosome in its own right.

Condensation and movement are shared features

During interphase the chromosomes are long, thin threads that cannot be seen with a light microscope. Long molecules like these would tangle and break if moved, so in both mitosis and meiosis the chromosomes condense. DNA is wound around groups of histone proteins to form nucleosomes, and this string of nucleosomes is then coiled again and again (supercoiling), until each chromosome is thousands of times shorter and much thicker.

Condensed chromosomes are then moved by microtubules, the protein fibres of the spindle. Microtubules attach to the centromeres at kinetochores. Microtubule motor proteins (such as dynein and kinesin) use ATP to “walk” along microtubules, and the microtubules themselves shorten by losing tubulin subunits, pulling chromosomes or chromatids to the poles. See D1.1 DNA replication for DNA replication.

Mitosis is one continuous process, divided for convenience into four phases. Interphase comes before it, and cytokinesis usually follows.

1. Prophase: chromosomes condense by supercoiling and become visible, each as two sister chromatids joined at a centromere. The nucleolus disappears. In animal cells the two centrosomes move to opposite poles and the spindle of microtubules forms. At the end of prophase the nuclear membrane breaks down.
2. Metaphase: microtubules from opposite poles attach to the centromere of each chromosome, one to each sister chromatid. The chromosomes are moved until they are lined up individually on the equator. A checkpoint ensures that every chromosome is attached before the cell continues.
3. Anaphase: the centromeres divide, so the sister chromatids separate. Each is now a chromosome. Spindle microtubules pull them, centromere first, to opposite poles, so each pole receives one copy of every chromosome.
4. Telophase: the chromosomes arrive at the poles and decondense. A nuclear membrane re-forms around each set, the nucleolus reappears and the spindle breaks down.

Why the daughter nuclei are genetically identical

Replication made the two sister chromatids of each chromosome identical. In metaphase every chromosome is attached to both poles, and in anaphase exactly one chromatid of each chromosome goes to each pole. So each daughter nucleus receives a complete set with the same number and type of chromosomes as the parent nucleus, carrying the same alleles.

Identifying the phases

In a stained root tip, most cells are in interphase (intact nucleus, grainy chromatin, a nucleolus). Then look at the chromosomes: long threads inside the nucleus (prophase); a single band across the middle (metaphase); two separating groups, often V-shaped (anaphase); two compact groups at opposite ends, perhaps with a cell plate between them (telophase).

A diploid nucleus (2n) has two copies of each type of chromosome, one from each parent, forming homologous pairs. A haploid nucleus (n) has just one of each type. In humans 2n = 46 and n = 23.

Why meiosis is needed

In a sexual life cycle two gametes fuse at fertilization. If the gametes were diploid, the chromosome number would double every generation. Meiosis halves the chromosome number, so haploid gametes (n) fuse to restore the diploid number (2n) in the zygote. Mitosis then copies this diploid set into every cell of the new organism. Gamete production and fertilization are covered in D3.1 Reproduction.

Two rounds of segregation

DNA is replicated once, before meiosis, and the nucleus then divides twice. Switch the 3D model to Meiosis to follow the steps.

Meiosis I (separating homologues):

• Prophase I: chromosomes condense, and homologous chromosomes pair up to form bivalents. Non-sister chromatids cross over.
• Metaphase I: bivalents line up on the equator, one homologue on each side. Both chromatids of each homologue are attached to the same pole.
• Anaphase I: homologous chromosomes are pulled to opposite poles; the centromeres do not divide.
• Telophase I and cytokinesis: two cells, each haploid, with one chromosome of each type, still made of two chromatids.

Meiosis II (separating sister chromatids), in each of the two cells, without any further DNA replication:

• Prophase II and metaphase II: a new spindle forms and the chromosomes line up individually on the equator.
• Anaphase II: centromeres divide and sister chromatids move to opposite poles.
• Telophase II and cytokinesis: four haploid cells, each with one chromatid of every chromosome type.

Meiosis sometimes goes wrong. In non-disjunction, chromosomes fail to separate correctly:

• in anaphase I, both homologues of a pair move to the same pole; or
• in anaphase II, both sister chromatids of a chromosome move to the same pole.

The resulting gametes have one chromosome too many (n + 1) or one too few (n − 1) of that type. Fertilization of an n + 1 gamete by a normal gamete gives a zygote with three copies of one chromosome, a trisomy.

Down syndrome

Down syndrome is caused by trisomy 21: the person has three copies of chromosome 21, so 47 chromosomes in each cell instead of 46. It is usually the result of non-disjunction during meiosis, and in most cases the extra chromosome comes from the egg. People with Down syndrome have some learning difficulties and characteristic facial features, and a higher risk of some conditions, such as heart defects; how strongly they are affected varies widely.

Maternal age

The proportion of babies born with Down syndrome rises with the mother’s age, slowly up to the early thirties and then steeply. One hypothesis: all of a woman’s oocytes begin meiosis before she is born and stay paused in prophase I until ovulation, so in older women the chromosomes have been held as bivalents for decades, and the structures that keep them together may deteriorate. The data show a correlation; on their own they do not show the mechanism.

The four cells produced by meiosis are genetically different from each other and from the parent cell. Two processes in meiosis cause this.

Random orientation of bivalents

In metaphase I each bivalent lines up on the equator, with the maternal homologue facing either pole. The orientation of each bivalent is random and independent of the others. So each gamete gets a random mixture of maternal and paternal chromosomes.

• With n pairs of homologous chromosomes there are 2ⁿ possible combinations.
• With n = 2 there are 2² = 4 combinations; humans (n = 23) can make 2²³, about 8.4 million.

Crossing over

In prophase I, while homologues are paired, a non-sister chromatid from each homologue breaks at the same position and the two exchange the corresponding sections. The places where they remain joined are chiasmata. Each bivalent usually has at least one. Crossing over produces recombinant chromatids that carry new combinations of alleles of genes on the same chromosome (linked genes), for example A with b instead of A with B. Because crossover positions vary, the number of possible combinations becomes effectively unlimited.

Random fertilization

On top of this, any one egg can be fertilized by any one of millions of genetically different sperm. Counting random orientation alone, two parents could produce 2²³ × 2²³, about 7 × 10¹³, genetically different zygotes. This variation is the raw material for natural selection (D3.2 Inheritance builds on how meiosis shuffles alleles in inheritance).