Multicellular organisms have various ways of producing a next generation. Some, like bacterial biofilms, are essentially aggregates of otherwise free-standing cells that
Genetics and the Logic of Evolution, by Kenneth M. Weiss and Anne V. Buchanan. ISBN 0-471-23805-8 Copyright © 2004 John Wiley & Sons, Inc.
reproduce by individual cell division. The biofilm does not reproduce as a single entity. Slime molds are another class of cells that normally live as individual single-celled organisms, although under some circumstances (to be described in Chapter 12), the cells are induced to come together to form multicellular aggregates, taking on many properties of multicellular organisms. At this slug stage, they have internal organization, signaling, various sensory behavior, cells that undergo apoptosis, and the like (e.g., Bonner 2000), and these reproduce basically as integrated organisms in which not all cells contribute directly to the reproductive process. Some simple multicellular animals, like sponges, can produce small aggregates of cells that shed as primitive larvae to become new organisms. A specific cell type is generally the precursor, but these are located throughout the parent's body. In plants, many or even most cells have the potential to produce an entirely new plant on their own, as even small pieces can regenerate complete new plants.
Most complex animals and plants have a specialized form of reproduction that we tend to treat as a standard for reproduction, although it is by no means the only or "best" form. In these, a type of cell, the gamete, is used only for reproduction. The organisms begin life as a single cell and grow through mitotic cell division. In many organisms, a lineage of germ line cells is set aside, usually early in development, and reserved for gamete production, while the rest of the organism's functions are carried out by other lineages of somatic (body) cells. Somatic cells may be replaced, lost, or even actively killed by apoptosis during life, and they can undergo natural selection that can even take the life of the organism (that is what cancer is). But although they are the product of a form of reproduction, and "beings" within the higher-order being known as an organism, somatic cells make no contribution to the next generation. Genetically, gametes are produced by meiosis and have half the chromosome complement of the somatic cells (e.g., are haploid vs. diploid).Typ-ically, reproduction occurs by the union of gametes from different individuals, but self-fertilization can occur and some species reproduce by parthenogenesis, in which the organism's genome is fully represented without a separate fertilization stage; there is wide variation in the ways this kind of reproduction occurs in nature.
Somatic cells experience mutation just as germ cells do; however, although mutations accumulating in somatic cells may affect the organism (and cancer and other changes associated with aging are examples), only those in the germ line affect the next generation. Hence, in organisms reproducing via single gamete cells, this is where the evolutionary memory lies. Of course, somatic mutations may have an evolutionary impact if they determine whether a particular organism's gametes are transmitted.
One theory explaining the early and active sequestration of primordial germ cells is that they can escape the chromosomal modification by methylation that is used to control differential gene expression in somatic cells. Before transmission, gamete genomes in some species (including humans) are systematically imprinted, sometimes differently in males and females, by methylation at specified sites on the chromosomes or by other means, affecting how and/or when genes transmitted from that parent are used in the offspring. Another common characteristic of germ line lineages is to undergo fewer cell divisions during the life of the individual (Buss 1987); we may think of this as protecting the patrimony from mutational damage, but it is not always true. Sperm-producing cells in mammals undergo many cell divisions, with evidence showing that this results in correspondingly more mutations occurring through the male lineage (e.g., Crow 2000).
However, there is no one rule about this. Plants and many other organisms do not even sequester a special germ line; flowers, for example, reproduce via sexual reproduction but can be produced on hundreds of different stems. Some species, including many plants, are hermaphroditic, that is, the individual contains, and functions as, both sexes. And not all reproduction occurs through single gametes.
Most cells in multicellular organisms toil their lives away at the evolutionary service of the few cells that will contribute to the next generation. However, in sexually reproducing organisms, only half the genome is transmitted to a given offspring anyway. The prevailing view is that, for evolution to produce individuals who make this genetic sacrifice, the cell and organism must gain some selfish advantage in exchange. A general explanation is that somatic cells are helping to advance the reproductive cause of their genetically identical germline kin. This is only partially true because accumulating mutations make each somatic cell different. We might thus expect furious natural selection among the varying cells in our bodies all the time, but, although there is much machinery to interfere with aberrantly behaving cells, there is no master genome controlling this. In regard to reproductive self-sacrifice, germ cells from the same organism are, at least, a somatic cell's closest genetic relative but no way has evolved for a really superior somatically mutated cell to make a germline copy of itself (though in principle something like this could occur in plants).
We can take a similar view of the colonies of certain organisms like ants and wasps, in which most individuals do not even contribute gametes. This has been seen as a theoretically instructive example of how sterility (e.g., in drones or workers) could evolve: their reproductive self-sacrifice leads to the high reproductivity of the queen they protect, to whom they are closely related by a special chromosomal system called haplodiploidy (see below). But if this is compatible with evolution it is not inevitable: termites and naked mole rates are similarly eusocial, forming colonies with one breeding female, a few breeding males, and the rest nonbreeding, but these species are diploid, not haplodiploid.
Because of their locked-in reproductive fate, some biologists have suggested that it is meaningful to consider such colonies as organisms themselves. The idea has been around for a long time (Bateson 1894; 1913). Indeed, in many ways, our own somatic cells can be likened to the drones in a beehive, and our germ line to the queen. What an ant, or the set of drones, are to a hive may not be so different from what a liver is to a human. A major difference is that the individual ants are not stuck together, whereas, for example, individual nephrons in a kidney are. We could even consider a species, like humans, as a single large superorganism, with various sets of cells, aggregated in hierarchical ways and exchanging genetic variation over time. Genetically, this is not far-fetched at all because the responsible mechanisms are similar in nature.
Sexual reproduction is a nearly ubiquitous part of life and has evolved countless forms, many presumably having arisen independently. The most plausible evolutionary advantage leading to this is provided by the added variation produced by recombination of homologous chromosomes from different individuals. The partial sequestration made possible by sexual reproduction makes a species more able to respond to changes in the environment, while keeping the species from devolving into a set of totally sequestered lineages.
That so many multicellular organisms reduce their reproductive activity to the shedding of only single cells raises the question of whether sexual reproduction is still a remnant of single-celled life. After all, reproduction could in principle occur by fusing many cells rather than one, and among more than two organisms, taking advantage of somatic mutations and natural selection within each organism's life history. This would allow a kind of semi-lamarckian evolution to take place: somatic cells used in reproductive fusions would be screened by selection for those cells that behaved well, including those carrying favorable somatic mutations. That can't happen now in lineages long committed to meiosis-based sexual reproduction, and problems could arise in terms of making a unified cellular tree of differentiation and development, or of self/nonself immune recognition. But different mechanisms might have evolved, and plants (for example) don't have a simple unified tree of development (see below).
Of course, some organisms do reproduce by shedding sets of somatic cells, and the rather fluid, context-specific nature of aggregation in otherwise single-celled organisms forming biofilms and slime molds shows that something of this kind can occur (Bonner 1998). Still, for whatever reason, the single-cell route has been repeatedly successful; whether it is "best" or not, it works and may be easiest to evolve. These are facts of importance to the definition of an individual from evolutionary perspectives (Buss 1987) and affect the tempo of evolutionary genetic change and perhaps also some aspects of its mode of action.
Viruses contain nucleic acids that have normal protein-coding function, but they are not alive and replicating when free-standing and must infect cellular organisms to reproduce. There, they shed their protein coat, allowing their RNA or DNA (depending on the type of virus) to interact with relevant enzymes in the host cell and to be copied. New viral coat proteins are coded for by the viral genome and translated into protein by the host cell. Multiple copies of viral DNA/RNA, and viral coat proteins, can circulate within the cell; thus a given virus can be assembled from components that did not all derive from a single parental ancestor. Viruses can exchange genes by recombination within the cell before new particles are formed, further adding to their repertoire of variation. Another way viruses reproduce is by integrating their DNA (or DNA copies of their RNA) into the host genome. If this occurs in somatic cells, the viral genome is inherited somatically; in a germ cell, the viral genome is transmitted to the next host generation via the horizontal transfer referred to in Chapter 2.
Bacterial plasmids, chloroplasts, mitochondria, and many chromosomally integrated genes are examples of the past or present infectious transfer of parasitic DNA that can be replicated as units within a cell and inherited through mitosis and meiosis.
THE SPECIAL PHENOMENON OF SEX DETERMINATION We have described sexual reproduction from its general genetic point of view, which is a source of shuffling of chromosomes (and recombination during meiosis). But sexual reproduction requires much more than mechanisms for the production of male and female reproductive cells (sperm, egg, pollen, ovule, and correspondingly different cell structures in yeast and bacteria). Making gametes is just the beginning: they (that is, the organisms that carry them) have to find each other. This burden clearly reflects the importance of whatever advantage sexual reproduction must have. It has led to so many differences in behavior and morphology that sexual dimorphism can be the greatest single source of variation among individuals in a species.
There is extensive variation in the genetic basis of sex determination. Even bacteria and yeast occasionally reproduce sexually, often induced to do so by nutritional or other environmental changes that we tend to characterize as stress. Morphological changes occur to denote "male" and "female" cells, which fuse to form a temporarily diploid organism, in which recombination can occur. In the case of bacteria, a small chromosomal element known as the fertility factor (F) can transfer between bacteria. F+ cells form short projections called pili on their surface, in a process called conjugation that joins F+ and F- cells; genetic information can then transfer through chromosomal recombination (see panels A-D in Figure 8-1A) (e.g., Griffiths et al. 1996; Lewin 2000; Suzuki et al. 1998). On rare occasions, the F element is incorporated directly into the bacterial genome and can thereafter be transferred to other bacteria in subsequent conjugation.
There are many variations among species, but the basic mating systems of yeast involves a "cassette" system. Figure 8-1B shows this for the brewer's yeast Saccha-romyces cerevisiae (e.g., Haber 1998). A gene in a chromosome III location called MAT codes for mating-related proteins. Depending on the mating type, one of two flanking genes, HMLa or HMRa, replaces the gene currently at MAT. Chromatin structures symbolized by gray ovals repress these genes, but at the appropriate time in the life cycle, a cut is made at MAT. Then, controlled in part by flanking
Was this article helpful?