Journal and Proceedings of The Royal Society of New South Wales Volume 120 Parts 1 and 2 [Issued September, 1987]

Дата канвертавання21.04.2016
Памер105.81 Kb.
1   2   3   4   5   6   7


Now, as I mentioned above, evolution can only continue for so long by the replacement of one existing allele by another. If the response to selection involves only the shuffling of preexisting genetic variation, without new alleles arising by mutation and being selected for, then evolution will not continue long. But this is not the case: the occurrence of favorable mutations has been repeatedly demonstrated. This demonstration is a classroom exercise for prokaryotes, because one can start with a single cell, known to lack a favorable trait (such as resistance to a given drug), and then demonstrate the occurrence of the trait in the progeny. Because prokaryotes have only one copy of each gene, the experimenter can be certain that the genes for resistance were not hidden in the initial bacterial cell. Long-running experiments of this kind can further produce significant improvements in the metabolic prowess of bacteria (Mortlake, 1983); these changes mimic the impressive powers of natural populations to evolve the abilities to use new carbon sources, of which a prime example is a wholly new enzyme attacking nylon (Olino, 1984)!

The detection of favorable mutations in eukaryotes is a little harder, because the habitual diploidy of many eukaryotes means that it is much harder to be certain that genes selected for were not present originally as concealed recessives, rather than newly-arisen by mutation. Yet, here too, the demonstration of favorable mutations has become both commonplace and economically valuable, through the induction of agriculturally-destrable mutations with radiation. Such induction is most easily done nowadays using large cultures of single cells from which whole plants can be grown, but has a long history. For animal populations, the study by Ayala (1966) is worth noting. Ayala inbred laboratory populations of the flies Drosophila errata and D. birchii, thus reducing both the level of variation and the average fitness of the population. Radiation led to the irradiated populations achieving larger sizes than non-irradiated controls (Figure 3), thus demonstrating that there are favorable mutations in amongst the harmful ones.


But is it worthwhile looking at the effects of single loci? It is a truism in genetics that the overall makeup of an organism results from a complex interaction between its genes and between them and the environment in which it develops. In fact, what evolves is not a collection of hair, skin, eyes (or cilia and flagella, or leaves and roots), but rather a developmental system that produces such characteristics under appropriate conditions. Given such complexity, will changes of one allele for another at one or just a few loci make any difference?

Yes. We have evidence from two sources that single gene changes are important in evolution.

Firstly, there are not as many functional genes as was once thought. The view of only two decades ago that organisms are determined by millions of genes appears quaint now, but it is easily understood. After all, there are about 2.7x109 (yes, only a little under three billion) nucleotide pairs in mammals such as the house mouse (Sang (1984) gives a table of genome sizes in various animals). An average protein is about 400 amino acids long, so that it could be encoded by 1,200 nucleotides, meaning that there is enough DNA in mammals for 2,250,000 genes.

But we now know that most eukaryote genes are made up to a large extent of introns, sequences within them that are cut out of the primary RNA transcripts before they leave the nucleus and which therefore do not code for amino acids. For example, the protein ovalbumin from chickens could be coded for by just 1,879 nucleotides (the length of the final messenger RNA), but the gene itself is stuffed with introns and is about 7,600 nucleotides long (Sang, 1984)! Furthermore, much of the genome is made up of spacer sequences between genes and various other noncoding sequences, so that the dramatic differences in DNA content sometimes found between relatively closely related organisms (e.g., the pea Pisurn sativum has more than nine times as much DNA as the mung bean Vigna radiata: Sang, 1984) reflect differences primarily in the non-coding and not in the coding sequences.

How many genes are there then coding for the complexity of organisms? And how can we find out? For the tightlyorganised genomes of prokaryotes, lacking the introns and general profligacy of non-coding DNA of higher organisms, dividing the amount of DNA by the average size of a gene is enough (Watson, 1977). To estimate the relatively rare coding sections in the comparative vastness of a eukaryote genome requires indirect methods, such as estimating the number of loci at which lethal mutations can occur (Raff and Kaufrnan, 1983) or by estimating the number of different messenger RNA molecules produced (Raff and Kaufman, 1983), and using this figure as an estimate of the number of active genes. By such means we can arrive at the following highly approximate numbers of genes in three organisms whose disparities are only matched by the affection that geneticists feet for them:



Escherichia coli (bacterium)


Drosphila melanogaster (fly)


Homo sapiens


It would be a mistake to think that these figures mean that humans are only ten times as complex as bacterial! The complexity of a developmental system depends on the interactions between genes, and is therefore dependent on some power of the number of loci, and is not a simple arithmetic function of the number. As an analogy, consider the complexity of a telephone exchange: the number of possible dyadic connections increases with the number of telephones, n, according to:

Σn (i - 1)

where i = 1 through n

so that, for three telephones there are three connections, but for six there are fifteen. The complexity of the system grows much more rapidly than the number of subscribers.

Fig. 4. Heads of males of (left) Drosophila heteroneura and (right) D. silvestris. Fom Kaneshiro and Val (1970), reprinted with permission

More direct evidence that single genes do matter comes from studies on interspecific differences. One rather spectacular example concerns the closely related flies Drosophila heteroneura and D. silvestris, whose males have dramatically different head shapes (Figure 4). These two species are extremely similar genetically, and can be induced to mate under laboratory conditions, when they yield viable and fertile hybrids (Vat, 1977; Templeton, 1977). Analysis of the head shape in the hybrids and backcrosses shows that the differences between the two species in male head shape are determined by only about six segregating units (Lande, 1981). Because these segregating units may contain two or more tightly linked genes, the number of gene loci is probably around about 12.

The case of these Hawaiian Drosophila, and others such as that of the peppered moth, show that major evolutionary changes may involve substitutions of one allele for another at as few as half a dozen loci, with further evolution involving “fine tuning” at other loci to ameliorate side effects of the major changes. There is, for example, some evidence that the heterozygotes for the allele for black coloration in Biston betularia have come more nearly to resemble the homozygotes for this gene today than was the case last century (Wallace, 1981).

1   2   3   4   5   6   7

База данных защищена авторским правом © 2016
звярнуцца да адміністрацыі

    Галоўная старонка