Thyroidal iodine metabolism in inbred and F1 hybrid mice

Thyroidal iodine metabolism was studied in four inbred strains of mice and two groups of F1 hybrids by use of radioiodine. Significant strain differences were found in both the 48-hour thyroidal retention of I131 and its output rate constant. Three closely related groups studied were C57BL/6, C57BR/cd and BBF1, their F1 hybrid. The other three groups included A/Jax, BALB/c and CAF1, their F1 hybrid. C57BL/6 mice had a significantly faster output rate and lower 48-hour I131 retention than any other group. The first three groups listed had a more rapid output rate and lower 48-hour retention than the last three. In intergroup comparisons and inverse relationship between I131 output rate and 48-hour retention was clearly shown. The results indicate that in mice pituitary TSH output, as indicated by thyroidal I131 output rate and thyroidal iodine pool size, is controlled by separate genetic factors.

Evidence of heterosis in man

In animal species the principle of balanced polymorphism due to heterosis has been accepted for many years, but the serious application of the same idea to man has been slow. Human experimental breeding is not possible, and effects, demonstrable in laboratory animals, may be concealed in a randomly breeding population. On the other hand, populations under natural selection may be suitable for certain kinds of studies not possible in experimental work. Evidence for the existence of some kind of hybrid vigour in man comes from both direct and indirect measurement. Most of the evidence is indirect, but the few examples of direct measurements on known or supposed heterozygotes may be considered initially.

Heterosis and morphism

I shall deal only with the heterosis concerned with the genetic balance-mechanisms underlying morphism (balanced polymorphism as defined by E. B. Ford). All morphisms found in nature must involve a selective balance between the morphs (R. A. Fisher), and most are either extrinsically adaptive (in adapting the species to a number of widely distinct conditions or habitats), or long-estabhshed (as in Primate blood-groups, Cepaea pattern, etc.), or both. Accordingly, they provide special opportunities for studying the origin, function and evolution of any heterotic mechanisms concerned.

On the biochemistry of heterosis, and the stabilization of polymorphism

The participants in this discussion have mainly considered two theories of heterosis. One approximates to the old view that heterosis increases the probability of the presence of at least one allele of the optimal type at each locus. This is probably true for loci where mutation is constantly producing suboptimal but not lethal alleles. These will only slowly be eliminated by natural selection, particularly when they are closely linked with optimal alleles. The second theory is that a heterozygote may be more versatile, that is to say, adapted to a wider range of environments, either internal, that is to say, in different tissues of the same organism, or external, than either of the corresponding homozygotes. I have held this view for some time (Haldane 1949 a ) and Lewis’s contribution strongly supports it.

Gene structure and action in relation to heterosis

The working hypotheses proposed for the interpretation of heterosis in terms of gene action can be reduced to two essential types. One is that of interaction between non-allelic genes; the other that of interaction between alleles. Examples of both types of action are, of course, known in physiological genetics. But while many examples of gene interactions have been analyzed in biochemical terms, only two of ‘one-gene heterosis’ have been so analyzed: sickling in man (Pauling, Itano, Singer & Wells 1949), and pab in Neurospora (Zalokar 1948). Physiological genetics does not provide any crucial argument to choose between a model of heterosis based on gene interactions and one based on allele interactions. Furthermore, advances in our knowledge of gene structure and action in recent years have led to the realization that there may be no absolute distinction between alleles of one gene and alleles of different genes. If this is so, the distinction between the two types of models for heterosis is one which has no longer a precise meaning, though it may still be useful at certain levels of approximation.

The expression of hybrid vigour in Drosophila subobscura

By hybrid vigour we mean the possession by outbred organisms of a number of characters which would confer fitness in a wide range of environments. This definition deliberately includes both cases where heterozygous individuals are fitter in a Darwinian sense in the wild, and cases of species or other distant hybrids showing vigour in, for example, efficient utilization of food, long life, or high resistance to disease, although of low fitness because of infertility or for some other reason. Both types of phenomenon are well known, but whether it is convenient to consider them under the same heading will depend on the aspect of the problem being studied. For example, Dobzhansky (1950) preferred to divide hybrid vigour into ‘euheterosis’ as found in wild populations, and ‘luxuriance’ as exemplified in species hybrids. Such a distinction is a helpful one when the aspect studied is the role of natural selection in maintaining a balanced polymorphism in the wild. However, the high level of heterozygosity maintained by natural selection in many wild populations suggests that for many pairs of alleles the heterozygote is fitter than either homozygote. Where we have some knowledge of the mode of action of the different alleles present in a wild population at a given locus, as we have, for example, for the blood groups in man, it appears that different alleles produc qualitatively different substances. This has led to the suggestion that outbred organisms may be biochemically more versatile (Haldane 1948, 1954; Robertson & Reeve 1952). It is such versatility that all ‘hybrids’ may have in common, whether they result from outbreeding in the wild or from distant crosses made in the laboratory.

Heterosis from the point of view of the chromosomes

In the present symposium we have seen the notion of heterosis expand from the vigour and cross-breeding of Darwin to the fitness and fertility of Dobzhansky and the balance of Mather. The processes by which this expansion has taken place have been illustrated with the help of almost every technical, material and theoretical resource of genetics. The complexity of the problem, it might seem, has been fully exploited. It has, indeed, reached new levels in Lewis’s experiments with the physiology of flowering in the tomato and in Rees’s experiments on chromosome behaviour with inbred rye which provide an elegant counterpart of those with out bred Allium of the cepa-fistulosum series. To get the full value out of the problem, however, one more point of view is worth noting. Two kinds of balance have been discussed in relation to heterosis: the balance between alleles and the balance between non-alleles. It has been difficult to say where one begins and the other ends. This difficulty has partly been clarified by Pontecorvo in terms of breeding analysis. It may be a little more clarified in terms of chromosome analysis. Within a narrow community such as Mendel’s peas the contrast between inbred and outbred individuals is a contrast between homozygotes and heterozygotes in respect of small differences, differences of an order that we may call intra-genic . In such narrow groups it is differences in balance between alleles that matter. Within a highly differentiated group such as cultivated maize or the human species the differences concerned can certainly include the results of intergenic or structural change. Indeed, within such larger groups we may expect to find different orders of hybridity. In such groups not only does the size of the unit of recombination expand and with it the scope of allelism, but the very nature of allelism may change. We may find two partners of structure ABCDEF one of which becomes ABCxDEF . The particle x must certainly come from some other part of the chromosome complement. Here the physiological distinction between allelism and non-allelism breaks down. Or, to put it in another way, allelism drops its physiological meaning and is left with a purely mechanical meaning.

Heterosis in chromosome behaviour

Most examples and measurements of heterosis are concerned with the more familiar features of the phenotype, with such characters as size and growth rate of the body or its outer parts. This account is concerned with less accessible, though not less important, parts of the phenotype, and describes heterosis as it is shown in certain aspects of the behaviour of chromosomes. We can consider and investigate the chromosomes in two ways. First, in so far as they consist of genic material, they are genotype . Secondly, because chromo­somes themselves are subject to the control of the genes they bear, as well as to the effects of external conditions, they are phenotype ; and since heterosis is a roperty of the phenotype it is with this phenotypic aspect of the chromosomes we shall mainly be dealing.

Gene interaction, environment and hybrid vigour

The manifestation of heterosis has been found mainly in characters which are the end-points of complex processes of growth and differentiation and controlled by many genes, such as weight, height, organ size or shape and rate of maturity. Even a character like the synthesis of a particular anthocyanin, which can be completely blocked by a single major gene, is no real exception, for the amount of synthesis is affected by a large number of interacting genes. Anthocyanin synthesis shows heterosis in Antirrhinum species hybrids (Dayton 1954). Again the bio­chemical models of heterosis in Neurospora are, in one example, the result of complementary effects of two dominant genes controlling amino-acid synthesis (Beadle & Coonradt 1944). The other example shows a heterotic effect of two alleles at one locus, one a sulphonamide-requiring mutant and the other its sup­pressor, but the effect is caused by a restoration of a delicate synthetic balance involving homocysteine, p -aminobenzoic acid and methionine (Emerson 1952). The two types of gene interaction generally postulated to explain heterosis— (1) complementary dominant genes which are non-allelomorphic and (2) over-dominance of alleles at the same locus—have different consequences in breeding behaviour, differences of great importance in a breeding program aimed at utilizing heterosis. But in terms of gene action there is, on the modern view, little if any difference between them. One criterion adopted in genetics which is useful in distinguishing two closely linked genes from alleles is that, when two hypomorphic genes together restore the wild phenotype—in other words show heterosis—then they are not allelomorphic but are closely linked genes in repulsion with complementary effects. On this basis a ‘single locus’ with an heterotic effect would be evidence not for over-dominance but for the compound structure of the locus, This complex locus would then be distinguished from two complementary dominant genes only in the strength of the linkage between its components. Thus even if we accept the evidence for ‘single-locus’ heterosis (cf. Schuler 1954), there is, as yet, nothing to contradict and much to support complementary dominant genes as the main cause.

Heterosis and variability in the mouse

The relation between heterosis and variabihty discussed by previous speakers i well illustrated by an example in the mouse where the structure of the second thoracic vertebra (vertebra prominens) is very variable; it may have a broad spatulat spinous process (+ + +), a round rod-shaped one (+ +), a mere spike (+), or then may be no osseous spinous process at all (—). As shown in table 20, the thre inbred strains CBA , C 57 BL and A differ much from each other ( means and variances have been calculated by ascribing arbitrary numerical values 3, 2, 1 an 0 to the four classes). In the cross CBA x C 57 BL , the variance in F 1 , F 2 and the backcross is much smaller than in either of the two parent strains. By contrast, i the cross between C 57 BL and A , the F 1 variance is between that of the pare strains, and F 2 is distinctly more variable. The same character thus has a great reduced variance in the first cross, but not in the second. The first cross led striking heterosis in F 1 as judged by the general vigour and the reproductive behaviour of the animals; the second cross showed little or no heterosis. While it is thus plausible to relate reduced variance to heterosis, nothing is known in the mouse as to why certain crosses between inbred strains result in marked heterosis while others do not.

Hybrid vigour in plant breeding

Hybrid vigour is of interest to the plant breeder because of its beneficial effects on the yield and quality of domestic plants. These benefits may be produced in various ways. Müntzing (1954) finds that the superior yield of F 1 hybrids of tetetraploid rye is the result of small increases in several components of yield; for example, a better stand of plants per plot, more flowers per plant and improved set of seed per inflorescence. Improved vigour may enable the plant to overcome environmental hazards and thereby enhance yield. In this connexion, it is note-Worthy that early-maturing hybrids have made possible the growing of corn in many new areas of Europe (Jugenheimer 1955). Maximal expression of heterosis is shown in F 1 hybrids, and hence, in crops which are raised from seed, the crosses must be repeated to keep up a continuous supply of planting material. This is an expensive process and one which is a serious obstacle to the commercial utilization of hybrid vigour, particularly with agricultural crops. More favourable economic conditions are found in some horticultural crops, notably tomatoes, egg plants, cucurbits and onions. Here the flowers are relatively large, easily handled and one pollination produces many seeds. In addition, the produce from one plant has a significant commercial value when viewed against the background of the total crop.

The genetical basis of heterosis

Appreciation of the practical value of hybrid vigour is as old as the mule, but its scientific investigation began only relatively recently. Vigour transcending that of the parents was observed in hybrids by a number of the early hybridizers and, indeed, Mendel himself records it as an incidental observation on his peas; but it remained for Darwin to make the first experiments designed to bring out the essential nature of the phenomenon. The results of his extensive investigations are set out in his book The effects of cross- and self-fertilization in the vegetable kingdom (1876), where he confirms the widespread occurrence of hybrid vigour. His inbreeding and crossing experiments enabled him also to show that the act of crossing was not itself sufficient to produce extra vigour in the offspring. Rather this vigour resulted from the ‘sexual elements (being) in some degree differentiated’. The differentiation he attributed to ‘individuals having been subjected during previous generations to different conditions or to their having varied in a manner commonly called spontaneous’. This is not quite the form of expression we should now choose; but while our genetical understanding amplifies these findings of Darwin, it brings no basic contradiction, for hybrid vigour depends on genetical dissimilarity of the gametes which fuse. Darwin also recognized what we would now call inbreeding depression, and attributed it to ‘the want of such differentiation in the sexual elements’, or to put it in our words, to the absence of genetical differences. The rise of Mendelian genetics in the present century made possible the re-examination of Darwin’s and similar findings in terms of the particulate factors or genes which we have come to recognize. Such a re-examination was made all the more urgent by the prospective practical value of hybrid vigour, especially to the breeder of cross-fertilizing crop such as maize, as the work of Shull, East and Jones had made clear. One question was particularly insistent. Did the extra vigour spring directly from the heterozygosity per se of the hybrids, or did it reflect simply the superior gene content possible in a hybrid where both parental gametes are bringing in desirable genes? The decision is of practical as well as theoretical significance, as on the one interpretation the vigour would be a direct property of heterozygosity and as such would never be fixable, whereas on the other view it could in fact be fixed—once all the desirable genes could be assembled m one set of chromosomes, so that male and female gametes could be produced like each other in that each included all the genes necessary for full vigour. Homozygotes could then be made with properties as desirable as the earlier hybrids.

Controlled heterozygosity in livestock

The word heterozygosity appears in the title of this paper in preference to heterosis for three reasons. First, heterosis has different usages and might be misleading. A simple illustration is provided by the colours of Shorthorn cattle which are determined by a pair of alleles. The heterozygotes, which are roan, may fairly be described as intermediate between the red and the white of the two homozygotes, and thus show no dominance and therefore no heterosis. Since, however, breeders prefer roans, they are maintained in breeding herds at a frequency greater than 0·5 (Rendel 1952).* On Dobzhansky’s (1952) definition of heterosis, this would be an example of single-locus heterosis and over-dominance. A similar situation obtains in respect of the white saddle of Wessex Saddleback pigs (Donald 1951). for neither the cattle nor the pigs is there evidence that breeders’ preferences for heterozygotes lead to higher production of meat or milk. Second, too little is known about metrical bias and scaling (Mather 1946) which are basic to a proper onception of the dominance relations involved; and third, heterosis in any current sense is not sufficiently comprehensive to describe the uses which controlled heterozygosity has or might have in livestock breeding. One simple and common method of increasing heterozygosity is to mate animals rawn from two distinct breeds. As may be judged from their appearance and their ross-bred progeny, many of these breeds differ substantially in genotype. The ncrease in heterozygosity consequent on crossing them has the assured merit of Adapting the progeny to market demands and the probable merit of improving Performance through heterosis.