Portugaliae Acta Biologica - 1949
A Cytophysiological Theory of the Gene, Gene Mutation
and Position Effect

J. A. Serra

Secção de Zoologia-Anthropologia, Faculdade de Ciências. Universidade, Coimbra



Normally gene mutation of the orthodox type, that is mutation without any detectable change in the chromosomes and with perfect allely within a well defined locus, is hypomorphic [A nomenclature on gene action similar to that of Müller but somewhat modified is followed here.] to the standard (generally the + allele) and by dosage of the mutant it is possible to obtain the same effect as the + allele. Hypo-hypermorphic factors are isomorphic, do the same work at different levels of intensity. In phenogenesis this must correspond to the hypomorphic gene producing the same catalyst as the hypermorphic or + allele but in a lower amount or, in the extreme (amorphs), none at all.


A logical corolary of the assumption that the genes act rather by modifying the balance of a system than by producing a single enzyme, is that the genes, if they are to have an unitary action, must not produce the enzymes through which they act. Many of the enzymes are known to require coenzymes of external (food) origin and therefore at least a part of the enzyme proceeds from the ambient. The proteic part (apoferment) of the enzymes must also in part be a cytoplasmic elaboration; well marked morpho-physiological changes in the cytoplasm of cells elaborating secretions and in active growth demonstrate that the cytoplasm actively partakes in their synthesis. Of the antibodies it is also known that the structure of a part of their milecule must be achieved in contact with the antigens, which serve as templets. It is unlikely that voluminous molecules of antigens may reach the genes; and even if this was possible the genes should produce the -globulins complementary to the haptene groups only when the antigen was "anchored" at the chromonema, which seems a very remote assumption. It is presumable that the cytoplasm produces the greater part of the proteins and that the nucleus only furnishes the basic units for the enzymatic systems which compose the synthetic apparatus that the nucleus only furnished the basic "Anlagen", which determined the order of the aminoacids, and the rest would be formed in the cytoplasm in a manner somewhat similar to the synthesis of antibodies.

The indirect, and by no means admittedly complete, evidence rather points out that the genes produce only a part of the enzymes through which they act, namely the basic templates, perhaps the minimum polypeptide chains or haptene groups which serve for the synthesis of the specific apoferments, achieved in the cytoplasm. The lack of production of one basic anlage, or its production in a different amount, will induce that the system of apoferments becomes modified and the action of the enzymatic system results unbalanced. The final means through which the genes act, according to this hypothesis, is a modification of the balance of an enzymic system; however, the primary gene product, synthesized by the gene itself, may be very different from the enzyme which finally is responsible for the modification of the character and do not necessitates of being a big molecule, a small but strictly specific haptogene (haptene group synthesized by the gene) being very probably sufficient.

If this is true, it is to be expected that by studying the antigens produced under the action of a certain gene it will not be easy to obtain a direct and immediate knowledge on the composition of the respective gene. In reality, the chain leading from the apoenzyme, or the antigen formed, back to the gene must include at least two other links: haptogene and nucleoplasm-cytoplasm; on the other hand, the synthetic processes taking part in this latter are very complicated, involving the formation of granula of lipo-proteins, or of nucleo-proteins, or even of lipo-protein-nucleic acit complexes and in animals also a series of non-essential aminoacids and other simple compounds must come from the food.

This interpretation of gene action is important for a discussion of gene nature. In effect, if the genes themselves were the complete templets for the primary gene-products, it would be necessary for the chromonema possessed as a kind of side-groups all the specific apoferments, antigens and other biologically specific protein compounds, which seems a rather improbable hypothesis in view of the big size of these molecules and the difficulties or even impossiility of they being reduplicated, at least if their molecules were in a globular or pliated state. On the other hand, the hypothesis of the haptogenes only requires that small regions of the chromonema can serve as templets where relatively simple but strongly specific polypeptide chantes are synthesized.

Another important corolary of the haptogene hypothesis is that gene reproduction and specific gene functioning (this latter in the meaning of a specific action in phenogenesis) may be different activities. This conclusion stands in contrast with a generally admitted hypothesis which supposes that synthesis of primary gene products and gene reduplication are the two sides of a same basic activity. Our conclusion, however,corresponds more closely to what is really known on mutation and position effect and is also in accord with the fact that genes and viruses cannot reproduce and function outside a cytoplasm, with which they form an integrated unit. Gene reproduction is achieved as an activity of the gene-cytoplasm complex, in which the gene furnishes the templets or haptogenes and the cytoplasm in its turn elaborates molecules of an higher order which serve in phenogenesis and for gene reproduction. This hypothesis is represented in Fig. 4.

Fig. 4 Scheme of the interactions between nucleus and cytoplasm, in haptogene production and gene reduplication. See the text.

According to this view, the chromonema and its genes may have a composition and structure very different from that of the primary and the final products, only small parts of the chromonema serving as the templets for haptogene formation. The chromonema may thus have a complicated structure with elaborately structured molecules of pliated polypeptide chains (and perhaps also of nucleic acids) which do not have necessarily to be reduplicated *en place*, as a sole activity of the genes themselves, starting from the component aminoacids and nuctides. The functions of the gene are at the same time simple in what concerns the primary synthesis of haptogenes, and complicated with respect to reproducing themselves, and this latter is a complement of the primary action of haptogene production, since only after the cytoplasm has elaborated complicated compounds on the basis of haptogene templets may gene reproduction be achieved. Gene reproduction is growth of the apposition type in which the cytoplasm actively intervenes, so that the cell (or the virus-host cell complex, in the case of viruses) works as an integrated complex and in fact gene reproduction must be secondary to cytoplasm synthesis and not the reverse, as is usually admitted.

The chief implications of this hypothesis of gene action which are of importance for an understanding of the theory of gene nature discussed in chapter 7 are: 1) The gene as a whole may have a structure very different from that of the primary gene product. This latter seems to be an haptogene of relatively small molecule, probably rather simple polypeptide chains with a certain specific structure. The haptogenes are further elaborated in the nucleoplasm and the cytoplasm to apoferments and other specific proteins and these serve afterwards as basic units for vital processes, including gene reduplication. 2) Gene reduplication and gene action are different activities, it being possible to have one without another; gene reproduction is a part of the growth of protoplasm, which depends on the cytoplasm as well as on the genes, these latter being unable to grow if relatively complicated components are not supplied to them from the cytoplasm. 3) The composition of the chromonema and the genes may approach that of the rest of the cell, and particularly that of the cytoplasm, without at the same time the genes losing their specificity, which resides only in limited pieces of their structure.


From the physiological point of view, one of the chief properties of the gene is that the great majority of the known mutations are recessive to a standard, usually the wild type or some common form. Gene changes are generally of an hypomorphic recessive type. On the other hand, the pure position effects are generally dominant, while the mimic ones, by their very nature, produce recessive phenotypes like those of gene mutations. The causes of the gene mutations being generally recessive to the standard are twofold, depending on the mutation process and on the interaction of the gene with the rest of the genotype. It seems that, according to the species and the genes in question, one or the other of the opposite theories (of Fischer and Müller; and of Haldane and Wright) of the evolution of dominance may hold true. In effect, it is found in many cases that there are + alleles of different valency, all of them producing homozygously the wild type or another normal phenotype, but showing different potencies in relation to some mutated allele of the factor. On the other hand, it is also certain that for many genes a series of dominance modifyers exist, scattered over several chromosome loci. While the explanation of the influence of dominance modifyers is to be sought for in the action of these factors upon the reaction chains which lead from the enzymes to the phenotype and therefore is a question for phenogenic analysis, the existence of different levels of efficiency of a same factor must be explained by a comprehensive theory of gene mutation.

Gene mutation of the hypermorphic-hypomorphic type may be explained by assuming that there are different levels of inactivation, corresponding to different efficiency values. Even in the most extense allelic series, never a continuous transition from factor to factor is found; mutation appears always to be a quantic process, which implicates a stepwise change in some efficiency. The conclusion to be drawn from the known facts is that gene mutation usually is a quantic inactivation or activation process, which may be represented schematically by:

g0 g1 g2 gq gD gQ

in which 1, 2, ... q, ... D, ... Q are the quanta of mutation; g0 represents the ground or zero level of total inactivation, q is a general term, D the minimum level for which dominance is attained or minimum + level, and Q is the maximum quantum number.

The quantum number is approximately equivalent to the valency of the factor and both numbers would be strictly equivalent if the rest of the genotype was unchanged, that is if the lines with mutants were isogenic. Studies of valency with isogenic lines are as yet so rare that the quantum numbers, even of the most thoroughly investigated genes, are known only with a gross approximation. The ground level g0 is fully realized in small deficiencies of the locus and only rarely is observed in practice as a gene mutation. For example, it is claimed that in Drosophila w deficiencies give an eye more deprived of pigment than the usual w mutation, but from this it can not be concluded that there exists no w mutation with an action so extreme as that of the corresponding deficiency: this mutation must be very rare and difficult to separate from the next level, which is the usual w gene. At present it may be assumed that g0 really may be produced by gene mutation as well as by a corresponding deficiency.

The case of the maximum quantum level Q is yet more complicated. If the efficiency of the gene is such that above and including the D level the maximum phenotype is produced, it follows that only in compounds with a suitable inferior allele may the quantic levels D ... Q be recognised as separate. Only in a very restricted number of cases has this been attempted. The necessity of using isogenic lines further complicates the obtention of reliable data on this point. It would be very important, however, to secure such data for an understanding of gene mutation. At present only some guesses at this important question of the definition of quantum levels may be attempted. For bobbed (data summarised in Stern 1933) D lies at about a valency of 30, which probably corresponds to a quantum of 4 or 5 and the Q number must correspond to 8-10. By dosage it is possible to obtain valencies very much above 30, but this has nothing to do with the Q number, which in diploid organism respects to a normal number of only two genes, without duplications or any other complication. The same may be said of the intermediary steps of dosage obtained by the addition of factors with different quantic numbers, for instance of g1 + g3, or g2 + g3, etc.

The Q number indicates the levels of activation of the gene; the possible states of the gene in what concerns its activity are therefore Q+1, including the ground level. These states may be due either to gene mutation or to position effects of the mimic type. In fact, it is known that many alleles in extense allelic series are no more than position effects. The circumstance that a certain level of activity is obtained by position effect has nothing to do with the definition of the respective quantum level; in fact, for a definition of quantum level, that is a certain level of activity, the process by which the level is attained is irrelevant, since the activation or inactivation, to have any effect, must finally manifest within the gene. The Q+1 number defines all the possible levels of gene action in any circumstance, either when the gene is in its normal chromonema «ambient» or when it is influenced by a different ambient brought about by chromonema rearrangements. In the next chapters the importance of the q numbers for gene structure will be discussed with some detail.

Quantum numbers are difficult to define for the neomorphic type of action, which corresponds to the pure type position effects; this is comprehensible on the ground of these being really genes in the making. It is only at the measure that these rearrangements become established and mutations occur in them that the quanta may be defined. For such extreme pure effects as Bar, with a low viability, the possibility of the new genes becoming established is somewhat remote, but probably this does not happen with many other pure effects whose action is only little detrimental, or even that enhance the viability. This point will again be dealt with in the following section.