Relationship Between Various Chromosomal Changes. By M. Demerec Carnegie Institution of Washington, Cold Spring Harbor, N. Y., U. S. A.

Transcription

1 Relationship Between Various Chromosomal Changes in Drosophila melanogaster By M. Demerec Carnegie Institution of Washington, Cold Spring Harbor, N. Y., U. S. A. In order to understand the mechanism instrumental in producing various genetic changes such as gene mutations, lethals, inversions, and translocations it is essential to have a knowledge of the interrela tionship of these changes. It is particularly important to know whether or not any of them occur simultaneously and if they do, the frequency with which they occur. Coincident occurrence of certain changes would place the responsibility for their origin on the same mechanism and would simplify the problem if such coinci dence were found to be a rule. In the course of our studies of various genetic changes an ex tensive collection of small deficiencies, gene mutations, transloca tions, and inversions has been accumulated. This material has been studied cytologically and genetically and the results of these studies will be reported here. By X-ray treatment a number of lethal and visible changes were induced in various X-chromosome loci. To detect these changes females carrying several recessive characters were mated with treated males. If any character carried by their mothers appeared in the Fl females this indicated that a change in that particular locus was induced by the treatment. Thus, if ec ct v g f females were mated with+treated males, the appearance of ct in one of the Fl females would indicate that a change in et was produced in the sperm of a treated male. All lethal changes obtained in the loci selected for observation were kept in stocks but only a few of the visible changes were saved. This material has been collected from experiments con ducted since The dosage in various treatments varied between 2500 and 3000 r-units. In this manner a collection of about 80 lethals affecting visible loci of the X-chromosomes has been accumulated. Salivary chromo somes of 61 of these lethals and of 15 visibles obtained in the same experiments were examined for chromosomal aberrations. In cases where such an aberration was found it has been determined through salivary chromosome analysis whether or not the chromosomal break

2 1126 M. DEMEREC Cytolcgia, Fujii jub. vol. is located in the region where the lethal or visible change occurred. A summary of the results is given in Table 1. It is evident from these data that out of a total of 61 lethal changes studied, in 26 or 42.6 per cent of the cases a chromosomal aberration occurred in the Table 1. Summary of data showing the frequency of coincidence between various genetic changes induced by X-ray treatment of about 2500 to 3000 r-units. (C. A.= chromosomal aberration; In=inversion; T=translocation) same chromosome with the lethal change. The significant fact brought out here is the high degree of coincidence between the two types of changes. In cases where both changes occurred in the same chromosome, one breakage point of the chromosomal aberration co incided with the region where the lethal change took place in 92.3 per cent of the times. This extremely high coincidence shows that the same mechanism must be responsible for the origin of both types of changes, viz. of lethals and chromosomal aberrations, at least in cases where they occurred together.

3 1937 Relationship between various chromosomal changes in Drosophila 1127 The situation seems to be different in the case of visible changes. These changes were accumulated in the same or similar experiments as lethal changes. The X-ray dosage applied in producing these changes was similar to that used in producing lethals. Only one out of the 30 visibles investigated carried a chromosomal aberration (Table 1). In this one case, moreover, neither of the breakage points of the aberration coincided with the visible change. In 15 of these cases the analysis was made by studying salivary chromosomes and in the other 15 cases by the genetic method of linkage relation studies.a similar relationship between induced lethals and chromosomal aberrations is indicated in the results of another experiment in which 100 treated sperms, taken at random, were tested for various changes. The wild-type males were treated with about 2500 r-units and mated to se ec cv et v g f females which carried markers almost throughout the whole length of the X-chromosome. Aberrations involving the X-chromosome were identified through reduction in crossing-over and lethals were identified through non-appearance of wild-type males. The following results were obtained: Total number of tests Lethals Lethals and chromosomal aberrations... 5 Chromosomal aberrations... 5 Visible changes... 2 In this experiment, also, a lethal change and a chromosomal aberration occurred in the same chromosome in a high proportion of cases. Though the cytological studies have not been made, it is justifiable to assume in this case that a large proportion of these changes coincided. When this evidence became available it was interesting to find out the frequency of chromosomal aberration among spontaneous lethals. In order to do that, 45 spontaneous lethals occurring in males of various wild type stocks were collected by the CIB method. During this work it was discovered that the Florida wild stock car ried a recessive factor in the second chromosome which increases the natural rate of mutability in this stock (Demerec, 1937). This factor is effective early in the embryonic development and thus if a lethal originates in its presence in a male, several sperms will carry it. In routine tests such a lethal change may be counted as several lethals. Special precautions were taken to be sure that all lethals used in these tests were of independent origin. The results of genetic tests show that none of the 45 spontaneous lethals were connected with chromosomal aberrations. Oliver (1932) made a similar study

4 1128 M. DEMEREC Cytolcgia, Fujii jub. vol. with 10 spontaneous lethals and Sakharov (1936) with 25 lethals. They obtained the same results. Thus tests are available for 80 spontaneous lethals showing no chromosomal aberration coincident with the lethal change. Available data show that spontaneous lethals are independent of chromosomal aberrations while the lethals induced by X-ray treat ment are frequently connected with chromosome breaks. In Table 2 Table 2. Frequency of chromosomal abnormalities (C. A.) among lethals obtained from treatments with different X-ray dosages data are given on the frequency of chromosomal aberrations in the material treated with different dosages. Although there is a dis crepancy in the per cent of aberrations obtained by Oliver and those obtained in my experiments, it is evident from the data that the frequency of chromosomal aberrations increases rapidly with the increased dosage. A large amount of data is available showing the relation between the X-ray dosage and the frequency of lethals. In only one case (Oliver 1932) were these lethals studied further and the proportion of chromosomal abnormalities occurring among them was determined. Oliver's data are given in Table 3 and shown graphically in figure 1. Table 3. Oliver's (1932) data on the frequency of various kinds of lethals induced by X-rays. (C. A.=Chromosomal aberration)

5 11937 Relationship between various chromosomal changes in Drosophila 1129 It is evident from these data that the dosage-fre quency curve for all lethals is a straight line. Such a curve, however, slopes down for lethals without somal chromo aberrations and slopes up for lethals with chro mosomal aberra Fig. 1. Curves showing incidence of chromosomal aberrations among X-chromosome lethals induced by various X-ray dosages. (Oliver's data, 1932). tions. The significance of this observation will be discussed later. Random lethals dealt with in all experiments on the mutation frequency have not been investigated cytologically and it is not directly known whether or not they are deficiencies. There is ample evidence, however, to show that lethals affecting known loci are deficiencies. We studied cytologically lethals affecting the following loci: yellow-achaete, prune, white, facet, cut, tan, lozenge, vermilion, miniature, dusky, wavy, sable, garnet, tiny, forked and beadex. In every case they were found to be deficiencies and, moreover, mostly deficiencies for more than one salivary band, and presumably for more than one locus. It seems justifiable to assume that random lethals are similar to lethals of visible loci, and that they are also deficiencies. There is no doubt that among random lethals there are many which affect visible loci but they have not been detected as such since special tests are required to do that. Estimates on the number of loci per chromosome supported by counts of salivary chromosome bands indicate that there are at least ten times as many loci in the X-chromosome as there are known visible changes. It is undoubtedly true that some of these as yet unknown loci will be identified after more new mutations are found. Since new mutations are becoming scarcer and scarcer, in spite of greatly enlarged work with Drosophila, it seems very probable that there are many loci in which changes (mutations) show no visible morphological effect. If the estimate of the total number of loci is correct, it appears probable that the majority of loci are of a phenotypically non-detectable type. It is equally probable that deficiencies of some of these loci will have a lethal effect. The groups of random lethals, therefore, may be com posed, in part, of deficiencies affecting visible loci and, in a larger part, of deficiencies affecting non-detectable loci. There might be

6 1130 M. DEMEREC Cytologia, Fujii jub. vol. some lethals, however, which are not due to deficiencies but to gene changes having a lethal effect. The relative proportion of these two types of lethals will have to be determined through study of salivary chromosomes. This is a laborious process, especially when dealing with deficiencies affecting very fine bands. In a recent paper, Sakharov (1936) states that he examined salivary chromosomes of heterozygous females of 25 X-chromosome lethals without finidng any deficiencies. A similar work is being carried on at this laboratory by Dr. B. Slizynski with results different from those obtained by Sakharov. Of 13 spontaneous lethals which have been carefully studied, at least half have detectable deficiencies. This work is still in progress. Evidence obtained so far is in favor of the contention outlined above. Discussion From the data presented in this paper it is evident that when a lethal and a chromosomal aberration occur in the same chromo some, one breakage point of the chromosomal aberration is at the place where the lethal change is located in a great majority of cases (92 per cent). Since such a coincidence cannot be expected if these two changes were occurring at random, it is evident that both of them must be induced by the same mechanism. The data are also ample to show that chromosomal aberrations do not occur with equal frequency among lethals induced by different factors. No chromosomal aberration was found among 80 spontaneous lethals studied by Oliver (1932), Sakharov (1936), and myself, while almost fifty per cent of these induced by the 3000 r-units X-ray treat ment carried chromosomal aberrations. The frequency of these aber rations increases with the X-ray dosage applied. All lethals are not connected with chromosomal aberrations, nor do all chromosomal aberrations carry lethals even at higher dosages. What is the mechanism which can account for such conditions? There is no doubt about the interrelation of these two types of changes, but at the same time there cannot be any doubt that the type of change responsible for the lethal effect may occur indepen dently of a chromosomal aberration. Is there any certainty that the opposite is true? In order to give a plausible answer to this question it is necessary to consider the problem of lethal changes. It has been pointed out earlier that the majority of lethals are probably deficien cies. If that assumption is taken for granted, then the statement made previously may be restated as follows: deficiencies frequently coincide with chromosomal aberrations. Since our methods for

7 1937 Relationship between various chromosomal changes in Drosophila 1131 determining chromosomal aberrations are reliable, it can be stated with certainty that a deficiency may occur without a chromosomal aberration. Such an independence in origin, however, may not hold true for chromosomal aberrations since they may regularly be con nected with deficiencies undetectable by our ordinary methods. It has been shown (Demerec and Hoover, 1936) that a homozygous deficiency affecting as many as four bands may be viable and show no morphological effect on the organism. Such a viable deficiency can be detected in salivary chromosomes only. If it affects one light band or a few of them and is, in addition, connected with a chromo somal abnormality, it may be physically impossible with methods available at present to detect it at all. The possibility is not excluded, therefore, that such chromosomal aberrations as inversions and trans locations not showing a lethal effect are connected with short deficiencies at their breakage points. Let us now see if it would be of any help toward the solution of our problem if an assumption were made that chromosomal breakages are usually connected with deficiencies. It must be made clear at this point that I am considering here small deficiencies affecting only a few bands, and that as a rule such deficiencies, when they affect more than one band, are produced by some sort of block change. Such a block change is no doubt produced regularly through a chemical process, though the possibility is not excluded that it may be some times produced by the pinching out of a small section of a chromo some. If we assume that breakages, as a rule, are connected with small deficiencies, it would seem probable that these deficiencies might be responsible for the breakages, viz. at a spot where such a defi ciency is produced a chromosome is either regularly or frequently broken. Because of the attraction between broken parts, broken chromosomes have a tendency to unite again thus showing a defi ciency only. If, however, breaks should occur at more than one place in a chromosomal complex, subsequent fusion of broken ends may produce various chromosomal aberrations such as inversions if two breaks should occur in the same chromosome, and translocations if they should occur in different chromosomes. It would be expected then, that when the frequency of deficiencies is low they would not be connected with chromosomal aberrations, and the proportion of aberrations would increase with an increase in the frequency of deficiencies. This interpretation agrees with the observed facts. None of the recessive visible gene mutations (induced by the same treatment which produced among lethals about 40 per cent of chromosomal aberrations) was connected with chromosomal aberra tions. This suggests that visible gene mutations are different from

8 1132 M. DEMEREC Cytolcgia, Fujii jub. vol lethals. It makes it probable, also, that gene mutations, rather than gene deficiencies, are responsible for the majority of recessive visible changes. Summary From a total of 61 lethals affecting known loci of the X-chromo some, and induced by X-ray treatment of 2500 to 3000 r-units, 26 or 42.6 per cent had chromosomal aberrations such as inversions and translocations. In 92.3 per cent of cases one breakage point of the chromosomal aberration coincided with the region where the lethal change took place. Of 30 visible mutations induced by a similar treatment only one carried a chromosomal aberration. This, however, did not coincide with the region where visible change took place. Data available on 80 spontaneous lethals show that none was connected with either an inversion or a translocation. Among lethals induced by X-ray treatment, the frequency of such chromosomal aber rations increases with the increase of dosage. It has been determined by salivary chromosome studies that all investigated lethals affecting known loci are minute deficiencies. It is suggested that either all or a great majority of random lethals are also deficiencies, and that some of the small deficiencies may not have a lethal effect. Coincidence between a breakage point of a chromosomal aberra tion and the place where a deficiency has occurred indicates that these two processes may be induced by the same mechanism. Chemical changes producing small deficiencies are responsible for breaks in chromosomes. Free ends produced by such breaks have a tendency to join again. If the frequency of breaks in a nucleus is low, as in case of spontaneous changes, chromosomal rearrangements have little chance to occur, but if it is high, as in case of high dosage X-ray treatment, the opportunity for the origin of chromosomal rearrangements (inversions and translocations) is also high. Literature Demerec, M., A mutability stimulating factor in the Florida stock of Droso phila melanogaster. Genetics 22: 190. Demerec, M. and M. E. Hoover, Three related X-chromosome deficiencies in Lrosophila. Jour. Hered. 27: Oliver, C. P An analysis of the effect of varying the duration of X-ray treat ment upon the frequency of mutations. Ztschr. ind. Abst. u. Vererb. 61: Sakharov, V. V., On the specificity of the action of factors causing mutations Bul. Biol. et Med. Exp. (Russian) 1: cited