DEPARTMENT OF GENETICS 181 MAIZE GENETICS Barsara McCLintTock Tue Benavior of “UNsaturatep” Broken Enps oF CHROMOSOMES In all cases involving rearrangements of segments of chromosomes which give rise to translocations, inversions, deficiencies, etc, it has been necessary to postulate some force that breaks a chromosome and some force that results in the permanent 2-by-2 fusion of the broken ends. Previous investigations in maize on the mitotic be- havior of ring-shaped chromosomes had suggested that fusions may occur between two recently broken ends of chromosomes which enter the same nucleus. Such broken ends may be considered “unsatu- rated,” i.e., capable of fusion with similar “unsaturated” broken ends, until fusion with another broken end occurs or until the end loses its capacity for fusion. To determine whether such an “unsaturated” state exists, male gametes containing a chromosome g whose short arm had been broken by mechanical pull at the previous anaphase were united with female gametes containing a similar recently broken chro- mosome g. The zygote formed received from each gamete nucleus a single chro- mosome with a single recently broken end. On the basis of published data, these two recently broken ends, derived from sepa- rate nuclei, are believed to be in the “un- saturated” state and therefore capable of fusion with each other. If some force exists that brings these unsaturated ends to- gether and results in fusion, a dicentric chromosome should be produced com- posed of the chromosome 9g contributed by the female gamete and the chromosome ' g contributed by the male gamete, fused at the ends of their short arms. Through the use of the endosperm markers J and C and through the aberrant mitotic behavior that reflects the presence of such broken chromosomes in the endosperm, it was possible to select the kernels from an ear whose zygote nucleus had received a chro- mosome with an unsaturated broken end from the male and female gamete nuclei, respectively. Out of a total of 18,243 kernels examined, 20 non-germless kernels were obviously of the type desired. These kernels were ger- minated. If fusion had occurred between the broken ends of the chromosomes g contributed by the two gametes, follow- ing chromosome reduplication, the dicen- tric chromosome should produce a double anaphase bridge configuration when the two centromeres of each chromatid passed to opposite poles. Breakage of the two ‘bridges would result in the entrance into each nucleus of two newly derived, unsatu- rated broken ends. Fusion of unsaturated broken ends could occur in each sister telophase nucleus. Again, the two chro- mosomes 9 would be joined to form one chromosome with two centromeres. Re- peated anaphase bridge configurations should be expected to follow from such a chromosomal type of breakage-fusion- bridge cycle. Plants having such a dicen- tric chromosome and undergoing this cycle should have cells with various types of heterozygous and homozygous duplica- tions and deficiencies of the short arm of chromosome g following nonmedian break- ages of the anaphase bridges. Because of this process, the plants should be conspicu- ously modified in appearance. The plants arising from 10 of the 20 kernels were obviously of the type expected if a di- centric chromosome g were present. Ex- amination of the early roots confirmed the presence of a dicentric chromosome. Some- 182 what less than one-half of the anaphase fig- ures showed contiguous double bridges. Owing to death or defective growth of many cells or sectors of tissue, 5 of these plants died in the seedling stage. Four of the remaining 5 plants continued to grow, because sectors of normal-appear- ing tissues developed. Gradually these sectors gained the ascendancy in growth, until the plant appeared quite normal. The fifth plant produced 3 normal shoots, which arose from the base of the de- cidedly aberrant and dying main shoot. Microsporocytes were obtained from the 4 recovered plants and from 2 of the 3 recovered shoots of the fifth plant. In all cases, pachytene analysis showed a bivalent chromosome 9. The two chromo- somes were not fused at the ends of their short arms. The two broken ends had healed in the ancestor cell which gave rise to the recovered sector. In most cases, the composition of the short arm of each mem- ber of the bivalent was greatly modified, although within a tassel sample all ex- amined sporocytes showed the same com- position for the individual member of the bivalent. In several of these plants, it was possible to determine the minimum num- ber of fusions, breakages, and bridges which must have occurred before healing of the two broken ends within a single nucleus had occurred. It is likewise known that the compositions of the short arms were entirely different in the sporocytes of the tassels of the 3 recovered shoots of the one original dicentric plant. The two chromosomes 9, however, had maintained their respective derived compositions with- in each shoot. This indicates that the mi- crosporocyte tissues of each shoot had originated from one individual cell whose cell ancestors had previously been under- going the chromosomal type of breakage- fusion-bridge cycle involving the original dicentric chromosome g. The root system CARNEGIE INSILLULION OF WASHILNGLUIN responded similarly. In the older roots of the surviving plants, no dicentric anaphase bridge configurations were observed. These experiments definitely show the existence of an “unsaturated” state of a recently broken end of a chromosome. Owing to causes as yet undetermined, however, such an end may become satu- rated (healed) without fusion. Following this, the end no longer takes part in any fusions. The remaining 10 of the original 20 kernels classified as having received a broken chromosome 9 from each parent gave rise to 9 normal-appearing plants and 1 pale-yellow plant which died in the seedling stage. None of these plants showed dicentric bridge configurations in the young roots. Examination of the sporo- cytes of the g surviving plants showed that 4 had received a broken chromosome 9 from each parent; but the morphology of the short arms gave no indication that fusions had occurred between these broken ends. In 1 plant one parent had con- tributed a broken chromosome 9, but it could not be determined whether the other parent had likewise contributed a broken chromosome 9. In the remaining 4 plants, each parent had contributed a broken chromosome 9, but one broken end had become saturated by fusion with a broken end other than that of the chromosome 9 contributed by the second gamete and pos- sibly before fusion of the gametes them- selves. Consequently, healing of the broken end of the second chromosome 9 had occurred. These results indicate that an unsaturated broken end produced by mechanical breakage of an anaphase bridge is capable of fusing with another unsaturated broken end arising from un- determined causes. A similar type of fusion has likewise been observed in sporocytes of 5 plants which were known to have been derived DEPARTMENT OF GENETICS from a gametophyte which had received a chromosome g with an unsaturated broken end. It is known that mechanical pull caused by an anaphase bridge will fre- quently break a chromosome at a knob or at the centromere. In 2 of the 5 cases, the centromere of the broken chromosome g was fused with the centromere of an- other chromosome of the complement. In one case, the fused chromosome was com- posed of the long arm of chromosome 9 and the short arm of chromosome 2. In the second case, it was composed of the long arm of chromosome g and the short arm of chromosome 10. In each case, the com- plementary arm was missing. In three cases, the fusions had occurred at other positions than centromeres. In one case, a segment from the long arm of chromo- some 4 had united with the broken end of the short arm of chromosome g. Since both chromosomes 4 in this plant were com- pletely normal, it is assumed that chroma- tid fusion in a gametophytic nucleus had occurred between the unsaturated broken end of chromosome 9 and a naturally aris- ing broken end terminating an acentric distal segment of chromosome 4. In the other two cases, both segments of the sec- ond broken chromosome were present. Pachytene analysis has led to the follow- ing interpretation: In the last two cases mentioned, a break occurred at one posi- tion in chromosomes 1 and 8, respectively. In both cases, this resulted in the presence of three unsaturated broken ends in the same nucleus, one of which was the broken end of the short arm of chromo- some g. Fusion occurred between the un- saturated broken end of chromosome g and the unsaturated broken end of the acentric segment of the second broken chromo- some. This left the centric segment of the second broken chromosome with a single unsaturated broken end, which thereafter healed. This healing of a single un- 133 saturated broken end, when introduced into sporophytic tissues, is in agreement with the results of similar investigations of this behavior. Puenorypic Errecrs of Homozycous Dr- FICIENCIES OF DisTAL SEGMENTS OF THE SHort ARM OF CHROMOSOME 9 The phenotypic effects in male gameto- phytes, and in endosperm and sporophytic tissues, of a series of homozygous defi- ciencies involving distal segments of the short arm of chromosome 9 are being in- vestigated. These deficiencies were ob- tained through meiotic breakage of a di- centric chromatid g which had been pro- duced following crossing over involving a duplicated segment of the short arm of chromosome 9. This method has been previously described (McClintock, 1941; see bibliography). A number of terminal deficiencies have been isolated, ranging in length from a fraction of the terminal chromomere to deficiencies of approxi- mately one-third of the short arm, in- cluding the locus of C. At pachytene, the short arm of chromosome 9 has approxi- mately 20 chromomeres. Those in the proximal third of the arm are large, those in the distal two-thirds of the arm are small. The effect of homozygous deficiencies on the functioning of female gametophytes. Plants heterozygous for these deficiencies produce female gametophytes which are totally deficient for the respective seg- ments of chromosome 9. Complete func- tioning of such gametophytes occurs in all cases of short deficiencies. Only in the case of longer deficiencies which include 4 or more chromomeres is there a reduc- tion in the functioning of such gameto- phytes. Environmental factors may be in- volved in this differential functioning. A preliminary test has indicated that, on a single ear, functioning of deficient female 184 gametophytes may be complete on one day and they may be totally nonfunctional on the succeeding day. Extensive tests are necessary to associate the effect with a par- ticular environmental condition. The effect of homozygous deficiencies on the appearance and functioning of male gametophytes. Plants heterozygous for these terminal deficiencies produce pollen grains one-half of which carry the deficient chromosome g. In all cases, homozygous deficient pollen grains are completely filled with starch. Only in the case of deficiencies that include the distal one-third of the short arm is it possible to distinguish any perceptible differences in the appearance of the normal and the homozygous de- ficient grains. The latter grains appear to be smaller, but an exact identification of each grain has not been possible. Only in the case of distal deficiencies that are greater than one-third of the short arm is there a classifiable visible effect on pollen development. Some starch develops even in pollen grains that are deficient for nearly all of the short arm of chromo- some 9. Pollen grains that are deficient for small terminal segments are completely func- tional. Those deficient for more than the terminal chromomere, although completely normal in appearance, are nonfunctional. The phenotypic effects of small terminal deficiencies on endosperm and sporophytic tissues: the deficiency mutants “pale-yel- low” and “white” and thetr dominance re- lationships. Plants that are heterozygous for small terminal deficiencies produce via- ble and functional male and female ga- metophytes. These plants were selfed to determine whether viable endosperms and embryos that were homozygous for these deficiencies could be obtained. In 5 of the 7 cases studied, the endosperm and embryo of kernels having the homozygous de- ficiencies were completely normal in ap- CARNEGIE INSTITUTION OF WASHINGTON pearance. In 2 cases, some but not all of the embryos that were homozygous de- ficient had died before the maturity of the kernel. The endosperm of these kernels, however, was completely normal. In all 5-cases with normal embryo development, pale-yellow seedlings, completely normal in gross morphology and growth rate, grew from these kernels. Although the coleop- tiles were light green, little chlorophyll developed in the leaves, and the seedlings died after exhaustion of the food reserves in the endosperm. The surviving embryos in the 2 cases where the homozygous de- ficiency resulted in early death of some embryos produced white seedlings com- pletely devoid of plastid pigments. Al- though the gross morphology of these seedlings was normal, the growth rate was considerably retarded. Proof of the asso- ciation of the pale-yellow and white seed- lings with the homozygous deficient state was obtained through cytological examina- tion of normal sibs, which had only homo- zygous normal and heterozygous deficient chromosomes; through crosses of these latter plants to plants heterozygous for longer deficiencies, where the mutant types appeared only from unions of the two re- spective deficient chromosomes; through close if not complete linkage with the mu- tant yg located near the end of the short arm of chromosome g; and through chro- mosomal examination within white sectors of sectorial plants. Intercrosses among all 7 cases have shown that the 5 pale-yellow mutants are allelic and that the 2 white-seedling mu- tants are allelic to pale-yellow, with pale- yellow dominant to white. The 5 defi- ciencies giving rise to pale-yellow do not include the yg locus, whereas the 2 def- ciencies giving rise to white seedlings may include this locus. The deficiencies giving rise to white seedlings are longer than those giving rise to pale-yellow seedlings. DEPAKIMENL UF GhIND Liwo although they have a deficient segment in common. This accounts for the allelic na- ture of the two mutants and the dominance of pale-yellow over white. The pale-yellow and white mutants represent typical Mendelizing mutants, which are associated with a state of homozygous deficiency. Dominance in these cases is an expression of the extent of the deficiency: no def- ciency produces green seedlings, a short terminal deficiency produces pale-yellow seedlings, and a longer terminal deficiency produces white seedlings, with dominance expressed in this order. The phenotypic effects of relatively long terminal homozygous deficiencies. Termi- nal deficiencies that include more than one chromomere do not give rise to functional pollen. Thus, the phenotypic effects of these deficiencies could not be studied by the direct method of selfing heterozygous plants. Instead, the variegation method, which produces sectors of tissue that are homozygous deficient, was introduced in these cases. This method utilizes the aber- rant mitotic behavior of recently broken chromosomes, which, in the endosperm, continuously deletes segments from the arm of the chromosome which has the broken end. If the female gametophyte contributed 2 deficient chromosomes, and the male gametophyte contributed a chro- mosome 9 whose short arm terminated in a recently broken end, the developing en- dosperm could be sectorial for homozygous deficient tissues. The endosperm mutants C (aleurone color), J (inhibitor of aleu- rone color, allelic and dominant to C), Sh (sh, shrunken endosperm), and Wx (wx, waxy starch) were used to mark the chromosomes contributed by the two par- ents. The preliminary investigations ‘on the effects of homozygous deficiencies on endosperm development may be summa- rized as follows: Endosperm development may be completely normal when homozy- 1v> gous deficiencies up to and including two terminal chromomeres are present. Be- yond this region, only patches of such homozygous deficient tissue, surrounded by normal tissues, will develop. As the homozygous deficiency becomes progres- sively longer, the rate of development within the sector is reduced. Although the C locus may still be present, aleurone- color development progressively dimin- ishes until only the rim of cells bordering normal cells shows color. Apparently, some substance or substances diffuse from the normal cells into these homozygous deficient cells, allowing them to develop normal aleurone color. This material, how- ever, either does not diffuse beyond a layer several cells deep or is used up before deeper penetration occurs. Starch develop- ment occurs in all the patches of homo- zygous deficient cells except when the de- ficiency approaches the distal third of the short arm and includes the locus of C. In the latter case, relatively extensive growth of the homozygous deficient cells occurs; but, owing to lack of starch forma- tion in these cells, a shrinkage leading to scar formation occurs after drying of the kernels. To study the effects of various homozy- gous deficiencies on sporophytic tissues, the method of covering a deficiency with a ring-shaped chromosome may be utilized. Frequent losses of the ring-shaped. chro- mosome during mitoses should produce cells that are homozygous deficient. Cells arising from these cells should produce sectors capable of expressing changes that could be related to the homozygous de- ficient state. Likewise, changes in consti- tution of ring chromosomes, which may delete segments from the ring, could pro- duce sectors that are homozygous deficient for various segments within the limits of the full deficiency. Only two such plants 186 have been produced. Both plants were characterized by numerous sectors of white, pale-yellow, and yellow-green tis- sues. Although these sectors probably rep- resent the expression of homozygous de- ficiencies, no conclusions will be drawn until this method receives more detailed and controlled analysis. A deficiency of one-third of the short arm of chromosome 9 is, relatively long, / CARNEGIE INSTITUTION OF WASHINGTON but none of these deficiencies have been cell lethal in any of the tissues studied. It is altogether possible that the observed effects of the homozygous deficiencies in the various tissues may be related to a few specific loci within the limits of the distal third of the short arm, rather than to the accumulative effect of a large number of such loci. This would be understandable if maize were a derived polyploid.