Genetic Fine Structure Analysis of The

On the basis of interallelic recombination frequencies measured in diallelic crosses of the 5 amylose-extender alleles in maize, ae, B I , B3, M2 and i1, it was possible to construct a unique linear genetic map ordering all 5 alleles within the locus. The reciprocal diallelic crosses each gave comparable frequency estimates. The relative order is ae, i l , B3, BY, A42 or the reverse. Even though F1 endosperms resulting from all possible diallelic crosses were phenotypically mutant, therefore non-complementing, no decision as to whether or not these alleles exhibit functional complementation should be made without biochemical characterization of the starches from the various heteroallelic genotypes. HE occurrence of intragenic recombination in at least two higher organisms Thas recently been confirmed. NELSON (1957) devised an efficient scoring system in maize whereby waxy (wx) mutant and revertant or recombinant normal ( W z ) genotypes could be distinguished in preparations of pollen grains on the basis of differential iodine staining properties. Following the measurements of Wx frequencies in numerous wx mutant strains as well as crosses among them, 24 non-complementing alleles could be ordered within the locus on the basis of frequencies of Wx gametes measured in both pollen (NELSON 1959, 1968) and conventional (NELSON 1962) analyses. Apparently the wx cistron is associated with a single enzyme, starch granule-bound glucosyl transferase (NELSON and RINES 1962; NELSON and TSAI 1964), but it is not clear whether wx is a structural or regulatory gene for this enzyme. In addition to other pollen analyses with wx alleles (BRIGGS and SMITH 1965; BRIGGS 1968; BIANCHI and CONTIN 1963) interallelic recombination has also been demonstrated at the glossy-l ( SALAMINI and BORGHI 1966), shrunken-2 (NELSON 1969), and sugary-1 (SALAMINI 1967) loci by conventional procedures. The best indication that intragenic recombination occurs in Drosophila was Authorized for publication on May 28, 1971 as paper Number 3988 in the Journal Series of The Pennsylvania Agricultural Experiment Station. Adapted from a thesis in Genetics in partial fulfillment of the requirements for degree of Doctor of Philosophy, The Pennsylvania State University. This investigation was supported in part by National Defense Education Act (Title IV) Grant Number 03-44-480720, US. Department of Agriculture Regional Researrh funds, and a grant from Corn Refmers Association. Present address: Department of Biology, University of Rochester, Rochester, N.Y. 14620. Genetics 70: 611-619 April 1972. 612 C. W. MOORE A N D R. G. CREECH obtained utilizing ingenious selective procedures by CHOVNICK and his collaborators (1962, 1964, 1970) who demonstrated recombination among several rosy (ry) mutants. No allelic complementation has been detected among these mutants, which completely lack xanthine dehydrogenase activity ( SCHALET, KERNAGHAN and CHOVNICK 1964). GRELL (1962) and YEN and GLASSMAN (1965) presented evidence that r y is the structural gene for this enzyme. Interallelic recombination has also been demonstrated for Notch ( WELSHONS 1965), rudimentary (GREEN 1963) and maroon-like ( CHOVNICK et al. 1969; FINNERTY, DUCK and CHOVNICK 1970), but each has been interpreted as being a single locus exhibiting some allelic complementation. Garnet mutants (HEXTER 1958; CHOVNICK 1961), as well as miniature and dusky (DORN and BURDICK 1962), and white mutants (GREEN 1959; JUDD 1959; STERN 1969) also seem to map within a single functional unit, although these loci generally have been considered complex. Recently, GREEN (1969) mapped a controlling element which he concluded is integrated at the white locus and reduces interallelic crossing over among white mutants. The genetic fine structure analysis of a second locus in maize which utilizes the pollen-scoring system. amylose-extender (ae) , is presented in this paper. It was made possible by the fact that while normal ( A e ) and mutant (ae) pollen starches both contain amylose and are indistinguishable when iodine-stained, red-staining Ae wx and dark-staining ae wx may be differentiated due to the presence of amylose in pollen starch of only the latter genotype. The ae gene alone is associated with the production of high amounts of amylose without greatly reducing total amounts of starch ( FERGASON, HELM and ZUBER 1966). On the basis of interallelic recombination frequencies measured in diallelic crosses of 5 ae mutants, a linear genetic map of the locus is proposed. MATERIALS AND METHODS Mutant alleles: The standard allele is ae, the one in general use by maize geneticists, and was obtained from Dr. H. H. KRAMER, Purdue University. The other 4 alleles have been provided with tentative symbols for experimental purposes and ease of presentation. Mr. R. P. BEAR of the Bear Hybrid Corn Company, Decatur, Illinois, provided a e B 1 (Bear-I) and a e B 3 (Bear-3). Dr. M. Z. ZUBER, University of Missouri, provided a&* (Missouri-2). The aeil mutation arose in a stock grown at Purdue University known to be carrying the activator-dissociator (Ac-Ds) system of genecontrolling elements described by MCCLINTWK (1965a, 1965b) ; thus, the designation induced-I (2 ) was assigned. Hereafter, these alleles will be denoted by their superscripts B I , B3, M2 and ii, respectively, and only the standard will be specificied ae. The waxy allele used in this study was the standard wx. Inbreeding: The ae and wx mutants were incorporated into the inbred line L317 in a backcross program at The Pennsylvania State University. All ae alleles were not backcrossed the same number of times to L317 (ae BC-3; BI BC-2; B3 BC-2; M2 B C I ; and il BC-1) since they were made aavilable at different times. It was desired that each stock be as nearly isogenic as possible to minimize background effects in the analysis; therefore, these 5 lines possessing the greatest number of backcrosses among the several available ae lines were selected. Pollen analyses; doubly mutant homoand heteroallelic plants were greenhouse grown during the spring of 1969. Their central spikes were removed early on days they expected to shed pollen. Tassels were rinsed with tap water to remove foreign pollen and stored in 75% ethyl alcohol for at least one month. This “curing” period was essential for pollen to retain the iodine stain. Amylose-Extender IN MAIZE 613 The pollen staining procedure was modified from the one CREECH and KRAMER described (1961). The staining solution was prepared by dissolving 3 0 g granular KI in 4 0 ml 25% dimethyl sulfoxide (DMSO); 6.5 g I, were then added before diluting the mixture with 60 ml 25% DMSO. Pollen was strained in the following manner: Twenty-seven to thirty-six anthers were removed from proximal florets and placed in a micro homogenizing flask with 1 ml of the iodine solution. Anthers were allowed to stand 2 min, homogenized for 30 sec with a VirTis “23” homogenizer, then permitted to stand 4.5 min. The suspension was filtered through Curity Grade 90 cheesecloth onto filter paper in a suction funnel. The flask and anthers were rinsed with 8 ml distilled H,O. After removing the staining solution by suction, 2 ml of 25% ethanol were added to the grains and allowed to remain for 30 sec. This procedure, unlike NELSON’S (1968), overstains all pollen in order that the 2 starch types of ae wx containing only about 10% amylose and Ae wx ki th no amylose can be reliably distinguished. In prellminary staining tests, it had been clear that some ae wx pollen failed to stain sufficiently to be unequivocally differentiated from Ae wx; thus the need for designing a method of overstaining, with controlled destaining during a subsequent step. The stained grains were washed from the filter paper into a watch glass with distilled H,O. Following removal of the H,O by vacuum, 0.7 ml of a gelatin medium, consisting of 0.6 g gelatin, 30 ml H,O and 3 drops Tween 80, was added to the pollen. The gel suspension was poured onto an 80 x 100 mm glass slide, evenly spread, and covered with a 4 0 x 60 mm cover glass. Once the mixture gelled, edges of the slip were sealed with colorless fingernail polish and preparations were stored at 8°C. The slides were examined after 18 hr with a Leitz stereo-scopic microscope with incident and transmitted light. The position of each Ae wx red-staining pollen grain, if any at this time, was marked on the cover slip with a red nylon tip marker. Each slide was then heated at 54°C for 2 min and scored again. This step was repeated 3 times if necessary, or until red grains were unequivocally differentiated. Unfortunately, since pollen is killed during this procedure, one cannot verify presumed Ae wx types as with the kernel phenotypic procedures. The total population of pollen grains on each slide was estimated from actual counts of 12 randomly preselected areas on the slide. Each sample area of 28.26 mm2 was photographed using a Leica 35 mm camera fitted on the microscope with the MIKAS microattachment plus an 1/3X intermediate adapter. The developed film was projected onto a grid screen and the grains were counted. The total number of pollen grains in the 12 sampling areas was multiplied by the constant 8.488, the number of equivalent areas on a slide, to obtain an estimate of the total number of grains per slide. RESULTS AND DISCUSSION The frequencies of red-stained pollen grains from plants produced following diallelic crossing of strains having the 5 ae alleles are listed in Table 1. Comparable recombinational estimates were obtained in each of 2 reciprocal crosses analyzed, BI x B3 and B3 x BI as well as BI x M 2 and M 2 x B I ; in the latter, the M 2 male and female parent, respectively, was the same plant. A third reciprocal cross, ae x M 2 and M 2 x ae, was sampled later in the year and also resulted in very similar estimates (Table 2) . When the 5 ae alleles were crossed in all possible combinations the phenotypes of F, kernels of diallelic constitutions could not be distinguished from those of homoallelic seeds. All possessed the translucent, semi-collapsed property characteristic of the ae ae ae wx wx wx genotype. From this standpoint, the 5 alleles did not exhibit complementation. The incidences of presumed Ae wx types in homoallelic lines are shown in Table 3. They are higher in every set of heteroallelic crosses, ranging from 12.2 to 87.4 x It than in homoallelic lines, which range from 2.3 to 3.3 x 614 C. W. MOORE AND R. G. C R E E C H