Position Effects of Ribosomal RNA Multigene Loci on Meiotic Recombination in Wheat

, and Jan Dvoák

Corresponding author: Jan Dvoák, Department of Agronomy and Range Science, University of California, Davis, CA 95616, jdvorak{at}ucdavis.edu (E-mail).

Communicating editor: J. A. BIRCHLER

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Homeologous wheat chromosome arms that differ by the presence or absence of a Nor locus or greatly differ in the numbers of copies of rRNA genes per Nor locus show conspicuous differences in the distribution of recombination. To assess directly the position effects of Nor loci on recombination across chromosome arms, a Triticum monococcum Nor9 haplotype was substituted for Triticum aestivum Nor9 haplotypes on two T. aestivum 1A chromosomes in the isogenic background of cv Chinese Spring. The numbers of rRNA genes in the 1A Nor9 haplotypes are greatly reduced relative to the T. monococcum haplotype. The substitution resulted in reduced recombination rate in the vicinity of the Nor9 locus. An intra-arm compensatory increase was observed in the proximal region of the arm so that the genetic length of the chromosome arm was unchanged. These findings suggest that Nor loci suppress recombination in their vicinity and change recombination patterns in Nor-bearing chromosome arms.

NUCLEOLUS organizing regions (NORs) are the sites of active 18S-5.8S-26S rRNA genes. In bread wheat, Triticum aestivum (2n = 6x = 42, genomes AABBDD), NORs are on the short arm of chromosome 1A in the A genome (Nor9), the short arm of chromosomes 1B (Nor1) and 6B (Nor2) in the B genome, and the short arm of chromosome 5D (Nor3) in the D genome (CROSBY 1957 ; LONGWELL and SVIHLA 1960 ; FLAVELL and SMITH 1974 ; FLAVELL and O'DELL 1976 ; APPELS et al. 1980 ; MILLER et al. 1980 ; MUKAI et al. 1991 ; DUBCOVSKY and DVORAK 1995 ). A majority of the wheat rRNA gene units are at the Nor1 and Nor2 loci and a minority are at the Nor3 and Nor9 loci (Table 1). Other minor chromosomal sites hybridizing with rRNA gene probes exist in T. aestivum (MUKAI et al. 1991 ; JIANG and GILL 1994 ), but there is no evidence that they contain active rRNA genes and are involved in the organization of nucleoli, and they will not be considered here.

Of the four wheat Nor loci, the lowest numbers of rRNA gene units are at the Nor9 locus. In situ hybridization of a rRNA gene repeated unit with T. aestivum metaphase chromosomes (MUKAI et al. 1991 ) showed that the Nor9 locus of Chinese Spring wheat has between a quarter to a third of the repeated-gene units of the Nor3 locus. Since the Nor3 locus was estimated to have 400 repeated gene units per 2C nucleus in Chinese Spring (LASSNER et al. 1987 ), the Chinese Spring Nor9 locus has approximately 100 to 140 repeated gene units per 2C nucleus. In cultivar Cheyenne, which, like Chinese Spring, has been extensively used in cytogenetic studies, there are similarly low numbers of rRNA gene units at the Nor9 locus. MILLER et al. 1980 reported two pairs of major in situ hybridization sites in Chinese Spring and two pairs of major sites and one pair of minor sites in Cheyenne. Cheyenne DNA fragments hybridizing with pTa250.15, a wheat nontranscribed rRNA gene-spacer clone (APPELS and DVORAK 1982 ), were mapped to the Nor1, Nor2, and Nor3 loci (DVORAK and CHEN 1984 ; DVORAK and APPELS 1986 ; LASSNER et al. 1987 ; DUBCOVSKY and DVORAK 1995 ). Therefore, the minor hybridization site observed by MILLER et al. 1980 in Cheyenne was the Nor3 locus. Although the precise mechanism that caused the great reduction in the number of rRNA genes at the bread wheat Nor9 locus is not known with certainty, unequal sister chromatid exchanges (PETES 1980 ) and intralocus deletions (DVORAK 1989 ) are potential causes.

Meiotic homologous exchanges tend to be underrepresented within rRNA loci in organisms as diverse as yeast, wheat, Drosophila, and maize (PETES 1980 ; DVORAK and APPELS 1986 ; WILLIAMS and ROBBINS 1992 ; SIMCOX et al. 1995 ). In yeast, homologous meiotic exchanges are suppressed within the rRNA locus and are replaced by frequent sister chromatid exchanges (PETES 1980 ). A hypothesis that a suppression of homologous recombination occurs also in the chromosomal neighborhoods of major RNA loci and, hence, that Nor loci exert position effects on recombination is tested here. To test this hypothesis, nearly isogenic chromosome pairs were constructed in the genetic background of Chinese Spring that differed by having either a minor Nor9 haplotype of T. aestivum or major Nor9 haplotype of T. monococcum L. (2n = 2x = 14, genomes A^mA^m). Although the genome of T. monococcum is closely related to the A genome of T. aestivum, T. monococcum chromosomes recombine poorly with the T. aestivum A genome chromosomes if the wheat suppressor of homeologous chromosome pairing Ph1 is active (PAULL et al. 1994 ; DUBCOVSKY et al. 1995 ). Meiotic pairing and recombination between these chromosomes are restored if the Ph1 locus is absent. Under those circumstances, 1A/1A^m recombinant chromosomes can readily be constructed (DUBCOVSKY et al. 1995 ). This ability to turn on and off recombination between the short arm of T. monococcum chromosome 1A^m and wheat chromosome 1A was exploited in the development of nearly isogenic chromosome pairs differing by Nor9 haplotypes.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Genetic stocks:
Recombinant substitution lines (RSLs) are wheat genetic stocks in which a single chromosome pair is replaced by a recombined homologous (or homeologous) chromosome pair. Hence, RSLs are expected to be isogenic with the recipient genotype except for a substituted chromosome segment. They can be produced by crossing a disomic substitution line with the recipient, backcrossing the F₁ to a respective monosomic or monotelosomic of the recipient genotype, and selecting monosomic progeny. In an RSL designated 1A^rec (Figure 1), a recombined chromosome 1A_cs/1A^m, composed of the entire short arm and most of the long arm of chromosome 1A^m of T. monococcum (accession G1777) and a distal part of the long arm of chromosome 1A of Chinese Spring, was substituted for 1A of Chinese Spring (DUBCOVSKY et al. 1995 ). From that RSL, a population of RSLs, harboring different 1A_cs/1A^m recombined chromosomes, was developed, and exchange points were mapped with molecular markers (DUBCOVSKY et al. 1995 ). RSL no. 21 was selected in that population for the present study (Figure 2). RSL21 had a 1A_cs/1A^m recombined chromosome composed of Chinese Spring 1A with a short-arm terminal 1A^m segment including the Nor9 locus (Figure 1). The length of the terminal segment was estimated to be less than 0.7 cM and 2.6 cM from maps reported by DUBCOVSKY et al. 1996 , and DUBCOVSKY et al. 1995 , respectively, and the exchange point was between the Gli1 locus and Nor9 (Figure 1) on the short arm map.

View larger version (22K): In this window In a new window Download PPT slide   Figure 1. —The structure of chromosomes used in the study. The line in which a chromosome is present is indicated in parentheses. A slash in a chromosome description indicates that the chromosome is recombined. Subscript cs stands for Chinese Spring genetic material and subscript cnn stands for Cheyenne genetic material. Superscript m indicates T. monococcum genetic material. Note that, for better illustration, the intervals are not proportional.

View larger version (22K): In this window In a new window Download PPT slide   Figure 2. —A crossing scheme by which F2 mapping populations homozygous for the 1Am Nor9 haplotype were developed. A slash in a chromosome description indicates that the chromosome is recombined. Subscript cs stands for Chinese Spring genetic material and subscript cnn stands for Cheyenne genetic material. Superscript m indicates T. monococcum genetic material. The sizes of the F2 mapping populations are specified in parentheses.

To study the effects of the 1A^m Nor9 on recombination in the adjacent 1A chromosome region, it was necessary to introduce polymorphism into the 1A segment. We elected to replace Chinese Spring genetic material in the 1A_cs/1A^m chromosome of RSL21 with the genetic material of Cheyenne 1A (Figure 2) because restriction fragment length polymorphisms (RFLPs) between Chinese Spring 1A and Cheyenne 1A had been identified at a number of loci (DUBCOVSKY et al. 1995 ). To accomplish that, RSL21 was crossed with a disomic substitution line in which Chinese Spring chromosome 1A was replaced by Cheyenne chromosome 1A (MORRIS et al. 1966 ; Figure 2). This substitution line will be henceforth designated DSCnn1A. F₁ progeny were crossed with a monotelosomic for the long arm of chromosome 1A (henceforth 1AL) isolated by us from ditelosomic 1AL, and 269 monosomic RSLs were selected; monotelodisomics were discarded (Figure 2). Crossovers in these monosomes were mapped with RFLP markers to identify chromosomes with distal crossovers in the short arm that replaced most of the 1A Chinese Spring genetic material with that of Cheyenne 1A. Three such RSLs were identified: RSL21-12, RSL21-103, and RSL21-139 (Figure 1). The exchange points between Chinese Spring and Cheyenne genetic material were within the XGlu3-Xmwg60 interval in RSL21-12 and RSL21-103 and within the XGli1-XGlu3 interval in RSL21-139 (Figure 3). In RSL21-103, the rest of the chromosome was Cheyenne. In RSL21-12 and RSL21-139, the Cheyenne segment was once more recombined with Chinese Spring in the XGlu1-Xbcd808 interval and the XTri-Glu1 interval, respectively (Figure 1). Because recombination between 1A^m and 1A is virtually eliminated from the short arm in the presence of the Ph1 locus (DUBCOVSKY et al. 1995 ), all four 1A/1A^m chromosomes were expected to have the same terminal segment of 1A^m including Nor9. That this was indeed the case was substantiated by hybridization of a 900-bp rRNA promoter fragment cloned in pTa250.15 (APPELS and DVORAK 1982 ) with SstI-digested DNAs of all RSLs (for an example see Figure 4).

View larger version (45K): In this window In a new window Download PPT slide   Figure 3. —Hybridization of gliadin clone pcP387 (FORDE et al. 1985 ), low-molecular-weight glutenin clone pTdUCD1 (CASSIDY and DVORAK 1991 ), and a random barley genomic clone MWG60 (GRANER et al. 1991 ) with Southern blots of DNAs of Chinese Spring (CS), the disomic substitution line in which Chinese Spring chromosome 1A is replaced by Cheyenne chromosome 1A (DSCnn1A), recombinant substitution lines 1Arec (RSL1Arec), recombinant substitution lines 21(RSL-21), ditelosomic 1AL (DT1AL), and RSL21-12, RSL21-103, and RSL21-139, illustrating the positions of crossovers between the 1Acs/1Am chromosome and Cheyenne chromosome 1A. DNA fragments unique to Chinese Spring chromosome 1A, Cheyenne chromosome 1A, and T. monococcum chromosome 1Am are indicated. Note that DNAs of all four RSLs have the Chinese Spring XGli1. DNA of RSL21-139 has the Cheyenne XGlu3 and Cheyenne Xmwg60 DNA fragments showing that the crossover between 1Acs/1Am and Cheyenne 1A is between XGli1 and XGlu3 in this chromosome. RSL21-12 and RSL21-103 have the Chinese Spring XGlu3 haplotype (they are missing the Cheyenne XGlu3 DNA fragment). They also have the Cheyenne Xmwg60 DNA fragment showing that the crossover is between XGlu3 and Xmwg60 in these two RSLs.

View larger version (58K): In this window In a new window Download PPT slide   Figure 4. —An autoradiogram of a Southern blot of genomic DNAs of Chinese Spring, RSL21, disomic substitution line Cheyenne 1A in Chinese Spring, and an example of F2 plants from mapping population RSL21 x RSL21-139 digested with SstI and hybridized with pTa250.15. The 1Am rDNA fragment is indicated. Note the 1Am band present in all F2 plants.

To investigate recombination in the vicinity of the Nor9 locus, RSL21-12, RSL21-103, and RSL21-139 were crossed with RSL21, and F₂ populations were produced by self-pollination (Figure 2). The numbers of plants in each F₂ population are indicated in Figure 2. Recombination in these populations was compared with that in a population of 162 F₂ plants from the cross Chinese Spring x DSCnn1A.

Recombination study:
Nuclear DNAs were isolated from the F₂ plants of the three RSL21 mapping populations (Figure 2) and the Chinese Spring x DSCnn1A population. DNAs of 91 F₂ plants from the cross Cheyenne-1 x Chinese Spring which were previously used for the construction of a map of chromosome 1B (DUBCOVSKY and DVORAK 1995 ) were used to make a map of wheat chromosome 1A based on an F₂ population. Restriction endonuclease-digested DNAs were electrophoretically fractionated in 1% agarose gels and transferred to Hybond N+ nylon membranes (Amersham, Arlington Heights, IL) by capillary transfer in 0.4 N NaOH overnight. The membranes were then rinsed in 2x SSC for 5 min and immediately prehybridized or stored wet. DNA inserts were isolated from plasmids by PCR amplification using plasmid primers and purification of the products by the Wizard PCR Columns (Promega, Madison, WI). Probes were ³²P-labeled by the random hexamer primer method. Prehybridization and hybridization were performed in a rotary hybridization chamber (National Labnet Company, Woodbridge, NJ) at 68° using hybridization solution described earlier (DUBCOVSKY et al. 1994 ). The membranes were washed in 2x SSC and 0.5% SDS from 30 min to 2 hr at 60°, 1x SSC and 0.5% SDS for 30 min at 65°, and 0.5x SSC and 0.5% SDS for 12 min at 65°. Maps were constructed with the computer program Mapmaker/EXP 3.0 (LANDER et al. 1987 ; LINCOLN et al. 1992 ) using KOSAMBI 1943 function. For statistical comparisons, recombination fractions from populations RSL21 x RSL21-12, RSL21 x RSL21-103, and RSL21 x RSL21-139 were tested for homogeneity (z-test) and, if homogeneous, they were combined. Recombination fractions in intervals across the short arm were statistically compared with those on the Chinese Spring x DSCnn1A map. To make these statistical comparisons, the interval lengths were converted from centimorgans to percentage recombination using Kosambi function and the maximum likelihood estimates of variance were computed (ALLARD 1956 ). The significance of differences between recombination fractions were determined by z-tests.

A genetic map of the T. aestivum chromosome arm 1AS was produced using a population of 91 F₂ plants from the cross Cheyenne-1 x Chinese Spring. Genetic maps of the T. aestivum 1BS chromosome arm based on the Cheyenne-1 x Chinese Spring F₂ population and those of the T. monococcum 1A^mS chromosome arm have been reported earlier (DUBCOVSKY and DVORAK 1995 ). Probes and their sources used in the development of these maps have also been described (DUBCOVSKY and DVORAK 1995 ).

Gene copy number estimation:
The following procedure was used to determine the number of rRNA gene units constituting the Nor9 haplotype of the T. monococcum accession G1777. One µg of DNA of Chinese Spring, ditelosomic 1AL and RSL 1A^rec were immobilized on a Hybond N+ membrane using a procedure recommended by the membrane manufacturer. A BioRad dot-blot apparatus was used. Also immobilized on the membrane were equivalents of 1000, 2000, 5000, 10,000, and 15,000 copies per 1C nucleus of the T. aestivum rRNA gene unit excised from pTa71 (GERLACH and BEDBROOK 1979 ). Each dot was repeated eight times on the membrane. Another set of dilutions of the genomic DNAs and the pTa71 plasmid insert were independently prepared and immobilized on another membrane. The two membranes were hybridized with a 3.6-kb gel-purified ³²P-labeled BamHI DNA fragment of the rRNA gene unit inserted in pTa71. The fragment contained large portions of the 18S and 26S rRNA genes, the 5.8S gene, and the internal transcribed spacer (GERLACH and BEDBROOK 1979 ). The numbers of disintegrations per minute (dpms) per dot were determined in a Phosphor Imaging System (Storm 860, Molecular Dynamics, Sunnyvale, CA). The hybridized probe was dissociated from the blots and the blots were rehybridized with the insert of pAS1, which is a highly repeated sequence present on the chromosomes of the D genome of wheat (RAYBURN and GILL 1986 ). Since the D genome was isogenic among the three lines and not perturbed by recombination or aneuploidy, the level of hybridization of pAS1 could be used as a standard to correct variation in DNA amounts loaded per dot. The corrected average dpms per dot were used to determine the numbers of copies per nucleus using a standard line constructed from dpms of copy number equivalents. These inferred copy numbers per genotype were used in analysis of variance for completely randomized design and Tukey's test to test the statistical significance of differences among the genetic stocks in the rRNA gene copy numbers.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

rRNA gene copies:
RSL1A^rec was produced by backcrossing an amphiploid monotelosomic 1AL x T. monococcum to monotelosomic 1AL. The genetic backgrounds of RSL1A^rec and those of monotelosomic and ditelosomic 1AL are therefore identical. It was estimated that there are totals of 5000 ± 450 (standard deviation) repeated gene units per 2C nucleus in monotelosomic 1AL and 6800 ± 106 repeated gene units in RSL1A^rec. The difference (statistically highly significant) between the two stocks, 1800 gene copies per 2C nucleus, estimates the number of the rRNA gene copies at the Nor9 locus in T. monococcum accession G1777. Chinese Spring was found to have 8400 ± 487 gene units (significantly different from the other two stocks at the 5% probability level).

Recombination study:
In T. monococcum F₂ populations G1117 x G2528 and DV92 x G3116, a distal interval near the Nor9 locus, XGli1-Xmwg60, was 10.7 and 4.8 cM long, respectively. In contrast, the same interval was 28 cM on the linkage map of chromosome 1A based on the Cheyenne-1 x Chinese Spring F₂ population (Figure 5) and 24.1 cM on the linkage map based on the Chinese Spring x DSCnn1A F₂ population (Figure 6). On a map of the T. aestivum chromosome arm 1BS, which does not have the Nor9 locus, this interval was 15.9 cM (Figure 5). Intervals in the vicinity of the Nor1 locus on chromosome 1B were shorter than those on the T. monococcum chromosome 1A^m, which does not have Nor1 (Figure 5). Markers X5SDna and Xabg500 were completely linked to Nor1 on 1B but were 19.1 cM apart on 1A^m (Figure 5). In all these comparisons, intervals near Nor loci were shorter than the same intervals on homeologous chromosomes that are devoid of Nor loci or have greatly reduced numbers of rRNA gene units per Nor locus.

View larger version (27K): In this window In a new window Download PPT slide   Figure 5. —Genetic maps of the short arms of T. aestivum chromosomes 1A, 1B, and T. monococcum chromosome 1Am. The 1A map is based on a population of 91 F2 plants from the cross Cheyenne-1 x Chinese Spring. The 1B map (DUBCOVSKY and DVORAK 1995 ) is based on a population of 91 F2 plants from the cross Cheyenne-1 x Chinese Spring. The 1Am G1777/G2528 map (DUBCOVSKY and DVORAK 1995 ) is based on a population of 76 F2 plants from the cross T. monococcum G1777 x T. monococcum G2528. The G3116/DV92 map (DUBCOVSKY et al. 1996 ) is based on 74 F2 plants from the cross T. monococcum G3116 x T. monococcum DV92. The distances between markers are in centimorgans and the centromeres are indicated by arrows.

View larger version (23K): In this window In a new window Download PPT slide   Figure 6. —Genetic maps based on populations of F2 plants from the following crosses: Chinese Spring x DSCnn1A, RSL21 x RSL21-12, RSL21 x RSL21-103, and RSL21 x RSL21-139. The solid lines indicate the regions of the chromosome pairs that were heterozygous and the dashed lines indicate the regions that were homozygous Chinese Spring. The distances between markers are in centimorgans and the centromeres are indicated by arrows. The statistical significance of difference in the interval lengths is indicated by * and ** for the 5 and 1% probability levels, respectively.

If these reductions in recombination in the vicinity of the Nor loci are position effects of the Nor loci on recombination, recombination should be reduced in the wheat 1AS arm if the 1A Nor9 haplotype is replaced by a T. monococcum 1A^mNor9 haplotype, since the latter has an order of magnitude more gene units. This hypothesis was tested by comparing recombination in the F₂ populations from crosses RSL21 x RSL21-12, RSL21 x RSL21-103 and RSL21 x RSL21-139 with recombination in a F₂ population from a cross Chinese Spring x DSCnn1A. The most distal interval in which recombination could be compared was XGlu3-Xmwg60 (Figure 6). This interval was 19.4 cM on the Chinese Spring x DSCnn1A F₂ map but only 10.7 cM on the RSL21 x RSL21-139 F₂ map (P < 0.01). Recombination in the neighboring proximal interval Xmwg60-XksuE18 could be compared in all populations. In populations RSL21 x RSL21-12, RSL21 x RSL21-103, and RSL21 x RSL21-139, this interval was 11.1, 10.8, and 12.5 cM, respectively. The same interval was 7.1 cM long on the Chinese Spring x DSCnn1A map (Figure 6). Since recombination in the former three populations was homogeneous, the populations were combined to increase the sample size. The length of the Xmwg60-XksuE18 interval in the combined population was 11.6 cM, which was significantly longer (P < 0.05) than the 7.1 cM in the Chinese Spring x DSCnn1A population. The length of proximal interval XksuE18-XTri did not statistically differ from the length of this interval in the Chinese Spring x DSCnn1A population even when the RSL21 x RSL21-12, RSL21 x RSL21-103, and RSL21 x RSL21-139 populations were combined. Nor did the XksuE18-XTri interval on the map based on the RSL21 x RSL21-139 population significantly differ from that based on the combined population RSL21 x RSL21-12 and RSL21 x RSL21-103. The entire proximal interval Xmwg60-XTri was, however, significantly longer (P < 0.01) in the combined population RSL21 x RSL21-12, RSL21 x RSL21-103, and RSL21 x RSL21-139 than in CS x DSCnn1A.

Populations RSL21 x RSL21-12 and RSL21 x RSL21-103 were also polymorphic at the XGlu1 locus on the long arm (Figure 6). The lengths of interval XTri-XGlu1, which includes the centromere, were similar and did not statistically differ.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The minor Nor9 locus was estimated to have 100 to 140 gene copies in Chinese Spring using in situ DNA hybridization reported by MUKAI et al. 1991 and copy number reconstructions reported by LASSNER et al. 1987 . Dot blot hybridization of a rRNA gene probe suggested that there were 8400 gene copies per 2C nucleus in Chinese Spring but only 5000 gene copies per 2C nucleus in Chinese Spring ditelosomic 1AL. According to these data, Chinese Spring Nor9 should have 3,400 gene copies. This estimate cannot be correct because the minor Nor9 locus would then have as many or more rRNA gene copies than the major Nor1 and Nor2 loci in the B genome (Table 1). The present copy number determinations for Chinese Spring and ditelosomic 1AL nevertheless agree with earlier observations made by FLAVEL and O'DELL (1979) who reported 9150 rRNA gene copies for Chinese Spring but only 5300 copies for ditelosomic 1AL. FLAVELL and O'DELL 1979 concluded that ditelosomic 1AL suffered a deletion at a Nor locus in the B genome. Spontaneous deletions occur relatively frequently at the wheat Nor loci (DVORAK 1989 ). In a study of the Nor2 locus, DVORAK and APPELS 1986 found one deletion (caused either by a greatly unequal sister chromatid exchange or some other mechanism) among 446 progeny plants of 6B monosomics. Another source of variation in the rRNA gene copy number among Chinese Spring stocks could be polymorphism in the initial Chinese Spring population that was used to found the various wheat cytogenetic stocks around the globe. Whatever is the cause of this variation, caution needs to be exercised in comparisons of rRNA gene copy numbers based on different Chinese Spring cytogenetic stocks.

Using cytogenetic stocks that share isogenic Chinese Spring background, the Nor9 was estimated to contain 1,800 rRNA gene units per 2C nucleus in T. monococcum accession G1777. Thus the T. monococcum and bread wheat Nor9 haplotypes differ by an order of magnitude in the number of rRNA gene copies.

In both T. monococcum F₂ populations, recombination was lower near the Nor9 locus than recombination near the Nor9 locus in the Cheyenne-1 x Chinese Spring or Chinese Spring x DSCnn1A F₂ populations. Recombination in the neighborhood of the Nor10 locus on chromosome 5A^m could not be compared with that in homeologous chromosomes because of the paucity of markers near the Nor10 locus (DUBCOVSKY and DVORAK 1995 ). In chromosome 1B, recombination rates were greatly reduced near the Nor1 locus compared to T. monococcum chromosome 1A^m and wheat chromosome 1A in which no Nor1 rRNA genes have been detected. Interval lengths near the Nor2 locus on chromosome 6B could be compared with those on homeologous chromosomes 6A and 6D from which rRNA repeated gene units have been largely or entirely eliminated (APPELS et al. 1980 ; MUKAI et al. 1991 ). There is little recombination between the Nor2 locus and the centromere (DVORAK and CHEN 1984 ; DVORAK and APPELS 1986 ; JIA et al. 1996 ). Recombination is also low on the distal side of the Nor2 locus. The Nor2 locus is tightly linked to Xpsr627 and Xpsr962 on the map of chromosome 6B (JIA et al. 1996 ). While the Xpsr627-Xpsr962 interval is 3.6 cM on the 6B map, it is 19.0 cM on the 6D map (JIA et al. 1996 ). Xpsr962 was not mapped on chromosome 6A but recombination rates between 6B and 6A, could be compared in interval Xpsr627-Xpsr605. This interval, which includes the centromere, was 9.2 cM on the map of chromosome 6B but was 32.2 cM on the map of chromosome 6A (JIA et al. 1996 ). Thus, comparisons of interval lengths near the Nor2 locus on chromosome 6B relative to those on homeologous chromosomes 6A and 6D suggest that recombination rates are greatly reduced in the vicinity of the Nor2 locus. These observations are consistent with the hypothesis that major Nor loci have the potential for reducing recombination in their neighborhoods. Although we are not aware of a systematic study on the position effects of Nor loci on recombination in their neighborhood in other organisms, a report of a complete linkage between the maize Mdm1 locus and the Nor locus (SIMCOX et al. 1995 ) suggests that the maize Nor locus may, like the wheat Nor loci, exert a suppressive effect on recombination in its neighborhood. Low recombination rates across Nor-bearing chromosomes were recently noted also in Arabidopsis (COPENHAVER et al. 1998 ).

While recombination in chromosomes 1A^m, 1A, and 1B and 6A, 6B, and 6D was compared in different genomes and inferences may be confounded by other factors, recombination in the RSL21 x RSL21-12, RSL21 x RSL21-103, and RSL21 x RSL21-139 F₂ populations and Chinese Spring x DSCnn1A F₂ population was compared in a nearly isogenic background of Chinese Spring and was measured between the same chromosomes, Chinese Spring 1A and Cheyenne 1A. The 1A^m Nor9 haplotype was introduced onto the Chinese Spring and Cheyenne chromosomes on an identical segment. Thus, pairing occurred between completely homologous chromosomes in the RSL21 x RSL21-12, RSL21 x RSL21-103 and RSL21 x RSL21-139 F₁ plants. Nevertheless, the length of an interval in the vicinity of the 1A^m Nor9 locus (XGlu3-Xmwg60) was reduced compared to its length in the Chinese Spring x DSCnn1A F₂ population. The length of this interval became comparable to its length on the maps of chromosome 1A^m in T. monococcum.

In spite of the high isogenicity of the RSL21 x RSL21-12, RSL21 x RSL21-103, RSL21 x RSL21-139 and Chinese Spring x DSCnn1A populations, two potentially confounding factors need to be considered. One factor is that the heterozygous segment in which recombination was measured in the crosses between RSLs was juxtaposed to a homozygous Chinese Spring segment (Figure 1). This was not true for the 1A chromosome pair in the parents of the control Chinese Spring x DSCnn1A population; in those plants, the chromosome pair was composed of the entire Chinese Spring 1A and entire Cheyenne 1A chromosomes. If crossovers had been allocated preferentially into the homozygous segments in chromosomes composed of heterozygous and homozygous segments, recombination in the heterozygous Chinese Spring/Cheyenne segment could have been reduced not by the position effect of the 1A^m Nor9 haplotype but by intrachromosomal effects of the homozygous Chinese Spring segment. Two lines of evidence argue against this hypothesis. First, recombination in interval XTri-XGlu1 was similar in populations RSLS21 x RSL21-12 and RSL21 x RSL21-103, 21.9 cM and 25.6 cM, respectively. While in the former population the interval was juxtaposed to a heterozygous segment, it was juxtaposed to a homozygous Chinese Spring segment in the latter population (Figure 1 and Figure 6). Second, the Xmwg60-XksuE18 interval in populations RSLS21 x RSL21-12 and RSL21 x RSL21-103 was juxtaposed to a homozygous segment but the lengths of the interval were not shorter than the length of the interval in population RSLS21 x RSL21-139 in which the juxtaposed segment was heterozygous (Figure 1 and Figure 6). These observations provide therefore no evidence for intra-chromosomal effects of homozygous segments on recombination in juxtaposed heterozygous segments in this material.

The second potentially confounding factor is that the replacement of 1A Nor9 in the Chinese Spring and Cheyenne chromosome pair by 1A^m Nor9 led to introgression of a T. monococcum chromosome segment. Although the segment is short, the possibility cannot be excluded that some other factor, the telomere in particular, is not responsible for the effects observed here. Comparisons of the distribution of recombination in the T. monococcum genome with the distribution of recombination in the wheat and barley genomes showed no reductions in recombination in the distal regions of T. monococcum chromosomes (DUBCOVSKY et al. 1996 ). The Nor9-bearing arm was the sole exception. Additionally, if the T. monococcum telomere or a T. monococcum gene caused poor pairing of the short arm of the 1A/1A^m chromosome, recombination in other intervals on the 1AS arm should had been reduced in addition to the most distal one, which was not the case.

Interval XGlu3-Xmwg60 in which the position effect of the Nor9 locus on recombination was investigated was 19.4 cM long on the Chinese Spring x DSCnn1A map. Because no polymorphic locus was found between Chinese Spring and Cheyenne in that region, a more precise determination of the distance at which the Nor9 exerted its effect upon recombination could not be made. Nevertheless, it appears that the position effect extends 17 or more cM away from the Nor9 locus. This is suggested by the following reasoning. The interval in which the effect was measured was at least 8 cM away from Nor9, 5.2 cM from XGlu3 to XGli1 (Figure 5 and Figure 6), and 2.3 cM from XGli1 to Bg (PANIN and NETSVETAEV 1986 ). The interval itself, which is 19.4 cM long in the Chinese Spring and Cheyenne chromosome pair, was reduced to 10.7 cM, i.e., by 8.7 cM. If recombination were entirely eliminated in the distal region of the interval, the effect would reach 8.7 cM into the XGlu3-Xmwg60 interval.

While the replacement of the Chinese Spring and Cheyenne Nor9 haplotypes by the T. monococcum 1A^m Nor9 haplotype resulted in a decrease in recombination in the XGlu3-Xmwg60 interval, i.e., near the Nor9 locus, an intra-arm compensatory increase in recombination occurred in the more proximal direction. As a result, the genetic length of the short arm, measured from XGlu3 to XTri, remained essentially unaltered, 39.5 and 36.3 cM in RSL21 x RSL21-139 and CS x DSCnn1A, respectively (Figure 6). The major effect of Nor loci on homologous meiotic exchanges is their redistribution in chromosome arms in wheat.

The mechanism by which Nor loci exert position effects on recombination in their vicinity is not clear. It is possible that the attachment of a chromosome arm to the nucleolus at the nucleolus organizing region (NOR) impairs the search for homologous sequences during the initial stages of homologous pairing (for review see, e.g., LOIDL 1990 ). The NOR-bearing arms are attached to the nucleolus for most of prophase I, which could interfere with other stages of the recombination process. Also unknown is whether Nor loci suppress recombination because of the expression of rRNA genes or because of the lack of expression of a portion (sometimes all) of the genes. Only a fraction of the rRNA genes of the wheat Nor loci are expressed; the rest of them form facultative heterochromatin associated with the NOR. Nor loci share a number of features, including peculiar recombination properties, with constitutive heterochromatin. Meiotic homologous exchanges are generally absent from heterochromatin (ROBERTS 1965 ). Likewise, homologous exchanges tend to be underrepresented within the Nor loci (PETES 1980 ; COEN et al. 1982 ; RANZANI et al. 1984 ; DVORAK and APPELS 1986 ; WILLIAMS and ROBBINS 1992 ; SIMCOX et al. 1995 ). Constitutive heterochromatin tends to have negative position effects on meiotic homologous exchanges in the vicinity and to cause redistribution of chiasmata within the chromosome arm (for a review see JOHN and MIKLOS 1979 ). Data in this paper show that the same is true for major Nor loci.

	ACKNOWLEDGMENTS

This project is a contribution to the International Triticeae Mapping Initiative (ITMI). The authors express their gratitude to the following: M. D. GALE, A. GRANER, M. E. SORRELLS, and B. S. GILL for supplying clones; J. DUBCOVSKY for assistance in the construction of the F₂ genetic map of chromosome 1A based on the Cheyenne-1 x Chinese Spring cross and for his critical reading of the manuscript and valuable suggestions; and D. LAVELLE for assistance in the measurement of Nor9 gene copy numbers. The authors acknowledge financial support from the United States Department of Agriculture National Research Initiative Competitive Grant Program by grant 96-35300-3822 to J. DVOÁK.

Manuscript received October 13, 1997; Accepted for publication March 4, 1998.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ALLARD, R. W., 1956  Formulas and tables to facilitate the calculation of recombination values in heredity. Hilgardia 24:235-278.

APPELS, R. and J. DVOÁK, 1982  The wheat ribosomal DNA spacer: its structure and variation in populations and among species. Theor. Appl. Genet. 63:337-348.

APPELS, R., W. L. GERLACH, E. S. DENNIS, H. SWIFT, and W. J. PEACOCK, 1980  Molecular and chromosomal organization of DNA sequences coding for the ribosomal RNAs in cereals. Chromosoma 78:293-311.

CASSIDY, B. G. and J. DVOÁK, 1991  Molecular characterization of a low-molecular-weight glutenin cDNA clone from Triticum durum.. Theor. Appl. Genet. 81:653-660.

COEN, E. S., J. M. THODAY, and G. A. DOVER, 1982  Rate of turnover of structural variants in the rDNA gene family of Drosophila melanogaster.. Nature 281:564-568.

COPENHAVER, G. P., K. L. NICKEL and D. PREUSS, 1998 Assaying centromere functions and genome-wide recombination in Arabidopsis, p. 47 in Proc. Plant and Animal Genome VI. International Conference on the Status of Plant and Animal Genome Research. Scherago International Inc., New York.

CROSBY, A. R., 1957  Nucleolar activity of lagging chromosomes in wheat. Am. J. Bot. 44:813-822.

DUBCOVSKY, J. and J. DVOÁK, 1995  Ribosomal RNA loci: Nomads in the Triticeae genomes. Genetics 140:1367-1377[Abstract].

DUBCOVSKY, J., A. F. GALVEZ, and J. DVOÁK, 1994  Comparison of the genetic organization of the early salt stress response gene system in salt-tolerant Lophopyrum elongatum and salt-sensitive wheat. Theor. Appl. Genet. 87:957-964.

DUBCOVSKY, J., M.-C. LUO, G. Y. ZHONG, R. BRANSTEITTER, and A. DESAI et al., 1996  Genetic map of diploid wheat, Triticum monococcum L., and its comparison with maps of Hordeum vulgare L. Genetics 143:983-999[Abstract].

DUBCOVSKY, J., M.-C. LUO, and J. DVOÁK, 1995  Differentiation between homoeologous chromosomes 1A of wheat and 1A^m of Triticum monococcum and its recognition by the wheat Ph1 locus. Proc. Natl. Acad. Sci. USA 92:6645-6649[Abstract/Free Full Text].

DVOÁK, J., 1989 Organization and evolution of multigene families: inferences from ribosomal RNA gene loci of wheat and related species, pp. 83–97 in Plant Population Genetics, Breeding, and Genetic Resources, edited by A. H. D. BROWN, M.T. CLEGG, A. L. KAHLER, and B. S. WEIR. Sinauer Associates Inc., Sunderland, MA.

DVOÁK, J. and R. APPELS, 1986  Investigation of homologous crossing over and sister chromatid exchange in the wheat Nor-2 locus coding for rRNA and Gli-B2 locus coding for gliadins. Genetics 113:1037-1056[Abstract/Free Full Text].

DVOÁK, J. and K.-C. CHEN, 1984  Distribution of nonstructural variation between wheat cultivars along chromosome arm 6Bp: Evidence from the linkage map and physical map of the arm. Genetics 106:325-333[Abstract/Free Full Text].

FLAVELL, R. B. and M. O'DELL, 1976  Ribosomal RNA genes on homoeologous chromosomes of groups 5 and 6 in hexaploid wheat. Heredity 37:377-385.

FLAVELL, R. B. and M. O'DELL, 1979  The genetic control of nucleolus formation in wheat. Chromosoma 71:135-152.

FLAVELL, R. B. and D. B. SMITH, 1974  The role of homoeologous group 1 chromosomes in the control of rRNA genes in wheat. Biochem. Genet. 12:271-279[Medline].

FORDE, B. G., M. KREIS, M. S. WILLIAMSON, R. P. FRY, and J. PYWELL et al., 1985  Short tandem repeats shared by B- and C-hordein cDNAs suggest a common evolutionary origin for two groups of cereal storage protein genes. EMBO J. 4:9-15[Medline].

GERLACH, W. L. and J. R. BEDBROOK, 1979  Cloning and characterization of ribosomal RNA genes from wheat and barley. Nucleic Acids Res. 7:1869-1885[Abstract/Free Full Text].

GRANER, A., A. JAHOOR, J. SCHONDELMEIER, H. SIEDLER, and K. PILLEN et al., 1991  Construction of an RFLP map of barley. Theor. Appl. Genet. 83:250-256.

JIA, J., K. M. DEVOS, S. CHAO, T. E. MILLER, and S. M. READER et al., 1996  RFLP-based maps of the homoeologous group-6 chromosomes of wheat and their application in the tagging of Pm12, a powdery mildew resistance gene transferred from Aegilops speltoides to wheat. Theor. Appl. Genet. 92:559-565.

JIANG, J. and B. S. GILL, 1994  New 18S-26S ribosomal RNA gene loci: chromosomal landmarks for the evolution of polyploid wheats. Chromosoma 103:179-185[Medline].

JOHN, B. and G. L. G. MIKLOS, 1979  Functional aspects of satellite DNA and heterochromatin. Int. Rev. Cytol. 58:1-114[Medline].

KOSAMBI, D. D., 1943  The estimation of map distances from recombination values. Ann. Eugen. 12:172-175.

LANDER, E. S., P. GREEN, J. ABRAHAMSON, A. BARLOW, and M. DALY et al., 1987  MAPMAKER: An integrated computer package for construction of primary linkage maps of experimental and natural populations. Genomics 1:174-181[Medline].

LASSNER, M., O. ANDERSON, and J. DVOÁK, 1987  Hypervariation associated with a 12-nucleotide direct repeat and inferences on intragenomic homogenization of ribosomal RNA gene spacer based on the DNA sequence of a clone from the wheat Nor-D3 locus. Genome 29:770-781.

LINCOLN, S., M. DALY and E. LANDER, 1992 Constructing genetic maps with MAPMAKER/EXP 3.0 Ed. 3. Whitehead Institute Technical Report, Cambridge, MA.

LOIDL, J., 1990  The initiation of meiotic chromosome pairing: the cytological view. Genome 33:759-778[Medline].

LONGWELL, A. C. and G. SVIHLA, 1960  Specific chromosomal control of the nucleolus and of cytoplasm in wheat. Exp. Cell Res. 20:294-312.

MILLER, T. E., W. L. GERLACH, and R. B. FLAVELL, 1980  Nucleolus organiser variation in wheat and rye revealed by in situ hybridization. Heredity 45:377-382.

MORRIS, R., J. W. SCHMIDT, P. J. MATTERN, and V. A. JOHNSON, 1966  Chromosomal location of genes for flour quality in the wheat variety Cheyenne using substitution lines. Crop Sci. 6:119-122[Abstract/Free Full Text].

MUKAI, Y., T. R. ENDO, and B. S. GILL, 1991  Physical mapping of the 18S.26S rRNA multigene family in common wheat: Identification of a new locus. Chromosoma 100:71-78.

PANIN, V. M. and V. P. NETSVETAEV, 1986  Genetic control of gliadins and some morphological characters of spike in durum winter wheats. Nauchno Tekhn. Bull. USGI Odessa 2:31-36.

PAULL, J. G., M. A. PALLOTA, P. LANGRIDGE, and T. T. THE, 1994  RFLP markers associated with Sr22 and recombination between chromosome 7A of bread wheat and the diploid species Triticum boeoticum.. Theor. Appl. Genet. 89:1039-1045.

PETES, T. D., 1980  Unequal meiotic recombination within arrays of yeast ribosomal DNA genes. Cell 19:765-774[Medline].

RANZANI, G. N., L. F. BERNINI, and M. CRIPPA, 1984  Inheritance of rDNA spacer length variants in man. Mol. Gen. Genet. 196:141-145[Medline].

RAYBURN, A. L. and B. S. GILL, 1986  Molecular identification of the D genome chromosomes of wheat. J. Hered. 77:253-255[Abstract/Free Full Text].

ROBERTS, P. A., 1965  Difference in behaviour of eu- and heterochromatin: crossing-over. Nature 205:725-726[Medline].

SIMCOX, K. D., M. D. MCMULLEN, and R. LOUIE, 1995  Co-segregation of the maize dwarf mosaic virus resistance gene, Mdm1, with the nucleolus organizer region in maize. Theor. Appl. Genet. 90:341-346.

WILLIAMS, S. M. and L. G. ROBBINS, 1992  Molecular genetic analysis of Drosophila rDNA arrays. Trends Genet. 8:335-340[Medline].