Natural variation in teosinte at the domestication locus teosinte branched1 (tb1)

The teosinte branched1(tb1) gene is a major QTL controlling branching differences between maize and its wild progenitor, teosinte. The insertion of a transposable element (Hopscotch) upstream of tb1 is known to enhance the gene’s expression, causing reduced tillering in maize. Observations of the maize tb1 allele in teosinte and estimates of an insertion age of the Hopscotch that predates domestication led us to investigate its prevalence and potential role in teosinte. We assessed the prevalence of the Hopscotch element across an Americas-wide sample of 837 maize and teosinte individuals using a co-dominant PCR assay. Additionally, we calculated population genetic summaries using sequence data from a subset of individuals from four teosinte populations and collected phenotypic data using seed from a single teosinte population where Hopscotch was found segregating at high frequency. Genotyping results indicate the Hopscotch element is found in a number of teosinte populations and linkage disequilibrium near tb1 does not support recent introgression from maize. Population genetic signatures are consistent with selection on the tb1 locus, revealing a potential ecological role, but a greenhouse experiment does not detect a strong association between the Hopscotch and tillering in teosinte. Our findings suggest the role of Hopscotch differs between maize and teosinte. Future work should assess tb1 expression levels in teosinte with and without the Hopscotch and more comprehensively phenotype teosinte to assess the ecological significance of the Hopscotch insertion and, more broadly, the tb1 locus in teosinte.

with the online version of this article). DNA was extracted from leaf tissue using a modified 25 CTAB approach (Doyle and Doyle, 1990;Maloof et al., 1984). We designed primers using   La Mesa (MSA) populations (Pyhäjärvi et al., 2013). We chose these populations because we had 10 both genotyping data for the Hopscotch as well as chromosome-wide SNP data for chromosome 1.

11
For each population we filtered the initial set of 5,897 SNPs on chromosome 1 to accept only 12 SNPs with a minor allele frequency of at least 0.1, resulting in 1,671, 3,023, 3,122, and 2,167 13 SNPs for SLO, EjuB, EjuA, and MSA, respectively. We then used Tassel (Bradbury et al., 2007) 14 to calculate linkage disequilibrium (r 2 ) across chromosome 1 for each population. 15 We examined evidence of introgression on chromosome 1 in these same four populations heterozygous for the insertion. We chose between 10-13 seeds from each of 23 sampling sites. We 30 treated seeds with Captan fungicide (Southern Agricultural Insecticides Inc., Palmetto, Florida, 31 USA) and germinated them in petri dishes with filter paper. Following germination, 206 32 successful germinations were planted into one-gallon pots with potting soil and randomly spaced 1 one foot apart on greenhouse benches. Plants were watered three times a day by hand and with 2 an automatic drip containing 10-20-10 fertilizer.

3
Starting on day 15, we measured tillering index as the ratio of the sum of tiller lengths to the 4 height of the plant (Briggs et al., 2007). Following initial measurements, we phenotyped plants for 5 tillering index every 5 days through day 40, and then on day 50 and day 60. On day 65 we 6 measured culm diameter between the third and fourth nodes of each plant. Culm diameter is not 7 believed to be correlated with tillering index or variation at tb1. Following phenotyping we 8 extracted DNA from all plants using a modified SDS extraction protocol. We genotyped 9 individuals for the Hopscotch insertion following the protocols listed above. Based on these initial  Table 1).

11
Although we found large variation in Hopscotch allele frequency among our populations, BayEnv 12 analysis did not indicate a correlation between the Hopscotch insertion and environmental 13 variables (all Bayes Factors < 1).
14 Sequencing-To investigate patterns of sequence diversity and linkage disequilibrium (LD) 15 in the tb1 region, we sequenced two small (<1kb)  Evidence of introgression-The highest frequency of the Hopscotch insertion in teosinte 3 was found in parviglumis sympatric with cultivated maize. Our initial hypothesis was that the 4 high frequency of the Hopscotch element in these populations could be attributed to introgression 5 from maize into teosinte. To investigate this possibility we examined overall patterns of linkage 6 disequilibrium across chromosome one and specifically in the tb1 region. If the Hopscotch is found 7 in these populations due to recent introgression we would expect to find large blocks of linked 8 markers near this element. We find no evidence of elevated linkage disequilibrium between the 9 Hopscotch and SNPs surrounding the tb1 region in our resequenced populations (Figure 2), and 10 r 2 in the tb1 region does not differ significantly between populations with (average r 2 of 0.085) 11 and without (average r 2 = 0.082) the Hopscotch insertion. In fact, average r 2 is lower in the tb1 12 region (r 2 = 0.056) than across the rest of chromosome 1 (r 2 = 0.083; Table 3).

13
The lack of clustering of Hopscotch genotypes in our NJ tree as well as the lack of LD around

32
After performing a repeated measures ANOVA between our transformed tillering index data and 1 Hopscotch genotype we find no correlation between genotype at the Hopscotch insertion and 2 tillering index (Fig. 4), tiller number, or culm diameter. 3 We performed a second grow-out of parviglumis from San Lorenzo (Phenotyping 2) to assess 4 whether lighting conditions or sample size may have affected our ability to detect an effect of tb1.

5
For the second grow-out we measured tillering index every five days through day 50 for 302 6 plants. We found the Hopscotch allele segregating at a frequency of 0.69, with a 0.6 frequency of 7 Hopscotch homozygotes, and a 0.2 frequency of both heterozygotes and homozygotes for the 8 teosinte allele. Results were similar to Phenotyping 1, with no significant correlation between 9 Hopscotch and any of the three phenotypes measured.

10
(2013) who found resistance to introgression from maize into teosinte around domestication loci.

11
We find no evidence of recent introgression in our analyses. Clustering in our NJ trees do not 12 reflect the pattern expected if maize alleles at the tb1 locus had introgressed into populations of 13 teosinte. Moreover, there is no signature of elevated LD in the tb1 region relative to the rest of 14 chromosome 1, and Bayesian assignment to a maize cluster in this region is both low and below 15 the chromosome-wide average (Fig. 3, Table 4). Together, these data point to an explanation other phenotypic traits (Kebrom and Brutnell, 2007;Clark et al., 2006). A recent study by Clues to the identity of these loci may be found in QTL studies that have identified loci 2 controlling branching architecture (e.g., Doebley andStec 1991, 1993 Table 3. mean r 2 values between SNPs on chromosome 1, in the broad tb1 region, within the 5' UTR of tb1 (Region 1), and 66kb upstream of tb1 (Region 2).