Populations of the "Louisiana iris" species Iris fulva, I. hexagona, and I. nelsonii were examined genetically to test for interspecific gene flow between I. fulva and I. hexagona, for pollen- versus seed-mediated introgression between these species, and for the presumed hybrid origin of I. nelsonii. Genetic markers were identified by using both a polymerase chain reaction-like method that allows the identification of random, nuclear markers and standard polymerase chain reaction experiments involving specific chloroplast DNA (cpDNA) oligonucleotides. Restriction endonuclease digestions of the cpDNA amplification products resolved diagnostic restriction site differences for I. fulva and I. hexagona. The distribution of the species-specific nuclear markers supports a hypothesis of bidirectional introgression between I. fulva and I. hexagona. Thus, individuals analyzed from a contemporary hybrid population demonstrate multilocus genotypes that are indicative of advanced-generation hybrid individuals. Furthermore, several markers from the alternate species were present in low frequency in one allopatric population each of I. fulva and I. hexagona. Data from the nuclear analysis also support the hypothesized hybrid origin of I. nelsonii from the interaction of I. fulva and I. hexagona. Finally, cpDNA data support the hypothesis that the localized and the dispersed introgression are largely due to pollen transfer. In addition to the biological implications, this study demonstrates the power of the polymerase chain reaction methodology for the rapid identification of random and specific genetic markers for testing evolutionary genetic hypotheses.

The transcriptional regulator Spo0A activates transcription from two types of promoters. One type of promoter is used by RNA polymerase containing sigma A, whereas the other type is used by RNA polymerase containing sigma H. There are Spo0A-binding sites near the -35 region of both types of promoters. It has been reported that some transcriptional regulators that bind near the -35 regions of promoters directly interact with the sigma subunit of RNA polymerase. Therefore, we looked for evidence that Spo0A interacts with both sigma factors by searching for single amino acid substitutions in these factors that specifically prevent expression from Spo0A-dependent promoters, but that do not decrease activity of Spo0A-independent promoters. Two such amino acid substitutions were isolated in sigma A and one was isolated in sigma H. The amino acid substitutions in sigma A prevented expression from the Spo0A-activated promoters, spoIIG and spoIIE, but expression was not impaired from the Spo0A-independent, sigma A-dependent promoter tms or from the Spo0A-activated, sigma H-dependent promoter, spoIIA. The amino acid substitution in sigma H prevented expression from the spoIIA promoter but not from the Spo0A-independent promoter, citGp2, which is used by sigma H-RNA polymerase. All of these amino acid substitutions occur in the carboxyl terminus of the sigma factors. These amino acid substitutions may define the sites of contact between the sigma factors and Spo0A. The ability of response regulators such as Spo0A to interact with multiple sigma factors may increase the variety of responses made by bacteria using a limited number of transcription factors.

A's ability to direct transcription in vivo and in vitro. We found that alanine substitutions at these positions specifically reduced expression from the A -dependent, Spo0A-dependent promoters, spoIIG and spoIIE, in vivo. Furthermore, we found that stimulation of spoIIG promoter activity by Spo0A in vitro was reduced by the single substitutions H359A and H359R in A .A growing body of evidence supports the idea that sitespecific DNA binding proteins activate promoters in bacteria by contacting RNA polymerase (reviewed in reference 4). The site on RNA polymerase that is contacted by the activator protein appears to depend in part on the site on DNA at which the activator protein is bound (16). The Escherichia coli cyclic AMP receptor protein (CAP) interacts with the C-terminal domain (CTD) of the RNA polymerase alpha subunit when CAP is bound at a site centered about 61 bp upstream from the transcription start point (Ϫ61) (reviewed in reference 11), whereas CAP bound at a site centered at about position Ϫ41 activates transcription by contacting the N-terminal domain of the ␣ subunit (25; reviewed in reference 5). Other activator proteins contact the subunit of RNA polymerase. FNR (a protein that is structurally similar to CAP), PhoB, MalT, and cI from phage lambda bind to sites that overlap the Ϫ35 region of promoters, and probably contact the subunit of RNA polymerase (reference 5 and references therein; 9, 22-24). Some promoters contain multiple binding sites for activators. In some of these cases the two coactivators appear to interact directly with the RNA polymerase (6,18,29). At other promoters one coactivator affects DNA binding by the second coactivator, which is the only one to make direct contact with RNA polymerase (26).Spo0A from Bacillus subtilis, which is required for the initiation of sporulation, binds the spoIIG promoter at two sites known as 0A boxes, centered at positions Ϫ37 and Ϫ87 (reviewed in reference 15; 1, 27). Binding of Spo0A at these sites stimulates utilization of the spoIIG promoter by RNA polymerase containing A , a homolog of E. coli 70 (1, 3, 20, 27, 28). It is not known whether spoIIG promoter activation involves interactions between Spo0A and RNA polymerase. Baldus et al. (2) found that spoIIG promoter activity was reduced in mutants of B. subtilis in which A contained one of two single amino acid substitutions (glutamate substituted for lysine at position 356, K356E, or arginine substituted for histidine at position 359, H359R). These observations led to the suggestion that spoIIG promoter activation involves an interaction between Spo0A and A . The two amino acid substitutions in A that reduced spoIIG promoter activity lie near the region of A that makes sequence-specific contacts with the base pairs at the Ϫ35 region of its cognate promoters (Fig. 1). Other amino acid substitutions in this region of A and other sigma factors have been shown to change the specificity of promoter utilization by RNA polymerase containing these mutant sigma factors (12,19,32). An alternative e...

An analysis of chloroplast DNA (cpDNA) variation was carried out for 106 individual plants from three natural populations of Louisiana irises. Two of the samples (59 individuals) represented I. brevicaulis populations. The third sample was from a population defined by allozyme markers as an area of contact between I. fidva, I. hexagona and I. brevicaulis. The cpDNA acts as a seed-specific genetic marker because it is inherited from the maternal parent. cpDNA markers were thus used to discriminate between (i) introgressive hybridization due to seed movement followed by pollen transfer and, (ii) introgression resulting from direct transfer of pollen between allopatric populations of the hybridizing taxa. Furthermore, the concurrent analysis of biparental and maternal markers for the same individuals allowed a test for any directionality in the introgression. A comparison of cpDNA results with data from previous nuclear analyses led to the conclusion that pollen flow occurred from allopatric populations of I. hexagona into an area of sympatry involving I. fulva and I. brevicaulis. In addition, the genotypes detected in the hybrid population indicate that 1. fulva and I. brevicaulis have acted as both pollen and seed parents to produce introgressant individuals. The results of the present study and those of previous nuclear and cpDNA analyses suggest that pollen dispersal is the most important avenue for gene flow between these Iris species.

Spo0A is a DNA binding protein in Bacillus subtilisrequired for the activation of spoIIG and other promoters at the onset of endospore formation. Activation of some of these promoters may involve interaction of Spo0A and the ςAsubunit of RNA polymerase. Previous studies identified two single-amino-acid substitutions in ςA, K356E and H359R, that specifically impaired Spo0A-dependent transcription in vivo. Here we report the identification of an amino acid substitution in Spo0A (S231F) that suppressed the sporulation deficiency due to the H359R substitution in ςA. We also found that the S231F substitution partially restored use of the spoIIG promoter by the ςA H359R RNA polymerase in vitro. Alanine substitutions in the 231 region of Spo0A revealed an additional amino acid residue important for spoIIG promoter activation, I229. This amino acid substitution in Spo0A did not affect repression of abrB transcription, indicating that the alanine-substituted Spo0A was not defective in DNA binding. Moreover, the alanine-substituted Spo0A protein activated the spoIIApromoter; therefore, this region of Spo0A is probably not required for Spo0A-dependent, ςH-directed transcription. These and other results suggest that the region of Spo0A near position 229 is involved in ςA-dependent promoter activation.