Genetic studies have established that members of the C1 / Pl gene family act to induce anthocyanin production in a tissue-specific manner. The c1 gene controls pigmentation in the aleurone and embryo of the kernel, and pl controls pigmentation in the vegetative portions of the plant, such as husk, stem, leaves, anthers, and roots. Two indirect types of evidence have led to the notion that this tissue-specificity is due to tissue-specific transcription of these regulatory genes. First, sequence analysis reveals that although the coding regions of the two genes are very highly conserved, the 5' promoter regions are strikingly different (Cone et al., Plant Cell 5:1795-1805, 1993). Second, tissue-specificity is lost if c1 or pl expression is driven by a "constitutive" promoter such as the 35S promoter from cauliflower mosaic virus (P. Cooper, unpublished). These results, together with the genetic behavior of the regulatory genes, have led to the conclusion that c1 controls pigmentation in the kernel because c1 is transcribed only in the kernel; similarly, pl controls pigmentation in tissues other than the kernel because pl is transcribed only in those tissues. In this communication, we describe results that call for a more conservative interpretation of the mechanism for tissue-specific pigmentation.

We have been using reverse transcriptase-polymerase chain reaction (RT-PCR) to examine levels of transcript in husks from plants carrying different alleles of pl. For the amplification, we typically use a pair of primers that span the first intron of the gene. Because the sequences of pl and c1 are very similar in this part of the gene, the primers can amplify both pl and c1 sequences. Recently, we used these non-specific primers to quantitate mRNA in husks from plants carrying a putative null pl allele (pl-987). The results indicated a very low level of transcript relative to the progenitor allele, pl-W22. To ask if the transcript might arise from "ectopic" expression of the functional C1-W22 allele carried by these plants, we repeated the experiment with pl- and c1- specific primer pairs (Table 1). Not unexpectedly, the pl-specific pair detected transcripts in both the pl-987 and pl-W22 husks, although the levels were much lower in the pl-987 husks. Surprisingly however, the c1-specific primer pair detected C1 transcripts in both samples. This result suggested that the C1-W22 allele in this line is normally expressed at low levels in husks, a tissue whose pigmentation is not genetically controlled by c1. We were also able to detect C1 transcripts in lines carrying a C1-McClintock (C1-McC) allele.

To test the possibility that pl might be expressed "ectopically" in kernels where pl , by genetic criteria, is not believed to act, we used gene-specific primer pairs to amplify transcripts from pigmented kernels from lines carrying pl-W22 or Pl-Rhoades alleles. In both cases, the c1-specific primer pair resulted in an amplification product but the pl-specific pair did not. These data indicate neither of these two pl alleles is expressed in kernels.

In an earlier publication (Cone et al., Plant Cell 5:1807-1816, 1993), we reported pl mRNA levels in husks from lines carrying different pl alleles--Pl-Rhoades and pl-McClintock (pl-McC).

Because those experiments were carried out using primers that would amplify both c1 and pl transcripts and both of the lines carried functional C1, the absolute levels of pl RNA reported earlier are probably overestimated. However, because the two lines carried the same C1 allele, the relative levels of pl mRNA are still an accurate indicator of the differences in levels of pl activity.

The results presented here prompt a re-interpretation of the idea that the tissue-specificity of anthocyanin pigmentation is determined by on/off patterns of transcription of the c1 and pl regulatory genes. Although we did not rigorously quantitate the level of C1 transcript in husks, it is roughly similar to the level of pl transcript. Because the husks were unpigmented, the levels of both transcripts must be below the threshold required to activate the anthocyanin structural genes. Thus, even though C1 expression is technically "on" in these husks, the expression is not high enough to be physiologically significant in this tissue. This sub-threshold level of expression in husks therefore does not contradict the genetic interpretation of the c1 gene's lack of influence on anthocyanin production in vegetative tissues.

For RT-PCR, RNA was purified from kernels harvested 19-25 days after pollination or from inner husk leaves harvested from field-grown plants at the time of silk emergence. In most cases, poly(A)+ (approximately 1 µg) was used for the RT-PCR. RNA was converted to first-strand cDNA using a Gibco BRL Superscript Preamplification Kit. For PCR, samples were denatured for 5 min at 94 C and amplification was achieved by 35 cycles of denaturation at 94 C for 1 min, annealing at 54 or 55 C (depending on primer set), and extension at 72 C for 2 min. Final extension was at 72 C for 10 min.

Two sets of pl-specific primers were used. The first set was designed from sequences in the 5' region of the Pl-Rhoades gene. The upstream primer is located in the 5' untranslated region of the first exon (5'-ACCCTGCTGCTAGCTAGCTG-3') and the downstream primer is located near the 3' end of the second exon (5'-CTGTTGCCGAGGAGCTTGTG-3'). Amplification of properly spliced RNA with these primers yields a 316 bp product. The second set of pl-specific primers was designed from sequences at the 3' end of the pl-W22 gene. The upstream primer is located 53 bp 5' of the stop codon (5'-TCTCGAGTCCGACGAGG-3') and the downsteam primer is in the 3' untranslated region (5'-GTATACATACGCATGGCTA-3'). Amplification with these primers results in a 107 bp product.

One set of c1-specific primers was designed based on the sequence of a C1-W22 allele (Paz-Ares et al., EMBO J. 6:3553-3558, 1987). The upstream primer is located 53 bp 5' of the stop codon and is identical to the upstream primer from pl-W22 (5'-TCTCGAGTCCGACGAGG-3'). The downstream primer is located 20 bp 3' of the stop codon (5'-CCTCGTGCTTATTGGACA-3'). Amplification with this primer pair produces a 93 bp product from C1-W22 and a 369 bp product from pl-McC RNA. The larger size of the latter product is presumably due to a polymorphic insertion.

Table 1. Pigmentation patterns and expression of various pl and c1 alleles in husk and kernel tissue.
 

	pigment		pl mRNA		c1 mRNA
Genotype	husk	kernel	husk	kernel	husk	kernel
pl-987, C1-W22	?	+	+	nd	+	nd
pl-W22, C1-W22	+	+	+	?	+	+
Pl-Rhoades, C1-McC	+	+	+	?	nd	+

Return to the MNL 71 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage