Fig. 1. Floral phenotype of mutant plants impaired in miR164 biogenesis. (A) Quantification of miR164 abundance in mir164 mutants. RNA blot analysis of the small RNA fraction isolated from wild-type Col-0, mir164a-4, mir164b-1, mir164c-1, mir164c-2, and mir164a-4 b-1 c-1 triple-mutant inflorescences hybridized with probes complementary to miR164a (upper blot) and U6 small RNA (middle blot), respectively. The ethidium bromide-stained agarose gel is shown beneath (numbers indicate fold-change of miR164 accumulation with respect to Col-0 wt, which was set to 1). As miR164a and miR164b differ from miR164c in a single nucleotide, different miR164 miRNAs cannot be distinguished on an RNA blot; thus, signals are derived from all three miR164 miRNAs. The experiment was repeated twice with the same result. The antisense MIR164 oligonucleotide probe hybridizes to two distinct RNA size classes, of 21 and 24 nt, in agreement with previous reports (Dunoyer et al., 2004; Valoczi et al., 2006). It has been proposed that the 21 nt form of miR164 is the functional entity sufficient to guide target cleavage, for which the 24 nt form, which has distinct requirements for its biogenesis, appears to be dispensable (Dunoyer et al., 2004). (B-D) Results of SEM analysis. (B) Mature (stage 13) wild-type flower of accession Ler. Flower stages were defined according to Smyth et al. (Smyth et al., 1990). (C) Stage 12 and (D) stage 13 flowers of mir164abc triple-mutant plants show variable organ numbers and unfused carpels. Sepals have been removed for better visibility of the inner organs. Scale bars: 200 µm in B; 100 µm in C,D. Abbreviations: pe, petals; ca, carpels; st, stamens. (E) A mir164abc triple-mutant inflorescence. (F,G) Charts representing organ counts from mir164abc triple-mutant (black) and mir164aAbBcC plants (gray), which served as the wild-type control to assess the potential influence of the mixed Ler/Col-0 background on the phenotypic changes. The average floral organ number ('Organ count') is plotted against each flower position along the stem ('Flower'). Numbers indicate the position of the flower along the stem from the oldest (1) to the youngest (25). Error bars represent s.d. in (F) sepal and (G) petal number. Stamen number was reduced with respect to Col-0 and slightly reduced with respect to the wild-type control. Notably, variability in stamen number, but not in sepal and petal number, increased in the mixed Ler/Col-0 background, when compared with the Col-0 background (data not shown). Carpel number is only weakly affected in mir164abc mutants.