364 C.L.H.Hord·H.Ma
SPL/NZZ is a central mediator for the promotion of microsporogenesis by
AG (Ito et al. 2004). These35S::SPL-GRflowers produced whorl 3 petals with
locules, instead of stamens; therefore,SPL/NZZalone was not sufficient to
specify stamens and other whorl-specific factors are involved in this process
(Ito et al. 2004).
ThereisevidencethatSPL/NZZis expressed at anther stage 2, consistent
with the idea thatSPL/NZZis necessary for promoting cellular division and
differentiation of another cell type (Schiefthaler et al. 1999). It is possible that
SPL/NZZpromotes the formation or differentiation of the PPCs and PSCs. Al-
ternatively,SPL/NZZmight not be necessary for the differentiation of PPCs
and PSCs, but is required for promoting the differentiation of the sporoge-
nous cells beginning at late stage 3. Due to the lack of specific molecular
markers for these early anther cell types, it is difficult to know whether PPCs
and PSCs are correctly formed at stage 3 and these hypotheses cannot yet be
distinguished.
Although the parietal cell types are affected in thespl/nzzmutants, it is
not known howSPL/NZZregulates the development of the parietal cell types.
A current theory in anther development is that the sporogenous cells pro-
mote the differentiation and development of the parietal cell layers possibly
through cell-cell communication (Albrecht et al. 2005; Scott et al. 2004; Yang
et al. 1999, 2003). This model would support an indirect role forSPL/NZZin
sporophytic development, by regulating genes necessary for the formation of
sporogenous cells, which then promote the differentiation of adjacent cells
into the parietal cell types. Alternatively, it is possible that thespl/nzzmutants
are not able to properly specify the archesporial cell type. Thus, the observa-
tion that the anther was filled with vacuolated parenchyma cells (Yang et al.
1999) may be the result of a defect in this specification. In this case,SPL/NZZ
would play an earlier and more direct role in specifying the parietal cells.
3
Tapetum Specification Requires Cell-Cell Signaling
The tapetum cell layer supports pollen development by producing and releas-
ing essential proteins (Izhar and Frankel 1971; Stieglitz 1977). When the tape-
tum cells are selectively destroyed the plant fails to produce pollen (Koltunow
et al. 1990; Mariani et al. 1990, 1991). Recently several genes have been shown
to be critical for tapetum formation and may be components of the same
signaling pathway. Two groups independently isolated mutant alleles of the
same gene calledEXCESS MICROSPOROCYTES1(EMS1)/EXTRA SPOROGE-
NOUS CELLS(EXS) (Canales et al. 2002; Zhao et al. 2002). Both theems1
andexsmutant alleles appear to lack the tapetal cell layer and produce an
increased number of PMCs (Fig. 1). This observation was further supported
by the results that the expression domain of the meiosis-specific markerSDS