A plant germline-specific integrator of sperm specification and cell cycle progression

Abstract

The unique double fertilisation mechanism in flowering plants depends upon a pair of functional sperm cells. During male gametogenesis, each haploid microspore undergoes an asymmetric division to produce a large, non-germline vegetative cell and a single germ cell that divides once to produce the sperm cell pair. Despite the importance of sperm cells in plant reproduction, relatively little is known about the molecular mechanisms controlling germ cell proliferation and specification. Here, we investigate the role of the Arabidopsis male germline-specific Myb protein DUO POLLEN1, DUO1, as a positive regulator of male germline development. We show that DUO1 is required for correct male germ cell differentiation including the expression of key genes required for fertilisation. DUO1 is also necessary for male germ cell division, and we show that DUO1 is required for the germline expression of the G2/M regulator AtCycB1;1 and that AtCycB1:1 can partially rescue defective germ cell division in duo1. We further show that the male germline-restricted expression of DUO1 depends upon positive promoter elements and not upon a proposed repressor binding site. Thus, DUO1 is a key regulator in the production of functional sperm cells in flowering plants that has a novel integrative role linking gametic cell specification and cell cycle progression.

Figure 1. Expression of male germline-specific genes in wild type and duo1 pollen.

Expression of AtMGH3-H2B::GFP (A), AtGEX2-GFP (B), AtGCS1-AtGCS1::GFP (C) and DUO1-DUO1::mRFP (D) during wild type pollen development, observed with CLSM. Panels are numbered 1 (left) to 5 (right). For all markers, fluorescence is not detected in microspores (MS; Panel 1), a weak signal is detected in the germ cell during or soon after engulfment (early-BC; Panel 2), fluorescence increases in mid-bicellular pollen (mid-BC; Panel 3) and remains in tricellular (TC; Panel 4) and mature pollen (MP; Panel 5). (E–L) Expression of germline expressed genes in heterozygous duo1 plants. The percentage pollen showing GFP or RFP in sperm cells of wild type (WT) pollen or the single germ cell in cdka;1 and duo1 mutant pollen in plants homozygous for AtMGH3-H2B::GFP (AtMGH3, E), AtGEX2-GFP (AtGEX2, F), AtGCS1-AtGCS1::GFP (AtGCS1, G) and DUO1-H2B::mRFP (DUO1, H). Individual examples viewed by fluorescence microscopy in I to L. AtMGH3-H2B::GFP (I), AtGEX2-GFP (J) and AtGCS1-AtGCS1::GFP (K) are not expressed, or have reduced expression in duo1 pollen while DUO1-H2B::RFP (L) is expressed. Each image has a wild type pollen grain to the left and a duo1 mutant grain to the right (see lower DAPI images).

Figure 2. Ectopic expression of DUO1 results in expression of male germline specific genes.

(A) RT-PCR analysis of mDUO1, AtMGH3, AtGEX2, AtGCS1 and AtCycB1;1 expression in whole seedlings transformed with the mDUO1 cDNA (see methods) under the control of an estradiol inducible promoter grown on media without estradiol (−) or with estradiol (+). Histone H3 was used as a control. (B, C) Mature pollen grains showing AtMGH3-H2B::GFP expression specifically in sperm cells in the absence of LAT52-DUO1::mRFP (B), or in both the vegetative cell nucleus and sperm cells in the presence of LAT52-DUO1::mRFP (C). Left and right panels correspond to RFP and GFP signals viewed by CLSM.

Figure 3. AtCycB1;1 expression in developing pollen.

(A–F), pCDG-dependent GUS staining (upper panel) and DAPI staining (lower panel) in isolated spores: (A, B), unicellular microspores, (C, D, E), early, mid-and late bicellular pollen and (F), tricellular pollen. (G, H) The frequency of pCDG-dependent GUS staining in microspores and vegetative cells close to mitosis is similar in duo1 heterozygotes and wild type plants (G), whereas GUS staining in germ cells, is reduced by approximately half in duo1 heterozygotes, where 50% of the pollen is WT and the other 50% mutant (H). The stage of pollen development is indicated below each graph and the approximate time of mitosis is indicated by grey squares with a dashed line. (I) DUO1-AtCycB1;1 is able to partially complement the bicellular phenotype of duo1 pollen. The amount of tricellular pollen (T) increases and the amount of bicellular pollen (B) decreases when heterozygous duo1 plants are transformed with DUO1-AtCycB1;1 (n = 31 T1 lines) compared with plants either not transformed (n = 3 individuals) or transformed with control constructs AtMGH3-AtCycB1;1::GFP (n = 17 T1 lines) or LAT52-AtCycB1;1 (n = 17 T1 lines). Bars represent the average percentage of pollen with error bars showing standard deviation. (J) Germline markers are not activated in the complemented tricellular pollen. In non-complemented plants ∼50% of the pollen is tricellular (T) with marker expression and ∼50% is bicellular (B) without marker expression. When the bicellular phenotype is partially complemented by DUO1-AtCycB1;1, ∼10% of pollen is tricellular without marker expression, while there is a decrease in the amount of bicellular pollen. Bars represent the average percentage of pollen from 3–6 individual plants with the error bars showing standard deviation.

Figure 4. Male germline specificity of DUO1 does not depend on putative GRSF binding sites.

(A) Schematic of the DUO1 promoter region illustrating the mutagenized putative GRSF binding site. (B,C) Expression of H2B::GFP in pollen driven by the native (B) or mutagenized DUO1 (C) promoters. Top panels show GFP signal, lower panels show DAPI staining. (D) RT-PCR analysis of native and mutagenized DUO1 promoter activity in seedlings. PCR was conducted on cDNA from wild type plants (1), control plants transformed with a constituitive HistoneH3 promoter-H2B::GFP fusion (2), and plants transformed with the native (3), or mutagenized (4), DUO1 promoters driving H2B::GFP expression. The primers used were specific for GFP (upper panel) or native Histone H3 transcripts (lower panel). The native or mutagenized DUO1 promoters showed no sporophytic expression of GFP transcripts. (E) Schematic representation of the of the DUO1 promoter 5′ deletion series used to drive expression of H2B::GFP. The first four deletions, including deletion 3 in which the putative GRSF binding site is removed, showed a similar expression pattern to that of the full-length DUO1 promoter, with GFP signal only observed in sperm cell nuclei. The same expression pattern was observed in all independent lines examined (n). GFP expression was not observed in any transformants harbouring the shortest promoter fragment (deletion 5).

Figure 5. Regulatory events in plant male germ cell production and specification.

Model integrating the role of DUO1 and SCF^FBL17 in plant germ cell production and specification. The germline-specific DUO1 protein (blue) activates the expression of several germline specific proteins (red). In parallel, the CDKA inhibitors KRP6 and KRP7 (green) are expressed in the vegetative cell and germ cell after asymmetric division, where they inhibit CDKA activity and S phase progression. The F-box protein FBL17 is then transiently expressed in the germline and forms an SCF^FBL17 complex (blue) that targets KRP6/7 for proteasome dependent proteolysis, licensing S-phase progression (green arrow). Further germ cell cycle progression is controlled by the DUO1-dependent G2/M phase expression of the CDKA regulatory subunit AtCYCB1;1 (red). Thus, while SCF^FBL17 and DUO1 promote male germ cell proliferation at successive stages of the cell cycle, DUO1 integrates germ cell specification and division to ensure the production of functional twin sperm cells that are essential for double fertilization. Arrows indicate a requirement for the protein rather than direct binding.