An ankyrin repeat-containing protein, characterized as a ubiquitin ligase, is closely associated with membrane-enclosed organelles and required for pollen germination and pollen tube growth in lily

Abstract

Exhibiting rapid polarized growth, the pollen tube delivers the male gametes into the ovule for fertilization in higher plants. To get an overall picture of gene expression during pollen germination and pollen tube growth, we profiled the transcription patterns of 1,536 pollen cDNAs from lily (Lilium longiflorum) by microarray. Among those that exhibited significant differential expression, a cDNA named lily ankyrin repeat-containing protein (LlANK) was thoroughly studied. The full-length LlANK cDNA sequence predicts a protein containing five tandem ankyrin repeats and a RING zinc-finger domain. The LlANK protein possesses ubiquitin ligase activity in vitro. RNA blots demonstrated that LlANK transcript is present in mature pollen and its level, interestingly contrary to most pollen mRNAs, up-regulated significantly during pollen germination and pollen tube growth. When fused with green fluorescent protein and transiently expressed in pollen, LlANK was found dominantly associated with membrane-enclosed organelles as well as the generative cell. Overexpression of LlANK, however, led to abnormal growth of the pollen tube. On the other hand, transient silencing of LlANK impaired pollen germination and tube growth. Taken together, these results showed that LlANK is a ubiquitin ligase associated with membrane-enclosed organelles and required for polarized pollen tube growth.

Figure 1.

Predicted structure of LlANK protein, generated by the SMART program (http://smart.embl-heidelberg.de/).

Figure 2.

Autoubiquitination of the LlANK fusion protein. Complete in vitro ubiquitination assays contained recombinant human 6× His-tagged E1 enzyme, recombinant Arabidopsis E2 enzyme GST-AtUBC8, recombinant GST-LlANK, and ubiquitin. Assay products were analyzed with western blot using either anti-ubiquitin or anti-GST antibodies. Omission of GST-AtUBC8 or GST-LlANK protein from the assay, as indicated above the lanes, resulted in a loss of protein ubiquitination. Asterisk (*), GST or GST-fusion fragments presumably due to partial proteolysis during protein purification and incubation.

Figure 3.

RNA gel-blot evaluation of LlANK expression during pollen germination and pollen tube growth. DH, Dehydrated; RH, rehydrated; PG, pregermination; GM, germinating; TG, tube growth. Bottom, 28 and 18 s rRNA.

Figure 4.

Cellular localization of transiently expressed LlANK-GFP fusion protein in lily pollen. A, Diagrammatic representation (not to scale) of both GFP control and LlANK-GFP fusion constructs. 35S-35S, Constitutive double-CaMV-35S promoter. B, LlANK-GFP associated with punctate or granular structures in pollen tube. C, Apical region of pollen tube free of the highlighted structure (asterisk (*), pollen apex). D, LlANK-GFP associated with a generative cell (Top, green fluorescence viewing; Bottom, generative nuclei visualized with Hoechst 33258). GFP control (left) and LlANK-GFP (right) images are juxtaposed for comparison. B and C are confocal images. Bar = 10 μm.

Figure 5.

Overexpression of LlANK in lily pollen. A, Diagrammatic representation (not to scale) of both GFP and LlANK-GFP constructs. Zm13, Pollen-specific strong promoter. B, Control GFP evenly distributed in pollen tube. C, Moderately expressed LlANK-GFP associated with granular structures. D, Increasing amount of LlANK-GFP DNA led to abnormal pollen tube growth. Bar = sd. Data were from three independent bombardments. Morphological abnormalities included budding (E), membrane invagination (F), tip ballooning (G), and tube swelling (H). In E to H, green fluorescence (left) and their corresponding transmitted images (right) are juxtaposed. Bar = 10 μm.

Figure 6.

Transient down-regulation of LlANK in lily pollen. A, Diagrammatic representation (not to scale) of both ihp and ihpLlANK constructs. Zm13, Pollen-specific strong promoter; pdk, a splicable intron. B, From one single pollen of different stage, LlANK, actin, and an exogenously applied synthetic tet transcript could be simultaneously amplified using conventional RT-PCR. RH, Rehydrated; PG, pregermination; GM, germinating; TG, tube growth. C, Quantitative real-time PCR evaluation of single pollen demonstrated the down-regulation of LlANK expression in ihpLlANK-transformed pollen (Student's t test, P = 0.01). Data are presented as the ratio in expression level normalized to tet and relative to untransformed pollen. D, Down-regulation of LlANK reduced in vitro pollen germination significantly. E, Effect of LlANK down-regulation on average tube length (Student's t test, P = 0.0046). For each pair of juxtaposed columns, data were from a same bombarded population. Numbers of grains/tubes scored are shown beneath each column. Bars = sd.