水稻花粉結鈣激活酵素OSCK1在花粉萌發中扮演重要的角色

Abstract

高等植物的成熟花粉在乾燥環境下自花藥釋出，必須落在可以相容的雌蕊柱頭上，花粉開始萌發，花粉管突出並伸長穿入柱頭，兩個精細胞核因此可分別與卵細胞及極核結合，完成授粉。已知鈣離子是調節花粉管生長的重要因子之ㄧ，卻很少知道在此過程中鈣離子的下游受動器為何，或鈣離子又是如何調節花粉管生長的機制？由於結鈣激活酵素（calcium dependent protein kinases，CDPKs）會受到鈣離子調控，故成為可能的候選基因。本實驗室過去曾在花粉成熟時期篩選到一個CDPK基因，命名為OSCK1，推測其功能與調控花粉發育、花粉管萌發或伸長相關，欲利用轉殖植物以了解OSCK1在花粉扮演的角色。實驗設計利用具有花粉專一性的Zm13啟動子，分別驅動OSCK1或其他三種突變型基因於水稻中大量表現，三種突變型基因分別可產生無法附著於細胞膜的G2A蛋白質、失去激活酵素活性的CI蛋白質、及不受鈣離子控制並且永遠具有酵素活性的CA蛋白質，與載體p35ST轉殖株相比較，發現各轉殖株的營養生長與植株外觀均無明顯差異。以掃描式電子顯微鏡觀察各轉殖品系之花粉形態，發現野生型與G2A轉植株花粉形態多正常，但CI與CA轉植株則多出現皺縮之花粉。檢查受粉後柱頭上的花粉萌發情形發現，CI轉殖株柱頭上雖佈滿花粉，卻多不萌發，且它的in vitro花粉萌發率與稔實率亦顯著低於p35ST轉殖株。因此提出OSCK1的功能可能影響花粉發育與後續之花粉管萌發。 為了進一步了解OSCK1如何影響花粉管萌發，利用基因槍將融合GFP之OSCK1質體送入正在萌發的百合花粉管中，進行暫時性基因表現分析。由於實驗室過去曾利用酵母菌雙雜交法在成熟花粉中篩選出與OSCK1互相結合的蛋白質，稱為OIP30，故此實驗除了送入單一OSCK1蛋白質表現外，亦觀察共同轉殖OIP30後，兩種蛋白質在次細胞分布位置的改變。單獨分析OSCK1或其突變型蛋白質明顯可知細胞膜及細胞質均有訊號，而G2A蛋白質則完全分布在細胞質中，因此判別OSCK1蛋白質主要是藉由荳蔻酸基附著於細胞膜上的；另一方面，單獨分析OIP30發現僅細胞質具有訊號，故OIP30明顯為單純的細胞質蛋白質。有趣的是當OIP30與OSCK1共同存在時，部份的OIP30蛋白質會結合在細胞膜上，因此提出在活體細胞中，OSCK1與OIP30可產生蛋白質間的交互作用。由於序列比對結果得知OIP30可能為一解旋酵素，預期在細胞核中方能執行它的酵素活性，故利用DAPI染劑追蹤被暫時性轉殖的百合花粉細胞核。結果顯示當OIP30/EYFP與G2A/ECFP或與CA/ECFP共同轉殖時，有部份的OIP30訊號會位於細胞核內。根據上述實驗結果，我們判斷花粉中OIP30可能是OSCK1的下游受質蛋白質，而前述轉殖水稻中花粉無法萌發的特殊性狀，可能是因不具酵素活性的CI蛋白質大量表現時，阻礙了正常OSCK1將OIP30磷酸化或送入細胞核的過程，因此抑制花粉管的萌發。

Pollen is released from the anther in a dehydrated state. Upon contact with a compatible stigma, pollen will germinate and pollen tube elongate through the style in order to have two sperm nuclei fused with the egg cell and the central cell, respectively, known as the double fertilization. Calcium is a key regulator of pollen tube growth, but little is known concerning the identity of downstream effectors of Ca2+ in this process or precisely how Ca2+ dynamics modulate the tip growth machinery. One class of likely candidates for such modulators is the calcium-dependent protein kinases (CDPKs). As OSCK1 (Oryza sativa CDPK 1) gene was previously identified to be expressed predominantly in the mature pollen of rice, it may involve in control of pollen development or tube elongation. Transgenic plant analysis was employed to investigate roles of OSCK1 in this study. Wild-type or mutated OSCK1 genes including G2A (non-membrane form), CI (catalytically inactive), or CA (constitutively active), driven by the maize pollen-specific promoter Zm13, were constructed and transformed into rice. No specific phenotypes were observed at the vegetative stage for all transgenic plants. Pollen morphology examined by scanning electron microscope was found to be basically normal for OSCK1 and G2A transgenic lines. However, severly shrunken pollen was found in various CI and CA lines. Interestingly, examination of stigma after pollination revealed that, among abundant pollen remained on stigma, almost none germinate in one CI transgenic line. Moreover, its in vitro pollen germination rate and fertility rate was significantly lower than that of control. These data indicate that OSCK1 may affect pollen development and consequencly, pollen germination processes. To further address how OSCK1 affect the pollen germination process, plasmids which can overproduce GFP-fused OSCK1 proteins were bombarded into the germinating lily pollen. The bombardments were carried out either alone or together with plasmid that encodes OIP30, which is a putative downstream substrate for OSCK1 previously identified by a yeast two-hybrid screen. Except for the OSCK1-G2A protein which remained in cytosol, wild-type and all other mutated OSCK1 proteins were mainly localized on the plasma membrane when bombarded alone, demonstrated that OSCK1 was bound to membrane via myristoylation. As predicted according to the sequence information, most of the OIP30 signals were in cytosol when expressed alone. However, its distribution was changed to be partially membrane-bound when co-bombarded with OSCK1, suggesting their protein-protein interactions in vivo. Since OIP30 is likely a helicase, it is expected to execute its enzyme activity in the nucleus. We therefore use DAPI to trace nucleus in the bombared pollen. Among various samples examined, OIP30-EYFP, when co-bombared with OSCK1-G2A or OSCK1-CA, were found to be partially located within nucleus. We conclude that OIP30 may be a downstream substrate for OSCK1 in pollen. Overexpression of the catalytically inactive OSCK1-CI protein in the rice pollen may interfere the phosphorylation or nucleus-import of OIP30 by OSCK1, therefore impede the pollen germination process.