Pollen tubes are a good model for the study of cell growth and morphogenesis because of their extreme elongation without cell division. Yet, knowledge about the genetic basis of pollen germination and tube growth is still lagging behind advances in pollen physiology and biochemistry. In an effort to reduce this gap, we have developed a new method to obtain highly purified, hydrated pollen grains of Arabidopsis through flowcytometric sorting, and we used GeneChips (Affymetrix, Santa Clara, CA; representing approximately 8,200 genes) to compare the transcriptional profile of sorted pollen with those of four vegetative tissues (seedlings, leaves, roots, and siliques). We present a new graphical tool allowing genomic scale visualization of the unique transcriptional profile of pollen. The 1,584 genes expressed in pollen showed a 90% overlap with genes expressed in these vegetative tissues, whereas one-third of the genes constitutively expressed in the vegetative tissues were not expressed in pollen. Among the 469 genes enriched in pollen, 162 were selectively expressed, and most of these had not been associated previously with pollen. Their functional classification reveals several new candidate genes, mainly in the categories of signal transduction and cell wall biosynthesis and regulation. Thus, the results presented improve our knowledge of the molecular mechanisms underlying pollen germination and tube growth and provide new directions for deciphering their genetic basis. Because pollen expresses about one-third of the number of genes expressed on average in other organs, it may constitute an ideal system to study fundamental mechanisms of cell biology and, by omission, of cell division.Pollen has been the subject of intense studies not only for its importance as the male partner in plant reproduction, but also as a model for the study of cell growth and morphogenesis in a broader sense (Feijó et al., 2001). Whereas early studies focused on pollen physiology and biochemistry (for review, see Mascarenhas, 1975), the last 20 years have been marked by increasing efforts to decipher the genetic basis of pollen development and functions (for reviews, see Scott et al., 1991;McCormick, 1993;Preuss, 1995;Taylor and Hepler, 1997; Franklin-Tong, 1999; Hepler et al., 2001;Lord and Russell, 2002). These efforts led to the identification of more than 150 pollenexpressed genes from more than 28 species (Twell, 2002). These genes encode proteins thought or known to be involved in pollen development, pollen germination, and pollen tube growth, as well as in interactions with the stigma/transmitting tissue or the female gametophyte. Many of these studies were conducted in Lilium sp. or in Solanaceae species, because their pollen is easily collected in sufficient quantities, and methods for in vitro germination are robust. However, the available genetic information for these species is limited. The availability of the genome sequence of Arabidopsis (The Arabidopsis Genome Initiative, 2000) and the concomitant increase in availab...

SummaryDespite much effort, a robust protocol for in vitro germination of Arabidopsis thaliana pollen has been elusive. Here we show that controlled temperatures, a largely disregarded factor in previous studies, and a simple optimized medium, solidified or liquid, yielded pollen germination rates above 80% and pollen tube lengths of hundreds of microns, with both Columbia and Landsberg erecta (Ler) ecotypes. We found that pollen germination and tube growth were dependent on pollen density in both liquid and solid medium. Pollen germination rates were not substantially affected by flower or plant age. The quartet1 mutation negatively affected pollen germination, especially in the Ler ecotype. This protocol will facilitate functional analyses of insertional mutants affecting male gametophyte function, and should allow detailed gene expression analyses during pollen tube growth. Arabidopsis thaliana can now be included on the list of plant species that are suitable models for physiological studies of pollen tube elongation and tip growth.

Fertilization in both animals and plants relies on the correct targeting of the male gametes to the female gametes. In flowering plants, the pollen tube carries two male gametes through the maternal reproductive tissues to the embryo sac, which contains two female gametes. The pollen tube then releases its two male gametes into a specialized receptor cell of the embryo sac, the synergid cell. The mechanisms controlling this critical step of gamete delivery are unknown. Here, data based on the new sirène (srn) mutant of Arabidopsis thaliana provide the first evidence for female control over male gamete delivery. Live imaging of fertilization shows that wild-type pollen tubes do not stop their growth and do not deliver their contents in srn embryo sacs.