Auxin is a key coordinative signal required for many aspects of plant development and its levels are controlled by auxin metabolism and intercellular auxin transport. Here we find that a member of PIn auxin transporter family, PIn8 is expressed in male gametophyte of Arabidopsis thaliana and has a crucial role in pollen development and functionality. Ectopic expression in sporophytic tissues establishes a role of PIn8 in regulating auxin homoeostasis and metabolism. PIn8 co-localizes with PIn5 to the endoplasmic reticulum (ER) where it acts as an auxin transporter. Genetic analyses reveal an antagonistic action of PIn5 and PIn8 in the regulation of intracellular auxin homoeostasis and gametophyte as well as sporophyte development. our results reveal a role of the auxin transport in male gametophyte development in which the distinct actions of ER-localized PIn transporters regulate cellular auxin homoeostasis and maintain the auxin levels optimal for pollen development and pollen tube growth.

Sexual plant reproduction depends on the production and differentiation of functional gametes by the haploid gametophyte generation. Currently, we have a limited understanding of the regulatory mechanisms that have evolved to specify the gametophytic developmental programs. To unravel such mechanisms, it is necessary to identify transcription factors (TF) that are part of such haploid regulatory networks. Here we focus on bZIP TFs that have critical roles in plants, animals and other kingdoms. We report the functional characterization of Arabidopsis thaliana AtbZIP34 that is expressed in both gametophytic and surrounding sporophytic tissues during flower development. T-DNA insertion mutants in AtbZIP34 show pollen morphological defects that result in reduced pollen germination efficiency and slower pollen tube growth both in vitro and in vivo. Light and fluorescence microscopy revealed misshapen and misplaced nuclei with large lipid inclusions in the cytoplasm of atbzip34 pollen. Scanning and transmission electron microscopy revealed defects in exine shape and micropatterning and a reduced endomembrane system. Several lines of evidence, including the AtbZIP34 expression pattern and the phenotypic defects observed, suggest a complex role in male reproductive development that involves a sporophytic role in exine patterning, and a sporophytic and/or gametophytic mode of action of AtbZIP34 in several metabolic pathways, namely regulation of lipid metabolism and/or cellular transport.

KEY MESSAGE : bZIP TF network in pollen. Transcriptional control of gene expression represents an important mechanism guiding organisms through developmental processes and providing plasticity towards environmental stimuli. Because of their sessile nature, plants require effective gene regulation for rapid response to variation in environmental and developmental conditions. Transcription factors (TFs) provide such control ensuring correct gene expression in spatial and temporal manner. Our work reports the interaction network of six bZIP TFs expressed in Arabidopsis thaliana pollen and highlights the potential functional role for AtbZIP18 in pollen. AtbZIP18 was shown to interact with three other pollen-expressed bZIP TFs-AtbZIP34, AtbZIP52, and AtbZIP61 in yeast two-hybrid assays. AtbZIP18 transcripts are highly expressed in pollen, and at the subcellular level, an AtbZIP18-GFP fusion protein was located in the nucleus and cytoplasm/ER. To address the role of AtbZIP18 in the male gametophyte, we performed phenotypic analysis of a T-DNA knockout allele, which showed slightly reduced transmission through the male gametophyte. Some of the phenotype defects in atbzip18 pollen, although observed at low penetrance, were similar to those seen at higher frequency in the T-DNA knockout of the interacting partner, AtbZIP34. To gain deeper insight into the regulatory role of AtbZIP18, we analysed atbzip18/- pollen microarray data. Our results point towards a potential repressive role for AtbZIP18 and its functional redundancy with AtbZIP34 in pollen.

Research investigating the dynamics of male gametophyte (MG) development has proven to be challenging for the plant science community. Here we describe our protocol for separating Arabidopsis MG developmental stages, which is based on the centrifugation of pollen through a discontinuous Percoll concentration gradient. This Percoll gradient can be formed using a pipette, and it does not require a gradient maker. The purity of the isolated developing spores is as high as 70%, and in most separations it is well above 80%. Using this protocol, we can separate four different stages of pollen development-uninucleate microspore (UNM), bicellular pollen (BCP), tricellular immature pollen (TCP) and mature pollen grain (MPG). The duration of the separation procedure, excluding the cutting of flower inflorescences, is 6 h. This is reduced to 4 h when using a vacuum cleaning method to remove the MPGs before the Percoll density separation.

Male gametophyte development leading to the formation of a mature pollen grain is precisely controlled at various levels, including transcriptional, post-transcriptional and post-translational, during its whole progression. Transcriptomic studies exploiting genome-wide microarray technologies revealed the uniqueness of pollen transcriptome and the dynamics of early and late successive global gene expression programs. However, the knowledge of transcription regulation is still very limited. In this study, we focused on the identification of pollen-expressed transcription factor (TF) genes involved in the regulation of male gametophyte development. To achieve this, the reverse genetic approach was used. Seventy-four T-DNA insertion lines were screened, representing 49 genes of 21 TF families active in either early or late pollen development. In the screen, ten phenotype categories were distinguished, affecting various structural or functional aspects, including pollen abortion, presence of inclusions, variable pollen grain size, disrupted cell wall structure, cell cycle defects, and male germ unit organization. Thirteen lines were not confirmed to contain the T-DNA insertion. Among 61 confirmed lines, about half (29 lines) showed strong phenotypic changes (i.e., ≥ 25% aberrant pollen) including four lines that produced a remarkably high proportion (70-100%) of disturbed pollen. However, the remaining 32 lines exhibited mild defects or resembled wild-type appearance. There was no significant bias toward any phenotype category among early and late TF genes, nor, interestingly, within individual TF families. Presented results have a potential to serve as a basal information resource for future research on the importance of respective TFs in male gametophyte development.