Protein markers reveal early polarity in Arabidopsis reproductive development

Finding direction in reproductive cells

Cell polarity means that different parts of a cell have distinct molecular identities. Proteins concentrated on one side can help organize directional transport, growth or unequal division. Much of what researchers know about plant polarity comes from the sporophyte: the diploid generation that forms the familiar flowering plant. Its much smaller, gamete-producing generation, the gametophyte, has been less thoroughly explored, apart from the specialized polarity of growing pollen tubes.

In a bioRxiv preprint posted on November 2, 2026, Samuel Koh Wee Han, Monique van Schie and colleagues at Wageningen University examine this less-charted territory in Arabidopsis thaliana. The study has not been certified by peer review. Using fluorescently tagged proteins and confocal microscopy, the team surveyed 14 proteins belonging to the SOSEKI family and an associated protein network during male and female reproductive development.

A crescent before pollen division

The male gametophyte begins as a single microspore. An asymmetric division produces a large vegetative cell and a smaller generative cell, which later divides to make two sperm cells.

Among the five SOSEKI proteins examined under their native expression controls, SOK1 showed a striking sequence of distributions. Before the first division, it accumulated in a crescent at the microspore’s periphery. The authors report that this pattern precedes nuclear migration, previously recognized as an early sign of polarization. After division, SOK1 outlined the generative cell and subsequently surrounded the sperm cells.

Comparison with a membrane marker supported localization to the peri-germ cell membrane, a membrane derived from the vegetative cell that encloses the germline compartment. However, the researchers could not exclude some signal coming from the closely adjacent germ cell membrane.

SOK1 also remained polarized in microspores with division defects in the gem1 mutant background. This supports the conclusion that establishing its polar domain does not require completion of the normal asymmetric division. It does not, by itself, show that SOK1 causes that division.

Two stages with different requirements

To ask whether SOK1’s behavior was unique, the researchers expressed the other four SOSEKI proteins using the SOK1 promoter, which controls where and when the gene is expressed. All four could form an early peripheral polar domain. Unlike SOK1, however, they did not retain membrane enrichment around the generative and sperm cells after division.

The result separates two targeting capabilities: forming the initial crescent and accumulating around the internalized germline. How the early domain relates physically to the later membrane remains unresolved; the authors propose live imaging to distinguish inheritance of the domain from dynamic rearrangement.

Several associated proteins showed similar developmental distributions. When SOK1 was examined alongside SIB1 or SICK, its polar accumulation appeared first. The authors interpret this as suggesting an upstream role for SOK1, while acknowledging that earlier, functionally relevant amounts of the other proteins could fall below detection limits.

Polarity inside the ovule

On the female side, the team found distinct protein patterns within the developing gametophyte and in surrounding sporophytic tissues. SOK1 marked the chalazal end of the developing embryo sac, while SOK5 later polarized in the two synergid cells. Other network proteins marked overlapping domains.

Broader experimental expression allowed SOK1 to label the female polar domains identified with other reporters. Expression specifically in the egg cell also produced polarized SOK1, whereas SOK5 remained mostly cytosolic. Together, these observations suggest a shared targeting system with protein-specific capabilities, rather than proving a universal mechanism for every polarity process.

A map, not yet a functional explanation

A comparison with the liverwort Marchantia found no polarized localization of the tested SOK and SICK proteins in egg cells. This contrasts with experimentally expressed Arabidopsis SOK1, but leaves open whether Marchantia eggs lack polarity or whether these particular proteins fail to reveal it.

The study’s central limitation is explicit: it documents polarity, not its developmental necessity. Mutant analysis is still needed to establish causal relationships and functions. The immediate contribution is a set of molecular landmarks that makes those questions more accessible.