When Plastidial HXK3 Is Lost, Pollen Development Collapses — A Reproducible Foundation Reveals a New Link Between Sugar Metabolism and ROS

In July 2026, a study published in The Plant Journal reported that disrupting the plastidial hexokinase CsHXK3 in cucumber severely impairs pollen development. CRISPR/Cas9 loss of CsHXK3 reduced sugar and starch accumulation and prevented normal pollen maturation. More interestingly, the phenotype did not stop at carbohydrate metabolism: hydrogen peroxide (H₂O₂) levels in the anther declined, and the timing of programmed cell death (PCD) in the tapetum was delayed.

Original paper: Lv L, Zhang L, Li H, et al. Loss of the plastidial hexokinase CsHXK3 triggers metabolic starvation and ROS imbalance leading to pollen abortion in cucumber. The Plant Journal. 2026;127:e71062. PubMed: https://pubmed.ncbi.nlm.nih.gov/42531135/ Full text on PMC: https://pmc.ncbi.nlm.nih.gov/articles/PMC13422659/

If you are already familiar with plant hexokinases, the first part of the story is not especially surprising. Knocking out an HXK can disturb sugar metabolism, and impaired carbohydrate metabolism can compromise pollen development. What makes this paper interesting is that it reconfirms a biological framework that has already been built across previous studies, using another plant and another HXK, and then extends that framework toward a new connection with ROS and tapetal PCD.

The connection between HXK and pollen was already visible in rice

In 2020, loss of rice OsHXK5 was shown to impair starch synthesis and utilization in pollen, reduce pollen germination and pollen tube growth, and compromise male fertility. That study did not rely on a single mutant. The phenotype was reproduced in Tos17 mutants and multiple CRISPR/Cas9 lines. Moreover, a catalytically inactive OsHXK5-G113D protein failed to rescue the phenotype, indicating that in OsHXK5 at least, the catalytic function of hexokinase itself—not merely a putative sugar-sensing role—is important for pollen development.

Reference: https://pubmed.ncbi.nlm.nih.gov/31671177/

So the basic idea that HXK-dependent sugar phosphorylation matters for male reproduction did not begin with the cucumber study.

That is why I think the CsHXK3 paper is more interesting when read not as a completely new phenomenon, but as a study that strengthens an existing biological model through a different experimental system.

The experimental chain is one of the strongest parts of the paper

The route from CsHXK3 to the final phenotype is particularly well constructed.

This research group has been studying sugar supply in cucumber reproductive tissues for years. In 2015 they analyzed the pollen-specific hexose transporter CsHT1; in 2019 the sucrose transporter CsSUT1; and in 2024 the cell wall invertase CsCWIN3. Across these studies, disrupting sugar transport or the conversion of sucrose into usable hexoses caused defects in pollen germination, pollen tube growth, fertilization, or male fertility.

CsHXK3 sits further downstream in that series. In fact, CsHXK3 expression was reduced when CsCWIN3 was suppressed. That creates a natural next question: after sucrose is cleaved into hexoses, how are those sugars processed inside the reproductive cells?

The authors first examined localization and showed that CsHXK3 is plastidial. In anthers, it is detected in tissues including the tapetum, microspores, and vascular-associated regions. They then tested recombinant CsHXK3 biochemically and confirmed that it has hexokinase activity, with a strong preference for glucose over fructose. The reported Km was 0.1059 mM for glucose and 43.14 mM for fructose. Replacing the putative catalytic residue Ser174 with alanine nearly abolished HXK activity.

Then came the genetics. The authors selected two independent Cshxk3 frameshift lines generated by CRISPR/Cas9 and examined predicted off-target sites for unintended mutations. Both lines showed defects in pollen development. That is much more reassuring than building the entire argument around a single edited plant.

From there, the study expands through anther morphology, pollen viability and fertility, soluble sugars, starch, gene expression, transcriptomics, ROS, and tapetal PCD.

This progression matters.

It is not simply:

Find a gene. Knock it out. See abnormal pollen. Stop.

The paper confirms that the protein is biochemically an HXK, reproduces the phenotype genetically in independent edited lines, measures the expected metabolic consequences, examines tissue development, and then extends the analysis to transcriptomic and ROS responses. Multiple experimental approaches converge on the same biological problem from different directions.

Strictly speaking, this is not a replication study. But the combination of reproducibility between independent mutant lines, convergence across different assays, and consistency with previous rice HXK work makes the central framework unusually robust.

From that robust foundation, a new ROS branch appears

Loss of CsHXK3 reduced glucose, fructose, sucrose, and starch. Several sugar transport-related genes, including members of the CsSWEET family and CsSTP13, were also downregulated. Given what is already known about HXKs and carbohydrate metabolism, this part is relatively easy to anticipate.

But H₂O₂ was also substantially reduced in the mutant anthers.

Expression of the NADPH oxidase-related gene CsRBOHB, which contributes to ROS production, decreased, whereas several peroxidases involved in ROS removal increased. At the same time, tapetal PCD was delayed.

The tapetum nourishes developing pollen, but it cannot simply remain intact indefinitely. Normal pollen maturation requires the tapetum to undergo PCD at the appropriate developmental stage. ROS involvement in this process is already well established in plant reproduction. The CsHXK3 results therefore raise a more interesting possibility: plastidial carbohydrate metabolism and the ROS system controlling tapetal death may be functionally connected.

This is the genuinely new part of the paper.

The authors discuss a possible connection among CsHXK3-derived glucose-6-phosphate (G6P), the oxidative pentose phosphate pathway (OPPP), NADPH production, RBOH-dependent ROS generation, and tapetal PCD. Conceptually, the model can be written as:

Glucose → CsHXK3 → G6P → OPPP → NADPH → RBOH → ROS → tapetal PCD

If this connection is correct, HXK is doing more than simply converting sugar into a metabolically usable form. It would imply that cellular carbon status can influence the timing of developmental cell death through ROS metabolism.

That is a compelling hypothesis.

However, it is not fully demonstrated by this paper. The study did not directly trace flux through the OPPP into NADPH and onward into RBOH-dependent ROS production, and it remains unclear how loss of CsHXK3 leads to reduced CsRBOHB expression. So the appropriate conclusion is not that the pathway has been solved, but that the study reveals a plausible mechanistic bridge between sugar metabolism and ROS regulation.

A good paper also makes its remaining gaps obvious

There are still important experiments to do.

For example, the study does not complete a classical genetic complementation test showing that reintroduction of wild-type CsHXK3 rescues fertility in the Cshxk3 mutant. It also does not test whether the catalytically inactive CsHXK3-S174A protein can rescue the phenotype in planta.

The OsHXK5 work went further in this respect by using a catalytically inactive rescue construct. Doing the equivalent experiment with CsHXK3 would make it possible to distinguish more strongly between catalytic and signaling roles. Likewise, testing a CsHXK3 version lacking its plastid transit sequence could help determine whether plastid localization itself is essential for the reproductive phenotype.

In other words, this paper does not close the mechanism completely.

Instead, it reconfirms the parts that previous research has already made plausible and then defines the next experiments needed to extend the model.

“It reproduces what we already knew” is not a weakness

When reading a new paper, it is easy to focus almost entirely on what has never been reported before. But the CsHXK3 story highlights a different kind of scientific value.

Loss of OsHXK5 in rice disrupts pollen starch metabolism and male fertility. Disrupting sugar transport and sucrose cleavage in cucumber also disrupts reproductive development. Now, removing a plastidial HXK in cucumber again causes carbohydrate defects and male reproductive failure.

The genes are different. The species are different. The experimental systems are different.

Yet the results converge in the same direction.

That convergence matters because it makes the newly observed ROS and tapetal PCD phenotype harder to dismiss as a peculiar feature of one mutant line. Once the underlying HXK–carbohydrate–male fertility relationship is supported across systems, the additional ROS phenotype becomes a much more credible starting point for the next biological question.

Science does not always advance through a single paper that overturns everything before it. Often, a model is tested again from another angle, its reliable parts become more secure, and a new branch grows from that foundation.

The CsHXK3 paper is a good example of that kind of cumulative research.

The relationship between hexokinases and pollen development has become a little stronger. And beyond it now sits another question: where, exactly, does sugar metabolism connect to the timing of developmental cell death?

Paper information

Lv L, Zhang L, Li H, et al. Loss of the plastidial hexokinase CsHXK3 triggers metabolic starvation and ROS imbalance leading to pollen abortion in cucumber. The Plant Journal. 2026;127:e71062. First published 30 July 2026.

Note on nomenclature: this paper names CsaV3_3G045830 as CsHXK3, whereas a separate cucumber HXK family analysis published in January 2026 names the same gene CsHXK4. HXK numbering is therefore not consistent across the literature; gene IDs should be checked when comparing studies.