Human Trabecular Meshwork and Hepatocyte Growth Factor Authors: Aparna Rao Published: 2026 We read with interest the recent article by Du et al.1 investigating hepatocyte growth factor (HGF) as a mediator of trabecular meshwork stem cell (TMSC) resilience and paracrine protection in glaucoma. The study explores if TMSCs can survive in vitro in a glaucomatous milieu (TGF stress in this paper) while protecting neighboring trabecular meshwork (TM) cells through secreted factors. Although the work gives useful insights to glaucoma research, especially on stem cell-based therapies, we welcome clarification on a few methodological concerns that need consideration and cautious interpretation of results. The study is well designed to address the question under concern using simple molecular methodologies like MTT viability assays, immunofluorescence, Western blotting, RNA sequencing, and cytokine array profiling. The authors have used cells from six independent human donors, which confers reproducibility. Validation of results with HGF receptor inhibitor (SGX-523) provides robust pharmacological evidence of causality. The authors used cells between passages 3 and 6; it is known that primary TM cells develop alternate signatures form passage 3 to 5.2 A stratified analysis or explicit statement of passage-matched comparisons would be helpful to understand the effect of inclusion of passage 6 in some experiments, given the chronic TGF exposure. We note that the senior author herself is part of recommendation committee for human TM cell culture, which suggests not to use beyond passage 7; it may be worth validating if different results are derived with use of passage 5 or lower, especially for the chronic glaucoma model. Donors were aged 43 to 69 years, a range spanning significant differences in TM cellularity, stem cell number, and oxidative stress tolerance. For these primary human TM experiments, it could be useful to know if donor age was adjusted for in the models used. Given that TMSC number declines with age and is associated with glaucoma onset, it may be worthwhile to explore if this could in any way confound results in this study (say studying cell signatures at passage 3 vs. passage 5 or higher). Further, it clarified if same the batch and passages were used for the same set of experiments (three different donors for MTT, cytokines, etc). This method can be mentioned to denote clearly which batches and passages were used for which experiments, because these variables have a direct impact on interpretation of results. The transcriptomic analyses were based on cells from three independent donors. Although this is acceptable for exploratory RNA sequencing, the authors have used an existing database (n = 5) to possibly offset these, whose donor details can be shared. It would be worth acknowledging this as a potential source of variation when drawing pathway-level conclusions (e.g., gene set enrichment analysis, Venn diagram intersections). The in-house study cell dataset versus the existing dataset may differ in cell signatures, and the generalizability of the transcriptomic findings could perhaps be framed more conservatively. The identification of HGF as the key differentially secreted cytokine rests largely on membrane-based cytokine array results, which are semiquantitative and subject to antibody cross-reactivity and signal saturation. Given that HGF enrichment in the TMSC secretome is the central mechanistic claim, independent quantitative confirmation may add credence, using perhaps ELISA.3,4 The authors rightly acknowledge that HGF is a “key, but partial, mediator” of TMSC–secretome protection. However, the article could quantify what fraction of the TMSC-Scr protection is attributable to HGF versus other secreted factors, given the fact that this was secondary to TGF exposure (different triggers, different stress response). The dose of recombinant HGF used (10 ng/mL) may not reflect the actual concentration contributed by HGF within the secretome, making direct comparisons between recombinant HGF and TMSC-Scr difficult to interpret mechanistically. An HGH immunodepletion experiment and assessing residual protection could establish the relative contribution of HGF; this could be added as a future directive by the authors, or perhaps added in the Discussion. SGX-523 was used to inhibit the HGF receptor (MET). Yet, could the authors clarify whether 100 nM is selective for MET in TM cells, and whether on-target engagement was confirmed through MET phosphorylation assays? The authors could then discuss if it is possible, therefore, that these results provide only circumstantial evidence of the protective effects of TMSC-Scr through HGF. The authors note that n-acetylcysteine fails to suppress connective tissue growth factor (CTGF), suggesting a reactive oxygen species–independent Smad axis, yet neither rHGF nor TMSC-Scr fully abolishes CTGF upregulation. Understanding why CTGF remains partially refractory to these treatments is of importance and the authors could include these points in the discussion. The authors propose TMSC-Scr and HGF as “promising candidates for cell-free therapies in glaucoma.” HGFs have other effects too, including mitogenic and angiogenic properties that need careful discussion before translational therapeutics are contemplated.5 The authors' own stated next steps (anterior segment perfusion cultures and mouse ocular hypertension models) are necessary. The discussion can be more precise and could convey caution regarding therapeutic readiness. The figure legends are comprehensive, but the authors could share some key numerical results which are missing (e.g., fold-changes in reactive oxygen species or TMRM intensity between groups); this would avoid inference of magnitudes from bar graphs alone and help the readers understand the results better. In Figure 1, the authors could add densitometry for OCT4 to the collage.