Single-cell RNA sequencing of murine limbal epithelia reveals Gas1 as a novel stem/progenitor cell marker for the corneal epithelium

doi:10.21203/rs.3.rs-4921718/v1

Download PDF

Article

Single-cell RNA sequencing of murine limbal epithelia reveals Gas1 as a novel stem/progenitor cell marker for the corneal epithelium

https://doi.org/10.21203/rs.3.rs-4921718/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

The corneal epithelium is endowed with a rare population of stem cells that reside within the limbus, a circumferential transition zone that partitions the cornea from the conjunctiva, thus referred to as limbal epithelial stem cells (LESC). Despite the surge in investigations using single-cell RNA sequencing (scRNA-seq) of the ocular surface, a unifying marker(s) that distinguish these cells from their progeny is yet to be identified. We used a keratin (K)-14-driven lineage-tracing system and SmartSeq-2 single-cell transcriptomics in 5-60-week-old mice to interrogate the identity of limbal epithelia.

Four cell clusters were identified, derived from both Confetti⁺ and Confetti⁻ cells (clusters 0–3), with cluster 3 designated to harbor progenitor cells. We found one gene of interest in cluster 3, growth arrest-specific gene 1 (Gas1) coding for a cell-surface protein. PCR, flow cytometry and immunofluorescence disclosed this gene to be rarely expressed in limbal epithelial cells. Gas1 was also co-expressed with K14 in both young and old mice and upregulated following a mild mechanical debridement injury to the central cornea.

The cell-surface expression of this antigen can be used to identify, extract and enrich progenitor cells for downstream molecular investigations and for generating better-quality cell-based grafts to treat severe corneal disease.

Biological sciences/Cell biology

Biological sciences/Computational biology and bioinformatics

Biological sciences/Stem cells

Health sciences/Biomarkers

Cornea

corneal epithelium

single-cell transcriptomics

limbus

limbal epithelial stem cells

Most mammalian organs contain a population of adult stem cells (SCs) endowed with life-long replenishing activity for the tissues they are programmed to sustain. Under steady-state, they remain quiescent, thereby expending little energy. However, their activity is pronounced after encountering an injury or other perturbation [1]. The cornea is a good model system for such interrogations; its transparency and external location are advantageous features which facilitate the study of SC function during homeostasis and disease progression with minimally invasive microscopy techniques [2]. This tissue is indulged with SCs that are nestled in a basal location within the narrow annular limbal region that partitions the cornea from the conjunctiva; these cells are otherwise known as limbal epithelial SCs (LESCs) [2]. Two recent studies employed transgenic mice, with intrinsic fluorescence reporters and two-photon intra-vital imaging to characterize the limbal niche, segregating it into inner and outer compartments [3, 4]. Both studies demonstrated that LESCs within the outer sector remain dormant until the cornea is severely injured, after which they participate in tissue restoration [4], while those in the inner perimeter give rise to progeny that replenishes the corneal epithelium under steady-state [3, 4]. This led to the conclusion that the outer limbus serves as a SC pool whilst the inner limbus is tasked with maintaining corneal homeostasis [3].

Depleted or dysfunctional LESCs can arise due to various causes, including physical trauma and chemical exposure, or through genetic mutations or inflammation of mucus membranes [5]. This predisposes tissues to severe, non-resolving, corneal epithelial defects or instigates a compensatory pathological wound healing response otherwise known as ‘conjunctivalization’, in which an inflamed fibrovascular conjunctival pannus invades the corneal surface resulting in blindness [6]. Unfortunately, therapies ranging from simple medical interventions to more complex corneal transplantation surgery are not viable options. Instead, a SC-based strategy is necessary to replenish the depleted SC pool and re-invigorate the corneal surface to restore tissue function and sight [7].

Since the earliest reports of the presence of regenerating cells within the cornea [8], and three decades after the first limbal tissue transplantation succeeded in treating patients with LSCD [9], a bona fide marker for these cells is yet to be discovered. This is important for several reasons. Firstly, to unequivocally map the distribution of LESCs, which has been an area of contention. Secondly, as a diagnostic/prognostic tool for patients with suspected LSCD, and thirdly, to enable the isolation, expansion and transplantation of a population of cells that can potentially improve clinical outcomes. Such a marker(s) could also be a surrogate for determining ‘stemness’ during in vitro expansion of primary LESCs or a gauge of lineage correctness for cells that are not of a corneal ancestry but reprogrammed to behave as if they were [10]. Notably, many methods have been employed to identify putative LESCs, including through the use of morphology [11], DNA binding agents to detect slow-cycling label-retaining cells [1], and laser capture microdissection to splice regions of the ocular surface for subsequent transcriptomic interrogation [12]. Fluorescence-activated cell sorting (FACS) and magnetic-activated cell sorting (MACS) have also been employed, although the location from which sorted cells originate is difficult to ascertain. To address this limitation, multiple transgenic approaches have been adopted to lineage trace limbal epithelia within the cornea [2, 3]. These paradigms have further incorporated single-cell transcriptomics (scRNA-seq) of labelled cells to unravel novel genetic signatures for LESCs [13, 14]. However, it remains contentious as to whether the cells being studied are bona fide progenitors.

Over the past decade, our laboratory has accrued evidence indicating the robustness of K14 as a marker of fetal [15] and adult [16] human and murine [17] LESCs. However, the cytosolic expression of K14 renders it inappropriate as a tethering tool for cell isolation via FACS or MACS, thereby preventing comprehensive omics assessments. Therefore, we adopted a similar inducible transgenic approach to that of Sartaj and colleagues [18] with a focus on K14⁺ cells. Using the K14CreERT2-Confetti (Confetti) system, multi-colored fluorescence derived from K14⁺ (Confetti⁺) epithelia were used as a surrogate marker for limbal progenitor cells. Confetti positive and negative cells were FACS sorted and scRNA-seq performed using the SmartSeq-2 platform to identify candidate genes that were subsequently validated at the mRNA and protein levels. One candidate gene piqued our interest, namely growth arrest-specific gene 1 (Gas1). Its localization and gene expression across the ocular surface is consistent with it being a marker of corneal precursor cells. Given its involvement in eye development [19], cell surface expression [20], and role in supporting SCs within other tissue compartments including muscle [21], neural [22] and skin [23], we anticipate using this marker for enrichment of stem/progenitor cells to enhance graft potency and improve clinical outcomes for individuals needing a SC therapy.

Cluster and gene ontology analysis of K14CreER^T2-Confetti mouse corneal tissue

Given that LESCs represent a subpopulation of epithelial cells residing within limbus, Confetti⁺ or Confetti⁻ (i.e., colored and non-colored) cells were isolated from mice aged between 6-60-weeks of age, after which they underwent scRNA-seq. Data was normalized using the SCTransform function of Seurat, and a standard workflow pipeline employed using the runPCA, runUMAP, FindNeighbours and FindClusters functions to generate and visualize cell clusters with Uniform Manifold Approximation and Projection (UMAP). The UMAP analysis identified four clusters (Fig. 1a), although not surprisingly, they were not clearly discriminated, due to the degree of similarity expected and modest number of cells analyzed (Fig. 1a).

When mouse age was considered, cells from each time-point were represented in all clusters (Fig. 1b and 1d). However, there were fewer cells from 6-week-old mice (T1) present in cluster 3, and a significant proportion of cells in cluster 1 were from 18-week-old mice (T2), while cells from T2 were under-represented in cluster 0 (Fig. 1d and Supplementary Table S3). Interestingly, most cells from T1 were evenly distributed throughout clusters 0–2 (Fig. 1d), which may reflect age, size, and differentiation state of the developing mouse cornea. Notably, when only Confetti status was considered (i.e., Confetti⁺ or Confetti⁻), epithelia that were positive and negative for transgene expression were found in each cluster (Fig. 1c). However, there was a significantly higher proportion of Confetti⁺ cells in clusters 2 and 3, whilst Confetti⁻ cells were significantly more abundant in cluster 0 (p < 0.05) (Fig. 1e and Supplementary Table S4). Given our previous studies have demonstrated Confetti⁺ cells to be confined to the limbus, and that this region is the repository for corneal progenitor cells, these results demonstrate that Confetti⁺ (i.e., K14⁺) cells display a progenitor-like phenotype. Conversely, Confetti⁻ (i.e., K14⁻) cells are more likely to be differentiated corneal epithelia, indicating that the lineage tracing approach employed is a viable method to resolve the genetic signature of corneal epithelial progenitor cells.

Key marker genes in each cluster were identified using the FindAllMarkers function of Seurat. The distribution of the top 12 genes discriminating each cluster is presented as violin plots (Fig. 2a-2d). Gene ontology (GO) enrichment (Supplementary Table T5) and KEGG pathway analysis were performed to identify clusters. Genes in clusters 0 and 1 have functions associated with tissue development and maintenance, and cell motility, features that characterize transient amplifying cells and differentiated corneal epithelia, respectively (Fig. 2a and 2b) [24]. Cluster 2 was enriched for genes involved in immune processes and exhibited high expression of Muc4 and Muc15, consistent with a conjunctival identity [13] (Fig. 2c). Cluster 3 was enriched for genes involved in metabolism and cell cycling, consistent with early stem/progenitor cells [25]. Six of the top 12 genes identified were distinctly expressed in cluster 3, namely Gas1, Bambi, Fosb, Zfp296, CDC20, and CDCA7 (Fig. 2d).

The expression of previously detected marker genes was then examined to aid in identifying the clusters [25] (Fig. 2e and 2f). This analysis revealed increased expression of Lrig1 (a cell surface protein that interacts with epidermal growth factor receptor tyrosine kinases), and decreased Gjb4 and Gjb3 (connexins that co-ordinate epithelial cell-cell signalling) in cluster 3 [14] (Fig. 2e), an expression pattern previously suggested to represent undifferentiated corneal SCs [25]. Additionally, Notch1, Sox9 and Egr1 were elevated in cluster 3 (Fig. 2g), all of which have been associated with SC identity [26–28]. Surprisingly, K14 mRNA was highly expressed in all clusters (Fig. 2g), although this protein is post-translationally regulated, the functional protein is likely not active in all clusters. K13 and Muc4 genes were predominately expressed in cluster 2, thereby identifying this cluster as bearing conjunctival cells (Fig. 2f). Although K19 has also been suggested as a conjunctival marker, only 10 cells in our dataset expressed this gene. Based on these data, we posited that cluster 0 predominantly comprised late transient amplifying cells (TACs), cluster 1 contained differentiated epithelia, cluster 2 supported conjunctival epithelia, and cluster 3 incorporated stem/early TACs.

Validating novel candidate markers of LESCs identified by scRNA-seq

We focussed on genes predominantly increased in cluster 3 (putative SC population) and those expressed on the cell surface, as this was deemed advantageous in future downstream isolation and cell-sorting strategies. One gene was found to fit this criterion, namely Gas1, which is primarily involved in development, differentiation and cell cycling activities [29]. Our initial validation was performed by RT-qPCR and flow cytometry (Fig. 3) to confirm gene and protein expression across the ocular surface. K12, K13, and K14 genes were expressed to varying degrees (Fig. 3a-3c). K12, a corneal epithelial marker, displayed significantly higher expression in the central cornea compared to limbal and conjunctival epithelia. Interestingly, K14 was detected at similar levels across the ocular surface (Fig. 3b), and similarly to that displayed in violin plots (Fig. 2g). However, K14 mRNA is translationally regulated depending on its location [30]. In contrast, K13 was highly expressed in conjunctival, and to a lesser extent, limbal epithelia (Fig. 3c).

Having defined the tissue boundaries with well-accepted marker genes, Gas1 was validated (Fig. 3d), together with a recently identified limbal marker, Lrig1 (Fig. 3e) [31]. Gas1 mRNA had similar expression in the central cornea and limbus but was significantly decreased in the conjunctiva (p < 0.01). As demonstrated in previous reports, Lrig1 was expressed at significantly higher levels in the limbus, compared to the cornea (p < 0.05) and conjunctiva (p < 0.001). Next, flow cytometry was performed on live cells from freshly disaggregated mouse corneas to determine the proportion of cells expressing these markers (Fig. 3f-3h). To this end, Gas1 and Lrig1 were expressed in 1.5 ± 0.63% and 3.6 ± 1.9% of cells respectively, implying they represent a rare population of the total cells analyzed.

Localization of Gas1 in the mouse cornea

Whole flat-mounted corneas from 15-18-week-old mice provided a holistic spatial appreciation for the location of Gas1 to other more established markers (Fig. 4). As expected, immunoreactivity for K12 was detected in the central and peripheral cornea, but not the limbus (Fig. 4a). Conversely, K14 expression predominated the limbus, with no staining in the peripheral or central cornea (Fig. 4b). Likewise, there was minimal reactivity for Gas1 in the central cornea, with most being confined to the limbal and peri-bulbar conjunctiva (if present) (Fig. 4c), with similar results recorded for Lrig1 (Fig. 4d). Notably, staining for Gas1 and Lrig1 appeared to overlap with that of K14, predominantly localizing to basal limbal epithelia. Quantitative image analysis disclosed that K12 was expressed in the central and peripheral cornea (Fig. 4a, histogram), while all other markers were highly expressed in the limbus compared to the peripheral and central regions (Fig. 4b-4d, histograms).

Once the spatial expression profile of Gas1 was established, co-expression with K14 was determined, as well as ascertaining whether this is temporally regulated. A key feature of SC markers is that they are maintained throughout life. To this end, we double immunostained Gas1 and Lrig1 with K14 in the corneas of mice aged 5-60-weeks (Fig. 5 and Supplementary Fig. 1). Although K14 was predominantly localized to the limbus [32] (Figs. 4 and 5), it was also sporadically detected throughout the central and peripheral cornea in younger mice (5 and 10-weeks of age) (Supplementary Fig. 1). By 20-weeks of age, K14 was confined to the limbal margin, and this pattern remained stable until 60-weeks. Likewise, Gas1 and Lrig1 immunoreactivity was found throughout the cornea (Supplementary Fig. 1a and 1b) and limbus (Fig. 5a-5d) in young mice, and by 20-weeks, expression remained stable and confined within the limbus and peripheral cornea until 60-weeks (Fig. 5e-5j and Supplementary Fig. 1e-1j). While being predominantly confined to the limbus and peripheral cornea, Gas1 and Lrig1 demonstrated wider circumferential staining compared to K14 (p < 0.0001 for each marker at all time points) (Fig. 5K). These results highlight that the identified markers display overlapping expression patterns, with similar distribution according to age and in relation to K14.

Expression of identified SC markers following central corneal injury

Mice aged 8 and 60-weeks were inflicted with a 2 mm central epithelial injury, then followed for 5-days to record any spatiotemporal change in Gas1 expression (Fig. 6a). Increased Gas1 expression was noted in young (8-week-old) mice after injury in both the peripheral and central cornea compared to the contralateral control eye (p < 0.0001) (Fig. 6b-6e), although this effect was absent in the 60-week-old mice (Fig. 6f-6i). These observations are congruent with a previous report that demonstrated young mice display upregulated SC marker expression following injury, which was absent from older animals [33].

Herein, single-cell transcriptomics using a lineage tracing modality was used to examine gene expression in murine corneolimbal epithelial cells. Several novel genes were uncovered that potentially identify putative stem/progenitor cells in cluster 3, one of which, namely Gas1, was validated using qPCR, flow cytometry and immunostaining to ascertain spatial and temporal expression patterns across the ocular surface. This gene is upregulated in muscle, neural and skin stem cells (references please), highlighting its wide-ranging potential to mark a multitude of progenitor cells, including those with the limbus. Additionally, this gene codes for a surface protein that could be used to tether antibodies for cell purification to study their biology and produce higher quality, efficacious SC grafts for patients with LSCD.

Bulk RNA sequencing has been performed for over two decades on many cell types. However, this technique lacks the resolution necessary to identify rare cells or those with genes expressed at low levels [34]. First performed in 2009 on mouse blastocysts [35], scRNA-seq has been widely used to study cell populations using a variety of methods (reviewed in [36] and [37]) and has even been explored, but not used, in diagnostics [38]. Several platforms are available to analyze the transcriptome of single cells, including SmartSeq-2 (utilized herein), 10X Chromium (10X Genomics, Pleasanton, CA), massively parallel RNA single-cell sequencing (MARS-seq) [39], the Fluidigm C1 systems SmartSeq [40] and mRNA Seq HT [37]. While Smart-seq v4 was recently used [41], the most common modality to evaluate the corneal epithelium has been 10X, likely due to its low cost, ability to analyze thousands of cells in a single reaction, and ease of sample preparation [13, 25, 42]. However, this approach precludes pre-selecting cells for downstream analyses, potentially omitting rare populations and thus suffering from selection bias, and the inability to detect more than 500–1000 genes/single cell [37]. The SmartSeq-2 technique we employed is a plate-based method that captures full-length transcripts [43], with the added capacity to produce ~ 1,000,000 reads/cell and the ability to amplify low-abundance genes [44]. A study that compared SmartSeq-2 and 10X with the same peripheral blood sample, discovered that while the former only captured ~ 10% of the number of cells compared to 10X, it produced 10-fold more reads/cell [44]. Nonetheless, Smartseq-2 suffers from inefficient reading of transcripts > 4 kb in length, preferentially amplifies abundant transcripts, and is limited by the number of cells that can be analyzed [45, 46]. As the majority of previous investigations on corneolimbal epithelia engaged either bulk sequencing [47–49] or 10X [13, 14, 25], notwithstanding its limitations, we employed, for the first time, Smartseq-2 to provide an alternative approach in an attempt to resolve an issue that continues to plague the field.

For the current investigation, potential marker genes for LESCs were considered based on two criteria. Firstly, they were expressed in cluster 3 (Fig. 2) which was deemed to harbor SCs [25–28], and secondly, the protein product of these genes was expressed on the cell surface. This prerequisite feature would facilitate downstream antibody tethering by either FACS or MACS [50] for isolating populations that contain more SCs, investigating their biology and function, and generating better quality, longer-lasting grafts for patients with LSCD. Certainly, it has been demonstrated that clinical success in patients with LSCD receiving a SC transplant is associated with grafts that comprise > 3% p63⁺ cells [51, 52]. With this in mind, several genes were identified, primarily residing in cluster 3 (Fig. 2). These genes included those involved in regulating embryonic (Zfp296) [53], mesenchymal (Fosb) [54], hematopoietic (Cdc20) [55], or cancer-associated (Cdca7) [56] SCs, although due to their nuclear or cytoplasmic localization, they were not pursued. Another gene of interest that was identified in cluster 3 was Bambi (Fig. 2). This gene has been identified in a subpopulation of mesenchymal [57] stem cells, and codes for a TGFβ pseudo receptor that acts as a signaling antagonist [58]. However, previous investigations have determined it to be a dispensable gene, as embryonic deletion does not affect fertility nor does it produce a discernible phenotype in mice [59].

Our approach also uncovered several genes that have been deemed by others as marking corneal epithelial progenitor cells. For example, Id1, a transcription factor that regulates bone morphogenetic protein (BMP) is associated with SC quiescence [60] and was expressed in cells from cluster 3. However, it has also been detected in immune cells [3, 61], thereby questioning its specificity or exclusivity for LESCs. Notably, both K14 [32, 62] and K15 [63] have been proposed as LESC markers, although they are also expressed by neighboring conjunctival epithelia [63, 64]. Whilst the Gpha2 protein was found to be expressed in the limbus by multiple investigators [13, 14, 65], our study detected the gene to be expressed in both Confetti⁺ and Confetti⁻ cells (data not shown) and in all clusters, suggesting it may not be as robust a marker as previously thought. Moreover, there are conflicting reports on the differentiation status of cells that express Gpha2, some describing them as being more mature, i.e., akin to an early TAC [14, 66], while others suggest they represent quiescent SCs [3, 13]. Similar to previous scRNA-seq investigations [42, 67, 68], LESC markers ABCB5 [69] and ABCG2 [70, 71] were not detected as highly expressed genes. Either they were not expressed in our data set (ABCB5), or they were found in multiple clusters (ABCG2) and were thus ruled out as they could not be used to accurately identify cells in cluster 3 (Fig. 1).

Reassuringly, the recently discovered LESC marker gene Lrig1 [25] was predominantly expressed in cluster 3, although it displayed low-level expression in clusters 0 and 1 (Fig. 2e). Importantly, this gene was not expressed in the neighboring conjunctival cluster 2. Numerous other limbal markers have also been identified in the conjunctiva, including K15 and K19 [63], ΔNp63 [72], and ABCG2 [73], suggesting they are not exclusive to cells in this region. Lrig1 was therefore used as a reference to confirm the limbal localization of novel genes (Figs. 3–5) and our scRNA-seq (Fig. 2) implies it identifies both stem/progenitor cells including TACs. Using qPCR, Lrig1 was predominantly expressed in limbal epithelia, and flow cytometry identified a rare population of Lrig1⁺ corneolimbal epithelial cells (Fig. 3). Furthermore, the Lrig1 protein was confined to the limbus and peripheral cornea, highlighting its expression in both the stem/progenitor cells and their early progeny (Figs. 3–5) [25]. Interestingly, both K14 and Lrig1 have been identified as progenitor cell markers in other organs such as skin [74] and gut [75]. These observations highlight that there may be a set of common markers that are potentially conserved in different, but related tissues.

We focused on Gas1 as this gene was predominantly found in cluster 3 and its protein product expressed on the cell surface (Fig. 2). To the best of our knowledge, this is the first report disclosing its expression in the normal mammalian cornea. Gas1 encodes a glycosylphosphatidylinositol (GPI)-anchored cell-surface protein [20, 76, 77] that has been described as a tumor suppressor [78, 79] due to its ability to negatively regulate the cell cycle. Interestingly, Gas1 modulates the growth and proliferation of muscle [21], neural [22], renal [80], and hair follicle [81] SCs. Relevant to the current topic, Gas1 plays a crucial role in eye development, with mutations in this gene resulting in micro-ophthalmia due to dysregulated cell proliferation in retinal pigmented epithelium and neural retina [19], and severe craniofacial deformations from dysregulated Sonic Hedgehog signaling [77].

We found Gas1 expression in cluster 3, which we identified as harboring putative LESCs, and flow cytometry confirmed that cells expressing this marker are indeed scarce, representing 1–3% of the total population (Fig. 3). Others have demonstrated a similar proportion of putative LESCs using label-retention [82, 83]. Furthermore, the stable limbal expression of Gas1 during aging, concurrent with K14 expression (Fig. 5), supports the proposition that at least some of these cells are bona fide corneal progenitors. However, some of these cells are also likely to be TACs, given that cells within the peripheral cornea also expressed this marker (Figs. 4 and 5). A key feature regarding SC markers is that they are expressed throughout life. Adult SCs appear to decline in both number and functional capacity during aging [84], as seen in neural [85], mesenchymal [86], hematopoietic [87] and muscle [88] compartments. This phenomenon is also apparent in both human [89–91] and mouse [92–94] corneas. Immunoreactivity for Gas1 in mouse corneas remained stable during ageing, suggesting that the limbus and its resident LESCs are relatively stable throughout life (from 20 to 60-weeks). While this provides reassurance that Gas1 is marking a long-lived population within the limbus, further functional experiments need to be performed to confirm this proposition.

Finally, Gas1 expression was found to be increased following a central corneal epithelial injury (Fig. 6), similar to other markers such as K15 [63], p63 and ABCG2 [95]. This likely reflects the function of the gene or protein being examined. For example, ABCG2 has anti-oxidative properties and is likely upregulated following injury to protect the proliferating cells from injury-induced oxidative stress [96]. Additionally, p63, and the ΔNp63α isoform in particular, facilitates the differentiation of corneal/limbal progenitor cells to TACs [97], a crucial element of the wound-healing process. Gas1 has specific roles in cellular growth arrest, and its upregulation during the post-injury period is likely to inhibit cellular proliferation and hyperplasia of the central cornea [98]. An interesting observation from the current investigation was that Gas1 was only upregulated in young (not old) mice following injury (Fig. 6). This data hints at an age-related decline in the ability to generate functional tissue following injury, a phenomenon observed previously [33]. While our group [32] and others [99] have observed a decline in LESC clones over age, this does not necessarily represent a decline in SC number, as neutral drift may reduce clone size over time [100]. However, while the number of SCs may remain stable during aging, our data represent a potential decrease in SC function in aged mice [93]. Interestingly, in humans, despite the reduction in limbal niche structures [89], no discernible decrease in SC markers was found in cultured cells, again demonstrating reduced function rather than the number of LESC during aging.

Using the convergence of linage tracing and single-cell transcriptomics [101] we identified Gas1 as a novel gene that can distinguish LESCs in the murine cornea. Gas1 was then validated at the mRNA and protein level, and co-localized with K14 within the limbus. Finally, this gene was upregulated following a mild central corneal epithelial injury. Our novel experimental design in concert with scRNA-seq provides a more targeted analysis of the identity of limbal epithelia. These data offer the foundation to further explore the functional consequences of these genes for maintaining a healthy cornea, what transpires during ocular surface injury and disease and whether cells bearing this signature can be exploited for downstream clinical use.

K14CreERT2-Confetti Mice

All procedures involving animals were approved by the UNSW Animal Care and Ethics Committee (approval numbers 17/81A, 20/59A) and performed in accordance with the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes. All authors complied with the ARRIVE guidelines. Confetti mice used in this study and the genotyping performed on them have been described elsewhere [2]. Briefly, the Confetti system generates irreversible and random expression of two fluorescent reporter proteins in cells that express K14. The fluorescent reporters do not provide any additional information regarding cell behavior; rather, in this instance they are used as a visual beacon for real-time in vivo monitoring [32]. For this investigation, they assisted in the identification of limbal epithelial cells. This feature facilitated their sorting without the need for exogenous labelling with antibodies. All cells expressing fluorescent proteins were considered equivalent.

Single-cell suspension for FACS

Male and female Confetti mice, aged 5, 17 and 59-weeks (n = 3/age), were administered Tamoxifen (5 mg/mL) (Sigma-Aldrich, St Louis, MI) in 10% ethanol (Sigma-Aldrich) and 90% sunflower seed oil (Sigma-Aldrich) by intraperitoneal injection on three consecutive days to induce expression of the multi-color fluorescent reporter. Seven days after the first injection, mice were euthanized by cervical dislocation, and their eyes immediately enucleated. Corneas, complete with the limbal ring, were dissected under a stereo microscope (Zeiss, Oberkochen, Germany), taking care to remove extraneous tissue. Samples were exposed to 0.5 U/mL Dispase II (Penzberg, Germany) in defined keratinocyte serum-free media (dKSFM) (ThermoFisher Scientific, Waltham, MA) for 2 hrs at 37°C, then trypsin-EDTA (ThermoFisher) for 10 min at 37°C to generate a single-cell suspension of corneolimbal epithelia. Cells underwent FACS for colored Confetti⁺ (CFP, GFP, YFP, RFP) and non-colored Confetti⁻ populations, taking care to only sort the epithelial population based on cell granularity, size and shape, and minimizing other cell types such as immune cells.

Library preparation, sequencing and transcriptomic analyses

Single-cell libraries were prepared using the SmartSeq-2 protocol [43], and sequencing was performed on the Illumina NextSeq platform (Ramaciotti Centre for Genomics, UNSW, Sydney, Australia). Mouse corneolimbal epithelia (n = 378 total), including Confetti⁺ and Confetti⁻ cells (n = 189/group) were segregated according to three time-points corresponding to mouse age (n = 63 cells/time-point) (T1 = 6, T2 = 18 and T3 = 60-weeks), then sequenced and analyzed. Reads were quantified with Salmon (v0.9.1) [102] using the mouse transcriptome GRCm38 to create an index before quantification. Read counts ranged from 288,000-1.4 million/cell with a median of 778,000/cell. The number of genes per cell ranged from 546–7,246 with a median of 4,758. Reads were summarized according to gene level using the R package Tximport [103]. The R package Seurat (v3.1.4) [104] was used to filter and normalize data, after which clustering and differential expression analysis were conducted. Cells with > 5% mitochondrial content and those with < 2000 genes were then removed, and the number of cells analyzed for Confetti⁺ at T1, T2, and T3, and Confetti⁻ at T1, T2, and T3 were 62, 55, and 41, and 61, 60 and 59, respectively. After filtering and normalization, 14,343 genes were identified across the data set in 338 cells.

Quantitative polymerase chain reaction (qPCR) analysis

Confetti mice (15-18-weeks-old) were euthanized by cervical dislocation, their eyes (n = 24) enucleated, and briefly washed in PBS prior to dissecting a 2 mm diameter central corneal button, limbal rim, and conjunctival tissue from each. Epithelium from each region was dissociated by incubating samples in dKSFM containing 0.5 U/mL Dispase II (Roche) for 2 hrs at 37^oC. Total RNA was extracted and purified using the RNeasy Plus Micro Kit (Qiagen, Valencia, CA). A total of 24 eyes were used to create six replicates; an independent replicate consisted of n = 4 pooled samples. Specific mRNAs were analyzed and quantified by reverse transcription and real-time PCR using SuperScript™ IV and SYBR Select Master Mix (Thermo Fisher Scientific) in a QuantStudio™ 12K Flex System (Applied Biosystems, Foster City, CA). Several primer sets were used (Supplementary Table T1) and no template controls were included to confirm specificity. Change in expression was determined using the 2^− ΔΔCT method in which genes of interest were normalized to β-actin mRNA.

Flow cytometry

C57BL/6 mice at 6 weeks of age (n = 5/experiment) were euthanized by cervical dislocation, their eyes enucleated, and their corneas dissected, taking care to remove all extraneous tissues. Corneas were then incubated in dKSFM containing 2.4 U/mL Dispase II (Sigma-Aldrich) overnight at 4^oC [67]. Epithelia were removed, pooled and washed in PBS by centrifugation (1200 rpm for 5 mins) at room temperature (RT). Next, samples were incubated in trypsin (ThermoFisher) for 15 mins at 37^oC, with agitation every 5 min to ensure a single cell suspension was generated. Cells were washed in PBS/10% FBS, then in PBS, and collected following each wash by centrifugation (1200 rpm for 5 min). Finally, they were incubated in dKSFM at 37^oC for 1 hr to facilitate the reconstitution of cell surface proteins that may have been cleaved by the enzymatic disaggregation. Cells were incubated with an appropriate primary antibody (Supplementary Table S2) for 5 min at RT in PBS/1% BSA, washed in PBS/1% BSA with 0.005% sodium azide, collected by centrifugation (1200 rpm for 3 min), then incubated with an appropriate secondary antibody (Supplementary Table S2) in PBS/1% BSA for 20 min at RT. Samples were counter-stained with propidium iodide (Sigma-Aldrich), washed in PBS/1% BSA and resuspended in dKSFM. Analysis was conducted on a FACS Melody (BD Biosciences, Franklin Lakes, NJ) and data visualized using FlowJo software v10.8.1 (BD Biosciences).

Immunofluorescence and quantitative analysis of wholemount corneas

Whole corneas (n = 4–5/target antibody) were dissected from 5-to-60-week-old transgenic mice, fixed in 2% paraformaldehyde (PFA) (ThermoFisher Scientific) in PBS for 2 hr at RT and stored in PBS at 4^oC until use. Samples were permeabilized with 1% Triton X-100 (Sigma-Aldrich) in TBS (TBS-T) with 2% bovine serum albumin (BSA) (Sigma-Aldrich) for 4 hr prior to incubation with an appropriate primary or relevant isotype control antibody (Supplementary Table S2) in TBS-T/2% BSA for 3-days at 4°C. Corneas were washed four times in TBS-T/2% BSA for 1 hr at RT, then incubated with an appropriate Alexa Fluor®-conjugated secondary antibody (10 µg/mL, Life Technologies) (Supplementary Table S2) at 4°C for 2-days and nuclei counterstained with Hoechst 33342 (2.5 µg/mL, Thermo Fisher Scientific). Additionally, corneas from mice aged 5, 10, 20, 40, and 60-weeks (n = 4/time-point/target) were stained for K14 (Supplementary Table S2) using the same protocol. Next, they were flat-mounted in ProLong Gold® anti-fade mounting media (Life Technologies, Carlsbad, CA) and visualized with an LSM780 confocal microscope (Zeiss, Oberkocken, Germany).

After immunostaining, regions of interest (ROI) (425 x 425 µm) were identified in the central and peripheral cornea, and limbus (12 ROIs total/ cornea) from mice aged between 15-to-18-weeks. Immunoreactivity concerning only basal epithelia was quantified using a custom-designed semi-automatic analytical tool (MatLab, MathWorks) as previously described [105]. Double-immunostained corneas were semi-quantitatively assessed for staining overlap between the target protein and K14 in the limbus/peripheral cornea and the central cornea.

Central corneal injury

Mice were anesthetized with an intraperitoneal injection of ketamine and xylazine (100 mg/mL and 10 mg/mL respectively), corneas were washed with PBS, then epithelial debridement was performed on the right eye as previously described [106]. In brief, the region of the epithelium to be removed was trephine-marked, and an Algerbrush II with a 1-mm rotating burr was used to remove the corneal epithelium within this region. Corneas were again washed with PBS to remove any remaining debris, and PBS instilled in both eyes to prevent desiccation. Left eyes served as an internal (uninjured) control. Mice were followed for 5-days, after which they were euthanized by cervical dislocation, and their eyes fixed in 2% PFA for 1 hr at room temperature before immunostaining.

Statistical Analysis

Comparison between groups was performed with PRISM v.9 software (GraphPad Software LLC, La Jolla, CA) using unpaired Students t-test and one-way ANOVA with Dunn’s multiple comparisons, respectively, and p < 0.05 was considered statistically significant.

Acknowledgements: The authors would like to thank the staff at the Katherine Gauss Light Microscope Facility. This work was supported by a grant from the Australian National Health and Medical Research Council (APP1156944) to NDG.

Author Contributions

Alexander Richardson: Conception and design, Collection and/or assembly of data, Data analysis and interpretation, Manuscript writing, Final approval of manuscript

Susan Corley: Collection and/or assembly of data, Data analysis and interpretation, Manuscript writing, Final approval of manuscript

Naomi Delic: Collection and/or assembly of data, Data analysis and interpretation, Manuscript writing, Final approval of manuscript

Hue Li: Collection and/or assembly of data, Final approval of manuscript

Andrew Lloyd: Financial support, Administrative support, Final approval of manuscript

Nick Di Girolamo: Conception and design, Financial support, Administrative support, Provision of study material, Data analysis and interpretation, Manuscript writing, Final approval of manuscript.

Disclosure of conflicts of interest: The authors declare no conflicts of interest.

Data availability statement: Data is available at Gene Expression Omnibus, accession number GSE274984.

Cotsarelis, G., Cheng, S. Z., Dong, G., Sun, T. T., and Lavker, R. M., Existence of slow-cycling limbal epithelial basal cells that can be preferentially stimulated to proliferate: Implications on epithelial stem cells. Cell, 1989. 57(2): p. 201–9.
Di Girolamo, N., et al., Tracing the fate of limbal epithelial progenitor cells in the murine cornea. Stem Cells, 2015. 33(1): p. 157–69.
Altshuler, A., et al., Discrete limbal epithelial stem cell populations mediate corneal homeostasis and wound healing. Cell Stem Cell, 2021. 28(7): p. 1248–1261 e8.
Farrelly, O., et al., Two-photon live imaging of single corneal stem cells reveals compartmentalized organization of the limbal niche. Cell Stem Cell, 2021. 28(7): p. 1233–1247 e4.
Deng, S. X., et al., Global consensus on definition, classification, diagnosis, and staging of limbal stem cell deficiency. Cornea, 2019. 38(3): p. 364–375.
Espana, E. M., et al., Characterization of corneal pannus removed from patients with total limbal stem cell deficiency. Invest Ophthalmol Vis Sci, 2004. 45(9): p. 2961–6.
Yazdanpanah, G., et al., Strategies for reconstructing the limbal stem cell niche. Ocul Surf, 2019. 17(2): p. 230–240.
Thoft, R. A. and Friend, J., The x, y, z hypothesis of corneal epithelial maintenance. Invest Ophthalmol Vis Sci, 1983. 24(10): p. 1442–3.
Kenyon, K. R. and Tseng, S. C., Limbal autograft transplantation for ocular surface disorders. Ophthalmology, 1989. 96(5): p. 709 – 22; discussion 722-3.
Casaroli-Marano, R. P., Nieto-Nicolau, N., Martinez-Conesa, E. M., Edel, M., and A, B. A.-P., Potential role of induced pluripotent stem cells (ipscs) for cell-based therapy of the ocular surface. J Clin Med, 2015. 4(2): p. 318–42.
Schlotzer-Schrehardt, U. and Kruse, F. E., Identification and characterization of limbal stem cells. Exp Eye Res, 2005. 81(3): p. 247–64.
Bath, C., et al., Transcriptional dissection of human limbal niche compartments by massive parallel sequencing. PLoS One, 2013. 8(5): p. e64244.
Collin, J., et al., A single cell atlas of human cornea that defines its development, limbal progenitor cells and their interactions with the immune cells. Ocul Surf, 2021. 21: p. 279–298.
Ligocki, A. J., et al., Molecular characteristics and spatial distribution of adult human corneal cell subtypes. Sci Rep, 2021. 11(1): p. 16323.
Eghtedari, Y., et al., Keratin 14 expression in epithelial progenitor cells of the developing human cornea. Stem Cells Dev, 2016. 25(9): p. 699–711.
Figueira, E. C., Di Girolamo, N., Coroneo, M. T., and Wakefield, D., The phenotype of limbal epithelial stem cells. Invest Ophthalmol Vis Sci, 2007. 48(1): p. 144–56.
Lobo, E. P., et al., Self-organized centripetal movement of corneal epithelium in the absence of external cues. Nat Commun, 2016. 7: p. 12388.
Sartaj, R., et al., Characterization of slow cycling corneal limbal epithelial cells identifies putative stem cell markers. Sci Rep, 2017. 7(1): p. 3793.
Lee, C. S., May, N. R., and Fan, C. M., Transdifferentiation of the ventral retinal pigmented epithelium to neural retina in the growth arrest specific gene 1 mutant. Dev Biol, 2001. 236(1): p. 17–29.
Del Sal, G., et al., Structure, function, and chromosome mapping of the growth-suppressing human homologue of the murine gas1 gene. Proc Natl Acad Sci U S A, 1994. 91(5): p. 1848–52.
Li, L., et al., Muscle stem cell renewal suppressed by gas1 can be reversed by gdnf in mice. Nat Metab, 2019. 1(10): p. 985–995.
Lopez-Ornelas, A., Vergara, P., and Segovia, J., Neural stem cells producing an inducible and soluble form of gas1 target and inhibit intracranial glioma growth. Cytotherapy, 2014. 16(7): p. 1011–23.
Tumbar, T., et al., Defining the epithelial stem cell niche in skin. Science, 2004. 303(5656): p. 359–63.
Vellonen, K. S., et al., Monocarboxylate transport in human corneal epithelium and cell lines. Eur J Pharm Sci, 2010. 39(4): p. 241–7.
Kaplan, N., et al., Single-cell rna transcriptome helps define the limbal/corneal epithelial stem/early transit amplifying cells and how autophagy affects this population. Invest Ophthalmol Vis Sci, 2019. 60(10): p. 3570–3583.
Kulkarni, B. B., et al., Comparative transcriptional profiling of the limbal epithelial crypt demonstrates its putative stem cell niche characteristics. BMC Genomics, 2010. 11: p. 526.
Menzel-Severing, J., et al., Transcription factor profiling identifies sox9 as regulator of proliferation and differentiation in corneal epithelial stem/progenitor cells. Sci Rep, 2018. 8(1): p. 10268.
Thomas, P. B., et al., Identification of notch-1 expression in the limbal basal epithelium. Mol Vis, 2007. 13: p. 337–44.
Hernandez, H., Millar, J. C., Curry, S. M., Clark, A. F., and McDowell, C. M., Bmp and activin membrane bound inhibitor regulates the extracellular matrix in the trabecular meshwork. Invest Ophthalmol Vis Sci, 2018. 59(5): p. 2154–2166.
Pajoohesh-Ganji, A., Pal-Ghosh, S., Tadvalkar, G., and Stepp, M. A., K14 + compound niches are present on the mouse cornea early after birth and expand after debridement wounds. Dev Dyn, 2016. 245(2): p. 132–43.
Parfitt, G. J., et al., Immunofluorescence tomography of mouse ocular surface epithelial stem cells and their niche microenvironment. Invest Ophthalmol Vis Sci, 2015. 56(12): p. 7338–44.
Richardson, A., et al., Keratin-14-positive precursor cells spawn a population of migratory corneal epithelia that maintain tissue mass throughout life. Stem Cell Reports, 2017. 9(4): p. 1081–1096.
Sagga, N., Kuffova, L., Vargesson, N., Erskine, L., and Collinson, J. M., Limbal epithelial stem cell activity and corneal epithelial cell cycle parameters in adult and aging mice. Stem Cell Res, 2018. 33: p. 185–198.
Toyooka, Y., Shimosato, D., Murakami, K., Takahashi, K., and Niwa, H., Identification and characterization of subpopulations in undifferentiated es cell culture. Development, 2008. 135(5): p. 909–18.
Tang, F., et al., Mrna-seq whole-transcriptome analysis of a single cell. Nat Methods, 2009. 6(5): p. 377–82.
Jovic, D., et al., Single-cell rna sequencing technologies and applications: A brief overview. Clin Transl Med, 2022. 12(3): p. e694.
See, P., Lum, J., Chen, J., and Ginhoux, F., A single-cell sequencing guide for immunologists. Front Immunol, 2018. 9: p. 2425.
Ketkar, S., Burrage, L. C., and Lee, B., Rna sequencing as a diagnostic tool. JAMA, 2023. 329(1): p. 85–86.
Jaitin, D. A., et al., Massively parallel single-cell rna-seq for marker-free decomposition of tissues into cell types. Science, 2014. 343(6172): p. 776–9.
Ramskold, D., et al., Full-length mrna-seq from single-cell levels of rna and individual circulating tumor cells. Nat Biotechnol, 2012. 30(8): p. 777–82.
Sasamoto, Y., et al., Epigenetic regulation of corneal epithelial differentiation by tet2. Int J Mol Sci, 2023. 24(3).
Li, D. Q., et al., Single-cell transcriptomics identifies limbal stem cell population and cell types mapping its differentiation trajectory in limbal basal epithelium of human cornea. Ocul Surf, 2021. 20: p. 20–32.
Picelli, S., et al., Smart-seq2 for sensitive full-length transcriptome profiling in single cells. Nat Methods, 2013. 10(11): p. 1096–8.
Wang, X., He, Y., Zhang, Q., Ren, X., and Zhang, Z., Direct comparative analyses of 10x genomics chromium and smart-seq2. Genomics Proteomics Bioinformatics, 2021. 19(2): p. 253–266.
Hahaut, V. and Picelli, S., Full-length single-cell rna-sequencing with flash-seq. Methods Mol Biol, 2023. 2584: p. 123–164.
Bhargava, V., Head, S. R., Ordoukhanian, P., Mercola, M., and Subramaniam, S., Technical variations in low-input rna-seq methodologies. Sci Rep, 2014. 4: p. 3678.
You, J., et al., Rna-seq analysis and comparison of corneal epithelium in keratoconus and myopia patients. Sci Rep, 2018. 8(1): p. 389.
Smits, J. G. A., et al., Identification of the regulatory circuit governing corneal epithelial fate determination and disease. PLoS Biol, 2023. 21(10): p. e3002336.
Abu-Romman, A., et al., Conditional deletion of cd25 in the corneal epithelium reveals sex differences in barrier disruption. Ocul Surf, 2023. 30: p. 57–72.
Norrick, A., et al., Process development and safety evaluation of abcb5(+) limbal stem cells as advanced-therapy medicinal product to treat limbal stem cell deficiency. Stem Cell Res Ther, 2021. 12(1): p. 194.
Pellegrini, G., et al., Biological parameters determining the clinical outcome of autologous cultures of limbal stem cells. Regen Med, 2013. 8(5): p. 553–67.
Rama, P., et al., Limbal stem-cell therapy and long-term corneal regeneration. N Engl J Med, 2010. 363(2): p. 147–55.
Miyazaki, S., et al., Zfp296 knockout enhances chromatin accessibility and induces a unique state of pluripotency in embryonic stem cells. Commun Biol, 2023. 6(1): p. 771.
Almada, A. E., et al., Fos licenses early events in stem cell activation driving skeletal muscle regeneration. Cell Rep, 2021. 34(4): p. 108656.
Mao, D. D., Cleary, R. T., Gujar, A., Mahlokozera, T., and Kim, A. H., Cdc20 regulates sensitivity to chemotherapy and radiation in glioblastoma stem cells. PLoS One, 2022. 17(6): p. e0270251.
Guiu, J., et al., Identification of cdca7 as a novel notch transcriptional target involved in hematopoietic stem cell emergence. J Exp Med, 2014. 211(12): p. 2411–23.
Chen, H., et al., Dissecting heterogeneity reveals a unique bambi(high) mfge8(high) subpopulation of human uc-mscs. Adv Sci (Weinh), 2022. 10(1): p. e2202510.
Liu, C., et al., Transcriptional repression of the transforming growth factor beta (tgf-beta) pseudoreceptor bmp and activin membrane-bound inhibitor (bambi) by nuclear factor kappab (nf-kappab) p50 enhances tgf-beta signaling in hepatic stellate cells. J Biol Chem, 2014. 289(10): p. 7082–7091.
Chen, J., Bush, J. O., Ovitt, C. E., Lan, Y., and Jiang, R., The tgf-beta pseudoreceptor gene bambi is dispensable for mouse embryonic development and postnatal survival. Genesis, 2007. 45(8): p. 482–6.
Zhang, G., et al., Bone morphogenetic protein signaling regulates id1-mediated neural stem cell quiescence in the adult zebrafish brain via a phylogenetically conserved enhancer module. Stem Cells, 2020. 38(7): p. 875–889.
Sachdeva, R., et al., Bmp signaling mediates glioma stem cell quiescence and confers treatment resistance in glioblastoma. Sci Rep, 2019. 9(1): p. 14569.
Amitai-Lange, A., et al., Lineage tracing of stem and progenitor cells of the murine corneal epithelium. Stem Cells, 2015. 33(1): p. 230–9.
Yoshida, S., et al., Cytokeratin 15 can be used to identify the limbal phenotype in normal and diseased ocular surfaces. Invest Ophthalmol Vis Sci, 2006. 47(11): p. 4780–6.
Kurpakus, M. A., Maniaci, M. T., and Esco, M., Expression of keratins k12, k4 and k14 during development of ocular surface epithelium. Curr Eye Res, 1994. 13(11): p. 805–14.
Maiti, G., et al., Single cell rna-seq of human cornea organoids identifies cell fates of a developing immature cornea. PNAS Nexus, 2022. 1(5): p. pgac246.
Finnegan, A., et al., Single-cell transcriptomics reveals spatial and temporal turnover of keratinocyte differentiation regulators. Front Genet, 2019. 10: p. 775.
Dou, S., et al., Molecular identity of human limbal heterogeneity involved in corneal homeostasis and privilege. Ocul Surf, 2021. 21: p. 206–220.
Catala, P., et al., Single cell transcriptomics reveals the heterogeneity of the human cornea to identify novel markers of the limbus and stroma. Sci Rep, 2021. 11(1): p. 21727.
Ksander, B. R., et al., Abcb5 is a limbal stem cell gene required for corneal development and repair. Nature, 2014. 511(7509): p. 353–7.
de Paiva, C. S., Chen, Z., Corrales, R. M., Pflugfelder, S. C., and Li, D. Q., Abcg2 transporter identifies a population of clonogenic human limbal epithelial cells. Stem Cells, 2005. 23(1): p. 63–73.
Watanabe, K., et al., Human limbal epithelium contains side population cells expressing the atp-binding cassette transporter abcg2. FEBS Lett, 2004. 565(1–3): p. 6–10.
Kitao, M., et al., Identification of bst2 as a conjunctival epithelial stem/progenitor cell marker. iScience, 2023. 26(7): p. 107016.
Stewart, R. M., Sheridan, C. M., Hiscott, P. S., Czanner, G., and Kaye, S. B., Human conjunctival stem cells are predominantly located in the medial canthal and inferior forniceal areas. Invest Ophthalmol Vis Sci, 2015. 56(3): p. 2021–30.
Vagnozzi, A. N., Reiter, J. F., and Wong, S. Y., Hair follicle and interfollicular epidermal stem cells make varying contributions to wound regeneration. Cell Cycle, 2015. 14(21): p. 3408–17.
Wong, V. W., et al., Lrig1 controls intestinal stem-cell homeostasis by negative regulation of erbb signalling. Nat Cell Biol, 2012. 14(4): p. 401–8.
Rosti, K., Goldman, A., and Kajander, T., Solution structure and biophysical characterization of the multifaceted signalling effector protein growth arrest specific-1. BMC Biochem, 2015. 16: p. 8.
Seppala, M., et al., Gas1 is a modifier for holoprosencephaly and genetically interacts with sonic hedgehog. J Clin Invest, 2007. 117(6): p. 1575–84.
Gobeil, S., Zhu, X., Doillon, C. J., and Green, M. R., A genome-wide shrna screen identifies gas1 as a novel melanoma metastasis suppressor gene. Genes Dev, 2008. 22(21): p. 2932–40.
Li, Q., et al., Gas1 inhibits metastatic and metabolic phenotypes in colorectal carcinoma. Mol Cancer Res, 2016. 14(9): p. 830–40.
van Roeyen, C. R., et al., Growth arrest-specific protein 1 is a novel endogenous inhibitor of glomerular cell activation and proliferation. Kidney Int, 2013. 83(2): p. 251–63.
Cancedda, R. and Mastrogiacomo, M., Transit amplifying cells (tacs): A still not fully understood cell population. Front Bioeng Biotechnol, 2023. 11: p. 1189225.
Pajoohesh-Ganji, A., Pal-Ghosh, S., Simmens, S. J., and Stepp, M. A., Integrins in slow-cycling corneal epithelial cells at the limbus in the mouse. Stem Cells, 2006. 24(4): p. 1075–86.
Zhao, J., Mo, V., and Nagasaki, T., Distribution of label-retaining cells in the limbal epithelium of a mouse eye. J Histochem Cytochem, 2009. 57(2): p. 177–85.
Oh, J., Lee, Y. D., and Wagers, A. J., Stem cell aging: Mechanisms, regulators and therapeutic opportunities. Nat Med, 2014. 20(8): p. 870–80.
Enwere, E., et al., Aging results in reduced epidermal growth factor receptor signaling, diminished olfactory neurogenesis, and deficits in fine olfactory discrimination. J Neurosci, 2004. 24(38): p. 8354–65.
Zhang, Y., Ravikumar, M., Ling, L., Nurcombe, V., and Cool, S. M., Age-related changes in the inflammatory status of human mesenchymal stem cells: Implications for cell therapy. Stem Cell Reports, 2021. 16(4): p. 694–707.
Pietras, E. M., Warr, M. R., and Passegue, E., Cell cycle regulation in hematopoietic stem cells. J Cell Biol, 2011. 195(5): p. 709–20.
Garcia-Prat, L., Sousa-Victor, P., and Munoz-Canoves, P., Functional dysregulation of stem cells during aging: A focus on skeletal muscle stem cells. FEBS J, 2013. 280(17): p. 4051–62.
Notara, M., Shortt, A. J., O'Callaghan, A. R., and Daniels, J. T., The impact of age on the physical and cellular properties of the human limbal stem cell niche. Age (Dordr), 2013. 35(2): p. 289–300.
Yang, S., Lee, H. J., Shin, S., Park, I. Y., and Chung, S. H., Limbal epithelial stem cell sheets from young donors have better regenerative potential. Sci Rep, 2022. 12(1): p. 14191.
Zheng, T. and Xu, J., Age-related changes of human limbus on in vivo confocal microscopy. Cornea, 2008. 27(7): p. 782–6.
Abu-Romman, A., et al., Age-related differences in the mouse corneal epithelial transcriptome and their impact on corneal wound healing. Invest Ophthalmol Vis Sci, 2024. 65(5): p. 21.
Douvaras, P., et al., Rare corneal clones in mice suggest an age-related decrease of stem cell activity and support the limbal epithelial stem cell hypothesis. Stem Cell Res, 2012. 8(1): p. 109–19.
Mort, R. L., Ramaesh, T., Kleinjan, D. A., Morley, S. D., and West, J. D., Mosaic analysis of stem cell function and wound healing in the mouse corneal epithelium. BMC Dev Biol, 2009. 9: p. 4.
Chang, C. Y., Green, C. R., McGhee, C. N., and Sherwin, T., Acute wound healing in the human central corneal epithelium appears to be independent of limbal stem cell influence. Invest Ophthalmol Vis Sci, 2008. 49(12): p. 5279–86.
Kubota, M., et al., The anti-oxidative role of abcg2 in corneal epithelial cells. Invest Ophthalmol Vis Sci, 2010. 51(11): p. 5617–22.
Novelli, F., et al., P63 in corneal and epidermal differentiation. Biochem Biophys Res Commun, 2022. 610: p. 15–22.
Park, M., et al., Peripheral (not central) corneal epithelia contribute to the closure of an annular debridement injury. Proc Natl Acad Sci U S A, 2019. 116(52): p. 26633–26643.
Dora, N. J., Hill, R. E., Collinson, J. M., and West, J. D., Lineage tracing in the adult mouse corneal epithelium supports the limbal epithelial stem cell hypothesis with intermittent periods of stem cell quiescence. Stem Cell Res, 2015. 15(3): p. 665–677.
West, J. D., Mort, R. L., Hill, R. E., Morley, S. D., and Collinson, J. M., Computer simulation of neutral drift among limbal epithelial stem cells of mosaic mice. Stem Cell Res, 2018. 30: p. 1–11.
Kester, L. and van Oudenaarden, A., Single-cell transcriptomics meets lineage tracing. Cell Stem Cell, 2018. 23(2): p. 166–179.
Patro, R., Duggal, G., Love, M. I., Irizarry, R. A., and Kingsford, C., Salmon provides fast and bias-aware quantification of transcript expression. Nat Methods, 2017. 14(4): p. 417–419.
Soneson, C., Love, M. I., and Robinson, M. D., Differential analyses for rna-seq: Transcript-level estimates improve gene-level inferences. F1000Res, 2015. 4: p. 1521.
Stuart, T., et al., Comprehensive integration of single-cell data. Cell, 2019. 177(7): p. 1888–1902 e21.
Zhang, R., et al., Dose-dependent benzalkonium chloride toxicity imparts ocular surface epithelial changes with features of dry eye disease. Ocul Surf, 2020. 18(1): p. 158–169.
Park, M., et al., Visualizing the contribution of keratin-14(+) limbal epithelial precursors in corneal wound healing. Stem Cell Reports, 2019. 12(1): p. 14–28.

No competing interests reported.

Supplementarymaterial.pptx
Supplementary Figure S1. Co-expression of K14 with Gas1 and Lrig1 in the central cornea. Whole corneas from mice aged 5 (a and b), 10 (c and d), 20 (e and f), 40 (g and h) and 60-weeks (i and j) were double immunostained for K14 (red) and Gas1 (green; a, c, e, g and i) and K14 and Lrig1 (b, d, f, h and j). Images were obtained as z-stacks from the central cornea and represent maximum intensity projections. Tissue was counterstained with Hoechst 33342 (blue). Scale bars = 50 µm.

Download PDF

Reviewers invited by journal
20 Aug, 2024
Editor assigned by journal
20 Aug, 2024
Editor invited by journal
20 Aug, 2024
Submission checks completed at journal
19 Aug, 2024
First submitted to journal
15 Aug, 2024

You are reading this latest preprint version

Single-cell RNA sequencing of murine limbal epithelia reveals Gas1 as a novel stem/progenitor cell marker for the corneal epithelium

Status:

Version 1

Abstract

Figures

INTRODUCTION

RESULTS

Cluster and gene ontology analysis of K14CreER^T2-Confetti mouse corneal tissue

Validating novel candidate markers of LESCs identified by scRNA-seq

Localization of Gas1 in the mouse cornea

Expression of identified SC markers following central corneal injury

DISCUSSION

CONCLUSION

MATERIALS AND METHODS

K14CreERT2-Confetti Mice

Single-cell suspension for FACS

Library preparation, sequencing and transcriptomic analyses

Quantitative polymerase chain reaction (qPCR) analysis

Flow cytometry

Immunofluorescence and quantitative analysis of wholemount corneas

Central corneal injury

Statistical Analysis

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1

Single-cell RNA sequencing of murine limbal epithelia reveals Gas1 as a novel stem/progenitor cell marker for the corneal epithelium

Status:

Version 1

Abstract

Figures

INTRODUCTION

RESULTS

Cluster and gene ontology analysis of K14CreERT2-Confetti mouse corneal tissue

Validating novel candidate markers of LESCs identified by scRNA-seq

Localization of Gas1 in the mouse cornea

Expression of identified SC markers following central corneal injury

DISCUSSION

CONCLUSION

MATERIALS AND METHODS

K14CreERT2-Confetti Mice

Single-cell suspension for FACS

Library preparation, sequencing and transcriptomic analyses

Quantitative polymerase chain reaction (qPCR) analysis

Flow cytometry

Immunofluorescence and quantitative analysis of wholemount corneas

Central corneal injury

Statistical Analysis

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1

Cluster and gene ontology analysis of K14CreER^T2-Confetti mouse corneal tissue