Next Article in Journal
Lipoprotein(a) and Cardiovascular Disease: From Genetic Risk Factor to Therapeutic Target
Previous Article in Journal
Rapid Drug Sensitivity Profiling via a Novel High-Success-Rate Culture Method for Patient-Derived Pancreatic Cancer: An Exploratory Preclinical Platform for Advancing Clinical Applications and Drug Development
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Cells
Volume 15
Issue 4
10.3390/cells15040314
cells-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Disruption of Cell-Adhesion Signaling Resolves Unwanted Progenitor Specification in Stem Cell-Derived α and β Cell Grafts

by
Kyle R. Knofczynski
1,
Ethan W. Law
1,
Sean Lewis-Brinkman
1,
Zenith Khashim
2,
Anna Marie R. Schornack
2,
Swikriti Shrestha
1,
Lauren T. Jennings
1 and
Quinn P. Peterson
2,*
1
Mayo Clinic Graduate School of Biomedical Sciences, Mayo Clinic, Rochester, MN 55905, USA
2
Department of Physiology and Biomedical Engineering, Mayo Clinic, Rochester, MN 55905, USA
*
Author to whom correspondence should be addressed.
Cells 2026, 15(4), 314; https://doi.org/10.3390/cells15040314
Submission received: 11 December 2025 / Revised: 16 January 2026 / Accepted: 28 January 2026 / Published: 7 February 2026
(This article belongs to the Section Stem Cells)
Browse Figures
Versions Notes

Highlights

What are the main findings?
  • An enrichment in cell-adhesion signaling is identified via single-cell RNA sequencing in the outgrowth-driving populations of transplanted stem cell-derived β and α cell products.
  • The interruption of cell-adhesion signaling via Notch inhibition or single-cell dissociation disrupts outgrowth-driving populations in transplant-ready cell populations.
What are the implications of the main findings?
  • The dispersion and reaggregation of stem cell-derived β and α cells enhances their safety profiles following transplantation in mice.

Abstract

Directed differentiation protocols have recently been developed to produce stem cell-derived α (SC-α) cells as a potential component of a complete cell-based therapy for T1D, to complement the more widely studied stem cell-derived β (SC-β) cells. Differentiation protocols for SC-β cells produce off-target cell populations implicated in the development of outgrowths in SC-β cell grafts, but outgrowths from SC-α cells have not been explored. This study identifies that engrafted SC-α cells generate outgrowths of similar composition to SC-β cell outgrowths. Both cell types share outgrowth-driving populations marked by SOX9, CDX2, or SOX2. Single-cell RNA sequencing was used to reveal an enrichment in cell-adhesion signaling events in outgrowth-driving populations. Small-molecule inhibition of the Notch pathway was insufficient to disrupt all three outgrowth-driving populations. A comprehensive disruption of cell-adhesion signaling via single-cell dispersion and reaggregation is found to reduce the outgrowth propensity in engrafted SC-α and SC-β cells. Together, these results suggest that disrupting residual progenitor cells with SC-α and SC-β cell clusters can enhance the safety profile of these cell therapy products for T1D therapy.

Graphical Abstract

1. Introduction

Type 1 diabetes (T1D) is an autoimmune disease resulting in the destruction of the insulin-producing β cells of the pancreas, affecting more than 8 million individuals globally [1]. Autoimmune destruction of β cells is often accompanied by dysfunction of glucagon-producing α cells, which can further exacerbate the episodes of hyper- and hypo-glycemia that occur in these patients [2,3,4,5,6]. Current treatment strategies focus on maintaining euglycemia primarily via the administration of exogenous insulin. Although the transplantation of cadaveric islets has only recently been FDA-approved as a therapeutic alternative to insulin injections, its clinical use remains limited to a specific subset of individuals with type 1 diabetes, specifically those experiencing severe hypoglycemia with impaired awareness. Similarly, whole-pancreas transplantation is a viable therapeutic option but is typically reserved for patients with end-stage renal disease who are simultaneously undergoing kidney transplantation [7,8,9]. Moreover, the shortage of donor organs and the need for lifelong immunosuppression limit the widespread implementation of both whole-pancreas and cadaveric islet transplantation therapy.
Directed differentiation of pluripotent stem cells is a promising, renewable alternative to cadaveric cell sources [7,10]. Initial efforts focused on the generation of pancreatic progenitor cells that could further mature into the three major islet endocrine cell types upon transplantation [11,12,13]. Unfortunately, while apparently safe, these cells did not demonstrate substantial therapeutic benefit upon transplantation into T1D patients [14,15]. Further research over the past decade has generated directed differentiation protocols to advance cells from the pancreatic progenitor stage into stem cell-derived pancreatic β (SC-β) cells prior to transplantation [16,17,18,19]; such cells are currently being clinically evaluated, with promising preliminary findings from Phase I clinical trials [20].
A major challenge for the clinical translation of SC-β cells is the development of graft outgrowths after transplantation in rodent models [21,22,23,24,25]. Common impurities found in SC-β cell populations generated via a variety of differentiation protocols include populations marked by the expression of SOX9 (pancreatic progenitor), SOX2 (foregut endoderm, stomach), CDX2 (midgut, intestine), and TPH1 (enterochromaffin) [25,26,27,28,29]. Recently, new protocols have been developed to generate stem cell-derived (SC-α) cells that model human pancreatic alpha cell identity and function [30,31]. It remains unclear if these SC-α differentiations generate the same impurities. Furthermore, the degree to which transplanted SC-α cell populations produce transplant outgrowths, and the properties of the cells contributing to such outgrowths, have not been previously explored.
Here, we demonstrate that transplanted SC-α cells generate graft outgrowths similar to those arising from SC-β cell transplants. Outgrowths from both transplants are shown to arise from SOX9+, CDX2+, or SOX2+ cells, each with a distinct outgrowth morphology. Each of the outgrowth-driving populations arises at key timepoints in the SC-α and SC-β differentiations, as a result of improper specification of the pancreatic or endocrine lineages. Single-cell RNA sequencing analysis of SC-α and SC-β cells highlights enriched cell-adhesion signaling in the outgrowth-driving populations. The targeting of the Notch pathway using pharmacological inhibitors specifically reduced SOX9 expression. Finally, we demonstrate that the outgrowth potential of these SC-α and SC-β products can be reduced by dispersion and reaggregation, pointing towards strategies for improving safety prior to use in T1D patients.

2. Materials and Methods

2.1. Stem Cell Culture and Differentiation

HUES8 (NIHhESC-09-0021) human pluripotent stem cells were cultured as spheroids and differentiated into SC-α or SC-β cells in an entirely 3D culture system as previously described [16,19,30,32]. Briefly, 150 million pluripotent stem cells were seeded into a 500 mL 3D spinner flask (Corning® Disposable Spinning Flask Bioreactor, Corning, NY, USA) with 300 mL of mTeSR (STEMCELL Technologies, Vancouver, BC, Canada) and 10 μM Y27632 (R&D Systems, Minneapolis, MN, USA) and stirred at 70 rpm in a humidified incubator at 37 °C and 5% CO2. After 48 h of culture, the medium was replaced with 300 mL of mTeSR without Y27632. After an additional 24 h of culture, directed differentiation was initiated by changing to S1 media supplemented with growth factors to induce differentiation toward definitive endoderm. Media changes and factor supplementation (Tables S1–S3) were performed either towards SC-α (Table S2) or SC-β cells (Table S3). All cell stocks were fingerprinted and karyotyped prior to banking, and mycoplasma was tested monthly during culture. The pluripotency of cell cultures was validated via the expression of OCT4, as measured by flow cytometry. The Mayo Clinic Stem Cell Research Oversight Committee reviewed and approved all work involving human pluripotent stem cells carried out in the present study, including review of compliance with informed consent.

2.2. Animals and Transplantation

All animal studies were conducted with approval from appropriate institutional oversight committees, including the Mayo Clinic Institutional Animal Care and Use Committee. Prior to surgery, animals were socially housed within sterile cages with unrestricted access to food and water. Ambient temperature was maintained between 18 and 25 °C, with humidity at 30–70% and 12 h light/dark cycles. Male SCID-beige mice, aged 8–10 weeks, were obtained from Charles River Laboratory. Five million of either SC-α or SC-β cells were transplanted under the kidney capsule of each mouse. Post-surgery, mice were single-housed and monitored for 12 weeks. At the conclusion of each study, animals were sacrificed, and the engrafted kidneys were collected for analysis as described below. Growth volume after 12 weeks was estimated by measuring the length and width of the growth and applying the formula: 0.5 × L × W2, where the width of the growth was the smaller of the two measurements.

2.3. Immunofluorescence and H&E Staining

Differentiated cell clusters or kidney grafts were fixed in 4% PFA at 21 °C for at least 1 h (clusters) or 24 h (kidneys). Cell clusters were subsequently embedded in HistoGel (Epredia, Kalamazoo, MI, USA). Cluster and kidney samples were embedded in paraffin and sectioned at 5 μm. Samples on glass slides were rehydrated, and antigen retrieval was performed in boiling sodium citrate buffer for 1 h. Slides were blocked with PBS + 0.1% Triton X-100 + 5% donkey serum for 1 h. Slides were incubated at room temperature with primary antibodies for 1 h and secondary antibodies with DAPI (4′,6-diamidino-2-phenylindole) for 1 h. Prior to imaging, slides were mounted using Fluoromount-G (Southern Biotech, Birmingham, AL, USA). Imaging was performed on a Zeiss Axio Observer 7 (Oberkochen, Germany). Images were processed using Zeiss Zen 3.9 software. Information on antibodies can be found in Table S4.
For hematoxylin and eosin (H&E) staining, rehydrated slides were stained with hematoxylin for 30 s and then rinsed with water. Slides were then counterstained with eosin for 90 s. Slides were subsequently dehydrated and mounted with Permount (Electron Microscopy Sciences, Hatfield, PA, USA). Automated imaging was performed with an Aperio Versa 200 (Leica, Wetzlar, Germany) slide scanner at 10 and 20×.

2.4. Quantitative PCR (qPCR)

Total RNA was isolated from frozen clusters from SC-α and SC-β cell differentiations at the end of differentiation using the Qiagen RNeasy® Mini Kit (Hilden, Germany). RNA concentration and quality were characterized with a NanoDrop™ One spectrophotometer (Thermo Fisher, Waltham, MA, USA). Complementary DNA (cDNA) for each sample was produced using the iScript cDNA Synthesis Kit (BioRad, Hercules, CA, USA). Quantitative PCR was performed with a 20 µL reaction mix containing 1× PowerUp SYBR™ Green PCR Master Mix (Thermo Fisher, Waltham, MA, USA), 200 nM forward primer, 200 nM reverse primer, and 20 ng cDNA. The forward and reverse primers for all genes analyzed were manufactured by Integrated DNA Technologies (Coralville, IA, USA). All primer sequences can be found in Table S5. Three replicates of each gene for each biological sample were run and measured using a Roche LightCycler® 96 System (Basal, Switzerland). The Ct values for the technical replicates were averaged, and gene expression was normalized to GAPDH to calculate ΔCq for each sample. ΔΔCq values were calculated by normalizing gene expression to the stage (Stage 1) or treatment (no XXi treatment) controls.

2.5. Flow Cytometry

Clusters of differentiated cells were dissociated using TrypLE (Thermo Fisher, Waltham, MA, USA) at 37 °C and fixed in 4% PFA for 1 h. Clusters were blocked using PBS + 0.1% Triton X-100 + 5% donkey serum for 1 h. Cells were incubated in primary antibodies for 1 h at room temperature. Cells were washed twice in PBST and incubated in secondary antibodies for 1 h at room temperature. Cells were again washed twice using PBST and analyzed using an Attune NxT flow cytometer (Thermo Fisher, Waltham, MA, USA). FlowJo v10 software was used to analyze the data. Information on antibodies can be found in Table S4. The gating strategy for flow cytometry plots is shown in Figure S1.

2.6. Dispersion and Reaggregation

SC-β and SC-α clusters were removed from the 3D culture system, washed twice with sterile PBS, and incubated with TrypLE for 10 min at 37 °C. Digested cells were dispersed to single cells via gentle pipetting and counted using trypan blue on a Countess 3 automated cell counter (Thermo Fisher, Waltham, MA, USA). Single cells were plated into round-bottom spheroid plates (Corning, Corning, NY, USA) at 20,000 cells/well and allowed to form into spheroids over 96 h, feeding with stage 6 media every 48 h. Prior to transplantation, cells were removed from the spheroid plates, and cell number and viability were assessed using Trypan Blue.

2.7. Notch Inhibition of SC-β and SC-α Cells

SC-β or SC-α clusters were removed from the 3D spinner flasks and split into 6-well plates at a concentration of ~4 million cells per well. Cells were fed with 3 mL of the respective stage 6 protocol media composition containing 0–1000 nM of Compound E (R&D Systems, Minneapolis, MN, USA) and cultured on a plate rocker (30 rpm). Plates were fed again 24 h later and harvested at 48 h for flow cytometry and qPCR analysis.

2.8. scRNA-seq Processing

SC-α cell sequencing data were obtained from GSE:138857 [30]. Data were analyzed using Seurat v5.3.0 [33]. Cells were filtered based on gene count, mitochondrial content, and total RNA counts. Raw counts were normalized and scaled using the 2000 most variable features and using cell cycle as a regression variable. A UMAP projection was carried out using the first 30 dimensions. Cell populations were manually identified based on upregulated gene expression in each population. Additional sequencing data for SC-β cells was obtained from GSE:114412 and analyzed in the same manner [19].

2.9. Statistical Analysis

Data were graphed and analyzed using Prism 10.0.0 software (GraphPad, San Diego, CA, USA). One-way repeated measures ANOVA with Dunnett’s multiple comparison tests were used for the analysis of qPCR and flow cytometry data from unique differentiation stages. Dose–response curves for gene expression were generated using a 4PL model in GraphPad. Paired Student’s t-tests were performed for gene expression following XXi treatment or reaggregation. Unpaired Student’s t-tests were performed for outgrowth size analysis. Data are reported as mean ± standard error of the mean with asterisks indicating statistical significance (* indicates a p-value < 0.05, ** indicates a p-value < 0.01).

3. Results

3.1. SC-α and SC-β Transplantation Results in Cystic Outgrowth of Grafts

While our prior studies evaluated the in vivo functionality of SC-α and SC-β cell populations [16,30], neither of these studies was specifically focused on the assessment of outgrowths from the engrafted cells. Previous studies focused on SC-β cell transplant outcomes identified populations of SOX9+, CDX2+, and/or SOX2+ cells that negatively affected transplant outcomes [26,27]. These cell populations were marked by transcription factors associated with distinct regions of the developing gut tube and pancreatic endoderm (Figure 1A). SOX2 and CDX2 expression corresponded to anterior foregut and midgut development, respectively. SOX9 expression was observed in pancreatic progenitors during early development, before its expression was restricted to ductal cells following pancreas maturation [28]. Given the similar strategy of pancreatic endoderm specification used in both SC-β and SC-α protocols, we wanted to investigate whether SC-α differentiations also give rise to the same adverse cell populations that may compromise transplantation outcomes. To evaluate the contribution of cellular impurities to graft outgrowth in both SC-α and SC-β cells, we differentiated SC-α and SC-β cells from HUES8 human embryonic stem cells as previously described [16,19,30]. Five million SC-α or SC-β cells were transplanted as clusters under the renal capsule of SCID-beige mice, and grafts were analyzed 12 weeks after transplantation.
Transplantation of either SC-α or SC-β cells resulted in graft outgrowths containing large cystic structures (Figure 1B). Hematoxylin and eosin (H&E) staining of outgrowths demonstrated that a large proportion of graft outgrowth was composed of cystic pockets. Immunofluorescence microscopy was utilized to characterize the presence of commonly observed byproducts of pancreatic endocrine differentiation in the cystic growths. Cells within the outgrowths stained positively for three key markers of differentiation byproducts: SOX9, CDX2, or SOX2 (Figure 1C,D) [26,27,28]. Each marker corresponded to a distinct morphology observable under H&E staining: SOX9-positive cells formed small, dense luminal clusters; CDX2-positive cells surrounded medium-sized cystic lumens with large columnar cells; and SOX2-positive cells lined large cystic lumens with a thin epithelial layer. Notably, TPH1 expression was absent in the cystic outgrowths, indicating that the enterochromaffin cells generated by the SC-β and SC-α protocols did not contribute to the outgrowth of the graft (Figure S2) [19,34].

3.2. Outgrowth-Driving Cell Populations Arise During Pancreatic and Endocrine Specification Windows in SC-α and SC-β Differentiations

To confirm that the presence of these off-target populations was associated with inadequate specification of pancreatic and endocrine lineages, we utilized immunofluorescence microscopy and qPCR to assess cellular specification through SC-α and SC-β differentiations. Marked expression of both CDX2 and SOX2 was observed following stage 3 in both differentiation protocols, corresponding to the specification of the PDX1+ pancreatic progenitors from the gut tube endoderm (Figure 2A). The expression of both CDX2 and SOX2 remained relatively consistent throughout the remainder of each protocol, highlighting the stable nature of these cell populations. SOX9 expression increased dramatically during stage 4 of both differentiation protocols during the specification of pre-endocrine and endocrine progenitors (Figure 2A). Decreased expression of SOX9 was seen after the completion of stage 5 of the SC-β protocol, while SOX9 expression gradually increased following continued SC-α differentiation. Immunofluorescence of fully differentiated SC-β and SC-α cell clusters demonstrated the presence of SOX9+, CDX2+, and SOX2+ cells in the pre-transplant spheroids (Figure 2B). None of the off-target cell markers were expressed in GCG+ SC-α cells or C-Peptide+ SC-β cells. The only exception was a small subset of cells in the SC-β differentiations that co-expressed CDX2 and C-Peptide that may differentiate into enterochromaffin cells. This indicates that SOX9, CDX2, and SOX2 were not labeling the desired hormone-producing cells generated by either differentiation protocol. Prior to transplantation, these populations did not display any unique morphological characteristics as they did in the 12-week transplant outgrowths. These results indicate that the SOX9+, CDX2+, and SOX2+ cell populations that give rise to growth following transplantation were a result of off-target differentiation products and incomplete specification of non-pancreatic cell fates.

3.3. Heightened Cell-Adhesion Signaling Patterns Mark Proliferative Outgrowth-Driving Cell Populations

To better understand the transcriptional identity of the growth-generating cell populations in our differentiations, we utilized single-cell RNA sequencing previously performed on an SC-α cell differentiation [30]. Clustering analysis revealed SOX9+ (orange) and CDX2+ (pink) cells distributed within a larger progenitor cluster as well as a distinct SOX2+ (blue) cluster within a larger endocrine population composed of SC-α and enterochromaffin cells (Figure 3A,B). The similarity of SOX2+ cells to the enterochromaffin population suggested a gastric-like endocrine identity, supporting the foregut-derived lineage of the SOX2+ cells. Cell cycle analysis revealed that the SOX9+ and CDX2+ populations were highly proliferative, while populations from the endocrine lineage, including SOX2+ cells, had lower rates of proliferation (Figure 3C).
To gain insight into how these different cell populations are regulated, we performed cell signaling analysis using CellChat (v2.2.0) [35]. The identification of the ten strongest incoming signals for each cell population revealed that SOX9+ and CDX2+ cells were highly enriched recipients of cell-adhesion signaling events, including both cell–ECM (Collagen, Fibronectin, Vitronectin) and cell–cell (Notch, Cadherin, Desmosome) signaling patterns. Signaling patterns in SOX2+ cells were similar to those of other endocrine cells with downregulated cell-adhesion signaling, especially that of Notch (Figure 3D). The distinct presence of Notch-signaling in the proliferative SOX9+ and CDX2+ progenitors and notable absence in the endocrine cell populations matches previous reports about the role of Notch in development, proliferation, and progenitor cell maintenance, including in pancreatic progenitors [36,37,38,39]. A closer look into Notch-signaling dynamics revealed that the SOX9+ and CDX2+ progenitor populations express high levels of both Notch receptors and ligands, particularly the non-canonical Notch ligand DLK1 (Figure 3E). SOX9+ progenitors also expressed the JAG1 ligand, which was absent in CDX2+ progenitors. Both progenitors had high expression of the Notch effector HES1. The endocrine cell populations did not display any notable expression of Notch receptors or ligands. Interestingly, despite the near-complete absence of Notch receptors and ligands in the SOX2+ population, it did display high expression levels of HES1 (Figure 3E), suggesting non-canonical activation of Notch effectors or recent fate transition from a pre-endocrine progenitor population [40].
The CellChat analysis of the progenitor populations also revealed the activation of other developmentally important signaling pathways, in addition to Notch. Of particular interest is the assumption of EGF and TGF-β signaling originating from the CDX2+ progenitor population and BMP signaling within the SOX2+ population (Figure S3). Because these signaling pathways are selectively active in the populations that drive outgrowth, they offer promising targets for future interventions to prevent off-target outgrowths. Comparing the cell populations of our SC-α sequencing data to previously published sequencing data from a similar SC-β differentiation protocol demonstrates that SC-β cells have similar off-target population distributions and signaling patterns, including heightened Notch, Fibronectin, Cadherin, and Laminin signaling. These overlapping signaling patterns indicate that the off-target populations of SC-α and SC-β differentiations are alike. Notably, the SC-β sequencing data lacked a distinct SOX2+ population that we had previously observed in our own SC-β differentiations (Figure S4) [19]. This discrepancy may be related to differentiation protocol differences or detection limits in the sequencing study used.

3.4. Inhibition of Notch Signaling Disrupts SOX9+ Pancreatic Progenitors but Fails to Modulate CDX2+ or SOX2+ Populations

The propensity of post-transplant outgrowth of transplanted SC-β or SC-α cells is a barrier towards clinical translation of these cell-based therapies. As such, we sought to find a method to eliminate the outgrowth-producing populations prior to their transplantation. Given that elevated Notch signaling marked the highly proliferative SOX9+ and CDX2+ progenitors, we investigated whether the inhibition of Notch signaling could disrupt the progenitor-like signaling patterns and eliminate these populations from SC-β and SC-α cells. To that end, cells obtained post SC-β and SC-α differentiation were treated with the gamma secretase inhibitor Compound E (XXi) for 48 h over a range of concentrations (0–1000 nM) to inhibit Notch signaling. The expressions of SOX9, CDX2, and SOX2 after treatment were measured via qPCR (Figure 4A).
SOX9 expression in SC-α and SC-β cells was reduced following XXi treatment, consistent with the well characterized requirement for Notch signaling in the maintenance of SOX9+ pancreatic progenitors [36,37]. The response of SOX9 expression to XXi concentration was modeled using a 4-parameter logistic dose–response curve for both SC-α and SC-β cells. The effect of XXi treatment on SOX9 expression had IC50 values of 56.1 nM and 34.8 nM in SC-α and SC-β cells, respectively. Notably, although both SOX9+ and CDX2+ cells showed elevated Notch-signaling patterns, treatment of SC-α and SC-β cells with XXi did not alter the expression of CDX2. Similarly, SOX2 expression was unaffected by XXi treatments in both differentiation protocols.
To ensure that the observed reductions in SOX9 transcript levels were due to the loss of the SOX9+ cell population, we performed flow cytometry to quantify the key cell populations following XXi-treatment of SC-α and SC-β cells (Figure 4B). Consistent with the dose–response curves, the proportion of SOX9+ cells in the SC-α cell product was significantly decreased at all tested concentrations of XXi greater than the IC50 value for SC-α cells (56.1 nM). Similarly, the proportion of SOX9+ cells in the SC-β cell product was decreased at all tested concentrations of XXi greater than the IC50 value for SC-β cells (34.8 nM). Importantly, XXi treatment did not impact the C-peptide+/NKX6.1+ fraction of SC-β cells or the GCG+/C-peptide- fraction of SC-α cells at any tested concentration, indicating that this treatment does not have deleterious effects on the therapeutic cell populations in potential transplant products (Figure 4C).

3.5. Dispersion and Reaggregation of SC-α and SC-β Cells Reduce Transplant Growth Potential

Although the inhibition of Notch signaling effectively reduced the SOX9+ cell population, it did not impact the other two growth-forming populations. Given the highly enriched nature of cell-adhesion signaling in the SOX9+ and CDX2+ progenitors, we hypothesized that dispersion and reaggregation of the SC-β or SC-α cells could simultaneously disrupt all adhesion signals. Consistent with this approach, dispersion and reaggregation of pancreatic progenitor cells, as well as various antibody-based sorting methods, have previously demonstrated the ability to enrich pancreatic endocrine cells while eliminating non-endocrine populations (Figure S5A) [19,26,41,42]. However, the specific effects of dispersion and reaggregation alone on unwanted cell growth following transplantation remain underexplored.
To this end, SC-α or SC-β cells were dispersed and reaggregated. Following reaggregation, markers specific to outgrowth-driving populations were reduced, along with genes contributing to cell adhesion, such as N-Cadherin (Figure S5B). Five million dispersed and reaggregated SC-α or SC-β cells were transplanted under the renal capsule of immunocompromised mice for 12 weeks (Figure 5A). At 12 weeks post-transplant, a significant reduction in transplant outgrowth size was observed in dispersed and reaggregated cell transplants as compared to the non-dispersed cell transplants for both SC-α and SC-β cells. None of the dispersed and reaggregated SC-α transplants formed any appreciable growths, while all of the non-dispersed control transplants formed noticeable growths (Figure 5B,D). Similarly, four out of nine dispersed and reaggregated SC-β transplants formed no outgrowths, and the remaining five out of nine outgrowths were significantly reduced as compared to non-dispersed control transplants (Figure 5C,E). Together, these results demonstrate that the dispersion and reaggregation of SC-β and SC-α cells prior to transplantation significantly reduce their potential to form outgrowths post-transplantation.

4. Discussion

Cellular outgrowth following stem cell-derived islet cell transplantation is a major concern as stem cell-derived therapies advance to clinical applications. In addition, a lack of clarity regarding the cell types that contribute to transplant outgrowth has hampered efforts to develop targeted strategies to remove or otherwise mitigate these cells. Here, we showed that outgrowths derived from SC-α and SC-β cell populations were the result of three distinct cell populations marked by SOX9, CDX2, or SOX2 expression. None of these populations marked the desired hormone-expressing cell types produced in either differentiation protocol. The developmental patterns of these growth-forming cell populations in SC-α and SC-β differentiations provided insight into the mechanisms underlying their emergence and informed strategies to prevent their formation in future applications.
The transcription factors CDX2 and SOX2 were the markers of the developing midgut and anterior foregut regions of the gut tube endoderm, respectively [43]. The expression of these markers during specification of the pancreatic progenitors in stage 3 of both SC-α and SC-β differentiation suggests insufficient control over the signaling required to either specify the pancreatic region or, perhaps more importantly, inhibit signaling associated with the development of lateral regions of the gut tube in these three-dimensional clusters.
Anterior–posterior patterning of the gut tube, with the associated specification of unique organ domains, was tightly regulated by the joint action of several different signaling pathways, particularly at the foregut/midgut boundary where the pancreatic domain lies. Altogether, WNT, FGF, and BMP signaling acted as posteriorizing signals promoting midgut specification, while the absence or antagonism of these signals promoted anterior foregut fate. Retinoic acid signaling helped to specify the anterior foregut region, from which the pancreas develops [44]. Both SC-α and SC-β protocols leveraged these pathways to achieve pancreatic cell fates, but the continued development of CDX2+ and SOX2+ cells highlights the difficulty of regulating all of these signaling events in a three-dimensional cluster. The identification of approaches to inhibit the expression of non-pancreatic transcription factors, like CDX2 and SOX2, during differentiation might mitigate the need to use disruptive purification protocols, such as antibody sorting or dispersion and reaggregation.
A population of cells marked by the expression of SOX9 also contributed to transplant outgrowth. SOX9 is a marker of progenitor cells in many different endodermal tissues, including the pancreas [28,45,46,47,48]. In early pancreatic development, SOX9 marks a multipotent pancreatic progenitor population along with PDX1, PTF1A, and CPA1 that gives rise to the entire pancreatic epithelium [46]. The multipotent pancreatic progenitor can then give rise to both PTF1A+ acinar-generating tip cells and SOX9+ bipotent trunk progenitor cells, which can differentiate into both ductal and endocrine lineages. While SOX9 is an indispensable transcription factor for pancreatic differentiation, its expression is subsequently restricted to pancreatic ductal cells [49]. The continued expression of SOX9 throughout the islet cell differentiation protocols highlights the limited ability of these differentiation protocols to force the differentiation of the SOX9+ progenitors towards the endocrine lineage. Instead, remnant SOX9+ progenitor cells in both SC-β and SC-α protocols continued to proliferate through the remainder of the differentiation protocols and were eventually transplanted, where they contributed to undesired overgrowth.
Attempting to find a strategy to eliminate these growth-generating populations, we explored the single-cell RNA sequencing data of our SC-α cells and found SOX9+ and CDX2+ cells to be highly proliferative progenitor-like populations enriched in cell-adhesion signaling, such as Notch. Using the Notch inhibitor XXi to disrupt the signaling niche of these cells, we expected a reduction in both SOX9+ and CDX2+ cell populations. Interestingly, while the CDX2+ population was not sensitive to Notch inhibition, treatment with XXi led to a robust reduction in the population of SOX9+ cells, which was consistent with previously identified regulation patterns of SOX9 in pancreatic progenitors [36,37,39].
While it is surprising that XXi treatment did not affect CDX2 expression, given the strong pattern of Notch signaling detected in our scRNA-seq analysis, it might have been the case that CDX2+ progenitors were less dependent on Notch-signaling dynamics for their maintenance as compared to SOX9+ progenitors. The developmental signals regulating the CDX2+ population remained unclear; however, further investigation into midgut-promoting signals may reveal a suitable strategy for disrupting CDX2.
The inhibition of Notch signaling did not impact the SOX2+ population of cells observed in SC-α and SC-β differentiation protocols either. Because this population of cells was not enriched in Notch signaling based on our scRNA-seq analysis, it was unsurprising that no effect was observed. However, similar to the CDX2+ population, future research will be needed to identify signals that might serve to prevent or disrupt this population. Our scRNA-seq analysis identified BMP signaling as a strong SOX2-associated signal, suggesting that the inhibition of this pathway might be one mechanism for reducing or eliminating the SOX2+ population. While both differentiation protocols utilized the BMP-inhibitor LDN 193189, the BMP signaling network is complex, and alternative inhibitors with both broad and specific targeting of this network should be explored in future efforts to prevent SOX2+ cell populations from forming.
Recent work has demonstrated successful enrichment of SC-β cells using sorting-based methods, resulting in reduced outgrowths following transplantation [26,42]. However, given that sorting-based enrichment methods negatively impact endocrine cell yield and viability, and are impractical to implement at scale, alterations to the differentiation protocol itself via the addition of protein factors or small molecules to more precisely specify desired cell populations while eliminating undesired populations are extremely intriguing. Although tremendous advances have been made over the past decade with regard to the purity of the cell populations that can be achieved by pancreatic differentiation protocols, it is unclear if the elimination of all off-target cell populations is technically feasible. Broader strategies, such as the use of bleomycin treatment to eliminate all proliferative cells from an SC-β differentiation, have recently been demonstrated [25]. Bleomycin treatment successfully ablated the SOX9+ population from SC-β cell differentiations. However, it remained unclear how this treatment strategy would impact the CDX2+ population, or other less proliferative off-target populations, such as the SOX2+ population. Thus, more general enrichment strategies that do not rely upon the proliferative state of the cells are desirable.
Here, we demonstrated that a simple mechanical approach, namely the dispersion and reaggregation of differentiated cell clusters, was sufficient on its own to improve the safety profile of both SC-β and SC-α cells. Based upon our scRNA-seq data, we propose that beyond simple enrichment, this process disfavored the continued survival of non-endocrine progenitors that had a high reliance on cell-adhesion signaling. Consistent with our findings, the endocrine enrichment effects of the dispersion and reaggregation of SC-β cells have previously been reported, and such a step is inherent in protocols that begin differentiation in 2D monolayers, followed by a transition to 3D suspension culture [18,19,41,50]. Current dispersion and reaggregation methods can significantly impact total cell yield, often recovering less than 50% of the targeted cell population [42]. Improvements in the efficiency of these processes will be necessary prior to use in translational applications.
Taken together, our results have identified distinct cell populations that contribute to undesired outgrowths from grafts of SC-α and SC-β cell preparations. The nature of these cells, as defined by expression of developmentally regulated transcription factors, provides clues as to how to most effectively modify current differentiation protocols to reduce or eliminate these mis-specified cells. Such modifications, when used in conjunction with simple cell purification techniques, such as dispersion and reaggregation, may result in cell preparations with an improved safety profile for use in research and in the treatment of diabetic patients.

5. Conclusions

This study is the first to demonstrate that transplantation of SC-α cells can give rise to graft outgrowths in a similar manner to SC-β cells. We found that the outgrowths from both cell types were driven by common cell populations that share an enriched cell-adhesion signaling niche. The use of a Notch inhibitor to interrupt cell–cell adhesion signaling led to the disruption of only a subset of the outgrowth-driving cell populations. However, when SC-α and SC-β cells were dissociated into single cells and then reaggregated prior to transplantation, their ability to form outgrowths was diminished. Together, these results suggest that broadly targeting cell-adhesion signaling pathways may more effectively suppress the cell populations that drive outgrowth in both SC-α and SC-β cell products, which could help lead to safer therapeutic interventions in future iterations of these therapies.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/cells15040314/s1. Figure S1: Flow cytometry gating strategy. Flow cytometry data are first gated along forward and side scatter to collect cellular material. Forward scatter height is then used to eliminate doublets before protein expression is characterized using fluorescent height. Figure S2: Enterochromaffin cells do not contribute to cystic outgrowths. Immunofluorescent staining of TPH1 and SOX2 in outgrowths from SC-α (left) or SC-β (right) cell transplants demonstrates a lack of TPH1+ enterochromaffin cells in the outgrowth tissue. Immunofluorescent images are from unique mice transplanted with SC-α or SC-β cells. Scale bars represent 100 μm. Figure S3: CellChat signaling analysis of SC-α cells. Heatmaps of predicted sender and receiver populations for key developmental signaling pathways reveal cell-type-specific signaling patterns in SC-α cells. Figure S4: Single-cell RNA-sequencing of SC-β cells from Veres et al. [19]. (A) UMAP clustering reveals 8 distinct populations, including SOX9+ and CDX2+ progenitor-like populations. (B) Transcriptomic markers of cell populations found in SC-β differentiations. (C) CellChat analysis of incoming signaling patterns reveals upregulation of cell adhesion patterns in SOX9+ and CDX2+ cell populations from SC-β differentiations. (D) SOX9+ and CDX2+ cells are enriched in Notch ligands, receptors, and effectors. Figure S5: Reduction in off-target cells following dispersion and reaggregation. Dispersion and reaggregation enrich for endocrine cell types and cause a reduction in outgrowth-driving cell populations. (A) Flow cytometry quantification of CHGA+ endocrine cells before and after reaggregation of SC-β cells. Reaggregation enriches endocrine cells from 74.85% to 95.47%. (B) Quantitative PCR for outgrowth markers, cadherins, and Notch-signaling members before and after reaggregation. A decreased expression of the outgrowth-drivers and N-Cadherin (CDH2) is observed following reaggregation. Data depicted as mean ± S.E.M. Statistical significance was measured using a paired student’s t-test against the control value of 1. Asterisks indicate p-values < 0.05 (*). Table S1: Media compositions for SC-α and SC-β cell differentiation. Table S2: Differentiation factors for SC-α cell differentiation. Table S3: Differentiation factors for SC-β cell differentiation. Table S4: Antibody information for immunofluorescence and flow cytometry. Table S5: PCR primers used in this study.

Author Contributions

K.R.K., E.W.L., S.L.-B. and Q.P.P. contributed to the conceptualization of experiments and design of methodology for experiments. K.R.K., E.W.L., Z.K., A.M.R.S., S.S. and L.T.J. conducted the investigations and data curation for all experiments performed in this manuscript. K.R.K. and Q.P.P. were responsible for the formal analysis and original writing of this manuscript. All authors were involved in the review and editing of this manuscript. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the generosity of the J.W. Kieckhefer Foundation, the Stephen and Barbara Slaggie Family, the Kenneth Aldridge Foundation, and the Khalifa Bin Zayed Al Nahyan Foundation.

Institutional Review Board Statement

The Mayo Clinic Stem Cell Research Oversight Committee reviewed and approved all work involving human pluripotent stem cells carried out in this manuscript, including the review of compliance with informed consent, under approval code 061917. (Approval Date: 19 July 2017) The cells used in this article were obtained from the Harvard Stem Cell Institute and were derived and consented at Harvard. These cells are available from the HSCI Stem Cell Core. This cell line is a registered NIH Embryonic Stem Cell Line (NIHhESC-09-0021) and, as such, adheres to all NIH guidelines on consent prior to registration. All animal studies were conducted with approval from the appropriate institutional oversight committees, including the Mayo Clinic Institutional Animal Care and Use Committee, under approval code A00006533-22-R25 (Approval Date: 26 January 2022; renewed: 22 January 2025).

Informed Consent Statement

Not applicable.

Data Availability Statement

This paper analyzes existing publicly available single-cell RNA-sequencing data accessible at GSE:138857 and GSE:114412. All other data reported in this paper will be shared by the corresponding author upon request. This paper does not report original code.

Acknowledgments

The authors thank Michael Q. Slama and Shaimaa Hassoun for technical assistance with cell transplants; Jennifer Westendorf and Katherine Arnold for assistance with H&E staining and analysis; Dena E. Cohen for editorial contributions during manuscript preparation; and Aleksey Matveyenko and all the members of the Peterson Lab for helpful discussions. E.W.L. was supported in part by the Clinical and Translational Science Award (CTSA), grant number UL1 TR002377, from the National Center for Advancing Translational Sciences (NCATS).

Conflicts of Interest

Q.P.P. is a scientific board member and shareholder in Mellicell. Q.P.P. is an inventor on intellectual property licensed by Vertex and on several relevant patents held by Harvard University and Mayo Clinic. L.T.J. is an inventor on a relevant patent held by Mayo Clinic. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Abbreviations

The following abbreviations are used in this manuscript:
SC-αStem cell-derived alpha
SC-βStem cell-derived beta
T1DType 1 diabetes
PBSPhosphate-buffered saline
PBSTPhosphate-buffered saline with Tween
PCRPolymerase chain reaction
PFAParaformaldehyde
DAPI4′,6-diamidino-2-phenylindole
H&EHematoxylin and Eosin
SCIDSevere combined immunodeficiency
UMAPUniform Manifold Approximation and Projection

References

  1. Gregory, G.A.; Robinson, T.I.G.; Linklater, S.E.; Wang, F.; Colagiuri, S.; de Beaufort, C.; Donaghue, K.C.; International Diabetes Federation Diabetes Atlas Type 1 Diabetes in Adults Special Interest Group; Magliano, D.J.; Maniam, J.; et al. Global incidence, prevalence, and mortality of type 1 diabetes in 2021 with projection to 2040: A modelling study. Lancet Diabetes Endocrinol. 2022, 10, 741–760. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  2. Ahren, B. Beta- and alpha-cell dysfunction in subjects developing impaired glucose tolerance: Outcome of a 12-year prospective study in postmenopausal Caucasian women. Diabetes 2009, 58, 726–731. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  3. Brissova, M.; Haliyur, R.; Saunders, D.; Shrestha, S.; Dai, C.; Blodgett, D.M.; Bottino, R.; Campbell-Thompson, M.; Aramandla, R.; Poffenberger, G.; et al. Alpha Cell Function and Gene Expression Are Compromised in Type 1 Diabetes. Cell Rep. 2018, 22, 2667–2676. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  4. Jamison, R.A.; Stark, R.; Dong, J.; Yonemitsu, S.; Zhang, D.; Shulman, G.I.; Kibbey, R.G. Hyperglucagonemia precedes a decline in insulin secretion and causes hyperglycemia in chronically glucose-infused rats. Am. J. Physiol. Endocrinol. Metab. 2011, 301, E1174–E1183. [Google Scholar] [CrossRef] [Scilit]
  5. Doliba, N.M.; Rozo, A.V.; Roman, J.; Qin, W.; Traum, D.; Gao, L.; Liu, J.; Manduchi, E.; Liu, C.; Golson, M.L.; et al. alpha Cell dysfunction in islets from nondiabetic, glutamic acid decarboxylase autoantibody-positive individuals. J. Clin. Investig. 2022, 132, e156243. [Google Scholar] [CrossRef] [Scilit]
  6. Pettway, Y.D.; Saunders, D.C.; Brissova, M. The human alpha cell in health and disease. J. Endocrinol. 2023, 258, e220298. [Google Scholar] [CrossRef] [Scilit]
  7. de Klerk, E.; Hebrok, M. Stem Cell-Based Clinical Trials for Diabetes Mellitus. Front. Endocrinol. 2021, 12, 631463. [Google Scholar] [CrossRef] [Scilit]
  8. Cayabyab, F.; Nih, L.R.; Yoshihara, E. Advances in Pancreatic Islet Transplantation Sites for the Treatment of Diabetes. Front. Endocrinol. 2021, 12, 732431. [Google Scholar] [CrossRef] [Scilit]
  9. Mailliez, A.; Ternynck, C.; Jannin, A.; Lemaitre, M.; Chevalier, B.; Le Mapihan, K.; Defrance, F.; Mackowiak, M.A.; Rollin, A.; Mehdi, M.; et al. Cognitive Outcome After Islet Transplantation. Transplant. Direct 2023, 9, e1493. [Google Scholar] [CrossRef] [Scilit]
  10. Ogieuhi, I.J.; Agbo, C.E.; Ajekiigbe, V.O.; Anthony, C.S.; Onyehalu, J.C.; Nwankwo, C.K.; Agu, M.C.; Lema, K.; Adewole, O.A.; Asade, O.; et al. Stem cell-derived pancreatic beta cells: A step closer to functional diabetes treatment? BMC Endocr. Disord. 2025, 25, 181. [Google Scholar] [CrossRef] [Scilit]
  11. D’Amour, K.A.; Bang, A.G.; Eliazer, S.; Kelly, O.G.; Agulnick, A.D.; Smart, N.G.; Moorman, M.A.; Kroon, E.; Carpenter, M.K.; Baetge, E.E. Production of pancreatic hormone-expressing endocrine cells from human embryonic stem cells. Nat. Biotechnol. 2006, 24, 1392–1401. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  12. D’Amour, K.A.; Agulnick, A.D.; Eliazer, S.; Kelly, O.G.; Kroon, E.; Baetge, E.E. Efficient differentiation of human embryonic stem cells to definitive endoderm. Nat. Biotechnol. 2005, 23, 1534–1541. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  13. Schulz, T.C.; Young, H.Y.; Agulnick, A.D.; Babin, M.J.; Baetge, E.E.; Bang, A.G.; Bhoumik, A.; Cepa, I.; Cesario, R.M.; Haakmeester, C.; et al. A scalable system for production of functional pancreatic progenitors from human embryonic stem cells. PLoS ONE 2012, 7, e37004. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  14. Shapiro, A.M.J.; Thompson, D.; Donner, T.W.; Bellin, M.D.; Hsueh, W.; Pettus, J.; Wilensky, J.; Daniels, M.; Wang, R.M.; Brandon, E.P.; et al. Insulin expression and C-peptide in type 1 diabetes subjects implanted with stem cell-derived pancreatic endoderm cells in an encapsulation device. Cell Rep. Med. 2021, 2, 100466. [Google Scholar] [CrossRef] [Scilit]
  15. Ramzy, A.; Thompson, D.M.; Ward-Hartstonge, K.A.; Ivison, S.; Cook, L.; Garcia, R.V.; Loyal, J.; Kim, P.T.W.; Warnock, G.L.; Levings, M.K.; et al. Implanted pluripotent stem-cell-derived pancreatic endoderm cells secrete glucose-responsive C-peptide in patients with type 1 diabetes. Cell Stem Cell 2021, 28, 2047–2061.e5. [Google Scholar] [CrossRef] [Scilit]
  16. Pagliuca, F.W.; Millman, J.R.; Gurtler, M.; Segel, M.; Van Dervort, A.; Ryu, J.H.; Peterson, Q.P.; Greiner, D.; Melton, D.A. Generation of functional human pancreatic beta cells in vitro. Cell 2014, 159, 428–439. [Google Scholar] [CrossRef] [Scilit]
  17. Rezania, A.; Bruin, J.E.; Arora, P.; Rubin, A.; Batushansky, I.; Asadi, A.; O’Dwyer, S.; Quiskamp, N.; Mojibian, M.; Albrecht, T.; et al. Reversal of diabetes with insulin-producing cells derived in vitro from human pluripotent stem cells. Nat. Biotechnol. 2014, 32, 1121–1133. [Google Scholar] [CrossRef] [Scilit]
  18. Hogrebe, N.J.; Maxwell, K.G.; Augsornworawat, P.; Millman, J.R. Generation of insulin-producing pancreatic beta cells from multiple human stem cell lines. Nat. Protoc. 2021, 16, 4109–4143. [Google Scholar] [CrossRef] [Scilit]
  19. Veres, A.; Faust, A.L.; Bushnell, H.L.; Engquist, E.N.; Kenty, J.H.; Harb, G.; Poh, Y.C.; Sintov, E.; Gurtler, M.; Pagliuca, F.W.; et al. Charting cellular identity during human in vitro beta-cell differentiation. Nature 2019, 569, 368–373. [Google Scholar] [CrossRef] [Scilit]
  20. Reichman, T.W.; Markmann, J.F.; Odorico, J.; Witkowski, P.; Fung, J.J.; Wijkstrom, M.; Kandeel, F.; de Koning, E.J.P.; Peters, A.L.; Mathieu, C.; et al. Stem Cell-Derived, Fully Differentiated Islets for Type 1 Diabetes. N. Engl. J. Med. 2025, 393, 858–868. [Google Scholar] [CrossRef] [Scilit]
  21. Jensen, N.K.; Ingvorsen, C.; Petersen, D.R.; Pereira, M.J.; Lu, T.T.H.; Alsted, T.J.; Kirkegaard, J.S.; Keane, K.A. Characterization of the Nonendocrine Cell Populations in Human Embryonic Stem Cell-Derived (hESC) Islet-Like Clusters Posttransplantation. Toxicol. Pathol. 2021, 49, 1269–1287. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  22. Rezania, A.; Bruin, J.E.; Riedel, M.J.; Mojibian, M.; Asadi, A.; Xu, J.; Gauvin, R.; Narayan, K.; Karanu, F.; O’Neil, J.J.; et al. Maturation of human embryonic stem cell-derived pancreatic progenitors into functional islets capable of treating pre-existing diabetes in mice. Diabetes 2012, 61, 2016–2029. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  23. Bruin, J.E.; Rezania, A.; Xu, J.; Narayan, K.; Fox, J.K.; O’Neil, J.J.; Kieffer, T.J. Maturation and function of human embryonic stem cell-derived pancreatic progenitors in macroencapsulation devices following transplant into mice. Diabetologia 2013, 56, 1987–1998. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  24. Pepper, A.R.; Bruni, A.; Pawlick, R.; O’Gorman, D.; Kin, T.; Thiesen, A.; Shapiro, A.M.J. Posttransplant Characterization of Long-term Functional hESC-Derived Pancreatic Endoderm Grafts. Diabetes 2019, 68, 953–962. [Google Scholar] [CrossRef] [Scilit]
  25. Shilleh, A.H.; Beard, S.; Russ, H.A. Enrichment of stem cell-derived pancreatic beta-like cells and controlled graft size through pharmacological removal of proliferating cells. Stem Cell Rep. 2023, 18, 1284–1294. [Google Scholar] [CrossRef] [Scilit]
  26. Aghazadeh, Y.; Sarangi, F.; Poon, F.; Nkennor, B.; McGaugh, E.C.; Nunes, S.S.; Nostro, M.C. GP2-enriched pancreatic progenitors give rise to functional beta cells in vivo and eliminate the risk of teratoma formation. Stem Cell Rep. 2022, 17, 964–978. [Google Scholar] [CrossRef] [Scilit]
  27. Cuesta-Gomez, N.; Verhoeff, K.; Jasra, I.T.; Pawlick, R.; Dadheech, N.; Shapiro, A.M.J. Characterization of stem-cell-derived islets during differentiation and after implantation. Cell Rep. 2022, 40, 111238. [Google Scholar] [CrossRef] [Scilit]
  28. Lynn, F.C.; Smith, S.B.; Wilson, M.E.; Yang, K.Y.; Nekrep, N.; German, M.S. Sox9 coordinates a transcriptional network in pancreatic progenitor cells. Proc. Natl. Acad. Sci. USA 2007, 104, 10500–10505. [Google Scholar] [CrossRef] [Scilit]
  29. Cote, F.; Thevenot, E.; Fligny, C.; Fromes, Y.; Darmon, M.; Ripoche, M.A.; Bayard, E.; Hanoun, N.; Saurini, F.; Lechat, P.; et al. Disruption of the nonneuronal tph1 gene demonstrates the importance of peripheral serotonin in cardiac function. Proc. Natl. Acad. Sci. USA 2003, 100, 13525–13530. [Google Scholar] [CrossRef] [Scilit]
  30. Peterson, Q.P.; Veres, A.; Chen, L.; Slama, M.Q.; Kenty, J.H.R.; Hassoun, S.; Brown, M.R.; Dou, H.; Duffy, C.D.; Zhou, Q.; et al. A method for the generation of human stem cell-derived alpha cells. Nat. Commun. 2020, 11, 2241. [Google Scholar] [CrossRef] [Scilit]
  31. Shrestha, S.; Jennings, L.T.; Knofczynski, K.; Shivakumar, S.B.; Peterson, Q.P. Modeling diabetic alpha cell dysfunction using stem cell-derived alpha cells. Stem Cell Rep. 2025, 20, 102504. [Google Scholar] [CrossRef] [Scilit]
  32. Salomon, T.; Knofczynski, K.R.; Shrestha, S.; Peterson, Q.P. Protocol to generate stem cell-derived alpha cells in 3D suspension culture. STAR Protoc. 2025, 6, 104063. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  33. Hao, Y.; Stuart, T.; Kowalski, M.H.; Choudhary, S.; Hoffman, P.; Hartman, A.; Srivastava, A.; Molla, G.; Madad, S.; Fernandez-Granda, C.; et al. Dictionary learning for integrative, multimodal and scalable single-cell analysis. Nat. Biotechnol. 2024, 42, 293–304. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  34. Augsornworawat, P.; Hogrebe, N.J.; Ishahak, M.; Schmidt, M.D.; Marquez, E.; Maestas, M.M.; Veronese-Paniagua, D.A.; Gale, S.E.; Miller, J.R.; Velazco-Cruz, L.; et al. Single-nucleus multi-omics of human stem cell-derived islets identifies deficiencies in lineage specification. Nat. Cell Biol. 2023, 25, 904–916. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  35. Jin, S.; Guerrero-Juarez, C.F.; Zhang, L.; Chang, I.; Ramos, R.; Kuan, C.H.; Myung, P.; Plikus, M.V.; Nie, Q. Inference and analysis of cell-cell communication using CellChat. Nat. Commun. 2021, 12, 1088. [Google Scholar] [CrossRef] [Scilit]
  36. Fujikura, J.; Hosoda, K.; Iwakura, H.; Tomita, T.; Noguchi, M.; Masuzaki, H.; Tanigaki, K.; Yabe, D.; Honjo, T.; Nakao, K. Notch/Rbp-j signaling prevents premature endocrine and ductal cell differentiation in the pancreas. Cell Metab. 2006, 3, 59–65. [Google Scholar] [CrossRef] [Scilit]
  37. Apelqvist, A.; Li, H.; Sommer, L.; Beatus, P.; Anderson, D.J.; Honjo, T.; Hrabe de Angelis, M.; Lendahl, U.; Edlund, H. Notch signalling controls pancreatic cell differentiation. Nature 1999, 400, 877–881. [Google Scholar] [CrossRef] [Scilit]
  38. Zhou, B.; Lin, W.; Long, Y.; Yang, Y.; Zhang, H.; Wu, K.; Chu, Q. Notch signaling pathway: Architecture, disease, and therapeutics. Signal Transduct. Target. Ther. 2022, 7, 95. [Google Scholar] [CrossRef] [Scilit]
  39. Seymour, P.A.; Collin, C.A.; Egeskov-Madsen, A.R.; Jorgensen, M.C.; Shimojo, H.; Imayoshi, I.; de Lichtenberg, K.H.; Kopan, R.; Kageyama, R.; Serup, P. Jag1 Modulates an Oscillatory Dll1-Notch-Hes1 Signaling Module to Coordinate Growth and Fate of Pancreatic Progenitors. Dev. Cell 2020, 52, 731–747.e8. [Google Scholar] [CrossRef] [Scilit]
  40. Dhanesh, S.B.; Subashini, C.; James, J. Hes1: The maestro in neurogenesis. Cell. Mol. Life Sci. 2016, 73, 4019–4042. [Google Scholar] [CrossRef] [Scilit]
  41. Nair, G.G.; Liu, J.S.; Russ, H.A.; Tran, S.; Saxton, M.S.; Chen, R.; Juang, C.; Li, M.L.; Nguyen, V.Q.; Giacometti, S.; et al. Recapitulating endocrine cell clustering in culture promotes maturation of human stem-cell-derived beta cells. Nat. Cell Biol. 2019, 21, 263–274. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  42. Parent, A.V.; Ashe, S.; Nair, G.G.; Li, M.L.; Chavez, J.; Liu, J.S.; Zhong, Y.; Streeter, P.R.; Hebrok, M. Development of a scalable method to isolate subsets of stem cell-derived pancreatic islet cells. Stem Cell Rep. 2022, 17, 979–992. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  43. Sherwood, R.I.; Chen, T.Y.; Melton, D.A. Transcriptional dynamics of endodermal organ formation. Dev. Dyn. 2009, 238, 29–42. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  44. Zorn, A.M.; Wells, J.M. Vertebrate endoderm development and organ formation. Annu. Rev. Cell Dev. Biol. 2009, 25, 221–251. [Google Scholar] [CrossRef] [Scilit]
  45. Jo, A.; Denduluri, S.; Zhang, B.; Wang, Z.; Yin, L.; Yan, Z.; Kang, R.; Shi, L.L.; Mok, J.; Lee, M.J.; et al. The versatile functions of Sox9 in development, stem cells, and human diseases. Genes Dis. 2014, 1, 149–161. [Google Scholar] [CrossRef] [Scilit]
  46. Seymour, P.A. Sox9: A master regulator of the pancreatic program. Rev. Diabet. Stud. 2014, 11, 51–83. [Google Scholar] [CrossRef] [Scilit]
  47. Jarc, L.; Bandral, M.; Zanfrini, E.; Lesche, M.; Kufrin, V.; Sendra, R.; Pezzolla, D.; Giannios, I.; Khattak, S.; Neumann, K.; et al. Regulation of multiple signaling pathways promotes the consistent expansion of human pancreatic progenitors in defined conditions. eLife 2024, 12, RP89962. [Google Scholar] [CrossRef]
  48. Kopp, J.L.; Dubois, C.L.; Hao, E.; Thorel, F.; Herrera, P.L.; Sander, M. Progenitor cell domains in the developing and adult pancreas. Cell Cycle 2011, 10, 1921–1927. [Google Scholar] [CrossRef] [Scilit]
  49. Seymour, P.A.; Freude, K.K.; Tran, M.N.; Mayes, E.E.; Jensen, J.; Kist, R.; Scherer, G.; Sander, M. SOX9 is required for maintenance of the pancreatic progenitor cell pool. Proc. Natl. Acad. Sci. USA 2007, 104, 1865–1870. [Google Scholar] [CrossRef] [Scilit]
  50. Liu, H.; Li, R.; Liao, H.K.; Min, Z.; Wang, C.; Yu, Y.; Shi, L.; Dan, J.; Hayek, A.; Martinez Martinez, L.; et al. Chemical combinations potentiate human pluripotent stem cell-derived 3D pancreatic progenitor clusters toward functional beta cells. Nat. Commun. 2021, 12, 3330. [Google Scholar] [CrossRef] [Scilit]
Cells 15 00314 g001
Figure 1. Transplantation of SC-α or SC-β cells into immunocompromised mice results in cystic outgrowth of the transplant site after 12 weeks. (A) Schematic depicting transcription factor expression of the developing gut tube endoderm and how inefficiencies in each differentiation protocol may lead to CDX2+- and SOX2+-expressing off-target cells. Created in BioRender. Knofczynski, K. (2026) https://BioRender.com/4q7kigf. (B) Gross images and H&E staining of outgrown kidneys from SC-α (left) and SC-β (right) transplants. Growths have large cystic structures of various morphologies. Gross images and H&E staining are representative images from n = 5 mice following SC-α or SC-β transplantation. Numbered regions in each image are magnified in panels (C,D). Magnified images of H&E-stained growths demonstrate 3 unique growth morphologies consistent with 3 off-target lineage markers (SOX9, CDX2, and SOX2) visualized via immunofluorescence in both SC-α (C) and SC-β (D) cells. IF images are representative images from n = 3 mice following SC-α or SC-β transplantation. H&E scale bars represent 5 mm (B) or 400 μm (C,D). Immunofluorescence scale bars represent 100 μm.
Figure 1. Transplantation of SC-α or SC-β cells into immunocompromised mice results in cystic outgrowth of the transplant site after 12 weeks. (A) Schematic depicting transcription factor expression of the developing gut tube endoderm and how inefficiencies in each differentiation protocol may lead to CDX2+- and SOX2+-expressing off-target cells. Created in BioRender. Knofczynski, K. (2026) https://BioRender.com/4q7kigf. (B) Gross images and H&E staining of outgrown kidneys from SC-α (left) and SC-β (right) transplants. Growths have large cystic structures of various morphologies. Gross images and H&E staining are representative images from n = 5 mice following SC-α or SC-β transplantation. Numbered regions in each image are magnified in panels (C,D). Magnified images of H&E-stained growths demonstrate 3 unique growth morphologies consistent with 3 off-target lineage markers (SOX9, CDX2, and SOX2) visualized via immunofluorescence in both SC-α (C) and SC-β (D) cells. IF images are representative images from n = 3 mice following SC-α or SC-β transplantation. H&E scale bars represent 5 mm (B) or 400 μm (C,D). Immunofluorescence scale bars represent 100 μm.
Cells 15 00314 g001
Cells 15 00314 g002
Figure 2. SOX9, CDX2, and SOX2 expression is observed in pre-transplant SC-β and SC-α cells. (A) Quantitative PCR results for SOX9, CDX2, and SOX2 for all stages of both SC-β and SC-α differentiation protocols. PCR was performed on 3 independent differentiations at all differentiation stages for SC-α and SC-β cells each. For each differentiation, PCR was performed in triplicate, with the average triplicate value being used to calculate ΔΔCq values. (B) Representative immunofluorescent expression patterns for SOX9, CDX2, and SOX2 against C-PEP+ SC-β and GCG+ SC-α cells at the end of each differentiation protocol. Images are representative of 3 independent differentiations for both SC-α and SC-β cells. Immunofluorescence scale bars represent 100 μm. Data depicted as mean ± S.E.M. Statistical significance was measured using one-way repeated measures ANOVA with Dunnett’s multiple comparison tests. Asterisks indicate p-values < 0.05 (*) or <0.01 (**).
Figure 2. SOX9, CDX2, and SOX2 expression is observed in pre-transplant SC-β and SC-α cells. (A) Quantitative PCR results for SOX9, CDX2, and SOX2 for all stages of both SC-β and SC-α differentiation protocols. PCR was performed on 3 independent differentiations at all differentiation stages for SC-α and SC-β cells each. For each differentiation, PCR was performed in triplicate, with the average triplicate value being used to calculate ΔΔCq values. (B) Representative immunofluorescent expression patterns for SOX9, CDX2, and SOX2 against C-PEP+ SC-β and GCG+ SC-α cells at the end of each differentiation protocol. Images are representative of 3 independent differentiations for both SC-α and SC-β cells. Immunofluorescence scale bars represent 100 μm. Data depicted as mean ± S.E.M. Statistical significance was measured using one-way repeated measures ANOVA with Dunnett’s multiple comparison tests. Asterisks indicate p-values < 0.05 (*) or <0.01 (**).
Cells 15 00314 g002
Cells 15 00314 g003
Figure 3. Transcriptomic characterization of growth-generating cell populations. (A) UMAP plot of SC-α cells. Seven unique cell populations are identified that fall into four broad categories, including progenitor-like, mesenchymal, pancreatic endocrine, and gastric endocrine. (B) Transcriptomic markers of cell populations found in SC-α differentiations. (C) Cell cycle analysis of SC-α cell populations reveals that progenitor-like populations are more proliferative than endocrine populations. (D) CellChat heatmap of the top 10 strongest incoming signals in SC-α cell populations. Among the top 10 signals are many cell-adhesion signaling pathways, which are enriched in progenitor-like populations. (E) Dot plot of Notch-signaling members in SC-α cell populations demonstrates upregulation of Notch ligands, receptors, and effectors in progenitor-like populations.
Figure 3. Transcriptomic characterization of growth-generating cell populations. (A) UMAP plot of SC-α cells. Seven unique cell populations are identified that fall into four broad categories, including progenitor-like, mesenchymal, pancreatic endocrine, and gastric endocrine. (B) Transcriptomic markers of cell populations found in SC-α differentiations. (C) Cell cycle analysis of SC-α cell populations reveals that progenitor-like populations are more proliferative than endocrine populations. (D) CellChat heatmap of the top 10 strongest incoming signals in SC-α cell populations. Among the top 10 signals are many cell-adhesion signaling pathways, which are enriched in progenitor-like populations. (E) Dot plot of Notch-signaling members in SC-α cell populations demonstrates upregulation of Notch ligands, receptors, and effectors in progenitor-like populations.
Cells 15 00314 g003
Cells 15 00314 g004
Figure 4. Notch inhibition reduces SOX9 but not CDX2 or SOX2 expression. (A) Dose response of SOX9, CDX2, and SOX2 gene expression, measured by qPCR, to XXi treatment in SC-α (red) and SC-β (purple) cells. Dose response curve depicted in solid lines with 95% confidence intervals depicted in dashed lines. Expression of SOX9 was dependent on XXi concentration with IC50 values of 56.1 nM (SC-α) and 34.8 nM (SC-β). Attempts to generate dose–response curves for CDX2 and SOX2 failed to produce meaningful curves. PCR was performed on 3 independent differentiations for SC-α and SC-β cells at each concentration tested. For each condition, PCR was performed in triplicate, with the average triplicate value being used to calculate ΔΔCq values. (B) Flow cytometry analysis of SOX9+ population of SC-α (triangle) and SC-β (circle) cell products following XXi treatment. A decline in the proportion of SOX9+ cells was observed following XXi treatments, consistent with the trends observed in the RNA. Flow cytometry was performed on 3 independent differentiations for SC-α and SC-β cells at each concentration tested. (C) Flow cytometry analysis of SC-α (triangle) and SC-β (circle) cell products from respective protocols. XXi treatment did not significantly alter the total number of targeted endocrine cell populations. Flow cytometry was performed on 3 independent differentiations for SC-α and SC-β cells at each concentration tested. Data depicted as mean ± S.E.M. Statistical significance was measured using a paired two-tailed Student’s t-test. Asterisks indicate p-values < 0.05 (*) or <0.01 (**).
Figure 4. Notch inhibition reduces SOX9 but not CDX2 or SOX2 expression. (A) Dose response of SOX9, CDX2, and SOX2 gene expression, measured by qPCR, to XXi treatment in SC-α (red) and SC-β (purple) cells. Dose response curve depicted in solid lines with 95% confidence intervals depicted in dashed lines. Expression of SOX9 was dependent on XXi concentration with IC50 values of 56.1 nM (SC-α) and 34.8 nM (SC-β). Attempts to generate dose–response curves for CDX2 and SOX2 failed to produce meaningful curves. PCR was performed on 3 independent differentiations for SC-α and SC-β cells at each concentration tested. For each condition, PCR was performed in triplicate, with the average triplicate value being used to calculate ΔΔCq values. (B) Flow cytometry analysis of SOX9+ population of SC-α (triangle) and SC-β (circle) cell products following XXi treatment. A decline in the proportion of SOX9+ cells was observed following XXi treatments, consistent with the trends observed in the RNA. Flow cytometry was performed on 3 independent differentiations for SC-α and SC-β cells at each concentration tested. (C) Flow cytometry analysis of SC-α (triangle) and SC-β (circle) cell products from respective protocols. XXi treatment did not significantly alter the total number of targeted endocrine cell populations. Flow cytometry was performed on 3 independent differentiations for SC-α and SC-β cells at each concentration tested. Data depicted as mean ± S.E.M. Statistical significance was measured using a paired two-tailed Student’s t-test. Asterisks indicate p-values < 0.05 (*) or <0.01 (**).
Cells 15 00314 g004
Cells 15 00314 g005
Figure 5. Dispersion and reaggregation of SC-α or SC-β cells decreases outgrowth post-transplantation. (A) Schematic detailing dispersion and reaggregation timeline prior to transplantation into immunocompromised mice. Created in BioRender. Knofczynski, K. (2026) https://BioRender.com/4q7kigf. (B,C) Representative images of outgrowths generated by undispersed SC-α (B) and SC-β (C) cells and those generated by dispersed and reaggregated cells prior to transplantation. (D,E) Quantification of outgrowth volumes in control (circle) or reaggregated (triangle) SC-α (D) and SC-β (E) transplants reveals significant decrease in outgrowth volume following dispersion and reaggregation. For panels (B–E), mouse numbers are as follows: n = 5 (SC-α control transplant), n = 5 (SC-β control transplant), n = 9 (SC-α reaggregated transplant), and n = 9 (SC-β reaggregated transplant). Tumor volume quantification strategy is reported in Section 2. Data depicted as mean ± S.E.M. Statistical significance was measured using an unpaired two-tailed Student’s t-test. Asterisks indicate p-values < 0.05 (*).
Figure 5. Dispersion and reaggregation of SC-α or SC-β cells decreases outgrowth post-transplantation. (A) Schematic detailing dispersion and reaggregation timeline prior to transplantation into immunocompromised mice. Created in BioRender. Knofczynski, K. (2026) https://BioRender.com/4q7kigf. (B,C) Representative images of outgrowths generated by undispersed SC-α (B) and SC-β (C) cells and those generated by dispersed and reaggregated cells prior to transplantation. (D,E) Quantification of outgrowth volumes in control (circle) or reaggregated (triangle) SC-α (D) and SC-β (E) transplants reveals significant decrease in outgrowth volume following dispersion and reaggregation. For panels (B–E), mouse numbers are as follows: n = 5 (SC-α control transplant), n = 5 (SC-β control transplant), n = 9 (SC-α reaggregated transplant), and n = 9 (SC-β reaggregated transplant). Tumor volume quantification strategy is reported in Section 2. Data depicted as mean ± S.E.M. Statistical significance was measured using an unpaired two-tailed Student’s t-test. Asterisks indicate p-values < 0.05 (*).
Cells 15 00314 g005
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Knofczynski, K.R.; Law, E.W.; Lewis-Brinkman, S.; Khashim, Z.; Schornack, A.M.R.; Shrestha, S.; Jennings, L.T.; Peterson, Q.P. Disruption of Cell-Adhesion Signaling Resolves Unwanted Progenitor Specification in Stem Cell-Derived α and β Cell Grafts. Cells 2026, 15, 314. https://doi.org/10.3390/cells15040314

AMA Style

Knofczynski KR, Law EW, Lewis-Brinkman S, Khashim Z, Schornack AMR, Shrestha S, Jennings LT, Peterson QP. Disruption of Cell-Adhesion Signaling Resolves Unwanted Progenitor Specification in Stem Cell-Derived α and β Cell Grafts. Cells. 2026; 15(4):314. https://doi.org/10.3390/cells15040314

Chicago/Turabian Style

Knofczynski, Kyle R., Ethan W. Law, Sean Lewis-Brinkman, Zenith Khashim, Anna Marie R. Schornack, Swikriti Shrestha, Lauren T. Jennings, and Quinn P. Peterson. 2026. "Disruption of Cell-Adhesion Signaling Resolves Unwanted Progenitor Specification in Stem Cell-Derived α and β Cell Grafts" Cells 15, no. 4: 314. https://doi.org/10.3390/cells15040314

APA Style

Knofczynski, K. R., Law, E. W., Lewis-Brinkman, S., Khashim, Z., Schornack, A. M. R., Shrestha, S., Jennings, L. T., & Peterson, Q. P. (2026). Disruption of Cell-Adhesion Signaling Resolves Unwanted Progenitor Specification in Stem Cell-Derived α and β Cell Grafts. Cells, 15(4), 314. https://doi.org/10.3390/cells15040314

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop