Abstract
Pancreatic progenitors derived from human embryonic stem cells (hESCs) are now in clinical trials for insulin replacement in patients with type 1 diabetes. Animal studies indicate that pancreatic progenitor cells can mature into a mixed population of endocrine cells, including glucose-responsive β cells several months after implantion. However, it remains unclear how conditions in the recipient may influence the maturation and ultimately the function of these hESC-derived cells. Here, we investigated the effects of (1) pregnancy on the maturation of human stage 4 (S4) pancreatic progenitor cells and (2) the impact of host sex on both S4 cells and more mature stage 7 (S7) pancreatic endocrine cells implanted under the kidney capsule of immunodeficient SCID-beige mice. Pregnancy led to increased proliferation of endogenous pancreatic β cells, but did not appear to affect proliferation or maturation of S4 cells at midgestation. Interestingly, S4 and S7 cells both acquired glucose-stimulated C-peptide secretion in females before males. Moreover, S4 cells lowered fasting blood glucose levels in females sooner than in males, whereas the responses with S7 cells were similar. These data indicate that the host sex may impact the maturation of hESC-derived cells in vivo and that this effect can be minimized by more advanced differentiation of the cells before implantation.
Type 1 diabetes (T1D) occurs when the pancreas does not produce sufficient insulin because of the autoimmune destruction of β cells. Islet transplantation is effective at restoring glucose homeostasis in patients with T1D (1, 2) but is limited by the scarceness of organ donors and need for chronic immunosuppression. Pluripotent human embryonic stem cells (hESCs) can be expanded and differentiated toward specialized cells in culture so, in theory, could be used to produce β cells on a large enough scale to overcome the lack of cadaveric islets (3). We previously published multistep protocols that efficiently convert hESCs into a population of immature pancreatic cells, including multipotent pancreatic progenitors [stage 4 (S4) pancreatic progenitor cells] in vitro (4–7). Several months after implantation in immunodeficient male rodents, these progenitor cells develop into mature glucose-responsive insulin-secreting cells and are capable of reversing diabetes in mice induced by either streptozotocin administration (5–7) or high-fat diet feeding (8). Alternatively, hESCs can be differentiated into more mature insulin-secreting cells in vitro [stage 7 (S7) endocrine cells] that are able to reverse diabetes even more quickly and with implantation of fewer cells (9). It is important to assess how in vivo conditions may promote or dysregulate the maturation of either S4 or S7 cells after implantation (10). For example, chronic hypothyroidism impaired hESC-derived S4 progenitor cell maturation in male mice (11), and S4 cells matured significantly faster in immunodeficient male rats compared with immunodeficient male mice (12).
Pregnancy is a unique period during adulthood in which pancreatic β-cell mass transiently expands to compensate for the increased metabolic demands of the mother and to support growth of the developing fetus (13). Pregnancy hormones (e.g., prolactin, placental lactogen, serotonin) act directly on adult mouse β cells to promote proliferation (14–16) and thus could act on differentiating human β cells in a similar manner. However, it has been shown that the prolactin receptor gene is enriched in mouse β cells compared with human β cells (17), meaning that prolactin-induced β-cell proliferation may not occur in humans. Although expansion of β-cell mass during pregnancy is transient in adults, expansion of hESC-derived pancreatic progenitor cells during pregnancy might have lasting effects (such as larger grafts with more β cells). Therefore, we investigated the impact of pregnancy on the development and function of hESC-derived S4 pancreatic progenitor cells in vivo.
This article also reports a detailed assessment of glucose-responsive insulin secretion from pancreatic progenitor cells implanted into male vs female mice. ViaCyte, Inc., is conducting clinical trials using macroencapsulated hESC-derived pancreatic progenitor cells in both male and female patients with T1D; therefore, we were interested in directly comparing the development and function of hESC-derived cells (both S4 and S7) after implantation into both male and female mice. We hypothesized that S4 cells might be more susceptible to variability in the host milieu than S7 cells because they are less differentiated and may be more responsive to in vivo signals that promote or hinder their development. At the time of implantation, S4 cells are composed of mainly immature endocrine progenitor cells (6), whereas S7 cells contain more mature endocrine cells that are mainly insulin positive (9). Our studies demonstrate that pregnancy (at 4 weeks after implantation) had no effect on the maturation of hESC-derived S4 cells in mice, but surprisingly the in vivo maturation of both S4 and S7 cells into glucose-responsive insulin-secreting cells was accelerated in female compared with male mice.
Materials and Methods
In vitro differentiation of hESCs and assessment of S4 pancreatic progenitor cells and S7 endocrine cells
All experiments with H1 cells at the University of British Columbia (UBC) were approved by the Canadian Stem Cell Oversight Committee and the UBC Clinical Research Ethics Board. Pluripotent H1 cells (WiCell Research Institute, Inc., Madison, WI) were differentiated into S4 and S7 according to our previously published S4 (5) and S7 (9) protocols. In the S4 protocol, embryonic stem cells were cultured in Tesr1 media (STEMCELL Technologies, Vancouver, BC, Canada) and differentiated to S4 and then released from monolayer as <100-μm clusters and reaggregated in suspension culture in spinner flasks for additional days (5). The S7 cells were prepared to reduce the number of polyhormonal cells and truncate the time for differentiation. In particular, stage 2 was 2 days instead of 3 used in our previous recipe (6), stage 3 in 2 days instead of 4, and S4 in 3 days instead of 5. Additional optimizations as compared to our previous recipes (6) included: (1) serum-free conditions; (2) the replacement of Activin A and Wnt3A with GDF8 and GSK3β inhibitor at stage 1; and (3) the replacement of Noggin with LDN, a small molecule inhibitor of the BMP receptor, and the addition of Activin A and FGF7 at stage 3 (9). In the new protocol, vitamin C is added during stages 2 through 4 (9). Further details about the growth factors and small molecules used for differentiation are described in our previously published protocols (5, 9). S4 and S7 cells were assessed by flow cytometry and immunostaining before implantation, as previously described (6); antibody information is provided in Supplemental Tables 1 and 2, where antibodies are organized based on their application (flow cytometry vs immunostaining).
Animals
Male and female SCID-beige mice (C.B-Igh-1b/GbmsTac-Prkdcscid-LystbgN7; Taconic, Hudson, NY) were maintained on a 12-hour light/dark cycle throughout the study. All mice received ad libitum access to a standard irradiated diet (Teklad Diet No. 2918; Harlan Laboratories, Madison, WI). All experiments were carried out in accordance with the Canadian Council on Animal Care guidelines and were approved by the UBC Animal Care Committee.
Implantation of hESC-derived cells
All mice were anesthetized with inhalable isoflurane and implant recipients received either ∼5 × 106 S4 cells or ∼1 × 106 S7 cells under the left kidney capsule. In the first cohort, cells were implanted in males at 8 to 9 weeks of age and females at 11 weeks. Mice were randomly assigned to either receive an S4 cell implant (n = 6 males and n = 30 females) or sham surgery (n = 8 females). Sham control mice received the same surgical procedure, but no cells were implanted. In the second cohort, cells were implanted in males and females at 8 weeks of age. Mice were randomly assigned to either receive an S4 (n = 11 males and n = 10 females) or S7 cell implant (n = 9 males and n = 9 females).
Breeding
In cohort 1, male breeders (n = 6) at 12 to 13 weeks old that did not undergo surgery were singly housed with 2 or 3 15-week-old female recipients of S4 cells (n = 9) at 4 weeks after implantation or sham females (n = 8). The gestational period was determined by checking the female mice for vaginal plugs; the day after a vaginal plug was detected was considered gestational day 1. The pregnant mice used for the experiments were matched for gestational age (gestational days 12 through 15). Male and female recipients of S4 cells that were not bred served as controls (n = 6 and 21, respectively). At 14 weeks after implantation, each of the male breeders (n = 3) at 22 to 23 weeks old were singly housed with 3 25-week-old female recipients of S4 cells (n = 9) from the control group. However, breeding was unsuccessful and data from this group of females were excluded from all further analyses.
Metabolic assessments
All metabolic analyses were performed in conscious, restrained mice. Blood samples were collected via the saphenous vein using heparinized microhematocrit tubes at the indicated time points. Body weight and blood glucose levels were assessed throughout the study following a 4-hour morning fast at the indicated time points. For all other metabolic tests, time 0 indicates the blood sample collected after fasting and before administration of glucose/arginine/insulin.
Mouse C-peptide levels were measured following a 4-hour morning fast prepregnancy (4 weeks after implantation), midpregnancy (6 weeks after implantation), and postpregnancy (8 weeks after implantation). Glucose-stimulated human C-peptide secretion from engrafted cells was assessed following an overnight (16 hour) fast and an intraperitoneal (IP) injection of glucose at either 2 g glucose/kg body weight (2 g/kg; 30% solution) or 4 g/kg (50% solution; Vétoquinol, Lavaltrie, QC, Canada). Oral glucose tolerance tests (OGTTs) were performed following a 6-hour morning fast and administration of glucose (2 g/kg) by oral gavage. Arginine tolerance tests were performed following a 4-hour morning fast and an IP injection of arginine (2 g/kg, 40% solution; Sigma-Aldrich, Saint Louis, MO). Insulin tolerance tests were performed following a 4-hour morning fast with an IP injection of human synthetic insulin (0.7 IU/kg body weight; Novolin ge Toronto, Novo Nordisk, Mississauga, ON, Canada).
A handheld glucometer was used to measure blood glucose levels (Lifescan; Burnaby, Canada). Insulin secretion from the endogenous pancreas was determined by measuring plasma mouse C-peptide levels [Mouse C-peptide enzyme-linked immunosorbent assay (ELISA); Alpco Diagnostics, Salem, NH]. Hormone secretion from engrafted hESC-derived cells was assessed by measuring plasma human C-peptide [C-peptide ELISA (80-CPTHU-E01.1; Alpco Diagnostics) for the first cohort and Ultrasensitive C-peptide ELISA (10-1141-01; Mercodia, Uppsala, Sweden) for the second cohort] and human insulin, glucagon, and GLP-1 levels (K15160C-2; Meso Scale Discovery, Gaithersburg, MD). Data generated from the Mercodia Ultrasensitive C-peptide ELISA were interpolated using an asymmetric sigmoidal, 5PL, curve. Any values that had absorbance levels above the highest standard of the assay are depicted as triangles instead of circles. Plasma samples were diluted 1:4 from 2 to 16 weeks after implantation; 1:25 from 20 to 24 weeks after implantation; and 1:50 at 35 weeks after implantation. A few males that received S7 cells had absorbance levels that were too high to be interpolated using the abovementioned curve at 4 (n = 2), 12 (n = 1), and 16 weeks (n = 2) after implantation.
Immunofluorescent staining and image quantification
Before implantation, a portion of S4 and S7 cells were collected for immunostaining. Kidney grafts and pancreas tissue were collected from all mice. Immunofluorescent staining was performed as previously described (18) with primary antibodies provided in Supplemental Table 2.
In cohort 1, kidney grafts and pancreas tissue were harvested from pregnant and nonpregnant recipients of S4 cells (n = 3 per group) during gestational days 12 through 15 at 6 weeks after implantation. Kidney grafts and pancreas tissue were collected from all mice at 31 weeks after implantation. In cohort 2, kidney grafts and pancreas tissue were collected from all mice at 38 weeks after implantation. All cells and tissues were fixed in 4% paraformaldehyde overnight. Tissues were stored in 70% ethanol and cells were embedded in 1% agarose before paraffin embedding. Wax-It Histology Services (Vancouver, BC, Canada) prepared all paraffin sections at a thickness of 5 μm.
To measure the number of proliferating β cells in the endogenous pancreata and the hESC-derived grafts, pancreas and kidney graft sections from pregnant and nonpregnant female recipients of S4 cells (n = 2 to 3 per group) were immunostained for insulin and proliferating cell nuclear antigen (PCNA). Whole slide immunofluorescence scanning was performed using an ImageXpressMicroTM Imaging System (Molecular Devices Corporation, Sunnyvale, CA), and images were stitched together and analyzed using MetaXpress Software. The total number of insulin-positive cells and the number of insulin-positive cells with nuclear PCNA staining were counted in every islet in each pancreas section or the entire hESC-derived graft. The average percentage of insulin-positive cells that coexpressed PCNA was then calculated per islet for each pancreas section or per graft for the implanted cells.
Statistical analysis
All statistics were performed using GraphPad Prism software (GraphPad Software Inc., La Jolla, CA). Two-way analysis of variance (ANOVA) was performed with a Tukey post hoc test to compare different treatment groups at different time points. One-way ANOVAs were performed with a Tukey post hoc test to compare different groups. Two-tailed unpaired t tests were used to compare two different groups and paired t tests were used when comparing samples preadministration and postadministration of glucose or arginine. Area under the curve (AUC) was calculated with y = 0 as the baseline. Area above the curve was calculated using the fasting blood glucose level for each mouse as the baseline. For all analyses, P < 0.05 was considered statistically significant. Data are either presented as mean ± standard error of the mean (SEM; line or bar graphs) or box-and-whisker plots showing individual data points or before-and-after line graphs.
Results
Characteristics of hESC-derived S4 and S7 cells before implantation
Following differentiation in vitro, hESC-derived S4 and S7 cells were assessed by fluorescence-activated cell sorting (Supplemental Figs. 1 and 2). The endocrine marker chromogranin was expressed in ∼27% of S4 cells (Supplemental Fig. 1) compared with ∼98% of S7 cells (Supplemental Fig. 2). Chromogranin and NKX6.1 were coexpressed in ∼6% of S4 cells vs ∼75% of S7 cells; chromogranin and NKX2.2 were coexpressed in ∼26% of S4 cells vs ∼93% of S7 cells (Supplemental Figs. 1 and 2, respectively). Although only ∼5% of S4 cells were insulin-positive, ∼59% of S7 cells were insulin-positive and ∼3% of S7 cells coexpressed insulin and glucagon (Supplemental Figs. 1 and 2). Before implantation, S4 and S7 cells were also assessed by immunofluorescent staining (Supplemental Fig. 3). Consistent with the fluorescence-activated cell sorting data, S7 cells were mostly immunoreactive for the endocrine marker synaptophysin, whereas the S4 cells had a greater portion of cells that were immunoreactive for cytokeratin 19 (a pancreatic epithelium and ductal marker; Supplemental Fig. 3). The S7 population contained more insulin-positive cells than the S4 population and S7 insulin-positive cells expressed more nuclear PDX1 and NKX6.1 (Supplemental Fig. 3). Additionally, although nuclear V-maf muscoloaponeurotic fibrosarcoma oncogene homolog A (MAFA) was evident in most insulin-positive S7 cells, MAFA was not detectable in S4 cells (Supplemental Fig. 3).
Implantation of hESC-derived S4 progenitor cells did not alter pregnancy outcomes or characteristics of offspring
To assess the impact of pregnancy on S4 graft development, a subset of female S4 cell recipients and female sham control mice were housed with male breeders at 4 weeks after implantation. Body weight, blood glucose, and mouse C-peptide levels were measured at 4, 6, and 8 weeks after implantation following a 4-hour fast (corresponding to pregestation, midgestation, and postgestation, respectively, in pregnant mice; Fig. 1a–1c). We collected blood and harvested grafts and pancreata from a subset of pregnant and nonpregnant mice (n = 3 per group) between gestational days 12 and 15 because this is when peak β-cell proliferation is observed in the mouse pancreas (14). As expected, pregnant mice had increased body weight midgestation compared with nonpregnant mice, irrespective of whether they received a cell implant (Fig. 1a; Supplemental Fig. 4a). Average blood glucose levels were similar between groups before, during, and after pregnancy (4, 6, and 8 weeks after implantation, respectively; Fig. 1b; Supplemental Fig. 4b). There were also no differences in litter size (Supplemental Fig. 4c) or average pup body weight at birth (Supplemental Fig. 4d) and at 3 weeks of age (Supplemental Fig. 4e) from mothers that were sham controls or recipients of S4 cell implants.
Pregnancy increases mouse β-cell proliferation but does not affect proliferation of hESC-derived, insulin-positive cells at 6 weeks after implantation. (a) Body weight, (b) blood glucose, and (c) mouse C-peptide levels were measured after a 4-hour morning fast at prepregnancy (4 weeks after implantation), midpregnancy (6 weeks after implantation), and postpregnancy (8 weeks after implantation) in nonpregnant female recipients of S4 cells (maroon; n = 8 at 4 and 6 weeks after implantation, n = 5 at 8 weeks), pregnant female recipients of S4 cells (purple; n = 9 at 4 and 6 weeks after implantation, n = 6 at 8 weeks after implantation), and pregnant females that underwent sham surgery (orange; n = 8 at 4, 6, and 8 weeks after implantation). (d) The average percentage of PCNA-positive, insulin-positive cells per islet in the pancreas or (e) per hESC-derived graft of nonpregnant (maroon; n = 3 pancreas; n = 3 graft) vs midpregnant (purple; n = 3 pancreas; n = 2 graft) female S4 recipients at 6 weeks after implantation. (f) Immunofluorescent staining of an islet and hESC-derived cells from nonpregnant and pregnant female S4 recipients at 6 weeks after implantation. Pregnant mouse islets had significantly increased nuclear PCNA (proliferative marker, green) staining in insulin-positive (red) cells compared with nonpregnant mouse islets. There were no differences between hESC-derived grafts in pregnant and nonpregnant mice. DAPI (nuclear marker, gray); white arrows indicate examples of PCNA and insulin-positive cells; scale bars = 100 μm. Data are presented as box-and-whisker plots or bar graphs as mean ± SEM with individual data points. For all box-and-whisker plots: *P < 0.05, two-way ANOVA. For all bar graphs: *P < 0.05, two-tailed unpaired t test (pregnant vs nonpregnant). (e) Statistics were not performed because the pregnant group had an n = 2. DAPI, 4′,6-diamidino-2-phenylindole.
Pregnancy increased proliferation of insulin-positive cells in the endogenous mouse pancreas but did not affect proliferation in hESC-derived grafts at 6 weeks after implantation
Mouse C-peptide levels were significantly higher at midgestation in pregnant mice compared with nonpregnant mice (Fig. 1c), suggesting that this time point likely corresponded to the reported window of endogenous β-cell proliferation during pregnancy (14). Therefore, at this stage (6 weeks after S4 cell implantation; between gestational days 12 and 15), both the endogenous pancreas and kidney bearing the S4 graft were harvested from a subset of pregnant and nonpregnant female mice (n = 3 per group; unfortunately, one graft from a pregnant female was damaged either during harvesting or during tissue processing and could not be analyzed). As predicted, pregnant females had significantly higher endogenous β-cell proliferation compared with nonpregnant females (an average of 8.5% PCNA-positive β cells per islet compared with 1.7% in nonpregnant S4 recipients; Fig. 1d and 1f). However, we observed no difference in the average percentage of insulin-positive cells that coexpressed PCNA in the analyzed hESC-derived grafts (Fig. 1e and 1f) from nonpregnant and pregnant mice (26.0% and 27.8%, respectively), although the limited sample size precludes us from performing statistical analysis.
Pregnancy did not affect the differentiation of hESC-derived S4 cells into glucose-responsive insulin-secreting cells, but grafts appeared to mature faster in females than males
During all glucose challenges from 13 to 28 weeks after implantation, females with S4 cell implants (“S4 females” or “S4 females, pregnancy”) displayed glucose-responsive human C-peptide secretion, irrespective of prior pregnancy (Fig. 2a–2c). At 24 weeks, all female recipients of S4 cells had significantly lower blood glucose levels at 30 and 60 minutes postglucose administration compared with female sham control mice, and pregnancy did not affect the glucose-lowering effect of implanted cells (Fig. 2d). Furthermore, hESC-derived grafts were examined in female S4 cell recipients at 31 weeks after implantation and no differences were observed in the populations of pancreatic endocrine cells that were exposed to pregnancy hormones at 4 weeks after implantation vs those from nulliparous females (Fig. 2e).
Pregnancy has no effect on the maturation of hESC into glucose-responsive insulin-secreting cells, but grafts appear to mature faster in females than in males. Blood glucose and human C-peptide levels after an overnight fast and 60 minutes postglucose (2 g/kg) administration via IP injection at (a) 13, (b) 21, and (c) 28 weeks after implantation: male recipients of S4 cells (green; n = 4 to 5 at 13 weeks; n = 3 to 4 at 21 weeks; n = 3 at 28 weeks), female recipients of S4 cells (maroon; n = 16 to 17 at 13 weeks; n = 5 at 21 weeks; n = 4 to 5 at 28 weeks), pregnant female recipients of S4 cells (purple; n = 6 at 13 and 21 weeks; n = 5 to 6 at 28 weeks), and pregnant females that underwent sham surgery (orange; n = 7 to 8 at 13 weeks; n = 8 at 21 weeks; n = 6 to 8 at 28 weeks). (d) Blood glucose levels during an OGTT after a 6-hour fast and 15, 30, 60, 90, and 120 minutes after glucose (2 g/kg) administration at 24 weeks after implantation: S4 males (green; n = 3), S4 females (maroon; n = 5), S4 females that underwent pregnancy (purple; n = 6), and sham females that underwent pregnancy (orange; n = 7). (e) Immunofluorescent staining of insulin (red), glucagon (green), somatostatin (blue), and DAPI nuclear stain (gray) in the hESC-derived grafts of female S4 cell recipients at 31 weeks after implantation. Scale bars = 100 μm. Data are presented as box-and-whisker plots with individual data points or line graphs as mean ± SEM. (a–c) Box-and-whisker plots: *P < 0.05, two-tailed paired t test (0 vs 60 minutes). For human C-peptide graphs: *P < 0.05, unpaired t test (S4 males vs S4 females and S4 females, pregnancy). (d) Box-and-whisker plot: *P < 0.05, one-way ANOVA. AUC was calculated with y = 0 as the baseline. (d) Mean ± SEM line graph: *P < 0.05, two-way ANOVA. DAPI, 4’,6-diamidino-2-phenylindole.
Interestingly, at 13 weeks after implantation, when female recipients of S4 cells demonstrated significant glucose responsiveness (irrespective of prior pregnancy), engrafted cells in male mice were not yet glucose responsive (Fig. 2a). At 21 weeks after implantation, male recipients of S4 cells (“S4 males”) displayed a trend toward glucose-responsive human C-peptide secretion, although they secreted significantly less C-peptide following a glucose stimulus compared with female S4 recipients, irrespective of prior pregnancy (Fig. 2b). Male S4 cell recipients continued to display a trend toward glucose responsiveness at 28 weeks after implantation, but secreted similar levels of human C-peptide as cells in females (Fig. 2c). These data (although in a small cohort) suggest that pancreatic progenitor cells may mature into glucose-responsive insulin-secreting cells faster in female mice than males.
S4 and S7 cells matured faster in female recipients compared with males
Although pregnancy did not appear to affect the maturation of pancreatic progenitor cells (Figs. 1and 2), the sex of the host recipient did appear to influence the maturation of S4 cells into glucose-responsive insulin-secreting cells (Fig. 2a and 2b). Therefore, we conducted a follow-up study to verify if S4 pancreatic progenitor cells matured differently in male and female implant recipients and to determine whether the sex of the host recipient would affect the maturation of more mature S7 cells. Because S7 cells are already committed to the β-cell fate and are almost fully mature at the time of implantation, we predicted that they would be less susceptible to in vivo signals compared with S4 cells that still require additional cues to reach the β-cell fate (as opposed to an α-cell fate, for example) and subsequently achieve glucose-responsive insulin secretion.
As in the previous study (Supplemental Fig. 4a), male SCID-beige mice generally weighed more than females (Fig. 3a and 3b), although notably, females had substantially more adipose tissue around the site of implantation in the kidney capsule than males (Supplemental Fig. 5). Females implanted with S4 cells had significantly lower fasting blood glucose levels than male recipients of S4 cells beginning around 7 weeks after implantation (Fig. 3a and 3c and Fig. 4a), whereas S7 cells lowered blood glucose levels to a similar degree in both male and female recipients (Fig. 3b and 3cand Fig. 4b). Female S4 cell recipients also had significantly lower postprandial blood glucose levels compared with males starting around 2 weeks after implantation (Fig. 3c and Fig. 4a) whereas there was no difference in postprandial blood glucose levels between male and female recipients of S7 cells except at 4 weeks after implantation (Fig. 3c and Fig. 4b).
Females with S4 cells have decreased blood glucose following a 4- to 6-hour fast and during an OGTT compared with males with S4 cells. Body weight and blood glucose levels were measured after a 4-hour fast in (a) male (gray; n = 11) and female (red; n = 10) recipients of S4 cells as well as (b) male (blue; n = 7) and female (green; n = 6) recipients of S7 cells throughout the duration of the study. Dashed line indicates when hESC-derived cells were implanted. (c) Blood glucose levels during an OGTT after a 6-hour fast and 15, 30, 60, 90, and 120 minutes postglucose (2 g/kg) administration at 24 weeks after implantation: S4 males (gray; n = 11), S4 females (red; n = 10), S7 males (blue; n = 7), and S7 females (green; n = 6). (c) AUC was calculated with y = 0 as the baseline and is shown in the box-and whisker-plot. Data are presented as box-and-whisker plots with individual data points or line graphs as mean ± SEM. For all box-and-whisker plots: *P < 0.05, one-way ANOVA. For all mean ± SEM line graphs: *P < 0.05, two-way ANOVA.
S4 and S7 cells develop glucose responsiveness in females before males, but cells in males secrete more human C-peptide than in females at 35 weeks after implantation following a glucose challenge. Blood glucose and human C-peptide levels were measured after an overnight fast (solid boxes) and 60 minutes (dashed boxes) after glucose (2 g/kg) administration via IP injection from 2 to 24 weeks after implantation in (a) male (gray; n = 10 to 11 for blood glucose; n = 8 to 11 for human C-peptide) and female (red; n = 9 to 10 for blood glucose; n = 5 to 10 for human C-peptide) recipients of S4 cells as well as (b) male (blue; n = 6 to 7 for blood glucose; n = 3 to 7 for human C-peptide) and female (green; n = 6 for blood glucose; n = 6 to 7 for human C-peptide) recipients of S7 cells. Points that were above the highest standard in the assay are represented as triangles. (a, b) Insets of human C-peptide data from 12 and 4 weeks after implantation are included to highlight the difference in glucose responsiveness. (c) Blood glucose and human C-peptide levels during a glucose challenge after an overnight fast and 15, 30, 60, 90, and 120 minutes after glucose (4 g/kg) administration via IP injection at 35 weeks after implantation: S4 males (gray; n = 11), S4 females (red; n = 9), S7 males (blue; n = 6), and S7 females (green; n = 6). (c) Dashed line indicates maximum detection limit of glucometer; AUC was calculated with y = 0 as the baseline. Data are presented as bar or box-and-whisker plots with individual data points or line graphs as mean ± SEM or before and after after line graphs. (a, b) Blood glucose bar plots: #P < 0.05, two-tailed unpaired t test (males vs females 0 minutes); *P < 0.05, two-tailed unpaired t test (males vs females 60 minutes). (a, b) Human C-peptide graphs: *P < 0.05; **P < 0.0001, two-tailed paired t test (0 vs 60 minutes) and unpaired t test (males vs females 60 minutes). (c) Box-and- whisker plots: *P < 0.05, two-tailed unpaired t test (male vs female per cell stage). For all mean ± SEM line graphs: *P < 0.05, two-way ANOVA.
Average plasma human C-peptide levels were above ∼0.5 ng/mL as early as 2 weeks after implantation in S7 cell recipients, whereas S4 cell recipients had levels above ∼0.3 ng/mL at 12 weeks after implantation (Fig. 4a and 4b). Plasma levels of fasted human C-peptide reached an average of ∼1 ng/mL by 4 weeks after implantation of S7 cells whereas it took around 16 weeks after implantation of S4 cells to reach similar levels (Fig. 4a and 4b). At 24 weeks after implantation, all S4 cell recipients and female S7 cell recipients had an average of ∼3 ng/mL of fasted plasma human C-peptide vs ∼8 ng/mL in male S7 cell recipients (Fig. 4a and 4b).
S4 cells produced statistically significant glucose-induced human C-peptide secretion at 12 weeks after implantation in female mice, whereas cells in male recipients did not display glucose responsiveness until the 16-week time point (Fig. 4a). Moreover, at 12 weeks after implantation (see inset in Fig. 4a), S4 cells secreted significantly more human C-peptide following a glucose stimulus in female recipients than in males. S7 cells in females were consistently glucose responsive at 4 weeks after implantation (see inset in Fig. 4b), whereas S7 cells in males did not display significant glucose-responsive human C-peptide secretion until the 8-week time point (Fig. 4b). We also noted that S4 cells were consistently glucose-responsive from 16 to 24 weeks after implantation in males and 12 to 20 weeks after implantation in females (Fig. 4a), whereas S7 cells were not consistently responsive to the 2 g/kg glucose challenge (Fig. 4b). We hypothesized that the glucose stimulus was insufficient to stimulate insulin production from the S7 cells because mice that received S7 cells had reduced postprandial blood glucose levels over time (Fig. 4b). At 2 weeks after implantation, postprandial blood glucose levels were more than 3× the fasted levels in all S4 and S7 cell recipients (Fig. 4a and 4b). However, postprandial blood glucose levels progressively declined during the study, particularly in recipients of S7 cells, to levels <5 mM (Fig. 4a and 4b). Since blood glucose levels hardly increased following a 2-g/kg glucose stimulus (Fig. 4b), the S7 cells were likely not stimulated to secrete insulin. Thus, at 35 weeks after implantation, all mice were subjected to a higher glucose dose (4 g/kg instead of 2 g/kg) to achieve greater postprandial blood glucose levels and consequently better challenge the grafts (Fig. 4c). As predicted, the mice produced robust human C-peptide secretion in response to the higher glucose dose in both males and females, including the recipients of S7 cells (Fig. 4c).
Interestingly, although at early time points after implantation, both S4 and S7 cells appeared to function better in female recipients than males [12 and 4 weeks after implantation (Fig. 4a and 4b), respectively], we observed the opposite at later stages [around 22 weeks after implantation for S4 cells and 10 weeks after implantation for S7 cells (Fig. 5a and 5b, respectively)]. At later stages, male recipients of either S4 or S7 cells had significantly higher plasma human C-peptide levels than females at 60 minutes postglucose administration and overall during the glucose challenge at 35 weeks after implantation (as indicated by AUC of the human C-peptide data; Fig. 4c). Additionally, S4 cells in male mice secreted significantly more human insulin than S4 cells in females both before and after an arginine stimulus, and only male mice exhibited arginine-induced human insulin secretion at 22 weeks after implantation (Fig. 5a). S7 cells in both male and female mice exhibited statistically significant arginine-stimulated human insulin secretion at 10 and 22 weeks after implantation (Fig. 5b). The S7 cells in males also secreted more human insulin than in females although this was only statistically significant following arginine administration (Fig. 5b). Although there were no significant differences in arginine-stimulated plasma GLP-1 levels between male and female recipients of S7 cells, females implanted with S4 cells had significantly higher arginine-stimulated GLP-1 levels than males with S4 cells at 22 weeks (Fig. 5a and 5b). There were no significant differences in plasma glucagon levels between males and females that received either S4 or S7 cells (Fig. 5a and 5b).
Males with S4 and S7 cells have higher arginine-stimulated plasma human insulin compared with females at 22 weeks after implantation. Blood glucose, human insulin, glucagon, and GLP-1 levels were measured after a 4-hour fast (solid bars) and 15 minutes postarginine (2 g/kg) administration (dashed bars) at 10 and 22 weeks after implantation in (a) male (gray; n = 9 to 11) and female (red; n = 8 to 10) recipients of S4 cells as well as (b) male (blue; n = 6 to 7) and female (green; n = 5 to 6) recipients of S7 cells. Data are presented as bar graphs with mean ± SEM plus individual data points. For all bar graphs: *P < 0.05, two-tailed paired t test (0 vs 15 minutes per group) and unpaired t test (males vs females). Dashed line indicates the limit of detection. Statistics were not performed on data that had values below the detection limit.
It is possible that differences in insulin sensitivity contributed to differences in blood glucose and human insulin/C-peptide levels between males and females. Indeed, at 25 weeks after implantation, females implanted with either S4 or S7 cells were slightly more insulin sensitive than their respective S4 or S7 males (significantly lower mean blood glucose levels in females between 60 and 120 minutes postinsulin administration; Supplemental Fig. 6a). At 38 weeks, there were no differences in insulin sensitivity between males and females (Supplemental Fig. 6b).
hESC-derived grafts were similar between males and females at 38 weeks after implantation
Synaptophysin-positive endocrine cells were similarly prevalent in grafts from male and female recipients of S4 and S7 cells and were more abundant than cytokeratin 19-positive ductal cells at 38 weeks after implantation (Fig. 6a). There were no exocrine (trypsin-positive) cells evident in any of the grafts (data not shown). Insulin-positive cells were the most predominant endocrine cell type in all grafts, followed by glucagon-positive cells and then somatostatin-positive cells (Fig. 6b). Additionally, all insulin-positive cells had nuclear NKX6.1 immunostaining regardless of sex in both S4 and S7 cell grafts (Fig. 6c). In contrast, some insulin-positive cells were not positive for nuclear MAFA and the intensity of MAFA immunostaining was more heterogeneous than the NKX6.1 immunostaining regardless of sex in both S4 and S7 grafts (Fig. 6d).
hESC-derived grafts were similar between males and females at 38 weeks after implantation. Immunostaining of whole kidney grafts from male and female recipients of S4 and S7 cells for: (a) synaptophysin (endocrine marker, red), cytokeratin19 (CK19, ductal marker, green), and DAPI (nuclear marker, gray); (b) insulin (red), glucagon (green), and somatostatin (blue); (c) insulin (red) and NKX6.1 (green); (d) insulin (red) and MAFA (green). Scale bars = 500 μm for low-magnification images and 100 μm for high-magnification images. White boxes in low-magnification images indicate where high-magnification images were taken. White dotted lines indicate the border of the kidney adjacent to the S4 or S7 graft. DAPI, 4’,6-diamidino-2-phenylindole.
Discussion
Although our studies ended up investigating intriguing sex differences in hESC-derived graft development in vivo, we had originally set out to examine the impact of pregnancy on the maturation of S4 progenitor cells in vivo because pregnancy hormones such as prolactin, placental lactogen, and serotonin promote β cell expansion (14). As expected, we observed higher mouse C-peptide levels in pregnant mice midgestation, along with ∼5× more proliferating β cells per islet in the endogenous pancreas compared with nonpregnant mice. Interestingly, pregnancy did not induce proliferation of insulin-positive cells in hESC-derived grafts at 6 weeks after implantation, although the reason for this remains unclear. It is possible that the mechanism underlying β-cell expansion during pregnancy may differ between mice and humans. For example, Butler and colleagues found no differences in β-cell proliferation in human pancreas samples from pregnant women (despite having a greater β-cell mass) compared with age-matched, nonpregnant women (19). The authors hypothesized that the increase in human β-cell mass during pregnancy was due to β-cell neogenesis, rather than the replication of existing β cells, which is thought to be the main mechanism in rodents (19). In support of this hypothesis, prolactin receptor expression is significantly reduced in human β cells compared with mouse β cells (17), and recent immunohistochemistry experiments indicate that prolactin receptor is mostly absent in adult human β cells but present in α and PP cells (20); thus, prolactin may not be promoting β-cell mass expansion in humans in a similar manner as in rodents. It is also possible that compared with the ∼36-week human gestational period, the mouse gestational period is too short (∼3 weeks) to have an impact on human cells. Alternatively, the immature human insulin-positive cells may have already been at their peak proliferation rate, irrespective of pregnancy, because proliferation was relatively high in graft cells and is generally high during human fetal pancreas development (21). Additionally, progenitor cells at the particular stage of development used in our experimental model may be resistant to pregnancy hormones. For example, human fetal islet cells have weak prolactin receptor immunoreactivity (22). Therefore, it is possible that pregnancy hormone receptors are not present in β cells early in development. Finally, human cells may generally be insensitive to mouse pregnancy hormones, but may be affected by the analogous human hormones (regardless of developmental stage). For example, mouse prolactin did not mimic human prolactin-induced cell proliferation in murine 32D cells transfected with human prolactin receptors (23).
After pregnancy, female mice implanted with S4 cells had similar plasma human C-peptide levels and glucose-dependent insulin secretion compared with nulliparous females throughout the study, indicating that there was no major impact of pregnancy at 4 weeks after implantation on the development of hESC-derived S4 cells in vivo. Interestingly, during this study, we noted that S4 cells in female recipients possessed robust glucose responsiveness at just 13 weeks after implantation, an age at which S4 cells in male recipients in this study and prior studies by our group (3, 5) had not yet developed regulated human insulin secretion. However, this study was designed to examine the effects of pregnancy, so the small group of male mice was insufficient to draw conclusions about possible sex differences. We subsequently followed up with a study containing a large cohort of male and female S4 cell recipients. We also examined the effect of host sex on S7 cells, which are at a more advanced stage of β-cell development at the time of implantation.
As in the pregnancy cohort, robust glucose-responsive human C-peptide secretion was observed at 12 weeks after implantation of S4 cells in female mice of our second study, whereas male mice again did not yet have regulated human C-peptide secretion. S4 cells in males were glucose responsive at 16 weeks after implantation, but displayed a blunted response compared with females (glucose-stimulated human C-peptide levels were ∼2.3× fasted levels in females compared with ∼1.7× in males). Interestingly, S7 cells also became glucose responsive in females before males, but at 4 weeks after implantation rather than at 12 weeks as in female S4 cell recipients. This is most likely from the advanced differentiation status of S7 cells compared with S4 cells at the time of implantation. Fasting blood glucose levels generally decreased over time in all recipients as hESC-derived cells matured and produced more insulin. However, blood glucose levels decreased faster in both female S4 and S7 recipients compared with male S4 and S7 recipients, respectively, which is likely a reflection of the better functioning hESC-derived grafts in females at early time points after implantation. It is possible that circulating estrogen in female recipients promotes the maturation of hESC-derived pancreatic progenitor cells. Inhibition of estrogen receptor-α reduced the expression of Neurogenin 3, an endocrine progenitor cell marker, and decreased β-cell proliferation in the developing mouse pancreas (24). Moreover, activation of estrogen receptor-β can enhance glucose-stimulated insulin secretion in both mouse and human islets (25). Female mice also had increased adipose tissue surrounding the kidney capsule where the cells were implanted, and adipokine signals may promote β-cell development. For example, the adipokine adipsin has recently been shown to play a role in maintaining β-cell function; isolated islets from adipsin-deficient mice have reduced glucose-stimulated insulin secretion (26). In potential support of a role of adipose tissue in enhancing the maturation of hESC-derived pancreatic progenitors, ViaCyte has reported routinely obtaining robust glucose-responsive human C-peptide at 12 weeks after implantation in male mice in which progenitor cells are implanted into the epididymal fat pad (27, 28). Further studies will be required to understand the mechanism underlying the accelerated maturation of S4 and S7 cells in female relative to male mice.
Although maturation of hESC-derived S4 and S7 cells was slower in male mice than female mice, both S4- and S7-derived cells appeared to eventually function better in male mice than female mice. Male mice exhibited higher human insulin levels than females in response to an arginine stimulus at 22 weeks after implantation and higher human C-peptide levels during a 4 g/kg glucose challenge at 35 weeks after implantation. Therefore, specific signals in the male host may influence the human insulin secretory response of hESC-derived cells. For example, testosterone was recently revealed to enhance glucose-stimulated insulin secretion in male mice and human β cells through interactions with androgen and GLP-1 receptors (29), which could explain why male recipients display a different secretory response compared with females. These observations could also be explained by differences in C-peptide/insulin clearance and in insulin sensitivity between males and females, leading to less secretory demand on the insulin producing cells in the grafts of females compared with male recipients. In fact, female humans and mice are known to be more insulin sensitive than their male counterparts (30, 31), although the increased insulin sensitivity we observed in female mice appeared to be transient.
In summary, pregnancy (at 4 weeks after implantation) had no apparent effect on the maturation of hESC-derived S4 pancreatic progenitor cells in mice, but we found that female mice promoted the development of glucose-responsive insulin-secreting cells at earlier stages after implantation compared with males in two separate cohorts. Additionally, we found that S7 cells, despite being more mature than S4 cells, also developed glucose responsiveness faster in female vs male mice, although the effect of sex differences on blood glucose levels were minimal with S7 cells compared with the results with S4 cells. Further studies need to be conducted to elucidate the mechanisms behind the faster maturation of hESC-derived cells in females. Additional studies are also required to determine if our observations are generally applicable to other stem cell lines (aside from H1 cells), other species, and other implantation sites. Regardless, our findings indicate that the differentiation and function of hESC-derived cells can be altered by the host sex, which should be considered in both preclinical rodent studies and ongoing clinical trials.
Antibody Table
| Peptide/Protein Target | Antigen Sequence (if Known) | Name of Antibody | Manufacturer, Catalog No. | Species Raised in; Monoclonal or Polyclonal | Dilution Used | RRID |
|---|---|---|---|---|---|---|
| C-peptide | Unknown | Rabbit anti-C-peptide | Cell Signaling Technology, 4593S | Rabbit; polyclonal | 1:10 | AB_10691857 |
| CDX2 | Unknown | Mouse anti-CDX2 | BD Biosciences, 560395 | Mouse; monoclonal | 1:250 | AB_1645405 |
| Chromogranin A | Unknown | Rabbit anti-Chromogranin A | Dako, Denmark, IS502 | Rabbit; polyclonal | 1:10 | AB_2716790 |
| Cytokeratin 19 (CK-19) | Unknown | Mouse anti-CK19 | Dako, Denmark, M 0888 | Mouse; monoclonal | 1:100 | AB_2234418 |
| Glucagon | Unknown | Mouse anti-Glucagon | Sigma-Aldrich, G 2654 | Mouse; monoclonal | 1:250; 1:1000 | AB_259852 |
| Glucagon (D16G10) | Residues surrounding Ala137 of human proglucagon protein | Rabbit anti-Glucagon | Cell Signaling Technology, 8233S | Rabbit; monoclonal | 1:500 | AB_10859908 |
| IgG, k | Unknown | Purified rabbit IgG, k Isotype | BD Biosciences, 550875 | Rabbit; polyclonal | 1:1000 | AB_393942 |
| IgG, k | Unknown | Purified mouse IgG, k Isotype | BD Biosciences, 557273 | Mouse; monoclonal | 1:50 | AB_396613 |
| IgG1 | Unknown | Alexa Fluor 647 IgG1, Isotype Control | BD Biosciences, 557732 | Mouse; monoclonal | 1:40 | AB_396840 |
| IgG1 k | Unknown | PE mouse IgG1, k, isotype control | BD Biosciences, 555749 | Mouse; monoclonal | 1:40 | AB_396091 |
| Insulin | Unknown | Rabbit insulin, AF647 | Cell Signaling Technology, 9008S | Rabbit; monoclonal | 1:80 | AB_2687822 |
| Insulin | Residues surrounding Val36 of human insulin | Mouse anti-insulin | Cell Signaling Technology, 8138S | Mouse; monoclonal | 1:250 | AB_10949314 |
| Insulin | Full-length human insulin | Guinea pig anti-insulin | Thermo Scientific, PA1-26938 | Guinea pig; polyclonal | 1:100 | AB_794668 |
| Insulin (C27C9) | Unknown | Rabbit anti-insulin | Cell Signaling Technology, 3014S | Rabbit; monoclonal | 1:10; 1:200 | AB_2126503 |
| Islet 1 | Unknown | PE mouse anti-Islet1 | BD Biosciences, 562547 | Mouse; monoclonal | 1:40 | AB_11154592 |
| Ki-67 | Unknown | Alexa Fluor 647 mouse antihuman Ki67 | BD Biosciences, 561126 | Mouse; monoclonal | 1:10 | AB_10611874 |
| NeuroD | Unknown | PE mouse anti-NeuroD | BD Biosciences, 563001 | Mouse; monoclonal | 1:40 | AB_2716791 |
| NKX2.2 | Unknown | Mouse anti-NKX2.2 | Developmental Studies Hybridoma Bank University of Iowa, 74.5A5 | Mouse; monoclonal | 1:100 | AB_531794 |
| NKX6.1 | Unknown | PE mouse anti-NKX6.1 | BD Biosciences, 563023 | Mouse; monoclonal | 1:40 | AB_2716792 |
| NKX6.1 | C-terminal a.a. 299-365 | Mouse anti-NKX6.1 | Developmental Studies Hybridoma Bank University of Iowa, F55A12 | Mouse; monoclonal | 1:50 | AB_532379 |
| NKX6.1 | Unknown | Rabbit anti-NKX6.1 | Betalogics (J&J), LP9878 | Rabbit; polyclonal | 1:1000 | AB_2716793 |
| OCT3.4 | Mouse OCT 3 a.a. 252-372 | Mouse OCT3/4, AF647 | BD Biosciences, 560329 | Mouse; monoclonal | 1:20 | AB_1645318 |
| Pax6 | Human Pax6 a.a. 406-422 | PE mouse antihuman Pax6 | BD Biosciences, 561552 | Mouse; monoclonal | 1:20 | AB_10714781 |
| PDX1 | Unknown | PE mouse anti-PDX1 | BD Biosciences, 562161 | Mouse; monoclonal | 1:40 | AB_10893589 |
| PDX1 | Binds preferentially the DNA motif 5′-[CT]TAAT[TG]-3′ | Guinea Pig anti-PDX1 | Abcam, ab47308 | Guinea pig; polyclonal | 1:1000 | AB_777178 |
| PCNA | Human PCNA a.a. 68-230 | Mouse anti-PCNA | BD Biosciences, 610665 | Mouse; monoclonal | 1:100 | AB_397992 |
| Somatostatin | Human somatostatin | Mouse anti-Somatostatin | Beta Cell Biology Consortium, AB1985 | Mouse; monoclonal | 1:500 | AB_10014609 |
| Somatostatin | APSDPRLRQFLQKSLAAAAGKQELAKYFLAELLSEPNQTENDALEPEDLSQAAEQDEMRLELQRSANSNPAMAPRERKAGCKN | Rabbit anti-Somatostatin | Sigma-Aldrich, HPA019472 | Rabbit; polyclonal | 1:500 | AB_1857360 |
| Synaptophysin (SP11) | C-terminus of human Synaptophysin | Rabbit anti-synaptophysin | Novus Biologicals, NB120-16659 | Rabbit; monoclonal | 1:50 | AB_792140 |
| Trypsin 1/2/3 | Mouse myeloma cell line NS0-derived recombinant human trypsin 2 Ala16-Ser247 | Sheep anti-trypsin | R&D Systems, AF3586 | Sheep; polyclonal | 1:25 | AB_884531 |
| MAFA | Unknown | Rabbit anti-MAFA | Betalogics (J&J), LP9872 | Rabbit; polyclonal | 1:1000 | AB_2665528 |
| Peptide/Protein Target | Antigen Sequence (if Known) | Name of Antibody | Manufacturer, Catalog No. | Species Raised in; Monoclonal or Polyclonal | Dilution Used | RRID |
|---|---|---|---|---|---|---|
| C-peptide | Unknown | Rabbit anti-C-peptide | Cell Signaling Technology, 4593S | Rabbit; polyclonal | 1:10 | AB_10691857 |
| CDX2 | Unknown | Mouse anti-CDX2 | BD Biosciences, 560395 | Mouse; monoclonal | 1:250 | AB_1645405 |
| Chromogranin A | Unknown | Rabbit anti-Chromogranin A | Dako, Denmark, IS502 | Rabbit; polyclonal | 1:10 | AB_2716790 |
| Cytokeratin 19 (CK-19) | Unknown | Mouse anti-CK19 | Dako, Denmark, M 0888 | Mouse; monoclonal | 1:100 | AB_2234418 |
| Glucagon | Unknown | Mouse anti-Glucagon | Sigma-Aldrich, G 2654 | Mouse; monoclonal | 1:250; 1:1000 | AB_259852 |
| Glucagon (D16G10) | Residues surrounding Ala137 of human proglucagon protein | Rabbit anti-Glucagon | Cell Signaling Technology, 8233S | Rabbit; monoclonal | 1:500 | AB_10859908 |
| IgG, k | Unknown | Purified rabbit IgG, k Isotype | BD Biosciences, 550875 | Rabbit; polyclonal | 1:1000 | AB_393942 |
| IgG, k | Unknown | Purified mouse IgG, k Isotype | BD Biosciences, 557273 | Mouse; monoclonal | 1:50 | AB_396613 |
| IgG1 | Unknown | Alexa Fluor 647 IgG1, Isotype Control | BD Biosciences, 557732 | Mouse; monoclonal | 1:40 | AB_396840 |
| IgG1 k | Unknown | PE mouse IgG1, k, isotype control | BD Biosciences, 555749 | Mouse; monoclonal | 1:40 | AB_396091 |
| Insulin | Unknown | Rabbit insulin, AF647 | Cell Signaling Technology, 9008S | Rabbit; monoclonal | 1:80 | AB_2687822 |
| Insulin | Residues surrounding Val36 of human insulin | Mouse anti-insulin | Cell Signaling Technology, 8138S | Mouse; monoclonal | 1:250 | AB_10949314 |
| Insulin | Full-length human insulin | Guinea pig anti-insulin | Thermo Scientific, PA1-26938 | Guinea pig; polyclonal | 1:100 | AB_794668 |
| Insulin (C27C9) | Unknown | Rabbit anti-insulin | Cell Signaling Technology, 3014S | Rabbit; monoclonal | 1:10; 1:200 | AB_2126503 |
| Islet 1 | Unknown | PE mouse anti-Islet1 | BD Biosciences, 562547 | Mouse; monoclonal | 1:40 | AB_11154592 |
| Ki-67 | Unknown | Alexa Fluor 647 mouse antihuman Ki67 | BD Biosciences, 561126 | Mouse; monoclonal | 1:10 | AB_10611874 |
| NeuroD | Unknown | PE mouse anti-NeuroD | BD Biosciences, 563001 | Mouse; monoclonal | 1:40 | AB_2716791 |
| NKX2.2 | Unknown | Mouse anti-NKX2.2 | Developmental Studies Hybridoma Bank University of Iowa, 74.5A5 | Mouse; monoclonal | 1:100 | AB_531794 |
| NKX6.1 | Unknown | PE mouse anti-NKX6.1 | BD Biosciences, 563023 | Mouse; monoclonal | 1:40 | AB_2716792 |
| NKX6.1 | C-terminal a.a. 299-365 | Mouse anti-NKX6.1 | Developmental Studies Hybridoma Bank University of Iowa, F55A12 | Mouse; monoclonal | 1:50 | AB_532379 |
| NKX6.1 | Unknown | Rabbit anti-NKX6.1 | Betalogics (J&J), LP9878 | Rabbit; polyclonal | 1:1000 | AB_2716793 |
| OCT3.4 | Mouse OCT 3 a.a. 252-372 | Mouse OCT3/4, AF647 | BD Biosciences, 560329 | Mouse; monoclonal | 1:20 | AB_1645318 |
| Pax6 | Human Pax6 a.a. 406-422 | PE mouse antihuman Pax6 | BD Biosciences, 561552 | Mouse; monoclonal | 1:20 | AB_10714781 |
| PDX1 | Unknown | PE mouse anti-PDX1 | BD Biosciences, 562161 | Mouse; monoclonal | 1:40 | AB_10893589 |
| PDX1 | Binds preferentially the DNA motif 5′-[CT]TAAT[TG]-3′ | Guinea Pig anti-PDX1 | Abcam, ab47308 | Guinea pig; polyclonal | 1:1000 | AB_777178 |
| PCNA | Human PCNA a.a. 68-230 | Mouse anti-PCNA | BD Biosciences, 610665 | Mouse; monoclonal | 1:100 | AB_397992 |
| Somatostatin | Human somatostatin | Mouse anti-Somatostatin | Beta Cell Biology Consortium, AB1985 | Mouse; monoclonal | 1:500 | AB_10014609 |
| Somatostatin | APSDPRLRQFLQKSLAAAAGKQELAKYFLAELLSEPNQTENDALEPEDLSQAAEQDEMRLELQRSANSNPAMAPRERKAGCKN | Rabbit anti-Somatostatin | Sigma-Aldrich, HPA019472 | Rabbit; polyclonal | 1:500 | AB_1857360 |
| Synaptophysin (SP11) | C-terminus of human Synaptophysin | Rabbit anti-synaptophysin | Novus Biologicals, NB120-16659 | Rabbit; monoclonal | 1:50 | AB_792140 |
| Trypsin 1/2/3 | Mouse myeloma cell line NS0-derived recombinant human trypsin 2 Ala16-Ser247 | Sheep anti-trypsin | R&D Systems, AF3586 | Sheep; polyclonal | 1:25 | AB_884531 |
| MAFA | Unknown | Rabbit anti-MAFA | Betalogics (J&J), LP9872 | Rabbit; polyclonal | 1:1000 | AB_2665528 |
Abbreviations: a.a., amino acid; IgG, immunoglobulin G; PE, phycoerythrin; RRID, Research Resource Identifier.
Antibody Table
| Peptide/Protein Target | Antigen Sequence (if Known) | Name of Antibody | Manufacturer, Catalog No. | Species Raised in; Monoclonal or Polyclonal | Dilution Used | RRID |
|---|---|---|---|---|---|---|
| C-peptide | Unknown | Rabbit anti-C-peptide | Cell Signaling Technology, 4593S | Rabbit; polyclonal | 1:10 | AB_10691857 |
| CDX2 | Unknown | Mouse anti-CDX2 | BD Biosciences, 560395 | Mouse; monoclonal | 1:250 | AB_1645405 |
| Chromogranin A | Unknown | Rabbit anti-Chromogranin A | Dako, Denmark, IS502 | Rabbit; polyclonal | 1:10 | AB_2716790 |
| Cytokeratin 19 (CK-19) | Unknown | Mouse anti-CK19 | Dako, Denmark, M 0888 | Mouse; monoclonal | 1:100 | AB_2234418 |
| Glucagon | Unknown | Mouse anti-Glucagon | Sigma-Aldrich, G 2654 | Mouse; monoclonal | 1:250; 1:1000 | AB_259852 |
| Glucagon (D16G10) | Residues surrounding Ala137 of human proglucagon protein | Rabbit anti-Glucagon | Cell Signaling Technology, 8233S | Rabbit; monoclonal | 1:500 | AB_10859908 |
| IgG, k | Unknown | Purified rabbit IgG, k Isotype | BD Biosciences, 550875 | Rabbit; polyclonal | 1:1000 | AB_393942 |
| IgG, k | Unknown | Purified mouse IgG, k Isotype | BD Biosciences, 557273 | Mouse; monoclonal | 1:50 | AB_396613 |
| IgG1 | Unknown | Alexa Fluor 647 IgG1, Isotype Control | BD Biosciences, 557732 | Mouse; monoclonal | 1:40 | AB_396840 |
| IgG1 k | Unknown | PE mouse IgG1, k, isotype control | BD Biosciences, 555749 | Mouse; monoclonal | 1:40 | AB_396091 |
| Insulin | Unknown | Rabbit insulin, AF647 | Cell Signaling Technology, 9008S | Rabbit; monoclonal | 1:80 | AB_2687822 |
| Insulin | Residues surrounding Val36 of human insulin | Mouse anti-insulin | Cell Signaling Technology, 8138S | Mouse; monoclonal | 1:250 | AB_10949314 |
| Insulin | Full-length human insulin | Guinea pig anti-insulin | Thermo Scientific, PA1-26938 | Guinea pig; polyclonal | 1:100 | AB_794668 |
| Insulin (C27C9) | Unknown | Rabbit anti-insulin | Cell Signaling Technology, 3014S | Rabbit; monoclonal | 1:10; 1:200 | AB_2126503 |
| Islet 1 | Unknown | PE mouse anti-Islet1 | BD Biosciences, 562547 | Mouse; monoclonal | 1:40 | AB_11154592 |
| Ki-67 | Unknown | Alexa Fluor 647 mouse antihuman Ki67 | BD Biosciences, 561126 | Mouse; monoclonal | 1:10 | AB_10611874 |
| NeuroD | Unknown | PE mouse anti-NeuroD | BD Biosciences, 563001 | Mouse; monoclonal | 1:40 | AB_2716791 |
| NKX2.2 | Unknown | Mouse anti-NKX2.2 | Developmental Studies Hybridoma Bank University of Iowa, 74.5A5 | Mouse; monoclonal | 1:100 | AB_531794 |
| NKX6.1 | Unknown | PE mouse anti-NKX6.1 | BD Biosciences, 563023 | Mouse; monoclonal | 1:40 | AB_2716792 |
| NKX6.1 | C-terminal a.a. 299-365 | Mouse anti-NKX6.1 | Developmental Studies Hybridoma Bank University of Iowa, F55A12 | Mouse; monoclonal | 1:50 | AB_532379 |
| NKX6.1 | Unknown | Rabbit anti-NKX6.1 | Betalogics (J&J), LP9878 | Rabbit; polyclonal | 1:1000 | AB_2716793 |
| OCT3.4 | Mouse OCT 3 a.a. 252-372 | Mouse OCT3/4, AF647 | BD Biosciences, 560329 | Mouse; monoclonal | 1:20 | AB_1645318 |
| Pax6 | Human Pax6 a.a. 406-422 | PE mouse antihuman Pax6 | BD Biosciences, 561552 | Mouse; monoclonal | 1:20 | AB_10714781 |
| PDX1 | Unknown | PE mouse anti-PDX1 | BD Biosciences, 562161 | Mouse; monoclonal | 1:40 | AB_10893589 |
| PDX1 | Binds preferentially the DNA motif 5′-[CT]TAAT[TG]-3′ | Guinea Pig anti-PDX1 | Abcam, ab47308 | Guinea pig; polyclonal | 1:1000 | AB_777178 |
| PCNA | Human PCNA a.a. 68-230 | Mouse anti-PCNA | BD Biosciences, 610665 | Mouse; monoclonal | 1:100 | AB_397992 |
| Somatostatin | Human somatostatin | Mouse anti-Somatostatin | Beta Cell Biology Consortium, AB1985 | Mouse; monoclonal | 1:500 | AB_10014609 |
| Somatostatin | APSDPRLRQFLQKSLAAAAGKQELAKYFLAELLSEPNQTENDALEPEDLSQAAEQDEMRLELQRSANSNPAMAPRERKAGCKN | Rabbit anti-Somatostatin | Sigma-Aldrich, HPA019472 | Rabbit; polyclonal | 1:500 | AB_1857360 |
| Synaptophysin (SP11) | C-terminus of human Synaptophysin | Rabbit anti-synaptophysin | Novus Biologicals, NB120-16659 | Rabbit; monoclonal | 1:50 | AB_792140 |
| Trypsin 1/2/3 | Mouse myeloma cell line NS0-derived recombinant human trypsin 2 Ala16-Ser247 | Sheep anti-trypsin | R&D Systems, AF3586 | Sheep; polyclonal | 1:25 | AB_884531 |
| MAFA | Unknown | Rabbit anti-MAFA | Betalogics (J&J), LP9872 | Rabbit; polyclonal | 1:1000 | AB_2665528 |
| Peptide/Protein Target | Antigen Sequence (if Known) | Name of Antibody | Manufacturer, Catalog No. | Species Raised in; Monoclonal or Polyclonal | Dilution Used | RRID |
|---|---|---|---|---|---|---|
| C-peptide | Unknown | Rabbit anti-C-peptide | Cell Signaling Technology, 4593S | Rabbit; polyclonal | 1:10 | AB_10691857 |
| CDX2 | Unknown | Mouse anti-CDX2 | BD Biosciences, 560395 | Mouse; monoclonal | 1:250 | AB_1645405 |
| Chromogranin A | Unknown | Rabbit anti-Chromogranin A | Dako, Denmark, IS502 | Rabbit; polyclonal | 1:10 | AB_2716790 |
| Cytokeratin 19 (CK-19) | Unknown | Mouse anti-CK19 | Dako, Denmark, M 0888 | Mouse; monoclonal | 1:100 | AB_2234418 |
| Glucagon | Unknown | Mouse anti-Glucagon | Sigma-Aldrich, G 2654 | Mouse; monoclonal | 1:250; 1:1000 | AB_259852 |
| Glucagon (D16G10) | Residues surrounding Ala137 of human proglucagon protein | Rabbit anti-Glucagon | Cell Signaling Technology, 8233S | Rabbit; monoclonal | 1:500 | AB_10859908 |
| IgG, k | Unknown | Purified rabbit IgG, k Isotype | BD Biosciences, 550875 | Rabbit; polyclonal | 1:1000 | AB_393942 |
| IgG, k | Unknown | Purified mouse IgG, k Isotype | BD Biosciences, 557273 | Mouse; monoclonal | 1:50 | AB_396613 |
| IgG1 | Unknown | Alexa Fluor 647 IgG1, Isotype Control | BD Biosciences, 557732 | Mouse; monoclonal | 1:40 | AB_396840 |
| IgG1 k | Unknown | PE mouse IgG1, k, isotype control | BD Biosciences, 555749 | Mouse; monoclonal | 1:40 | AB_396091 |
| Insulin | Unknown | Rabbit insulin, AF647 | Cell Signaling Technology, 9008S | Rabbit; monoclonal | 1:80 | AB_2687822 |
| Insulin | Residues surrounding Val36 of human insulin | Mouse anti-insulin | Cell Signaling Technology, 8138S | Mouse; monoclonal | 1:250 | AB_10949314 |
| Insulin | Full-length human insulin | Guinea pig anti-insulin | Thermo Scientific, PA1-26938 | Guinea pig; polyclonal | 1:100 | AB_794668 |
| Insulin (C27C9) | Unknown | Rabbit anti-insulin | Cell Signaling Technology, 3014S | Rabbit; monoclonal | 1:10; 1:200 | AB_2126503 |
| Islet 1 | Unknown | PE mouse anti-Islet1 | BD Biosciences, 562547 | Mouse; monoclonal | 1:40 | AB_11154592 |
| Ki-67 | Unknown | Alexa Fluor 647 mouse antihuman Ki67 | BD Biosciences, 561126 | Mouse; monoclonal | 1:10 | AB_10611874 |
| NeuroD | Unknown | PE mouse anti-NeuroD | BD Biosciences, 563001 | Mouse; monoclonal | 1:40 | AB_2716791 |
| NKX2.2 | Unknown | Mouse anti-NKX2.2 | Developmental Studies Hybridoma Bank University of Iowa, 74.5A5 | Mouse; monoclonal | 1:100 | AB_531794 |
| NKX6.1 | Unknown | PE mouse anti-NKX6.1 | BD Biosciences, 563023 | Mouse; monoclonal | 1:40 | AB_2716792 |
| NKX6.1 | C-terminal a.a. 299-365 | Mouse anti-NKX6.1 | Developmental Studies Hybridoma Bank University of Iowa, F55A12 | Mouse; monoclonal | 1:50 | AB_532379 |
| NKX6.1 | Unknown | Rabbit anti-NKX6.1 | Betalogics (J&J), LP9878 | Rabbit; polyclonal | 1:1000 | AB_2716793 |
| OCT3.4 | Mouse OCT 3 a.a. 252-372 | Mouse OCT3/4, AF647 | BD Biosciences, 560329 | Mouse; monoclonal | 1:20 | AB_1645318 |
| Pax6 | Human Pax6 a.a. 406-422 | PE mouse antihuman Pax6 | BD Biosciences, 561552 | Mouse; monoclonal | 1:20 | AB_10714781 |
| PDX1 | Unknown | PE mouse anti-PDX1 | BD Biosciences, 562161 | Mouse; monoclonal | 1:40 | AB_10893589 |
| PDX1 | Binds preferentially the DNA motif 5′-[CT]TAAT[TG]-3′ | Guinea Pig anti-PDX1 | Abcam, ab47308 | Guinea pig; polyclonal | 1:1000 | AB_777178 |
| PCNA | Human PCNA a.a. 68-230 | Mouse anti-PCNA | BD Biosciences, 610665 | Mouse; monoclonal | 1:100 | AB_397992 |
| Somatostatin | Human somatostatin | Mouse anti-Somatostatin | Beta Cell Biology Consortium, AB1985 | Mouse; monoclonal | 1:500 | AB_10014609 |
| Somatostatin | APSDPRLRQFLQKSLAAAAGKQELAKYFLAELLSEPNQTENDALEPEDLSQAAEQDEMRLELQRSANSNPAMAPRERKAGCKN | Rabbit anti-Somatostatin | Sigma-Aldrich, HPA019472 | Rabbit; polyclonal | 1:500 | AB_1857360 |
| Synaptophysin (SP11) | C-terminus of human Synaptophysin | Rabbit anti-synaptophysin | Novus Biologicals, NB120-16659 | Rabbit; monoclonal | 1:50 | AB_792140 |
| Trypsin 1/2/3 | Mouse myeloma cell line NS0-derived recombinant human trypsin 2 Ala16-Ser247 | Sheep anti-trypsin | R&D Systems, AF3586 | Sheep; polyclonal | 1:25 | AB_884531 |
| MAFA | Unknown | Rabbit anti-MAFA | Betalogics (J&J), LP9872 | Rabbit; polyclonal | 1:1000 | AB_2665528 |
Abbreviations: a.a., amino acid; IgG, immunoglobulin G; PE, phycoerythrin; RRID, Research Resource Identifier.
Abbreviations:
- ANOVA
analysis of variance
- AUC
area under the curve
- ELISA
enzyme-linked immunosorbent assay
- hESC
human embryonic stem cell
- IP
intraperitoneal
- MAFA
V-maf muscoloaponeurotic fibrosarcoma oncogene homolog A
- OGTT
oral glucose tolerance test
- PCNA
proliferating cell nuclear antigen
- S4
stage 4
- S7
stage 7
- SEM
standard error of the mean
- T1D
type 1 diabetes
- UBC
University of British Columbia
Acknowledgments
The authors thank Dr. Nina Quiskamp, Dr. Cara Ellis, and Ali Asadi for technical assistance and Dr. Kevin D’Amour for contributing to manuscript revisions.
Financial Support: This work was funded by the Canadian Institutes of Health Research, Juvenile Diabetes Research Foundation (JDRF), the Stem Cell Network, Chuck and Peter Allard, and STEMCELL Technologies. N.S. received a graduate scholarship from the National Sciences and Engineering Research Council of Canada. J.E.B. was supported by a JDRF postdoctoral fellowship, a Canadian Diabetes Association postdoctoral fellowship, and a L’Oréal Canada for Women in Science Research Excellence Fellowship. H.S. received a Deutscher Akademischer Austauschdienst (DAAD) Research Internships in Science and Engineering (RISE) worldwide international research internship.
Clinical Trial Information: ClinicalTrials.gov no. NCT02239354 (registered 12 September 2014).
Author Contributions: N.S. wrote the manuscript. N.S., J.E.B., A.R., and T.J.K. contributed to the conception and design of experiments. N.S., J.E.B., S.O., H.S., A.R., and T.J.K. were involved in the acquisition, analysis, and interpretation of data; contributed to manuscript revisions; and approved the final version of the manuscript.
Current Affiliations: J.E. Bruin’s current affiliation is the Department of Biology and Institute of Biochemistry, Carleton University, Ottawa, Ontario K1S 5B6, Canada. A. Rezania’s current affiliation is ViaCyte, Inc., San Diego, California 92121.
Disclosure Summary: A.R. is an employee and shareholder of Janssen R&D, LLC. T.J.K. received financial support from Janssen R&D, LLC, for the research described in this article. The remaining authors have nothing to disclose.
References
