β-Cell failure underlies much of the pathogenesis of type 1 and type 2 diabetes, and restoring β-cell mass, for example by stimulating proliferation, has been an active area of investigation. β-Cells are generally, although not universally, quiescent. Numerous studies have documented the low but nonzero replicative capacity of both rodent and human β-cells (1,2). Further, in adult rodent models, it appears that β-cell proliferation can occur in response to injury (3), nutrients (4), or pregnancy (5,6). In humans, the matter may be more complicated, with many pathways under consideration (7). Therefore, understanding what regulates β-cell proliferation could have important implications for regenerative medicine, as the mechanisms underlying this process may lead to new avenues for therapeutic intervention.

Gene-expression profiling of β-cells has been challenging because islets comprise a heterogeneous cell population. Early efforts profiled intact islets without regard to cell type; for example, Affymetrix profiling of islets from pregnant mice (8,9) pointed to pathways potentially responsible for the proliferative response during pregnancy. For the most part, such a strategy is reasonable in rodent islets as they are typically composed of >80% β-cells. In human islets, however, β-cells barely make up 50% of the population. Transcriptomic analysis of individual cell types in the human islet, based on sorting cells by surface marker antibodies (10), helped define each cell type genetically. More recent advances in single-cell profiling technologies have enabled the examination of individual cells from human islets (11); although current methods limit profiling to relatively few cells, those numbers should increase over time with advances in microfluidics. In general, surprisingly few studies have used next-generation sequencing to analyze the genetic program of β-cells, and none thus far have looked at a comparison of phenotypic subtypes of β-cells.

In this issue of Diabetes, Klochendler et al. (12) asked the question of whether there were clear gene-expression differences between replicating and quiescent β-cells in mice. They leveraged a transgenic mouse model they had created previously to study hepatocyte replication, in which green fluorescent protein (GFP) was fused with the N-terminal destruction box of cyclin B1 (13). GFP is constitutively expressed but rapidly degraded by the proteasome during quiescence, as an E3 ubiquitin ligase targets cyclin B1 for destruction. However, as cells enter S phase and proceed to mitosis, cyclin B1 is no longer ubiquitinated and remains stable, causing an increase in GFP expression in replicating cells. An advantage of this approach is that it can be used to analyze live cells. Because of cellular contamination from acinar tissue, the authors took two independent two-step sorting strategies to enrich the population for GFP-positive β-cells (Fig. 1). These efforts resulted in the first examination of the specific gene-expression landscape of dividing β-cells isolated in vivo.

Figure 1

Schematic illustrating the methods used to isolate proliferating β-cells. Pancreatic islets were isolated from transgenic mice expressing a fusion construct of cyclin B1 destruction box and GFP. β-Cells were isolated either by staining with the zinc dye N-(6-Methoxy-8-quinolyl)-p-toluenesulfonamide (TSQ) (top) or by expression of Tomato in doubly transgenic mice expressing an insulin-driven Tomato gene (bottom). Proliferating cells were then isolated based on their expression of GFP; this two-step process eliminated contamination from proliferating acinar cells. RNA-seq was then performed on two populations of β-cells, proliferating and quiescent, from these two models, and the results form the basis of the work of Klochendler et al. (12).

Figure 1

Schematic illustrating the methods used to isolate proliferating β-cells. Pancreatic islets were isolated from transgenic mice expressing a fusion construct of cyclin B1 destruction box and GFP. β-Cells were isolated either by staining with the zinc dye N-(6-Methoxy-8-quinolyl)-p-toluenesulfonamide (TSQ) (top) or by expression of Tomato in doubly transgenic mice expressing an insulin-driven Tomato gene (bottom). Proliferating cells were then isolated based on their expression of GFP; this two-step process eliminated contamination from proliferating acinar cells. RNA-seq was then performed on two populations of β-cells, proliferating and quiescent, from these two models, and the results form the basis of the work of Klochendler et al. (12).

Klochendler et al. (12) found that thousands of genes were differentially expressed between the two populations of β-cells. As one might expect, genes involved in mitotic processes were at the top of this list. Gene ontology (GO) analysis found sets of genes involved in DNA synthesis and repair, mitosis, and chromatin remodeling highly enriched in the set of GFP-positive cells, consistent with a proliferative state. Perhaps more interestingly, gene sets involved in RNA processing and splicing, which were also observed in proliferative hepatocytes (13), were enriched in the GFP-positive cells, leading to new hypotheses and mechanisms that can be followed up in terms of their relation to the cell cycle.

A key question in the β-cell community has been whether β-cell proliferation and function are mutually exclusive. Klochendler et al. (12) found that genes involved in β-cell function, specifically insulin processing, insulin granule formulation, and glucose-stimulated insulin secretion, were repressed in the GFP-positive population. These findings coincided with gene-expression analysis of a human β-cell line designed to be conditionally quiescent. When the authors observed that most of the repressed genes were decreased approximately twofold, they reasoned that one possibility was a doubling of the total RNA content in proliferative cells. Such findings had been previously noted in global gene-expression analyses of cell lines (14). Using single-molecule RNA in situ hybridization (smRNA-FISH) fluorescence, they indeed found that to be the case: most genes in the GFP-positive population had doubled in their expression levels per cell. Interestingly, of those genes that failed to double, genes involved in β-cell function were enriched, suggesting that this set of transcripts may have some lag time in returning to a basal state, which would be required for maintaining differentiation after exit from the cell cycle.

This work represents a key step forward in our ability to analyze the cellular characteristics of proliferating β-cells in vivo. By using an elegant genetic approach to isolate replicating cells, the authors have uncovered many novel genetic factors that differentiate proliferative from quiescent β-cells. Further, the finding that β-cell functional genes are depleted during proliferation relative to most other genes will lead to new avenues for understanding the cellular maturation process and its interplay with cellular expansion. Of course, the use of mouse models leaves open the question of whether these findings will translate to human β-cells, but this work sets the stage for these future investigations. We should anticipate that advances in single-cell technologies will enable more sophisticated evaluation of cellular heterogeneity in the islet, both in the basal state and in response to various biological and pharmacological perturbations. Proteomics analyses of the β-cell are also indicated to determine whether gene-expression changes are manifest at the protein level and what other posttranscriptional processes are involved in regulating biological processes such as proliferation. The work of Klochendler et al. (12) will help the diabetes field refine the design of such high-dimensionality experiments and improve the questions it asks of the islet.

See accompanying article, p. 2081.

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

1.
Meier
JJ
,
Butler
AE
,
Saisho
Y
, et al
.
β-Cell replication is the primary mechanism subserving the postnatal expansion of β-cell mass in humans
.
Diabetes
2008
;
57
:
1584
1594
2.
Dor
Y
,
Brown
J
,
Martinez
OI
,
Melton
DA
.
Adult pancreatic beta-cells are formed by self-duplication rather than stem-cell differentiation
.
Nature
2004
;
429
:
41
46
3.
Nir
T
,
Melton
DA
,
Dor
Y
.
Recovery from diabetes in mice by beta cell regeneration
.
J Clin Invest
2007
;
117
:
2553
2561
4.
Porat
S
,
Weinberg-Corem
N
,
Tornovsky-Babaey
S
, et al
.
Control of pancreatic β cell regeneration by glucose metabolism
.
Cell Metab
2011
;
13
:
440
449
5.
Karnik
SK
,
Chen
H
,
McLean
GW
, et al
.
Menin controls growth of pancreatic beta-cells in pregnant mice and promotes gestational diabetes mellitus
.
Science
2007
;
318
:
806
809
6.
Kim
H
,
Toyofuku
Y
,
Lynn
FC
, et al
.
Serotonin regulates pancreatic beta cell mass during pregnancy
.
Nat Med
2010
;
16
:
804
808
7.
Bernal-Mizrachi
E
,
Kulkarni
RN
,
Scott
DK
,
Mauvais-Jarvis
F
,
Stewart
AF
,
Garcia-Ocaña
A
.
Human β-cell proliferation and intracellular signaling part 2: still driving in the dark without a road map
.
Diabetes
2014
;
63
:
819
831
8.
Layden
BT
,
Durai
V
,
Newman
MV
, et al
.
Regulation of pancreatic islet gene expression in mouse islets by pregnancy
.
J Endocrinol
2010
;
207
:
265
279
9.
Rieck
S
,
White
P
,
Schug
J
, et al
.
The transcriptional response of the islet to pregnancy in mice
.
Mol Endocrinol
2009
;
23
:
1702
1712
10.
Dorrell
C
,
Schug
J
,
Lin
CF
, et al
.
Transcriptomes of the major human pancreatic cell types
.
Diabetologia
2011
;
54
:
2832
2844
11.
Li
J
,
Klughammer
J
,
Farlik
M
, et al
.
Single-cell transcriptomes reveal characteristic features of human pancreatic islet cell types
.
EMBO Rep
2016
;
17
:
178
187
12.
Klochendler
A
,
Caspi
I
,
Corem
N
, et al
.
The genetic program of pancreatic β-cell replication in vivo
.
Diabetes
2016
;
65
:
2081
2093
13.
Klochendler
A
,
Weinberg-Corem
N
,
Moran
M
, et al
.
A transgenic mouse marking live replicating cells reveals in vivo transcriptional program of proliferation
.
Dev Cell
2012
;
23
:
681
690
14.
Lovén
J
,
Orlando
DA
,
Sigova
AA
, et al
.
Revisiting global gene expression analysis
.
Cell
2012
;
151
:
476
482