SA JOURNAL OF DIABETES & VASCULAR DISEASE
REVIEW
VOLUME 11 NUMBER 1 • MARCH 2014
25
conditions recorded a high frequency of spontaneous
in vitro
differentiation into cells with an immunohistochemical phenotype
akin to insulin-secreting islet cells. Accordingly, insulin secretion into
the culture medium was observed in a differentiation-dependent
manner. This was associated with the emergence of other islet
cell markers such as mRNA for islet
GK
and
GLUT2
genes and
the transcription factors
PDX1
and
Ngn3
, which are critical for
pancreatic neogenesis.
18
Leading on from this, research has focused on refining
in vitro
differentiation protocols to generate longterm functionally viable
islet cells from human ES cells for transplantation. Accordingly,
protocols that yield large numbers of insulin producing cells (~65%)
for transplantation have been developed that initially culture
human ES cells into embryoid bodies. The embryoid bodies are then
differentiated in a biphasic process: first into definitive endoderm
using the cytokines activin A and retinoic acid, and subsequently
into islet cell clusters, which mature under 3D conditions into insulin
producing islet cells that express functional islet cell markers such
as
PDX1
, C-peptide, insulin and
MafA
. This latter stage requires
amino acid-rich media supplemented with the incretins liraglutide
and exendin-4. When transplanted into rodent models of diabetes
these ES cellderived islets significantly improved hyperglycaemia for
up to 96 days.
19
The evidence on the utility of ES cells as a source of islet cell
regeneration is predominantly preclinical and currently limited in its
application towards ameliorating hyperglycaemia in animal models;
there are no clinical trial data on human diabetic patients.
Adult stem cells
Evidence suggests the pancreas has the capacity to regenerate,
which relates to its harbouring endogenous resident adult stem
cells or progenitor cells.
20
Possible progenitor cells include pancreatic
ductal cells that express the transcription factor PDX1.
21
It has been
demonstrated that long-term culture of rodent pancreatic ductal
tissue generates islet-like clusters; transplantation of these islet-like
clusters significantly ameliorates hyperglycaemia in rodent models
of diabetes.
22
Similar results have been reported using human
pancreaticductal tissue,whichcoined the term‘adult pancreatic stem
cell’.
23
Therefore, adult stem cells offer two therapeutic approaches
to the management of diabetes, namely stimulating endogenous
progenitor stem cell replication and differentiation
in situ
, and
generating adult stem cell-derived islets for transplantation.
Research on the former has been hindered by the lack of effective
markers to stimulate, expand, differentiate and enrich endogenous
adult pancreatic stem cells
in vitro
. However, research on the latter
has been more promising. Adult stem cells isolated from the liver,
bone marrow, central nervous system, adipose tissue and intestine
have been shown to differentiate into insulin-producing islets using
special culture conditions
in vitro
or, alternatively, by genetically
modifying them.
24
In particular, MSCs appear to have a favourable
propensity and developmental plasticity towards differentiating
into insulin secreting islets.
25
Transplantation of these adult stem
cell-derived insulin-secreting cells into rodent models of diabetes
has been shown to restore euglycaemia.
26
It has been demonstrated using a three-stage protocol that
bone marrow-derived MSCs from diabetic patients can generate
functional insulin producing cells
in vitro
. These can be used as an
autologous source of insulin producing cells for transplantation.
27
Furthermore, human placenta-derived MSCs have been used in a
pilot trial in 10 patients with longstanding type 2 diabetes, high
insulin dose requirements, islet cell dysfunction and poor glycaemic
control. Each patient received three infusions of human placenta-
derived MSCs and were followed up for three months. After
transplantation the daily mean dose of insulin was reduced from
63.7 IU to 34.7 IU (
p
< 0.01) and the C-peptide level was increased
from 4.1 to 5.6 ng/ml
(p
< 0.05), with an associated improvement
in cardiac and renal function. No adverse effects were reported. The
results of this small trial suggested that human placenta-derived
MSCs may potentially offer a simple, effective and safe treatment
option for diabetes.
28
A small-scale pilot study assessed the safety and efficacy of
umbilical cord blood, which contains a large population of stem
cells, for the treatment of recently diagnosed type 1 diabetes in
15 patients. Preliminary results suggest that autologous umbilical
cord blood infusions are safe. Furthermore, infusions of umbilical
cord blood ameliorated the rate of loss of endogenous insulin
production, reduced insulin requirements and HbA
1c
and increased
C-peptide levels in children with type 1 diabetes. These effects
occurred through an immunomodulatory mechanism that involved
regulatory T-lymphocytes that restored immune tolerance.
29
The mainstay of adult stem cell research for the treatment of
diabetes remains in the pre-clinical phase with no clinical trial data
published to date.
Induced pluripotent stem cells
A number of protocols, based on
in vivo
pancreatic development,
have been devised for stimulation of differentiation of iPS cells
into functional
β
-cells. This was first achieved in 2008 using a
four-stage procedure to differentiate human dermal fibroblast
derived-iPS cells into functional islet cell clusters.30 Functional
analysis of these cells using RT-PCR and immunohistochemistry
revealed the differentiated iPS cells positively expressed stage-
specific genes and antigen markers for each stage of development
–
Foxa2
and
Sox17
marked the expression of definitive endoderm;
PDX1
marked pancreatic endoderm development;
NKX6.1
and
Ptf1
marked the development of exocrine/endocrine cell clusters;
finally, expression of insulin, C-peptide and glucagon marked
the genesis of insulin-producing cells. However, the efficiency of
differentiation was limited with significant clonal variability in the
potential of iPS cells to differentiate into pancreatic endocrine
lineage cells.
A study in a rodent model of type 2 diabetes involving intra-
portal venous administration of iPS cell-derived insulin-producing
β
-like cell transplants resulted in the attainment of euglycaemia
for > 3 months with significant concomitant increases in insulin
concentration. This was sustained for 56 days post-transplantation
and further maintained for ~20–30% of the life expectancy of the
diabetic rodent model.
31
The risk of graft rejection was minimal;
the iPS cells were derived from fibroblasts from background mice
that were close relatives of the diabetic mice used in the study.
Limitations on the use of iPS cell-derived insulin-producing
β
-like cells include: low reprogramming efficiency with a typical
reprogramming event occurring in 0.01–0.1% of cultured cells;
low differentiation efficiency into functional insulin-producing
β
-like cells; autoimmune mediated destruction of newly generated
β
-cells. Furthermore, the mainstay of iPS cell research in the
treatment of diabetes remains in the pre-clinical phase with no
clinical trial data published to date.
Future challenges to optimise the utility of iPS cellderived insulin-
producing
β
-like cells include: generating safe iPS cells with no risk of
