You are on page 1of 7

Cell Metabolism

Short Article

Loss of Autophagy Diminishes Pancreatic b Cell

Mass and Function with Resultant Hyperglycemia
Hye Seung Jung,1 Kun Wook Chung,1 Jeong Won Kim,1 Jin Kim,2 Masaaki Komatsu,3,4,5 Keiji Tanaka,3
Yen Hoang Nguyen,6 Tong Mook Kang,6 Kun-Ho Yoon,7 Ji-Won Kim,7 Yeon Taek Jeong,1 Myoung Sook Han,1
Moon-Kyu Lee,1 Kwang-Won Kim,1 Jaekyoon Shin,8 and Myung-Shik Lee1,*
1Department of Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, 50 Irwon-dong Kangnam-ku,

Seoul 135-710, Korea

2Department of Anatomy and Cell Death Disease Research Center, College of Medicine, The Catholic University of Korea,

505 Banpo-dong, Seocho-ku, Seoul 137-701, Korea

3Laboratory of Frontier Science, Tokyo Metropolitan Institute of Medical Science, Bunkyo-ku, Tokyo 113-8613, Japan
4Department of Biochemistry, Juntendo University School of Medicine, Bunkyo-ku, Tokyo 113-8421, Japan
5PRESTO, Japan Science and Technology Corporation, Kawaguchi 332-0012, Japan
6Department of Physiology, Sungkyunkwan University School of Medicine, Suwon 440-746, Korea
7Department of Endocrinology and Metabolism, College of Medicine, The Catholic University of Korea, 505 Banpo-dong,

Seocho-ku, Seoul 137-701, Korea

8Department of Molecular Cell Biology, Sungkyunkwan University School of Medicine, Suwon 440-746, Korea

DOI 10.1016/j.cmet.2008.08.013

SUMMARY amino acid retrieval (Lum et al., 2005). Paradoxically, autophagy

also represents a form of nonapoptotic cell death and is called as
Autophagy is a cellular degradation-recycling sys- ‘‘type 2 programmed cell death’’ (Boya et al., 2005). Thus, au-
tem for aggregated proteins and damaged organ- tophagy may either promote cell death through excessive degra-
elles. Although dysregulated autophagy is implicated dation of cellular constituents or protect cells from cell death by
in various diseases including neurodegeneration, its providing essential nutrients and removing damaged organelles
role in pancreatic b cells and glucose homeostasis during cellular stress, depending on the cellular and environmen-
tal context. Because autophagy degrades unnecessary or dys-
has not been described. We produced mice with
functional organelles such as mitochondria and rejuvenates their
b cell-specific deletion of Atg7 (autophagy-related 7).
function, autophagy is expected to play important roles in the
Atg7 mutant mice showed impaired glucose toler- development and the maintenance of physiological function of
ance and decreased serum insulin level. b cell normal tissues (Klionsky and Emr, 2000). In fact, disruption of
mass and pancreatic insulin content were reduced autophagic process leads to failure of cavitation during embryo-
because of increased apoptosis and decreased pro- genesis (Qu et al., 2007) or accumulation of abnormal mitochon-
liferation of b cells. Physiological studies showed dria in adult tissues (Komatsu et al., 2005). In addition to the
reduced basal and glucose-stimulated insulin secre- physiological roles, dysregulated autophagy may play pathoge-
tion and impaired glucose-induced cytosolic Ca2+ netic roles in diverse disease processes. Targeted disruption of
transients in autophagy-deficient b cells. Morpho- autophagy genes in specific tissues induced neurodegeneration
logic analysis revealed accumulation of ubiquitinated (Hara et al., 2006; Komatsu et al., 2006) or cardiomyopathy, par-
ticularly when cellular stress is increased, which is probably due
protein aggregates colocalized with p62, which was
to accumulation of damaged molecules and organelles (Nakai
accompanied by mitochondrial swelling, endoplas-
et al., 2007). Other studies revealed important roles for auto-
mic reticulum distension, and vacuolar changes in phagy in eradication of microbes, antigen presentation, tumor
b cells. These results suggest that autophagy is suppression (Levine and Kroemer, 2008), and endoplasmic retic-
necessary to maintain structure, mass and function ulum (ER) stress responses (Yorimitsu et al., 2006; Ogata et al.,
of pancreatic b cells, and its impairment causes 2006).
insulin deficiency and hyperglycemia because of ab- Type 2 diabetes (T2D) is characterized by abnormal regulation
normal turnover and function of cellular organelles. of nutrients and their metabolites that develops as a conse-
quence of combined insulin resistance and relative insulin defi-
ciency. Insulin and its downstream molecules such as mTOR/
INTRODUCTION S6K1 are well-known inhibitors of autophagy (Blommaart et al.,
1995), whereas glucagon, a counterregulatory hormone of insu-
Macroautophagy (here referred to as autophagy) is a dynamic lin, induces autophagy (Ashford and Porter, 1962). Because T2D
process involving the rearrangement of subcellular membranes critically depends on the metabolism of nutrients, insulin action,
to sequester cytoplasm and organelles for delivery to lyso- and its counterregulation, dysregulated autophagy may play
somes, where the sequestered material is degraded and re- a role in the pathogenesis of T2D. Furthermore, cellular organ-
cycled (Klionsky and Emr, 2000). Autophagy is activated during elles that play crucial roles in b cell survival, death, insulin secre-
nutrient deprivation to promote cell survival and to increase tion, and insulin action, or sensitivity, such as mitochondria

318 Cell Metabolism 8, 318–324, October 8, 2008 ª2008 Elsevier Inc.

Cell Metabolism
Autophagy in Pancreatic b Cells

Figure 1. Production of b Cell-Specific Atg7 Knockout (Atg7Db cell) Mice

(A) Atg7F/F mice were bred with RIP-Cre mice to produce Atg7Db cell mice. Top: PCR analysis with genomic DNA from isolated pancreatic islet cells and primers
flanking the floxed region was done to confirm Cre-mediated recombination in Atg7Db cell mice. In Atg7F/F mice, the size of the PCR product was larger than that in
wild-type mice because of the inserted Flox sequence. Bottom: RT-PCR was conducted with total RNA from primary islet cells and specific primers to examine
Atg7 expression.
(B) Number and area of autophagosomes with double-membrane vacuole-like structure representing steady-state autophagy level were measured in pancreatic
b cells of overnight fasted animals by EM (n = 2 each). Twenty random spots (total 2000–2400 mm2) were scanned per mouse.
(C) Body weight of Atg7Db cell mice and littermates. Representative data from mice between 6 and 20 weeks of age are presented (n = 15 each).
(D) Random blood glucose level was measured with a glucometer (*, p < 0.05 versus Atg7F/F and Atg7F/W:Cre+ mice; n = 15 each).
(E) IPGTT was done at 20 weeks of age by intraperitoneal injection of 1 g/kg glucose (*, p < 0.01 versus Atg7F/F and Atg7F/W:Cre+ mice; #, p < 0.01 between Atg7F/F
and Atg7F/W:Cre+ mice; n = 15 each).
(F) Fasting serum insulin level in vivo was measured by radioimmunoassay (*, p < 0.05 versus Atg7F/F and Atg7F/W:Cre+ mice; n = 15 each).
In (B)–(F), error bars represent the SEM.

(Petersen et al., 2004; Rodriguez-Enriquez et al., 2006; Maechler Atg7Db cell mice (Figure 1A, top). When we studied Atg7 mRNA
and Wollheim, 1999) and ER (Bernales et al., 2007; Laybutt et al., expression by RT-PCR analysis, it was undetectable in pancre-
2007; Ozcan et al., 2006), rely on autophagy for proper function. atic islet cells of Atg7Db cell mice. In islet cells of Atg7F/W:Cre+
All these findings suggest the possibility that autophagy is in- mice, Atg7 expression was slightly decreased compared to lit-
volved in diabetes by controlling hormone action and organelle termate Atg7F/F mice (Figure 1A, bottom). Accordingly, the num-
function. However, the relationship between autophagy and dia- ber and area of starvation-induced autophagosomes measured
betes has been hardly studied (Marsh et al., 2007; Kaniuk et al., by electron microscopy (EM) were markedly decreased in
2007), partly because of lack of proper animal models. We con- pancreatic b cells of Atg7Db cell mice compared to Atg7F/F mice
ducted this investigation to examine the role of autophagy in (Figure 1B).
b cell function and glucose homeostasis using a tissue-specific
knockout mouse model. Impaired Glucose Tolerance in b Cell-Specific Atg7
Knockout Mice
RESULTS AND DISCUSSION Up to 1 yr after birth, Atg7Db cell mice were indistinguishable in
appearance and growth from age-matched controls: Atg7F/F
Production of b Cell-Specific Atg7 Knockout Mice and Atg7F/W:Cre+ littermates (Figure 1C). However, significant
By crossing Atg7-floxed mice (Atg7F/F) (Komatsu et al., 2005, increases in random glucose level were detected in Atg7Db cell
2006) with RIP-Cre mice, we generated b cell-specific Atg7 mice since 8 weeks of age (Figure 1D). Intraperitoneal glucose
knockout mice (Atg7Db cell). PCR analysis confirmed deletion of tolerance test (IPGTT) at 20 weeks of age disclosed significant
floxed sequence in genomic DNA from primary islet cells of glucose intolerance in Atg7Db cell mice. Atg7F/W:Cre+ mice

Cell Metabolism 8, 318–324, October 8, 2008 ª2008 Elsevier Inc. 319

Cell Metabolism
Autophagy in Pancreatic b Cells

Figure 2. Morphological and Functional

Changes in b Cells of Atg7Db cell Mice
(A) H&E staining showing variable-sized vacuo-
lated cells in pancreatic islets of Atg7Db cell mice
(arrows) that are similar in morphology to the de-
generation of autophagy-deficient hepatocytes
(Komatsu et al., 2005). Some vacuolated cells
stained positive for insulin.
(B) Paraffin-embedded pancreatic sections were
subjected to insulin IHC, and relative b cell area
was determined by point counting (*, p < 0.05 ver-
sus Atg7F/F and Atg7F/W:Cre+ mice; n = 9 each).
(C) Combined insulin IHC and TUNEL staining was
done to count apoptotic b cell number/islet (*, p <
0.05 versus Atg7F/F and Atg7F/W:Cre+ mice; n = 9
each). Representative apoptotic b cells are shown
(arrow heads).
(D) Double IHC for insulin and BrdU after intraper-
itoneal injection of BrdU was conducted for the
evaluation of b cell proliferation (p < 0.05 by
ANOVA; *, p < 0.05 versus Atg7F/F mice by post
hoc analysis; n = 9 each).
(E) Pancreatic insulin was extracted by acid-
ethanol, and insulin content was measured by
radioimmunoassay (*, p < 0.001 versus Atg7F/W:
Cre+ mice; n = 8 each).
(F) Insulin secretion ex vivo. Pancreatic islets from
20-week-old mice (n = 3 each) were incubated in
KRBB containing 3.0 mM glucose for 1 hr and
then in KRBB containing 16.7 mM glucose for
1 hr. Buffer was collected for insulin radioimmuno-
assay. Results are representative of two indepen-
dent experiments performed in triplicate (*, p <
0.05 versus Atg7F/F and Atg7F/W:Cre+ mice; #,
p < 0.05 versus Atg7F/F mice).
(G) Glucose-induced Ca2+ transients in isolated is-
lets (n = 5 each). [Ca2+]c was determined in Tyrode
solution containing 3.0 or 16.7 mM glucose (36 C).
In (B)–(F), error bars represent the SEM.

showed no difference from Atg7F/F mice in random or fasting glu- bined with insulin IHC to study the mechanism of decreased
cose level. However, IPGTT revealed impaired glucose tolerance b cell mass in Atg7Db cell mice, a significant increase of b cell ap-
in Atg7F/W:Cre+ mice, although less prominent than that of optosis was found in Atg7Db cell mice compared to control mice
Atg7Db cell mice (Figure 1E). Fasting serum insulin concentration (0 ± 0/islet in Atg7F/F, 0.012 ± 0.009/islet in Atg7F/W:Cre+, and
was significantly lower in Atg7Db cell mice than in control mice (p < 0.079 ± 0.029/islet in Atg7Db cell mice; p < 0.05 versus Atg7F/F
0.05 versus Atg7F/F and Atg7F/W:Cre+ mice) (Figure 1F), suggest- and Atg7F/W:Cre+ mice) (Figure 2C). We also studied whether au-
ing that insulin deficiency is responsible for the impaired glucose tophagy is involved in b cell proliferation that occurs at a low level
tolerance in Atg7Db cell mice. We also measured body weight and but is important for the maintenance of b cell mass (Dor et al.,
glucose and insulin levels of RIP-Cre mice, which were not signif- 2004). BrdU+ b cell number/islet representing b cell proliferation
icantly different from those of Atg7F/F mice (data not shown) as was significantly reduced in Atg7Db cell mice (0.100 ± 0.039/islet
reported (Fex et al., 2007; Kim et al., 2007). in Atg7F/F, 0.072 ± 0.016/islet in Atg7F/W:Cre+, and 0.008 ±
0.006/islet in Atg7Db cell mice; p < 0.05 versus Atg7F/F mice)
Morphological Changes of Atg7-Deficient b Cells (Figure 2D), suggesting that both increased apoptosis and de-
Next, we conducted morphological analysis of pancreatic islets creased proliferation contribute to the decreased b cell mass
of Atg7Db cell mice. Microscopic examination after hematoxylin of Atg7Db cell mice. Consistent with the decreased b cell mass, in-
and eosin (H&E) staining showed frequent vacuolated cells in sulin content in the pancreas of Atg7Db cell mice was significantly
islets of Atg7Db cell mice (Figure 2A), which were not found in con- lower than that in the pancreas of control Atg7F/W:Cre+ mice
trol mice. Some of the vacuolated cells contained insulin+ mate- (Figure 2E). In contrast to the b cell changes, a and d cells of
rials. Relative b cell mass estimated by point counting after insu- Atg7Db cell mice evaluated by IHC for glucagon and somatostatin
lin immunohistochemistry (IHC) was significantly reduced to were not different in morphology or distribution from those of
about 40% that of control mice (1.20% ± 0.32% in Atg7F/F, control mice (data not shown). The molecular mechanism of
1.32% ± 0.09% in Atg7F/W:Cre+, and 0.56% ± 0.06% in Atg7Db cell the increased apoptosis and decreased proliferation of auto-
mice; p < 0.05 versus Atg7F/F and Atg7F/W:Cre+ mice) (Figure 2B). phagy-deficient b cells is unknown. The inhibition of autophagy
When we examined b cell apoptosis using TUNEL staining com- may results in a shortage of bioenergetic sources such as

320 Cell Metabolism 8, 318–324, October 8, 2008 ª2008 Elsevier Inc.

Cell Metabolism
Autophagy in Pancreatic b Cells

Figure 3. Accumulation of Ubiquitin Aggregates in b Cells of Atg7Db cell Mice

(A) Pancreatic sections were subjected to IHC for ubiquitin and p62.
(B) Double immunofluorescent staining for ubiquitin (red) and p62 (green) was done, and confocal microscopy was employed so that their colocalization would be
examined. Hoechst staining was applied for identification of nuclei (blue). (Arrowheads, yellow color after merge, indicating colocalization of ubiquitin and p62.)

NADPH2 and ATP that may trigger apoptosis, presumably vestigate whether autophagy affects intracellular pathway regu-
through a direct effect on mitochondria, or deinhibit the activa- lating insulin secretion in response to glucose. Tracing of [Ca2+]c
tion of caspases or permeabilization of mitochondrial outer showed that glucose-induced cytosolic Ca2+ transients were
membrane (Maiuri et al., 2007). p27 activated by metabolic defective in primary islet cells from Atg7Db cell mice (Figure 2G),
stress and energy deficiency may also lead to cell-cycle arrest suggesting that b cell function is impaired even in apparently
and decreased proliferation (Liang et al., 2007). Morphological healthy viable b cells from Atg7Db cell mice and that autophagy
or functional changes of ER leading to ER stress response ob- is necessary for physiological function of b cells. Some islet cells
served in autophagy-deficient cells (Nakai et al., 2007; see EM isolated from Atg7Db cell mice showing ‘‘damaged’’ appearance
pictures below) may also play a role in the increased apoptosis had increased basal [Ca2+]c (data not shown), suggesting more
and decreased proliferation of b cells observed in Atg7Db cell severe defects in Ca2+ handling.
mice (Harding et al., 2001; Wu and Kaufman, 2006; Zhang
et al., 2006). Hence, constitutive turnover of proteins and organ- Ubiquitinated Aggregates Colocalized with p62
elles might be important not only for the maintenance of organ- in Atg7-Deficient b Cells
elle function and cell viability but also for supply of basic nutrients We next conducted ubiquitin IHC to study possible accumula-
that are necessary for low-grade proliferation of b cells. tion of ubiquitinated proteins by genetic ablation of b cell
autophagy because polyubiquitinated proteins are reportedly
Functional Changes of Atg7-Deficient b Cells degraded by autophagy (Hara et al., 2006; Nakai et al., 2007; Ko-
We next conducted functional analysis of pancreatic islets from matsu et al., 2007). Large ubiquitin+ cytoplasmic inclusions were
Atg7Db cell mice. When we measured insulin secretion ex vivo us- found in pancreatic islet cells of Atg7Db cell mice but not in those
ing the same number of viable primary islets of similar size after of control mice (Figure 3A), similar to the findings in autophagy-
excluding apparently damaged or nonviable cells, basal insulin deficient hepatocytes (Komatsu et al., 2007). We also investi-
secretion was significantly decreased in Atg7Db cell mice com- gated expression of p62 that links ubiquitinated proteins to au-
pared to control mice (p < 0.05 versus Atg7F/F and Atg7F/W: tophagy apparatus (Bjorkoy et al., 2005). In islet cells of Atg7Db cell
Cre+ mice). High glucose-stimulated insulin secretion from pri- mice, p62 accumulated probably because p62 is degraded by
mary islets of Atg7Db cell mice was also significantly reduced autophagy (Komatsu et al., 2007) (Figure 3A). Confocal
compared to control mice (p < 0.05 versus Atg7F/F and Atg7F/W: microscopy at higher magnifications showed that ubiquitin and
Cre+ mice) (Figure 2F). Glucose-stimulated insulin secretion was p62 were colocalized (Figure 3B). These results suggest the pos-
also compromised in primary islets of Atg7F/W:Cre+ mice (p < sibility that accumulation of ubiquitinated proteins associated
0.05 versus Atg7F/F mice), while less pronounced compared to with p62 contributes to increased apoptosis-decreased prolifer-
Atg7Db cell mice. We next measured glucose-induced changes ation of b cells, decreased b cell mass, and impaired insulin
of cytosolic Ca2+ concentration ([Ca2+]c) in viable islet cells to in- secretion from autophagy-deficient b cells; however, the role of

Cell Metabolism 8, 318–324, October 8, 2008 ª2008 Elsevier Inc. 321

Cell Metabolism
Autophagy in Pancreatic b Cells

Figure 4. EM Analysis of the Pancreas at

20 Weeks of Age
(A) b cells of Atg7Db cell mice with markedly re-
duced number of insulin granules (left, white ar-
row) compared to those of Atg7F/F mice with nu-
merous insulin granules dispersed in the
cytoplasm (right).
(B) Ubiquitin aggregates (top, asterisk) and large
cystic structures (bottom, double asterisks) were
found in b cells of Atg7Db cell mice, which were
not found in those of Atg7F/F mice. The cystic
structure had an opening to the plasma membrane
(C) Ultrastructural changes in b cells of Atg7Db cell
mice at higher magnifications: swelling of mito-
chondria (white arrowheads) and cisternal disten-
sion of ER (circle) and Golgi complex (rectangle).
(D) Organelles surrounded by double membrane
representing autophagosomes (arrows) were
seen in pancreatic b cells of Atg7F/F mice.

ER dysfunction or unbalanced control of

insulin degradation secondary to autoph-
agy deficiency (Uchizono et al., 2007)
might account for the decreased num-
bers of insulin granules.
Our results demonstrate that constitu-
tive autophagy plays important roles in
the maintenance of pancreatic b cell
mass by regulating cell survival and
proliferation. Glucose-induced insulin se-
inclusion body formation in those processes is yet to be eluci- cretion was also impaired in b cells lacking autophagy, which is
dated (Casas et al., 2007; Komatsu et al., 2007). Although diabe- attributable to the decreased glucose-induced Ca2+ transients
tes state itself could induce formation of ubiquitin aggregates in caused by dysfunction of cellular organelles such as mitochon-
b cells (Kaniuk et al., 2007), contribution of hyperglycemia to the dria. Consequently, b cell mass was decreased, serum insulin
ubiquitin aggregates in b cells of Atg7Db cell mice would be minor level decreased, and hyperglycemia developed in mice with ge-
because the size and number of ubiquitin aggregates in b cells netic ablation of b cell autophagy. We speculate that accumula-
were much smaller in other types of diabetes models such as tion of ubiquitinated proteins and dysfunction of organelles such
db/db mice than those in Atg7Db cell mice (M.S.H. and M.-S.L., un- as ER and mitochondria contribute to these changes. Autophagy
published data), and hyperglycemia was mild in Atg7Db cell mice. could be a key regulator of cellular organelles in b cells, where
vigorous protein synthesis occurs, abundant energy is required,
Electron Microscopy and therefore function of ER and mitochondria may be critical,
We finally studied ultrastructural changes associated with ge- particularly when increased demand for insulin production exists
netic ablation of b cell autophagy. EM revealed that some b cells such as insulin resistance.
of Atg7Db cell mice contained remarkably reduced numbers of in-
sulin granules (Figure 4A, left) compared to b cells of Atg7F/F EXPERIMENTAL PROCEDURES
mice (Figure 4A, right). Ubiquitin aggregates (Figure 4B, top)
Production of Atg7Db cell Mice
and large cystic structure (Figure 4B, bottom) that correspond
Atg7F/F mice were crossed with RIP-Cre mice (Jackson Laboratory) for the
to a part of vacuolated islet cells (Figure 2A) were also found in generation of Atg7Db cell mice. Cre-mediated recombination was confirmed by
b cells of Atg7Db cell mice. Some cystic structures were open to PCR with genomic DNA from isolated pancreatic islet cells and primers flanking
plasma membrane (Figure 4B, bottom). Further magnification the floxed region (forward, TGGCTGCTACTTCTGCAATGATGT; reverse,
revealed more ultrastructural abnormalities such as swelling of CTAAGCAGGTGAGATCTACACTCA). Blood glucose level and body weight
mitochondria and cisternal distension of rough ER and Golgi were monitored weekly. Similar number of male and female mice in each group
was employed for all experiments. All animal experiments were conducted in
complex (Figure 4C), even in apparently normal-looking b cells
accordance with the institutional guidelines of Samsung Medical Center.
of Atg7Db cell mice at lower magnifications. In b cells of Atg7F/F
mice, such abnormal changes were not found but occasional au-
tophagosomes were found (Figure 4D). The cellular mechanism Expression of Atg7 was examined by PCR with total RNA from primary islet
responsible for the sparse insulin granules in b cells of Atg7Db cell cells and specific primers (forward, ATGCCAGGACACCCTGTGAACTTC;
mice is unclear. Defective biogenesis of insulin granules due to reverse, CAGGACAGAGACCATCAGCTCCAC).

322 Cell Metabolism 8, 318–324, October 8, 2008 ª2008 Elsevier Inc.

Cell Metabolism
Autophagy in Pancreatic b Cells

Glucose Tolerance Test at 3 5000. Number and area of autophagosomes with double-membrane vac-
IPGTT was carried out after overnight fasting by intraperitoneal injection of uole-like structure in the scanned area were measured with a software pro-
1 g/kg glucose to 20-week-old mice. Blood glucose concentrations were de- gram (ImagePro, Media Cybernetics).
termined with an Accu-Check glucometer (Roche) before (0 min) and 15, 30,
60, and 120 min after glucose injection. Serum insulin concentration was mea- Statistical Analysis
sured with a commercial radioimmunoassay kit for rat and mouse insulin mea- All values were expressed as means ± SEM. Student’s t test and one-way
surement (Linco). ANOVA with post hoc Tukey test were employed for two-group and multiple
comparisons, respectively. p values less than 0.05 were considered to repre-
Measurement of Pancreatic Insulin Content sent statistically significant differences.
Pancreatic insulin was extracted by a previously published method (Yoon
et al., 2003). Insulin content was measured by radioimmunoassay as de-
scribed above and normalized to pancreas weight.
We thank Hong-Lim Kim, Dong-Sik Ham, and Seung Hoon Oh for technical as-
Isolation of Pancreatic Islets sistance. This work was supported by a Nano/Bio Science Program Grant
Islets were isolated from overnight fasted mice via the collagenase digestion (2004-00716) and by the 21C Frontier Functional Proteomics Project of the Ko-
technique (Chang et al., 2004). rean Ministry of Science and Technology (FPR05C1-160). M.-S.L. is the recip-
ient of SRC Grant R11-2000-080-11003-0 from the Korea Science and Engi-
Glucose-Stimulated Insulin Secretion Ex Vivo neering Foundation.
After isolation, medium-sized (150 mm in diameter) islets were picked, and 20
islets of each genotype were placed on the 12 mm-Millicell inserts (Millipore) in
Received: March 31, 2008
24-well plates. They were starved in Krebs-Ringer bicarbonate buffer (KRBB)
Revised: June 29, 2008
containing 3.0 mM glucose for 1 hr, and wells were replenished with 800 ml of
Accepted: August 15, 2008
the same buffer. After incubation for additional 1 hr, the replenished buffer was
Published: October 7, 2008
collected for measurement of basal insulin release by radioimmunoassay.
Then, islets were stimulated with 800 ml of KRBB containing 16.7 mM glucose
for 1 hr. The buffer was collected for assay of glucose-stimulated insulin
Ashford, T.P., and Porter, K.R. (1962). Cytoplasmic components in hepatic cell
2+ lysosomes. J. Cell Biol. 12, 198–202.
Measurement of [Ca ]c
[Ca2+]c was measured as published with modifications (Chang et al., 2004). In Bernales, S., Schuck, S., and Walter, P. (2007). ER-phagy: Selective auto-
brief, medium-sized islets were loaded with 4 mM Fura 2-AM for 30 min at room phagy of the endoplasmic reticulum. Autophagy 3, 285–287.
temperature. After islets were washed twice with Tyrode solution, [Ca2+]c was Bjorkoy, G., Lamark, T., Brech, A., Outzen, H., Perander, M., Overvatn, A.,
recorded by a microfluorometric system comprising an inverted fluorescence Stenmark, H., and Johansen, T. (2005). p62/SQSTM1 forms protein aggre-
microscope (Olympus IX-70) equipped with a dry-type fluorescence objective gates degraded by autophagy and has a protective effect on huntingtin-in-
lens, a photomultiplier tube (Hamamatsu), and Deltascan illuminator (Photon duced cell death. J. Cell Biol. 171, 603–614.
Technology International). [Ca2+]c was calculated from the fluorescence emis- Blommaart, E.F., Luiken, J.J., Blommaart, P.J., van Woerkom, G.M., and
sion ratio at 340 nm/380 nm excitation. Meijer, A.J. (1995). Phosphorylation of ribosomal protein S6 is inhibitory for au-
tophagy in isolated rat hepatocytes. J. Biol. Chem. 270, 2320–2326.
Histological Analysis
Boya, P., Gonzalez-Polo, R.A., Casares, N., Perfettini, J.L., Dessen, P., Lar-
After fixation and embedding, the pancreata were sectioned to a thickness of
ochette, N., Metivier, D., Meley, D., Souquere, S., Yoshimori, T., et al. (2005).
3 mm for H&E staining. IHC for insulin, glucagons, somatostatin, ubiquitin, and
Inhibition of macroautophagy triggers apoptosis. Mol. Cell. Biol. 25, 1025–
p62 was done as described (Kim et al., 1999). Relative b cell mass was esti-
mated by point counting after insulin IHC (Montana et al., 1993). An average
Casas, S., Gomis, R., Gribble, F.M., Altirriba, J., Knuutila, S., and Novials, A.
of 7000 points/mouse was counted. The number of test points was chosen us-
(2007). Impairment of the ubiquitin-proteasome pathway is a downstream en-
ing a nomogram to set the probable error of volume density to less than 10% of
doplasmic reticulum stress response induced by extracellular human islet am-
the calculated value (Weibel, 1978). TUNEL staining (Roche) combined with in-
yloid polypeptide and contributes to pancreatic beta-cell apoptosis. Diabetes
sulin IHC was conducted for the identification of apoptotic b cells (Kim et al.,
56, 2284–2294.
1999). So that b cell proliferation could be measured, BrdU incorporation
into b cells 16 hr after intraperitoneal injection of 100 mg/kg BrdU was evalu- Chang, I., Cho, N., Kim, S., Kim, J.Y., Kim, E., Woo, J.E., Nam, J.H., Kim, S.J.,
ated by double IHC for insulin and BrdU as described (Kim et al., 1999). and Lee, M.S. (2004). Role of calcium in pancreatic islet cell death by IFN-g/
TNF-a. J. Immunol. 172, 7008–7014.
Confocal Microscopy Dor, Y., Brown, J., Martinez, O.I., and Melton, D.A. (2004). Adult pancreatic
Colocalization of p62 and ubiquitin was evaluated by confocal microscopy (Ni- beta-cells are formed by self-duplication rather than stem-cell differentiation.
kon E600) after double immunofluorescent staining and Hoeschst33342 stain- Nature 429, 41–46.
ing (Molecular Probes). Fex, M., Wierup, N., Nitert, M.D., Ristow, M., and Mulder, H. (2007). Rat insulin
promoter 2-Cre recombinase mice bred onto a pure C57BL/6J background
Antibodies exhibit unaltered glucose tolerance. J. Endocrinol. 194, 551–555.
Primary antibodies used for IHC or confocal microscopy were as follows:
Hara, T., Nakamura, K., Matsui, M., Yamamoto, A., Nakahara, Y., Suzuki-Mi-
anti-insulin, anti-somatostatin, anti-ubiquitin (from DAKO), anti-glucagon
gishima, R., Yokoyama, M., Mishima, K., Saito, I., Okano, H., and Mizushima,
(Chemicon), and anti-p62 (Abnova). Secondary antibodies with or without fluo-
N. (2006). Suppression of basal autophagy in neural cells causes neurodegen-
rescence were from Zymed.
erative disease in mice. Nature 441, 885–889.

Electron Microscopy Harding, H.P., Zeng, H., Zhang, Y., Jungries, R., Chung, P., Plesken, H., Saba-
After fixation in 2.5% glutaraldehyde-0.1 M NaH2PO4/Na2HPO4 phosphate tini, D.D., and Ron, D. (2001). Diabetes mellitus and exocrine pancreatic dys-
buffer (pH 7.2) and PBS washing, pancreas tissues were postfixed with 1% function in perk / mice reveals a role for translational control in secretory cell
OsO4 and embedded in Epon812. Ultrathin sections were mounted and survival. Mol. Cell 7, 1153–1163.
stained with uranyl acetate/lead citrate (JEM-1010, Jeol). They were randomly Kaniuk, N.A., Kiraly, M., Bates, H., Vranic, M., Volchuk, A., and Brumell, J.H.
scanned at 20 different spots (total 2000–2400 mm2) per each mouse (2007). Ubiquitinated-protein aggregates form in pancreatic beta-cells during

Cell Metabolism 8, 318–324, October 8, 2008 ª2008 Elsevier Inc. 323

Cell Metabolism
Autophagy in Pancreatic b Cells

diabetes-induced oxidative stress and are regulated by autophagy. Diabetes mone content in secretory-deficient pancreatic endocrine beta-cells. Mol.
56, 930–939. Endocrinol. 21, 2255–2269.
Kim, S., Millet, I., Kim, H.S., Kim, J.Y., Han, M.S., Lee, M.K., Kim, K.W., Sher- Montana, E., Bonner-Weir, S., and Weir, G.C. (1993). Beta cell mass and
win, R.S., Karin, M., and Lee, M.S. (2007). NF-kappa B prevents beta cell death growth after syngeneic islet cell transplantation in normal and streptozotocin
and autoimmune diabetes in NOD mice. Proc. Natl. Acad. Sci. USA 104, 1913– diabetic C57BL/6 mice. J. Clin. Invest. 91, 780–787.
1918. Nakai, A., Yamaguchi, O., Takeda, T., Higuchi, Y., Hikoso, S., Taniike, M.,
Kim, Y.H., Kim, S., Kim, K.A., Yagita, H., Kayagaki, N., Kim, K.W., and Lee, Omiya, S., Mizote, I., Matsumura, Y., Asahi, M., et al. (2007). The role of au-
M.S. (1999). Apoptosis of pancreatic beta-cells detected in accelerated tophagy in cardiomyocytes in the basal state and in response to hemodynamic
diabetes of NOD mice: No role of Fas-Fas ligand interaction in autoimmune stress. Nat. Med. 13, 619–624.
diabetes. Eur. J. Immunol. 29, 455–465. Ogata, M., Hino, S., Saito, A., Morikawa, K., Kondo, S., Kanemoto, S., Mura-
Klionsky, D.J., and Emr, S.D. (2000). Autophagy as a regulated pathway of cel- kami, T., Taniguchi, M., Tanii, I., Yoshinaga, K., et al. (2006). Autophagy is ac-
lular degradation. Science 290, 1717–1721. tivated for cell survival after endoplasmic reticulum stress. Mol. Cell. Biol. 26,
Komatsu, M., Waguri, S., Chiba, T., Murata, S., Iwata, J., Tanida, I., Ueno, T., 9220–9231.
Koike, M., Uchiyama, Y., Kominami, E., and Tanaka, K. (2006). Loss of auto- Ozcan, U., Yilmaz, E., Ozcan, L., Furuhashi, M., Vaillancourt, E., Smith, R.O.,
phagy in the central nervous system causes neurodegeneration in mice. Gorgun, C.Z., and Hotamisligil, G.S. (2006). Chemical chaperones reduce
Nature 441, 880–884. ER stress and restore glucose homeostasis in a mouse model of type 2 diabe-
Komatsu, M., Waguri, S., Ueno, T., Iwata, J., Murata, S., Tanida, I., Ezaki, J., tes. Science 313, 1137–1140.
Mizushima, N., Ohsumi, Y., Uchiyama, Y., et al. (2005). Impairment of starva- Petersen, K.F., Dufour, S., Befroy, D., Garcia, R., and Shulman, G.I. (2004). Im-
tion-induced and constitutive autophagy in Atg7-deficient mice. J. Cell Biol. paired mitochondrial activity in the insulin-resistant offspring of patients with
169, 425–434. type 2 diabetes. N. Engl. J. Med. 350, 664–671.
Komatsu, M., Waguri, S., Koike, M., Sou, Y.S., Ueno, T., Hara, T., Mizushima, Qu, X., Zou, Z., Sun, Q., Luby-Phelps, K., Cheng, P., Hogan, R.N., Gilpin, C.,
N., Iwata, J., Ezaki, J., Murata, S., et al. (2007). Homeostatic levels of p62 con- and Levine, B. (2007). Autophagy gene-dependent clearance of apoptotic cells
trol cytoplasmic inclusion body formation in autophagy-deficient mice. Cell during embryonic development. Cell 128, 931–946.
131, 1149–1163. Rodriguez-Enriquez, S., Kim, I., Currin, R.T., and Lemasters, J.J. (2006).
Laybutt, D.R., Preston, A.M., Akerfeldt, M.C., Kench, J.G., Busch, A.K., Bian- Tracker dyes to probe mitochondrial autophagy (mitophagy) in rat hepato-
kin, A.V., and Biden, T.J. (2007). Endoplasmic reticulum stress contributes to cytes. Autophagy 2, 39–46.
beta cell apoptosis in type 2 diabetes. Diabetologia 50, 752–763. Uchizono, Y., Alarcon, C., Wicksteed, B.L., Marsh, B.J., and Rhodes, C.J.
Levine, B., and Kroemer, G. (2008). Autophagy in the pathogenesis of disease. (2007). The balance between proinsulin biosynthesis and insulin secretion:
Cell 132, 27–42. Where can imbalance lead? Diabetes Obes. Metab. 9, 56–66.
Liang, J., Shao, S.H., Xu, Z.X., Hennessy, B., Ding, Z., Larrea, M., Kondo, S., Weibel, E.R. (1978). Sterologic methods. In Practical Methods for Biologic
Dumont, D.J., Gutterman, J.U., Walker, C.L., et al. (2007). The energy sensing Morphometry, Volume 1, E.R. Weibel, ed. (London: Academic Press),
LKB1-AMPK pathway regulates p27(kip1) phosphorylation mediating the de- pp. 101–161.
cision to enter autophagy or apoptosis. Nat. Cell Biol. 9, 218–224. Wu, J., and Kaufman, R.J. (2006). From acute ER stress to physiological roles
Lum, J.J., Bauer, D.E., Kong, M., Harris, M.H., Li, C., Lindsten, T., and Thomp- of the Unfolded Protein Response. Cell Death Differ. 13, 374–384.
son, C.B. (2005). Growth factor regulation of autophagy and cell survival in the Yoon, K.H., Ko, S.H., Cho, J.H., Lee, J.M., Ahn, Y.B., Song, K.H., Yoo, S.J.,
absence of apoptosis. Cell 120, 237–248. Kang, M.I., Cha, B.Y., Lee, K.W., et al. (2003). Selective beta-cell loss and al-
Maechler, P., and Wollheim, C.B. (1999). Mitochondrial glutamate acts as pha-cell expansion in patients with type 2 diabetes mellitus in Korea. J. Clin.
a messenger in glucose-induced insulin exocytosis. Nature 402, 685–689. Endocrinol. Metab. 88, 2300–2308.
Maiuri, M.C., Zalckvar, E., Kimchi, A., and Kroemer, G. (2007). Self-eating and Yorimitsu, T., Nair, U., Yang, Z., and Klionsky, D.J. (2006). Endoplasmic retic-
self-killing: Crosstalk between autophagy and apoptosis. Nat. Rev. Mol. Cell ulum stress triggers autophagy. J. Biol. Chem. 281, 30299–30304.
Biol. 8, 741–752. Zhang, W., Feng, D., Li, Y., Iida, K., McGrath, B., and Cavener, D.R. (2006).
Marsh, B.J., Soden, C., Alarcon, C., Wicksteed, B.L., Yaekura, K., Costin, A.J., PERK EIF2AK3 control of pancreatic b cell differentiation and proliferation is
Morgan, G.P., and Rhodes, C.J. (2007). Regulated autophagy controls hor- required for postnatal glucose homeostasis. Cell Metab. 4, 491–497.

324 Cell Metabolism 8, 318–324, October 8, 2008 ª2008 Elsevier Inc.