PhD Program in Clinical and Translational Science
Human beta cell dysfunction induced by
different lipoglucotoxic conditions may be
transient or persistent and associates with
specific transcriptomic changes which are
shared in type 2 diabetes
PhD Program Director
Prof. Stefano Del Prato
Tutor:
Prof. Piero Marchetti
PhD Candidate:
Mara Suleiman
KEY WORDS
Pancreatic islets
Beta-cells
Lipotoxicity
Glucotoxicity
Lipoglucotoxicity
Type 2 diabetes
Beta-cell rescue
Insulin secretion
RNA sequencing
eQTL
SUMMARY
Pancreatic beta-cell failure is determined by the interplay of genetic and acquired
factors and represent the key event that leads to the development and progression of
type 2 diabetes (T2D). The cause of this derangement is not completely clarified. Work
in vivo shows evidence that alleviation of metabolic stress, through low calorie diet,
administration of glucose-lowering drugs or bariatric surgery, can improve beta-cell
function. Studies in vitro demonstrated that prolonged exposure to saturated fatty acids
(lipotoxicity), high glucose (glucotoxicity) or combinations thereof (lipoglucotoxicity)
may contribute to beta-cell failure, possibly via endoplasmic reticulum (ER) stress,
oxidative stress, loss of identity and/or other mechanisms. In addition, the molecular
mechanisms underlying the persistence or transience of human beta-cell defects are
still to be investigated to unveil to which extent the functional and molecular
modifications (and possible recovery) in non-diabetic (ND) islets reflect those of islets
from T2D individuals.
The present thesis aimed first to assess the direct impact of different lipoglucotoxic
treatments on human beta-cell function, and then to evaluate if the deleterious effects
were persistent or reversible after washout. Finally, the associated transcriptomic
changes were analyzed and compared with T2D islet gene expression signature. Islets
obtained from 26 ND organ donors were cultured in M199 medium, containing 5.5
mmol/l glucose, for 2 days (D2); then batches of islets were cultured for 2 additional
days (D4) either in absence (ctrl) or in the presence of: 0.5 palmitate (P), 11.1 mmol/l
glucose (g), 22.2 mmol/l glucose (G), 0.5 mmol/l palmitate + 11.1 mmol/l glucose
(P+g), 0.5 mmol/l palmitate + 22.2 mmol/l glucose (P+G), 1.0 mmol/l palmitate + oleate,
(1:2 molar ratio, P+O), 1.0 mmol/l palmitate + oleate + 11.1 mmol/l glucose (P+O+g),
and 1.0 mmol/l palmitate + oleate + 22.2 mmol/l glucose (P+O+G). At D4, islets were
washed and incubated with plain M199 medium for 4 additional days (D8 control and
D8 washout, according to the incubation condition). Furthermore, 28 ND and 58 T2D
islet preparations were studied. Glucose-stimulated insulin secretion (GSIS) from all
the preparations was assessed, and islets were also prepared for RNA extraction and,
in selected cases, for histology.
No significant change occurred in GSIS with ctrl, g, P+O and P+O+g islets throughout
the study period. However, GSIS at D4 declined (p<0.05 or less) with P, G, P+g, P+G
and P+O+G exposure. Normalization of GSIS was observed at D8 washout vs D4 with
P, G and P+g, but not with P+G and P+O+G. For P, G and P+G conditions, islet
transcriptome and genome features were analyzed by RNA-sequencing and eQTL,
respectively, to unveil the molecular mechanisms underlying beta-cell damage and its
reversal. For the conditions where islets were exposed to P (functionally impaired at
D4 and rescued at D8 washout), treated islets compared to control islets at D4 had
646 differentially (FDR<0.05) expressed genes (272 up- and 374 downregulated). Of
these, 595 were protein-coding genes (248 up- and 347 downregulated by palmitate),
including genes involved in lipid metabolism, inflammation and other cell functions.
Enrichment analysis identified several P-modified functional categories including
upregulation of unfolded protein response, acyl-CoA biosynthesis and fatty acid
metabolism and others. Comparison of D8 washout vs D4 palmitate-treated islets
identified 714 genes differentially expressed (167 up- and 547 down-). Of these, 656
were protein-coding (142 up- and 514 down-) comprising genes with a role in
lipid/glucose metabolism, transcription, inflammation, beta-cell function and others.
The Enrichment Map regarding D8 washed out islets showed downregulation of fatty
acid metabolism and upregulation of carbohydrate catabolic processes. As for human
islets exposed to G (impaired GSIS at D4 with recovery after washout), the comparison
of G-exposed islets vs control islets at D4 showed 50 differentially expressed genes
(38 up- and 12 down-); among these, 42 were protein-coding (32 up- and 10
downregulated by high glucose) including genes involved in metabolic pathways,
gated channel activity and other cell functions. Enrichment analysis showed inhibition
of cell junction and metabolic processes. D8 washout vs D4 G-treated islets RNA-seq
data comparison resulted in 341 differentially expressed genes (81 up- and 260
down-regulated). Of these, 320 were protein-coding (70 up- and 250 down-), also involved
in extracellular organization, establishment of protein localization to ER and
glycerophospholipid metabolism. The relative Enrichment Map showed
downregulation of response to wounding, myeloid cell differentiation and chemotaxis.
More profound changes in islet transcriptome were observed with combined P+G
(beta-cell dysfunction at D4 and persistence after washout). D4 P+G-exposed islets
resulted in differential expression of 1,498 genes (756 up- and 742 down-). Of these,
1,386 were protein-coding (699 up- and 687 down-) and they were mainly related to
transcription, inflammation, cell turnover, ion channels/transporters, mitochondrial
function, and redox balance. The Enrichment Map showed clustering of interrelated
gene-sets upregulated for most and comprising the unfolded protein response, protein
degradation, mRNA splicing regulation and ER stress-induced apoptosis. The
RNA-seq data of human islets at P+G D8 washout vs D4 treated identified only 322 genes
differentially expressed (120 up- and 202 down-). Of them, 292 were protein-coding
(102 up- and 190 down-).
Finally, the molecular changes associated with persistent or transient beta-cell insulin
secretion defects were correlated with those of human islets from T2D donors
compared with ND donors. The Rank-Rank Hypergeometric Overlap (RRHO)
approach was used, which allows to compare differentially expressed transcriptomes
between independent experiments. Overall, the RRHO analyses showed that
persistent or transient human beta-cell dysfunction induced by metabolic stress was
accompanied by specific gene expression signatures that were shared with T2D, with
the greatest concordant overlap between conditions that induce beta-cell dysfunction
and fail to recover after washout, namely P+G and P+O+G.
These results, obtained during the PhD course, show that certain lipoglucotoxic
conditions may induce persistent or reversible beta-cell dysfunction, depending on the
type, concentration and combination of the stressors. This associates with specific
molecular changes that overlap with T2D islet traits. Identification of novel mechanisms
responsible for human beta-cell functional deterioration and rescue, which are shared
in T2D, could provide novel insights into T2D pathogenesis and should foster the
development of improved beta-cell specific therapeutic approaches.
ABBREVIATIONS
ABS
Adult Bovine Serum
ATP
Adenosine triphosphate
cis-eQTL
Cis-Expression quantitative trait loci
Ctrl
Control
DDA
Data Dependent Acquisition
dTTP
Deoxythymidine triphosphate
dUTP
2´-Deoxyuridine, 5´-Triphosphate
D2
Day 2
D4
Day 4
D8
Day 8
DPP-IV
Dipeptidyl peptidase IV
ER
Endoplasmic reticulum
eQTL
Expression quantitative trait loci
FBS
Fetal bovine serum
FCccorr
Corrected fold change
FDR
False discovery rate\
FFA
Free-fatty acids
g
11.1 mmol/l glucose
G
22.2 mmol/l glucose
GSEA
Gene Set Enrichment Analysis
GSIS
Glucose-stimulated Insulin Secretion
GWAS
Genome-wide association study
HBSS
Hanks Balanced Salt Solution
ISI
Insulin Stimulation Index
KRBS
Krebs Ringer Bicarbonate Solution
KRBH
Krebs Ringer BicarbonateHepes
LC-MS/MS
Liquid chromatography–mass spectrometry
MOI
Multiplicity of infection
mRNA
messenger Ribonucleic acid
ND
Non-diabetic
NEFA
Non-esterified fatty acids
P
0.5 mmol/l palmitate
PBS
Phosphate Buffered Saline
PBS-T0.1
Phosphate Buffered Saline, Tween 0.1%
P+g
0.5 mmol/l palmitate + 11.1 mmol/l glucose
P+G
0.5 mmol/l palmitate + 22.2 mmol/l glucose
P+O
1.0 mmol/l palmitate + oleate, (1:2 molar ratio)
P+O+g
0.5 mmol/l palmitate + 11.1 mmol/l glucose
P+O+G
0.5 mmol/l palmitate + 22.2 mmol/l glucose
qRT-PCR
Quantitative Reverse Transcription PCR
RIN
RNA Integrity Number
RIPA
Radioimmunoprecipitation assay buffer
RNA-seq
RNA sequencing
RPKM
Reads Per Kilobase of exon per Million mapped reads
RRHO
Rank-Rank Hypergeometric Overlap
SDS-PAGE Sodium Dodecyl Sulphate - PolyAcrylamide Gel Electrophoresis
siRNA
Short interfering RNA
TABLE OF
CONTENTS
KEY WORDS
i
SUMMARY
ii
ABBREVIATIONS
v
INTRODUCTION
1
PRESENT
PROJECT
7
MATERIALS AND
METHODS
8
1. Experimental model
8
1a. Human pancreatic islets
8
1b. Cell lines
9
2. Functional studies
9
2a Culture and incubation media
9
2b Culture conditions
10
2c Glucose-stimulated insulin secretion
11
3. Immunocytochemistry and electron microscopy
11
4. RNA extraction
12
5. Library preparation and sequencing
13
6. RNA-seq data analysis
14
7. Evaluation of functional enrichment
15
8. Validation studies
15
8a Proteomics and proteomic data analysis
15
8b The role of ANKRD23
17
8c The role of Sec61G
18
9. Expression quantitative trait loci (eQTL)
18
RESULTS
20
1. Impact of different lipoglucotoxic conditions on
non-diabetic (ND) islet beta-cell function
20
2. Morphometric and ultrastructural assessments
22
3. RNA-seq data of palmitate-treated human islets
22
4. RNA-seq of high glucose-treated human islets
23
5. RNA-seq of palmitate plus high glucose-treated
human islets
24
6. Comparison of palmitate, high glucose and
palmitate plus high glucose exposure
25
7. Evidence for beta-cell dedifferentiation with
lipotoxicity
25
8. Validation studies
26
9. Association with T2D genome-wide association
loci
27
10. RNA-seq of T2D vs ND human islets
28
11. Comparison of lipoglucotoxic and T2D islet gene
expression
signatures
by
Rank-Rank
Hypergeometric Overlap (RRHO)
29
CONCLUSIONS
32
TABLES AND
FIGURES
37
REFERENCES
279
INTRODUCTION
Pancreatic beta-cell failure, due to the interplay of genetic and acquired factors, is key
to the onset and progression of diabetes. Beta-cells are unique endocrine cells that
synthetize, store and secrete insulin under the control of multiple and integrated
signals, thus tightly regulating blood glucose concentrations. They have a diameter of
10 μm on average, and each of them contains approximately 20 pg insulin. Beta-cells
are the most represented endocrine cell type in pancreatic islets, comprising 50-80%
of islet cells. Studies with autoptic samples, organ donor specimens and surgical cases
have found that beta cell mass in the human pancreas may vary from 0.6 to 2.1 g, with
beta cell volume and area (relative to the pancreatic tissue) ranging 1.1-2.6% and
0.6-1.6%, respectively. In normal individuals, beta-cells release about 30–70 units of
insulin per day (essentially depending on body weight, physical activity and nutritional
habits), half of which under basal condition and the remaining in response to meals.
The most important regulator of insulin release is glucose, that acts as both a trigger
as well as an amplifier of insulin secretion. Several other physiological molecules
regulate insulin secretion, including non-carbohydrate nutrients, hormones and
neurotransmitters.
In type 2 diabetes (T2D), the most common form of diabetes (representing
approximately 80-90% of all cases), beta-cell incompetence is due to both reduced
mass and functional impairment. Studies in-vitro and in-vivo have recently proposed
that dysfunction, rather than death, is the prevalent defect of the beta-cells in T2D.
Since the UK Prospective Diabetes Study [1], it has been assumed that the decline of
beta-cell functional mass begins before the onset of T2D and proceeds relentlessly
thereafter, leading to worsening of glycemic control and requiring progressive
intensification of diabetes therapy. The causes of this deterioration are not completely
understood, but prolonged exposure to saturated fatty acids (lipotoxicity), high glucose
(glucotoxicity) or combinations thereof (lipoglucotoxicity) have been proposed to
contribute to beta-cell failure both in vivo and ex vivo, probably via endoplasmic
reticulum (ER) stress, oxidative stress, loss of identity and other mechanisms [2–13].
Of interest, growing evidence shows that alleviation of metabolic stress can improve
beta-cell function and even induce remission of T2D. In the 1990’, a study was
performed aimed to investigate short- and long-term effects of lipid infusion on insulin
secretion. Twelve healthy individuals underwent a 24-h Intralipid (10% triglyceride
emulsion) infusion. After an overnight fast (baseline), at 6 and at 24 h of i.v. Intralipid
administration and 24 h after Intralipid discontinuation (recovery test), all subjects
underwent an intravenous glucose tolerance test. Intralipid infusion caused a threefold
increase of circulating plasma non-esterified fatty acid (NEFA) concentrations, with no
difference between the 6- and the 24-h time points. Compared to baseline acute insulin
response (AIR), short-term (6-h) Intralipid infusion was associated with a significant
increase of insulin secretion, whereas long-term (24-h) Intralipid administration
induced a significant reduction of AIR. The recovery test showed that after 24h from
Intralipid infusion discontinuation fasting plasma NEFA concentrations and AIR values
had returned to baseline values, demonstrating reversibility of beta-cell functional
alterations induced by the “lipotoxic” condition in vivo [14].
A few years later, insulin secretion was evaluated during a 4-day lipid infusion in normal
glucose tolerant individuals with or without a family history positive for T2D [15].
Thirteen and 8 subjects in the first and second group were studied, and received in
random order a lipid (Liposyn III, 20% triglyceride emulsion) or saline infusion. On days
1 and 2 insulin and C-peptide levels were measured after standardized mixed meals,
and a hyperglycemic clamp was performed on day 3. Similar concentrations of NEFA
were observed in the two groups during the study. In control subjects, the lipid infusion
determined a significant increase of insulin secretion after mixed meals and during the
hyperglycemic clamps. On the contrary, beta cell function decreased markedly in the
group with T2D family history, which affected both first and second phase insulin
release. These alterations became even more evident after correction for insulin
sensitivity. The results indicated therefore that subjects who are genetically at risk to
develop T2D, are particularly susceptible to the beta-cell functional damage induced
by increased plasma fatty acids (lipotoxicity), resulting in reduced insulin secretion in
response to mixed meals and intravenous glucose challenge. In a follow-up study, the
same group assessed the effects of pharmacological reduction of NEFA in 9
non-diabetic volunteers with a strong predisposition to T2D [16]. They used Acipimox, an
antilipolytic nicotinic acid derivative. Subjects were admitted twice and received in
random order either the drug or placebo for two days, in a double-blind design. In
addition to plasma glucose, insulin and C-peptide measurements during the days, a
hyperglycemic clamp was performed on day 3. Acipimox reduced 48-h plasma FFA by
about one third, which was associated with improved beta-cell function during the day.
First and, more evidently, second-phase insulin secretion during the hyperglycemic
clamp also improved, which was particularly remarkable after adjustment for the
prevailing insulin resistance. This study and its results provide further evidence for the
relevant role of lipotoxicity in the impairment of beta-cell function, at least in individuals
predisposed to T2D. More importantly, they demonstrate that beta-cell functional
damage can be rescued, provided the metabolic insult is attenuated.
If and how the possible recovery of beta-cell function observed in non-diabetic
individuals may apply to patients with overt T2D and impact on diabetes clinical
trajectory in still matter of debate. Remission of diabetes of variable duration can be
achieved in proportions of T2D subjects by carbohydrate restriction, low-calorie diets,
pharmacological treatments and bariatric surgery [17–22]. A few studies have
assessed if changes at the beta-cell level play a role in T2D remission following
low-calorie diets and bariatric surgery. In an early study [23] eleven subjects with T2D were
examined before and after 1, 4 and 8 weeks of a low-calorie (600 kcal)/day) diet. The
results showed that after 1 week of restricted energy intake fasting plasma glucose
normalized in the diabetic group. The first-phase insulin secretion increased during the
study period and approached control values. Maximal insulin response became
supranormal at 8 weeks. This was accompanied by significant decrease of pancreatic
triacylglycerol content. In a more recent article [24] detailed metabolic studies were
performed in subgroups of the Diabetes Remission Clinical Trial (Direct), in which
remission of T2D and persistence of non-diabetic blood glucose control were achieved
in 46% of patients treated with a low-calorie diet [17]. This specific study included 64
and 26 people respectively from the intervention and control groups). The authors
found that pancreatic fat content decreased whether or not glycemic values
normalized. However, recovery of first-phase insulin release identified those
individuals with remission of diabetes, which was durable at one year. These results
demonstrated that beta cell function can be rescued in a sizeable proportion of T2D
subjects, to achieve and sustain normoglycemia.
In the case of bariatric surgery, there is evidence that Roux-en-Y gastric bypass
(RYGB), vertical sleeve gastrectomy (VSG) and biliopancreatic diversion (BPD)
improve glucose control and can promote remission of T2D independently from weight
reduction [25,26]. The beneficial effects on glycemic indices may persist over the years
in several patients and become more evident as weight loss progresses [27–30].
Although bariatric surgery has distinct effects on several organs and functions,
post-surgical subjects with previous T2D show rapid and definite improvements of insulin
secretion, that persist over the time [31–33]. Restoration of first-phase insulin release
to intra-venous glucose administration (one to four weeks after RYGB or BPD) has
been consistently observed, occurs before any significant weight loss and reasonably
contributes to diabetes improvement or remission after the bariatric surgery procedure
[34–36]. Although the mechanisms determining better beta-cell function after bariatric
surgery are still largely unclear, increased GLP-1 release and potentiated incretin
effect are believed to play a major role [37,38].
In the past few years some information has been generated on the reversal of
beta-cell damage directly from isolated human islets, under different experimental
conditions. The use of human islets, prepared from the pancreas of organ donors,
allow assessment of beta-cell features independently from biological factors that
in-vivo could confound the scenario, and, importantly, can shed light on islet cell
morphological and molecular traits.
In an early study, [39] islets prepared from seven ND human pancreata were cultured
for 48h in normal or high glucose (16.7 mmol/l) containing medium. Islets were then
perifused with different glucose concentrations (3.3 and 16.7 mmol/l) or 10 mmol/l
L-arginine. It was seen that islets cultured in the high glucose medium lost acute
glucose-induced insulin release. However, insulin release in response to arginine was
preserved, supporting the concept of a selective loss of beta-cell sensitivity to glucose
stimulation induced by glucotoxicity. Notably, after additional 48h culture in normal (at
5.5 mmol/l glucose concentration) medium, the islets partially recovered from their
insulin secretion dysfunction. A few years later, these findings were confirmed and
implemented in a study in which ND islets were exposed to 33 mmol/l glucose for four
and nine days [40]. After these glucotoxic incubations, the authors observed a severe
decrease of insulin content (that was correlated with the reduction of insulin mRNA
and the transcriptional activity of PDX1) and the loss of glucose-stimulated insulin
release. Interestingly, most of these beta-cell alterations were partially reversible when
islets previously cultured for 6 days in high glucose were transferred to normal glucose
(5.5 mmol/l) concentrations for 3 additional days.
A few studies have assessed whether the defects of islets prepared from T2D donors
can be counteracted. The first data on this issue were generated with six T2D islet
preparations [41]. The authors found that, compared to ND islets, T2D islets showed
reduced insulin content, decreased amount of mature insulin granules, impaired
glucose-induced insulin secretion, reduced insulin mRNA expression, increased
apoptosis, together with higher expression of nitrotyrosine (a marker of oxidative
stress) and genes involved in the redox balance. Remarkably, the study showed that
24h incubation of T2D islets with a therapeutical concentration of metformin was
associated with increased insulin content, increased number and density of mature
insulin granules, improved glucose-induced insulin release, increased insulin mRNA
expression and reduced apoptosis. These effects were associated with decreased
oxidative stress, as suggested by lower levels of nitrotyrosine and reduced expression
of NADPH oxidase, catalase and GSH peroxidase after exposure to metformin.
On a similar line, some evidence has been provided to show that molecules of the
incretin system may have direct beneficial effects on T2D islets. In a study with islets
from seven T2D and eleven ND donors, the cells were exposed for 48h to 10 nmol/l
exendin-4, a DPP-4 resistant GLP-1 mimetic [42]. The authors observed that
incubation with exendin-4 caused a significantly better glucose-stimulated insulin
release from both T2D and ND islets, in comparison with untreated cells. In diabetic
islets this was associated with increased expression of a few key beta-cell genes,
including glucokinase, PDX-1, E2F1 and Cyclin D1.
A couple of years later the same group assessed the direct effect of GLP-1 and GIP
(alone or in combination) on isolated human T2D (four donors) and ND (eleven donors)
islets. Islets were exposed to incretins for 45 min (acute exposure: 0.1, 1.0, 10 or 100
nmol/l) or 2 days (prolonged exposure: 10 nmol/l), and functional as well as molecular
features were then assessed. After acute exposure (at concentrations from 1.0 to 100
nmol/l), both GLP-1 and, more markedly so, GIP, significantly improved
glucose-stimulated insulin release in ND islets, with no apparent synergic action. Similar effects
were observed with T2D islets acutely treated with 10 nmol/l GLP-1 or 100 nmol/l GIP.
Following prolonged exposure, improved insulin secretion was observed with T2D
islets cultured with GIP and GLP-1 in combination with GIP. Transcription of insulin,
PDX-1 and Bcl-2 tended to be higher after incretin culture in both T2D and ND islets,
with the combination of GLP-1 and GIP showing more significant effects.
Finally, more recent work [43] investigated the effects of compounds involved in the
modulation of autophagy (a process that leads to the degradation and recycling of
intracellular components) on beta-cell function, survival and ultrastructure. Islets
isolated from seventeen ND and nine T2D organ donors were cultured for 1–5 days
with 10 ng/ml rapamycin (an autophagy inducer), 5 mM 3-methyladenine (3-MA) or 1.0
nM concanamycin-A (autophagy blockers), either in the presence or not of metabolic
(0.5 mM palmitate) or chemical (0.1 ng/ml brefeldin A) endoplasmic reticulum (ER)
stressors. In ND islets glucose-stimulated insulin secretion was reduced by palmitate
and brefeldin; rapamycin prevented palmitate-, but not brefeldin-mediated cytotoxic
damage. Palmitate (and in similar way, brefeldin A) exposure increased beta-cell
apoptosis in ND islets, which was significantly prevented by rapamycin and worsened
by 3-MA. Both palmitate and brefeldin induced the expression of ER stress markers
(PERK, CHOP BiP) which was partially, but significantly prevented by rapamycin. In
T2D islets, the exposure to rapamycin improved insulin secretion, reduced the
proportion of beta-cells with signs of apoptosis and preserved insulin granules,
mitochondria and ER ultrastructure compared to untreated T2D islets; this was
associated with a significant reduction of PERK, CHOP and BiP gene expression.
These findings are in line with the concept that supporting autophagy can improve
human beta-cell function and survival under ER stress conditions and in the case of
T2D.
All these in-vivo and ex-vivo data support the concept that beta-cell dysfunction
induced by metabolic stress and in the case of T2D may be rescued, at least in part.
However, extensive assessments of these issues are lacking, and very little is known
on the associated molecular features.
In this PhD thesis a comprehensive study was performed with a large number of human
islet preparations to assess the direct impact of different lipoglucotoxic treatments on
beta-cell function and evaluated if the deleterious effects were persistent or reversible
after washout. The ex-vivo lipoglucotoxic treatments were selected to reflect
biologically relevant concentrations of the most common saturated and
mono-unsaturated fatty acid and glucose. For key conditions islet transcriptome and genome
were analyzed by RNA-sequencing and eQTL to examine the possibly involved
molecular mechanisms. Finally, the molecular changes associated with persistent or
transient beta-cell dysfunction were correlated with those of islets from T2D donors
compared with ND subjects. Mechanisms associated with human beta-cell functional
deterioration and rescue were identified, and their overlap with T2D islet traits unveiled,
to provide insights into T2D pathogenesis and foster the development of improved
β-cell targeted therapeutic strategies.
PRESENT PROJECT
With this project I present a comprehensive study (Figure 1) with a large series (n=26)
of ND human islet preparations (Table 1) in which the direct impact of several
lipoglucotoxic treatments on beta-cell function was assessed, aiming to evaluate if the
deleterious effects were persistent or reversible after the removal of the stressors
(washout). The ex vivo lipoglucotoxic treatments were selected to reflect biologically
relevant concentrations of the most common saturated and mono-unsaturated fatty
acid and glucose [44]. For specific treatments following which persistence of the
damage or recovery was observed, islet transcriptome and genome were analyzed by
RNA-sequencing and eQTL to shed light on the molecular mechanisms possibly
involved. In addition, the molecular changes associated with persistent or transient
beta-cell dysfunction were correlated with those found by comparing the transcriptomic
profile of islets from 28 T2D donors with 58 ND subjects (Tables 2a, 2b), to provide
insights into T2D pathogenesis and possibly, towards the development of therapeutic
strategies beta-cell specific.
MATERIALS AND METHODS
1. Experimental models
1a. Human pancreatic islets
Islets were prepared from the pancreas of 26 ND organ donors (16 females, 10 males,
age: 72±3 years, BMI: 24.6±0.7 Kg/m
2) [45,46] at the Pancreatic Islets Laboratory, with
permission by the Ethics Committee of the University of Pisa, upon written consent of
donors’ next-of-kin. The main clinical characteristics of each donor are provided in
Table 1 together with features of the processed pancreases and isolated islets. In a
separate set of experiments, islets were obtained from the pancreas of 28 T2D donors
(9 females and 19 males, age: 73.6±1.5 years, BMI: 26.3±0.7 Kg/m
2) and 58 ND (30
females and 28 males, age: 64.4±2.1 years, BMI: 25.3±0.5 Kg/m
2) (detailed
information is provided in Table 2a, 2b). Islets were isolated by enzymatic digestion
and density gradient purification; briefly, the gland was cleaned by careful dissection
of the surrounding fat tissue, duodenum portion, lymph nodes, vessels and spleen (if
present). After removing the head of the organ to access the Wirsung duct, body and
tail portion was used for islet preparation. The pancreatic duct was cannulated with a
18-gauge angio catheter and 200 ml of Hank’s Balanced Salt Solution (HBSS,
Sigma-Aldrich MO - USA), completed with 10% Adult bovine serum (ABS, Sigma-Sigma-Aldrich) and
3.6 mg/ml collagenase P (Roche Diagnostics, Switzerland) was injected into the
pancreatic duct until it completely distended the gland. Then, the pancreas was placed
into a sterile glass beaker in a water bath (37°C for 10 minutes). Successively, the
gland was gently hand-shaken at room temperature until free islets were observed in
a sample. At this point, the digestate was sequentially passed through 300-μm and
90-μm mesh stainless steel filters. The resulting suspension and the tissue trapped on the
300-μm mesh filter were placed back into the beaker with HBSS for further digestion.
The tissue remaining on the 90-μm filter was washed with HBSS supplement with 10%
of ABS. The same procedures of filtration, washing, and settling in the HBSS solution
was repeated at 5- to 8-minute intervals up to 40 minutes. For the purification, the
digested suspension was aliquoted in to 50-ml Falcon conical tubes and centrifuged at
1000 rpm for 2 minutes. The supernatant was discarded and the pellet suspended in
15 ml of a solution containing Lympholyte (Euro-Clone, Milano – Italy): HBSS (with
10% ABS) with different ratio (for example 80:20 or 70:30). Then 10 ml of HBSS was
layered over the Lympholyte-HBSS solution. The tubes were centrifuged at 1800 rpm
for 5 minutes at 4°C. Islets recovered from the interface between the two layers were
again centrifuged at 1800 rpm for 2 minutes, following which the supernatant was
discarded and the islet pellets were aliquoted into 75-cm
2Corning® cell culture flasks
(Sigma-Aldrich) containing complemented M199 medium (Euro-Clone); the medium
was refreshed after 24h culture. The islet yield was evaluated by staining a sample
with dithizone (DTH, Sigma-Aldrich). The purity of the islets was calculated as the ratio
between DTH-stained and total (stained and unstained) cell clusters [47].
1b. Cell lines
For validation studies, three different beta cell-lines were used. The EndoC-βH1 cell
line, a beta-cell line derived from human fetal pancreas, was cultured in DMEM
containing 5.6 mmol/l glucose, 2% BSA fraction V, 50 µmol/l 2-mercaptoethanol
(Sigma-Aldrich, Poole, UK), 10 mmol/l nicotinamide (Calbiochem, Darmstadt,
Germany), 5.5 µg/ml transferrin, 6.7 ng/ml selenite (Sigma-Aldrich), 100 units/ml
penicillin and 100 µg/ml streptomicin (Lonza, Leusden, The Netherlands). The same
medium, but with 2% FBS, was used after transfection with siRNAs. INS-1E beta-cells,
were cultured in RPMI 1640, containing 11 mmol/l glucose, 10% fetal bovine serum,
10 mmol/l HEPES, 50 μM 2-mercaptoethanol, and 1 mmol/l sodium pyruvate at 37°C
with 5% CO2 in a humidified atmosphere. MIN6B1 cells were cultured in 25 mmol/l
glucose DMEM supplemented by 15% fetal bovine serum, 2 mmol/l L-glutamine, 20
mmol/l HEPES, 50 μmol/l β-mercaptoethanol, plus penicillin (100 units/mL) and
streptomycin (0.1 mg/mL) at 37˚C in an atmosphere of humidified air (95%) and CO2
(5%).
2. Functional studies
2a. Culture and incubation media
The culture medium (M199 medium) was prepared from M199 medium (Euro-Clone)
containing 5.5 mmol/l glucose, supplemented with 0.1 mg/ml L-glutamine (Sigma
Aldrich), 10% ABS (Sigma-Aldrich), 100 UI/l penicillin (Sigma-Aldrich), 100 mg/l
streptomycin (Sigma-Aldrich), 750 μg/l amphotericin B (Sigma-Aldrich) and 50 mg/l
gentamycin (Sigma-Aldrich). pH was adjusted to 7.35-7.40 and the final solution was
filtered with a 0.2 Vacuum filter, PES (Sarstedt, Germany). The palmitate and the
oleate incubation media were obtained by adding the free-fatty acids (FFA) solutions
to M199 medium. FFA solutions were obtained by dissolving palmitate (sodium salt,
Sigma-Aldrich) and oleate (sodium salt, Sigma-Aldrich) in 90% volume/volume
ethanol, heated to 60°C and 1:100 diluted to a final concentration of 0.5 mmol/l.
Palmitate was used at a molar ratio of palmitate:bovine serum albumin of 3.3 [48]. The
11.1 mmol/l and 22.2 mmol/l glucose incubation media were prepared by dissolving
5.5 mmol (990 mg) and 16.7 mmol (3,006 mg) glucose to 1L M199 culture medium.
The Krebs Ringer Bicarbonate Solution (KRBS), used for insulin secretion studies was
prepared by combining equal volumes of 4 separate solutions:
Solution 1: KRBS containing 0.5 mol/l NaCl
Solution 2: KRBS containing 20 mmol/l KCl, 96 mM NaHCO
3, 4 mM MgCl
2•6H2O.
Solution 3: KRBS containing 10 mmol/l CaCl
2Solution 4: KRBS containing 2% Bovine Serum Albumin Fraction V (BSA, Roche
Diagnostics, Germany)
2b. Culture conditions
After the isolation, around 2000 ND islets were cultured in M199 medium [46,49] for 2
days (D2), to allow recovery from the isolation process. Then, based on previously
reported procedures [10,45,49], batches of islets were cultured for 2 additional days
(D4) under one of the following metabolically stressful conditions: 0.5 palmitate (P),
11.1 mmol/l glucose (g), 22.2 mmol/l glucose (G), 0.5 mmol/l palmitate + 11.1 mmol/l
glucose (P+g), 0.5 mmol/l palmitate + 22.2 mmol/l glucose (P+G), 1.0 mmol/l palmitate
+ oleate, (1:2 molar ratio, P+O), 1.0 mmol/l palmitate + oleate + 11.1 mmol/l glucose
(P+O+g), and 1.0 mmol/l palmitate + oleate + 22.2 mmol/l glucose (P+O+G). Islet
batches from the same donor were also cultured in normal M199 medium in parallel.
At D4, both control islets and treated islets were washed and incubated with M199
medium for 4 additional days (indicated respectively as D8 control and D8 washout)
(schematically represented in Figure 1A). The number of separate islet preparations
per condition ranged from 3 to 6 and replicates for each series ranged from 26 to 54,
for a total number of 277.
ND and T2D islets from separate experiments were cultured in normal M199 medium
for 2-3 days, before the experimental procedures, to allow recovery from the islet
preparation process (Figure 1B).
2c. Glucose-stimulated insulin secretion
Insulin secretion in response to acute glucose stimulation was assessed with
handpicked islets by batch incubation, and islet insulin content was measured after
acid-alcohol extraction [46,50]. Fifteen islets of similar size (approximate diameter 150
mm) were hand-picked and pre-incubated with KRBS containing 3.3 mmol/l glucose
(2 ml) at 37°C for 45 min, followed by a short wash and an incubation with 3.3 mmol/l
glucose (2 ml) at 37°C for 45 min (basal). The islets were then challenged with 16.7
mmol/l glucose (2 ml) at 37°C for additional 45 min (stimulated). After centrifuge step,
the respective supernatants were collected for insulin quantification; the bottom islet
pellets were incubated with an acid-alcohol solution (2 ml) at 4°C overnight for
subsequent insulin content measurement. Basal and stimulated insulin release, and
insulin content were quantified by a radioimmunometric assay (DIAsource
ImmunoAssays S.A., Nivelles, Belgium).
Insulin release was expressed as actual and normalized (to insulin content) values
respectively; insulin stimulation index was calculated as ratio of insulin release at 16.7
mmol/l glucose over release at 3.3 mmol/l glucose.
3. Immunocytochemistry and electron microscopy
Isolated islets, from selected samples, were studied by light microscopy
immunocytochemistry
[45]
and/or
electron
microscopy
[52-53].
For
immunocytochemistry, islet pellets were first 4% paraforlmaldehyde-fixed at room
temperature for 1 h, then washed with 1X-phosphate buffered saline (PBS,
Sigma-Aldrich) and resuspended in HistoGel
TM(Thermo Scientific, Waltham, MA, USA), a
liquid embedding medium used to encapsulate and retain small and friable specimens.
After the complete solidification of the gel containing the islets, the block was handled
according to tissue processing and paraffin-embedding [51]. Then, sequential 2 mm
sections were cutted from the gel block; hematoxylin-eosin staining was performed to
assess the section quality; sequential sections were stained by applying the following
primary antibodies: guinea pig anti-insulin antibody, (1:100, Invitrogen, Carlsbad, CA,
USA); polyclonal rabbit anti-human glucagon antibody, (1:3,000, Dako, Carpinteria,
CA, USA); mouse monoclonal anti-chromogranin A antibody, (1 μg/ml, Ventana, Oro
Valley, AZ, USA). Biotinylated secondary antibody, which reacts with mouse, rabbit,
guinea pig and rat primary antibodies, was purchased as Histostain-Plus kits
(Invitrogen). Sections were analyzed using Leica DM5500 B microscope (Leica,
Wetzlar, Germany) equipped with the DFC310 FX camera (Leica). Images were
acquired using Leica HCX PL FLUOTAR objective lenses at 40X magnification and
processed by the Leica MetaMorph® software, v1.8.0.
For electron microscopy, 30 hand-picked islet pellets were fixed with 2.5% (vol./vol.)
glutaraldehyde in 0.1 mmol/l cacodylate buffer, pH 7.4 for 1h at 4°C, and then postfixed
in 1% (vol./vol.) cacodylate-buffered osmium tetroxide for 2 h at room temperature.
Samples were dehydrated in a graded series of ethanol, transferred to propylene oxide
and embedded in Epon-Araldite. Ultrathin sections (60–80 nm thick) were cut with a
diamond knife, placed on formvar/carbon-coated copper grids (200 mesh), and stained
with uranyl acetate and lead citrate [52,53]. Apoptotic beta-cells were identified based
on the presence of marked chromatin condensation in the nucleus and/or blebs, as
described by Masini et al. 2009, Galluzzi et al. 2007 [52,54].
4. RNA extraction
For the lipoglucotoxic experiments, approximately 120 islets per condition were used.
They were handpicked, washed with sterile PBS (Sigma-Aldrich), centrifuged at 3,000
× g for 5 min to remove the washing buffer, snap frozen and then stored at -80 °C until
RNA extraction [46,55]. Total RNA was prepared using the RNeasy MINI Kit +
QIAshredder (Qiagen, Hilden, Germany); samples were lysed and homogenized in the
presence of a highly denaturing guanidine-thiocyanate-containing buffer which
immediately inactivates RNases. Ethanol was added to provide total RNA appropriate
binding conditions to the membrane of the speed column. Genomic DNA
contamination was removed by treating samples with DNase I (RNase-Free DNase
Set; Qiagen GmbH, Hilden, Germany); each sample was incubated with 80 ml of
solution containing 10 ml DNase I (27 units) + 70 ml buffer RDD at room temperature
for 15 min. Consecutive washing steps were finally performed, RNA was eluted in 30
ml water and concentration and integrity measurements were assessed.
RNA extraction from ND and T2D islets was performed using 100–120 islets, rinsed in
sterile PBS (Sigma-Aldrich) and centrifuged at 3,000 × g for 5 min to. The washing
PBS was disposed and 100 μl of extraction buffer (PicoPure RNA Isolation Kit; Life
Technologies, Foster City, CA) was added. Samples were incubated at 42°C for 30
min, centrifuged at 3,000 × g for 2 min, and the supernatant was collected and stored
at −80°C until RNA isolation. RNA isolation was performed according to the
manufacturer’s protocol. Briefly, 100 μl of 70% ethanol were added to the cell extract,
the mixture was added to the purification columns, washed, and subjected to DNase
treatment by incubation with 40 μl of DNase I solution (RNase-Free DNase Set, Qiagen
GmbH) for 15 min. Two additional washes were performed and RNA was eluted in 30
μl of elution buffer. RNA concentration (absorbance 260) and purity were evaluated
using the NanoDrop™ 2000c Spectrophotometer (Thermo Fisher Scientific) which
allowed the use of minimum volume-size, and provided information on purity utilizing
spectral output to detect possible protein (absorbance 280) or phenols (absorbance
230) contamination. RNA quality was assessed using Agilent Bioanalyzer 2100
(Agilent Technologies, Wokingham, UK) and Agilent RNA Nano Chips (Agilent
Technologies). The RNA integrity number (RIN), provided by the instrument, is based
on an algorithm which extracts the relationship between different characteristics of the
measurement. RIN values from the samples of the lipoglucotoxic series were: mean
8.9; SEM 0.1; median 8.8, range 7.9–9.7; the RIN values from ND and T2D samples
used for the T2D vs ND islet comparison had a RIN value of 7.5 and above. All the
samples had RIN values ensuring suitability for sequencing.
5. Library preparation and sequencing
Libraries were prepared using the TruSeq Stranded mRNA Library Prep Kit (Illumina,
San Diego, CA, USA). Briefly, mRNA was purified from 0.40 µg total RNA using
poly-T oligo attached magnetic beads. Following purification, mRNA was fragmented using
divalent cations and exposure at 94°C for 1 min, followed by 4°C hold temperature, in
order to obtain insert sizes ranging from 130 and 310 nucleotides in length. Cleaved
RNA fragments were copied into first strand cDNA using reverse transcriptase and
random primers. Second strand cDNA synthesis was performed by replacing dTTP
with dUTP in the second strand, using DNA Polymerase I and RNase H to generate
blunt-ended ds cDNA. The incorporation of dUTP quenches the second strand during
amplification. Adenylation of the 3’ ends of the blunt fragments and adaptor ligation
were performed to prepare the ds cDNA for hybridization onto a flow cell. Selective
enrichment of DNA fragments with adapter molecules on both ends and DNA
amplification of the first strand were obtained by PCR. The polymerase used does not
amplify the second strand incorporating dUTP. The generated libraries were quantified
using qPCR and quality was assessed using Agilent Bioanalyzer 2100 (Agilent
Technologies) and Agilent DNA 1000 chips (Agilent Technologies). All the libraries had
concentration and size (300 bp) suitable for sequencing which was performed on the
Illumina HiSeq 2500 instrument (Illumina, Inc. San Diego, California, U.S) for the
lipoglucotoxic conditions, and on an Illumina Genome Analyzer II (Illumina) for the T2D
and ND samples (Figure 2A). Paired-end sequencing (2x100bp) at 170 million reads
was performed.
6. RNA-seq data analysis
Sequenced reads from lipo/glucotoxicity experiments were mapped to the human
genome (assembly version GRCh37/hg19) using TopHat 2 (v2.0.13) [56] with default
parameters (Figure 2B, C). Reads were assigned to their corresponding exons and
genes based on GENCODE annotation v18 [57] using Flux Capacitor [58] with default
parameters. Resulting raw gene counts were normalized to gene length (sum of exons)
and sequencing depth, i.e., reads per kilobase per million (RPKM) mapped reads
(Figure 2D). The differentially expressed genes were identified by the R/Bioconductor
package EdgeR [59] using the raw counts with a false discovery rate (FDR) <0.05 as
cut-off for significance. The p-value for EdgeR differential analysis were corrected
according to the Benjamini and Hochberg 1995 method, as implemented by R p.adjust
function [60] (Figure 2E). For the comparison of differential expression of
lipo/glucotoxicity versus T2D, sequence reads were quantified with Salmon 13 [61]
using the transcriptome GRCh38 version 95 in quasi-mapping mode with sequence
and GC bias corrections and mapping validation. Differential analysis was done using
R DESeq2 package version 1.24.0 [62]. The enrichment map and gene list and plots
were generated by a modified two-tailed Rank-Rank Hypergeometric Overlap (RRHO)
method [63,64]. The logarithm of fold change was used to rank differentially expressed
genes. For the within-group comparisons over time, time-induced effects were first
evaluated by performing differential transcription analysis for the control conditions
between D4 vs D8, which reflects time-induced effects in the absence of a metabolic
insult. Next, the number of differentially expressed genes common to treatment and
control, and the magnitude of change over time were examined. Differentially
expressed genes with larger fold change of expression by treatment than control/time
were considered as “true” differentially expressed genes. The time effect was also
taken in to account in the RRHO analyses, through computing a corrected fold change
(FCcorr). From the fold change of D4 P±G (FCa) vs D4 control and D8 P±G vs D8
control (FCb), FCcorr is given by FCb/FCa. The RRHO software was modified to better
take in to account the multiplicity of minimal p-values, null p-values and the asymmetry
between the number of genes up- or down-regulated in the two datasets. The p-value
for RRHO was controlled for multiple testing with the Benjamini and Yekutieli method
(2001) as implemented by R p.adjust function [65]. The functional enrichment analysis
for the RRHO gene lists was generated with the gprofiler2 R package [66].
7. Evaluation of functional enrichment
Functional enrichment was generated using Gene Set Enrichment Analysis (GSEA)
softwar0e v3.0 [67]. Standard parameters were applied, except for minimum and
maximum size of functional category values that were adjusted to 5 and 500,
respectively. Enrichment maps of significantly modified biological processes (Gene
Ontology) were generated using the plugin “Enrichment Map” v3.1 [68] and visualized
within Cytoscape v3.6, using as geneset similarity cutoff a Jacard Index of 0.25 and a
FDR <0.05. To identify genes most likely to be relevant to the clusters of genesets
produced by Enrichment Mapping, the genes most frequently appearing across their
leading-edge analysis were identified [67] (Figure 2F). The leading-edge subset
represents the core of genes accounting for the gene set’s enrichment signal in GSEA.
Heatmaps show GSEA rank metric scores from these genes; higher absolute values
identify critical genes (Figure 2F).
8. Validation studies
8a. Proteomics and proteomic data analysis
Proteomics experiments [69] were accomplished with human islets to verify the protein
expression of some of the differentially expressed genes; functional as well as
molecular experiments were conducted to validate the role of genes associated with
changes in insulin secretion in the human islet incubation conditions described above.
For shotgun proteomics analysis, experiments were performed similar to what
previously described [69–72]. Briefly, islets were collected and washed twice with PBS
(37 °C). Cells were suspended in the buffer solution (7 M urea, 2 M thiourea, 4%
CHAPS, 60 mM dithiothreitol (DTT), 0.002% bromophenol blue) added with 50 mM
NaF, 2 mM Na3VO4, 1 μL/106 cells protease inhibitors, 1 µM trichostatin A, 10 mM
nicotinamide. After stirring and sonication (4 seconds, 5 times) cells were allowed to
rehydrate for 1 h at room temperature with occasional stirring. Thereafter, the solution
was centrifuged at 17,000 g for 5 min at room temperature. Protein concentration of
the resulting supernatant was determined using the Bio-Rad RC/DC-protein assay
(Bio-Rad). BSA was used as a standard. [69]. Aliquots of 40 µg of human islets protein
extracts were loaded onto 12% acrylamide resolving gel and subjected to
1D-electrophoresis. After protein visualization by Comassie blue, gel bands (16 bands for
lane) were excised, destained and subjected to in-gel reduction, alkylation and
overnight trypsin digestion at 37°C. The resulting peptides were analyzed in technical
triplicates by LC-MS/MS using a Proxeon EASY-nLCII (Thermo Fisher Scientific,
Milan, Italy) chromatographic system coupled to a Maxis HD UHR-TOF (Bruker
Daltonics GmbH, Bremen, Germany) mass spectrometer equipped with a nanoESI
spray source. Peptides were loaded on the EASY-Column C18 trapping column (2 cm
L., 100 µm I.D, 5 µm ps, Thermo Fisher Scientific), and subsequently separated on an
Acclaim PepMap100 C18 (75 µm I.D., 25 cm L, 5 µm ps, Thermo Fisher Scientific)
nano scale chromatographic column. The flow rate was set to 300 nL/min and the
gradient was from 3 to 35% of B in 80 min followed by 35 to 45% in 10 min and from
45 to 90% in 11 min. Mobile phase A was 0.1% formic acid in H2O and mobile phase
B was 0.1% formic acid in acetonitrile. The mass spectrometer was operated in positive
ion polarity and Auto MS/MS mode (Data Dependent Acquisition - DDA), using N2 as
collision gas for CID fragmentation. Precursors in the range 350 to 2,200 m/z
(excluding 1,220.0–1,224.5 m/z) with a preferred charge state +2 to +5 (excluding
singly charged ions) and absolute intensity above 4,706 counts were selected for
fragmentation in a maximum cycle time of 3 seconds. After acquiring one MS/MS
spectrum, the precursors were actively excluded from selection for 30 seconds.
Isolation width and collision energy for MS/MS fragmentation were set according to the
mass and charge state of the precursor ions (from 3 to 9 Da and from 21 eV to 55 eV).
In-source reference lock mass (1,221.9906 m/z) was acquired online throughout the
runs.
For bioinformatics processing, raw data were processed using PEAKS Studio v7.5
software (Bioinformatic Solutions Inc, Waterloo, Canada) using the ‘correct precursor
only’ option. The resulting mass lists were searched against nextprot database
(including isoforms as of June 2017; 42,151 entries). Carbamidomethylation of
cysteines was selected as fixed modification, oxidation of methionines and
deamidation of asparagine and glutamine were set as variable modifications.
Non-specific cleavage was allowed to one end of the peptides, with a maximum of 2 missed
cleavages. The highest error mass tolerances for precursors and fragments were set
at 10 ppm and 0.05 Da, respectively.
8b. The role of ANKRD23
To assess the effect of Ankrd23 overexpression on AMPK activation, the MIN6B1 cells
were plated on a 12-well plate, in duplicate for each condition and infected by an
adenovirus expressing either GFP (control) or mouse Ankrd23 (Vector Biolabs,
Malvern, PA, USA) at a multiplicity of infection (MOI) of 10 virus particle/cell. Two days
post-infection, the cells were preincubated for 1h in Krebs Ringer Bicarbonate Hepes
(KRBH) buffer (40 mM NaCl, 3.6 mM KCl, 0.5 mM NaH2PO4, 0.2 mM MgSO4, 1.5 mM
CaCl2, 10 mM HEPES, 25 mM NaHCO3), saturated with 95% O2/5% CO2 and
adjusted to pH 7.4 with the addition of 3 mM glucose before incubation for 30 min in
low (3 mM) or high (30 mM) glucose. The cells were then lysed in 100 μL
Radioimmunoprecipitation assay buffer (RIPA) containing protease (cOmplete™
Protease Inhibitor Cocktail, Roche) and phosphatase inhibitors (phosphatase inhibitor
cocktail 2, Sigma). Protein concentration was assessed using a Pierce BCA protein
assay kit (Thermo Fisher Scientific) according to manufacturer’s instructions. Proteins
were separated by SDS-PAGE under reducing conditions and transferred onto a
polyvinylidene fluoride membrane (0.45 μm, Amersham). The membrane was blocked
in 5% milk in PBS-T0.1 (Phosphate Buffer Saline, Tween 0.1%) for 1h and incubated
overnight at 4˚C with the primary antibodies anti-pAMPK (Abcam 2535S, dilution
1/1000), pRaptor (Abcam 2083S, dilution 1/1000), pACC (Abcam 3661S, dilution
1/1000), AMPK (Abcam 2532S, dilution 1/1000), Raptor (Abcam 2280S, dilution
1/1000), ACC (Abcam 3662S, dilution 1/1000), ANKRD23 (Abcam ab118210, dilution
1/500). The expression of α-tubulin (dilution 1/20000; T5168, Sigma) was used as a
loading control. After washing, the membrane was incubated for 1h in horseradish
peroxidase coupled secondary antibodies (Goat Anti-Rabbit-HRP, Abcam, AB6721,
dilution 1/8000) and the signal was detected on Amersham Hyperfilm ECL, after
exposure to Amersham ECL WB Detection Reagent. Band intensity was measured
using ImageJ.
8c. The role of Sec61G
The role of Sec61G was evaluated in INS-1E and EndoC-βH1 cells, cultured as
described [73–75] and further validated in dispersed human islets. Cells were
transfected overnight with 30 nM control siRNA (Qiagen) or siRNAs targeting rat or
human Sec61G as previously described [76]. Messenger RNA was isolated using the
Dynabeads mRNA DIRECT kit (Invitrogen, Paisley, UK) and reverse transcribed [77].
qRT-PCR was done using iQ SYBR Green Supermix (BIO-RAD, Nazareth Eke,
Belgium) on a LightCycler (Roche Diagnostics, Mannheim, Germany) or iCycler MyiQ
Single Color (BIORAD). Data were expressed as number of copies generated from the
standard curve method and corrected for the reference gene β-actin or
glyceraldehyde-3phosphate dehydrogenase (GAPDH). Primers used for qRT-PCR are
listed in the table below. Cells were exposed to 0.5 mM palmitate pre-complexed to
0.67% FFA-free BSA (16h for INS-1E cells, 24h for human islet cells) or to thapsigargin
for 24h. Apoptosis was assessed by fluorescence microscopy in propidium iodide (5
mg/ml) and Hoechst 33342 (5 mg/ml) stained cells as described [48]. For Western
blotting, INS-1E cells were lysed in Laemmli buffer and immunoblotted with cleaved
caspase 3 (Asp175, Cell Signaling Cat# 9661, dilution 1/1000), cleaved caspase 9
(Asp353, Cell Signaling Cat# 9507, 1/1000), and tubulin (Sigma Cat# T9026, 1/5000)
antibodies. Membranes were subsequently exposed to secondary
peroxidase-conjugated antibody (1/5000), and developed using SuperSignal West Femto
chemiluminescent substrate (Thermo Scientific) in a Bio-Rad chemi DocTM XRS +
(Bio-Rad laboratories).
Primers for rat and human GAPDH, ACTB, SEC61G
9. Expression quantitative trait loci (eQTL)
Specific details of the study were recently published [78]. Briefly, an eQTL analysis
was performed in a total of 100 islets from ND organ donor subjects. Genotyping was
performed using the 2.5 M Omniarray Beadchip and RNA analysis was performed
using Human Genome U133 Plus 2.0 Array to generate the eQTL analysis. Gender
and age were used as covariated and a cis-window of 500 kb was used. A nominal
p-value of p<0.05 was considered significant.
10. Statistical analysis
The results from the lipoglucotoxic conditions functional experiments are presented as
mean±SE and were analyzed with the R software or the GraphPad Prism Software
v8.4.2. Comparisons between two sets of data were performed by the two-tailed
Student’s t-test. Results from more than two groups were analyzed by ANOVA followed
by the Turkey correction. Multiple linear regression was used to take in to account the
effects of incubation time and culture conditions on multiple groups of islets. A p-value
< 0.05 was considered statistically significant.
Application Forward (5´-3´) Reverse (5´-3´) Species Length
GAPDH ST Standard curve ATGACTCTACCCACGGCAAG TGTGAGGGAGATGCTCAGTG Rat 930 bp
ACTB ST Standard curve AAATCTGGCACCACACCTTC CCGATCCACACGGAGTACTT Human 805 bp
SEC61G ST Standard curve TGGATCAGGTAATGCAGTTTG TAGGGATGTGGATCAGTTTCA Rat 180bp
SEC61G ST Standard curve TGGATCAGGTAATGCAGTTTG CAGCCACCAACAATGATGTT Human 205 bp
GAPDH RT Real time AGTTCAACGGCACAGTCAAG TACTCAGCACCAGCATCACC Rat 136 bp
ACTB RT Real time CTGTACGCCAACACAGTGCT GCTCAGGAGGAGCAATGATC Human 127 bp
SEC61G RT Real time AGT CGG CAG TTT GTA AAG GA GAA CCC CAT GAT AGC GAA T Rat 117 bp
