Articles in PresS. Am J Physiol Endocrinol Metab (February 24, 2015). doi:10.1152/ajpendo.00447.2014
1
1
Liver Protein Profiles in Insulin Receptor Knockout Mice Reveal Novel Molecules
2
Involved in the Diabetes Pathophysiology
3
4
Barbara Capuani1, David Della-Morte1,6, Giulia Donadel1, Sara Caratelli1, Luca Bova1,
5
Donatella Pastore1, Michele De Canio4,5, Simona D’Aguanno5, Andrea Coppola1,
6
Francesca Pacifici1, Roberto Arriga1, Alfonso Bellia1,3, Francesca Ferrelli1, Manfredi
7
Tesauro1,3, Massimo Federici1,3, Anna Neri2,3, Sergio Bernardini3,4, Paolo Sbraccia1,3,
8
Nicola Di Daniele1,3, Giuseppe Sconocchia7, Augusto Orlandi2, Andrea Urbani4,5, Davide
9
Lauro1,3.
10
11
1
12
Department of Biomedicine and Prevention University of Rome Tor Vergata Rome, Italy.
13
3
14
and Surgery, University of Rome Tor Vergata, Rome, Italy.5 Laboratory of Proteomics and
15
Metabonomics, S. Lucia Foundation—IRCCS, Rome, Italy. 6IRCCS San Raffaele Pisana,
16
Rome, Italy. 7Institute of Traslational Pharmacology, National Research Council Rome.
17
Corresponding author:
18
Davide Lauro, MD, Ph.D
19
Montpellier Street 1, 00133 Rome, Italy
20
Tel. +39-06-20904662 and +39-06-20904666
21
Fax. +39-0620904668
22
Email: [email protected]
23
Short title: Liver Protein Profiles in Diabetes
Department of Systems Medicine, University of Rome Tor Vergata, Rome, Italy.
2
Policlinico Tor Vergata Foundation, Rome, Italy. 4Department of Experimental Medicine
24
Copyright © 2015 by the American Physiological Society.
2
25
ABSTRACT
26
Liver has a principal role in glucose regulation and lipids homeostasis. It is under a
27
complex control by substrates such as hormones, nutrients and neuronal impulses. Insulin
28
promotes glycogen synthesis, lipogenesis and lipoprotein synthesis, and inhibits
29
gluconeogenesis, glycogenolysis, and VLDL secretion by modifying the expression and
30
enzymatic activity of specific molecules. To understand the pathophysiologic mechanisms
31
leading to metabolic-liver disease, we analyzed liver protein patterns expressed in a mice
32
model of diabetes by proteomic approaches. We used insulin receptor knockout (IR-/-) and
33
heterozygous (IR+/-) mice as a murine model of liver metabolic dysfunction associated with
34
diabetic ketoacidosis and insulin resistance. We evaluated liver fatty acid levels by
35
microscopic examination and protein expression profiles by orthogonal experimental
36
strategies using protein 2-DE MALDI-TOF/TOF and peptic nLC-MS/MS shotgun
37
profiling. Identified proteins were then loaded into Ingenuity Pathways Analysis to find
38
possible molecular networks. Twenty-eight proteins identified by 2-DE analysis and 24
39
identified by nLC-MS/MS shotgun, were differentially expressed among the 3 genotypes.
40
Bioinformatic analysis revealed a central role of High Mobility Group Box 1/2 and
41
huntigtin never reported before in the association with metabolic and related liver disease.
42
A different modulation of these proteins both in blood and hepatic tissue further suggests
43
their role in these processes. These results provide new insight into pathophysiology of
44
insulin resistance and hepatic steatosis, and could be useful in identifying novel biomarkers
45
to predict risk for diabetes and its complications.
46
47
48
49
50
Key Words: Insulin Resistance, Huntigtin, HMGB1, Proteomics.
3
51
INTRODUCTION
52
Type 2 Diabetes (T2D) is a complex metabolic disorder characterized by increased level of
53
insulin resistance and impaired beta cell function with reduced insulin secretion (35). In
54
T2D, modification of hepatic metabolism is associated with glucose and lipids
55
overproduction, leading to overt hyperglycemia and diabetic dyslipidemia. Glucose
56
overproduction is a physiological response in T2D, while the overproduction of lipids
57
linked with insulin resistance remains a phenomenon to clarify since insulin increases
58
abundance of lipogenic enzymes, and insulin resistance diminishes lipogenesis (41) (31).
59
Furthermore, insulin resistance and T2D are linked with hepatic steatosis, which is an
60
exacerbation of liver dysfunction generated by accumulation of lipids, mainly triglycerides
61
(23).
62
Several etiological mechanisms have been proposed to explain the pathologic link between
63
insulin resistance, T2D and liver disease, including inflammation and oxidative stress (36).
64
Recently, novel anti-inflammatory proteins called chaperones, and toll-like receptors
65
(TLRs) have been suggested to play a significant role in this pathological process (27).
66
However, to date the molecular mechanisms linking the impairment in glucose
67
homeostasis with liver disease are not fully understood.
68
Therefore, elucidation of molecular alterations, which regulate metabolic liver dysfunction
69
in insulin resistance conditions, is essential to increase our knowledge in the field and to
70
develop new therapeutic approaches for T2D. With this final objective, among the different
71
experimental approaches, proteomic technique, for its specific characteristics, may be
72
helpful in identifying new liver tissue peptides and proteins involved in this pathological
73
loop (43). In the present study by using two different orthogonal proteomic approaches and
74
protein ontologies pathway analysis, we aimed to further understand molecular features
75
mechanisms underlying the pathophysiology of liver disease triggered by high levels of
76
insulin resistance (insulin receptor heterozygous (IR+/-) mice) and diabetic ketoacidosis
77
conditions (insulin receptor knockout (IR-/-) mice). We also aimed to investigate lipid and
4
78
inflammatory patterns linked with insulin resistance to further understand processes
79
underlay liver degeneration in diabetic patients.
5
80
MATERIALS AND METHODS
81
Animals
82
Mutant mice bred on C57BL/6J used for this study have been kindly donated by Prof.
83
D. Accili, Columbia University College of Physicians and Surgeons, New York, USA. All
84
experimental procedures were approved by the European animal welfare authorities and
85
performed in accordance with the institutional animal care guidelines.
86
Mice generation and genotyping have been clearly described in previous publication
87
(22) (Supplementary materials). Mice employed for experiments were knockout (IR-/-),
88
heterozygous (IR+/-), and wild type (wt) (IR+/+) for the IR gene. A sample of n=5 of
89
animals per group was used in each experiment.
90
91
Ultrastructural Histological study
92
Liver tissue were extracted and placed in 10% buffered formalin, dehydrated, and
93
embedded in paraffin sections or frozen and embedded in OCT cryostatic sections. The
94
percentage of steatotic cells and fatty liver area were determined by Haematoxylin-Eosin
95
stained, blue-toluidin (TB) semi-thin and in Oil Red O solution-stained- sections. HMGB1,
96
CD3, CD19 and F4/80 protein expression was performed by immunohistochemistry (32),
97
on 4-μm-thick paraffin sections using antibodies from Abcam. The positive
98
immunostaining was assessed using a semiquantitative scale modified as follows: 3,
99
strong; 2, moderate; 1, weak; 0, absent (30). Measurements were performed by 2 different
100
investigators in at least 10 fields at X200 magnification, for each case, with an inter
101
observer variability less than 5%.
102
103
Liver samples preparation
104
Proteins were extracted from a pool of 5 different livers having the same genotype by
105
Griding Kit (GE Healthcare), in lysis buffer containing 8M Urea, 2% Chaps, 10mM NaF,
6
106
0.5mM DTT, 1mM Na3VO4, 0.5mM PMSF, 2µg/µl leupeptin, 2µg/µl aprotinin, 2µg/µl
107
pepstatin. Protein concentration was determined by Thricloraoacetic Acid (TCA) method.
108
109
ELISA assay
110
Blood samples withdrawn by vein tail from 5 to 8 mice for each genotype analysis were
111
tested to determine circulating HMGB1, (Shino Test), Insulin (Mercodia) and C-peptide
112
(Mercodia) by ELISA kits, used according to the manifacturer’s instructions (42).
113
114
qRT PCR
115
Total RNA was extracted from liver tissue by Trizol (Life Technologies). cDNA was
116
synthetized with a high-capacity cDNA archive kit according to manifacture’s instructions.
117
Fifty nanograms of cDNA was amplified by real-time polymerase chain reaction; cDNA
118
expression was analyzed by relative 2-
119
18S. The primers used were: Cd3d Mm00442746_m1;Emr1 Mm00802529_m1;Cd19 Mm
120
00515420_m1.
∆∆ CT
method, using as endogen control ribosomal
121
122
Western blotting analysis and 2D gel electrophoresis
123
Hepatic lysates (100 µg/sample) were mixed with Laemmli buffer 5x, run on 10% SDS-
124
PAGE bis-tris polyacrylamide gels and transferred to nitrocellulose membrane (Protoran;
125
Schleicher&Schuell). Blots were probes with the following polyclonal antisera: HMGB1,
126
HMGB2, Sin 3A, HTT, PSMA5, PDIA3, NF-kB (from Abcam). One hundred ug liver
127
protein lysates were separated in the first dimension on IPG gels (pH 3–10 non linear GE
128
Healthcare) (8). The 2D separation was performed on 4-20% SDS polyacrylamide gels.
129
Next, gels were stained with silver using a protocol compatible to mass spectrometry
130
analysis, and scanned to analyze through Image Master Platinum 5.0 software (GE
131
Healthcare). Spot detection and normalization were performed by the automated tools of
132
the software. To reduce the number of possible “false positives," protein relative profile
7
133
was set to 500 ppm. We considered only spots with a p<0.05 according to the Student’s t
134
test and with relative standard deviation (SD) less than 30%. The resulting list of
135
modulated spots (28 spots) was screened considering only spots with a ratio above 1.3
136
between 2 conditions.
137
138
Mass spectrometry analysis (MALDI TOF/TOF)
139
Protein spots were excised from gels and digested with trypsin to obtain peptides which
140
were separated in a MALDI TOF/TOF. Mass spectra were acquired with an Ultraflex III
141
MALDI TOF/TOF spectrometer (BrukerDaltonics). After removing contaminant ions from
142
the peak list, database search was conducted by MASCOT 2.2.06 algorithm
143
(www.matrixscience.com) interrogating the NCBInr_20100116 database restricted to the
144
Mus musculus taxonomy (144908sequences). MALDI-TOF MS/MS analysis was
145
performed in LIFT mode; chosen ions were selected manually and analyzed by Flex
146
Analysis 3.0 software (11).
147
148
Quantitative proteomics by nLC-MS/MS
149
One hundred μg of proteins from each sample were combined in order to create 3 different
150
pools representative of the 3 distinct murine genotypes. Reduction and alkylation of
151
proteins were obtained by adding 100mM DTT (1h at 37°C) and 200mM iodoacetamide
152
(1h at R.T). Protein samples, at final concentration of 2μg/μL, were digested with 1:20
153
(w/w) sequence grade porcine trypsin (Promega) at 37°C overnight, and loaded onto a
154
Proxeon Easy-nLC II (Thermo Scientific) chromatographic system coupled to a Q-TOF
155
mass spectrometer (micrOTOF-Q II, BrukerDaltonics) for protein identification and
156
quantification (10).
157
158
159
8
160
Functional annotation and Pathway analysis
161
Expressed proteins identified by the 2 different proteomic approaches were characterized
162
in more detail by using various bioinformatics tools, including the “Panther classification
163
system” (www.pantherdb.org) for functional annotation of biological process and the
164
“Ingenuity Pathway Analysis” (IPA) (www.ingenuity.com) for network analysis (9).
165
166
Data and Statistical Analysis
167
Two-D gel electrophoresis analysis was performed by Image Master Platinum software
168
(GE Healthcare), version 5.0. Statistical analysis was performed by GraphPad Prism 5 (La
169
Jolla, CA, USA). Statistical evaluation of the data was performed using ANOVA test,
170
followed by Bonferroni’s post hoc test. Differences were considered statistical significant
171
at p<0.05. All data are expressed as mean of standard error (SEM).
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
9
187
188
Abbreviations: ACOX 1 Peroxisomal Acyl-Coenzyme A Oxidase 1; CBP CREB Binding
189
Protein; DAMP Damage-Associated Molecular Pattern; DM Diabetes Mellitus; ER
190
Endoplasmic Reticulum; E2A E Protein Family of Trascription Factors; GLUT4 Glucose
191
Transporter
192
Hydroxymethylglutaryl-CoA Synthase 2; HMGB1/2 High Mobility Group Box 1/2; HTT
193
Huntingtin; IPA Ingenuity Pathway Analysis; IPG Immobilized pH Gradient; IR Insulin
194
Receptor; MALDI TOF TOF Matrix- Assisted Laser Desorption/Ionization Time Of
195
Flight; MIB mitochondrial inter-membrane space bridging; NAFLD Non Alcoholic Fatty
196
Liver Disease; nLC-MS/MS liquid chromatography tandem mass spectrometry; PDIA 3
197
Protein Disulfide-Isomerase A3; PDX1 Pancreatic and duodenal homeobox 1; PSD-95
198
Post Synaptic Density Protein-95; PSMA5 Proteasome Subunit Alpha Type; Sin3 A Paired
199
Amphipathic Helix Protein Sin3a; TLR Toll Like Receptor; T2D Type 2 Diabetes;
200
YWHAB 14-3-3 protein beta/alpha; 2-DE 2 Dimensional Electrophoresis.
201
4;
HD
Huntington
Disease;
HFD
High
Fat
Diet;
HMCS2
10
202
RESULTS
203
204
IR-/- mice present different metabolic and inflammatory patterns
205
Hepatic insulin resistance is associated with nonalcoholic fatty liver disease (NAFLD) and
206
is among the major factors in the pathogenesis of T2D. Hepatic insulin resistance is caused
207
by defect in intracellular insulin signalling and inflammation, activation of the endoplasmic
208
reticulum (ER) stress pathways, and accumulation of hepatocellular lipids which can
209
decrease hepatic insulin sensitivity. To evaluate fatty acid levels in the liver, we performed
210
intracellular staining using Oil Red dye in each different group of mice (data not shown).
211
Since IR-/- mice died in less than a week (mean=3±1 days) (1), we carried out these set of
212
experiments in newborn animals livers. Microscopic examination revealed in all livers
213
presence of abundant hematopoietic and progenitor cells, according to results reported in a
214
previous study (18). Differences in microscopic appearance of hepatic tissue between
215
genotypic distinct mice are reported in Fig. 1 (A). We observed an increase by 50-fold of
216
liver fatty area in IR-/- mice compared to IR+/+ mice (p<0.00001; Fig.1 (B)). Liver steatosis
217
was also 10-fold higher in the heterozygous genotype (p<0.005) than IR+/+ group,
218
suggesting a pivotal role of IR in liver disease associated with the glucose homeostasis
219
alteration.
220
We also detected an increase in the percentage of hepatocytes with microvescicular and
221
macrovescicular steatosis in the heterozygous genotype compared to IR+/+ hepatocytes
222
(p<0.005), and more markedly in IR-/- genotype compared with IR+/- and IR+/+ mice
223
(p<0.00001, and p<0.005, respectively). Differences in liver steatosis area between the
224
three groups were also confirmed by means of Oil Red-O staining of cryostatic sections
225
(data not shown).
226
Moreover, since Hribal et al. (20) showed a negative correlation between IGF1 sera
227
levels and steatosis grade in humans we analyzed whether there was a correlation between
228
IGF1 mRNA expression and steatosis in our models. In agreement with previous studies
11
229
we observed a significant decreasing of IGF1 mRNA in mice lacking of IR compared with
230
the other two groups of mice (p<0.01) (Fig. 1 (C)).
231
Since in the IR-/- mice has been demonstrated a profound inflammatory component in
232
the liver (28) we also investigated the difference in the inflammation state in our mice
233
models by characterizing the activation of specific immune cells implicated in this
234
mechanism. Firstly, we found increased levels of Nuclear Factor- Kappa –light chain-
235
enhancer of activated B cells, NF-kB protein expression in IR-/- compared with IR+/+ and
236
IR+/- (p<0.01 and p<0.001) (Fig. 2 (A)), which is an important transcription factor linking
237
inflammation to metabolic diseases, such as T2D (19) (38). Then we showed a higher
238
activation of macrophages and B cells, typical mediator of inflammation, in IR-/- compared
239
to other genotypes of mice, by measuring surface markers CD19 (Fig. 2 (C,F)), and F4/80
240
(Fig. 2 (D,G)) (p<0.05 and p<0.005 respectively) by qRT-PCR and immunostaining
241
analysis.
242
243
Hyperinsulinemia in IR-/-mice did not involve insulin clearance
244
To further provide a biochemical characterization of the mice employed in this study we
245
evaluated serum levels of insulin and C-peptide (Fig. 3 (A) (B)). We observed that,
246
accordingly with previous results from Accili D. et al (1), IR-/- mice had ~20 folds higher
247
levels of insulin compared with IR+/+ and IR+/- mice (p<0.01). Similarly, IR-/- mice showed
248
~10 folds higher levels of C-peptide compared with IR+/+ and IR+/- mice. We also estimated
249
insulin clearance amount, calculated by the ratio of the serum levels of C-peptide and
250
insulin (28). Nevertheless, IR-/- mice were hyperinsulinemic, with a non significant
251
reduction of insulin clearance (Fig. 3 (C)), suggesting that the absence of IR lead to a lower
252
muscle glucose uptake, hyperglycemia and subsequent increase levels of insulin secretion.
253
254
255
12
256
Liver of IR-/- Mice Exhibits Altered Protein Expression Profiles
257
To better characterize different liver protein expressions between IR+/+, IR+/- and IR-/- mice
258
we performed proteomic analysis by using 2D electrophoresis coupled with mass
259
spectrometry MALDI–TOF/TOF technique and shotgun approach. Two dimensional gel
260
maps showed a total of (mean ± SEM) 451±52, 547±12, and 635±28 protein spots, in
261
IR+/+, IR+/- and IR-/-, respectively (Fig. 4 (A), (B) and (C)). Comparative analysis revealed
262
28 significantly modulated protein spots across IR-/-/IR+/+, IR+/-/IR+/+ and IR-/-/IR+/- with a
263
ratio>1.5 (p≤0.05) according to statistic tests (see Supplementary Material).Spot
264
identifications are shown in Table 1. In IR-/- vs. IR+/+ seven spots were up-regulated, while
265
10 were down-regulated. In IR-/-/IR+/- ratio 15 proteins were down-regulated while in IR+/-/
266
IR+/+ ratio only 4 proteins were up-regulated.
267
To further investigate proteome characterization we also performed a peptide-centric
268
shotgun analysis by loading tryptic peptides of hepatic protein lysate on nLC-MS/MS. A
269
total of 127 proteins were identified between different mice genotypes. Among them, 24
270
proteins were significantly modulated considering the 3 comparisons previously described
271
(Table 2) with a regulation ratio >1.3 or <0.7 and % coefficient variation (CV) of
272
regulation ratio<30%. Seventeen proteins were up-regulated in IR+/+ vs. IR-/- while only
273
one Hydroxymethylglutaryl-CoA synthase (HMCS2) was up-regulated in IR-/-. IR-/-/IR+/-
274
ratio showed 17 proteins up-regulated in IR+/- mice and 2 in IR+/-/IR+/+ ratio, although IR+/+
275
revealed only 1 protein up-regulated in IR+/-/IR+/+ ratio (Acyl Coenzime A oxidase 1,
276
ACOX1).
277
The preponderance of the identified proteins spots belonging to 8 different functional
278
patterns based on their Gene Ontology annotations which explain their molecular function.
279
These proteins can be involved in specific biological processes and/or belong to structural
280
cellular component:
281
1. Lipid metabolism (RBP2, APOA1, FABP5, ACOX1, HMCS2, Vimentin, Regucalcin);
282
2. Oxidative stress (GSTM1, PRDX1, GSTA3, PARK7, ATPB); 3. Protein synthesis and
13
283
degradation (IF5A1, EIF1A1, PSMA5, SIN3A); 4. Chaperone and endoplasmic reticulum
284
(ER) stress (HSP7C, PDIA3, PDIA1, GRP78, PPIA, Calmodulin); 5. Cellular amino acids
285
biosynthetic process (NDKA, NDKB, ASSY, ARGI1, CPSM, BHTM1, FAA); 6. Glucose
286
metabolism (LDHA, ALDOB, ENOB, G3P, F16P1, 6PGL, ACON); 7. Blood circulation
287
(HBE, HBA, HBB, TRFE, FETA, ALB); 8. Cellular component and morphogenesis
288
(TBA1B). In Fig. 5 is reported the impact of each protein category overall, particularly we
289
analyzed the up- (A) and down-regulated (B) proteins for each different IR genotype. In
290
IR-/- phenotype we observed up-regulation of proteins involved in lipid and glucose
291
metabolism, oxidative stress and protein synthesis and degradation while none of the
292
proteins belonging to amino acid biosynthetic process, blood circulation, chaperone and
293
ER stress, cellular component and morphogenesis were expressed.
294
295
Novel Pathway is linking Diabetes Mellitus with Liver Steatosis
296
Identified proteins by proteomic studies were were used to interrogate IPA software to
297
explore possible biological interactions among them. IPA database search returned 3 main
298
ranked networks. As expected, we observed that the most accurate network (higher score)
299
resulted from the link between IR and proteins implicated in mechanisms associated with
300
glucose homeostasis (Fig. 6). Among this analysis, the most interesting association was
301
found between IR, HMGB1/2, and (Huntingtin) HTT. The two orthogonal proteomic
302
analyses of IR-/- mice liver tissue, in agreement with evidences of the hepatic steatosis
303
occurrence in these animals (21), suggested a central role of lipid metabolism (data not
304
shown), and oxidative stress (data not shown) in this process.
305
306
Differential expression of HMGBs and HTT in liver from IR-/- mice
307
Since HMGBs and HTT had a central role in our protein network generated by pathway
308
analysis we further explored their expressions in mice liver by immunoblotting and
309
immunohistochemical analyses. Western Blot of HMGB1 showed 1.5-fold increase levels
14
310
in IR-/- vs. either IR+/+, and IR+/- mice (p <0,01) (Fig. 7 (A)). Likewise, HMGB2 was up-
311
regulated in IR-/- and IR+/- vs. IR+/+ mice (Fig. 7 (B)). Higher presence of HMGBs proteins
312
in IR-/- mice further support the hypothesis of their role as a modulator of the inflammatory
313
response during the development of multifactorial diseases, such as diabetes (29).
314
Interesting, we found that HTT was 2-fold lower expressed in IR-/- vs. IR+/+ mice (p<0,05),
315
and 1.5-fold in IR-/- vs. IR+/- mice (p<0,01) (Fig. 7 (C)). Lower levels of HTT in IR-/- mice
316
suggest its role in regulating glucose homeostasis and liver disease as reported by recent in
317
vivo study (3). Immunohistochemical analysis confirms results from WB for HMGB1 and
318
HTT proteins (Fig. 8 (A), and Fig. 8 (B)). To further validate results from WB for HMGBs
319
and HTT proteins we also perform the same analysis for other proteins identified by
320
proteomic analysis. The results of WB were in agreement with data from proteomic
321
analysis for Paired Amphipathic Helix Protein Sin3a (SIN3A), Proteasome Macropain
322
subunit alpha type (PSMA5), and Protein disulfide isomerase family A member 3 (PDIA3)
323
(Fig. 7 (D), Fig. 7 (E) and Fig. 7 (F)). Since another important site of insulin resistance,
324
relevant for development of T2D and its complications, is the skeletal muscle we aimed to
325
confirm present results in this organ. After western blot analysis of specific proteins
326
we found modulated in the liver of IR+/+, IR+/- and IR-/- mice, we observed that expression
327
levels of HTT, HMGB1, HMGB2 and PSMA5 had similar trend of expression in liver and
328
muscle, while PDIA3 and SIN3A presented a pattern variations, suggesting a tissue
329
modulation response, which deserve further investigation (data not shown).
that
330
331
Higher level of HMGB1 in IR-/- mice sera
332
Since HMGB1 is the only protein among those found in the principal network which is
333
secreted in the serum (25), we evaluated its level in sera across each animal genotype by
334
using ELISA assay. We detected that sera HMGB1 was 1.4 -fold increase in IR-/- vs. IR+/+
335
(p<0.05), and 2.6-fold vs. IR+/- (p<0.05) (Fig. 9).
15
336
DISCUSSION
337
Hepatic insulin actions are regulated principally at transcriptional levels by blocking
338
gluconeogenesis and activating lipogenesis (26). Liver insulin receptor knockout mice
339
showed fasting and postprandial hyperglycemia associated with hyperinsulinemia and
340
hepatic and skeletal muscle insulin resistance (5). These findings are then suggesting that
341
hepatic insulin resistance could be the first step in the development of peripheral insulin
342
resistance. In the present study, we used IR+/- and IR-/- mice to analyze liver dysfunction
343
and protein expressions either in conditions of diabetic ketoacidosis and higher levels of
344
insulin resistance. We first detected microvescicular and macrovescicular steatosis in both
345
IR+/- and IR-/- mice, and by using two different proteomic approaches we identified altered
346
protein expression profiles in IR+/- and IR-/- mice compared to littermate wt. These proteins
347
mainly belong to the classical biological patterns linked to organ injury such as oxidative
348
stress, lipid metabolism, and glucose homeostasis. We further confirmed these associations
349
by ontological network analysis where we found a central role of HTT protein in animals
350
lacking IR. An interesting role in this network was also found for HMGB1 and 2 (Fig. 6).
351
Increase of HMGB1, and a significant decrease of HTT protein expressions were found in
352
liver and sera sample from IR+/- and IR-/- mice compared to wt, further suggesting their
353
involvement in metabolic-liver disorders.
354
Several models of genetically modified animals have been previously employed to
355
investigate mechanisms underlay liver disease in association with diabetes and insulin
356
resistance.. However, at the best of our knowledge, we are the first showing in IR+/- and IR-
357
/-
358
quantification.. Impairment in these regulatory mechanisms in the liver could explain, at
359
least in part, the anomalous hepatic accumulation of lipids, as previously reported (40).
360
Mainly, the present data are confirmatory of the evidence that insulin resistance is an
361
important and early factor in the pathogenesis of dyslipidemia and steatosis in subjects
mice a liver structural alteration linked with metabolic disorder by a specific lipid
16
362
prone to developing metabolic syndrome and T2D. Furthermore, we identified novel
363
proteins implicated in the association between insulin resistance and liver disease using
364
two different orthogonal proteomic strategies. Proteomic investigations revealed different
365
functional classes of proteins, with a different expressions between IR-/-, IR+/-, and IR+/+
366
mice (Fig. 5). These proteins mainly belong to the classical patterns implicated in the liver
367
disease such as lipid and glucose metabolism, and oxidative stress. Considerably IR-/- mice
368
showed an elevated number of up-regulated proteins belonging to the classes of protein
369
synthesis and degradation, and oxidative stress pathway. Particularly, among those
370
proteins, 30% were involved in the oxidative stress pathways. It is worth to highlight that
371
several of identified peptides (ATP synthase, calmodulin, Albumin NADPH- flavin
372
reductase, Hemoglobin subunits, peroxiredoxins), have been already associated with
373
metabolic diseases and with different forms of diabetes, such as T1D (15). These data
374
confirm that maintaining a stable redox state, which underlay the oxidative stress damage,
375
is pivotal to counteract metabolic alterations, and therefore to prevent metabolic diseases
376
and their micro and macrovascular complications (14), (13).
377
To further increase the significance of the present study, we performed bioinformatics
378
analysis to explore biological interactions between the different expressed proteins. We
379
showed a central role of HTT protein in the network with a higher score associated with
380
glucose homeostasis (Fig. 7). The HTT protein is required for human development and
381
healthy brain function (33). HTT is found in many of the body's tissues; however, the
382
complete understandings of its function have to be clearly established. It is subject to
383
posttranslational modification, and some events, such as phosphorylation, can play an
384
enormous role in regulating HTT function. Mainly, it has been involved in cellular
385
signalling, transporting materials, binding proteins, and apoptosis. Mutation in the IT-15
386
gene, that expands abnormally the number of CAG nucleotide repeats, results in a mutated
387
HTT, which contains expansions of glutamines (polyQ) that make it prone to aggregate,
388
leading to a neurological disorder called Huntington Disease (HD) (39). The result of this
17
389
study are in line with a new field of research linking T2D, insulin resistance and related-
390
liver diseases with cognitive impairment and neurodegeneration (12). Insulin resistance
391
can be a pathogenic factors in HD since it represents a metabolic stressor which may
392
induce neurodegeneration (34). It has been shown that HD patients despite having a low
393
body mass index tend to develop insulin resistance and increased risk for T2D (25).
394
However, human studies have explored the association between HD and metabolic
395
disorders including T2D, with controversial results (24). Recently, increasing PolyQ length
396
repeats has been shown to decrease insulin secretion and glucose responsiveness, but only
397
in early stage of HD onset (4). A strong link between HTT and insulin resistance has been
398
proposed by a study which demonstrated as pancreatic islets from HD transgenic mice
399
express reduced levels of the insulin, somatostatin, and glucagon and exhibit intrinsic
400
defects in insulin production, suggesting a role of HTT protein in all these processes (6). It
401
is also known that T2D is common in other triplet repeat disorders including Friedreich’s
402
ataxia and myotonic dystrophy (16). In both of these diseases, the diabetic phenotype is
403
associated with hyperinsulinemia and insulin resistance. Since the function of HTT is not
404
well understood, we only know that when mutated the protein induced insulin resistance by
405
affecting pathways involving expression of key regulators of insulin gene transcription,
406
including the pancreatic homeoprotein PDX-1, E2A proteins, and the coactivators CBP
407
and p300 (2). However, to the best of our knowledge, we are the first reporting that non-
408
mutated form of HTT is involved in pathways regulating glucose homeostasis, further
409
supporting its role in metabolic control mechanisms.
410
Moreover, results from immunoblotting and immunohistochemical analyses, showing a
411
significant decrease of HTT in IR-/- compared to IR+/- and IR+/+ mice liver, further
412
corroborating the HTT involvement in metabolic/liver disease. We may speculate that HTT
413
when mutated, or dysfunctional is implicated in mechanisms leading to insulin resistance
414
and dyslipidaemia, such as in HD patients (37). In a similar fashion, lower levels of HTT
415
(lower activity) may be associated to impairment of glucose homeostasis and higher
18
416
predisposition of liver disease in diabetic patients. Additional studies are imperative to
417
fully understand this hypothesis.
418
On the contrary, we found higher levels of HMGB1 in IR-/- mice either in the
419
cytoplasm of hepatocytes than in blood sera. HMGB1 and HMGB2, along with HTT, were
420
present in the glucose homeostasis pathway resulted from networking analysis. This
421
finding is not extremely surprising since HMGB1 and 2, as damage-associated molecular
422
pattern (DAMP) molecules, are late mediators of noninfectious systemic inflammation.
423
Recently, a study conducted in humans demonstrated as the intracellular distribution of
424
HMGB1 is modified in a state of insulin resistance, and as HMGB1 is a stimulatory factor
425
of β-pancreatic cells insulin secretion, supporting a role of inflammation regulators in
426
glucose homeostasis (17). Moreover, a strong association between HMGB1 and liver
427
disease has been previously reported (7) even if the role of HMGB1 in the association
428
between metabolic distress and liver disease are not fully understood.
429
Strengths of this study include: 1. Use of well characterized animal model comparing wt,
430
knockout and heterozygotes mice. 2. The employment of two different proteomic
431
techniques and validation of the results by immunoblotting analysis. 3. The utilization of
432
sophisticated bioinformatics programs to perform functional and networking analysis.
433
Limitations to acknowledge for this study are mainly associated with the not-controllable
434
bias typical of the techniques used. For instance, the lack in the evaluation of
435
compensatory mechanisms characteristic of knockout mice, which may reflect on different
436
protein patterns after proteomic analysis. Another important limitation to acknowledge for
437
this study is the absence of the mechanistic data, which is typical charateristic of
438
proteomics studies, and that reduces, at least in part, the strenght of the hypothesys.
439
However IPA allows a better speculation of the results that must be supported by further
440
experiments.
441
In conclusion, in the present study we confirmed as defect of liver insulin action is strictly
442
linked with the development of liver disease. In addition, we found novel molecules and
19
443
pathways associated with impairment in glucose homeostasis such as HTT protein and
444
HMGB1. Altered levels of these proteins were also found either in the liver and blood in a
445
state of insulin resistance and DM. Further studies are imperative to fully understand the
446
mechanisms beyond their role/link in DM and liver diseases, and their possible application
447
as biomarkers to predict in patients the risk for DM and its complications.
448
449
450
451
Grant Support: Research Project 2009 grant, Fondazione Roma; PRIN 2010 and 2011
452
grants from the Ministero dell'Istruzione, dell' Università e della Ricerca (D.L. and P.S.);
453
Fondazione Umberto Di Mario; 2010 Grant from Associazione Italiana per la Ricerca sul
454
Cancro; AIRC grant (G.S.):IGI0555; The Ministry of Education, University and Research
455
(PRIN), grant 2010AX2JX7_005 (G.S.); ASI N 2013-084-R0 COREA Research Project
456
Italian Space Agency
457
Disclosure: The authors declare no conflict of interest
458
Acknowledgements We thank Prof Porzio O.for data analysis;
459
460
461
462
463
464
465
20
466
REFERENCES
467
1.
468
Taylor SI, and Westphal H. Early neonatal death in mice homozygous for a null allele of the
469
insulin receptor gene. Nature genetics 12: 106-109, 1996.
470
2.
471
Ferrante RJ, Habener JF, Beal MF, and Thomas MK. Huntington's disease of the endocrine
472
pancreas: insulin deficiency and diabetes mellitus due to impaired insulin gene expression.
473
Neurobiology of disease 11: 410-424, 2002.
474
3.
475
Li JY, Sundler F, Brundin P, and Mulder H. The R6/2 transgenic mouse model of Huntington's
476
disease develops diabetes due to deficient beta-cell mass and exocytosis. Human molecular genetics
477
14: 565-574, 2005.
478
4.
479
Norremolle A, Nielsen JE, and Hasholt L. Huntington's disease does not appear to increase the
480
risk of diabetes mellitus. Journal of neuroendocrinology 21: 770-776, 2009.
481
5.
482
L. Severe impairment in liver insulin signaling fails to alter hepatic insulin action in conscious
483
mice. The Journal of clinical investigation 115: 1306-1313, 2005.
484
6.
485
XJ. Loss of huntingtin-associated protein 1 impairs insulin secretion from pancreatic beta-cells.
486
Cellular and molecular life sciences : CMLS 69: 1305-1317, 2012.
487
7.
488
mobility group box 1 (HMGB1) in liver diseases. Molecular medicine 19: 357-366, 2013.
489
8.
490
Volpe S, D'Agostino D, Todaro M, Stassi G, Di Ilio C, Urbani A, and De Laurenzi V. p63
491
isoforms regulate metabolism of cancer stem cells. Journal of proteome research 13: 2120-2136,
492
2014.
Accili D, Drago J, Lee EJ, Johnson MD, Cool MH, Salvatore P, Asico LD, Jose PA,
Andreassen OA, Dedeoglu A, Stanojevic V, Hughes DB, Browne SE, Leech CA,
Bjorkqvist M, Fex M, Renstrom E, Wierup N, Petersen A, Gil J, Bacos K, Popovic N,
Boesgaard TW, Nielsen TT, Josefsen K, Hansen T, Jorgensen T, Pedersen O,
Buettner C, Patel R, Muse ED, Bhanot S, Monia BP, McKay R, Obici S, and Rossetti
Cape A, Chen X, Wang CE, O'Neill A, Lin YF, He J, Xu XS, Yi H, Li H, Li S, and Li
Chen R, Hou W, Zhang Q, Kang R, Fan XG, and Tang D. Emerging role of high-
D'Aguanno S, Barcaroli D, Rossi C, Zucchelli M, Ciavardelli D, Cortese C, De Cola A,
21
493
9.
494
A. Shotgun proteomics and network analysis of neuroblastoma cell lines treated with curcumin.
495
Molecular bioSystems 8: 1068-1077, 2012.
496
10.
497
M, Bernardini S, Federici G, and Urbani A. New insights into neuroblastoma cisplatin
498
resistance: a comparative proteomic and meta-mining investigation. Journal of proteome research
499
10: 416-428, 2011.
500
11.
501
Longone P, Spalloni A, De Laurenzi V, Bernardini S, Federici G, and Urbani A. Protein
502
repertoire impact of Ubiquitin-Proteasome System impairment: insight into the protective role of
503
beta-estradiol. Journal of proteomics 75: 1440-1453, 2012.
504
12.
505
inhibitor of the insulin/IGF signaling pathway protects from age-onset, neurodegeneration-linked
506
proteotoxicity. Aging cell 13: 165-174, 2014.
507
13.
508
and its role in diabetes mellitus related cardiovascular diseases. Current pharmaceutical design 19:
509
5695-5703, 2013.
510
14.
511
G. The role of oxidative stress in the pathogenesis of type 2 diabetes mellitus micro- and
512
macrovascular complications: avenues for a mechanistic-based therapeutic approach. Current
513
diabetes reviews 7: 313-324, 2011.
514
15.
515
C, Lauro D, Tripathy D, Jenkinson C, Paroni R, Orsenigo E, Cighetti G, Gregorini L,
516
Staudacher C, Secchi A, Bachi A, Brownlee M, and Fiorina P. Proteomics reveals novel
517
oxidative and glycolytic mechanisms in type 1 diabetic patients' skin which are normalized by
518
kidney-pancreas transplantation. PloS one 5: e9923, 2010.
D'Aguanno S, D'Agnano I, De Canio M, Rossi C, Bernardini S, Federici G, and Urbani
D'Aguanno S, D'Alessandro A, Pieroni L, Roveri A, Zaccarin M, Marzano V, De Canio
D'Alessandro A, D'Aguanno S, Cencioni MT, Pieroni L, Diamantini A, Battistini L,
El-Ami T, Moll L, Carvalhal Marques F, Volovik Y, Reuveni H, and Cohen E. A novel
Fiorentino TV, Prioletta A, Zuo P, and Folli F. Hyperglycemia-induced oxidative stress
Folli F, Corradi D, Fanti P, Davalli A, Paez A, Giaccari A, Perego C, and Muscogiuri
Folli F, Guzzi V, Perego L, Coletta DK, Finzi G, Placidi C, La Rosa S, Capella C, Socci
22
519
16.
520
Friedreich ataxia (FA) associated with diabetes mellitus type 1 and hyperthrophic cardiomyopathy.
521
Bosnian journal of basic medical sciences / Udruzenje basicnih mediciniskih znanosti =
522
Association of Basic Medical Sciences 9: 107-110, 2009.
523
17.
524
Navarro S, Giralt M, Garcia-Rios A, Cobo-Padilla D, Tinahones FJ, Lopez-Miranda J,
525
Villarroya F, Fruhbeck G, Fernandez-Real JM, and Malagon MM. Alarmin high-mobility
526
group B1 (HMGB1) is regulated in human adipocytes in insulin resistance and influences insulin
527
secretion in beta-cells. International journal of obesity 2014.
528
18.
529
hematopoietic stem cells in late fetal and newborn mice: models for human umbilical cord blood.
530
Blood 90: 174-181, 1997.
531
19.
Hotamisligil GS. Inflammation and metabolic disorders. Nature 444: 860-867, 2006.
532
20.
Hribal ML, Procopio T, Petta S, Sciacqua A, Grimaudo S, Pipitone RM, Perticone F,
533
and Sesti G. Insulin-like growth factor-I, inflammatory proteins, and fibrosis in subjects with
534
nonalcoholic fatty liver disease. The Journal of clinical endocrinology and metabolism 98: E304-
535
308, 2013.
536
21.
537
disorders of insulin receptor-deficient mice by transgenic overexpression of glucokinase in the liver.
538
Diabetologia 45: 1292-1297, 2002.
539
22.
540
D. Targeted disruption of the insulin receptor gene in the mouse results in neonatal lethality. The
541
EMBO journal 15: 1542-1547, 1996.
542
23.
543
Clinical and molecular hepatology 19: 210-215, 2013.
Gucev Z, Tasic V, Jancevska A, Popjordanova N, Koceva S, Kuturec M, and Sabolic V.
Guzman-Ruiz R, Ortega F, Rodriguez A, Vazquez-Martinez R, Diaz-Ruiz A, Garcia-
Harrison DE, and Astle CM. Short- and long-term multilineage repopulating
Jackerott M, Baudry A, Bucchini D, Jami J, and Joshi RL. Improved metabolic
Joshi RL, Lamothe B, Cordonnier N, Mesbah K, Monthioux E, Jami J, and Bucchini
Koo SH. Nonalcoholic fatty liver disease: molecular mechanisms for the hepatic steatosis.
23
544
24.
545
Lukic L, and Kostic VS. Glucose homeostasis in Huntington disease: abnormalities in insulin
546
sensitivity and early-phase insulin secretion. Archives of neurology 65: 476-480, 2008.
547
25.
548
significance of serum HMGB1 in colorectal carcinomas. PloS one 7: e34318, 2012.
549
26.
550
mTORC1 required for stimulation of lipogenesis, but not inhibition of gluconeogenesis.
551
Proceedings of the National Academy of Sciences of the United States of America 107: 3441-3446,
552
2010.
553
27.
554
Famulla S, Roden M, Eckel J, and Habich C. Heat shock protein 60 as a mediator of adipose
555
tissue inflammation and insulin resistance. Diabetes 61: 615-625, 2012.
556
28.
557
Kahn CR. Loss of insulin signaling in hepatocytes leads to severe insulin resistance and
558
progressive hepatic dysfunction. Molecular cell 6: 87-97, 2000.
559
29.
560
RAGE: a functional tripod that leads to diabetic inflammation. Expert opinion on therapeutic
561
targets 15: 1023-1035, 2011.
562
30.
563
Spagnoli LG, and Czernobilsky B. Cellular retinol binding protein-1 expression in endometrial
564
hyperplasia and carcinoma: diagnostic and possible therapeutic implications. Modern pathology :
565
an official journal of the United States and Canadian Academy of Pathology, Inc 19: 797-803,
566
2006.
567
31.
568
hepatic insulin resistance and type 2 diabetes. Nature 510: 84-91, 2014.
569
32.
570
Ferrara FF, Guglielmi MB, Carminati P, Dallavalle S, Morini G, Merlini L, Orlandi A, and
Lalic NM, Maric J, Svetel M, Jotic A, Stefanova E, Lalic K, Dragasevic N, Milicic T,
Lee H, Song M, Shin N, Shin CH, Min BS, Kim HS, Yoo JS, and Kim H. Diagnostic
Li S, Brown MS, and Goldstein JL. Bifurcation of insulin signaling pathway in rat liver:
Marker T, Sell H, Zillessen P, Glode A, Kriebel J, Ouwens DM, Pattyn P, Ruige J,
Michael MD, Kulkarni RN, Postic C, Previs SF, Shulman GI, Magnuson MA, and
Nogueira-Machado JA, Volpe CM, Veloso CA, and Chaves MM. HMGB1, TLR and
Orlandi A, Ferlosio A, Ciucci A, Francesconi A, Lifschitz-Mercer B, Gabbiani G,
Perry RJ, Samuel VT, Petersen KF, and Shulman GI. The role of hepatic lipids in
Pisano C, De Cesare M, Beretta GL, Zuco V, Pratesi G, Penco S, Vesci L, Fodera R,
24
571
Zunino F. Preclinical profile of antitumor activity of a novel hydrophilic camptothecin, ST1968.
572
Molecular cancer therapeutics 7: 2051-2059, 2008.
573
33.
574
Mikus CR, Laye MJ, Laughlin MH, Booth FW, and Ibdah JA. Mitochondrial dysfunction
575
precedes insulin resistance and hepatic steatosis and contributes to the natural history of non-
576
alcoholic fatty liver disease in an obese rodent model. Journal of hepatology 52: 727-736, 2010.
577
34.
578
molecular medicine 82: 510-529, 2004.
579
35.
580
metabolism. Nature 414: 799-806, 2001.
581
36.
582
missing links. Cell 148: 852-871, 2012.
583
37.
584
current state of research with peripheral tissues. Experimental neurology 219: 385-397, 2009.
585
38.
586
insulin resistance. Endocrinology and metabolism clinics of North America 35: 525-549, viii, 2006.
587
39.
588
Huntington's disease. Drug discovery today 2014.
589
40.
590
Brockmann GA. Fat storage capacity in growth-selected and control mouse lines is associated with
591
line-specific gene expression and plasma hormone levels. International journal of obesity and
592
related metabolic disorders : journal of the International Association for the Study of Obesity 23:
593
586-594, 1999.
594
41.
595
perspective. Current opinion in pharmacology 10: 684-691, 2010.
Rector RS, Thyfault JP, Uptergrove GM, Morris EM, Naples SP, Borengasser SJ,
Ristow M. Neurodegenerative disorders associated with diabetes mellitus. Journal of
Saltiel AR, and Kahn CR. Insulin signalling and the regulation of glucose and lipid
Samuel VT, and Shulman GI. Mechanisms for insulin resistance: common threads and
Sassone J, Colciago C, Cislaghi G, Silani V, and Ciammola A. Huntington's disease: the
Schwartz EA, and Reaven PD. Molecular and signaling mechanisms of atherosclerosis in
Sepers MD, and Raymond LA. Mechanisms of synaptic dysfunction and excitotoxicity in
Timtchenko D, Kratzsch J, Sauerwein H, Wegner J, Souffrant WB, Schwerin M, and
Wong RH, and Sul HS. Insulin signaling in fatty acid and fat synthesis: a transcriptional
25
596
42.
597
protein 1 (HMGB1) quantified by ELISA with a monoclonal antibody that does not cross-react with
598
HMGB2. Clinical chemistry 49: 1535-1537, 2003.
599
43.
600
and Hua ZC. Role of Fas-associated death domain-containing protein (FADD) phosphorylation in
601
regulating glucose homeostasis: from proteomic discovery to physiological validation. Molecular &
602
cellular proteomics : MCP 12: 2689-2700, 2013.
Yamada S, Inoue K, Yakabe K, Imaizumi H, and Maruyama I. High mobility group
Yao C, Zhuang H, Du P, Cheng W, Yang B, Guan S, Hu Y, Zhu D, Christine M, Shi L,
603
604
605
FIGURE LEGENDS
606
Fig. 1. Histological evaluation of hepatic samples in IR+/+ (+/+), IR+/- (+/-) and IR-/- (-/-
607
) mice. (A) Representative images of Haematoxylin&Eosin-stained (H&E) liver sections
608
showing a slight lipid microvesiculation of hepatocytes in IR+/- mice, whereas evident
609
macrovesciculation is also present in IR-/- mice (left and the center of the picture,
610
respectively). Same results were obtained in toluidine blue-stained (TB) sections (right of
611
the picture). (B) Bar graphs showing the percentages of positive hepatocytes with
612
microvesiculation (left) and macrovesiculation (centre); n=10, *p<0.05 (IR+/- vs. IR+/+). On
613
the right, the percentage of liver area occupied by fat; n=10, *p<0.05 (IR+/- vs IR+/+).
614
(C) qRT-PCR showed a dramatic decreased levels of IGF1 mRNA expression in IR+/- and
615
IR-/- mice; n=5 *p<0.05 (IR+/+ vs. IR+/-, IR+/+ vs. IR-/- ).
616
617
Fig. 2. Inflammation activation. A) NF-kB expression analyzed by Western Blot.
618
Protein extracts were obtained from 5 animals and bar graph were used to analyze the
619
expression. Actin has been used for protein normalization. Results are reported as
620
means±SEM. *p<0.05; **p<0.01; ***p <0.005.
621
mRNA expression of immune cell marker: B) CD3 (lymphocytes T); C) CD19 (B cells);
qRT-PCR was performed to detect
26
622
D) F4/80 (Macrophages). mRNAwas extracted from 5 mice and bar graph was used to
623
represent the relative expression normalized with 18S RNA, as endogenous control. A.U
624
indicates arbitrary units. Statistical analysis was reported as mean±SEM, by 1 way Anova,
625
*p<0.05; **p<0.01; ***p <0.005.
626
627
Fig. 3. Sera levels of Insulin and C peptide. A) B) Sera quantification of Insulin and
628
Cpeptide among the 3 different genotype performed by ELISA kits. C) Insulin Clearance
629
calculated by ratio between sera Insulin and Cpeptide. Data are expressed as means ±
630
SEM; *p<0.01.
631
632
Fig. 4. 2D Gel electrophoresis of liver proteins. Representative protein maps of hepatic
633
lysate extracted from (A) IR+/+, (B) IR+/-, and (C) IR-/- mice. Arrows and numbered spots
634
refer to proteins with different expression profiles between the three different genotypes
635
identified by MALDI TOF spectrometer. Spot numbers correspond to proteins listed in
636
Table 1.
637
638
Fig. 5. Functional classification of different class of proteins identified by proteomic
639
analysis. Fiftytwo identified proteins were categorized into eight class of pathways, and
640
the percentage contribution of each category was calculated for each phenotype. (A) The
641
overall percentage of up- regulated proteins in the three phenotypes; (B) The percentage of
642
down-regulated proteins in the three phenotypes.
643
644
Fig. 6. Ingenuity Pathway Analysis. Proteins networking analysis of differentially
645
expressed molecules in three mice models, was obtained by Ingenuity Pathway analysis.
646
Selected network has a higher score. Proteins colored in red and green refer to Tables 1 and
647
2.
648
27
649
Fig. 7. Validation of selected liver proteins by western blot analysis. (A) HMGB1; (B)
650
HMGB2; (C) Huntingtin (HTT); (D) Sin 3A, (E) PSMA 5, (F) PDIA3. Protein extracts
651
were obtained from 5 animals and bar graph were used to analyze the expression. Actin has
652
been used for protein normalization. Results are reported as means±SEM. *p<0.01;
653
**p<0.05; ***p <0.005.
654
655
Fig 8. Immunostaining analysis. Immunostaining of 4-μm-thick paraffin liver sections
656
from IR+/+, IR+/-, and IR-/- newborn mice. The expression of HMGB1 (A) was detected by
657
using a rabbit polyclonal anti-HMGB1 antibody while HTT
658
Huntingtin antibody.
(B) were stained using
659
660
Fig 9. HMGB1 ELISA assay. Chart reporting levels of HMGB1 sera quantification among the 3
661
different genotype. Sera levels of HMGB1 tested in mice IR+/+, IR+/-e IR-/- by ELISA assay. Data
662
are expressed as means ± SEM; *p<0.01.
663
664
665
666
667
668
669
670
671
672
673
674
Table 1. Identification of the differential 2DE spots by MALDI-TOF-TOF*.
Spot
no.
ratio
ratio
ratio
KO/WT KO/ET ET/WT Entry Name
#
Protein Description
GO Biological Process
§
1
-
-
0.5
HSP7C_MOUSE
Heat Shock Cognate 71 kDa Protein
immune system process, protein folding, protein
2
-
-
4.4
ACON_MOUSE
Aconitate Hydratase
complex assembly, response to stress
tricarboxylic acid cycle, carbohydrate metabolic process
3
-
-
0.5
PDIA1_MOUSE
Protein Disulfide-Isomerase
protein modification process
4
0.6
-
0.3
PDIA3_MOUSE
Protein Disulfide-Isomerase A3
protein modification process
5
-
-
2.2
ALDOB_MOUSE
Fructose-Bisphosphate Aldolase B
glycolysis
6
-
-
2.1
Q3TUI9_MOUSE
Proteasome Subunit Alpha Type
proteolysis
7
-
1.6
-
BLVRB_MOUSE
Flavin Reductase (NADPH)
-
8
-
2.2
-
IF5A1_MOUSE
translation
9
-
0.6
-
NDKA_MOUSE
Eukaryotic Translation Initiation
Factor 5A-1
Nucleoside Diphosphate Kinase A
10
-
0.5
-
PPIA_MOUSE
Peptidyl-Prolyl cis-trans Isomerase A immune system process, intracellular protein transport,
nuclear transport, protein folding
11
-
0.5
-
Q059R7_MOUSE
Retinol-Binding Protein 2
lipid transport, vitamin transport, signal transduction,
cellular component morphogenesis,
ectoderm development
cellular component morphogenesis,
ectoderm development
glycolysis
pyrimidine base metabolic process
12
2.8
-
-
VIME_MOUSE
Vimentin
13
0.5
-
-
VIME_MOUSE
Vimentin
14
0.2
-
-
G3P_MOUSE
Glyceraldehyde-3-Phosphate
Dehydrogenase
15
3.5
-
1.8
6PGL_MOUSE
6-Phosphogluconolactonase
pentose-phosphate shunt
16
3.2
-
-
SIN3A_MOUSE
Paired Amphipathic Helix Protein
Sin3a
regulation of transcription from RNA polymerase II
promoter
Flavin Reductase (NADPH)
-
Nucleoside Diphosphate Kinase B
pyrimidine base metabolic process
17
0.3
-
-
BLVRB_MOUSE
18
0.4
-
0.6
NDKB_MOUSE
19
0.5
-
0.7
20
0.4
0.4
21
0.4
0.5
22
3.8
23
0.3
24
2.2
PPIA_MOUSE
Peptidyl-Prolyl cis-trans Isomerase A immune system process, intracellular protein transport,
nuclear transport, protein folding
-
FAAA_MOUSE
Fumarylacetoacetase
cellular amino acid catabolic process
-
ARGI1_MOUSE
Arginase-1
cellular amino acid catabolic process
5.7
-
RGN_MOUSE
Regucalcin
0.5
0.6
ATPB_MOUSE
ATP Synthase Subunit Beta
cation transport, calcium-mediated signaling,
carbohydrate metabolic process
respiratory electron transport chain, purine base
3.8
-
APOA1_MOUSE
Apolipoprotein A-I
ATPB_MOUSE
ATP Synthase Subunit Beta
PARK7_MOUSE
Protein DJ-1
0.5
PDIA3_MOUSE
Protein Disulfide-Isomerase A3
lipid transport, lipid metabolic process
metabolic process
respiratory electron transport chain,
purine base metabolic process
immune system process, regulation of transcription
from RNA polymerase II promoter, response to stress
protein modification process
0.2
ENOB_MOUSE
Beta-Enolase
glycolysis
25
-
2.1
0.3
26
3.8
3.2
-
27
0.1
0.3
28
2.4
10.9
*Significantly modulated proteins with a ratio above 1.5 across two conditions and p-value less than 0.05 assessed by Mann-Whitney test; #Entry name
§
according to UniProtKB/Swiss-Prot database; Biological Process derived from Gene Ontologies by PANTHER Classification System.
Table 2. Label free protein quantitation by nLC-MS/MS analysis*.
ratio
no.
1
ratio
ratio
KO/WT KO/ET ET/WT Entry Name
-
#
§
Protein Description
GO Biological Process
HMCS2_MOUSE
Hydroxymethylglutaryl-CoA
Synthase
coenzyme metabolic process, cholesterol metabolic
process
1.5
-
2
0.7
0.7
-
GSTA3_MOUSE
Glutathione S-Transferase A3
immune system process, response to toxin
3
0.7
-
-
HBE_MOUSE
Hemoglobin Subunit Epsilon-Y2
blood circulation, transport
4
0.7
0.7
-
EF1A1_MOUSE
Elongation Factor 1-Alpha 1
translation
5
0.7
-
-
HBA_MOUSE
Hemoglobin Subunit Alpha
blood circulation, transport
6
0.7
0.7
-
HBB1_MOUSE
Hemoglobin Subunit Beta-1
blood circulation, transport
7
0.7
0.7
-
GSTM1_MOUSE
Glutathione S-Transferase Mu 1
immune system process, response to toxin
8
0.7
0.7
-
PRDX1_MOUSE
Peroxiredoxin-1
immune system process, oxygen and reactive oxygen
species, metabolic process
9
0.7
-
-
TBA1B_MOUSE
Tubulin Alpha-1B Chain
intracellular protein transport, mitosis, cell motion,
chromosome segregation, cellular component
morphogenesis
10
0.7
0.7
-
LDHA_MOUSE
L-Lactate Dehydrogenase A Chain
glycolysis
11
0.7
0.7
-
ALDOB_MOUSE
Fructose-Bisphosphate Aldolase B
glycolysis
12
0.6
0.7
-
BHMT1_MOUSE
Betaine-Homocysteine
S-Methyltransferase 1
cellular amino acid metabolic process
13
0.6
-
-
PDIA3_MOUSE
Protein Disulfide-Isomerase A3
protein modification process
14
0.4
0.7
-
TRFE_MOUSE
Serotransferrin
transport
15
0.4
0.3
1.5
CPSM_MOUSE
Carbamoyl-Phosphate Synthase
pyrimidine base metabolic process, cellular amino acid
biosynthetic process
16
0.4
-
-
PDIA1_MOUSE
Protein Disulfide-Isomerase
protein modification process
17
0.2
0.4
-
GRP78_MOUSE
78 kDa Glucose-Regulated Protein
immune system process, protein folding, protein complex
assembly, response to stress
18
0.1
-
-
FABP5_MOUSE
Fatty Acid-Binding Protein 5
lipid transport, vitamin transport, signal transduction,
lipid metabolic process, ectoderm development
transport, mesoderm development
19
-
0.7
-
FETA_MOUSE
Alpha-Fetoprotein
20
-
0.7
-
ALBU_MOUSE
Serum Albumin
transport
21
-
0.7
-
F16P1_MOUSE
Fructose-1,6-Bisphosphatase 1
gluconeogenesis
22
-
0.3
-
ASSY_MOUSE
Argininosuccinate Synthase
nitrogen compound metabolic process, cellular amino acid
biosynthetic process
23
-
-
1.4
CALM_MOUSE
Calmodulin
cell cycle, calcium-mediated signaling
24
-
-
0.6
ACOX1_MOUSE
Peroxisomal Acyl-Coenzyme A
Oxidase 1
fatty acid beta-oxidation
*Significantly modulated proteins with a ratio above 1.3 across two conditions and %CV less than 30%; #Entry name according to UniProtKB/Swiss-Prot
§
database; Biological Process derived from Gene Ontologies by PANTHER Classification System.