Many efforts have been undertaken to unravel the role of the immune system in the pathogenesis of Nonalcoholic Steatohepatitis (NASH) [
Th17 are a subset of effector T cells that express the nuclear receptor retinoid-related orphan receptor
Tregs derive from CD4+ Th0 cells and constitutively express CD25 (the IL2 receptor
Moreover, Tregs were specifically reduced in the AAT of preclinical insulin-resistant models of obesity [
Other key factors in the pathogenesis of NASH are the adipokines, which are mainly produced in the white adipose tissue. Leptin is able to decrease appetite and concomitantly increases energy expenditure [
In order to further elucidate the differential role of Th17 and Tregs at different anatomical sites in the pathogenesis of NASH, we analyze in the present study the Treg and Th17 populations in mice fed HFD and their location-specific modifications with particular interest in liver and AAT and SAT. Moreover we analyze the possible correlation between the tissue-specific variations of these cells and the diet-induced disturbances of metabolic homeostasis.
C57BL6/J six-to-eight-week-old male mice were purchased (Charles River Laboratories, Brussels, Belgium) and kept at the animal facility of the University of Antwerp in temperature- and light-controlled conditions. Mice were allowed to consume water and pellet chow ad libitum.
Mice were fed for 36 weeks with either HFD (with 60% of kcals derived from fat, 20% from proteins, and 20% from carbohydrate) (Research Diets) or normal chow, in line with what was previously described [
The protocols were approved by the Antwerp University Ethical Committee on Animal Experiments (permit number: 2012-47). The animals received human care and were treated according to the Helsinki declaration, the national guidelines for animal protection, and the “Guide for the Care and Use of Laboratory Animals” (National Institutes of Health, 1985).
Mice were anesthetized with pentobarbital. Blood samples on ethylenediaminetetraacetic acid (EDTA) were collected from anesthetized mice by cardiac puncture. SAT and AAT and liver were excised after flushing organs through the right ventricle with 5 mL of cold phosphate buffered saline (PBS) (Sigma) and cut into small pieces of 1-2 mm. Liver tissue was smashed through a sieve and intrahepatic immune cells were harvested after centrifugation with 33% Percoll (Sigma) and heparin [
The cells were incubated with the following fluorochrome conjugated antibodies: CD45-eF450 (BD), CD3-PerCPCy5.5 (Bio-Legend), CD4-FITC (e-Bioscience), and CD25-AlexaFluor700 (e-Bioscience). The cells were fixed in Phosflow, Lyse Fix (Becton Dickinson). Intracellular stains for FOXP3-APC (e-Bioscience) and ROR
Samples were measured on a FASCanto II flow cytometer with DIVA software (BD) and analyzed with Kaluza software (Beckman Coulter). Lymphocytes were selected based on side scatter and CD45 positivity. T lymphocytes were selected as CD45+CD3+. Among this population, Tregs were defined as CD4+CD25+FOXP3+ and Th17 cells as CD4+ROR
After fixation and sectioning, the liver tissue samples were stained with Haematoxylin-Eosin and Trichrome-Masson. Steatosis, lobular inflammation, and ballooning were scored in a blinded way by one single experienced hepatologist according to the NASH Clinical Research Network Scoring System. The fibrosis grade was also assessed [
Total RNA was extracted from stored frozen liver samples using the column-based technique (RNeasy Minikit, Qiagen, KJ Venlo, Netherlands), according to the manufacturer’s instructions. The total RNA extraction for the SAT and AAT samples was based on the Trizol (Invitrogen, Life Technology, Belgium) procedure from W. M. Keck Foundation Biotechnology Microarray Resource Laboratory at Yale University. Purified total RNAs were treated with DNase to obtain DNA-free RNA (Turbo DNase free, Ambion, Life Technology, Belgium). cDNA was synthesized using a Transcriptor First Strand cDNA Synthesis Kit (Roche Applied Science, Indianapolis, IN).
TaqMan Gene expression assay for IL10 (Life Technologies, Belgium, Gene ID 16153, reference Mm00439614_m1), IL17A (Life Technologies, Belgium, Gene ID 16171, reference Mm00439618_m1), and leptin (Gene ID 16846, reference Mm00434759_m1) was performed on an ABI Prism 7300 sequence detector system (Applied Biosystems, Life Technology, Belgium) in 25
Alanine aminotransferase (ALT) was measured by means of enzyme-linked immunosorbent assay (ELISA). All samples were analyzed on the same day. Fasting glucose was measured with a blood glucose-meter (Accu Check).
The data are presented as median (range). The Mann-Whitney
Table
Main characteristics of the model.
ND ( |
HFD ( |
| |
---|---|---|---|
|
5.6 (3–9) | 30.6 (12–42) |
0.001 |
Fasting glucose (mg/dL) | 101 (85–148) | 166 (77–291) | 0.02 |
ALT (UI/L) | 30 (20–33) | 223 (179–250) | 0.01 |
NAS | 0 | 4.5 (1–6) | 0.000004 |
Steatosis ( |
|||
0 | 7 | 1 | |
1 | 0 | 4 | |
2 | 0 | 3 | 0.000451 |
Lobular inflammation ( |
|||
0 | 7 | 0 | |
1 | 0 | 6 | |
2 | 0 | 2 | |
3 | 0 | 0 | 0.00002 |
Ballooning ( |
|||
0 | 7 | 1 | |
1 | 0 | 0 | |
2 | 0 | 7 | 0.000008 |
Fibrosis ( |
|||
0 | 7 | 5 | |
1 | 0 | 3 | |
2 | 0 | 0 | 0.08 |
Validation of the model:
Liver histology in mice fed normal diet (ND) ((a) Hematoxylin/Eosin stain; (b) Masson’s Trichrome stain, original magnification ×10) or high fat diet (HFD) ((c) Hematoxylin/Eosin stain; (d) Masson’s Trichrome stain, original magnification ×10). Mice fed ND showed a normal liver histology while mice fed HFD showed steatosis, lobular inflammation, and ballooning and therefore proved the presence of Nonalcoholic Steatohepatitis (NASH).
These data confirmed the validity of the experimental model to obtain HFD-induced NASH together with an impairment of glucose metabolism and obesity-like disturbances.
Flow cytometry analysis is summarized in Table
Flow cytometry analysis.
Blood % | Liver % | Abdominal fat % | Subcutaneous fat % | |||||
---|---|---|---|---|---|---|---|---|
(median (range)) | (median (range)) | (median (range)) | (median (range)) | |||||
ND | HFD | ND | HFD | ND | HFD | ND | HFD | |
CD45+ | 81 (42.3–78.4) | 86.1 (69.2–96.3) | 66.8 (54.2–78.4) | 60 (48.6–71) | 35.6 (26.8–73.6) | 30.3 (19.8–78.6) | 73.9 (32.5–95.61) | 74.2 (45.8–46.62) |
CD45+CD3+ | 14 (8.4–11.3) | 9.5 (4.9–17.1) | 13.6 (7.9–32.4) | 18.8 (11.7–46.9) | 14.8 (6–61.5) | 11.7 (5.4–27.6) | 33.1 (13.4–51.6) | 24.7 (11.9–28.3) |
CD45+CD3+CD4+ | 42.6 (14.6–39.7) | 39.7 (30.1–52.5) | 33 (27.4–35.6) | 29.1 (9.2–41.4) | 45.9 (40.2–56.2) | 29.4 (14.7–39.8) |
37.4 (28.1–41.6) | 41.5 (31.1–46.6) |
CD45+CD3+CD4+FOXP3+ | 9.6 (6.3–10.9) | 10 (7.6–15.3) | 5.4 (3.3–32.6) | 8.5 (4.6–45) | 22.8 (13.5–45.6) | 16 (10.7–20.1) |
8.2 (0.8–13.2) | 11.7 (7.6–18.2) |
Tregs (CD4+CD25+FOXP3+) | 3.5 (1.7–5.5) | 3 (2–4.4) | 0.14 (0.06–0.6) | 0.2 (0.6–3.3) | 1.4 (0.6–6.7) | 3 (0.6–4.3) | 0.6 (0.3–1.3) | 1.3 (0.8–3.5) |
CD4+ROR |
1.7 (0.5–6.7) | 2.2 (0.7–5) | 2.4 (0.9–5.1) | 9.7 (1.5–24.6) |
2.3 (0.7–5.8) | 8.7 (2.7–15.4) |
3.6 (1–18.5) | 4.5 (2–10.2) |
Flow cytometric analysis of blood, liver, abdominal fat, and subcutaneous fat in mice fed high fat diet (HFD) and normal diet (ND). Cells are represented as % [(median (range)] of the gated population: CD45+CD3+ were gated on CD45+, CD45+CD3+CD4+ were gated on CD45+CD3+, CD45+CD3+CD4+FOXP3+ were gated on the CD45+CD3+CD4+, Tregs (CD45+CD3+CD4+CD25+FOXP3+) were gated on the CD45+CD3+CD4+, and CD45+CD3+CD4+ROR
Tregs were significantly increased in the SAT of mice fed HFD
(a) Representative flow cytometry plots of CD4+ROR
RT-PCR showed a positive cytokine response to HFD: both IL10 and IL17A were expressed in liver (ΔCt values, resp., 15.5 (15.4–16.5) and 21.7 (21.2–22.3)) and both the AAT (resp., 10.8 (5.1–10.9) and 8 (6.9–8.2)) and the SAT (resp., 11.3 (8–11.7) and 6.4 (5.5–6.9)) of mice fed HFD, while they were nondetectable in mice fed ND. Moreover leptin gene expression was increased in the adipose tissue of mice fed HFD compared to mice fed ND. This increase was more pronounced in the SAT district: in the abdominal district leptin increased 2-fold versus 986-fold in the SAT (Figure
In addition the significant correlations (Table
Significant correlations between immune cells and metabolic and histological parameters.
CD4+ROR |
|
CD4+ROR |
|
CD4+ROR |
|
CD4+ROR |
|
CD4+ROR |
|
CD4+ROR |
|
Treg subcutaneous fat and |
|
Tregs subcutaneous fat and fasting glucose |
|
Treg subcutaneous fat and ballooning |
|
Treg subcutaneous fat and inflammation |
|
Treg subcutaneous fat and NAS |
|
|
|
|
|
|
|
Significant correlations with the respective correlation coefficient (Spearman’s rank correlation coefficient,
There is a growing interest in the role of the immune system as a key contributor to the pathogenesis of NASH and the metabolic syndrome [
We used a 36-week HFD mouse model to induce NASH. The histologic features fulfilled the histologic criteria for NASH (defined as the simultaneous presence of steatosis, inflammation, and ballooning). There was also a grade 1 fibrosis in some of them [
When considering the different organs investigated (Figure
Immune cells, cytokines, and adipokines modifications after high fat diet (HFD). Th17 were significantly increased in liver and abdominal adipose tissue (AAT) of mice fed a HFD while Tregs were significantly increased in the subcutaneous adipose tissue (SAT). After HFD there was an increase in the leptin gene expression in the adipose tissue, more pronounced in the SAT. Moreover there was a positive cytokine response to HFD in liver and adipose tissue (both AAT and SAT).
In the AAT, similar to the liver, we observed an increase of the CD4+ROR
In the SAT the Tregs played a predominant role as we observed no statistically significant variations of the CD4+ROR
As far as the Th17 are concerned, our results are in agreement with Tang et al., who demonstrated an increased number of hepatic Th17 and an increased hepatic gene expression in a HFD mouse model, as well as an attenuation of the LPS-induced liver injury after neutralization of IL17 [
Regarding the Tregs, previous preclinical studies reported that AAT is a preferential site of accumulation of Tregs in mice fed a ND and that they decrease after HFD [
Finally we found a leptin upregulation in mice fed a HFD, mainly in the SAT, which, through its systemic effect, may have contributed to the stimulation of immune cells, and particularly the Th17, at distance. This upregulation of the leptin gene expression is in agreement with previous findings [
In addition leptin has shown proinflammatory, profibrogenic, and prodiabetogenic effects [
In line with the ability of leptin to stimulate the inflammatory immune response [
In contrast to what could be expected, however, this increase in the proinflammatory IL17A was not counterbalanced by a decrease of the IL10 mRNA in the same sites. Moreover we observed a positive correlation between adipose tissue-derived IL10 and hepatic inflammation at histology. IL10 is an anti-inflammatory cytokine. Its role, however, in diet-induced steatohepatitis and insulin resistance is controversial [
In contrast to the flow-cytometric data, the gene expression analysis showed a cytokine upregulation in the liver and in both the AAT and SAT. This discrepancy can be explained by the contribution to cytokine secretion by other cell types, which cannot be discriminated by this method. IL10 increase could be explained, at least in part, due to a tissue-recruitment of M2 macrophages in response to HFD, as previously described [
In conclusion our results show a stimulation of both the Th17 and the Treg axis in NASH, also in correlation with its metabolic complications. These pathways seem to act differently at different sites, the Th17 axis being upregulated at the level of the AAT and the liver, and the Treg axis being upregulated in the SAT. It has recently been shown that SAT expresses genes that are implicated in inflammation that correlate with liver damage and is therefore also potentially implicated in NASH pathogenesis [
The authors declare that there is no conflict of interests regarding the publication of this paper.
The authors would like to thank A. Jürgens for the technical support for RT-PCT, P. Aerts and M. Vinckx for support in the tissue preparation, and C. Mertens for the technical support for flow cytometry.