Abstract
Background:
G protein coupled receptor kinase 2 (GRK2) inhibitor, paroxetine, has been approved to ameliorate diabetic cardiomyopathy (DCM). GRK2 is also involved in regulating T cell functions; the potential modifications of paroxetine on the immune response to DCM is unclear.
Methods and Results:
DCM mouse was induced by high-fat diet (HFD) feeding. A remarkable reduction in the regulatory T (Treg) cell subset in DCM mouse was found by flow cytometry, with impaired cardiac function evaluated by echocardiography. The inhibited Treg differentiation was attributable to insulin chronic stimulation in a GRK2-PI3K-Akt signaling-dependent manner. The selective GRK2 inhibitor, paroxetine, rescued Treg differentiation in vitro and in vivo. Furthermore, heart function, as well as the activation of excitation-contraction coupling proteins such as phospholamban (PLB) and troponin I (TnI) was effectively promoted in paroxetine-treated DCM mice compared with vehicle-treated DCM mice. Blockade of FoxP3 expression sufficiently inhibited the proportion of Treg cells, abolished the protective effect of paroxetine on heart function as well as PLB and TnI activation in HFD-fed mice. Neither paroxetine nor carvedilol could effectively ameliorate the metabolic disorder of HFD mice.
Conclusions:
The impaired systolic heart function of DCM mice was effectively improved by paroxetine therapy, partially through restoring the population of circulating Treg cells by targeting the GRK2-PI3K-Akt pathway.
More than half of the diabetic patients died of heart failure (HF) caused by diabetic cardiopathy, such as coronary artery disease (CAD), diabetic cardiomyopathy (DCM), and diabetic hypertensive heart disease. With the application of an insulin complementary strategy, the prevalence of DCM increases rapidly.1
We have characterized that chronic insulin stimulation promotes phosphodiesterase (PDE4D) expression through the insulin receptor (IR)-G protein coupled receptor kinase 2 (GRK2)-βarrestin2-ERK pathway, leading to impressed cAMP and PKA activity, which results in the impaired heart function.2
Besides the direct effect of insulin to the heart, cardiac dysfunction induced by insulin through disturbing the homeostasis of the organism should also be emphasized. A large amount of evidence indicates that the activation of the immune system plays an important role in the pathogenesis of diabetes as well as cardiovascular diseases.3
Type 2 diabetes is accepted as a chronic inflammatory disease.4
Long-term inflammatory stress damaging islet β cells is an important mechanism for the development of insulin resistance in type 2 diabetes. Previous studies have shown this inflammatory injury mainly involves the innate immunity, with the release of inflammatory factors by innate immune cells under the microenvironment of hyperinsulinemia.5
However, adaptive immune regulation is also involved, especially CD4 positive helper T (Th) cells. Diabetes patients are reported to suffer from weakened immune defenses.6
To further reveal the pathogenesis of heart dysfunction in diabetes, the effect of insulin on the immune system of the DCM model needs to be elucidated.
In the present study, 2 important circulating Th lymphocyte subsets, Th17 and regulatory T (Treg) cells, were determined in a high-fat diet (HFD) feeding-induced DCM mouse model and we found a remarkable reduction of Treg, which is an immune-inhibitory subset. Subsequently, the attenuated Treg differentiation by insulin stimulation was revealed to be GRK2-PI3K-Akt-signaling dependent. Paroxetine treatment restored the differentiation of Treg cells and rescued the heart function of DCM mice, confirming the pivotal role of immune disorders in diabetic heart dysfunction and suggesting an immunotherapy strategy for treating DCM.
Methods
Animals
Male C57BL/6J mice (5–6 weeks old) were purchased from Slack Laboratory Animals Co., Ltd. (Shanghai), license number is SCXK (Zhejiang) 2015-0009. All mice were raised in a specific pathogen-free animal laboratory at Anhui Medical University with a 12-h light/dark cycle.
High Fat Feeding and Treatment
C57BL/6 mice were fed a HFD (Research Diets, D12492, 5.24 kcal/g with 20% energy derived from protein, 60% from fat, and 20% from carbohydrate) for 4 months to establish DCM. Mice fed a low-fat diet (LFD, Research Diets, D12450J, 3.85 kcal/g with 20% of calories from protein, 10% of calories from fat, and 70% of calories from carbohydrate) were set as normal chow (NC) controls. The HFD mice were treated with vehicle (5% DMSO), carvedilol (2.5 mg·kg−1·day−1), or paroxetine (2.5 mg·kg−1·day−1) by gavage for another one month with continuous feeding.2
In a specific experiment, HFD mice treated with paroxetine were intraperitoneally injected with 100 μg of penetrating peptide P60 (RDFQSFRKMWPFFAM, FoxP3 blocker) or P301 (MKMFFDAFPQRRSWF, negative control) daily.7
Organ Collection and Weight
The mice were weighted and euthanized. Organs including the heart, liver, and lung were collected and rinsed in PBS without calcium to remove the blood. The organs then were dried on paper and weighed. Tibia length was measured as a reference to analyze the heart index.
Blood Glucose and Intraperitoneal Glucose Tolerance Test (IPGTT)
Blood glucose level and IPGTT were determined after 6-h of fasting with a blood glucose meter (Bayer Contour, Basel, Switzerland) by snipping the tail tip. Blood glucose level was monitored before and 15, 30, 60, and 120 min after 1 g/kg glucose intraperitoneal injection to assess the glucose tolerance of individual mice. The glucose concentration curves were plotted and the areas under curves (AUC) were analyzed.
Flow Cytometry
Circulating Th17 and Treg cells were determined using flow cytometry, as previously described.8
The frequency of Treg cells was detected by using an eBioscience mouse regulatory T cell staining kit (Thermo Fisher Scientific, CA, USA) according to the manufacturer’s instructions. Briefly, 100 μL of T cells were incubated with FITC-CD4, PE-CD25 for 3 min in the dark at room temperature, then were fixed and permeabilized before staining with PE-Cy5-FoxP3 for 30 min. Th17 cells were detected after culture with phorbol myristate acetate and ionomycin for 5 h. The cell samples were recorded on a flow cytometer (FC500, Beckman Coulter, Fullerton, CA, USA) and analyzed using CXP Analysis software (Beckman Coulter). The peripheral lymphocytes were permeabilized and indirectly labeled with p-Akt, p-signal transduction and activator of transcription 5 (STAT5), p-Drosophila mothers against decapentaplegic protein 2 (smad2), and p-smad3 to detect the expression of the mentioned proteins. Cells only incubated with fluorescently labeled secondary antibody were used as isotype controls. Left ventricular tissues were removed and digested in 2.5 mL of 0.2% type II collagenase (Worthington Biochemical, Lakewood, NJ, USA) at 37℃ in Gentle MACS Octo with Heaters (Miltenyi Biotec, Germany). DMEM with 10% FBS was added to stop the digestion. Samples were spun at 300 g for 5 min at 4℃. The pellet was incubated with 2 mL of red blood cell lysis buffer for 2 min on ice. After centrifugation, the cell pellet was resuspended in 100 μL of PBS without calcium. And 1 μL of anti-CD4-FITC antibody was used to show CD4-positive cells in heart tissue.
Treg Cell Differentiation
Splenic lymphocytes were collected from spleen suspension using lymphocyte separation liquid.9
Splenic lymphocytes were incubated with anti-CD4-FITC and anti-CD62L-PE antibodies for 30 min at 4℃ and were then sorted in a FACS Aria cell sorting system (FACSARIA II, BD, MD, USA). After sorting, CD4+
CD62L+
naïve T cells were enriched to more than 95% as analyzed by flow cytometry. Naïve T cells were suspended in RPMI-1640 complete medium with 5% FBS, and were seeded in 48-well plates at a concentration of 2×106
cells/mL. After culturing with anti-CD3/28 mAb (BD Bio-sciences, CA, USA) 50 ng/mL, IL-2 (PeproTech, NJ) 100 U and TGF-β1 (PeproTech, NJ) 5 ng/mL for 5 days, the cells were collected and the population of Treg cells, the expression of p-smad2, p-smad3, p-STAT5 and p-Akt was determined by flow cytometry, and the mRNA expression of FoxP3 and CTLA-4 were detected by real-time qRT-PCR.
Real-Time qRT-PCR
The mRNA expression of FoxP3, RORγt, CTLA-4, ANP and BNP were detected by applying real-time qRT-PCR as reported.10
Briefly, purified T cells were lysed in TRI-Reagent (Sigma-Aldrich, St. Louis, MO, USA) and total RNA was extracted to perform reverse transcription with a First Strand cDNA Synthesis Kit (Promega, WI, USA). The primers for detecting mentioned genes are listed in
Table 1. The genes were amplified on a real-time fluorescent quantitative PCR system (Applied Biosystems 7500) using SYBR Green PCR Master Mix (Thermo Fisher Scientific, MA, USA).
Table 1.
Primers for Real-Time qRT-PCR
| Gene |
Forward |
Reverse |
| FoxP3 |
5’-CAGCTGCCTACAGTGCCCCTAG-3’ |
5’-CATTTGCCAGCAGTGGGTAG-3’ |
| RORγt |
5’-CCGCTGAGAGGGCTTCAC-3’ |
5’-TGCAGGAGTAGGCCACATTACA-3’ |
| CTLA-4 |
5’-AGAACCATGCCCGGATTCTG-3’ |
5’-CATCTTGCTCAAAGAAACAGCAG-3’ |
| ANP |
5’-TCGTCTTGGCCTTTTGGCT-3’ |
5’-TCCAGGTGGTCTAGCAGGTTCT-3’ |
| BNP |
5’-CTCCTGAAGGTGCTGTCC-3’ |
5’-GCCATTTCCTCCGACTTT-3’ |
| GAPDH |
5’-CATGGCCTTCCGTGTTCCTA-3’ |
5’-CCTGCTTCACCACCTTCTTGAT-3’ |
The concentration of serum TGF-β1, IL-17A and insulin were determined using specific ELISA kits (Elabscience, TX, USA) according to the manufacturer’s instructions. Briefly, 40 μL of serum from mice was diluted with 280 μL Reference Standard & Sample Diluent, 40 μL of Activator reagent I, and 40 μL of Activator reagent II sequentially, and the ELISA kit was equilibrated to room temperature before determination. Biotinylated primary TGF-β1, IL-17A or insulin antibodies (1:100) and HRP-conjugated secondary antibodies (1:100) were applied to detect the concentration of serum TGF-β1, IL-17A or insulin using a standard curve. The ELISA plates were read on a Microplate reader (ELx808, BioTek, VT) at 450 nm.
Echocardiography
After treatment, the cardiac function of mice was recorded using an ultrasound imaging system (Vevo 2100, FUJIFILM VisualSonics, Canada). Mice inhaled anesthesia with isoflurane and were placed on the console in a supine position. The MS550D probe was placed on the chest to record the short-axis consecutive imaging of M-mode for ejection fraction (EF); fractional shortening (FS); interventricular septum end-diastole (IVS; d) and interventricular septum end-systole (IVS; s); left ventricular internal diameter end-diastole (LVID; d) and left ventricular internal diameter end-systole (LVID; s); left ventricular posterior wall end-diastole (LVPW; d) and left ventricular posterior wall end-systole (LVPW; s); left ventricular mass (LV mass); left ventricular volume end-diastole (LV Vol; d) and left ventricular volume end-systole (LV Vol; s) analysis.
Heart Index and Histology
The hearts of mice were collected and weighted. Heart index was shown as heart weight (mg) against Tibia length (cm). Heart tissues were fixed in 4% formalin for 24 h and were embedded in paraffin. Four-μm sections were cut and fixed with H&E staining. The area of papillary muscles which were cross-sectioned was analyzed by Image J software and normalized against NC mouse (Version 1.52q, NIH).
Statistical Analysis
All data were shown as the mean±standard deviation (SD) and the significance was analyzed using GraphPad Prism 6. An unpaired parametric t-test was used for 2-group comparison, and the differences between multiple groups were evaluated by using a one-way ANOVA. P≤0.05 was considered to be significant.
This animal study was approved by the Ethical Review Committee of Animal Experiments, Institute of Clinical Pharmacology, Anhui Medical University (Certificate No. LLSC2017002).
Results
Circulating Treg Cells Were Reduced in HFD-Fed Induced DCM Mice
The imbalanced ratio of Treg/Th17 is a general characteristic of immune disorders in many chronic diseases. With HFD-fed induced DCM mice, we found that the circulating Treg cells were remarkably decreased, whereas the proportion of Th17 cells were slightly upregulated but with no significance, suggesting the immunologic dissonance in DCM mouse primarily due to a downregulated Treg cell pool (Figure 1A,B). As a result, the ratio of Th17 against Treg cells was obviously raised, indicating that an immunologic derangement was triggered by long-term HFD feeding (Figure 1C). In accordance with the percentage of immune cells, the gene expression of transcription factor, Forkhead box p3 (FoxP3), was substantially decreased, and the mRNA level of RORγt was minimally increased in peripheral lymphocytes (Figure 1D,E). Transforming growth factor β1 (TGF-β1) is the main inducer for Treg cell differentiation; furthermore, Treg cells exert their immunosuppressive function by secreting TGF-β1 and IL-10.9
As we detected, plasma TGF-β1 expression was significantly reduced in DCM mice compared to normal controls feeding with NC (Figure 1F). In contrast, the level of IL-17A, which is primarily produced by Th17 cells, was mildly elevated in plasma of HFD-fed mice (Figure 1G). Currently, the potential role of a T cell-mediated immune response is gradually stressed in elucidating the pathogeneses of DCM. We revealed a reduced Treg subpopulation in DCM mice; however, what contributes to the diminishment of Treg cells in DCM mice was totally unclear.
CD4+C62L+
naïve T cells were isolated from splenic cells by fluorescence-activated cell sorting and then were treated with TGF-β1 and IL-2 for 5 days in the presence of anti-CD3 and anti-CD28 antibodies. Under the stimulation of both cytokines, ∼58% of T cells were differentiated into FoxP3-expressing Treg cells. The percentage of Treg cells were decreased to ∼33% when co-treated with insulin and Treg-induction media, demonstrating that insulin substantially prevented Treg differentiation in vitro (Figure 2A,B). Further analysis showed a reduced Foxp3 expression in induced T cells, with insulin administration relative to without insulin (Figure 2C). In addition, after 5 days’ induction, we detected a reduced FoxP3 mRNA expression by qRT-PCR in insulin treated cells (Figure 2D). The external domain of cytotoxic T-lymphocyte antigen 4 (CTLA-4), also named CD152, is abundantly expressed on Treg cells and contributes to the direct suppressive effect of Treg.11
In agreement with the expression of FoxP3, the induction culture media containing both TGF-β1 and IL-2, as well as anti-CD3/ anti-CD28 antibody, promoted the production of CTLA-4 on naïve T cells; in vitro insulin pretreatment remarkably decreased CTLA-4 expression (Figure 2E). A significant decrease in the mRNA level of CTLA-4 was also found in response to insulin stimulation comparing to control-differentiated T cells (Figure 2F). The mechanisms of the differentiation of Treg cells from resting T cells have been intensively studied. As reported, anti-CD28 or IL-2 receptor-triggered phosphoinositide 3-kinase (PI3K)-Akt signaling, TGF-β-activated Smad2/3 pathway and IL-2 receptor-mediated JAK3-STAT5 signaling are basically involved in FoxP3 expression and Treg cell development.12
Therefore, we checked the effect of insulin on the activation of smad2, smad3, STAT5 or Akt respectively to reveal the potential pathways that would be targeted by insulin during the process of Tregs differentiation. Interestingly, the addition of insulin did not obviously inhibit the phosphorylation of smad2, smad3 and STAT5; however, it significantly reduced Akt activation induced by Treg cell differentiation media (Figure 2G–J). The data suggest that insulin restrains Treg cell induction from native T cells in vitro through the PI3K-Akt pathway.
Conversely, it is well known that PI3K is activated by insulin stimulation and is important in glucose transportation and cell growth; however, the reason that insulin inhibited PI3K-Akt signaling in the procedure of Treg cell development and the prevention of Tregs differentiation has not been investigated.13
Previous researches revealed that PTEN, a tumor suppressor, is a natural negative regulator of PI3K/Akt pathway.14
We then investigated whether chronic insulin treatment inhibited PI3K through PTEN by applying a selective PTEN inhibitor, VO-OHpic (Vo). As shown in
Figure 3A, the reduced activity of Akt induced by insulin could not be rescued with additional VO-OHpic treatment, indicating that insulin inhibits PI3K-Akt signaling independent of PTEN. Accumulating researches have revealed a direct interaction of insulin receptor with G protein coupled receptor in cardiac myocytes, adipocytes and hepatocytes, etc. Moreover, the crosstalk between two receptors generally exerts a negative effect on both downstream signaling transductions by a key regulatory molecule, G protein coupled receptor kinase 2 (GRK2).2
We then pretreated the differentiating cells with paroxetine, a specific GRK2 inhibitor, and clearly observed a restoration of AKT phosphorylation comparable to normal developing Treg cells (Figure 3B). In agreement, insulin-reduced Treg differentiation was significantly rescued by paroxetine co-stimulation but not VO-OHpic (Figure 3C). These data confirming that insulin inhibited PI3K-Akt signaling and reduced Treg differentiation is primarily due to trans-activating GRK2 in T cells.
As we observed a GRK2-mediated insulin blockage of Treg differentiation in vitro, the mice fed with HFD for 4 months were then treated with the GRK2 inhibitor, paroxetine, or a classical cardiac protector, carvedilol, as a positive control for 1 month in vivo. As expected, the remarkably reduced Treg populations in HFD-fed mice were successfully recovered in paroxetine-treated mice; however, carvedilol was not able to rescue the production of Treg cells (Figure 4A,B). In accordance, paroxetine significantly increased T cells mRNA expression of FoxP3, which was substantially decreased in the T cells from HFD-fed mice treated with vehicle (Figure 4C). Among mice that were HFD fed, the circulating cytokine, TGF-β1, was greatly increased by paroxetine treatment relative to vehicle treatment (Figure 4D), but the plasma IL-17A expression in HFD-fed mice was not changed by paroxetine (Figure 4E). In contrast, carvedilol did not effectively modulate both cytokines production in DCM mice (Figure 4D,E). Furthermore, the population of cardiac CD4-positive cells from treated mice was detected; data revealed an obvious increment of infiltrated CD4 subset in heart tissue from DCM mice compared with that of NC mice. Paroxetine treatment, but not carvedilol therapy, efficiently reduced the CD4 cell cardiac deposit (Figure 4F).
To investigate whether cardiac function benefits from the improved immune balance in DCM mouse, we determined the contractile function of vehicle-, paroxetine-, or carvedilol-treated mice by echocardiography. As shown in
Figure 5A, 5-month HFD feeding impaired both cardiac EF and FS, which is consistent with findings from previous reports.15
Either paroxetine or carvedilol therapy successfully rescued the systolic heart function of HFD-fed mice (Figure 5A, Table 2A), indicating that carvedilol improves heart function via an immune-independent pathway, and instead might be improved through direct regulation of β-AR function on myocytes. Both treatments did not improve the hypertrophy of hearts in HFD-fed mice; there was an unchanged ratio of heart weight to Tibia length, LV mass, as well as the area of cardiomyocytes (Figure 5B,C, Table 2). Neither paroxetine nor carvedilol could affect the body, lung or liver weight of mice fed with HFD (Table 2B). HFD feeding significantly upregulated the fasting blood glucose level and serum insulin concentration compared with that of NC-fed mice. Paroxetine or carvedilol therapy slightly reduced the increases of fasting blood glucose and insulin level with no significant difference (Figure 5D,E). Consistently, either paroxetine or carvedilol was not found to effectively improve the impaired glucose tolerance of HFD mice when an IPGTT was performed (Figure 5F,G), indicating that both drugs only slightly ameliorated the metabolic disorder of the diabetic mouse. However, the activation of important excitation-contraction (E-C) coupling proteins such as phospholamban (PLB) and troponin I (TnI) were effectively promoted by paroxetine compared to vehicle treatment in DCM mice (Figure 5H). Carvedilol slightly increased the phosphorylation of PLB at serine 16 with no significance, and hardly recovered the phosphorylation of TnI at serine 23/24 (Figure 6A). Nevertheless, neither carvedilol nor paroxetine was able to reduce the mRNA levels of ANP and BNP, both of which are diagnostic and predictive markers of systolic heart dysfunction (Figure 5I,J). These results revealed a novel immune regulation effect of paroxetine in HFD-fed-induced DCM mice that eventually improves the impaired heart function.
Table 2.
Systolic Heart Function and Organ Weight of HFD-Fed Mice After Treatment
| (A) Systolic heart function of HFD-fed mice after treatment |
| |
NC |
HFD-Veh |
HFD-Paro |
HFD-Carv |
| Heart rate |
517.30±59.53 |
447.00±38.64 |
531.80±56.22* |
435.20±23.36 |
| IVS; d (mm) |
0.68±0.10 |
0.82±0.05# |
0.73±0.03 |
0.74±0.09 |
| IVS; s (mm) |
1.17±0.11 |
1.05±0.14 |
1.09±0.09 |
1.06±0.17 |
| LVID; d (mm) |
4.30±0.33 |
3.87±0.12 |
4.06±0.42 |
4.05±0.29 |
| LVID; s (mm) |
2.87±0.17 |
3.21±0.24# |
2.87±0.19* |
2.88±0.29* |
| LVPW; d (mm) |
0.71±0.11 |
0.93±0.04## |
0.80±0.10 |
0.82±0.14 |
| LVPW; s (mm) |
1.20±0.05 |
0.94±0.13# |
1.14±0.18* |
1.14±0.15* |
| EF (%) |
52.78±2.46 |
38.86±6.71### |
55.12±4.10** |
56.24±4.39** |
| FS (%) |
29.01±2.04 |
21.23±2.42### |
28.26±1.59** |
28.32±1.42** |
| LV mass (mg) |
109.20±10.59 |
129.2±12.70# |
112.80±14.51 |
112.90±17.22 |
| Mass (corre) |
82.86±10.60 |
103.00±11.81# |
96.40±14.19 |
98.36±14.27 |
| LV vol; d (μL) |
90.04±16.05 |
64.48±6.27# |
72.99±15.00 |
85.55±15.72 |
| LV vol; s (μL) |
31.42±3.66 |
41.43±7.88# |
32.59±5.77* |
33.19±6.61* |
| (B) Body and organ weight of HFD-fed mice after treatment |
| |
NC |
HFD-Veh |
HFD-Paro |
HFD-Carv |
| Body weight (g) |
37.21±2.98 |
51.26±1.62** |
50.80±5.16 |
51.31±4.38 |
| Tibia length (cm) |
2.31±0.09 |
2.25±0.08 |
2.27±0.16 |
2.28±0.08 |
| Heart weight (g) |
0.13±0.02 |
0.16±0.02* |
0.15±0.02 |
0.15±0.02 |
| Heart weight/Tibia length (mg/cm) |
56.86±5.18 |
66.74±6.49* |
61.37±6.99 |
62.83±7.45 |
| Lung weight (g) |
0.17±0.02 |
0.19±0.01 |
0.18±0.01 |
0.18±0.02 |
| Liver weight (g) |
1.42±0.17 |
2.07±0.38** |
2.17±0.50 |
2.05±0.49 |
| (C) Organ weight and blood glucose level of paroxetine-treated DCM mice with FoxP3 blockage |
| |
NC |
HFD-Veh |
HFD-Paro-P301 |
HFD-Paro-P60 |
| Body weight (g) |
33.83±2.27 |
46.32±3.60### |
47.97±4.69 |
47.58±4.34 |
| Tibia length (cm) |
2.31±0.11 |
2.31±0.12 |
2.37±0.10 |
2.35±0.11 |
| Heart weight (g) |
0.13±0.01 |
0.16±0.01## |
0.15±0.01 |
0.15±0.02 |
| Heart weight/Tibia length (mg/cm) |
57.79±4.52 |
70.80±8.03# |
63.47±6.75 |
65.30±9.03 |
| Lung weight (g) |
1.62±0.20 |
1.78±0.16 |
1.80±0.20 |
1.77±0.35 |
| Liver weight (g) |
1.39±0.27 |
2.28±0.40## |
2.36±0.63 |
2.41±0.43 |
| FBG (mg/dL) |
150.70±19.13 |
214.20±42.71## |
181.80±27.28 |
183.80±31.37 |
Carv, carvedilol; corre, corrected; d, end-diastole; DCM, diabetic cardiomyopathy; EF, ejection fraction; FoxP3, Forkhead box p3; FBG, fasting blood glucose; FS, factional shortening; HFD, high-fat diet; IVS, interventricular septum; LV, left ventricular; LVID, left ventricular internal diameter; LVPW, left ventricular posterior wall; NC, normal chow; Paro, paroxetine; s, end-systole; Veh, vehicle; vol, volume. (A) Data are presented as mean±SD, n=10. #P<0.05, ##P<0.01, ###P<0.001 vs. NC mice; *P<0.05, **P<0.001 vs. vehicle-treated HFD mice. (B) Data are presented as mean±SD. *P<0.05, **P<0.001 vs. NC mice. (C) Data are presented as mean±SD. #P<0.05, ##P<0.01, ###P<0.001 vs. NC mice.
To confirm that restoring the percentage of circulating Treg cells is an important target of paroxetine in ameliorating cardiac function of DCM mice, we treated the DCM mice with paroxetine in the presence of a cell-penetrating peptide, P60, which blocks Foxp3 expression as reported, or a negative control peptide, P301.7
As expected, P60 administration substantially abolished paroxetine-induced circulating Treg cell expansion in relative to P301 combined therapy (Figure 6A). The rescued FoxP3 mRNA expression in peripheral T cells from DCM mice under paroxetine treatment was identically blunted by P60, but not by P301 treatment (Figure 6B). Paroxetine and P301 treatment effectively recovered the impaired systolic cardiac function of DCM mice with remarkable increases in EF and FS; however, these increments induced by paroxetine were significantly attenuated by P60 combined treatment, indicating that Treg cell formation is an important target for paroxetine in treating DCM mice (Figure 6C,D). Of note, the data showed that even in DCM mice treated with both paroxetine and P60, their decreased cardiac function was significantly improved (Figure 6C,D). In accordance with the change of circulating Treg cells, FoxP3 expression in heart tissues from DCM mice was distinctly reduced. Paroxetine treatment successfully upregulated cardiac FoxP3 expression in the presence of P301, but failed when in combination with P60 (Figure 6E). Moreover, P60 partially abolished paroxetine-induced recovery of PLB and TnI activities in cardiac tissue of DCM mice (Figure 6E). However, neither treatment was able to substantially influence the body weight, heart weight, heart index against Tibia length, lung or liver weight, or the fasting blood glucose concentration (Table 2C). These data indicate that paroxetine may exert its cardiac protective function through, but not restricted to, immunoregulation; however, Treg cell expansion is a pivotal target of paroxetine treatment.
Discussion
It has been reported that the mRNA level and activity of GRK2 is significantly increased in left ventricular of patients with DCM and other vascular diseases that can cause HF.2
Subsequently, GRK2 activity in lymphocytes of HF patients was revealed to be upregulated in accordance with that in myocardium, and was positively associated with the loss of β-adrenergic receptor response as well as heart dysfunction. Therefore, lymphocytes GRK2 expression and activity can be used as a surrogate biomarker for monitoring cardiac GRK2 expression in HF patients.16
In the present work, we found that the subset of Treg cells was remarkably reduced in DCM mice due to hyperinsulinemia, and this process was GRK2 dependent, indicating that the increased GRK2 expression in circulating lymphocytes impaired Treg differentiation and influenced the immune homeostasis of DCM mice.
Although the molecular mechanisms mediating the role of Treg cells in the pathogenesis of cardiovascular dysfunction is still to be elucidated, accumulating evidence has strongly indicated that a decreased percentage of Treg, which is involved in the imbalance of immune homeostasis and tolerance, participates in the development and progression of chronic HF. Of note, a lower frequency of Treg is associated with cardiovascular complications in patients with type 1 diabetes mellitus.17
In case of pregnancy or after myocardial infarction, Treg cells promote cardiomyocyte proliferation.18
Moreover, Treg cells may infiltrate the heart tissue under chronic stress to exert an anti-inflammatory effect and protect cardiomyocytes against inflammatory damage.19
Therefore, restoration of proper Treg proportion and function might provide a promising and novel therapeutic strategy to the immunomodulation and treatment of cardiovascular diseases such as DCM.20
The differentiation of Treg is dependent on the stimulation of specific cytokines including TGF-β and IL-2. TGF-β triggers the activation of smad2 and smad3 via its receptor complex and promotes the binding of smad4 with the smad binding site on DNA, which regulates downstream genes transcription.21
In addition, TGF-β and IL-2 promote demethylation of the regulatory region for the Foxp3 gene and facilitates Foxp3 expression. IL-2 induces Treg differentiation by phosphorylating STAT5 through both the IL-2 receptor α chain (CD25) and β chain (CD122).22
Moreover, researches have revealed that the PI3K-Akt pathway is important in Treg cell development, function, and stability.23
Both Akt and STAT5 cascades mediate the early metabolic response to TCR stimulation and contribute to naïve human CD4+
T cell differentiation.24
We explored the underlining mechanism of insulin-impaired Treg differentiation and demonstrated that hyperinsulinemia prevents Treg cell induction in a GRK2-PI3K-Akt-dependent manner by using an individual pathway inhibitor.
GRK2 is well established as an important serine-threonine kinase; however, it is a negative regulator of the PI3K-Akt pathway. Increased GRK2 expression in the heart of type 2 diabetic rats leads to the impairment of Akt/eNOS signaling and drives myocyte apoptosis and heart dysfunction.25
Inhibiting GRK2 in the liver significantly recovers glucose homeostasis and activates the Akt/eNOS-mediated insulin signal in type 2 diabetic mice.2
Moreover, in white blood cells, overexpression of GRK2 prevented CCL2-induced Akt phosphorylation.26
As a consequence, blocking GRK2 activity may promote insulin sensitivity of metabolic disorders, rescue the contractile function of myocytes under chronic cardiovascular diseases, and facilitate the migration and function of immunocytes to inflammatory responses.
Paroxetine binds to GRK2 directly and blocks kinase activity, with a 60-fold selectivity for GRK2 over other GRKs, and it has been conformed to be beneficial for the heart function of DCM mice.27
Paroxetine treatment rescued systolic heart function and restored the activation of PLB and TnI, which are important E-C coupling proteins without influencing heart hypertrophy. Previously, we reported that paroxetine directly inhibits the interaction between the insulin receptor and β-AR on cardiomyocytes and decreases β-arrestin2-ERK-induced production of PDE4D, resulting in improved cardiac contractile function of HFD mice.3
The present data confirms that besides the direct action on myocytes, paroxetine may rescue heart function of DCM mice through restoring immune balance and tolerance. It successfully increased the circulating Treg subset in DCM mice in vivo and promoted naïve T cell differentiation to Treg cells in vitro.
Taken together, we found a downregulated Treg subset in DCM mice and the impaired Treg differentiation was mediated by hyperinsulinemia through the GRK2-PI3K-Akt pathway. Paroxetine treatment effectively restored the Treg subpopulation and differentiation both in vivo and in vitro and rescued the heart function of mice with DCM (Figure 6F).
Acknowledgments
This work was supported by the National Natural Science Foundation of China (81202541, 81973314), the Anhui Provincial Natural Science Foundation (1808085J28, 1408085MH173), the Key Research and Development Program of Anhui Province (1804h08020291), and the Young Outstanding Doctor Research Program of Anhui Provincial Hospital (2015).
Disclosures
The authors declare no potential conflicts of interest.
References
- 1.
Borghetti G, von Lewinski D, Eaton DM, Sourij H, Houser SR, Wallner M. Diabetic cardiomyopathy: Current and future therapies. Beyond Glycemic Control. Front Physiol 2018; 9: 1514.
- 2.
Wang Q, Liu Y, Fu Q, Xu B, Zhang Y, Kim S, et al. Inhibiting insulin-mediated β2-adrenergic receptor activation prevents diabetes-associated cardiac dysfunction. Circulation 2017; 135: 73–88.
- 3.
Lutgens E, Atzler D, Döring Y, Duchene J, Steffens S, Weber C. Immunotherapy for cardiovascular disease. Eur Heart J 2019; 40: 3937–3946.
- 4.
Keane KN, Calton EK, Carlessi R, Hart PH, Newsholme P. The bioenergetics of inflammation: Insights into obesity and type 2 diabetes. Eur J Clin Nutr 2017; 71: 904–912.
- 5.
Donath MY, Dinarello CA, Mandrup-Poulsen T. Targeting innate immune mediators in type 1 and type 2 diabetes. Nat Rev Immunol 2019; 19: 734–746.
- 6.
Mahabadi-Ashtiyani E, Sheikh V, Borzouei S, Salehi I, Alahgholi-Hajibehzad M. The increased T helper cells proliferation and inflammatory responses in patients with type 2 diabetes mellitus is suppressed by sitagliptin and vitamin D3 in vitro. Inflamm Res 2019; 68: 857–866.
- 7.
Moreno Ayala MA, Gottardo MF, Imsen M, Asad AS, Bal de Kier Joffé E, Casares N, et al. Therapeutic blockade of Foxp3 in experimental breast cancer models. Breast Cancer Res Treat 2017; 166: 393–405.
- 8.
Zhang M, Hu S, Tao J, Zhou W, Wang R, Tai Y, et al. Ginsenoside compound-K inhibits the activity of B cells through inducing IgD-B cell receptor endocytosis in mice with collagen-induced arthritis. Inflammopharmacology 2019; 27: 845–856.
- 9.
Hu S, Wang R, Zhang M, Liu K, Tao J, Tai Y, et al. BAFF promotes T cell activation through the BAFF-BAFF-R-PI3K-Akt signaling pathway. Biomed Pharmacother 2019; 114: 108796.
- 10.
Han Y, Wang Q, Fan X, Chu J, Peng J, Zhu Y, et al. Epigallocatechin gallate attenuates overload-induced cardiac ECM remodeling via restoring T cell homeostasis. Mol Med Rep 2017; 16: 3542–3550.
- 11.
owshanravan B, Halliday N, Sansom DM. CTLA-4: A moving target in immunotherapy. Blood 2018; 131: 58–67.
- 12.
Iizuka-Koga M, Nakatsukasa H, Ito M, Akanuma T, Lu Q, Yoshimura A. Induction and maintenance of regulatory T cells by transcription factors and epigenetic modifications. J Autoimmun 2017; 83: 113–121.
- 13.
Rai U, Kosuru R, Prakash S, Singh SP, Birla H, Tiwari V, et al. Tetramethylpyrazine prevents diabetes by activating PI3K/Akt/GLUT-4 signalling in animal model of type-2 diabetes. Life Sci 2019; 236: 116836.
- 14.
Zhu HH, Wang XT, Sun YH, He WK, Liang JB, Mo BH, et al. MicroRNA-486-5p targeting PTEN protects against coronary microembolization-induced cardiomyocyte apoptosis in rats by activating the PI3K/AKT pathway. Eur J Pharmacol 2019; 885: 244–251.
- 15.
West TM, Wang Q, Deng B, Zhang Y, Barbagallo F, Reddy GR, et al. Phosphodiesterase 5 associates with β2 adrenergic receptor to modulate cardiac function in type 2 diabetic hearts. J Am Heart Assoc 2019; 8: e012273.
- 16.
Iaccarino G, Barbato E, Cipolletta E, De Amicis V, Margulies KB, Leosco D, et al. Elevated myocardial and lymphocyte GRK2 expression and activity in human heart failure. Eur Heart J 2005; 26: 1752–1758.
- 17.
El-Shabrawy RM, Ahmed AM, Selim FO, Said NM. Association between CD4+, CD25+, FOXP3+ regulatory T-cells and cardiovascular complications in diabetic patients type 1. Egypt J Immunol 2019; 26: 129–139.
- 18.
Zacchigna S, Martinelli V, Moimas S, Colliva A, Anzini M, Nordio A, et al. Giacca, Paracrine effect of regulatory T cells promotes cardiomyocyte proliferation during pregnancy and after myocardial infarction. Nat Commun 2018; 9: 2432.
- 19.
Blanton RM, Carrillo-Salinas FJ, Alcaide P. T-cell recruitment to the heart: Friendly guests or unwelcome visitors? Am J Physiol Heart Circ Physiol 2019; 317: H124–H140.
- 20.
Meng X, Yang J, Dong M, Zhang K, Tu E, Gao Q, et al. Regulatory T cells in cardiovascular diseases. Nat Rev Cardiol 2016; 13: 167–179.
- 21.
Liao H, Peng X, Gan L, Feng J, Gao Y, Yang S, et al. Protective regulatory T cell immune response induced by intranasal immunization with the live-attenuated pneumococcal vaccine SPY1 via the transforming growth factor-β1-Smad2/3 pathway. Front Immunol 2018; 9: 1754.
- 22.
Freudenberg K, Lindner N, Dohnke S, Garbe AI, Schallenberg S, Kretschmer K. Critical role of TGF-β and IL-2 receptor signaling in Foxp3 induction by an inhibitor of DNA Methylation. Front Immunol 2018; 9: 125.
- 23.
Pompura SL, Dominguez-Villar M. The PI3K/AKT signaling pathway in regulatory T-cell development, stability, and function. J Leukoc Biol 2018; 103: 1065–1076.
- 24.
Jones N, Vincent EE, Cronin JG, Panetti S, Chambers M, Holm SR, et al. Akt and STAT5 mediate naïve human CD4+ T-cell early metabolic response to TCR stimulation. Nat Commun 2019; 10: 2042.
- 25.
Wan ZH, Zhang YJ, Chen L, Guo YL, Li GH, Wu D, et al. G protein-coupled receptor kinase 2 inhibition improves erectile function through amelioration of endothelial dysfunction and oxidative stress in a rat model of type 2 diabetes. Asian J Androl 2018; 21: 74–79.
- 26.
Han CC, Ma Y, Li Y, Wang Y, Wei W. Regulatory effects of GRK2 on GPCRs and non-GPCRs and possible use as a drug target (Review). Int J Mol Med 2016; 38: 987–994.
- 27.
Thal DM, Homan KT, Chen J, Wu EK, Hinkle PM, Huang ZM, et al. Paroxetine is a direct inhibitor of g protein-coupled receptor kinase 2 and increases myocardial contractility. ACS Chem Biol 2012; 7: 1830–1839.
