- Email: [email protected]
Activation of brown adipocytes by placental growth factor
Biochemical and Biophysical Research Communications xxx (2018) 1e8
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Activation of brown adipocytes by placental growth factor Jing Zhou a, 1, Nan-Nan Wu a, 1, Rui-Li Yin a, 1, Wei Ma b, Cen Yan a, Ying-Mei Feng a, Chuan-Hai Zhang c, **, Dong Zhao a, * a
Beijing Key Laboratory of Diabetes Prevention and Research, Endocrinology Centre, Lu He Hospital, Capital Medical University, Beijing, 101149, China2 Gynecology and Obstetrics Centre, Lu He Hospital, Capital Medical University, Beijing, 101149, China2 c Beijing Advanced Innovation Centre for Food Nutrition and Human Health, College Food Science and Nutritional Engineering, China Agricultural University, Beijing, 100083, China3
b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 8 August 2018 Accepted 14 August 2018 Available online xxx
Gestational diabetes mellitus (GDM) is a type of diabetes and occurs during pregnancy. Brown adipose tissue (BAT) improves glucose homeostasis and mitigates insulin resistance, however, its activity is reduced in GDM. Placenta growth factor (PlGF) is an angiogenic factor produced by placental trophoblasts. Nevertheless, whether and how PlGF could affect BAT function in GDM are not defined. To investigate this question, 91 non-diabetic pregnant participants and 73 GDM patients were recruited to Gynaecology and Obstetrics Centre in Lu He hospital. Serum levels of PlGF were quantified by ELISA. Skin temperature was measured by far infrared thermography in the supraclavicular region where classical BATs were located. The direct effect of PlGF on BAT function was explored using the established human preadipocyte differentiation system. Thereby, we demonstrated that serum levels of PlGF were lower in GDM patients compared with controls, which was accompanied by decreased skin temperature in the supraclavicular region. By qPCR and western blot, mRNA and protein expression of UCP1 and OXPHOS were elevated in differentiated adipocytes treated with PlGF. PlGF stimulated mitochondrion transcription and increased copy number of mitochondrial. When subjected for respirometry, PlGF-treated differentiated adipocytes showed higher oxygen consumption rates than controls. PlGF induced AMPK phosphorylation and blockade of AMPK phosphorylation blunted UCP1 and OXPHOS expression in differentiated adipocytes. PlGF administration reduced cholesterol and triglyceride content in the liver and improved insulin sensitivity in db mice compared with control. In Conclusion, PlGF could activate BAT function. Downregulation of PlGF might contribute to the reduced BAT activity in GDM. © 2018 Published by Elsevier Inc.
Keywords: Gestational diabetes mellitus Brown adipose tissue Placenta growth factor
1. Introduction Gestational diabetes mellitus (GDM) is a special type of diabetes and mostly occurs after 24th gestational week. Over the years, the prevalence of GDM is approximately 9.3e25.5% worldwide and 3.9e18.9% in China [1,2]. GDM is not only associated with pre- and perinatal complications but also exposes high risk of type 2 diabetes mellitus (T2DM) for mothers after pregnancy [3]. Although the aetiology of GDM is not yet clear, it shares common features
* Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (C.-H. Zhang), (D. Zhao). 1 Equal contribution to the study. 2 Tel.: þ86 10 6954 3901; fax: þ86 10 6953 1069 3 Tel.: þ86 10 6273 7751 3901; fax: þ86 10 6273 7751
[email protected]
with T2DM such as adiposity [4], insulin resistance [5,6] and inflammation [7,8]. Activation of brown adipocytes and beige cells improves glucose homeostasis and attenuates excessive nutrientsinduced obesity and insulin resistance [9]. Unfortunately, BAT activity is decreased in both T2DM and GDM patients [10,11]. Norepinephrine is the classical stimulator for brown adipocyte and beige cell activation. Binding norepinephrine to b-adrenergic receptor adipocytes induce AMPK phosphorylation that in turn upregulates UCP1 expression for mitochondrial biogenesis and thermogenesis [12]. Except norepinephrine, brown fat activation could be activated by adipokines such as lipocalin 2 [13] and hormones such like incretin [14]. Placental trophoblasts secret a large amount of proteins, growth factors, cytokines and hormones by which the cross-talk between placenta and peripheral tissues is established [15]. Nevertheless, how placenta communicates with BAT remains unknown.
https://doi.org/10.1016/j.bbrc.2018.08.106 0006-291X/© 2018 Published by Elsevier Inc.
Please cite this article in press as: J. Zhou, et al., Activation of brown adipocytes by placental growth factor, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.08.106
2
J. Zhou et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
Placenta growth factor (PlGF) is a member of vascular endothelial growth factor (VEGF) family and mainly produced by trophoblasts in placenta. PlGF levels are reported reduced in subjects with pregnancy complications, which is associated with early pregnant losses [16]. In line with this study, downregulation of PlGF in GDM renders the physical interaction between beta cells and endothelial cells in islets, resulting in impaired beta cells proliferation [17]. We hypothesized that PlGF could be the messenger between BAT function and placenta. To investigate this question, we compared serum PlGF levels in healthy pregnant subjects and GDM patients. We studied the effect of PlGF on BAT marker expression and function in preadipocytes isolated from human embryos [18]. 2. Materials and methods 2.1. Human study This case-control study was performed among pregnant women who admitted in the Gynaecology and Obstetrics Centre in Lu He hospital between November 2015 till December 2016. The subjects who were diagnosed preeclampsia, retinopathy or nephropathy were excluded. Finally, 91 non-diabetic pregnant participants and 73 GDM patients were analysed in the study. Our study complied with the Helsinki Declaration for investigation of human subjects. The study protocol was obtained ethical approval from the competent Institutional Review Boards of Capital Medical University. All participants provided written informed consent. Screening of GDM was performed at 24e28 gestation weeks by an oral glucose tolerance test. The criteria of GDM diagnosis was adapted from American Diabetes Association guideline as following: 1) fasting blood glucose levels 5.3 mmol/L or 2) 1-h oral 100-g glucose tolerance test 10.0 mmol/L [19]. General information including age, disease history and medications was obtained. Body mass index (BMI) at 24e28 gestation weeks were obtained by weight in kilograms divided by the square of height in meters (kg/m2).
study. They were randomized and blindly assigned to receive totalled 1.5 mg murine recombinant PlGF (RD, Bio-Techne, USA) for 4 weeks or equivalent volume of saline through a subcutaneous implanted osmotic minipump (Durect, Cupertino, CA, USA) as described before [23]. At sacrifice, after overnight fasting, blood was bled and liver and fat tissues were dissected. 2.5. Cell culture Primary human embryo adipocytes were isolated from human embryos during abortion and cultivated as described before [18]. Once they reached 100% confluence, cells were induced adipogenic differentiation in the absence or presence of recombinant human PlGF (0e1 ng/mL) for 8 days. Medium was refreshed every two days. For inhibitor experiments, cells were differentiated with 1 mM AMPK phosphorylation inhibitor-Compound C. The detailed information of measurement of oxygen consumption and Phosphorylation pathways using Luminex technology were shown in Supplementary Materials and methods. 2.6. Real-time PCR and western blotting Real-time PCR and western blotting were performed as described previously. The primer sequences are listed in Supplementary Table 1. The specific primary antibodies were shown in Supplementary Materials. 2.7. Immunofluorescence staining Differentiated adipocytes were stained with 1 mg/ml antihuman UCP1, followed by Alexa 488-conjugated secondary antibody (Invitrogen), BODIPY (Thermo) and DAPI (Leagene) complying with the procedure. Brown adipocytes were positive for both UCP1 and BODIPY. Negative controls were stained with omission of primary antibody. Images were taken by Zeiss laser scanning confocal microscopy (LSM780, Germany).
2.2. Clinical measurement
2.8. ELISA
At the antenatal examination, blood pressure was recorded by auscultation of the Korotkoff sounds, using a standard mercury sphygmomanometer. Fasting blood samples were collected and glucose, insulin, total cholesterol, triglyceride and HDL-cholesterol (HDL-c) and creatinine were measured in the central laboratory in the hospital [20]. LDL-cholesterol (LDL-c) was computed from serum total cholesterol and HDL-c and serum triglycerides by the Friedewald equation [21].
Serum levels of PlGF were quantified by enzyme-linked immunosorbent assay (ELISA) according to the manual (MLBio, Shanghai, China).
2.3. Far infrared thermography Because PET-CT could not be applied to measure BAT activity in pregnant women, infrared thermography was used alternatively. As described before [22], the study subject was seated in the upright position with head positioned in a neutral position and the subject looked straight ahead. Images were acquired by a thermal imaging camera (FLIR B425, 3.1Mpixel, FLIR Systems Australia Pty Ltd, Melbourne, Vic., Australia) in the anterior neck and upper chest. Skin temperature in the supraclavicular region where classical BATs were located was analysed by two experienced investigators who were blind to the study using the software (Version 1.2, Wilsonville, OR). 2.4. Mice study Diabetic db mice at the age of 8 weeks old were used in the
2.9. Statistical analysis Data were expressed as mean ± SEM. In the experiments where there were 2 experimental groups, unpaired, 2-tailed Student's ttest was used for data with normal distribution. For data that did not fit normal distribution, nonparametric Mann Whitney analysis was used. In the experiments containing more than 2 experimental groups, one-way ANOVA with Dunnett was used when comparing treated groups against control and ANOVA with Bonferroni was applied when comparing all groups. Statistics analysis was performed using GraphPad Prism (GraphPad Software Inc, La Jolla, CA, U.S.A.). A P value less than 0.05 was considered significant. 3. Results 3.1. Serum PlGF levels were reduced in GDM patients At 24e28 gestation weeks when GDM was diagnosed, BMI was already higher in GDM patients than healthy pregnant subjects (26.24 ± 0.64 kg/m2 vs 24.38 ± 0.87 kg/m2, p ¼ 0.09). General characterization of the study subjects at the antenatal visit was listed in Table 1. Compared with non-diabetic pregnant controls, GDM
Please cite this article in press as: J. Zhou, et al., Activation of brown adipocytes by placental growth factor, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.08.106
J. Zhou et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
3
Table 1 General characterization of the study subjects.
N Age (years) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Fasting blood glucose (mmol/L) Fasting insulin (pmol/L) Haemoglobin (g/L) Total cholesterol (mmol/L) Total triglyceride (mmol/L) LDL-cholesterol (mmol/L) HDL-cholesterol (mmol/L) Homa-b (%) Homa-IR (%)
Non-diabetic
Diabetic
p
91 30.0 (0.6) 117.2 (1.2) 72.3 (1.0) 4.57 (0.05) 15.4 (0.9) 120.6 (1.7) 5.38 (0.13) 1.57 (0.09) 2.49 (0.47) 1.82 (0.16) 184.1 (15.3) 1.68 (0.16)
73 31.6 (0.6) 118.4 (2.1) 76.3 (1.8) 5.11 (0.13) 20.8 (3.4) 116.7 (2.7) 5.45 (0.20) 1.98 (0.3) 2.64 (0.17) 1.86 (0.11) 199.5 (25.7) 2.81 (0.39)
0.06 0.58 0.13 <0.0001 0.04 0.21 0.78 0.07 0.40 0.82 0.59 0.003
Data are expressed by mean ± SEM. LDL, low-density lipoprotein; HDL, high-density lipoprotein; Homa-b and Homa-IR were computed by Homeostasis Model Assessment algorithm (http://www.dtu.ox.ac.uk/homacalculator/) using fasting insulin and fasting blood glucose.
patients were older (p ¼ 0.06) and had higher fasting glucose and insulin levels in the peripheral blood (p < 0.05 for both). The index of insulin resistance was 1.7- fold greater in GDM patients than controls (p ¼ 0.003). Quantified by ELISA, serum PlGF levels were 381.8 (42.5) pg/mL in non-diabetic pregnant participants but reduced to 296.7 (17.16) pg/mL in GDM patients (p < 0.05) (Fig. 1A). Quantified by Far infrared thermography, the skin temperature in the supraclavicular region was lower in GDM patients than controls, indicating reduced BAT activity (Fig. 1B and C). 3.2. PlGF induced activation of human brown adipocytes Previously, we have shown that cells in SVF fraction contained enriched brown preadipocytes that could differentiate into brown adipocytes under desirable condition [18]. To explore whether PlGF had any direct effect on brown adipocyte activation, stromal vascular fraction was isolated from intracapsular BAT of human embryos and cultivated in vitro. Cells were differentiated under PlGF concentrations (0e1 ng/mL) for 8 days. By qPCR, PlGF treatment increased mRNA expression of UCP1, PGC1a, PGC1b, CPT1a and CTP1b in adipocytes in a dose-dependent manner and peaked between 0.5 ng/mL and 1 ng/mL PlGF (Figure s1, A-E). However, PlGF did not alter mRNA expression of PPAR (Figure s1F). Next, we evaluated the protein expression profile in brown adipocytes exposed to PlGF. Since we observed the dose-dependent effect of PlGF on BAT activation, 1 ng/mL PlGF was used in the following experiments. Consistent with previous findings, PlGF stimulated a series of thermogenic gene expression including UCP1, PGC1a, PGC1b, CPT1a and CPT1b (Fig. 2A) in cultivated adipocytes at mRNA levels. In line with increased mRNA expression, protein expression of UCP1 and mitochondrial-specific OXPHOS protein, including ATP5A, ubiquinol-cytochrome C reductase core protein II (UQCRC2), succinate dehydrogenase complex and subunit B (SDHB) and NADH dehydrogenase (ubiquinone) 1b subcomplex (NDUFB8) were significantly elevated in cells treated with PlGF (Fig. 2B and C). The increased UCP1 expression in PlGF-treated adipocytes was further confirmed by dual-immunofluorescence staining (Fig. 2D). 3.3. Mitochondrial function in brown adipocytes induced by PlGF Quantified by qPCR, mRNA levels of mitochondrial transcription factor A (TFAM) and nuclear respiratory factor 1 (NRF-1) were elevated in PlGF-treated cells, suggesting enhanced mitochondrion replication (Fig. 3A). Indeed, the increased mitochondrial replication by PlGF was further confirmed by quantification of
Fig. 1. Serum levels of PlGF and BAT activity in GDM. (A) Quantification of serum levels of PlGF in non-diabetic pregnant participants (n ¼ 58) and GDM patients (n ¼ 76). (B) Skin temperature in the supraclavicular region was analysed by the software. (C) Images were acquired by a thermal imaging camera in the anterior neck and upper chest. The circles indicated the positions where skin temperature was measured. Data are expressed as mean ± SEM.
mitochondrial copy number by qPCR (Fig. 3B). To further examine the function of differentiated brown adipocytes, oxygen consumption rates (OCR) were assessed by respirometry. Terminally differentiated adipocytes were seeded in 96well plate to measure basal respiration and ATP turnover (Fig. 3C). Quantification of the area under the curve (AUC) showed that the basal level of respiration rates was 1.8- fold higher in PlGFtreated cells than non-treated ones (Fig. 3D). Accordingly, the leak respiration rates of PlGF-treated cells were 3.5- fold higher than non-treated ones (Fig. 3D). 3.4. PlGF stimulated AMPKa phosphorylation in brown adipocytes To explore the signalling pathways underlying the activation of brown adipocytes by PlGF, proteins were extracted from differentiated brown adipocytes treated without or with PlGF. Equal
Please cite this article in press as: J. Zhou, et al., Activation of brown adipocytes by placental growth factor, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.08.106
4
J. Zhou et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
Fig. 2. Thermogenic proteins expression in differentiated adipocytes. Preadipocytes were differentiated with 1 ng PlGF for 8 days. (A) Relative mRNA expression of UCP1, PGC1a, PGC1b, CPTa and CPTb in differentiated adipocytes. (B) The protein expression of UCP1, ATP5A, UQCRC2, SDHB and NDUFB8 in differentiated adipocytes. GAPDH was used as loading control. (C) Quantification of protein expression normalized by GAPDH. (D) Stained with UCP1 and BODIPY in terminally differentiated adipocytes (x200). Data are expressed as mean ± SEM, n ¼ 6.
amount of proteins was subjected for total target assay to quantify phosphorylated AKT, BAD, GSK-3, IRS-1, m-TOR, c-Jun, P70 S6 kinase, PTEN and S6 ribosomal protein using Bio-Plex Pro cell signalling assays. In parallel, b-actin was measured in all samples. After normalization by b-actin, these pathways did not alter between two groups (Figure s2). Phosphorylation of AMPKa has been shown to regulate BAT activity [24]. By western blot, our data demonstrated that phosphorylated AMPKa was 1.4- fold higher in cells treated with PlGF compared with non-treated cells whereas phosphorylation of Akt did not differ between two groups (Fig. 4A and B). To further testify whether PlGF-induced BAT activity was mediated through AMPKa phosphorylation, pre-adipocytes were induced differentiation with 1 ng/mL PlGF in the absence or presence of 1mm pAMPKa inhibitor
Compound C (CC). BAT marker expression was evaluated at the end of differentiation. Western blot illustrated that blockade of AMPKa phosphorylation abrogated PlGF-induced UCP1 and OXPHOS expression in brown adipocytes (Fig. 4C). 3.5. PlGF administration reduced lipid content in the liver and increased brown adipose tissue activity of db/db mice Finally, the effects of PlGF were evaluated in db/db mice. At the age of eight weeks old, db/db mice were implanted mini-pump containing saline or 1.5 mg PlGF. After six weeks of administration, cholesterol levels in the blood and liver were markedly reduced in PlGF-treated mice compared with controls (Figure s3, A and B). Serum triglyceride levels did not differ, its content in liver was
Please cite this article in press as: J. Zhou, et al., Activation of brown adipocytes by placental growth factor, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.08.106
J. Zhou et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
5
Fig. 3. The functionality of differentiated adipocytes. (A) Relative UCP1 mRNA expression of TFAM and NRF1 after treated with 0 or 1 ng PlGF for 8 days. (B) To assess mitochondrial replication, DNAs were extracted from terminally differentiated adipocytes. The mitochondrial copy number was quantified by qPCR. (C) After 24 h of non-treated or PlGFtreated differentiation, cells onto 24-well plate. The O2 consumption was measured with a Seahorse Bioscience XF24 extracellular flux analyser. *p 0.05 and ł p 0.01 when compared with controls. (D) Quantitation of the OCRs from four parts, n ¼ 4.
decreased in PlGF-treated mice compared with controls (Figure s3, C and D). Although fasting blood glucose was similar between two groups, fasting C peptide levels were lower in PlGF-treated mice, indicating improved insulin sensitivity. In addition, the activity of BAT was obviously increased in db/db mice after six weeks of PlGF administration compared with saline control (Figure s4). 4. Discussion The key findings of this study are summarized as follow: (1) Serum PlGF levels were lower in GDM patients than non-diabetic pregnant participants at the end of gestation, which was accompanied with reduced activity of BAT, as evidenced by decreased skin temperature in supraclavicular region detected by infrared
thermography; (2) in vitro, PlGF was able to induce favorable preadipocytes differentiation toward brown adipocytes with BAT function; (3) AMPKa phosphorylation was required for PlGFinduced brown adipocyte activation; and (4) PlGF administration decreased cholesterol and triglyceride contents in the liver and improved insulin sensitivity in db/db mice. T2DM and GDM share some common features of metabolic disorders such as obesity and reduced BAT activity. Nonetheless, T2DM develops chronically and experiences from hyperinsulinemia to insulin deficiency due to beta cell dysfunction. By contrast, beta cell function was comparable between healthy pregnant women and GDM patients but insulin resistance index was much higher in GDM patients than controls (Table 1). When we screened GDM at 24e28 gestation weeks, BMI was already 1.1- fold higher in GDM
Please cite this article in press as: J. Zhou, et al., Activation of brown adipocytes by placental growth factor, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.08.106
6
J. Zhou et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
Fig. 4. Activation of AMPKa by PlGF in differentiated adipocytes. Preadipocytes were differentiated with 0 or 1 ng PlGF as described above. At the end of differentiation, cells were harvested for western blot. (A) Protein expression in differentiated adipocytes. (B) The intensities of immunoblotting bands were expressed as arbitrary units. Data were expressed as mean ± SEM. (CeD) Preadipocytes were subjected for differentiation with 1 mM CC, the AMPKa inhibitor in the absence or presence of 1 ng PlGF. Protein expression of P-AMPKa, AMPKa, UCP1, OXPHOS and GAPDH in differentiated preadipocytes. Data were expressed as mean ± SEM,n ¼ 6.
patients than healthy pregnant subjects. Taken together, these data indicate that GDM is more featured as obesity and insulin resistance rather than impaired beta cell function. Adipose tissue is not only a reservoir of energy but also an endocrine organ for production of hormones, adipokines, cytokines and growth factors. Similarly, the placenta is an endocrine organ besides nourishing fetus. The cross-talk between placenta and white adipose tissue enhances inflammatory cytokine production
and substantially accelerates insulin resistance in GDM [15]. Distinct from white adipose tissue, activation of BAT and beige cells improves glucose homeostasis and mitigates insulin resistance. Previously, we and others have illustrated that transplantation of BAT to high-fat diet mice could reduce obesity and insulin resistance and improve glucose and cholesterol homeostasis [25]. Up to date, the underlying mechanisms why BAT activity is reduced in GDM are not fully understood.
Please cite this article in press as: J. Zhou, et al., Activation of brown adipocytes by placental growth factor, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.08.106
J. Zhou et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
Among all the factors produced by placenta, PlGF is prominently produced during pregnancy. PlGF is a well-known angiogenic factor [26] belonging to VEGF family. Systemic administration of PlGF improved myocardial blood flow and systolic function recovery in a porcine chronic myocardial ischemia model [27]. PlGF is also expressed in human atherosclerotic plaque. Till now, controversy results have been shown to the role of PlGF in plaque progression [28,29] and thus the nature of PlGF in atherosclerosis remains to be defined. Serum levels of PlGF has been reported reduced in the subjects with pregnancy complications which was associated with early pregnant losses [16]. Furthermore, Xu et al. showed that the decrease of PlGF in GDM disrupted the communication between endothelial cells and pancreatic beta cells, resulting in impaired beta cell proliferation [17]. Thus, we initiated this study to explore whether PlGF could have any impact on brown adipocyte activation. Using the simplified in vitro system, our data illustrated the beneficial role of PlGF in the production of functional brown adipocytes. We further demonstrated that PlGF-induced BAT activation was independent of b-adrenergic signalling but required AMPK phosphorylation. Collectively, our results suggest that the decrease of PlGF might contribute to the reduced BAT activity in GDM. The present study must be interpreted within the context of some potential limitations. First, it is not practical to measure serum PlGF levels in the entire pregnant period. Second, PET-CT is the most sensitive method to study the function of BAT, however, it was not feasible for pregnant women in the study. In conclusion, we demonstrated the direct regulation of PlGF on pre-adipocytes differentiation toward brown adipocytes with functionality. Compared to non-diabetic pregnant women, downregulation of PlGF levels might partially contribute to reduced BAT activity in GDM patients. Declaration of interest There is no conflict of interest. Funding This work was supported by The National Natural Science Foundation of China (# 81470566 & 81670765 to Ying-Mei Feng; #81501141 to Jing Zhou; and #81700684 to Chuan-Hai Zhang). Ying-Mei Feng had full access to all the data of the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Author contributions J Zhou, C H Zhang and D Zhao conceived and designed the study; Y M Feng, C H Zhang and R L Yin were involved in the concept and design of the study; C Yan, W Ma and N N Wu were involved in drafting the manuscript; J Zhou, C H Zhang and R L Yin collected the data. Acknowledgements All authors contributed substantially to the study in the aspects of acquisition and/or interpretation of the data. All authors have read and approved the manuscript. Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.bbrc.2018.08.106.
7
Transparency document Transparency document related to this article can be found online at https://doi.org/10.1016/j.bbrc.2018.08.106. References [1] C. Jeppesen, H.T. Maindal, J.K. Kristensen, P.G. Ovesen, D.R. Witte, National study of the prevalence of gestational diabetes mellitus among Danish women from 2004 to 2012, Scand. J. Publ. Health 45 (2017) 811e817. [2] T. Xu, Y. He, L. Dainelli, K. Yu, P. Detzel, I. Silva-Zolezzi, S. Volger, H. Fang, Healthcare interventions for the prevention and control of gestational diabetes mellitus in China: a scoping review, BMC Pregnancy Childbirth 17 (2017) 171. [3] C. Song, Y. Lyu, C. Li, P. Liu, J. Li, R.C. Ma, X. Yang, Long-term risk of diabetes in women at varying durations after gestational diabetes: a systematic review and meta-analysis with more than 2 million women, Obes. Rev. 19 (3) (2018 Mar) 421e429. [4] W. Bao, A. Baecker, Y. Song, M. Kiely, S. Liu, C. Zhang, Adipokine levels during the first or early second trimester of pregnancy and subsequent risk of gestational diabetes mellitus: a systematic review, Metabolism 64 (2015) 756e764. [5] A. Jakubik, Theoretical and methodologic bases of systems in psychiatry, Psychiatr. Pol. 23 (1989) 1e9. [6] M.I. Anghebem-Oliveira, B.R. Martins, D. Alberton, E.A.S. Ramos, G. Picheth, F.G.M. Rego, Type 2 diabetes-associated genetic variants of FTO, LEPR, PPARg, and TCF7L2 in gestational diabetes in a Brazilian population, Arch. Endocrinol. Metabol. 61 (2017) 238e248. [7] P. Yin, P. Shao, H. Liu, W. Li, L. Wang, J. Wang, S. Zhang, J. Leng, N. Li, H. Tian, X. Yang, Z. Yu, G. Hu, C-peptide levels and the risk of diabetes and prediabetes among Chinese women with gestational diabetes, J. Diabet. Complicat. 31 (2017) 1658e1662. [8] M. Bublitz, S. De La Monte, S. Martin, L. Larson, G. Bourjeily, Childhood maltreatment and inflammation among pregnant women with gestational diabetes mellitus: a pilot study, Obstet. Med. 10 (2017) 120e124. [9] N. Nagata, L. Xu, S. Kohno, Y. Ushida, Y. Aoki, R. Umeda, N. Fuke, F. Zhuge, Y. Ni, M. Nagashimada, C. Takahashi, H. Suganuma, S. Kaneko, T. Ota, Glucoraphanin ameliorates obesity and insulin resistance through adipose tissue browning and reduction of metabolic endotoxemia in mice, Diabetes 66 (2017) 1222e1236. [10] S. Kralisch, A. Hoffmann, J. Kratzsch, M. Bluher, M. Stumvoll, M. Fasshauer, T. Ebert, The Brown-fat-secreted Adipokine Neuregulin 4 Is Decreased in Gestational Diabetes Mellitus, Diabetes Metab, 2017. [11] M. Thuzar, W.P. Law, J. Ratnasingam, C. Jang, G. Dimeski, K.K.Y. Ho, Glucocorticoids Suppress Brown Adipose Tissue Function in Humans: a Doubleblind Placebo-controlled Study, Diabetes Obes Metab, 2017. [12] T.J. Bartness, C.H. Vaughan, C.K. Song, Sympathetic and sensory innervation of brown adipose tissue, Int. J. Obes. (Lond) 34 (Suppl 1) (2010) S36eS42. [13] Y. Zhang, H. Guo, J.A. Deis, M.G. Mashek, M. Zhao, D. Ariyakumar, A.G. Armien, D.A. Bernlohr, D.G. Mashek, X. Chen, Lipocalin 2 regulates brown fat activation via a nonadrenergic activation mechanism, J. Biol. Chem. 289 (2014) 22063e22077. [14] J.E. Campbell, D.J. Drucker, Pharmacology, physiology, and mechanisms of incretin hormone action, Cell Metabol. 17 (2013) 819e837. [15] N. Jayabalan, S. Nair, Z. Nuzhat, G.E. Rice, F.A. Zuniga, L. Sobrevia, A. Leiva, C. Sanhueza, J.A. Gutierrez, M. Lappas, D.J. Freeman, C. Salomon, Cross talk between adipose tissue and placenta in obese and gestational diabetes mellitus pregnancies via exosomes, Front. Endocrinol. (Lausanne) 8 (2017) 239. [16] H.R. Nejabati, Z. Latifi, T. Ghasemnejad, A. Fattahi, M. Nouri, Placental growth factor (PlGF) as an angiogenic/inflammatory switcher: lesson from early pregnancy losses, Gynecol. Endocrinol. 33 (2017) 668e674. [17] X. Xu, J. Shen, Reduction in placental growth factor impaired gestational betacell proliferation through crosstalk between beta-cells and islet endothelial cells, Am. J. Trans. Res. 8 (2016) 3912e3920. [18] N.N. Wu, C.H. Zhang, H.J. Lee, Y. Ma, X. Wang, X.J. Ma, W. Ma, D. Zhao, Y.M. Feng, Brown adipogenic potential of brown adipocytes and peri-renal adipocytes from human embryo, Sci. Rep. 6 (2016) 39193. [19] B.E. Metzger, D.R. Coustan, Summary and recommendations of the fourth international workshop-conference on gestational diabetes mellitus. The organizing committee, Diabetes Care 21 (Suppl 2) (1998) B161eB167. [20] Y.M. Feng, D. Zhao, N. Zhang, C.G. Yu, Q. Zhang, L. Thijs, J.A. Staessen, Insulin resistance in relation to lipids and inflammation in Type-2 diabetic patients and non-diabetic people, PLoS One 11 (2016), e0153171. [21] R. Johnson, P. McNutt, S. MacMahon, R. Robson, Use of the Friedewald formula to estimate LDL-cholesterol in patients with chronic renal failure on dialysis, Clin. Chem. 43 (1997) 2183e2184. [22] C. Jang, S. Jalapu, M. Thuzar, P.W. Law, S. Jeavons, J.L. Barclay, K.K. Ho, Infrared thermography in the detection of brown adipose tissue in humans, Phys. Rep. 2 (2014). [23] S. Van de Veire, I. Stalmans, F. Heindryckx, H. Oura, A. Tijeras-Raballand, T. Schmidt, S. Loges, I. Albrecht, B. Jonckx, S. Vinckier, C. Van Steenkiste, S. Tugues, C. Rolny, M. De Mol, D. Dettori, P. Hainaud, L. Coenegrachts, J.O. Contreres, T. Van Bergen, H. Cuervo, W.H. Xiao, C. Le Henaff, I. Buysschaert,
Please cite this article in press as: J. Zhou, et al., Activation of brown adipocytes by placental growth factor, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.08.106
8
J. Zhou et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
B. Kharabi Masouleh, A. Geerts, T. Schomber, P. Bonnin, V. Lambert, J. Haustraete, S. Zacchigna, J.M. Rakic, W. Jimenez, A. Noel, M. Giacca, I. Colle, J.M. Foidart, G. Tobelem, M. Morales-Ruiz, J. Vilar, P. Maxwell, S.A. Vinores, G. Carmeliet, M. Dewerchin, L. Claesson-Welsh, E. Dupuy, H. Van Vlierberghe, G. Christofori, M. Mazzone, M. Detmar, D. Collen, P. Carmeliet, Further pharmacological and genetic evidence for the efficacy of PlGF inhibition in cancer and eye disease, Cell 141 (2010) 178e190. [24] S. Wang, X. Liang, Q. Yang, X. Fu, M. Zhu, B.D. Rodgers, Q. Jiang, M.V. Dodson, M. Du, Resveratrol enhances brown adipocyte formation and function by activating AMP-activated protein kinase (AMPK) alpha1 in mice fed high-fat diet, Mol. Nutr. Food Res. 61 (2017). [25] X. Liu, S. Wang, Y. You, M. Meng, Z. Zheng, M. Dong, J. Lin, Q. Zhao, C. Zhang, X. Yuan, T. Hu, L. Liu, Y. Huang, L. Zhang, D. Wang, J. Zhan, H. Jong Lee, J.R. Speakman, W. Jin, Brown adipose tissue transplantation reverses obesity in Ob/Ob mice, Endocrinology 156 (2015) 2461e2469.
[26] A. Luttun, M. Tjwa, P. Carmeliet, Placental growth factor (PlGF) and its receptor Flt-1 (VEGFR-1): novel therapeutic targets for angiogenic disorders, Ann. N. Y. Acad. Sci. 979 (2002) 80e93. [27] X. Liu, P. Claus, M. Wu, G. Reyns, P. Verhamme, P. Pokreisz, S. Vandenwijngaert, C. Dubois, J. Vanhaecke, E. Verbeken, J. Bogaert, S. Janssens, Placental growth factor increases regional myocardial blood flow and contractile function in chronic myocardial ischemia, Am. J. Physiol. Heart Circ. Physiol. 304 (2013) H885eH894. [28] R. Khurana, L. Moons, S. Shafi, A. Luttun, D. Collen, J.F. Martin, P. Carmeliet, I.C. Zachary, Placental growth factor promotes atherosclerotic intimal thickening and macrophage accumulation, Circulation 111 (2005) 2828e2836. [29] M. Wu, P. Pokreisz, M. Swinnen, E. Caluwe, H. Gillijns, N. Vanden Driessche, A. Casazza, E. Verbeken, D. Collen, S. Janssens, Sustained placental growth Factor-2 treatment does not aggravate advanced atherosclerosis in ischemic cardiomyopathy, J. Cardiovasc. Trans. Res. 10 (2017) 348e358.
Please cite this article in press as: J. Zhou, et al., Activation of brown adipocytes by placental growth factor, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.08.106
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Activation of brown adipocytes by placental growth factor Jing Zhou a, 1, Nan-Nan Wu a, 1, Rui-Li Yin a, 1, Wei Ma b, Cen Yan a, Ying-Mei Feng a, Chuan-Hai Zhang c, **, Dong Zhao a, * a
Beijing Key Laboratory of Diabetes Prevention and Research, Endocrinology Centre, Lu He Hospital, Capital Medical University, Beijing, 101149, China2 Gynecology and Obstetrics Centre, Lu He Hospital, Capital Medical University, Beijing, 101149, China2 c Beijing Advanced Innovation Centre for Food Nutrition and Human Health, College Food Science and Nutritional Engineering, China Agricultural University, Beijing, 100083, China3
b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 8 August 2018 Accepted 14 August 2018 Available online xxx
Gestational diabetes mellitus (GDM) is a type of diabetes and occurs during pregnancy. Brown adipose tissue (BAT) improves glucose homeostasis and mitigates insulin resistance, however, its activity is reduced in GDM. Placenta growth factor (PlGF) is an angiogenic factor produced by placental trophoblasts. Nevertheless, whether and how PlGF could affect BAT function in GDM are not defined. To investigate this question, 91 non-diabetic pregnant participants and 73 GDM patients were recruited to Gynaecology and Obstetrics Centre in Lu He hospital. Serum levels of PlGF were quantified by ELISA. Skin temperature was measured by far infrared thermography in the supraclavicular region where classical BATs were located. The direct effect of PlGF on BAT function was explored using the established human preadipocyte differentiation system. Thereby, we demonstrated that serum levels of PlGF were lower in GDM patients compared with controls, which was accompanied by decreased skin temperature in the supraclavicular region. By qPCR and western blot, mRNA and protein expression of UCP1 and OXPHOS were elevated in differentiated adipocytes treated with PlGF. PlGF stimulated mitochondrion transcription and increased copy number of mitochondrial. When subjected for respirometry, PlGF-treated differentiated adipocytes showed higher oxygen consumption rates than controls. PlGF induced AMPK phosphorylation and blockade of AMPK phosphorylation blunted UCP1 and OXPHOS expression in differentiated adipocytes. PlGF administration reduced cholesterol and triglyceride content in the liver and improved insulin sensitivity in db mice compared with control. In Conclusion, PlGF could activate BAT function. Downregulation of PlGF might contribute to the reduced BAT activity in GDM. © 2018 Published by Elsevier Inc.
Keywords: Gestational diabetes mellitus Brown adipose tissue Placenta growth factor
1. Introduction Gestational diabetes mellitus (GDM) is a special type of diabetes and mostly occurs after 24th gestational week. Over the years, the prevalence of GDM is approximately 9.3e25.5% worldwide and 3.9e18.9% in China [1,2]. GDM is not only associated with pre- and perinatal complications but also exposes high risk of type 2 diabetes mellitus (T2DM) for mothers after pregnancy [3]. Although the aetiology of GDM is not yet clear, it shares common features
* Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (C.-H. Zhang), (D. Zhao). 1 Equal contribution to the study. 2 Tel.: þ86 10 6954 3901; fax: þ86 10 6953 1069 3 Tel.: þ86 10 6273 7751 3901; fax: þ86 10 6273 7751
[email protected]
with T2DM such as adiposity [4], insulin resistance [5,6] and inflammation [7,8]. Activation of brown adipocytes and beige cells improves glucose homeostasis and attenuates excessive nutrientsinduced obesity and insulin resistance [9]. Unfortunately, BAT activity is decreased in both T2DM and GDM patients [10,11]. Norepinephrine is the classical stimulator for brown adipocyte and beige cell activation. Binding norepinephrine to b-adrenergic receptor adipocytes induce AMPK phosphorylation that in turn upregulates UCP1 expression for mitochondrial biogenesis and thermogenesis [12]. Except norepinephrine, brown fat activation could be activated by adipokines such as lipocalin 2 [13] and hormones such like incretin [14]. Placental trophoblasts secret a large amount of proteins, growth factors, cytokines and hormones by which the cross-talk between placenta and peripheral tissues is established [15]. Nevertheless, how placenta communicates with BAT remains unknown.
https://doi.org/10.1016/j.bbrc.2018.08.106 0006-291X/© 2018 Published by Elsevier Inc.
Please cite this article in press as: J. Zhou, et al., Activation of brown adipocytes by placental growth factor, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.08.106
2
J. Zhou et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
Placenta growth factor (PlGF) is a member of vascular endothelial growth factor (VEGF) family and mainly produced by trophoblasts in placenta. PlGF levels are reported reduced in subjects with pregnancy complications, which is associated with early pregnant losses [16]. In line with this study, downregulation of PlGF in GDM renders the physical interaction between beta cells and endothelial cells in islets, resulting in impaired beta cells proliferation [17]. We hypothesized that PlGF could be the messenger between BAT function and placenta. To investigate this question, we compared serum PlGF levels in healthy pregnant subjects and GDM patients. We studied the effect of PlGF on BAT marker expression and function in preadipocytes isolated from human embryos [18]. 2. Materials and methods 2.1. Human study This case-control study was performed among pregnant women who admitted in the Gynaecology and Obstetrics Centre in Lu He hospital between November 2015 till December 2016. The subjects who were diagnosed preeclampsia, retinopathy or nephropathy were excluded. Finally, 91 non-diabetic pregnant participants and 73 GDM patients were analysed in the study. Our study complied with the Helsinki Declaration for investigation of human subjects. The study protocol was obtained ethical approval from the competent Institutional Review Boards of Capital Medical University. All participants provided written informed consent. Screening of GDM was performed at 24e28 gestation weeks by an oral glucose tolerance test. The criteria of GDM diagnosis was adapted from American Diabetes Association guideline as following: 1) fasting blood glucose levels 5.3 mmol/L or 2) 1-h oral 100-g glucose tolerance test 10.0 mmol/L [19]. General information including age, disease history and medications was obtained. Body mass index (BMI) at 24e28 gestation weeks were obtained by weight in kilograms divided by the square of height in meters (kg/m2).
study. They were randomized and blindly assigned to receive totalled 1.5 mg murine recombinant PlGF (RD, Bio-Techne, USA) for 4 weeks or equivalent volume of saline through a subcutaneous implanted osmotic minipump (Durect, Cupertino, CA, USA) as described before [23]. At sacrifice, after overnight fasting, blood was bled and liver and fat tissues were dissected. 2.5. Cell culture Primary human embryo adipocytes were isolated from human embryos during abortion and cultivated as described before [18]. Once they reached 100% confluence, cells were induced adipogenic differentiation in the absence or presence of recombinant human PlGF (0e1 ng/mL) for 8 days. Medium was refreshed every two days. For inhibitor experiments, cells were differentiated with 1 mM AMPK phosphorylation inhibitor-Compound C. The detailed information of measurement of oxygen consumption and Phosphorylation pathways using Luminex technology were shown in Supplementary Materials and methods. 2.6. Real-time PCR and western blotting Real-time PCR and western blotting were performed as described previously. The primer sequences are listed in Supplementary Table 1. The specific primary antibodies were shown in Supplementary Materials. 2.7. Immunofluorescence staining Differentiated adipocytes were stained with 1 mg/ml antihuman UCP1, followed by Alexa 488-conjugated secondary antibody (Invitrogen), BODIPY (Thermo) and DAPI (Leagene) complying with the procedure. Brown adipocytes were positive for both UCP1 and BODIPY. Negative controls were stained with omission of primary antibody. Images were taken by Zeiss laser scanning confocal microscopy (LSM780, Germany).
2.2. Clinical measurement
2.8. ELISA
At the antenatal examination, blood pressure was recorded by auscultation of the Korotkoff sounds, using a standard mercury sphygmomanometer. Fasting blood samples were collected and glucose, insulin, total cholesterol, triglyceride and HDL-cholesterol (HDL-c) and creatinine were measured in the central laboratory in the hospital [20]. LDL-cholesterol (LDL-c) was computed from serum total cholesterol and HDL-c and serum triglycerides by the Friedewald equation [21].
Serum levels of PlGF were quantified by enzyme-linked immunosorbent assay (ELISA) according to the manual (MLBio, Shanghai, China).
2.3. Far infrared thermography Because PET-CT could not be applied to measure BAT activity in pregnant women, infrared thermography was used alternatively. As described before [22], the study subject was seated in the upright position with head positioned in a neutral position and the subject looked straight ahead. Images were acquired by a thermal imaging camera (FLIR B425, 3.1Mpixel, FLIR Systems Australia Pty Ltd, Melbourne, Vic., Australia) in the anterior neck and upper chest. Skin temperature in the supraclavicular region where classical BATs were located was analysed by two experienced investigators who were blind to the study using the software (Version 1.2, Wilsonville, OR). 2.4. Mice study Diabetic db mice at the age of 8 weeks old were used in the
2.9. Statistical analysis Data were expressed as mean ± SEM. In the experiments where there were 2 experimental groups, unpaired, 2-tailed Student's ttest was used for data with normal distribution. For data that did not fit normal distribution, nonparametric Mann Whitney analysis was used. In the experiments containing more than 2 experimental groups, one-way ANOVA with Dunnett was used when comparing treated groups against control and ANOVA with Bonferroni was applied when comparing all groups. Statistics analysis was performed using GraphPad Prism (GraphPad Software Inc, La Jolla, CA, U.S.A.). A P value less than 0.05 was considered significant. 3. Results 3.1. Serum PlGF levels were reduced in GDM patients At 24e28 gestation weeks when GDM was diagnosed, BMI was already higher in GDM patients than healthy pregnant subjects (26.24 ± 0.64 kg/m2 vs 24.38 ± 0.87 kg/m2, p ¼ 0.09). General characterization of the study subjects at the antenatal visit was listed in Table 1. Compared with non-diabetic pregnant controls, GDM
Please cite this article in press as: J. Zhou, et al., Activation of brown adipocytes by placental growth factor, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.08.106
J. Zhou et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
3
Table 1 General characterization of the study subjects.
N Age (years) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Fasting blood glucose (mmol/L) Fasting insulin (pmol/L) Haemoglobin (g/L) Total cholesterol (mmol/L) Total triglyceride (mmol/L) LDL-cholesterol (mmol/L) HDL-cholesterol (mmol/L) Homa-b (%) Homa-IR (%)
Non-diabetic
Diabetic
p
91 30.0 (0.6) 117.2 (1.2) 72.3 (1.0) 4.57 (0.05) 15.4 (0.9) 120.6 (1.7) 5.38 (0.13) 1.57 (0.09) 2.49 (0.47) 1.82 (0.16) 184.1 (15.3) 1.68 (0.16)
73 31.6 (0.6) 118.4 (2.1) 76.3 (1.8) 5.11 (0.13) 20.8 (3.4) 116.7 (2.7) 5.45 (0.20) 1.98 (0.3) 2.64 (0.17) 1.86 (0.11) 199.5 (25.7) 2.81 (0.39)
0.06 0.58 0.13 <0.0001 0.04 0.21 0.78 0.07 0.40 0.82 0.59 0.003
Data are expressed by mean ± SEM. LDL, low-density lipoprotein; HDL, high-density lipoprotein; Homa-b and Homa-IR were computed by Homeostasis Model Assessment algorithm (http://www.dtu.ox.ac.uk/homacalculator/) using fasting insulin and fasting blood glucose.
patients were older (p ¼ 0.06) and had higher fasting glucose and insulin levels in the peripheral blood (p < 0.05 for both). The index of insulin resistance was 1.7- fold greater in GDM patients than controls (p ¼ 0.003). Quantified by ELISA, serum PlGF levels were 381.8 (42.5) pg/mL in non-diabetic pregnant participants but reduced to 296.7 (17.16) pg/mL in GDM patients (p < 0.05) (Fig. 1A). Quantified by Far infrared thermography, the skin temperature in the supraclavicular region was lower in GDM patients than controls, indicating reduced BAT activity (Fig. 1B and C). 3.2. PlGF induced activation of human brown adipocytes Previously, we have shown that cells in SVF fraction contained enriched brown preadipocytes that could differentiate into brown adipocytes under desirable condition [18]. To explore whether PlGF had any direct effect on brown adipocyte activation, stromal vascular fraction was isolated from intracapsular BAT of human embryos and cultivated in vitro. Cells were differentiated under PlGF concentrations (0e1 ng/mL) for 8 days. By qPCR, PlGF treatment increased mRNA expression of UCP1, PGC1a, PGC1b, CPT1a and CTP1b in adipocytes in a dose-dependent manner and peaked between 0.5 ng/mL and 1 ng/mL PlGF (Figure s1, A-E). However, PlGF did not alter mRNA expression of PPAR (Figure s1F). Next, we evaluated the protein expression profile in brown adipocytes exposed to PlGF. Since we observed the dose-dependent effect of PlGF on BAT activation, 1 ng/mL PlGF was used in the following experiments. Consistent with previous findings, PlGF stimulated a series of thermogenic gene expression including UCP1, PGC1a, PGC1b, CPT1a and CPT1b (Fig. 2A) in cultivated adipocytes at mRNA levels. In line with increased mRNA expression, protein expression of UCP1 and mitochondrial-specific OXPHOS protein, including ATP5A, ubiquinol-cytochrome C reductase core protein II (UQCRC2), succinate dehydrogenase complex and subunit B (SDHB) and NADH dehydrogenase (ubiquinone) 1b subcomplex (NDUFB8) were significantly elevated in cells treated with PlGF (Fig. 2B and C). The increased UCP1 expression in PlGF-treated adipocytes was further confirmed by dual-immunofluorescence staining (Fig. 2D). 3.3. Mitochondrial function in brown adipocytes induced by PlGF Quantified by qPCR, mRNA levels of mitochondrial transcription factor A (TFAM) and nuclear respiratory factor 1 (NRF-1) were elevated in PlGF-treated cells, suggesting enhanced mitochondrion replication (Fig. 3A). Indeed, the increased mitochondrial replication by PlGF was further confirmed by quantification of
Fig. 1. Serum levels of PlGF and BAT activity in GDM. (A) Quantification of serum levels of PlGF in non-diabetic pregnant participants (n ¼ 58) and GDM patients (n ¼ 76). (B) Skin temperature in the supraclavicular region was analysed by the software. (C) Images were acquired by a thermal imaging camera in the anterior neck and upper chest. The circles indicated the positions where skin temperature was measured. Data are expressed as mean ± SEM.
mitochondrial copy number by qPCR (Fig. 3B). To further examine the function of differentiated brown adipocytes, oxygen consumption rates (OCR) were assessed by respirometry. Terminally differentiated adipocytes were seeded in 96well plate to measure basal respiration and ATP turnover (Fig. 3C). Quantification of the area under the curve (AUC) showed that the basal level of respiration rates was 1.8- fold higher in PlGFtreated cells than non-treated ones (Fig. 3D). Accordingly, the leak respiration rates of PlGF-treated cells were 3.5- fold higher than non-treated ones (Fig. 3D). 3.4. PlGF stimulated AMPKa phosphorylation in brown adipocytes To explore the signalling pathways underlying the activation of brown adipocytes by PlGF, proteins were extracted from differentiated brown adipocytes treated without or with PlGF. Equal
Please cite this article in press as: J. Zhou, et al., Activation of brown adipocytes by placental growth factor, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.08.106
4
J. Zhou et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
Fig. 2. Thermogenic proteins expression in differentiated adipocytes. Preadipocytes were differentiated with 1 ng PlGF for 8 days. (A) Relative mRNA expression of UCP1, PGC1a, PGC1b, CPTa and CPTb in differentiated adipocytes. (B) The protein expression of UCP1, ATP5A, UQCRC2, SDHB and NDUFB8 in differentiated adipocytes. GAPDH was used as loading control. (C) Quantification of protein expression normalized by GAPDH. (D) Stained with UCP1 and BODIPY in terminally differentiated adipocytes (x200). Data are expressed as mean ± SEM, n ¼ 6.
amount of proteins was subjected for total target assay to quantify phosphorylated AKT, BAD, GSK-3, IRS-1, m-TOR, c-Jun, P70 S6 kinase, PTEN and S6 ribosomal protein using Bio-Plex Pro cell signalling assays. In parallel, b-actin was measured in all samples. After normalization by b-actin, these pathways did not alter between two groups (Figure s2). Phosphorylation of AMPKa has been shown to regulate BAT activity [24]. By western blot, our data demonstrated that phosphorylated AMPKa was 1.4- fold higher in cells treated with PlGF compared with non-treated cells whereas phosphorylation of Akt did not differ between two groups (Fig. 4A and B). To further testify whether PlGF-induced BAT activity was mediated through AMPKa phosphorylation, pre-adipocytes were induced differentiation with 1 ng/mL PlGF in the absence or presence of 1mm pAMPKa inhibitor
Compound C (CC). BAT marker expression was evaluated at the end of differentiation. Western blot illustrated that blockade of AMPKa phosphorylation abrogated PlGF-induced UCP1 and OXPHOS expression in brown adipocytes (Fig. 4C). 3.5. PlGF administration reduced lipid content in the liver and increased brown adipose tissue activity of db/db mice Finally, the effects of PlGF were evaluated in db/db mice. At the age of eight weeks old, db/db mice were implanted mini-pump containing saline or 1.5 mg PlGF. After six weeks of administration, cholesterol levels in the blood and liver were markedly reduced in PlGF-treated mice compared with controls (Figure s3, A and B). Serum triglyceride levels did not differ, its content in liver was
Please cite this article in press as: J. Zhou, et al., Activation of brown adipocytes by placental growth factor, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.08.106
J. Zhou et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
5
Fig. 3. The functionality of differentiated adipocytes. (A) Relative UCP1 mRNA expression of TFAM and NRF1 after treated with 0 or 1 ng PlGF for 8 days. (B) To assess mitochondrial replication, DNAs were extracted from terminally differentiated adipocytes. The mitochondrial copy number was quantified by qPCR. (C) After 24 h of non-treated or PlGFtreated differentiation, cells onto 24-well plate. The O2 consumption was measured with a Seahorse Bioscience XF24 extracellular flux analyser. *p 0.05 and ł p 0.01 when compared with controls. (D) Quantitation of the OCRs from four parts, n ¼ 4.
decreased in PlGF-treated mice compared with controls (Figure s3, C and D). Although fasting blood glucose was similar between two groups, fasting C peptide levels were lower in PlGF-treated mice, indicating improved insulin sensitivity. In addition, the activity of BAT was obviously increased in db/db mice after six weeks of PlGF administration compared with saline control (Figure s4). 4. Discussion The key findings of this study are summarized as follow: (1) Serum PlGF levels were lower in GDM patients than non-diabetic pregnant participants at the end of gestation, which was accompanied with reduced activity of BAT, as evidenced by decreased skin temperature in supraclavicular region detected by infrared
thermography; (2) in vitro, PlGF was able to induce favorable preadipocytes differentiation toward brown adipocytes with BAT function; (3) AMPKa phosphorylation was required for PlGFinduced brown adipocyte activation; and (4) PlGF administration decreased cholesterol and triglyceride contents in the liver and improved insulin sensitivity in db/db mice. T2DM and GDM share some common features of metabolic disorders such as obesity and reduced BAT activity. Nonetheless, T2DM develops chronically and experiences from hyperinsulinemia to insulin deficiency due to beta cell dysfunction. By contrast, beta cell function was comparable between healthy pregnant women and GDM patients but insulin resistance index was much higher in GDM patients than controls (Table 1). When we screened GDM at 24e28 gestation weeks, BMI was already 1.1- fold higher in GDM
Please cite this article in press as: J. Zhou, et al., Activation of brown adipocytes by placental growth factor, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.08.106
6
J. Zhou et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
Fig. 4. Activation of AMPKa by PlGF in differentiated adipocytes. Preadipocytes were differentiated with 0 or 1 ng PlGF as described above. At the end of differentiation, cells were harvested for western blot. (A) Protein expression in differentiated adipocytes. (B) The intensities of immunoblotting bands were expressed as arbitrary units. Data were expressed as mean ± SEM. (CeD) Preadipocytes were subjected for differentiation with 1 mM CC, the AMPKa inhibitor in the absence or presence of 1 ng PlGF. Protein expression of P-AMPKa, AMPKa, UCP1, OXPHOS and GAPDH in differentiated preadipocytes. Data were expressed as mean ± SEM,n ¼ 6.
patients than healthy pregnant subjects. Taken together, these data indicate that GDM is more featured as obesity and insulin resistance rather than impaired beta cell function. Adipose tissue is not only a reservoir of energy but also an endocrine organ for production of hormones, adipokines, cytokines and growth factors. Similarly, the placenta is an endocrine organ besides nourishing fetus. The cross-talk between placenta and white adipose tissue enhances inflammatory cytokine production
and substantially accelerates insulin resistance in GDM [15]. Distinct from white adipose tissue, activation of BAT and beige cells improves glucose homeostasis and mitigates insulin resistance. Previously, we and others have illustrated that transplantation of BAT to high-fat diet mice could reduce obesity and insulin resistance and improve glucose and cholesterol homeostasis [25]. Up to date, the underlying mechanisms why BAT activity is reduced in GDM are not fully understood.
Please cite this article in press as: J. Zhou, et al., Activation of brown adipocytes by placental growth factor, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.08.106
J. Zhou et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
Among all the factors produced by placenta, PlGF is prominently produced during pregnancy. PlGF is a well-known angiogenic factor [26] belonging to VEGF family. Systemic administration of PlGF improved myocardial blood flow and systolic function recovery in a porcine chronic myocardial ischemia model [27]. PlGF is also expressed in human atherosclerotic plaque. Till now, controversy results have been shown to the role of PlGF in plaque progression [28,29] and thus the nature of PlGF in atherosclerosis remains to be defined. Serum levels of PlGF has been reported reduced in the subjects with pregnancy complications which was associated with early pregnant losses [16]. Furthermore, Xu et al. showed that the decrease of PlGF in GDM disrupted the communication between endothelial cells and pancreatic beta cells, resulting in impaired beta cell proliferation [17]. Thus, we initiated this study to explore whether PlGF could have any impact on brown adipocyte activation. Using the simplified in vitro system, our data illustrated the beneficial role of PlGF in the production of functional brown adipocytes. We further demonstrated that PlGF-induced BAT activation was independent of b-adrenergic signalling but required AMPK phosphorylation. Collectively, our results suggest that the decrease of PlGF might contribute to the reduced BAT activity in GDM. The present study must be interpreted within the context of some potential limitations. First, it is not practical to measure serum PlGF levels in the entire pregnant period. Second, PET-CT is the most sensitive method to study the function of BAT, however, it was not feasible for pregnant women in the study. In conclusion, we demonstrated the direct regulation of PlGF on pre-adipocytes differentiation toward brown adipocytes with functionality. Compared to non-diabetic pregnant women, downregulation of PlGF levels might partially contribute to reduced BAT activity in GDM patients. Declaration of interest There is no conflict of interest. Funding This work was supported by The National Natural Science Foundation of China (# 81470566 & 81670765 to Ying-Mei Feng; #81501141 to Jing Zhou; and #81700684 to Chuan-Hai Zhang). Ying-Mei Feng had full access to all the data of the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Author contributions J Zhou, C H Zhang and D Zhao conceived and designed the study; Y M Feng, C H Zhang and R L Yin were involved in the concept and design of the study; C Yan, W Ma and N N Wu were involved in drafting the manuscript; J Zhou, C H Zhang and R L Yin collected the data. Acknowledgements All authors contributed substantially to the study in the aspects of acquisition and/or interpretation of the data. All authors have read and approved the manuscript. Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.bbrc.2018.08.106.
7
Transparency document Transparency document related to this article can be found online at https://doi.org/10.1016/j.bbrc.2018.08.106. References [1] C. Jeppesen, H.T. Maindal, J.K. Kristensen, P.G. Ovesen, D.R. Witte, National study of the prevalence of gestational diabetes mellitus among Danish women from 2004 to 2012, Scand. J. Publ. Health 45 (2017) 811e817. [2] T. Xu, Y. He, L. Dainelli, K. Yu, P. Detzel, I. Silva-Zolezzi, S. Volger, H. Fang, Healthcare interventions for the prevention and control of gestational diabetes mellitus in China: a scoping review, BMC Pregnancy Childbirth 17 (2017) 171. [3] C. Song, Y. Lyu, C. Li, P. Liu, J. Li, R.C. Ma, X. Yang, Long-term risk of diabetes in women at varying durations after gestational diabetes: a systematic review and meta-analysis with more than 2 million women, Obes. Rev. 19 (3) (2018 Mar) 421e429. [4] W. Bao, A. Baecker, Y. Song, M. Kiely, S. Liu, C. Zhang, Adipokine levels during the first or early second trimester of pregnancy and subsequent risk of gestational diabetes mellitus: a systematic review, Metabolism 64 (2015) 756e764. [5] A. Jakubik, Theoretical and methodologic bases of systems in psychiatry, Psychiatr. Pol. 23 (1989) 1e9. [6] M.I. Anghebem-Oliveira, B.R. Martins, D. Alberton, E.A.S. Ramos, G. Picheth, F.G.M. Rego, Type 2 diabetes-associated genetic variants of FTO, LEPR, PPARg, and TCF7L2 in gestational diabetes in a Brazilian population, Arch. Endocrinol. Metabol. 61 (2017) 238e248. [7] P. Yin, P. Shao, H. Liu, W. Li, L. Wang, J. Wang, S. Zhang, J. Leng, N. Li, H. Tian, X. Yang, Z. Yu, G. Hu, C-peptide levels and the risk of diabetes and prediabetes among Chinese women with gestational diabetes, J. Diabet. Complicat. 31 (2017) 1658e1662. [8] M. Bublitz, S. De La Monte, S. Martin, L. Larson, G. Bourjeily, Childhood maltreatment and inflammation among pregnant women with gestational diabetes mellitus: a pilot study, Obstet. Med. 10 (2017) 120e124. [9] N. Nagata, L. Xu, S. Kohno, Y. Ushida, Y. Aoki, R. Umeda, N. Fuke, F. Zhuge, Y. Ni, M. Nagashimada, C. Takahashi, H. Suganuma, S. Kaneko, T. Ota, Glucoraphanin ameliorates obesity and insulin resistance through adipose tissue browning and reduction of metabolic endotoxemia in mice, Diabetes 66 (2017) 1222e1236. [10] S. Kralisch, A. Hoffmann, J. Kratzsch, M. Bluher, M. Stumvoll, M. Fasshauer, T. Ebert, The Brown-fat-secreted Adipokine Neuregulin 4 Is Decreased in Gestational Diabetes Mellitus, Diabetes Metab, 2017. [11] M. Thuzar, W.P. Law, J. Ratnasingam, C. Jang, G. Dimeski, K.K.Y. Ho, Glucocorticoids Suppress Brown Adipose Tissue Function in Humans: a Doubleblind Placebo-controlled Study, Diabetes Obes Metab, 2017. [12] T.J. Bartness, C.H. Vaughan, C.K. Song, Sympathetic and sensory innervation of brown adipose tissue, Int. J. Obes. (Lond) 34 (Suppl 1) (2010) S36eS42. [13] Y. Zhang, H. Guo, J.A. Deis, M.G. Mashek, M. Zhao, D. Ariyakumar, A.G. Armien, D.A. Bernlohr, D.G. Mashek, X. Chen, Lipocalin 2 regulates brown fat activation via a nonadrenergic activation mechanism, J. Biol. Chem. 289 (2014) 22063e22077. [14] J.E. Campbell, D.J. Drucker, Pharmacology, physiology, and mechanisms of incretin hormone action, Cell Metabol. 17 (2013) 819e837. [15] N. Jayabalan, S. Nair, Z. Nuzhat, G.E. Rice, F.A. Zuniga, L. Sobrevia, A. Leiva, C. Sanhueza, J.A. Gutierrez, M. Lappas, D.J. Freeman, C. Salomon, Cross talk between adipose tissue and placenta in obese and gestational diabetes mellitus pregnancies via exosomes, Front. Endocrinol. (Lausanne) 8 (2017) 239. [16] H.R. Nejabati, Z. Latifi, T. Ghasemnejad, A. Fattahi, M. Nouri, Placental growth factor (PlGF) as an angiogenic/inflammatory switcher: lesson from early pregnancy losses, Gynecol. Endocrinol. 33 (2017) 668e674. [17] X. Xu, J. Shen, Reduction in placental growth factor impaired gestational betacell proliferation through crosstalk between beta-cells and islet endothelial cells, Am. J. Trans. Res. 8 (2016) 3912e3920. [18] N.N. Wu, C.H. Zhang, H.J. Lee, Y. Ma, X. Wang, X.J. Ma, W. Ma, D. Zhao, Y.M. Feng, Brown adipogenic potential of brown adipocytes and peri-renal adipocytes from human embryo, Sci. Rep. 6 (2016) 39193. [19] B.E. Metzger, D.R. Coustan, Summary and recommendations of the fourth international workshop-conference on gestational diabetes mellitus. The organizing committee, Diabetes Care 21 (Suppl 2) (1998) B161eB167. [20] Y.M. Feng, D. Zhao, N. Zhang, C.G. Yu, Q. Zhang, L. Thijs, J.A. Staessen, Insulin resistance in relation to lipids and inflammation in Type-2 diabetic patients and non-diabetic people, PLoS One 11 (2016), e0153171. [21] R. Johnson, P. McNutt, S. MacMahon, R. Robson, Use of the Friedewald formula to estimate LDL-cholesterol in patients with chronic renal failure on dialysis, Clin. Chem. 43 (1997) 2183e2184. [22] C. Jang, S. Jalapu, M. Thuzar, P.W. Law, S. Jeavons, J.L. Barclay, K.K. Ho, Infrared thermography in the detection of brown adipose tissue in humans, Phys. Rep. 2 (2014). [23] S. Van de Veire, I. Stalmans, F. Heindryckx, H. Oura, A. Tijeras-Raballand, T. Schmidt, S. Loges, I. Albrecht, B. Jonckx, S. Vinckier, C. Van Steenkiste, S. Tugues, C. Rolny, M. De Mol, D. Dettori, P. Hainaud, L. Coenegrachts, J.O. Contreres, T. Van Bergen, H. Cuervo, W.H. Xiao, C. Le Henaff, I. Buysschaert,
Please cite this article in press as: J. Zhou, et al., Activation of brown adipocytes by placental growth factor, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.08.106
8
J. Zhou et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
B. Kharabi Masouleh, A. Geerts, T. Schomber, P. Bonnin, V. Lambert, J. Haustraete, S. Zacchigna, J.M. Rakic, W. Jimenez, A. Noel, M. Giacca, I. Colle, J.M. Foidart, G. Tobelem, M. Morales-Ruiz, J. Vilar, P. Maxwell, S.A. Vinores, G. Carmeliet, M. Dewerchin, L. Claesson-Welsh, E. Dupuy, H. Van Vlierberghe, G. Christofori, M. Mazzone, M. Detmar, D. Collen, P. Carmeliet, Further pharmacological and genetic evidence for the efficacy of PlGF inhibition in cancer and eye disease, Cell 141 (2010) 178e190. [24] S. Wang, X. Liang, Q. Yang, X. Fu, M. Zhu, B.D. Rodgers, Q. Jiang, M.V. Dodson, M. Du, Resveratrol enhances brown adipocyte formation and function by activating AMP-activated protein kinase (AMPK) alpha1 in mice fed high-fat diet, Mol. Nutr. Food Res. 61 (2017). [25] X. Liu, S. Wang, Y. You, M. Meng, Z. Zheng, M. Dong, J. Lin, Q. Zhao, C. Zhang, X. Yuan, T. Hu, L. Liu, Y. Huang, L. Zhang, D. Wang, J. Zhan, H. Jong Lee, J.R. Speakman, W. Jin, Brown adipose tissue transplantation reverses obesity in Ob/Ob mice, Endocrinology 156 (2015) 2461e2469.
[26] A. Luttun, M. Tjwa, P. Carmeliet, Placental growth factor (PlGF) and its receptor Flt-1 (VEGFR-1): novel therapeutic targets for angiogenic disorders, Ann. N. Y. Acad. Sci. 979 (2002) 80e93. [27] X. Liu, P. Claus, M. Wu, G. Reyns, P. Verhamme, P. Pokreisz, S. Vandenwijngaert, C. Dubois, J. Vanhaecke, E. Verbeken, J. Bogaert, S. Janssens, Placental growth factor increases regional myocardial blood flow and contractile function in chronic myocardial ischemia, Am. J. Physiol. Heart Circ. Physiol. 304 (2013) H885eH894. [28] R. Khurana, L. Moons, S. Shafi, A. Luttun, D. Collen, J.F. Martin, P. Carmeliet, I.C. Zachary, Placental growth factor promotes atherosclerotic intimal thickening and macrophage accumulation, Circulation 111 (2005) 2828e2836. [29] M. Wu, P. Pokreisz, M. Swinnen, E. Caluwe, H. Gillijns, N. Vanden Driessche, A. Casazza, E. Verbeken, D. Collen, S. Janssens, Sustained placental growth Factor-2 treatment does not aggravate advanced atherosclerosis in ischemic cardiomyopathy, J. Cardiovasc. Trans. Res. 10 (2017) 348e358.
Please cite this article in press as: J. Zhou, et al., Activation of brown adipocytes by placental growth factor, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.08.106