INTRODUCTION
Cachexia is a complex metabolic syndrome frequently associated with chronic illnesses such as cancer, chronic kidney disease (CKD), and heart failure. Characterized by an elevated basal energy expenditure, cachexia leads to a debilitating wasting of both adipose tissue and skeletal muscle. This wasting arises from an acceleration of fat and protein catabolism, significantly impacting patient prognosis. Alarmingly, cachexia affects approximately half of all cancer patients (Argiles et al., 2014), and up to 75% of CKD patients undergoing dialysis exhibit signs of this wasting syndrome (Mak et al., 2011). The presence of cachexia is a strong predictor of adverse outcomes and increased mortality. Distinct from simple malnutrition, cachexia cannot be reversed solely by nutritional supplementation, and effective therapies remain scarce (Fearon et al., 2013; Penna et al., 2010).
Recent advancements in understanding adipose tissue biology have revealed the existence of at least two types of uncoupling protein 1 (UCP1)-expressing adipocytes (Peirce et al., 2014). Classical brown fat, primarily located in the interscapular region in rodents, and beige adipocytes, found within white adipose tissues upon cold exposure or hormonal stimulation, are now recognized. Beige adipocytes originate from a distinct lineage compared to classical brown fat (Wu et al., 2012). The thermogenic activity of both brown and beige fat significantly contributes to energy expenditure, particularly in rodents. Importantly, both types of thermogenic fat cells are also present and functional in humans, playing a role in energy homeostasis (Cypess et al., 2013; Virtanen et al., 2009). Several studies have documented the activation of brown fat in rodent models of cancer cachexia, and anecdotal evidence suggests similar activation in some cachectic patients (Bianchi et al., 1989; Bing et al., 2000; Brooks et al., 1981; Roe et al., 1996; Shellock et al., 1986; Tsoli et al., 2012). More recently, the “browning” of white fat depots has been identified as a key driver of wasting in rodent models of cancer cachexia (Kir et al., 2014; Petruzzelli et al., 2014; Tsoli et al., 2012).
Our prior research identified parathyroid hormone-related protein (PTHrP), a polypeptide secreted by tumors, as a potent inducer of thermogenic gene expression and adipose tissue wasting (Kir et al., 2014). Intriguingly, neutralizing PTHrP with a specific antibody effectively reduced wasting in both fat tissue and skeletal muscle in tumor-bearing mice (Kir et al., 2014). PTHrP is frequently overexpressed in various tumors, and its circulating levels correlate with the severity of wasting in patients with metastatic cancer (Kir et al., 2014). Both PTHrP and parathyroid hormone (PTH) exert their effects through the same cell surface receptor, the PTH/PTHrP receptor (PTHR) (Vilardaga et al., 2011). While PTH-secreting tumors are rare, secondary hyperparathyroidism is a common complication in CKD patients (Bayne and Illidge, 2001; Levin et al., 2007; Tentori et al., 2015). In this study, we utilized fat-specific PTHR knockout mice to investigate the role of the adipose PTH/PTHrP pathway in cachexia associated with both CKD and cancer. Our findings demonstrate that mice lacking PTHR specifically in their fat tissue are remarkably resistant to cachexia induced by both renal failure and tumors.
RESULTS
5/6 Nephrectomy Induces Adipose Tissue Browning and Cachexia
The 5/6 nephrectomy model, involving the removal of one kidney and two-thirds of the other, is a well-established experimental model for kidney failure (Deboer, 2009). Mice subjected to this procedure develop cachexia and exhibit elevated circulating PTH levels. We employed this model to investigate the role of PTH in CKD-associated cachexia. Nephrectomized mice developed uremia, as indicated by elevated blood urea nitrogen (BUN) levels, and showed increased circulating PTH (Figure 1A, 1B and S1A). These mice also exhibited a significant reduction in body weight compared to sham-operated controls (Figure 1C). This weight loss phenotype was accompanied by an increase in energy expenditure, evidenced by elevated oxygen consumption and heat production (Figure 1D and S1B). Carbon dioxide production was also increased, without significant changes in the respiratory quotient (Figure S1C and S1D). Importantly, the observed weight loss was not attributable to increased physical activity or reduced food intake (Figure 1E and S1E). To further characterize the cachexia phenotype, we measured the weight of fat and skeletal muscle tissues. Epididymal white adipose tissue (eWAT, a visceral fat depot), inguinal white fat (iWAT, a subcutaneous fat depot), interscapular brown fat (iBAT), and gastrocnemius muscle all showed significant reductions in mass (Figure 1F, S2A and S2B). Gene expression analysis revealed the underlying mechanisms. The expression of thermogenic genes, including Ucp1, Dio2, Cidea, and Pgc1a, was significantly induced in iWAT, iBAT, and to a lesser extent in eWAT (Figure 1G, 1H, and S2C). Skeletal muscle wasting in nephrectomized mice was also associated with a decrease in the expression of the pro-growth hormone Igf1 and an induction of muscle atrophy-related genes such as Murf-1, Atrogin-1, and Myostatin (Figure 1I).
Figure 1. 5/6 nephrectomized mice develop cachexia with adipose tissue browning and skeletal muscle atrophy.
(A-I) Mice were subjected to sham or 5/6 nephrectomy (5/6 Nx) surgery and sacrificed 5 weeks post-surgery (n = 5-7). (A) Blood Urea Nitrogen (BUN) levels, (B) plasma PTH levels, and (C) body weight were measured. Metabolic cages were used to assess mice between post-surgery weeks 2 and 3, when body weight differences were minimal. (D) Oxygen consumption (VO2) and (E) physical activity were monitored. (F) Fat and muscle tissues were dissected and weighed. mRNA levels in (G) iWAT, (H) iBAT, and (I) gastrocnemius muscle were determined by RT-qPCR. Data are presented as mean ± SEM. P < 0.05, P < 0.005, P* < 0.0005. Refer to Figure S1 and S2 for supplementary data.
PTH Stimulates Thermogenic Gene Expression in Fat Tissues
PTHrP is a known potent inducer of thermogenic gene expression (Kir et al., 2014). Given the sequence homology in the first 34 amino acids and the shared receptor, PTH and PTHrP are expected to have similar functions. The elevated PTH levels observed in nephrectomized mice prompted us to investigate whether PTH administration could induce a thermogenic program in adipose cultures and in vivo. Similar to PTHrP(1-34), both PTH(1-34) and full-length PTH(1-84) stimulated the mRNA levels of Ucp1, Dio2, and Pgc1a when applied to primary inguinal fat cells (Figure 2A). Both PTH and PTHrP peptides also increased UCP1 protein levels and cellular respiration, including uncoupled respiration, via a mechanism involving the PKA signaling cascade, as previously reported (Kir et al., 2014). Furthermore, PTH administration to mice resulted in the induction of these thermogenic genes in various fat tissues (Figure 2B, 2C and 2D). These results suggest that elevated circulating PTH levels can indeed induce thermogenesis in fat.
Figure 2. PTH treatment elevates the thermogenic gene program.
(A) Primary cultures of inguinal adipocytes were treated with 10 ng/ml PTHrP(1-34), 10 ng/ml PTH(1-34), or 25 ng/ml PTH(1-84) for 2 hours (n = 3). (B-D) Mice were administered a single dose of PTH(1-34) (1 mg/kg body weight; SubQ) and sacrificed 2 hours later (n = 6). Gene expression changes were assessed by RT-qPCR. Data are presented as mean ± SEM. P < 0.05, P < 0.005, P* < 0.0005.
To explore the clinical relevance of thermogenic gene expression in adipose tissues and PTH levels, we examined samples from patients with primary hyperparathyroidism (PHPT). Subcutaneous and deep cervical fat samples were collected from PHPT patients undergoing parathyroidectomy. Control samples were obtained from patients undergoing thyroidectomy for benign pathologies (Graves’ disease, benign goiter, and Hurtle-cell neoplasm). Crucially, none of the patients had hyperthyroidism, which is known to affect thermogenic regulation (Table S1). Initially, we compared subcutaneous and deep cervical fat samples across all patients, examining mRNA levels of marker genes previously identified in a study profiling gene expression in these anatomical locations (Cypess et al., 2013). We observed that UCP1 and LHX8 expression was elevated in deep cervical fat (Figure 3A), while LEPTIN and SHOX2 expression was enriched in subcutaneous samples (Figure 3B). Next, we compared thermogenic gene expression in PHPT and control groups. While subcutaneous fat gene expression did not differ significantly between the groups (Figure 3C), we found a significant upregulation of CIDEA and PGC1A, and a trend towards increased UCP1 and DIO2 expression in deep cervical fat samples from PHPT patients (Figure 3D). Deep cervical fat is known to possess characteristics of both brown and beige fat, with considerable thermogenic capacity (Cypess et al., 2013). Therefore, PTH-induced thermogenesis in these fat depots may contribute significantly to hypermetabolism in PHPT.
Figure 3. The thermogenic gene program is increased in primary hyperparathyroidism patients.
(A-D) Subcutaneous and deep cervical fat samples were collected from patients undergoing neck surgeries. Gene expression levels were compared between subcutaneous and deep cervical samples (n = 23) (A-B) or within subcutaneous fat (C) and deep cervical fat (D) of control and primary hyperparathyroidism patients (PHPT) (n = 11-12). mRNA levels were determined by RT-qPCR. Statistical analysis was performed using a two-tailed t-test.
PTHrP and PTH Signal through PTHR to Induce Ucp1 Gene Expression
To investigate the role of PTHR in cachexia in the context of kidney failure and cancer models, we generated fat cell-specific PTHR knockout mice (Adipo-PTHR-KO) by crossing PTHR-floxed mice (Pth1rlox/lox) (Kobayashi et al., 2002) with Adiponectin-Cre mice (Figure S3). We isolated primary fat cells from these mice and treated them with PTHrP, PTH, or norepinephrine. While norepinephrine, a known inducer of thermogenesis, robustly stimulated Ucp1 mRNA expression in knockout cells, PTHrP and PTH completely failed to upregulate Ucp1 (Figure 4A). Similarly, PTHrP treatment of knockout mice did not induce thermogenic gene expression in white and brown fat depots (Figure 4B, 4C and 4D). These findings clearly demonstrate that both PTH and PTHrP require the PTHR receptor to activate the thermogenic gene program in adipose tissue. Notably, attempts to generate skeletal muscle-specific PTHR knockout mice using the human skeletal alpha (HSA)-Cre driver were unsuccessful in depleting PTHR in skeletal muscle tissue, despite robust Cre expression. This observation aligns with previous reports indicating PTHR localization in Pax7+ satellite cells and CD34+ hematopoietic stem cells, but not in mature myotubes (Kimura and Yoshioka, 2014).
Figure 4. PTH and PTHrP are unable to induce thermogenic genes in the absence of PTHR.
(A) Primary adipocytes from wild-type and mutant cells were treated with 10 ng/ml PTHrP(1-34), 10 ng/ml PTH(1-34), or 10 nM Norepinephrine (NE) for 2 hours (n = 3). (B-D) Mice were administered a single dose of PTHrP(1-34) (1 mg/kg body weight; SubQ) and sacrificed 2 hours later (n = 5-6). Gene expression changes were determined by RT-qPCR. (D) Total UCP1 and ERK1/2 protein levels in inguinal fat tissue were measured by western blotting. Data are presented as mean ± SEM. (*) indicates significant differences compared to the control group. (#) indicates significant differences between the WT and KO groups. P < 0.05, P < 0.0005, ###P* < 0.0005. See also Figure S3.
Adipo-PTHR-KO Mice Are Resistant to 5/6 Nephrectomy-driven Cachexia
Next, we investigated the contribution of elevated circulating PTH to cachexia associated with 5/6 nephrectomy using fat-specific PTHR knockout mice (Adipo-PTHR-KO) and their wild-type littermates (WT). Both groups developed comparable uremia and secondary hyperparathyroidism following 5/6 nephrectomy (Figure 5A, 5B and S4A). As observed in Figures 1D and S1B, WT mice exhibited a significant increase in energy expenditure and heat production, along with a decrease in physical activity, after nephrectomy (Figure 5C, 5D and S4B). Remarkably, nephrectomy-induced hypermetabolism was significantly suppressed in Adipo-PTHR-KO mice, and their physical activity was improved (Figure 5C and 5D and S4B). Carbon dioxide production, respiratory quotient, and food intake remained unchanged in Adipo-PTHR-KO mice (Figure S4C, S4D and S4E). Consistent with these metabolic phenotypes, nephrectomized Adipo-PTHR-KO mice experienced significantly less weight loss compared to the WT nephrectomy group, which exhibited severe cachexia (Figure 5E). Furthermore, wasting of both fat tissue and, surprisingly, skeletal muscle was ameliorated in the Adipo-PTHR-KO group (Figure 5F).
Figure 5. Adipo-PTHR-KO mice are resistant to 5/6 nephrectomy-driven cachexia.
(A-F) Mice were subjected to sham or 5/6 nephrectomy (5/6 Nx) surgery and sacrificed 5 weeks post-surgery (n = 5-6). (A) Blood Urea Nitrogen (BUN) and (B) plasma PTH levels were measured. Metabolic cages were used to assess mice between post-surgery weeks 2 and 3, when body weight differences were minimal. (C) Oxygen consumption (VO2), (D) physical activity, and (E) body weight were monitored. (F) Fat and muscle tissues were dissected and weighed. Data are presented as mean ± SEM. (*) indicates significant differences between the Sham and 5/6 Nx groups. (#) indicates significant differences between the WT-5/6 Nx and KO-5/6 Nx groups. P < 0.05, P* < 0.005, ***P < 0.0005, #P < 0.05, ##P < 0.005. See also Figure S4.
Histological examination of adipose tissues and skeletal muscle revealed that PTHR deletion in fat tissue prevented the fat droplet shrinkage and muscle fiber atrophy observed in the WT nephrectomy group (Figure 6A). PTHR depletion also blocked the upregulation of thermogenic genes in fat tissues (Figure 6B, 6C and 6D). In addition to preserving muscle mass, nephrectomized Adipo-PTHR-KO mice maintained muscle strength, as evidenced by improved grip strength compared to nephrectomized WT mice (Figure 6E). Consistently, the induction of atrophy-related genes Murf-1, Atrogin-1, and Myostatin was inhibited in Adipo-PTHR-KO mice (Figure 6F). These findings indicate a PTH-PTHR signaling axis mediating cachexia in kidney failure and suggest an important crosstalk mechanism between fat and skeletal muscle wasting. One plausible explanation for this crosstalk is that PTHrP-induced, fat tissue-derived circulating factors may trigger muscle atrophy. Microarray analysis of fat cells treated with PTH or PTHrP identified 13 genes encoding secreted proteins that were upregulated by both PTH and PTHrP (Figure S5). These include circulating bioactive cytokines such as IL6, IL33, Cxcl1, Cxcl5, and Cxcl14.
Figure 6. Adipose tissue browning and skeletal muscle atrophy are suppressed in 5/6 nephrectomized Adipo-PTHR-KO mice.
(A-F) Mice were subjected to sham or 5/6 nephrectomy (5/6 Nx) surgery (n = 5-6). (A) H&E staining of adipose tissues and gastrocnemius muscle. mRNA levels in (B) iWAT, (C) iBAT, and (F) gastrocnemius muscle were measured by RT-qPCR. (D) Total UCP1 and ERK1/2 protein levels in inguinal fat tissue were determined by western blotting. (E) Muscle function was analyzed by grip strength. Data are presented as mean ± SEM. (*) indicates significant differences between the Sham and 5/6 Nx groups. (#) indicates significant differences between the WT-5/6 Nx and KO-5/6 Nx groups. P < 0.05, P* < 0.005, ***P < 0.0005, #P < 0.05. See also Figure S5.
Adipo-PTHR-KO Mice Are Resistant to LLC-tumor-driven Cachexia
Building upon the findings from the 5/6 nephrectomy study, we examined tumor-driven cachexia in Adipo-PTHR-KO mice using the Lewis Lung Carcinoma (LLC) model. Remarkably, 16 days post-tumor inoculation, Adipo-PTHR-KO mice did not experience significant weight loss, while WT mice exhibited evident cachexia (Figure 7A). PTHR depletion in fat tissues attenuated wasting of both adipose tissue and skeletal muscle without altering average tumor mass (Figure 7B and 7C). Histological analysis revealed larger fat droplets and muscle fibers in tumor-bearing Adipo-PTHR-KO mice compared to WT controls (Figure 7D). Similar results were observed in Adipo-PTHR-KO mice sacrificed 14 days post-tumor inoculation, when WT controls were experiencing moderate cachexia (Figure S6). Gene expression analysis indicated that Adipo-PTHR-KO mice were resistant to browning induced by LLC tumors (Figure 7E, 7F and 7G). Muscle function in knockout mice was also significantly improved, with suppressed atrophy-related gene expression in muscle tissue (Figure 7H and 7I). These observations parallel our previously published findings on PTHrP neutralization using a systemic antibody (Kir et al., 2014). These new data strongly suggest that the cachectic effects of tumor-derived PTHrP are primarily mediated by PTHR expressed in fat tissues.
Figure 7. Adipo-PTHR-KO mice are resistant to LLC tumor-driven cachexia.
(A-D) Mice inoculated with LLC cells were sacrificed 16 days later (n = 6). (A) Carcass weight (total weight minus tumor weight) and (B) tumor weight are shown. (C) Fat and muscle tissues were dissected and weighed. (D) H&E staining of adipose tissues and gastrocnemius muscle. (E-G) Mice inoculated with LLC cells were sacrificed 14 days later (n = 4-5). mRNA levels in (E) iWAT and (F) iBAT were measured by RT-qPCR. (G) Total UCP1 and ERK1/2 protein levels in inguinal fat tissue were determined by western blotting. (H-I) Mice inoculated with LLC cells were sacrificed 16 days later (n = 6). (H) Muscle function was analyzed by grip strength. (I) mRNA levels in gastrocnemius muscle were measured by RT-qPCR. Data are presented as mean ± SEM. (*) indicates significant differences between the LLC and non-tumor-bearing groups. (#) indicates significant differences between the WT-LLC and KO-LLC groups. P < 0.05, P* < 0.005, ***P < 0.0005, #P < 0.05, ##P < 0.005, ###P < 0.0005. See also Figure S6.
DISCUSSION
Cachexia is a severe and debilitating syndrome that significantly worsens the prognosis of cancer and other chronic diseases. The urgent need for effective therapies against this wasting syndrome is widely recognized. PTHrP, frequently produced by tumors, is often implicated in hypercalcemia of malignancy (Iguchi et al., 2001; Mundy and Edwards, 2008). Our previous work demonstrated that tumor-derived PTHrP induces adipose tissue browning and cachexia in tumor-bearing mice, even at levels insufficient to elevate plasma calcium. Given that PTH and PTHrP share receptor specificity, both hormones may contribute to cachexia. Elevated circulating PTH levels have been reported in numerous chronic diseases, particularly CKD (Childs et al., 2012; Dolecek et al., 2003; Jackson et al., 2013; Levin et al., 2007; Visser et al., 2003). Our current findings highlight a critical role for PTH in CKD-related cachexia. Similar to PTHrP, PTH effectively stimulates thermogenic gene expression in fat tissue. Importantly, and somewhat unexpectedly, the selective deletion of PTHR in fat tissues effectively prevents both fat and muscle wasting in CKD and cancer cachexia models. The observation that loss of PTHR in fat tissue also improves muscle mass and strength strongly suggests indirect mechanisms, mediated by PTHR function in fat tissue, that transmit pathological signals from fat to skeletal muscle.
Chronic kidney failure is a multifaceted disease. While malnutrition and comorbidities can contribute to weight loss in CKD patients, increased energy expenditure in certain CKD patients is associated with increased mortality and cardiovascular disease (Neyra et al., 2003; Wang et al., 2004). Interestingly, dialysis patients with hyperparathyroidism have been shown to exhibit increased resting energy expenditure, which can be reduced by parathyroidectomy (Cuppari et al., 2004). Increased Ucp1 expression in brown fat has previously been implicated in weight loss in 5/6 nephrectomized mice (Cheung et al., 2014; Cheung et al., 2007). Our findings reinforce the involvement of adipose tissue thermogenesis in this cachexia model and identify PTH as a significant driver of thermogenesis and hypermetabolism. As demonstrated, PTH treatment potently induces thermogenic genes in mouse fat tissue. Furthermore, primary hyperparathyroidism in humans, although not directly linked to cachexia, is associated with increased thermogenic gene expression in deep cervical fat, a depot with brown/beige fat characteristics and significant thermogenic capacity. Therefore, secondary hyperparathyroidism in CKD and other chronic diseases may lead to inappropriate thermogenesis, triggering wasting in the presence of other contributing factors. Our research also reveals that the PTH/PTHrP pathway stimulates the expression of several secreted factors in fat cells. This suggests that this pathway exerts multiple effects on fat tissue, potentially contributing to hypermetabolism and skeletal muscle atrophy. Further research is needed to elucidate the roles of these secreted factors in PTH/PTHrP-driven muscle atrophy.
In conclusion, our work underscores the PTH/PTHrP pathway as a key player in cachexia, suggesting that targeting this pathway may be a promising therapeutic approach. Indeed, our earlier study demonstrated that neutralizing PTHrP with a specific antibody effectively blocks tumor-driven cachexia (Kir et al., 2014). Investigating the potential benefits of similar approaches against CKD-driven cachexia is warranted. However, it is important to consider that PTH, unlike PTHrP, also plays a crucial role in mineral metabolism. Therefore, therapeutic targeting of PTH or PTHR may be limited by their essential functions in calcium and phosphate regulation.
EXPERIMENTAL PROCEDURES
Reagents
Synthetic mouse PTHrP(1-34), rat PTH(1-34), and rat PTH(1-84) were obtained from Bachem. The mouse intact PTH (1-84) ELISA assay kit was purchased from ALPCO. Norepinephrine was acquired from Sigma. Total-ERK1/2 antibody (#9102) and anti-Ucp1 antibody (ab10983) were from Cell Signaling and Abcam, respectively.
Animal Studies
All animal experiments were conducted with the approval of the Institutional Animal Care and Use Committee of Beth Israel Deaconess Medical Center. Lean, 6-8-week-old, male, C57BL/6 mice were used in all studies. Pth1r-floxed mice, a gift from Dr. Henry Kronenberg (MGH) (Kobayashi et al., 2002), and Pth1r-floxed mice (± Adiponectin-Cre) were maintained on a pure C57BL/6 background. All other mice were sourced from Charles River Laboratories. Mice were housed under 12-hour light/dark cycles (6 am-6 pm) at 24°C and fed a standard irradiated rodent chow diet. For tumor studies, 5 million LLC cells per mouse were injected subcutaneously into the flank. Control mice received vehicle (PBS) only. PTHrP and PTH peptides were administered via subcutaneous injections, and all mice were sacrificed between 4 pm and 7 pm. Plasma was collected into EDTA tubes for PTH ELISA assay and into heparin tubes for Blood Urea Nitrogen (BUN) measurements, performed using a Vitros analyzer. Whole-body energy metabolism was assessed using a Comprehensive Lab Animal Monitoring System (CLAMS, Columbia Instruments). Mice were placed in metabolic cages between post-surgery weeks 2 and 3 (and sacrificed at week 5), when body weight differences between sham and 5/6 nephrectomy groups were minimal. CO2 and O2 data were collected every 32 minutes per mouse, normalized to total body weight. Activity, heat generation, and food intake data were measured at shorter intervals. For hematoxylin and eosin staining, tissues were fixed in 10% formalin, paraffin-embedded, and sectioned at 6 μm thickness. Whole-body fat composition was determined using an EchoMRITM analyzer.
5/6 Nephrectomy
The 5/6 nephrectomy procedure involved a two-phase surgical removal of one kidney and two-thirds of the other. Mice were anesthetized with 1% isoflurane. An incision was made lateral to the spine, the right kidney was freed, decapsulated (preserving the adrenal gland), and the upper and lower poles were partially resected, leaving approximately 1/3 of the kidney. Bleeding was controlled using electrocautery. The abdomen was closed with sutures and wound clips. After one week of recovery, the procedure was repeated on the left side, removing the entire kidney after cauterizing renal blood vessels and the ureter. Control mice underwent sham operations involving kidney decapsulation but not removal. Buprenorphine was administered as an analgesic for two days post-surgery.
Grip Strength
Forelimb grip strength was assessed on the day of sacrifice. Mice were allowed to grasp a bar attached to a force transducer (Model DFX II; Chatillon) and pulled horizontally by the tail away from the bar (Cabe et al., 1978).
Cell Culture
Inguinal stromal-vascular (SV) fractions were isolated from 30-35 day-old male mice. Inguinal fat tissue was dissected, washed, minced, and digested for 45 min at 37°C in PBS containing 10 mM CaCl, 2.4 U/mL dispase II (Roche), and 1.5 U/mL collagenase D (Roche). Digested tissue was filtered through a 100-μm cell strainer, centrifuged at 600g for 5 min to pellet SV cells, resuspended in adipocyte culture medium (DMEM/F12 plus glutamax (1:1; Invitrogen), pen/strep, and 10% FBS), filtered through a 40-μm cell strainer, centrifuged, resuspended again, and plated. SV cells were grown to confluency and differentiated using an adipogenic cocktail (1 μM dexamethasone, 5 μg/mL insulin, 0.5 μM isobutylmethylxanthine (DMI), and 1 μM rosiglitazone in adipocyte culture medium). Two days post-induction, cells were maintained in adipocyte culture medium with 5 μg/mL insulin and 1 μM rosiglitazone. From day 6, cells were maintained in adipocyte culture medium only, treated with PTHrP and PTH peptides for 2 hours, and harvested at day 8.
RT-qPCR
RNA was extracted from cultured cells or frozen tissues using TRIzol (Invitrogen), purified with Qiagen RNeasy minicolumns, and reverse transcribed using the High Capacity cDNA Reverse Transcription kit (Applied Biosystems). qPCR was performed using SYBR® GreenER™ PCR Master Mix (Invitrogen) in a 384-well format using an ABI PRISM® 7900HT instrument (Applied Biosystems). Relative mRNA levels were calculated using the comparative CT method, normalized to cyclophilin mRNA. Mouse primer sequences were previously published (Kir et al., 2014). Human primer sequences are provided in the original article.
Microarray
Affymetrix mouse genome 430A v2.0 arrays were used for gene expression profiling, analyzed using dChip software. The microarray dataset is available under GEO accession number GSE74082.
Western Blotting
Frozen tissues were homogenized in lysis buffer (50 mM Tris (pH 7.4), 500 mM NaCl, 1% NP40, 20% glycerol, 5 mM EDTA, 1 mM PMSF, protease and phosphatase inhibitors). Homogenates were centrifuged, and supernatants (whole cell lysates) were collected. Protein concentration was determined by Bio-Rad Protein assay. 30 μg of protein lysate per sample was used for SDS-PAGE. PVDF membranes were blotted with antibodies in TBS containing 0.05% Tween and 5% BSA. Secondary antibody incubation used TBS-T with 5% milk. ECL western blotting substrates (Pierce) were used for visualization.
Human Study
This study was approved by the Human Studies Institutional Review Board of Massachusetts General Hospital (MGH). Patients undergoing neck surgery at MGH were identified by K.P.E. and W.L., and written informed consent was obtained prior to surgery. Subcutaneous and deep cervical fat samples were collected from patients undergoing parathyroidectomy for PHPT and from control patients undergoing thyroidectomy for benign pathologies. Patients with thyroid or parathyroid cancer and hyperthyroidism were excluded. Subjects were not excluded based on demographics.
Statistical Analysis
Data are presented as mean ± SEM. Statistical significance was determined using a two-tailed, unpaired t test for single variables and two-way ANOVA followed by Bonferroni post-tests for multiple variables. Two-way ANOVA with repeated measures was used for analyzing body weight and metabolic data. Graphpad Prism software was used for ANOVA analysis.
Supplementary Material
ACKNOWLEDGEMENTS
S.K. was supported by a Robert Black Fellowship from the Damon Runyon Cancer Research Foundation (DRG-2153-13). H.K. was supported by a JSPS Postdoctoral Fellowship for Research Abroad. This work was funded by NIH grants (DK31405 to B.M.S. and DK097105 to B.L.) and the JPB Foundation to B.M.S.
Footnotes
Publisher’s Disclaimer: This is a PDF file of an unedited manuscript accepted for publication. This version is provided as a service to customers and will undergo copyediting, typesetting, and proof review before final publication. Errors may be discovered during production that could affect content, and all legal disclaimers apply to the journal.
REFERENCES
[List of references as in the original article]
