Lipodystrophy in methylmalonic acidemia associated with elevated FGF21 and abnormal methylmalonylation

A distinct adipose tissue distribution pattern was observed in patients with methylmalonyl-CoA mutase deficiency, an inborn error of branched-chain amino acid (BCAA) metabolism, characterized by centripetal obesity with proximal upper and lower extremity fat deposition and paucity of visceral fat, that resembles familial multiple lipomatosis syndrome. To explore brown and white fat physiology in methylmalonic acidemia (MMA), body composition, adipokines, and inflammatory markers were assessed in 46 patients with MMA and 99 matched controls. Fibroblast growth factor 21 levels were associated with acyl-CoA accretion, aberrant methylmalonylation in adipose tissue, and an attenuated inflammatory cytokine profile. In parallel, brown and white fat were examined in a liver-specific transgenic MMA mouse model (Mmut–/– TgINS-Alb-Mmut). The MMA mice exhibited abnormal nonshivering thermogenesis with whitened brown fat and had an ineffective transcriptional response to cold stress. Treatment of the MMA mice with bezafibrates led to clinical improvement with beiging of subcutaneous fat depots, which resembled the distribution seen in the patients. These studies defined what we believe to be a novel lipodystrophy phenotype in patients with defects in the terminal steps of BCAA oxidation and demonstrated that beiging of subcutaneous adipose tissue in MMA could readily be induced with small molecules.

Ultrasound of the abdomen showed normal spleen, kidneys and liver, with increased subcutaneous fat.
Ultrasound of the dorsocervical fat area showed increased vascularization (Supplemental Figure 1D).Whole body DEXA was performed with a BMD corrected for height Z-score of 1.3.
Subcutaneous tissue obtained during a port-a-cath removal showed mildly positive staining for uncoupling protein -1 expression (UCP1) and rare mitochondria with abnormal ultrastructure (Supplemental Figure 1E and F).
Patient 13 F-FDG-PET study: Approximately 60min after the intravenous administration of 5.788mCi F-18fluoro-2-deoxyglucose, PET/CT was performed form the base of the skull to the thighs.The blood glucose was 97mg/dl.The patient had hemodialysis 5hrs prior to injection.Great care was taken to ensure that no glucose containing fluids were used during dialysis.The patient's gastrostomy tube feeds were also withheld.PET and POCT fingerstick glucose were performed in Nuclear Medicine, 300 Longwood Ave, Boston, MA.
Dual-Energy X-ray Absorptiometry (DXA) analysis: Whole body scans were performed using DXA (Hologic Delphi A; Hologic, Bedford, MA) to quantify the full-body mass distribution for fat, fat-free (lean) mass, bone and total.Subtotal mass represents full body minus the head measurements.Using the subregion analysis tool, customized regions of interest (ROIs) were drawn to capture 4 segments: right upper arm, right forearm, right thigh and right shank (Figure 2A and Supplemental Table 1).To assess the between-day intra-tester and inter-tester reliability for the cross-section analysis of appendicular segments, all DXA scans subregions were drawn and analyzed twice by the same observer, while a subset of 50 scans (20 patient and 25 controls) were re-analyzed separately by two different observers.Coefficients of variation (CV) were calculated using IBM SPSS Statistics Version 21.0, Chicago, IL.Coefficients of variation below 5% were considered highly reliable.
(Supplemental Table 1) SomaScan assay: Large-scale proteomic analysis was performed using the SomaScan 1.3k Assay (SomaLogic, Boulder, CO, USA) at the Trans-NIH Center for Human Immunology, Autoimmunity, and Inflammation (CHI), as described previously (2).This aptamer-based assay can detect 1305 protein analytes in human serum, including cytokines, hormones, growth factors, receptors, kinases, proteases, protease inhibitors, and structural proteins.A complete list of analytes measured can be found at http://somalogic.com/wp-content/uploads/2017/06/SSM-045-Rev-2-SOMAscan-Assay-1.3k-Content.pdf.Concentrations are measured as relative fluorescence units.A subset of 20 MMA patient samples were used, 10 with high and 10 with low elevations in plasma FGF21 concentrations, previously measured by quantikine ELISA (R&D Systems, DF2100); control set included plasma samples from 20 age, gender, BMI matched controls.Specimens (50mL) were diluted to three concentrations (0.005%,1%, and 40%) to separate groups of high, medium, and low abundance proteins, respectively, and then combined with dilution-specific SOMAmers.Quality controls (QC) and calibrators provided by SomaLogic were run together with internal site QC samples.Data generated from these samples were used to assess interassay variability, as described previously.Analysis of the data was performed with Partek platform and Ingenuity Pathway Analysis (IPA).Data were analyzed using SomaSuite version 1.0.3 (NEC Corporation, Minato, Tokyo, Japan) and web-tools developed by CHI (https://foocheung.shinyapps.io/adat_v02/andhttps://foocheung.shinyapps.io/plotterII/)(3,4).Unpaired t-tests for two group comparison false discovery rate (FDR) for correction for multiple comparisons were performed.Pearson correlation was performed using an online tool developed by CHI.

MOUSE STUDIES:
Mice were maintained on a 12:12-h light-dark cycle, and all experiments were performed in agreement with National Institutes of Health guidelines and with the approval of the Animal Care and Use Committee of NHGRI and NIDDK.Metabolic phenotyping.Indirect calorimetry was performed using an eight-chamber Oxymax system (Columbus Instruments, Columbus, OH) with one mouse/chamber and by testing transgenic Mmut -/-;Tg INS-Alb- Mmut mice (n=5), simultaneously with littermate controls, Mmut +/-;Tg INS-Alb-Mmut (n=12) and wild type littermates (n=6).Female mice, 2.8-3.1 months old, were used for these experiments.Experimental design was as follows: Day 1: mice were adapted to metabolic chambers at 24°C; day 2: data were collected for 24h at 24°C; day 3: data were obtained for 24h at 30°C; day 4: mice were injected intraperitoneally with the β3-selective adrenergic agonist CL-316,243 (1mg/kg in saline) as described previously (5).Activity (total and ambulatory) was determined by infrared beam interruption (Opto-Varimex mini, Columbus Instruments, Columbus, OH).
Mice had free access to regular chow (7022 NIH-07 diet, 29% kcal from protein, 56% from carbohydrates and 15% , Harlan Laboratories, Madison, WI) and water.Total O2 consumption was calculated as the average of the points with fewer than six ambulating beam breaks per interval and omission of the first hour of the experiment.The respiratory exchange ratio (RER; the ratio of CO2 produced to O2 consumed) was calculated using the same data points.Oxidation of carbohydrate produces an RER of 1.00, whereas FA oxidation results in an RER of 0.70.Results for O2 consumption were normalized to (body weight) ^ 0.75 .The effect of CL-316,243 on metabolic rate was measured with each mouse serving as its own control.Mice were placed into the calorimetry chambers (prewarmed to 30°C) at 9:00 AM and baseline data were collected.CL-316,243 was injected (1 mg/kg ip) 3h later, and after a 1h delay data were collected for 3h.Body composition was measured in non-anesthetized mice with time domain Echo MRI 3-in-1 analyzer (Echo Medical Systems, Houston, TX).
Cold exposure experiment: Core body temperature and blood glucose were monitored hourly or every 30min if the temperature was reduced, by a rectal temperature probe (RET3 probe connected to a Thermocouple thermometer #BAT-12R, World Precision Instruments, FL).Three hours after cold exposure mice were anaesthetized inside the cold room, by injection of sodium pentobarbital (30 mg/ml in 0.9% NaCl; 1.5 ml/kg of body mass, intraperitoneal) and, once locomotor activity had ceased, retroorbital bleed was obtained, and brown and white fat pads were dissected and snap frozen for subsequent gene expression analysis.Mean temperature change (%change from baseline) was monitored at hourly intervals.Temperature drop was not significantly different between the different genotypes, although the trend was bigger for female mutant mice at the 3hrs time point.Another set of mice kept at room temperature (24 ± 1°C; controls) were euthanized for tissue collection as controls.
18 F-FDG-PET studies: FDG-PET studies were performed using an Inveon small-animal PET scanner (Siemens Preclinical Solutions). 18F-FDG (18F-fluoro-D-glucose) was administered intraperitoneally to experimental mice with a dose of 11.1 MBq.Mice were kept fasting 3h prior to the PET studies.One hour after the tracer had been injected, a 15 min static PET scan was performed under isoflurane anesthesia with a heating pad to keep anesthetized animals at 23-25 0 C. The images were reconstructed using a two-dimensional ordered-subset expectation maximum (2D OSEM) algorithm.The mean pixel value of each ROI was measured with Inveon Research Workshop software (Siemens Preclinical Solution).The value was then converted to the concentration of radioactivity in units of megabecquerels per milliliter.The image-derived tissue uptake, presented as percent injected dose per gram (%ID/g), was obtained with tissue radioactivity divided by injected dose assuming a tissue density of 1 g/mL.
Quantitative real-time PCR analysis.Total RNA from frozen tissue was extracted using the RNeasy Mini Kit (74104; Quiagen, Valencia, CA).DNase digestion was performed using DNA-free (AM1906; Ambion, Austin, TX) and 2µg of RNA was reverse transcribed using the High-Capacity cDNA Kit (4368814; Applied Biosystems, Foster City, CA).TaqMan gene expression assays were performed in triplicate according to the manufacturer's instructions using the Fast Universal PCR Master Mix (4352042, Applied Biosystems) and the Applied Biosystems 7500 Fast Real-Time PCR System.TaqMan probes specific to murine PpargC1α (Mm01208835_m1), Ucp1 (Mm01244861_m1), Dio2 (Mm00515664_m1), Lcn2 (Mm01324470_m1) were used and mRNA expression was normalized to housekeeping Gapdh (Mm99999915_g1) (Applied Biosystems, Foster City, CA).Quantification of relative gene expression was calculated using the 2 -ΔΔCT (comparative threshold) method.
Histology, immunohistochemistry and electron microscopy.Tissues were fixed in 10% formalin, embedded in paraffin, sectioned, stained with hematoxylin and eosin following standard procedures (Histoserv), and examined by light microscopy.Sections of white fat, inguinal or subcutaneous were stained for UCP1 (ab-23841; Abcam) by immunohistochemistry, following the manufacturers' instructions [Ready-to-Use Vectastain Universal ABC Kit (Vector Labs)].Tissue slides were analyzed with an Olympus microscope at a 200× magnification.Transmission electron microscopy was performed on tissues fixed at 4 0 C in 2% glutaraldehyde in 0.1M cacodylate buffer (pH 7.4).The tissues were postfixed in 2% OsO4 for 2h, washed again with 0.1M cacodylate buffer three times, subsequently washed with water and placed in 1% uranyl acetate for 1h.The tissues were serially dehydrated in ethanol and propylene oxide and embedded in EMBed 812 resin (Electron Microscopy Sciences, Hatfield, PA, USA).Thin sections, 80 nm thick, were obtained by utilizing an ultramicrotome (Leica, Deerfield, IL, USA), placed onto 300 mesh copper grids and stained with saturated uranyl acetate in 50% methanol and then with lead citrate.The grids were viewed in the JEM-1200EXII electron microscope (JEOL Ltd, Tokyo, Japan) at 80 kV and images were recorded on the XR611M, mid mounted, 10.5Mpixel, CCD camera (Advanced Microscopy Techniques Corp, Danvers, MA, USA).Since the SOMASCAN platform did not include FGF21, we measured GDF15, that is closely correlated with FGF21 in our MMA patient cohort (7) similar to observations in various mitochondrial disorders (8).Both markers were significantly elevated in the subset of MMA patients included in the current report compared to matched controls (P<0.0001for both, Supplemental Figure 6A) (r=0.541,P=0.0052, R 2 =0.293).In addition to the dysregulation of several inflammatory pathways, significant differences were observed in a number of complement factors and members of the coagulation cascade, which will require further validation and exploration in the MMA patient cohort.Although not typically associated with a hypercoagulable state, there Manoli I et al.
Lipodystrophy in methylmalonic acidemia 10 have been reports of cases with venous thrombosis, especially in the perioperative setting, especially after a liver transplantation procedure .