While others may be dieting and hitting the gym hard to stay in shape, some people stay slim effortlessly no matter what they eat.
In a study publishing May 21 in the journal Cell, researchers use a genetic database of more than 47,000 people in Estonia to identify a gene linked to thinness that may play a role in resisting weight gain in these metabolically healthy thin people.
They show that deleting this gene results in thinner flies and mice and find that expression of it in the brain may be involved in regulating energy expenditure.
“We all know these people: it’s around one percent of the population,” says senior author Josef Penninger, the director of the Life Sciences Institute and professor of the department of medical genetics at the University of British Columbia.
“They can eat whatever they want and be metabolically healthy. They eat a lot, they don’t do squats all the time, but they just don’t gain weight.
“Everybody studies obesity and the genetics of obesity,” he says. “We thought, ‘Let’s just turn it around and start a new research field.’ Let’s study thinness.”
Penninger’s team looked at data from the Estonian Biobank, which includes 47,102 people aged 20 to 44 years old. The team compared the DNA samples and clinical data of healthy thin individuals with normal-weight individuals and discovered genetic variants unique to thin individuals in the ALK gene.
Scientists have known that the ALK gene frequently mutates in various types of cancer, and it gained a reputation as an oncogene, a gene that drives the development of tumors.
The role of ALK outside of cancer has remained unclear. But this new finding suggested that the gene may play a role as a novel thinness gene involved in weight-gain resistance.
The researchers also found that flies and mice without ALK remained thin and were resistant to diet-induced obesity. Furthermore, despite having the same diet and activity levels as normal mice, mice with deleted ALK have lower body weight and body fat.
The team’s mouse studies also suggested that ALK, which is highly expressed in the brain, plays a part there by instructing the fat tissues to burn more fat from food.
The researchers say that therapeutics targeting the gene might help scientists fight obesity in the future.
“If you think about it, it’s realistic that we could shut down ALK and reduce ALK function to see if we did stay skinny,” says Penninger. “ALK inhibitors are used in cancer treatments already. It’s targetable. We could possibly inhibit ALK, and we actually will try to do this in the future.”
Further research will be required to see if these inhibitors are effective for this purpose. The team also plans to further study how neurons that express ALK regulate the brain at a molecular level to balance metabolism and promote thinness.
The Estonian Biobank that the team studied was ideal because of its wide age range and its strong phenotype data. But one limitation for replicating these findings is that biobanks that collect biological or medical data and tissue samples don’t have a universal standard in data collection, which makes comparability a challenge.
The researchers say they will need to confirm their findings with other data banks through meta-analysis. “You learn a lot from biobanks,” says Penninger. “But, like everything, it’s not the ultimate answer to life, but they’re the starting points and very good points for confirmation, very important links and associations to human health.”
The team says that its work is unique because of how it combines exploration of the genetic basis of thinness on a population- and genome-wide scale with in vivo analyses in mice and flies of the gene’s function.
“It’s great to bring together different groups, from nutrition to biobanking, to hardcore mouse and fly genetics,” says Penninger. “Together, this is one story including evolutionary trees in metabolism, the evolutionary role of ALK, human evidence, and hardcore biochemistry and genetics to provide causal evidence.”
Given the ever-growing epidemics of obesity and weight-related health complications such as type 2 diabetes and cardiovascular disease, many studies have focused on understanding metabolic changes associated with overweight and obesity.
However, novel insights on metabolic differences in humans might also come from studying people on the extreme opposite side of obesity, a state called constitutional thinness (CT).
CT is characterized by a stable low BMI, generally <18 kg/m2 (1, 2). Unlike in anorexia nervosa or other eating disorders associated with low weight, the nonpathological state of CT is defined by the lack of psychological features such as food aversion, refusal to eat alternated with binge eating, lower self-esteem, perfectionism, and body dissatisfaction (3).
In contrast to lipodystrophic and anorexic individuals, CT subjects have body fat mass percentages considered in the healthy range (4), albeit lower than people generally considered lean (BMI: 19–24 kg/m2).
Hormonal profiles are similar to those of lean subjects: they have no detectable abnormalities of cortisol, growth hormone, insulin-like growth factor 1, or free triiodothyronine secretory patterns (2) and normal menstrual cycles (5).
Lower leptin levels were detected in CT individuals compared with lean women (3). Daily calorie consumption is consistently similar in CT individuals and lean controls, while absolute resting metabolic rate is lower in CT individuals.
In addition, CT women were found to be resistant to weight gain upon fat overfeeding (6). To date, however, none of the studies on CT shed light on the possible molecular mechanisms involved in this thin, weight-resistant phenotype.
Moreover, clinical characterization has focused on female CT individuals only, while the CT phenotype also exists in males. In this study, we evaluated plasma and/or urinary proteins and metabolites before and after a 2-wk overfeeding intervention in both female and male CT individuals compared with lean, normal-weight participants.
We also analyzed mitochondrial function and transcriptomic profiles in the adipose tissue and muscle from CT and control subjects. Our results postulate that CT individuals are characterized by an adipose tissue–specific higher mitochondrial function as well as distinct transcriptomic profiles, testifying for altered lipid metabolism in CT.
This observation suggests that altered adipose tissue metabolic activity is a hallmark of the CT phenotype.
Clinical characterization of CT individuals at baseline
Table 1 summarizes the results of the clinical characterization of the subjects at baseline. As a sex interaction was not observed for most variables, we combined both sexes within each group. Total fat and lean mass were lower in CT individuals compared with controls (both in absolute and relative terms).
We also identified lower systolic and diastolic blood pressure in CT individuals compared with controls. To exclude that the lower weight in CT individuals might be related to defects in gastrointestinal fat absorption, such as steatorrhea, we evaluated potential excess accumulation of fat in the feces (24).
However, neither visual inspection by the Bristol stool scale nor fat quantification in the stool showed any differences between the groups. Food intake was not different between the 2 groups, both in free-living as well as in the calorimetric chambers, and the proportions for each macronutrient category were similar between the groups (consisting on average of 48% carbohydrates, 37% lipids, and 15% protein relative caloric intake).
Snacking frequency and snacking caloric intake were significantly higher in CT individuals than controls, confirming previous results (6). Based on Actiheart measurements, we did not observe higher spontaneous physical activity in CT compared with control individuals. Rather, CT individuals had lower physical activity compared with controls.
Both REE and TEE were significantly lower in CT individuals compared with controls in both free-living and calorimetric chamber conditions, but this difference was blunted when correcting for fat-free mass (P = 0.92 in canopy and P = 0.12 in the calorimetric chamber) (Supplemental Figure 2).
Fasting RQ by canopy measurement was similar between CT individuals and controls while fasting fat oxidation rate was higher in CT individuals than controls, and the carbohydrate oxidation index was highly variable and nominally higher in CT individuals.
To complement standard energy expenditure measures, we also tracked internal Tc for 3 h postingestion with a noninvasive Jonah capsule temperature sensor.
Overall results show a slightly higher Tc in CT individuals compared with controls at baseline (Supplemental Figure 3A; ANOVA, P = 0.0087), with an average increase of +0.14°C pre- and postmeal.
TABLE 1 – Clinical variables in CT (n = 29, 14 females and 15 males) compared with control (n = 29, 15 females and 14 males) individuals at baseline 1
|Characteristic||CT||Control||CT status, P||CT status × sex, P|
|Age, y||24.97 ± 4.7||22.63 ± 2.93|
|BMI, kg/m2||16.96 ± 0.74||22.99 ± 1.03|
|Waist-hip ratio||0.77 ± 0.06||0.78 ± 0.07||0.042||0.28|
|Total fat, kg||9.71 ± 1.86||18.71 ± 5.20||1.7 × 10−23*||0.98|
|Total fat body mass, %||19.59 ± 4.97||27.09 ± 7.89||4.5 × 10−08*||0.53|
|Total lean body mass, kg||39.28 ± 6.70||49.79 ± 9.70||2.0 × 10−26*||0.72|
|Total lean body mass, %||80.41 ± 4.97||72.56 ± 8.13||4.4 × 10−08*||0.61|
|Heart rate, bpm||69.21 ± 10.68||65.28 ± 10.38||0.033||0.031|
|Systolic blood pressure, mm Hg||115.07 ± 11.86||125.72 ± 9.83||4.2 × 10−05*||0.17|
|Diastolic blood pressure, mm Hg||66.21 ± 9.36||71.21 ± 6.67||6.0 × 10−04*||0.079|
|Fat stool, g/24 h||3.16 ± 2.13||4.14 ± 2.31||0.25||0.39|
|Total caloric intake, kcal/24 h||2179.62 ± 433.09||2004.55 ± 482.75||0.45||0.062|
|Carbohydrates, g||264.14 ± 65.22||227 ± 58.01||0.16||0.036|
|Protein, g||79.48 ± 19.53||80.34 ± 20.24||0.3||0.053|
|Lipids, g||88.52 ± 18.64||81.38 ± 20.24||0.77||0.23|
|Snacking calories, kcal/24 h||328.2 ± 192.67||177.08 ± 138.82||2.1 × 10−03*||0.82|
|Snacking frequency||1.47 ± 0.72||0.95 ± 0.48||7.7 × 10−03*||0.88|
|Physical activity level||1.55 ± 0.21||1.75 ± 0.21||9.1 × 10−04*||0.016|
|Fasting respiratory quotient||0.82 ± 0.04||0.8 ± 0.05||0.6||0.07|
|Fat oxidation index, mg/min/kg||1.00 ± 0.31||0.91 ± 0.27||0.016*||0.027|
|Carbohydrate oxidation index, mg/min/kg||1.34 ± 0.84||0.80 ± 0.59||0.022||0.16|
|REE, kcal/24 h||1255.14 ± 228.91||1466.83 ± 239.82||1.4 × 10−10*||0.62|
|TEE, kcal/24 h||1966.14 ± 458.79||2561.8 ± 591.65||2.5 × 10−10*||0.15|
|Total caloric intake, kcal/24 h||1919.71 ± 267.5||2007.39 ± 235.53||0.025||0.800|
|REE, kcal/24 h||1442.53 ± 386.89||1653.98 ± 331.45||8.10 × 10−08*||0.62|
|TEE, kcal/24 h||1819.77 ± 243.94||2180.2 ± 284.21||9.10 × 10−20*||0.88|
|Cholesterol, mmol/L||4.46 ± 0.78||4.13 ± 0.78||0.74||0.29|
|HDL-C, mmol/L||1.55 ± 0.36||1.36 ± 0.35||0.18||0.82|
|LDL-C, mmol/L||2.48 ± 0.76||2.30 ± 0.72||0.83||0.55|
|Triglycerides, mmol/L||0.93 ± 0.26||0.87 ± 0.29||0.59||0.33|
|Glycerol, μmol/L||31.24 ± 28.47||23.78 ± 20.08||0.55||0.19|
|NEFAs, μmol/L||389.97 ± 202.38||372 ± 223.82||0.29||0.71|
|Fasting glucose, mmol/L||4.66 ± 0.34||4.68 ± 0.44||0.81||0.46|
|Fasting insulin, mUI/L||6.07 ± 3.67||7.51 ± 2.75||5.7 × 10−04*||0.36|
|Blood biochemical parameters|
|IGF-1, µg/L||241.62 ± 86.59||254.8 ± 52.51||0.033||0.097|
|Leptin, µg/L||3.76 ± 2.23||9.34 ± 6.77||0.34||0.27|
|Metadrenaline, nmol/L||0.20 ± 0.09||0.16 ± 0.06||0.18||0.75|
|Normetadrenaline, nmol/L||0.38 ± 0.11||0.34 ± 0.1||0.11||0.063|
|TSH, mUI/L||2.38 ± 1.31||2.06 ± 1.01||0.38||0.33|
|Free T3, pmol/L||5.98 ± 1.99||5.44 ± 0.66||0.66||0.52|
|Free T4, pmol/L||16.83 ± 3.38||17.21 ± 1.68||0.99||0.2|
|Creatinine, mmol/L||10.05 ± 2.78||13.75 ± 3.6||4.7 × 10−11*||0.56|
|Urea, mmol/L||263.11 ± 95.08||385.52 ± 143.25||3.5 × 10−09*||0.17|
|Uric acid, mmol/L||2.62 ± 0.78||3.53 ± 0.94||7.0 × 10−09*||0.92|
Data are presented as means ± SDs. P values are assessed by repeated-measures ANOVA for outcomes adjusted for sex and age. A Bonferroni correction was applied for each set of measure to define a result as significant (defined by *). bpm, beats per minute; CT, constitutional thinness; HDL-C, HDL cholesterol; IGF-1, insulin-like growth factor 1; LDL-C, LDL cholesterol; NEFA, nonesterified fatty acid; REE, resting energy expenditure; TEE, total energy expenditure; TSH, thyrotropin; T3, triiodothyronine; T4, thyroxine.
Assessment of bone quality of the radius and tibia by HR-pQCT demonstrated that many bone quality parameters were lower in CT individuals compared with controls, with the main differences observed in tibia (Supplemental Table 2), confirming previous results in females (25). The lower bone quality was observed in both sexes.
At baseline, on the molecular level, circulating lipid metabolites, including cholesterol, triglycerides, free fatty acids, and fasting and postprandial kinetics of glucose and kinetics of insulin, were similar between the groups (Table 1, Supplemental Figure 3B and 3C), confirming previous results (2), while fasting insulin levels were lower in CT individuals.
Measured urinary metabolites, including urea, creatinine, and uric acid, were significantly lower in CT individuals than control subjects (FC 0.68, 0.73, and 0.74, respectively). Untargeted proteomic analyses of the plasma samples revealed 11 proteins that were differentially abundant between CT individuals and controls at baseline (Supplemental Table 3), including insulin-like growth factor binding protein 2 and dopamine β-hydroxylase (DBH) among the top plasma proteins with increased levels in CT individuals compared with controls.
Molecular analyses of adipose tissue and muscle demonstrate a distinct adipose-specific mitochondrial signature in CT
To characterize the metabolic state of CT individuals in more depth, we next performed analyses on 2 key metabolic tissues, adipose tissue and muscle. H&E staining of the sWAT did not reveal any visual difference in terms of extracellular matrix remodeling or inflammation. CT individuals had a marked smaller adipocyte size (black bars, 2174 ± 142 μm2) compared with controls (white bars, 3586 ± 216 μm2) at baseline (P < 0.001) (Figure 1A).
This decrease was reflected by a higher frequency of small adipocytes in CT individuals (Supplemental Figure 4). In addition, males overall had on average smaller adipocytes than their female counterparts (males, 2666 ± 253 μm2; females, 3067 ± 197 μm2) (P < 0.05) (Figure 1A).
We then evaluated the mitochondrial function through high-resolution respirometry analyses in both sWAT and skeletal muscle. The oxygen fluxes during leak respiration—independent of ADP-phosphorylating activity—were similar between CT individuals and controls (Figure 1B), suggesting that there was no difference in uncoupling. Immunohistochemical staining of UCP1 also revealed the lack of presence of UCP1 in the sWAT samples compared with a human BAT sample used as positive control (Supplemental Figure 5).
Interestingly, we observed significantly higher mitochondrial CII activity in CT individuals (FC 2.17, P = 0.001), as well as CI + CII activity (FC 1.71, P = 0.03) and maximal ETS activity (FC 2.07, P = 0.001) compared with controls. No significant difference in mitochondrial activity was detected between males and females.
To evaluate if the increase in CII and ETS activity was due to an increase in mitochondrial number, we performed a quantitative mtDNA assay (defined as ratio of mitochondrial compared with nuclear DNA).
The results showed a significantly higher amount of mtDNA per adipocyte (CT 784 ± 27 mtDNA molecules compared with control 675 ± 30 mtDNA molecules per cell) (P = 0.02) (Figure 1C), with no sex difference observed. This suggests that adipocytes of CT individuals have an increased number of mitochondria and that this contributes, at least in part, to the increased CII and ETS activity observed in CT.
In skeletal muscle, uncoupled respiration (leak) was not different in CT individuals compared with controls (Supplemental Figure 6). Similarly, maximal respiration in the coupled state, with electron input through CI alone, CI and CII, CII alone, or maximal ETS capacity, was comparable between CT individuals and controls, even though there was a tendency toward lower CI and CII and ETS capacity, but it was not statistically significant.
This suggests that the higher mitochondrial activity and content observed in sWAT of CT individuals are not due to a systemic difference in mitochondrial regulation, but rather are specific to the white adipose tissue.
Transcriptomic profiling of adipose tissue and muscle confirm altered lipid metabolism in CT
To gain further insights into the molecular basis of the observed differences in mitochondrial respiration and content, we profiled the RNA expression levels of the sWAT and muscle by total RNA-seq. Comparison of the expression profiles in sWAT revealed significant differences between the 2 groups (Figure 2A).
We did not observe any sex differences in the gene expression profiles. Using an FDR of 0.1 and a log2 FC threshold of 0.75, we identified 88 Ensembl gene IDs (ENSG) differentially expressed between CT individuals and controls, of which 62 corresponded to annotated genes.
To validate the gene expression results, we designed probes for those 62 genes and assessed gene expression by Nanostring technology. RNA-seq and Nanostring data amply correlated (Figure 2B, Spearman’s r = 0.8926, P < 0.0001), and 54 genes passed the threshold FDR of 0.1, corresponding to a validation of 87% of the RNA-seq results. Given the higher number of mitochondria found in CT, we first performed a GO cellular component enrichment analysis and found 21 terms that had a significant adjusted P value lower than 0.01, of which 9 belonged to the mitochondria (Supplemental Table 4), showing indeed an enrichment in genes belonging to the mitochondrial cellular component.
In parallel, pathway analyses of the entire data set revealed increased oxidative metabolism as evidenced by significant upregulation of fatty acid oxidation but strikingly also triglyceride biosynthesis (Figure 2C).
In addition, the IL-8 signaling pathway was reduced. IL-8 is a key proinflammatory mediator, known to be associated with obesity (26). Enrichment using the MGI Mammalian Phenotype identified nominal enrichment in abnormal lipid levels (MP0001547), decreased body fat amount (MP0010025), and other interesting enrichments such as altered angiogenesis (MP0000260) and improved glucose tolerance (MP0005292) (Supplemental Table 5).
Consistently, by zooming in on the specific top differentially expressed genes (Table 2), we found that many of the differences were associated with known positive metabolic outcomes in CT. These included decreased LEP, EGFL6, MMP9, and HSD11B1; elevated APOC1; and increased fatty acid synthesis, including elevated expression of FASN, SCD, ELOVL6, and AACS. Several antiangiogenic factors (THBS1, TNMD, and EGFL6), as well as oxidative stress factors (HSPB7, CRYAB, and UCHL1), were significantly downregulated. In addition, we identified multiple differentially expressed genes that have not been previously described for a potential role in metabolism. RNA-seq profiling in skeletal muscle samples did not reveal any significantly differentially expressed genes between CT individuals and controls (passing FDR of 0.1) (Supplemental Figure 7).
The lack of robust transcriptomic alterations in the muscle is in line with the fact that respiration in CT muscle is not significantly affected either. We therefore postulate that the molecular signature correlating with the CT phenotype is most likely specifically linked to an alteration in lipid metabolism in the adipose tissue rather than a systemic difference in mitochondrial biogenesis affecting other metabolically active tissues as well.
TABLE 2 – List of genes differentially expressed in adipose tissue at baseline from CT (n = 30) compared with control (n = 30) individuals at baseline1
|Ensembl gene ID||Symbol||Entrez gene name||Expression log2 FC||Expression FDR||Expression log2 FC||Expression FDR||Correlation FDR|
|ENSG00000198759||EGFL6||Epidermal GF-like domain multiple 6||–2.422||7.78 × 10−02||–1.829||1.17 × 10−02||7.10 × 10−10|
|ENSG00000119125||GDA||Guanine deaminase||–2.392||7.78 × 10−02||–1.947||1.87 × 10−02||2.50 × 10−13|
|ENSG00000123119||NECAB1||N-terminal EF-hand calcium binding protein 1||–2.355||7.78 × 10−02||–3.142||7.00 × 10−03||6.00 × 10−31|
|ENSG00000100985||MMP9||Matrix metallopeptidase 9||–1.83||8.02 × 10−02||–1.423||1.47 × 10−02||1.20 × 10−10|
|ENSG00000113389||NPR3||Natriuretic peptide receptor 3||–1.519||7.78 × 10−02||–1.683||5.49 × 10−03||2.20 × 10−16|
|ENSG00000079689||SCGN||Secretagogin, EF-hand calcium binding protein||–1.293||7.78 × 10−02||–0.627||5.43 × 10−02||2.20 × 10−16|
|ENSG00000174697||LEP||Leptin||–1.243||7.78 × 10−02||–1.214||3.98 × 10−03||2.20 × 10−16|
|ENSG00000154277||UCHL1||Ubiquitin C-terminal hydrolase L1||–1.218||7.78 × 10−02||–1.081||4.25 × 10−03||7.40 × 10−31|
|ENSG00000182255||KCNA4||Potassium voltage-gated channel subfamily A m4||–1.132||7.78 × 10−02||–1.407||6.41 × 10−03||2.20 × 10−19|
|ENSG00000183117||CSMD1||CUB and Sushi multiple domains 1||–1.122||8.02 × 10−02||–0.748||1.87 × 10−02||2.80 × 10−05|
|ENSG00000113739||STC2||Stanniocalcin 2||–1.108||7.78 × 10−02||–1.116||6.61 × 10−03||2.50 × 10−08|
|ENSG00000172995||ARPP21||cAMP-regulated phosphoprotein 21||–1.078||8.17 × 10−02||–1.089||5.49 × 10−03||1.30 × 10−04|
|ENSG00000137801||THBS1||Thrombospondin 1||–1.041||7.78 × 10−02||–1.020||3.98 × 10−03||2.20 × 10−16|
|ENSG00000163075||CFAP221||Cilia flagella-associated protein 221||–1.024||7.78 × 10−02||–1.135||3.98 × 10−03||2.20 × 10−16|
|ENSG00000000005||TNMD||Tenomodulin||–0.975||7.78 × 10−02||–0.768||5.36 × 10−03||2.20 × 10−16|
|ENSG00000158714||SLAMF8||SLAM family member 8||–0.975||9.79 × 10−02||–0.892||5.58 × 10−02||4.80 × 10−07|
|ENSG00000186439||TRDN||Triadin||–0.972||9.70 × 10−02||–1.188||1.24 × 10−02||2.20 × 10−16|
|ENSG00000178826||TMEM139||Transmembrane protein 139||–0.943||7.78 × 10−02||–1.049||1.47 × 10−02||8.90 × 10−14|
|ENSG00000099957||P2RX6||Purinergic receptor P2 × 6||–0.919||7.78 × 10−02||–1.229||5.22 × 10−03||1.60 × 10−10|
|ENSG00000110092||CCND1||Cyclin D1||–0.893||7.78 × 10−02||–0.804||4.30 × 10−03||2.20 × 10−16|
|ENSG00000164188||RANBP3L||RAN binding protein 3 like||–0.868||7.78 × 10−02||–1.020||6.41 × 10−03||2.20 × 10−16|
|ENSG00000109846||CRYAB||Crystallin αB||–0.858||7.78 × 10−02||–0.935||3.98 × 10−03||2.20 × 10−16|
|ENSG00000117154||FCGBP||Immunoglobin superfamily member 21||–1.3181||8.97 × 10−02||–0.921||4.49 × 10−02||8.40 × 10−08|
|ENSG00000188778||ADRB3||Adrenoceptor β3||–0.824||9.09 × 10−02||–0.720||4.82 × 10−03||2.20 × 10−16|
|ENSG00000102359||SRPX2||Sushi repeat containing protein, X-linked 2||–0.816||7.78 × 10−02||–0.852||3.98 × 10−03||4.90 × 10−31|
|ENSG00000173641||HSPB7||Heat shock protein family B (small) m7||–0.811||7.78 × 10−02||–0.813||3.98 × 10−03||2.20 × 10−16|
|ENSG00000117594||HSD11B1||Hydroxysteroid 11-β dehydrogenase 1||–0.803||8.99 × 10−02||–0.704||2.54 × 10−02||2.20 × 10−16|
|ENSG00000147852||VLDLR||Very-low-density lipoprotein receptor||–0.767||8.17 × 10−02||–0.883||4.82 × 10−03||2.20 × 10−16|
|ENSG00000077274||CAPN6||Calpain 6||0.76||7.78 × 10−02||0.777||3.98 × 10−03||2.20 × 10−16|
|ENSG00000169710||FASN||Fatty acid synthase||0.769||8.58 × 10−02||0.727||5.49 × 10−03||2.20 × 10−16|
|ENSG00000206384||COL6A6||Collagen type VI α6 chain||0.772||7.78 × 10−02||0.903||4.55 × 10−03||2.20 × 10−16|
|ENSG00000183798||EMILIN3||Elastin microfibril interfacer 3||0.811||9.27 × 10−02||0.449||5.96 × 10−02||9.30 × 10−05|
|ENSG00000081760||AACS||Acetoacetyl-CoA synthetase||0.816||8.60 × 10−02||0.783||7.52 × 10−03||2.20 × 10−16|
|ENSG00000166126||AMN||Amnion-associated transmembrane protein||0.855||7.78 × 10−02||0.568||7.63 × 10−02||1.70 × 10−04|
|ENSG00000112394||SLC16A10||Solute carrier family 16 member 10||0.866||7.78 × 10−02||1.447||3.98 × 10−03||8.60 × 10−13|
|ENSG00000060709||RIMBP2||RIMS binding protein 2||0.87||9.10 × 10−02||0.708||1.12 × 10−02||2.20 × 10−16|
|ENSG00000253379||EYA1||EYA transcriptional coactivator and phosphatase 1||0.902||7.78 × 10−02||1.419||3.98 × 10−03||1.50 × 10−17|
|ENSG00000099194||SCD||Stearoyl-CoA desaturase||0.911||7.78 × 10−02||1.100||3.98 × 10−03||2.20 × 10−16|
|ENSG00000124003||MOGAT1||Monoacylglycerol O-acyltransferase 1||0.912||8.02 × 10−02||1.213||3.98 × 10−03||2.20 × 10−16|
|ENSG00000168702||LRP1B||LDL receptor-related protein 1B||0.919||7.78 × 10−02||0.717||5.44 × 10−02||2.20 × 10−16|
|ENSG00000130208||APOC1||Apolipoprotein C1||0.92||7.78 × 10−02||0.870||2.76 × 10−02||2.20 × 10−16|
|ENSG00000140284||SLC27A2||Solute carrier family 27 member 2||0.926||9.88 × 10−02||1.132||1.32 × 10−02||2.20 × 10−16|
|ENSG00000113494||PRLR||Prolactin receptor||0.979||7.78 × 10−02||1.080||5.27 × 10−03||2.20 × 10−16|
|ENSG00000198963||RORB||RAR-related orphan receptor B||0.986||7.78 × 10−02||0.877||5.49 × 10−03||2.20 × 10−16|
|ENSG00000138678||GPAT3||Glycerol-3-phosphate acyltransferase 3||1.065||7.78 × 10−02||1.087||9.85 × 10−03||2.20 × 10−16|
|ENSG00000164879||CA3||Carbonic anhydrase 3||1.072||7.78 × 10−02||1.327||7.52 × 10−03||2.20 × 10−16|
|ENSG00000135744||AGT||Angiotensinogen||1.087||7.78 × 10−02||1.116||4.82 × 10−03||2.20 × 10−16|
|ENSG00000143365||RORC||RAR-related orphan receptor C||1.098||7.78 × 10−02||1.660||3.98 × 10−03||2.20 × 10−16|
|ENSG00000145428||RNF175||Ring finger protein 175||1.098||7.78 × 10−02||1.119||1.87 × 10−02||7.60 × 10−09|
|ENSG00000170522||ELOVL6||ELOVL fatty acid elongase 6||1.167||7.78 × 10−02||1.204||8.19 × 10−02||2.20 × 10−16|
|ENSG00000160339||FCN2||Ficolin 2||1.181||7.78 × 10−02||1.294||5.36 × 10−03||2.20 × 10−16|
|ENSG00000015520||NPC1L1||NPC1-like intracellular cholesterol transporter 1||1.276||7.78 × 10−02||0.765||3.50 × 10−02||5.50 × 10−05|
|ENSG00000103460||TOX3||TOX high-mobility group box family member 3||1.282||7.78 × 10−02||1.378||8.73 × 10−02||1.70 × 10−11|
|ENSG00000172425||TTC36||Tetratricopeptide repeat domain 36||1.295||7.78 × 10−02||1.260||6.41 × 10−03||3.50 × 10−10|
Genes passing an FDR of 0.1 and a log2 FC threshold of 0.75 are reported here. Differential expression was tested by using a logistic regression model correcting for age and sex followed by a Wald test to assess the statistical significance of the coefficient of interest. As no sex interaction was found, association was evaluated without interaction term. Benjamini–Hochberg multiple testing correction was applied within the fitted model on the derived P values (FDR). Correlation tests between RNA-seq and Nanostring gene quantification were performed using a nonparametric Spearman correlation test. CT, constitutional thinness; FC, fold change; FDR, false discovery rate; RNA-seq, RNA sequencing.
Short-term overfeeding did not result in a different metabolic response in CT subjects
To investigate whether the metabolic response to overfeeding was different in CT compared with controls, the participants were next given a 600-kcal surplus to their diet for 14 d (300-mL shake in the evening, containing 30 g protein, 72 g carbohydrate, and 21 g fat). Overall compliance to the overfeeding regimen was confirmed by a significant increase in urea excretion in both groups relative to the baseline (FC 1.22, P < 0.001) (Supplemental Table 6).
Both groups gained a small but significant amount of weight during the overfeeding (P < 0.001), but no significant difference in weight gain was observed between the 2 groups (Supplemental Figure 8).
Even though some clinical and energy balance parameters were altered upon overfeeding (Supplemental Table 6), no differences in response to the overfeeding were observed in CT individuals compared with controls. Similarly, tissue respiration (Figure 1B, Supplemental Figure 6) and transcriptomic profiling (Supplemental Figures 9 and 10), as well as plasma proteomics (Supplemental Table 3), all showed some differences upon the intervention, but none demonstrated a differential response to the overfeeding between CT individuals compared with controls.
Together, these results demonstrate that the molecular response to the overfeeding is similar in CT individuals compared with controls and that the main molecular differences observed at baseline remain present after the overfeeding.
nqz144_Supplemental_File – pdf file
- Apfelbaum M, Sachet P. [Constitutional thinness]. Rev Prat. 1982;32(3):245–7.(in French).
- Germain N, Galusca B, Le Roux CW, Bossu C, Ghatei MA, Lang F, Bloom SR, Estour B. Constitutional thinness and lean anorexia nervosa display opposite concentrations of peptide YY, glucagon-like peptide 1, ghrelin, and leptin. Am J Clin Nutr. 2007;85(4):967–71.
- Estour B, Marouani N, Sigaud T, Lang F, Fakra E, Ling Y, Diamondé A, Minnion JS, Galusca B, Germain N. Differentiating constitutional thinness from anorexia nervosa in DSM 5 era. Psychoneuroendocrinology. 2017;84:94–100Gallagher D, Heymsfield SB, Heo M, Jebb SA, Murgatroyd PR, Sakamoto Y. Healthy percentage body fat ranges: an approach for developing guidelines based on body mass index. Am J Clin Nutr. 2000;72(3):694–701.
- Bossu C, Galusca B, Normand S, Germain N, Collet P, Frere D, Lang F, Laville M, Estour B. Energy expenditure adjusted for body composition differentiates constitutional thinness from both normal subjects and anorexia nervosa. Am J Physiol Endocrinol Metab. 2007;292(1):E132–7.
- Germain N, Galusca B, Caron-Dorval D, Martin JF, Pujos-Guillot E, Boirie Y, Khalfallah Y, Ling Y, Minnion JS, Bloom SR et al. . Specific appetite, energetic and metabolomics responses to fat overfeeding in resistant-to-bodyweight-gain constitutional thinness. Nutr Diabetes. 2014;4:e126.
- Ling Y, Galusca B, Hager J, Feasson L, Valsesia A, Epelbaum J, Alexandre V, Wynn E, Dinet C, Palaghiu R et al. . Rational and design of an overfeeding protocol in constitutional thinness: understanding the physiology, metabolism and genetic background of resistance to weight gain. Ann Endocrinol (Paris). 2016;77(5):563–9.
- Crouter SE, Churilla JR, Bassett DR Jr. Accuracy of the Actiheart for the assessment of energy expenditure in adults. Eur J Clin Nutr. 2008;62(6):704–11.
- Blond E, Maitrepierre C, Normand S, Sothier M, Roth H, Goudable J, Laville M. A new indirect calorimeter is accurate and reliable for measuring basal energy expenditure, thermic effect of food and substrate oxidation in obese and healthy subjects. e-SPEN. 2011;6(1):e7–15.
- Lazzer S, Boirie Y, Montaurier C, Vernet J, Meyer M, Vermorel M. A weight reduction program preserves fat-free mass but not metabolic rate in obese adolescents. Obes Res. 2004;12(2):233–40.
- Montaurier C, Morio B, Bannier S, Derost P, Arnaud P, Brandolini-Bunlon M, Giraudet C, Boirie Y, Durif F. Mechanisms of body weight gain in patients with Parkinson’s disease after subthalamic stimulation. Brain. 2007;130(Pt 7):1808–18.
- Liu Y, Zhu SH, Wang GH, Ye F, Li PZ. Validity and reliability of multiparameter physiological measurements recorded by the Equivital LifeMonitor during activities of various intensities. J Occup Environ Hyg. 2013;10(2):78–85.
- Dayon L, Núñez Galindo A, Cominetti O, Corthésy J, Kussmann M. A highly automated shotgun proteomic workflow: clinical scale and robustness for biomarker discovery in blood. Methods Mol Biol. 2017;1619:433–49.
- Lan J, Núñez Galindo A, Doecke J, Fowler C, Martins RN, Rainey-Smith SR, Cominetti O, Dayon L. Systematic evaluation of the use of human plasma and serum for mass-spectrometry-based shotgun proteomics. J Proteome Res. 2018;17(4):1426–35.
- Canto C, Garcia-Roves PM. High-resolution respirometry for mitochondrial characterization of ex-vivo mouse tissues. Curr Protoc Mouse Biol. 2015;5(2):135–53.
- Eigentler AE, Fontana-Ayoub M, Fasching M, Lassnig B, Stadlmann S, Rieger G, Haffner B, Lemieux H, Gnaiger E. Selected media and chemicals for respirometry with mitochondria and permeabilized cells. Mitochondrial Physiol Netw. 2012;3:1–9.
- Fasching M, Fontana-Ayoub M, Gnaiger E. Mitochondrial respiration medium—MiR6. Mitochondrial Physiol Netw. 2013;14:1–5.
- Galarraga M, Campión J, Muñoz-Barrutia A, Boqué N, Moreno H, Martínez JA, Milagro F, Ortiz-de-Solórzano C. Adiposoft: automated software for the analysis of white adipose tissue cellularity in histological sections. J Lipid Res. 2012;53(12):2791–6.
- Marquis J, Lefebvre G, Kourmpetis YAI, Kassam M, Ronga F, De Marchi U, Wiederkehr A, Descombes P. MitoRS, a method for high throughput, sensitive, and accurate detection of mitochondrial DNA heteroplasmy. BMC Genomics. 2017;18(1):326.
- Atger F, Gobet C, Marquis J, Martin E, Wang J, Weger B, Lefebvre G, Descombes P, Naef F, Gachon F. Circadian and feeding rhythms differentially affect rhythmic mRNA transcription and translation in mouse liver. Proc Natl Acad Sci USA. 2015;112(47):E6579–88.
- Armenise C, Lefebvre G, Carayol J, Bonnel S, Bolton J, Di Cara A, Gheldof N, Descombes P, Langin D, Saris WH et al. . Transcriptome profiling from adipose tissue during a low-calorie diet reveals predictors of weight and glycemic outcomes in obese, nondiabetic subjects. Am J Clin Nutr. 2017;106(3):736–46.
- Law CW, Chen Y, Shi W, Smyth GK. voom: precision weights unlock linear model analysis tools for RNA-seq read counts. Genome Biol. 2014;15(2):R29.
- Kuleshov MV, Jones MR, Rouillard AD, Fernandez NF, Duan Q, Wang Z, Koplev S, Jenkins SL, Jagodnik KM, Lachmann A et al. . Enrichr: a comprehensive gene set enrichment analysis web server 2016 update. Nucleic Acids Res. 2016;44(W1):W90–7.
- Heizer WD. Normal and abnormal intestinal absorption by humans. Environ Health Perspect. 1979;33:101–6.
- Galusca B, Zouch M, Germain N, Bossu C, Frere D, Lang F, Lafage-Proust MH, Thomas T, Vico L, Estour B. Constitutional thinness: unusual human phenotype of low bone quality. J Clin Endocrinol Metab. 2008;93(1):110–7.
- Sharabiani MT, Vermeulen R, Scoccianti C, Hosnijeh FS, Minelli L, Sacerdote C, Palli D, Krogh V, Tumino R, Chiodini P et al. . Immunologic profile of excessive body weight. Biomarkers. 2011;16(3):243–51.
- Galusca B, Verney J, Meugnier E, Ling Y, Edouard P, Feasson L, Ravelojaona M, Vidal H, Estour B, Germain N. Reduced fibre size, capillary supply and mitochondrial activity in constitutional thinness’ skeletal muscle. Acta Physiol (Oxf). 2018;224(3):e13097.
- Boudina S, Graham TE. Mitochondrial function/dysfunction in white adipose tissue. Exp Physiol. 2014;99(9):1168–78.
- Arner P, Andersson DP, Bäckdahl J, Dahlman I, Rydén M. Weight gain and impaired glucose metabolism in women are predicted by inefficient subcutaneous fat cell lipolysis. Cell Metab. 2018;28(1):45–54.e3.
- Galgani J, Ravussin E. Energy metabolism, fuel selection and body weight regulation. Int J Obes (Lond). 2008;32(Suppl 7):S109–19.
- Guarente L. Mitochondria—a nexus for aging, calorie restriction, and sirtuins?. Cell. 2008;132(2):171–6.
- Flachs P, Rossmeisl M, Kuda O, Kopecky J. Stimulation of mitochondrial oxidative capacity in white fat independent of UCP1: a key to lean phenotype. Biochim Biophys Acta. 2013;1831(5):986–1003.
- Dulloo AG, Jacquet J, Montani JP, Schutz Y. Adaptive thermogenesis in human body weight regulation: more of a concept than a measurable entity?. Obes Rev. 2012;13(Suppl 2):105–21.
- Landsberg L. Core temperature: a forgotten variable in energy expenditure and obesity?. Obes Rev. 2012;13(Suppl 2):97–104.
- Heikens MJ, Gorbach AM, Eden HS, Savastano DM, Chen KY, Skarulis MC, Yanovski JA. Core body temperature in obesity. Am J Clin Nutr. 2011;93(5):963–7.
- Hoffmann ME, Rodriguez SM, Zeiss DM, Wachsberg KN, Kushner RF, Landsberg L, Linsenmeier RA. 24-h core temperature in obese and lean men and women. Obesity (Silver Spring). 2012;20(8):1585–90.
- Monnard CR, Fares EJ, Calonne J, Miles-Chan JL, Montani JP, Durrer D, Schutz Y, Dulloo AG. Issues in continuous 24-h core body temperature monitoring in humans using an ingestible capsule telemetric sensor. Front Endocrinol (Lausanne). 2017;8:130.
- Arner P, Bernard S, Salehpour M, Possnert G, Liebl J, Steier P, Buchholz BA, Eriksson M, Arner E, Hauner H et al. . Dynamics of human adipose lipid turnover in health and metabolic disease. Nature. 2011;478(7367):110–3.
- Bruss MD, Khambatta CF, Ruby MA, Aggarwal I, Hellerstein MK. Calorie restriction increases fatty acid synthesis and whole body fat oxidation rates. Am J Physiol Endocrinol Metab. 2010;298(1):E108–16.
- Vallerand AL, Zamecnik J, Jones PJ, Jacobs I. Cold stress increases lipolysis, FFA Ra and TG/FFA cycling in humans. Aviat Space Environ Med. 1999;70(1):42–50.
- Rush RA, Geffen LB. Dopamine β-hydroxylase in health and disease. CRC Crit Rev Clin Lab Sci. 1980;12(3):241–77.
- Ryoichi I, Furuno T, Ohnishi T, Okada A. Obesity and serum dopamine-β-hydroxylase activity. Jpn J Health Hum Ecol. 1985;51(5):202–6.
- Zeng W, Pirzgalska RM, Pereira MM, Kubasova N, Barateiro A, Seixas E, Lu YH, Kozlova A, Voss H, Martins GG et al. . Sympathetic neuro-adipose connections mediate leptin-driven lipolysis. Cell. 2015;163(1):84–94.
- Pasanisi F, Pace L, Fonti R, Marra M, Sgambati D, De Caprio C, De Filippo E, Vaccaro A, Salvatore M, Contaldo F. Evidence of brown fat activity in constitutional leanness. J Clin Endocrinol Metab. 2013;98(3):1214–8.
- Rosen ED, Spiegelman BM. What we talk about when we talk about fat. Cell. 2014;156(1–2):20–44.
- Alligier M, Meugnier E, Debard C, Lambert-Porcheron S, Chanseaume E, Sothier M, Loizon E, Hssain AA, Brozek J, Scoazec JY et al. . Subcutaneous adipose tissue remodeling during the initial phase of weight gain induced by overfeeding in humans. J Clin Endocrinol Metab. 2012;97(2):E183–92.
More information:Cell, Orthofer et al.: “Identification of ALK in thinness” www.cell.com/cell/fulltext/S0092-8674(20)30497-9 , DOI: 10.1016/j.cell.2020.04.034