Jensen, Michael D.

Objectives: To determine whether meal fatty acid (FA) metabolism differs by sex and obesity phenotypes and to examine how weight loss (WtL) through diet and exercise affects meal FA trafficking. Methods: Before WtL (study 1) 13 obese men (ObM), 9 upper body obese (UbO) and 8 lower body obese (LbO) women consumed a standard meal containing [ 3 H]triolein followed by subcutaneous abdominal and gluteal fat biopsies. Indirect calorimetry was measured by 5 h substrate oxidation. Whole/regional fat distribution was measured using DXA and CT scan. Participants then enrolled in a structured WtL program, after which 23 returned for study 2. Results: There were no differences (p=0.32) in meal FA oxidation between groups. Abdominal uptake of meal FA in LOb was greater than in UOb (p=0.026) and ObM (p<0.001). LOb took up more meal FA in the lower body than ObM (p=0.001). Participants who returned for study 2 lost 14 ± 1.1 kg. After WtL, uptake of meal FA in upper body subcutaneous fat increased (p=0.028) in UOb and ObM (p=0.046) and decreased in lower body fat (p=0.025) in ObM. WtL led to reduced 5 h postprandial total fat oxidation and increased 24 h meal FA oxidation. Conclusion: Changes in uptake of meal FA after WtL followed a pattern indicative of body fat distribution maintenance. These results highlight the potential importance of meal fat uptake in modulating body fat distribution in obesity. Funding: NIH RO1 DK45343, DK50456 and R00585.

This study examined the plasma FFA, intramuscular long chain acylcarnitine (LCAC) fatty acids and imTG stores during basal and hyperinsulinemic conditions. Methods: Endurance trained (TR, n=8) and sedentary individuals (SED, n=15) were studied under overnight postabsorptive and euglycemic hyperinsulinemic clamp (1.0mU·kg FFM −1 ·min −1 ) conditions. FFA kinetics were measured using a 5‐h infusion of [U‐ 13 C]oleate, for labeling of imTG and LCAC pools. Muscle biopsies were performed to measure imTG and imLCAC enrichment at 1400 h after 5‐h of infusion. imTG concentrations were greater in TR than SED individuals (P<0.01). Table SED Trained Saline** Clamp Saline** Clamp plasma oleate FFA (mpe) 0.210±0.018* 0.271±0.036* 0.189±0.026* 0.361±0.050* imLCAC (mpe) 0.025±0.003 0.005±0.002 0.016±0.005 0.002±0.002 imTG (mpe) 0.018±0.003† 0.011±0.001† 0.024±0.004† 0.013±0.001† P<0.001 saline vs. clamp; *P<0.000 between compartment (plasma vs. imLCAC vs. imTG); †P<0.000 imLCAC and imTG Methods: Endurance trained (TR, n=8) and sedentary individuals (SED, n=15) were studied under overnight postabsorptive and euglycemic hyperinsulinemic clamp (1.0mU·kg FFM −1 ·min −1 ) conditions. FFA kinetics were measured using a 5‐h infusion of [U‐ 13 C]oleate, for labeling of imTG and LCAC pools. Muscle biopsies were performed to measure imTG and imLCAC enrichment at 1400 h after 5‐h of infusion. Conclusion: The imTG contributes more to the imLCAC pools than plasma FFA both in the basal and hyperinsulinemic state, and hyperinsulinemia suppressed oleate FFA entry into the imLCAC pools, more so in trained than untrained adults. Muscle fatty acid oxidation is largely via obligatory cycling of fatty acids through imTG or other intracellular complex lipid pools.

An upper body (UB) fat distribution is correlated with insulin resistance (IR). It is not known whether IR is caused by an UB fat distribution or if IR predisposes to UB fat gain. We recruited 25 lean volunteers to gain ~2 kg of body fat over ~8 weeks. We hypothesized that IR individuals would gain more UB fat than insulin sensitive individuals. At baseline, insulin sensitivity was assessed by 24‐hour insulin AUC over a meal controlled inpatient stay. Meal fat disposal was measured by the meal fat tracer/fat biopsy approach. Participants were grouped into tertiles based on their AUC insulin: least insulin sensitive tertile 1 (T1) = 37,449±2600; middle tertile 2 (T2) = 22,11±3002; most insulin sensitive tertile 3 (T3)= 13,233±2452 uU/ml/24h (mean±sem). At baseline, volunteers in the different tertiles had significantly different % body fat (T1=29±2%, T2=22±2%, and T3=18±2%; P<0.005). The T1 had more upper body subcutaneous (UBSQ) fat at baseline (8.3±0.5 kg) than the T3 (6.1±0.5 kg; P<0.05). Visceral and lower body subcutaneous (LBSQ) fat at baseline did not differ significantly between tertiles. Baseline meal fat storage in LBSQ fat was greater in T1 than T3 (20.9±3.4% vs. 9.0±1.9%, P < 0.05). Changes in regional fat gain as a percentage of total fat gain was not different by tertiles (53.3±2.7%, 8.3±2.0%, and 37.8±1.8%, for UBSQ, visceral, and LBSQ). Baseline insulin sensitivity was not predictive of the proportion of fat gain in the visceral or UBSQ regions. This argues against the hypothesis that insulin resistance results in UB fat gain. The alternative explanation is that those who gain UB fat eventually develop IR. T3 stored the least meal fat in the LBSQ region, suggesting that regional fat depots have differing responses to the prevailing insulinemia, perhaps through regulation of LPL activity.

Women differ from men by the larger lower‐body subcutaneous fat mass and the tendency to gain more fat in that depot if obesity develops. We hypothesized that these differences in fat distribution may depend on preadipocyte availability. We identified the proportion of preadipocytes from the stromovascular (SV) cells in adipose tissue by immunofluorescent staining for aP2 (the adipocyte fatty acid trafficking protein) and CD68 (a macrophage marker). Macrophages are the other cell type, aside from preadipocytes, in the SV fraction of adipose tissue that express aP2. We determined the percent preadipocytes in the SV fraction of subcutaneous abdominal and femoral adipose tissue of 17 premenopausal lean women and 10 age‐matched lean men. Body fat compartments were measured by DEXA and CT scans of the abdomen. Compared to men, women had less visceral fat and more subcutaneous fat mass, predominantly in the lower‐body compartment: 7.0 ± 0.3 kg vs. 5.2 ± 0.4 kg, respectively (p < 0.01). Femoral adipocytes were significantly larger than abdominal adipocytes in both groups (lipid per fat cell in women and men: abdomen, 0.46 ± 0.03 and 0.44 ± 0.06 μg/cell; thigh, 0.66 ± 0.04 and 0.55 ± 0.08 μg/cell, respectively). The percent preadipocytes in the SV fraction from femoral subcutaneous adipose tissue was greater in women than in men (29.7 ± 2.6% vs. 19.3 ± 2.0%, p < 0.02) but similar when isolated from the abdominal depot (29.9 ± 3.3% vs. 27.2 ± 1.9%). Our findings suggest that the fraction of femoral stromovascular cells as preadipocytes is a potential factor for sex‐dependent differences in regional fat distribution. The study was funded by RO1 DK 45343‐13S to M. Jensen.

The thermic effect of food (TEF) is partially mediated by β‐adrenergic receptors (βAR) and is an important component of daily energy expenditure. Greater sympathetic activity has been associated with decreased BAR responsiveness. We hypothesized that muscle sympathetic nerve activity (MSNA) would be inversely correlated with the βAR component of TEF (TEF β ). We measured MSNA (microneurography) and energy expenditure (EE) (indirect calorimetry) before and for 5 h after a mixed meal (40% daily EE). Measurements were made during systemic βAR blockade with propranolol (BB) and saline (NS) infusion on separate days in 10 healthy subjects (2 female, age 28±2 y, BMI 24.4±0.6 kg/m 2 ). Resting MSNA was 39±3 bursts/100 heartbeats (hb). Within 1 h after the meal, MSNA significantly increased during both NS (12±3 bursts/100 hb) and BB (18±4) but was not significantly different between the two days, (P=0.375). The maximum increase in REE after the meal was 269±29 (NS) and 198±22 (BB) kcal/24 h (P<0.05). TEF β (TEF β = AUC TEF NS – AUC TEF BB ) was 48±20% of total TEF. TEF β was positively correlated with resting MSNA (R 2 =0.70, P<0.05). Contrary to our hypothesis, we found that greater levels of MSNA were associated with greater TEF β . These results indicated that resting MSNA may be an important contributor to daily energy expenditure. Studies are needed to determine if greater MSNA is associated with a resistance to weight gain. The project described was supported by the Mayo Foundation for Medical Education and Research and NIH Grant Numbers K23 DK82424, P30 DK50456, and UL1 RR024150.

Background Expression of adiponectin, an anti‐atherogenic and insulin‐sensitizing adipokine, is decreased in obesity. Mechanisms underlying this reduction are not well understood. Increased adiponectin in obesity may protect against development of cardiac and metabolic disorders. Objective To investigate the role of leptin in regulating adiponectin expression in humans. Design 1) determining the relationship between leptin and adiponectin in healthy humans during modest weight gain; 2) in‐vitro validation of the role of leptin in adiponectin expression. Results Modest weight gain of 3.8 ± 1.2 kg over 8 weeks of overfeeding in normal weight healthy subjects (n=33, 22 men, 30±7 years) increased both leptin and adiponectin. Changes in leptin during weight gain correlated with changes in adiponectin (rho=0.35, p=0.045), suggesting that leptin may regulate adiponectin expression. In cultured differentiated human white preadipocytes, leptin increased adiponectin mRNA and protein expression. Conclusions Leptin induces adiponectin expression. In obesity, adipose tissue may be resistant to leptin signaling thus preventing increases in adiponectin. Therefore, impaired leptin signaling may contribute to low adiponectin in obesity. Targeting leptin resistance may be beneficial in preventing obesity‐associated cardiac and metabolic disorders. Support : AHA 11SDG7260046, R21 DK81014

Introduction: Senescent cells with essentially irreversible loss of replicative potential accumulate in various tissues with aging. Cellular senescence is associated with a pro‐inflammatory senescence‐associated secretory phenotype which the Janus‐associated kinase (JAK) pathway plays a role in. We hypothesized that senescent preadipocytes contribute to age‐related dysfunction of healthy cells. Methods: Preadipocytes from lean subjects were X‐irradiated to induce senescence. We co‐cultured non‐senescent cells with senescent cells or cultured them in conditioned medium (CM) from senescent preadipocytes to test the inflammation, differentiation and insulin sensitivity. Macrophage migration in response to CM was assayed as well. Moreover, we treated old or obese mice with Ruxolitinib, a JAK1/2 inhibitor, to test the beneficial effect. Results: The adipogenesis of preadipocytes was impaired by exposure to senescent cells. Macrophage migration was greater in response to senescent preadipocytes. Inflammation was increased in both adipose tissue and preadipocytes exposed to CM from senescent cells. All these adverse effect could be partially rescued by JAK inhibitor. In addition, senescent preadipocytes inhibited insulin‐induced phosphor‐AKT in adipose tissue and HepG2 cells. Insulin‐induced glycogen synthesis was also inhibited in HepG2 cells. Two months of Ruxolitinib treatment reduced inflammation, reversed age‐related lipodystrophy and improved physical function in frail, 24 month old mice. It also improved insulin sensitivity in diet‐induced obese mice. Conclusion: Senescent preadipocytes contribute to age‐related impairment in adipogenesis, fat tissue inflammation, and insulin resistance. Grant Funding Source : supported by NIH AG13925 and Robert and Arlene Kogod Center on Aging in Mayo Clinic.

Insulin has been shown to have major impact on muscle mitochondrial function. Recent data show that ceramide can decrease oxygen consumption and ATP production in isolated mitochondria. The aim of the present study was to determine whether insulin deficiency alters ceramide (CER) and its metabolites ‐ sphinganine (SPA), sphingosine (SPH) and sphingosine‐1‐phosphate (S1P) ‐ in mitochondrial (mt) and sarcoplasmic (src) fractions from quadriceps muscle of streptozotocin‐diabetic C57BL/6 male mice. We studied non‐diabetic controls (C); diabetic, insulin‐treated (D+); and diabetic, insulin‐deprived (D−). D− and D+ group received LinBit insulin implants, whereas C group received blank implants. Implants from D− group were removed 7 days prior to sacrifice. CER and its metabolites were measured in a tandem mass spectrometer connected to a UPLC system. Insulin deprivation increased CER in muscle src although decreased in mt. Insulin treatment normalized CER content in src, but further lowered CER content in mt, which was accompanied by elevation in mitochondrial SPH content. Changes in SPA and S1P levels in both the src and mt were non‐significant. In conclusion, diabetes and insulin have profound effects on CER concentrations in muscle subcellular fractions. C D− D+ src mt src mt src mt SPA Mean 0.34 nd 0.32 nd 0.34 nd SEM ±0.04 ±0.02 ±0.04 SPH Mean 0.44 1.63 0.61 2.15 0.47 2.55 SEM ±0.07 ±0.23 ±0.10 ±0.14 ±0.03 ±0.21 S1P Mean 0.32 1.14 0.27 1.06 0.23 0.91 SEM ±0.03 ±0.2 ±0.05 ±0.14 ±0.01 ±0.1 CER Mean 85.7 326.0 162.5 267.3 93.0 199.7 SEM ±17.4 ±38.4 ±19.8 ±24.0 ±4.4 ±11.2

AbstractObjectiveAdipose tissue (AT) senescence is associated with AT dysfunction in rodents, but little is known about human AT senescence. The study goal was to define the distribution of senescent cells in two subcutaneous depots and understand relationships with adiposity and inflammation.MethodsSixty‐three volunteers (48 females) underwent abdominal and femoral subcutaneous fat biopsies. Fat cell size, senescent cells using senescence‐associated β‐galactosidase staining per 100 nucleated cells (percentage), and mRNA expression of four cytokines were measured.ResultsThere was a larger proportion of senescent cells in femoral than abdominal subcutaneous AT (mean difference 1.6% [95% CI: 0.98%‐2.3%], p < 0.001), and the percentage of femoral AT senescent cells was greater in women than men (median 3.9% vs. 2.1%, p < 0.01). There was a positive correlation between senescence and fat cell size in abdominal (rs = 0.44, p < 0.001) and femoral (rs = 0.35, p = 0.007) AT depots. Abdominal AT tumor necrosis factor alpha (rs = 0.49, p < 0.01) and interleukin‐1β (rs = 0.44, p = 0.01) expression was positively correlated with abdominal, but not femoral, AT senescence.ConclusionsIn human subcutaneous AT, there are more senescent cells in femoral than abdominal depots; abdominal AT senescent cells are more associated with inflammatory signals than femoral AT senescent cells.

Adipose inflammation is proposed to be at the nexus of the systemic inflammation and metabolic derangements associated with obesity. The present study provides evidence to support adipose inflammation as a central feature of the pathophysiology of obesity. Adipose inflammation is associated with systemic and peripheral metabolic derangements, including increased systemic inflammation, reduced insulin sensitivity and reduced skeletal muscle mitochondrial respiration.