ABSTRACT

De novo lipogenesis (DNL) converts carbon substrates to lipids. Increased hepatic DNL could contribute to pathogenic liver triglyceride accumulation in nonalcoholic steatohepatitis (NASH) and therefore may be a potential target for pharmacological intervention. Here, we measured hepatic DNL using heavy water in 123 NASH patients with fibrosis or cirrhosis, calculated the turnover of hepatic triglycerides to allow repeat labeling studies and determined the associations of hepatic DNL with metabolic, fibrotic, and imaging markers. We found that hepatic DNL was higher in fibrotic NASH patients [median (IQR), 40.7% contribution to palmitate (32.1, 47.5), n=103] than has been previously reported in healthy volunteers and remained elevated [median (IQR), 36.8% (31.0, 44.5), n=20] in patients with cirrhosis, despite lower liver fat content. We also showed that turnover of intrahepatic triglyceride pools was slow (t½ >10 days). Furthermore, DNL contribution was determined to be independent of liver stiffness by magnetic resonance imaging, but was positively associated with the number of large very low-density lipoprotein (VLDL) particles, the size of VLDL, the lipoprotein insulin resistance score, and levels of ApoB100 (r=0.6, p=0.07), and trended towards negative associations with the fibrosis markers FIB-4, FibroSure and APRI. Finally, we found treatment with the acetyl-CoA carboxylase inhibitor firsocostat reduced hepatic DNL at 4 and 12 weeks, using a correction model for residual label that accounts for hepatic triglyceride turnover. Taken together, these data support an important pathophysiological role for elevated hepatic DNL in NASH, and demonstrate that response to pharmacological agents targeting DNL can be correlated with pre-treatment DNL.

INTRODUCTIONHepatic de novo lipogenesis (DNL) is an essential biosynthetic pathway through which non-lipid energy substrates are converted to lipid species via the activities of several key enzymes, including acetyl-CoA carboxylase (ACC), ATP-citrate lyase, and fatty acid synthase ((1)Jensen-Urstad A.P.L. Semenkovich C.F. Fatty acid synthase and liver triglyceride metabolism: Housekeeper or messenger?.,(2)Hellerstein M.K. Schwarz J.M. Neese R.A. Regulation of hepatic de novo lipogenesis in humans.). Newly synthesized lipids generated by hepatic DNL are stored or excreted by hepatocytes and play key roles in hepatic energy homeostasis. The physiological roles and regulation of hepatic DNL in humans remain incompletely characterized, but elevated hepatic DNL may be important in a variety of metabolic disorders.The most prevalent of these disorders is nonalcoholic fatty liver disease (NAFLD) ((3)Lambert J.E. Ramos–Roman M.A. Browning J.D. Parks E.J. Increased De Novo Lipogenesis Is a Distinct Characteristic of Individuals With Nonalcoholic Fatty Liver Disease., (4)Lawitz E.J. Coste A. Poordad F. Alkhouri N. Loo N. McColgan B.J. Tarrant J.M. Nguyen T. Han L. Chung C. Ray A.S. McHutchison J.G. Subramanian G.M. Myers R.P. Middleton M.S. Sirlin C. Loomba R. Nyangau E. Fitch M. Li K. Hellerstein M. Acetyl-CoA Carboxylase Inhibitor GS-0976 for 12 Weeks Reduces Hepatic De Novo Lipogenesis and Steatosis in Patients With Nonalcoholic Steatohepatitis., (5)Kim C. Addy C. Kusunoki J. Anderson N.N. Deja S. Fu X. Burgess S.C. Li C. Ruddy M. Chakravarthy M. Previs S. Milstein S. Fitzgerald K. Kelley D.E. Horton J.D. Acetyl CoA Carboxylase Inhibition Reduces Hepatic Steatosis but Elevates Plasma Triglycerides in Mice and Humans: A Bedside to Bench Investigation.). The global prevalence of NAFLD and its progressive form, nonalcoholic steatohepatitis (NASH), are estimated to be approximately 24% and 1.5-6.5%, respectively, and are rising due to the epidemics of obesity and diabetes mellitus ((6)Younossi Z.M. Koenig A.B. Abdelatif D. Fazel Y. Henry L. Wymer M. Global epidemiology of nonalcoholic fatty liver disease—Meta‐analytic assessment of prevalence, incidence, and outcomes.,(7)Younossi Z. Anstee Q.M. Marietti M. Hardy T. Henry L. Eslam M. George J. Bugianesi E. Global burden of NAFLD and NASH: trends, predictions, risk factors and prevention.). NASH is also emerging as a growing cause of hepatocellular carcinoma (HCC) and will soon become the leading indication for liver transplantation ((8)Younossi Z. Stepanova M. Ong J.P. Jacobson I.M. Bugianesi E. Duseja A. Eguchi Y. Wong V.W. Negro F. Yilmaz Y. Romero-Gomez M. George J. Ahmed A. Wong R. Younossi I. Ziayee M. Afendy A. Nonalcoholic Steatohepatitis Is the Fastest Growing Cause of Hepatocellular Carcinoma in Liver Transplant Candidates.). Hepatic steatosis, or accumulation of triacylglycerols (TG), defines NAFLD and can in principle derive from three sources: fatty acids released through lipolysis of TG stored in adipose tissues, dietary TG carried in circulating lipoproteins, and hepatic DNL. Donnelly et al. reported that hepatic DNL in patients with NAFLD contributes up to 26% of liver TG compared to 2–5% in normal patients, and thus, may be an important contributor to liver TG accumulation in this condition ((9)Donnelly K.L. Smith C.I. Schwarzenberg S.J. Jessurun J. Boldt M.D. Parks E.J. Sources of fatty acids stored in liver and secreted via lipoproteins in patients with nonalcoholic fatty liver disease.).Progress in the development of pharmacologic inhibitors of key enzymes in the DNL pathway, such as ACC, positions hepatic DNL in humans as a target with practical importance for the treatment of NASH ((4)Lawitz E.J. Coste A. Poordad F. Alkhouri N. Loo N. McColgan B.J. Tarrant J.M. Nguyen T. Han L. Chung C. Ray A.S. McHutchison J.G. Subramanian G.M. Myers R.P. Middleton M.S. Sirlin C. Loomba R. Nyangau E. Fitch M. Li K. Hellerstein M. Acetyl-CoA Carboxylase Inhibitor GS-0976 for 12 Weeks Reduces Hepatic De Novo Lipogenesis and Steatosis in Patients With Nonalcoholic Steatohepatitis.). Accurate direct measurement of hepatic DNL in vivo in humans for clinical trials is, however, challenging for several reasons. First, this is an intracellular biosynthetic process in the liver but needs to be assessed noninvasively; second, any in vivo metabolic labeling study requires knowledge of the label content in the true biosynthetic precursor pool in the tissue of interest ((10)Hellerstein M.K. Neese R.A. Mass isotopomer distribution analysis at eight years: theoretical, analytic, and experimental considerations.); and, third, the turnover of the intrahepatic TG storage pool is slow ((11)Vedala A. Wang W. Neese R.A. Christiansen M.P. Hellerstein M.K. Delayed secretory pathway contributions to VLDL-triglycerides from plasma NEFA, diet, and de novo lipogenesis in humans.), which means that there may be label remaining in hepatic TG from a prior metabolic labeling study that could confound interpretation of a therapeutic intervention trial that involves a test/retest protocol.A method for accurately measuring DNL using stable isotope tracers and mass spectrometry based on mass isotopomer distribution analysis (MIDA) and mathematical modeling was developed by our laboratory in the 1990s ((12)Hellerstein M.K. Christiansen M. Kaempfer S. Kletke C. Wu K. Reid J.S. Mulligan K. Hellerstein N.S. Shackleton C.H.L. Measurement of de novo hepatic lipogenesis in humans using stable isotopes.) and has recently been modified for longer-term labeling studies with heavy water (2H20, deuterated water) ((4)Lawitz E.J. Coste A. Poordad F. Alkhouri N. Loo N. McColgan B.J. Tarrant J.M. Nguyen T. Han L. Chung C. Ray A.S. McHutchison J.G. Subramanian G.M. Myers R.P. Middleton M.S. Sirlin C. Loomba R. Nyangau E. Fitch M. Li K. Hellerstein M. Acetyl-CoA Carboxylase Inhibitor GS-0976 for 12 Weeks Reduces Hepatic De Novo Lipogenesis and Steatosis in Patients With Nonalcoholic Steatohepatitis.,(9)Donnelly K.L. Smith C.I. Schwarzenberg S.J. Jessurun J. Boldt M.D. Parks E.J. Sources of fatty acids stored in liver and secreted via lipoproteins in patients with nonalcoholic fatty liver disease.,(13)Esler W.P. Tesz G.J. Hellerstein M.K. Beysen C. Sivamani R. Turner S.M. Watkins S.M. Amor P.A. Carvajal-Gonzalez S. Geoly F.J. Biddle K.E. Purkal J.J. Fitch M. Buckeridge C. Silvia A.M. Griffith D.A. Gorgoglione M. Hassoun L. Bosanac S.S. Vera N.B. Rolph T.P. Pfefferkorn J.A. Sonnenberg G.E. Human sebum requires de novo lipogenesis, which is increased in acne vulgaris and suppressed by acetyl-CoA carboxylase inhibition.). By this approach, very low-density lipoprotein (VLDL)-TG secreted from the liver into the plasma is used as a window into the metabolic contribution from DNL to TG synthesized in the liver. The combinatorial isotopic labeling pattern in VLDL-TG-fatty acids, calculated by MIDA ((12)Hellerstein M.K. Christiansen M. Kaempfer S. Kletke C. Wu K. Reid J.S. Mulligan K. Hellerstein N.S. Shackleton C.H.L. Measurement of de novo hepatic lipogenesis in humans using stable isotopes.) after exposure to a stable isotope label such as 13C-acetate (incorporated via acetyl-CoA) or deuterated water (incorporated through the NADPH pool), allows accurate calculation of the fractional contribution from DNL to non-essential fatty acids. To account for potential delays in reaching plateau or near-plateau values for DNL due to the slow turnover of intrahepatic TG pools in NAFLD, heavy water is optimally administered as oral doses for days or weeks ((4)Lawitz E.J. Coste A. Poordad F. Alkhouri N. Loo N. McColgan B.J. Tarrant J.M. Nguyen T. Han L. Chung C. Ray A.S. McHutchison J.G. Subramanian G.M. Myers R.P. Middleton M.S. Sirlin C. Loomba R. Nyangau E. Fitch M. Li K. Hellerstein M. Acetyl-CoA Carboxylase Inhibitor GS-0976 for 12 Weeks Reduces Hepatic De Novo Lipogenesis and Steatosis in Patients With Nonalcoholic Steatohepatitis.,(11)Vedala A. Wang W. Neese R.A. Christiansen M.P. Hellerstein M.K. Delayed secretory pathway contributions to VLDL-triglycerides from plasma NEFA, diet, and de novo lipogenesis in humans.). Heavy water labeling is safe, relatively inexpensive, does not require an intravenous infusion, and can be readily administered for long periods on an outpatient basis, making it an attractive approach in this setting. Currently, there are no validated plasma lipid markers to replace direct measurement of hepatic DNL, although some studies have suggested that circulating fatty acid ratios might be useful as indirect markers of DNL ((14)Volk B.M. Kunces L.J. Freidenreich D.J. Kupchak B.R. Saenz C. Artistizabal J.C. Fernandez M.L. Bruno R.S. Maresh C.M. Kraemer W.J. Phinney S.D. Volek J.S. Effects of step-wise increases in dietary carbohydrate on circulating saturated fatty acids and palmitoleic acid in adults with metabolic syndrome., (15)Jacobs S. Jäger S. Jansen E. Peter A. Stefan N. Boeing H. Schulze M.B. Kröger J. Associations of Erythrocyte Fatty Acids in the De Novo Lipogenesis Pathway with Proxies of Liver Fat Accumulation in the EPIC-Potsdam Study., (16)Ma W. Wu J.H.Y. Wang Q. Lemaitre R.N. Mukamal K.J. Djoussé L. King I.B. Song X. Biggs M.L. Delaney J.A. Kizer J.R. Siscovick D.S. Mozaffarian D. Prospective association of fatty acids in the de novo lipogenesis pathway with risk of type 2 diabetes: the Cardiovascular Health Study., (17)Zong G. Zhu J. Sun L. Ye X. Lu L. Jin Q. Zheng H. Yu Z. Zhu Z. Li H. Sun Q. Lin X. Associations of erythrocyte fatty acids in the de novo lipogenesis pathway with risk of metabolic syndrome in a cohort study of middle-aged and older Chinese.).

In the current study, we measured hepatic DNL using heavy water labeling in a large cohort of patients with NASH and fibrosis as baseline studies prior to participating in a 12-week clinical trial of pharmacological therapies. This approach addressed several important metabolic and study design questions that had not previously been resolved: the turnover time of the intrahepatic TG storage pool, based on the kinetics of label incorporation and die-away in plasma TG-palmitate; the contribution of hepatic DNL to circulating TG-palmitate at plateau after long-term metabolic labeling; and the optimal timing and calculation approach for repeat labeling studies in the setting of therapeutic interventions. We report that the hepatic TG storage pool in NASH turns over slowly; that the plateau DNL contribution to TG-palmitate is considerably higher than previously reported in studies that used shorter metabolic labeling periods; that DNL remains elevated in patients with cirrhosis despite reduced liver fat; and that use of a correction model for hepatic TG turnover permits repeat labeling studies after 4 weeks of an intervention. In addition, we explore associations of hepatic DNL with several biomarkers in patients with NASH.

MATERIALS AND METHODSPatients and Study DesignWe enrolled 123 patients with NASH-related fibrosis from 9 sites in the U.S. and 1 site in New Zealand in an open-label, proof-of-concept study including 9 treatment groups that evaluated the safety and efficacy of the ACC inhibitor firsocostat, the farnesoid X receptor (FXR) agonist cilofexor, and/or the apoptosis signal-regulating kinase 1 (ASK1) inhibitor selonsertib, given alone or in combination for 12 weeks (NCT02781584; full eligibility criteria and study design are detailed in the Supplementary Material and Supplementary Figure 1). In the present analysis, we report baseline data from the 123 patients from all 9 cohorts as well as the post-treatment results for 10 patients with NASH-related cirrhosis and 10 patients with NASH-related fibrosis who were treated with firsocostat monotherapy. Some of the data (baseline and week 12) from the 10 non-cirrhotic patients who received firsocostat monotherapy have been previously reported ((4)Lawitz E.J. Coste A. Poordad F. Alkhouri N. Loo N. McColgan B.J. Tarrant J.M. Nguyen T. Han L. Chung C. Ray A.S. McHutchison J.G. Subramanian G.M. Myers R.P. Middleton M.S. Sirlin C. Loomba R. Nyangau E. Fitch M. Li K. Hellerstein M. Acetyl-CoA Carboxylase Inhibitor GS-0976 for 12 Weeks Reduces Hepatic De Novo Lipogenesis and Steatosis in Patients With Nonalcoholic Steatohepatitis.,(18)McCabe B.J. Bederman I.R. Croniger C. Millward C. Norment C. Previs S.F. Reproducibility of gas chromatography–mass spectrometry measurements of 2H labeling of water: Application for measuring body composition in mice.). For the non-cirrhotic cohorts (n=103), we enrolled patients 18 to 75 years of age with suspected NASH based on a clinical diagnosis of NAFLD plus a historical biopsy consistent with F2 to F3 fibrosis, according to the NASH Clinical Research Network (CRN) classification (or equivalent), or a magnetic resonance imaging-estimated proton density fat fraction (MRI-PDFF) of ≥10% and liver stiffness by magnetic resonance elastography (MRE) of >2.88 kPa. Only patients without cirrhosis were eligible for enrollment in these cohorts, confirmed either by a FibroSure/FibroTest (LabCorp, Burlington, NC) result of

For the cirrhosis cohorts (n=20), we enrolled patients with a clinical diagnosis of NAFLD plus a historical biopsy consistent with F4 fibrosis by NASH CRN criteria (or equivalent), a historical liver stiffness by vibration-controlled transient elastography (FibroScan, Echosens, Paris, France) ≥14 kPa, or a liver stiffness by MRE ≥4.67 kPa during screening. Patients with body mass index (BMI) <18 kg/m2, serum alanine aminotransferase (ALT) concentration >5 times the upper limit of normal, serum creatinine concentration ≥2 mg/dL, or documented weight loss exceeding 5% between the date of the liver biopsy and screening were excluded (see Supplementary Appendix for full eligibility criteria).

Heavy Water Labeling ProtocolPatients underwent three cycles of heavy water administration: first during the 2 weeks prior to baseline and then prior to weeks 4 and 12 of treatment. Heavy water was administered orally in 1-week loading cycles (see Figure 1A). The labeling protocol involved consumption of 50 mL of 70% deuterated water three times daily for one week, with blood samples collected 3, 7, and 14 days after the start of labeling. To confirm patient compliance, enrichment of deuterated water in blood samples was determined by gas chromatography–mass spectroscopy (GCMS) after reacting with acetone, as previously described ((18)McCabe B.J. Bederman I.R. Croniger C. Millward C. Norment C. Previs S.F. Reproducibility of gas chromatography–mass spectrometry measurements of 2H labeling of water: Application for measuring body composition in mice.). Body water 2H2O enrichment curves for the entire study are shown in Figure 1A.

Figure 1Isotopic enrichments in body water and plasma palmitate in NASH patient cohorts with and without cirrhosis. A. Oral administration of heavy water occurred over 3 labeling periods in the course of the study, resulting in 2H2O enrichment over natural abundance (EM1) in plasma. B. The isotopic enrichment of the palmitate precursor pool was determined by mass isotopomer analysis of the palmitate enrichment in the M1 and M2 isotopomers. C. 2H-enrichment of M1 isotopomer of palmitate in plasma samples collected over time was measured to calculate the palmitate precursor pool enrichment and hepatic DNL. D. 2H-enrichment of M2 isotopomer (%EM2) of palmitate in plasma samples collected over time was measured to calculate the palmitate precursor pool enrichment. (Datapoints are mean ± SD).

Determination of Hepatic DNLThe fractional contribution from hepatic DNL to palmitate (%DNL) in plasma was determined at the University of California, Berkeley in fasting blood samples collected during 14-day periods of heavy water exposure. To simplify sample processing, labeling of total plasma palmitate was used to represent hepatic DNL, based on a subset of fasting plasma samples collected in this study in which we observed that the isotopic enrichment of total plasma palmitate was quantitatively almost identical to that of palmitate isolated from VLDL-TG (details provided in Supplementary Material). The time course of %DNL represents cumulative label incorporation across the circadian cycle in free-living patients over 2 weeks. Plasma palmitate was esterified and analyzed for mass isotopomer abundances by GCMS, using MIDA to determine the effective body water deuterium exposure (precursor pool enrichment) for the calculation of fractional DNL, as previously described ((4)Lawitz E.J. Coste A. Poordad F. Alkhouri N. Loo N. McColgan B.J. Tarrant J.M. Nguyen T. Han L. Chung C. Ray A.S. McHutchison J.G. Subramanian G.M. Myers R.P. Middleton M.S. Sirlin C. Loomba R. Nyangau E. Fitch M. Li K. Hellerstein M. Acetyl-CoA Carboxylase Inhibitor GS-0976 for 12 Weeks Reduces Hepatic De Novo Lipogenesis and Steatosis in Patients With Nonalcoholic Steatohepatitis.,(10)Hellerstein M.K. Neese R.A. Mass isotopomer distribution analysis at eight years: theoretical, analytic, and experimental considerations.,(19)Turner S.M. Roy S. Sul H.S. Neese R.A. Murphy E.J. Samandi W. Roohk D.J. Hellerstein M.K. Dissociation between adipose tissue fluxes and lipogenic gene expression in ob/ob mice.). Briefly, equal parts chloroform and methanolic HCl were added to plasma samples and incubated for 1 hr at 55° C to trans-esterify fatty acids to fatty acid methyl esters (FAME). 3 ml of hexane and 2 ml of water were added, mixed well and centrifuged for 10 min at 2500 RPM. The top organic layer containing the FAME was transferred to another tube, to which was added another 3 ml hexane. The tube was centrifuged again, and the top organic layer was pooled with the organic phase of the previous extraction, dried under stream of nitrogen, and resolubilized in toluene for subsequent analysis by GCMS to quantify isotopic enrichment of both the M1 isotopomer and the M2 isotopomer due to the incorporation of deuterium from heavy water into palmitate. Specifically, a DB-17 (Restek RTX-50) or equivalent column was used, with electron impact ionization with single ion monitoring. Methyl-palmitate ions from palmitate-methyl esters were monitored at a mass-to-charge ratio of 270–272, representing the parent M0 through the M2 isotopomers. Excess M2 (EM2) and excess M1 (EM1) enrichments were determined by subtraction of natural abundance values in unlabeled standards (run in parallel) from the sample enrichment. The proportion of plasma palmitate that originated from the DNL pathway was then calculated from the EM1 and EM2 of palmitate using MIDA to determine both the biosynthetic precursor enrichment and the corresponding isotopic enrichment of newly-synthesized palmitate molecules ((10)Hellerstein M.K. Neese R.A. Mass isotopomer distribution analysis at eight years: theoretical, analytic, and experimental considerations.). The precursor pool enrichment (p) was determined from the ratio of EM2/EM1 in the experimental data. Knowledge of the calculated metabolic precursor pool enrichment and the known n (number of repeating subunits in the polymer = 21 for palmitate ((20)Diraison F. Moulin P. Beylot M. Contribution of hepatic de novo lipogenesis and reesterification of plasma non esterified fatty acids to plasma triglyceride synthesis during non-alcoholic fatty liver disease.)) allows calculation of the theoretical asymptote enrichment of the single-labeled mass isotopomer species (EM1*), representing the maximum possible enrichment when palmitate is newly synthesized at this deuterium precursor pool enrichment, as described previously ((10)Hellerstein M.K. Neese R.A. Mass isotopomer distribution analysis at eight years: theoretical, analytic, and experimental considerations.). For samples collected during the baseline labeling period, the fractional synthesis of palmitate was then calculated by comparing the experimentally measured enrichment to the calculated asymptote:

FractionalPalmitateDNL=EM1/EM1*

(1)

Samples collected after an additional labeling period (W4 and W12 timepoints) required an adjustment to Equation 1 to account for labeled palmitate remaining from previous labeling periods (see below).Non-linear Regression Analysis of the Time Course of DNL in NASHThe time courses of DNL during the during the baseline periods for each patient cohort, averaged across the 103 non-cirrhotic patients and 20 cirrhotic patients, were analyzed by nonlinear regression (GraphPad Prism 8, GraphPad Software, San Diego, CA), using a 2-phase exponential association model:
We used this 2-phase exponential model to fit to the baseline study DNL datapoints and to characterize the relatively slow turnover pool that determines the half-life (t1/2, Slow, in days) by which %DNL approaches an equilibrium or plateau value (%DNLplateau) as well as the relative size of the fast (%Fast) and slow (%Slow) turnover pools. To constrain the 2-pool model, a fast turnover pool with a fixed 3.3-hour half-life (t1/2, Fast = 0.14 days) was introduced to reflect the rapid turnover of TG in plasma ((21)Mittendorfer B. Yoshino M. Patterson B.W. Klein S. VLDL Triglyceride Kinetics in Lean, Overweight, and Obese Men and Women.).Measurement of the Label Die-Away Rate of the Hepatic TG Storage PoolTo quantitatively estimate in each subject the relatively slow hepatic TG turnover that was apparent in the label rise-to-plateau kinetics, we measured the die-away kinetics of labeled plasma palmitate after cessation of label exposure. Because the half-life of water in the human body is slow (Figure 1A), we made measurements in blood samples between day 56 and day 70 of the study (Week 8 and Week 10, respectively, in Figure 1), after body water deuterium enrichments had fallen to relatively low values. By this means, the die-away kinetics of the slow turnover pool of palmitate released from liver into blood can be assessed after a minimal correction for contamination by ongoing direct incorporation of new deuterium label from body water into palmitate in blood TG. In this model, the availability of heavy water tracer in body water at any given time contributes to plasma palmitate enrichment by a direct or immediate (fast turnover) pool and also much more slowly after passage through a slow turnover pool, as seen in the rise-to-plateau data here and as also previously shown by the delayed incorporation of 13C-acetate and labeled fatty acid tracers into blood TG in human subjects ((11)Vedala A. Wang W. Neese R.A. Christiansen M.P. Hellerstein M.K. Delayed secretory pathway contributions to VLDL-triglycerides from plasma NEFA, diet, and de novo lipogenesis in humans.). Since the heavy water tracer present at the Day 56 and 70 time-points was small but not zero, it was necessary to impute the portion of the residual isotopic label found in plasma palmitate that was due to rapid palmitate labeling from the prevailing heavy water tracer (direct or immediate pool), such that the remainder can be attributed to having come from a slow turnover tissue storage pool. The 2-phase exponential curve fit shown in Figure 2 provided a determination of the relative contributions of the observed fast and slow pools to the isotopic labeling of plasma palmitate. Specifically, at both the Week 8 and Week 12 timepoints, the amount of isotopic enrichment attributed to the slow storage pool was determined by the following equation:

%EM1slow,t=%EM1t-(EM1*BWx%DNLW12x%Fast)

(3)

where

%EM1t = the isotopic enrichment in plasma palmitate from plasma samples taken at time t (Weeks 8 and 12 of treatment), which represents the summed combination of the isotopic label from newly synthesized palmitate and the residual isotopic label from labeling period at Week 4.

%EM1slow, t = the isotopic enrichment in plasma palmitate derived from a slow turnover tissue storage pool, at a given timepoint t.

EM1*BW = the asymptote enrichment of the single-labeled mass isotopomer species, representing the maximum possible enrichment when palmitate is newly synthesized, as a function of measured heavy water enrichment in the body water (BW) at timepoint t ((10)Hellerstein M.K. Neese R.A. Mass isotopomer distribution analysis at eight years: theoretical, analytic, and experimental considerations.).

% DNLW12 = the fractional contribution of new hepatic DNL as determined from the corrected increase in isotopic enrichment (EM1) in plasma palmitate after the onset of repeat labeling with heavy water.

%Fast = the average relative size of the fast and slow pools, determined for the study cohort by 2-phase exponential line fitting of the DNL time course at baseline (Figure 2).

Figure 2Baseline de novo lipogenesis (DNL) in NASH patient cohorts with and without cirrhosis rises over 14 days of isotopic labeling. A 2-phase exponential curve fit (shown as the connecting line) was used to derive the rise-to-plateau kinetic parameters for the model equation shown (%DNLplateau = 55%, %Fast = 44%, %Slow = 56%, t1/2, slow = 13 days). Data are mean with error bars showing SD.

This calculation model assumes that fractional DNL contribution is stable from the week 8 and week 12 time-points after an intervention. Using these values of %EM1slow, t, the half-life of labeled palmitate derived from the slow pool present at day 56 through day 70 was calculated using Equation 4, based on simple exponential decay:

t1/2(days)=-14*ln(2)/ln(%EM1slow,day70/%EM