5963Automated quantification of epicardial adipose tissue from non-contrast CT on multi-center and multi-vendor data using deep learning

Background

Epicardial adipose tissue (EAT), a metabolically active visceral fat depot surrounding the coronary arteries, has been shown to promote the development of atherosclerosis in underlying coronary vasculature.

Purpose

We evaluate the performance of deep learning (DL), a sub-group of machine learning algorithms, for robust and fully automated quantification of EAT on multi-center cardiac CT data.

Methods

In this study, 850 non-contrast calcium scoring CT scans, from multiple cohorts, scanners and protocols, with manual measurements of EAT from 3 different readers were considered. The DL method was based on a convolutional neural network trained to reproduce the expert measurement. DL global performance was first assessed using all the scans, and then compared to inter-observer variability on a subset of 141 scans. Finally, automated EAT progression was compared to manual measurement using baseline and follow-up serial scans available for 70 subjects. The proposed model was validated using 10-fold cross validation.

Results

Automated quantification was performed in 1.57±0.49 seconds compared to 15 minutes for manual measurement. DL provided high agreement with expert manual quantification for all scans (R=0.974, p<0.001) with no significant bias (0.53 cm3, p=0.13). EAT volume was higher in patients with hypertension (+18.02 cm3, p<0.001, N=442), with diabetes (+18.33 cm3, p<0.001, N=75) and with hypercholesterolemia (+7.33 cm3, p=0.039, N=508). Manual EAT volumes measured by two experienced readers on 141 scans were highly correlated (R=0.984, p<0.001) but presented a significant difference of 4.35 cm3 (p<0.001). On these 141 scans, DL quantifications were highly correlated to both experts' measurements (R=0.973, p<0.001; R=0.979, p<0.001) with significant and non-significant bias for readers 1 and 2 (5.19 cm3, p<0.001; 0.84 cm3, p=0.26), respectively. In 70 subjects, EAT progression quantified by DL correlated strongly with EAT progression measured by the expert reader (R=0.905, p<0.001) with no significant bias (0.64 cm3, p=0.43), and was related to increased non-calcified plaque burden quantified from coronary CT angiography (5.7% vs 1.8%, p=0.026).

Automated vs. manual EAT volume

Conclusion

Deep learning allows rapid, robust and fully automated quantification of EAT from calcium scoring CT. It performs as an expert reader and can be implemented for routine cardiovascular risk assessment.

Acknowledgement/Funding

1R01HL133616/01EX1012B/Adelson Medical Research Foundation

P6151Fully automated epicardial adipose tissue volume and density measured from non-contrast CT predict major adverse cardiovascular events in asymptomatic subjects

Background

Epicardial adipose tissue (EAT) volume and density has shown to correlate with standard markers of coronary artery disease (CAD) and may predict major adverse cardiovascular events (MACE).

Purpose

We aimed to evaluate the prognostic value of EAT volume and density measured by fully automated deep-learning software from non-contrast cardiac computed tomography (CT).

Methods

We assessed 2071 consecutive asymptomatic subjects (age 56±9 years, 59% male) from the EISNER (Early Identification of Subclinical Atherosclerosis by Noninvasive Imaging Research) trial with long-term follow-up after coronary artery calcium (CAC) measurement. EAT volume and mean density were quantified using automated deep-learning software from non-contrast cardiac CT. MACE was defined as myocardial infarction (MI), cardiac death, late (>90 days) revascularization and acute coronary syndrome (ACS). EAT volume and density were systematically compared to CAC score and atherosclerotic cardiovascular disease (ASCVD) risk score using Cox proportional hazards regression for MACE prediction.

Results

At 14±3 years, 217 subjects suffered MACE. In age-and-gender-adjusted multivariate analysis, ASCVD risk score, CAC (two-fold increase) and EAT volume (two-fold increase) were associated with increased risk of suffering MACE [Hazard Ratio (HR) (95% CI): 1.03 (1.01–1.04); 1.25 (1.19–1.30); and 1.36 (1.08–1.70) respectively, p<0.01 for all] (Figure); the corresponding Harrell's C-statistic was 0.76. The area-under-the curve from receiver-operator characteristic analysis for MACE prediction increased significantly from 0.69 to 0.77 (p<0.0001) when EAT volume and CAC were added to the current clinical standard (ASCVD, family history and obesity measures BMI and BSA). Both in men and women, increase in EAT volume was associated with increased risk of MACE, with HR 1.14 (1.06–1.22), p<0.001 in men vs. 1.15 (1.01–1.31), p=0.03 in women, for each 20 cubic centimeter increase in volume. EAT density (HU) was independently inversely associated with MACE [HR: 0.96 (0.93–0.99), p=0.01].

MACE Prediction

Conclusions

EAT volume and density measurements improve prediction of MACE in asymptomatic populations over the current clinical standard. Fully automated EAT volume and density quantification by deep-learning from non-contrast cardiac CT can provide additional prognostic value for the asymptomatic patient.

Acknowledgement/Funding

1R01HL133616, Forschungsstiftung Medizin Universitätsklinikum Erlangen, grant from Dr Miriam and Sheldon G. Adelson Medical Research Foundation

Phase II safety and clinical comparison with single-photon emission computed tomography myocardial perfusion imaging for detection of coronary artery disease: flurpiridaz F 18 positron emission tomography

OBJECTIVES

This was a phase II trial to assess flurpiridaz F 18 for safety and compare its diagnostic performance for positron emission tomography (PET) myocardial perfusion imaging (MPI) with Tc-99m single-photon emission computed tomography (SPECT) MPI with regard to image quality, interpretative certainty, defect magnitude, and detection of coronary artery disease (CAD) (≥50% stenosis) on invasive coronary angiography (ICA).

BACKGROUND

In pre-clinical and phase I studies, flurpiridaz F 18 has shown characteristics of an essentially ideal MPI tracer.

METHODS

One hundred forty-three patients from 21 centers underwent rest-stress PET and Tc-99m SPECT MPI. Eighty-six patients underwent ICA, and 39 had low-likelihood of CAD. Images were scored by 3 independent, blinded readers.

RESULTS

A higher percentage of images were rated as excellent/good on PET versus SPECT on stress (99.2% vs. 88.5%, p < 0.01) and rest (96.9% vs. 66.4, p < 0.01) images. Diagnostic certainty of interpretation (percentage of cases with definitely abnormal/normal interpretation) was higher for PET versus SPECT (90.8% vs. 70.9%, p < 0.01). In 86 patients who underwent ICA, sensitivity of PET was higher than SPECT (78.8% vs. 61.5%, respectively, p = 0.02). Specificity was not significantly different (PET: 76.5% vs. SPECT: 73.5%). Receiver-operating characteristic curve area was 0.82 ± 0.05 for PET and 0.70 ± 0.06 for SPECT (p = 0.04). Normalcy rate was 89.7% with PET and 97.4% with SPECT (p = NS). In patients with CAD on ICA, the magnitude of reversible defects was greater with PET than SPECT (p = 0.008). Extensive safety assessment revealed that flurpiridaz F 18 was safe in this cohort.

CONCLUSIONS

In this phase 2 trial, PET MPI with flurpiridaz F 18 was safe and superior to SPECT MPI for image quality, interpretative certainty, and overall CAD diagnosis.

Misfolding of mutant aquaporin-2 water channels in nephrogenic diabetes insipidus

We reported that several aquaporin-2 (AQP2) point mutants that cause nephrogenic diabetes insipidus (NDI) are retained in the endoplasmic reticulum (ER) of transfected mammalian cells and degraded but can be rescued by chemical chaperones to function as plasma membrane water channels (Tamarappoo, B. K., and Verkman, A. S. (1998) J. Clin. Invest. 101, 2257-2267). To test whether mutant AQP2 proteins are misfolded, AQP2 folding was assessed by comparative detergent extractability and limited proteolysis, and AQP2 degradation kinetics was measured by label-pulse-chase and immunoprecipitation. In ER membranes from transfected CHO cells containing [(35)S]methionine-labeled AQP2, mutants T126M and A147T were remarkably detergent-resistant; for example wild-type AQP2 was >95% solubilized by 0.5% CHAPS whereas T126M was <10% solubilized. E258K, an NDI-causing AQP2 mutant which is retained in the Golgi, is highly detergent soluble like wild-type AQP2. The mutants and wild-type AQP2 were equally susceptible to digestion by trypsin, thermolysin, and proteinase K. Stopped-flow light scattering measurements indicated that T126M AQP2 at the ER was fully functional as a water channel. Pulse-chase studies indicated that the increased degradation rates for T126M (t((1)/(2)) 2.5 h) and A147T (2 h) compared with wild-type AQP2 (4 h) involve a brefeldin A-resistant, ER-dependent degradation mechanism. After growth of cells for 48 h in the chemical chaperone glycerol, AQP2 mutants T126M and A147T became properly targeted and relatively detergent-soluble. These results provide evidence that NDI-causing mutant AQP2 proteins are misfolded, but functional, and that chemical chaperones both correct the trafficking and folding defects. Strategies to facilitate protein folding might thus have therapeutic efficacy in NDI.

Vasopressin regulated trafficking of a green fluorescent protein-aquaporin 2 chimera in LLC-PK1 cells

Aquaporin 2 (AQP2) transfected into LLC-PK1 cells functions as a vasopressin-regulated water channel that recycles between intracellular vesicles and the plasma membrane upon vasopressin stimulation. The green fluorescent protein (GFP) of the jellyfish, Aequorea victoria, was used as an autofluorescent tag to monitor AQP2 trafficking in transfected LLC-PK1 cells. Two chimeras were constructed, one in which GFP was fused to the amino-terminus of AQP2 [GFP-AQP2(NT)] and the second in which it was fused to the carboxyl-terminus [AQP2-GFP(CT)]. The GFP-AQP2(NT) chimera trafficked in a regulated pathway from intracellular vesicles to the basolateral plasma membrane in response to vasopressin or forskolin stimulation of cells. In contrast, the AQP2-GFP(CT) chimera expressed in LLC-PK1 cells was localized constitutively on both apical and basolateral plasma membranes. The cellular location of this chimera was not modified by vasopressin or forskolin. Thus, while the GFP-AQP2(NT) chimera will be useful to study AQP2 trafficking in vitro, the abnormal, constitutive membrane localization of the AQP2-GFP(CT) chimera suggests that one or more trafficking signals exist on the carboxyl-terminus of the AQP2 protein.

Defective aquaporin-2 trafficking in nephrogenic diabetes insipidus and correction by chemical chaperones

Five single-point aquaporin-2 (AQP2) mutations that cause non-X-linked nephrogenic diabetes insipidus (NDI) were characterized to establish the cellular defect and to develop therapeutic strategies. In Xenopus oocytes expressing AQP2 cRNAs, single-channel water permeabilities of mutants L22V, T126M, and A147T were similar to that of wild-type AQP2, whereas R187C and C181W were nonfunctional. In [35S]methionine pulse-chase experiments in transiently transfected CHO cells, half-times for AQP2 degradation were approximately 4 h for wild-type AQP2 and L22V, and mildly decreased for T126M (2.7 h), C181W (2.4 h), R187C (2.0 h), and A147T (1.8 h). Immunofluorescence showed three distinct AQP2-staining patterns: plasma membrane and endosomal staining (wild-type, L22V), endoplasmic reticulum (ER) staining (T126M > A147T approximately R187C), or a mixed pattern of reticular and perinuclear vesicular staining. Immunoblot of fractionated vesicles confirmed primary ER localization of T126M, R187C, and A147T. To determine if the AQP2-trafficking defect is correctable, cells were incubated with the "chemical chaperone" glycerol for 48 h. Immunoblot showed that glycerol produced a nearly complete redistribution of AQP2 (T126M, A147T, and R187C) from ER to membrane/endosome fractions. Immunofluorescence confirmed the cellular redistribution. Redistribution of AQP2 mutants was also demonstrated in transfected MDCK cells, and using the chaperones TMAO and DMSO in place of glycerol in CHO cells. Water permeability measurements indicated that functional correction was achieved. These results indicate defective mammalian cell processing of mutant AQP2 water channels in NDI, and provide evidence for pharmacological correction of the processing defect by chemical chaperones.

Reconstitution of a regulated transepithelial water pathway in cells transfected with AQP2 and an AQP1/AQP2 hybrid containing the AQP2-C terminus

Transepithelial water permeability was measured in LLC-PK1 cells stably transfected with aquaporins (AQPs): AQP1, AQP2, and a chimera of AQP1 and AQP2 containing 41 amino acids of the C-terminus of AQP2. Transepithelial water fluxes (Jw) were not previously reported in cells transfected with aquaporins. Jw were now recorded each minute using a specially developed experimental device. A significant increase in Posm after forskolin (FK) plus vasopressin (VP) was found in AQP2 transfected cells (39.9 +/- 8.2 vs. 12.5 +/- 3.3 cm.sec-1.10(-3)), but not in cells transfected with AQP1 (15.3 +/- 3.6 vs. 13.4 +/- 3.6 cm.sec-1.10(-3)). In the case of the AQP1/2 cells (chimera) the FK plus VP induced Posm was smaller than in AQP2 cells but significantly higher than in mock cells at rest (18.1 +/- 4.8 vs. 6.7 +/- 1.0 cm.sec-1.10(-3)). The increases in Posm values were not paralleled by increases in 14C-Mannitol permeability. HgCl2 inhibited the hydrosmotic response to FK plus VP in AQP2 transfected epithelia. Results were comparable to those observed, in parallel experiments, in a native ADH-sensitive water channel containing epithelial barrier (the toad urinary bladder). Electron microscopy showed confluent LLC-PK1 cells with microvilli at the mucosal border. The presence of spherical or elongated intracellular vacuoles was observed in AQP2 transfected cells, specially after FK plus VP stimulus and under an osmotic gradient. These results demonstrate regulated transepithelial water permeability in epithelial cells transfected with AQP2.

Identification and characterization of aquaporin-2 water channel mutations causing nephrogenic diabetes insipidus with partial vasopressin response

Congenital nephrogenic diabetes insipidus (NDI) is a rare disease caused most often by mutations in the vasopressin V2 receptor (AVPR2). We studied a family which included a female patient with NDI with symptoms dating from infancy. The patient responded to large doses of desmopressin (dDAVP) which decreased urine volume from 10 to 4 I/day. Neither the parents nor the three sisters were polyuric. The patient was found to be a compound heterozygote for two novel recessive point mutations in the aquaporin-2 (AQP2) gene: L22V in exon 1 and C181W in exon 3. Residue Cys181 in AQP2 is the site for inhibition of water permeation by mercurial compounds and is located near to the NPA motif conserved in all aquaporins. Osmotic water permeability (Pf) in Xenopus oocytes injected with cRNA encoding C181W-AQP2 was not increased over water control, while expression of L22V cRNA increased the Pf to approximately 60% of that for wild-type AQP2. Co-injection of the mutant cRNAs with the wild-type cRNA did not affect the function of the wild-type AQP2. Immunolocalization of AQP2-transfected CHO cells showed that the C181W mutant had an endoplasmic reticulum-like intracellular distribution, whereas L22V and wild-type AQP2 showed endosome and plasma membrane staining. Water permeability assays showed a high Pf in cells expressing wild-type and L22V AQP2. This study indicates that AQP2 mutations can confer partially responsive NDI.

Identification of a system N-like Na(+)-dependent glutamine transport activity in rat brain neurons

Glutamine is a primary precursor for the biosynthesis of the neurotransmitters glutamate and gamma-aminobutyric acid. It is proposed that glutamine, synthesized and released by astrocytes, is transported into the neuron for subsequent conversion to neurotransmitters. To provide a more complete characterization of this process, we have delineated the transport systems for glutamine uptake in primary cultures of brain neuronal cells from 1-day-old rats. The Na(+)-dependent glutamine entry is mediated by system A, system ASC, and a third, previously unidentified, activity that has been tentatively designated as system Nb. System Nb activity can be monitored by assaying Na(+)-dependent [3H]glutamine uptake in the presence of 2 mM concentrations of both 2-(methylamino) isobutyric acid and threonine to block uptake by systems A and ASC, respectively. The newly identified transport activity exhibits an apparent substrate specificity that is unique compared with the hepatic system N, because it is inhibited by glutamine and asparagine, but not by histidine. Also, the affinity of system Nb for glutamine, as estimated from K(m) values, is significantly greater than that observed for the hepatic and muscle Na(+)-dependent glutamine transporters, systems N and Nm. In sharp contrast to the hepatic system N transporter, system Nb exhibits a relative insensitivity to pH and does not permit Li+ substitution for Na+ as the cosubstrate. The substrate specificity, kinetic analysis, pH sensitivity, and cation dependence of this transport activity indicate that it represents a glutamine transport system not previously identified.

Expressed human hippocampal ASCT1 amino acid transporter exhibits a pH-dependent change in substrate specificity

In mammalian cells, the basal Na+-dependent uptake for many of the neutral amino acids is mediated by a transport activity designated System ASC. A cloned human brain cDNA sequence, ASCT1, encodes a Na+-dependent neutral amino acid transport activity that exhibits a substrate specificity similar to that commonly associated with System ASC. However, the characteristics of ASC activity varies significantly between cell types and not all tissues contain detectable levels of ASCT1 mRNA. A unique property of System ASC activity is an altered substrate selectivity such that at pH values below 7.4 anionic amino acids function as inhibitors and substrates. The experiments in this report were designed to determine if the cloned ASCT1 transporter exhibited this pH-dependent anionic transport. Following transfection of HeLa cells with the ASCT1 cDNA, transport strongly favored neutral zwitterionic) amino acids when uptake was measured at a physiologic pH value of 7.5. However, lowering the assay pH to 5.5 significantly enhanced the interaction of the ASCT1 carrier with anionic amino acids such as cysteate, in a pH-dependent manner. The apparent pK for the titratable group was in the range of 6.5-7.0. These results provide evidence that the human brain ASCT1 transporter exhibits the most distinguishing characteristic known for System ASC and provides a model system to investigate the molecular basis for this shift in substrate acceptance.

Water transport across mammalian cell membranes

This review summarizes recent progress in water-transporting mechanisms across cell membranes. Modern biophysical concepts of water transport and new measurement strategies are evaluated. A family of water-transporting proteins (water channels, aquaporins) has been identified, consisting of small hydrophobic proteins expressed widely in epithelial and nonepithelial tissues. The functional properties, genetics, and cellular distributions of these proteins are summarized. The majority of molecular-level information about water-transporting mechanisms comes from studies on CHIP28, a 28-kDa glycoprotein that forms tetramers in membranes; each monomer contains six putative helical domains surrounding a central aqueous pathway and functions independently as a water-selective channel. Only mutations in the vasopressin-sensitive water channel have been shown to cause human disease (non-X-linked congenital nephrogenic diabetes insipidus); the physiological significance of other water channels remains unproven. One mercurial-insensitive water channel has been identified, which has the unique feature of multiple overlapping transcriptional units. Systems for expression of water channel proteins are described, including Xenopus oocytes, mammalian and insect cells, and bacteria. Further work should be directed at elucidation of the role of water channels in normal physiology and disease, molecular analysis of regulatory mechanisms, and water channel structure determination at atomic resolution.

Protein modification of glutamine transporters in SV40-transformed hepatocytes and immunodetection of proteins associated with hepatic system N transport activity

The secondary active transport of glutamine into hepatocytes is mediated by the sodium-dependent System N amino acid transporter. In SV40-transformed fetal or adult hepatocytes, System N activity was increased by culturing the cells at the growth-permissive temperature of 33 degrees C compared with 40 degrees C. The sulfhydryl specific protein modification reagent p-chloromercuribenzene sulfonate (PCMBS) inactivated all of System N activity in fetal cells, whereas in the adult cells PCMBS inactivated only the elevated System N activity expressed at 33 degrees C; basal System N activity remained unaffected. Also, the inactivation of transport in the adult cells could be blocked by the presence of the substrate glutamine, whereas that in the fetal cells was completely resistant to protection. To characterize the System N transport, murine monoclonal antibodies System N transporter, murine monoclonal antibodies were generated. Two hybridomas (3E1-2 and 1E7-3) produced antibodies that inhibit System N activity in hepatocytes and can immunoprecipitate the activity from a solubilized protein fraction. Based on two-dimensional PAGE, the molecular size of the carrier protein has been shown to be approximately 100 kDa.

Cloning and expression of a novel Na(+)-dependent neutral amino acid transporter structurally related to mammalian Na+/glutamate cotransporters

A cDNA has been isolated from human hippocampus that appears to encode a novel Na(+)-dependent, Cl(-)-independent, neutral amino acid transporter. The putative protein, designated SATT, is 529 amino acids long and exhibits significant amino acid sequence identity (39-44%) with mammalian L-glutamate transporters. Expression of SATT cDNA in HeLa cells induced stereospecific uptake of L-serine, L-alanine, and L-threonine that was not inhibited by excess (3 mM) 2-(methylamino)-isobutyric acid, a specific substrate for the System A amino acid transporter. SATT expression in HeLa cells did not induce the transport of radiolabeled L-cysteine, L-glutamate, or related dicarboxylates. Northern blot hybridization revealed high levels of SATT mRNA in human skeletal muscle, pancreas, and brain, intermediate levels in heart, and low levels in liver, placenta, lung, and kidney. SATT transport characteristics are similar to the Na(+)-dependent neutral amino acid transport activity designated System ASC, but important differences are noted. These include: 1) SATT's apparent low expression in ASC-containing tissues such as liver or placenta; 2) the lack of mutual inhibition between serine and cysteine; and 3) the lack of trans-stimulation. SATT may represent one of multiple activities that exhibit System ASC-like transport characteristics in diverse tissues and cell lines.

Glucocorticoid regulation of splanchnic glutamine, alanine, glutamate, ammonia, and glutathione fluxes

Interorgan glutamine and associated metabolite fluxes were measured across the gut and liver to delineate splanchnic bed fluxes secondary to enhanced arterial loads mobilized in the periphery by glucocorticoid. Experiments were performed on adrenalectomized rats since adrenalectomy doubled the hepatic glucocorticoid receptor population compared with intact animals. Under these conditions, triamcinolone supplement (40 micrograms.day-1.100 g body wt-1) enhanced the combined net glutamine uptake by gut and liver eightfold, whereas combined gut and liver unidirectional breakdown and synthesis fluxes both increased (3.4- and 7.4-fold, respectively). Triamcinolone supplement also altered the pattern of metabolite released; gut released predominantly ammonium and some alanine, whereas the liver removed more alanine along with glutamine and released more urea, glutamate, and glutathione. Mechanistically, enhanced cellular glutamine uptake could be attributed to a three- to fourfold acceleration of glutamine transport associated with a rise in intracellular glutamine content. However, uptake by isolated membrane vesicles revealed only a small (27%) increase in System N activity, whereas extraction and reconstitution of the transporter into proteoliposomes failed to demonstrate increased transporter activity. Similarly, activity of phosphate-dependent glutaminase and glutamate dehydrogenase increased in crude homogenates (2-fold), but the former disappears in completely disrupted preparations. Furthermore, whereas messenger RNA and assayable enzymic activity for glutamate dehydrogenase clearly increased with glucocorticoid, glutaminase message was less significantly increased. Thus glucocorticoid appears directly capable of accelerating hepatic glutamine extraction primarily by modulating transporter activity that is closely coupled to glutamine utilization.

Identification of the protein responsible for hepatic system N amino acid transport activity

In the liver, glutamine utilization may be limited by the rate of transport across the plasma membrane by the System N carrier. System N-mediated transport activity has been solubilized from rat liver plasma membrane, partially purified, and then reconstituted into proteoliposomes. To identify the System N carrier protein, monoclonal antibodies were generated against the protein fraction enriched for System N activity. Two antibodies , 3E1-2 and 1E7-3, inhibited System N activity in hepatocytes. These antibodies also immunoprecipitated System N activity from a mixture of solubilized proteins and were specific for antigen recognition in that neither immunoprecipitated System A activity. The antibody recognized a single protein of molecular size 100 kDa by immunoblot analysis. Recognition of this protein by the antibody increased in parallel with the enrichment of System N activity in solubilized membrane fractions. These data suggest that a 100-kDa plasma membrane protein mediates System N transport activity in rat hepatocytes.

Characteristics and regulation of hepatic glutamine transport

Glutamine is an important amino acid because of its key role in the transfer of both carbon and nitrogen between tissues in the body. Specific tissues are usually associated with either net synthesis or net utilization of glutamine, but the liver plays a central role in glutamine homeostasis, in that it can shift to function in either capacity. This capability, along with the localization of urea biosynthesis in the periportal hepatocytes, focuses attention on the transport mechanisms in hepatocytes for uptake and release of glutamine. Active transport of glutamine by hepatocytes is mediated by a Na(+)-dependent activity termed system N, which exhibits a rather narrow substrate specificity mediating uptake of histidine and asparagine as well as of glutamine. This secondary active transport system allows for the net accumulation of glutamine against a concentration gradient and maintenance of intracellular concentrations of glutamine between 4 and 8 mM in the face of a plasma concentration of 0.6 mM. Utilization of the Na+ electrochemical gradient as a driving force ensures that the system N carrier catalyzes a unidirectional transport event favoring the cytoplasm. It is obvious from the glutamine gradient across the plasma membrane that efflux of this amino acid is typically slower than accumulation; measurement of saturable, Na(+)-independent glutamine transport by system L substantiates this proposal. However, it is clear that under certain metabolic conditions the liver represents a source of glutamine for other tissues in the body and net efflux must occur. The system N transport activity in hepatocytes is regulated by hormones such as insulin, glucagon, and glucocorticoids, as demonstrated both in vivo and in vitro.(ABSTRACT TRUNCATED AT 250 WORDS)