Cholesteryl ester depletion from the ovaries of superovulated female rats fed a normal or essential fatty acid deficient diet

The cholesteryl ester content of the ovaries was determined in rats diets containing corn oil or hydrogenated coconut oil (essential fatty acid (EFA) deficient) and subjected to superovulation by injection of luteinizing hormone and follicle-stimulating hormone. Superovulation increased ovarian weight; the effect was greater in animals fed corn oil. Superovulation significantly decreased total ovarian cholesteryl ester concentration in animals fed corn oil, with disproportionately large decreases occurring in the esters of 20:1, 20:2, 22:5w6, and 22:6w3. Significant decreases were observed in these esters when the data were expressed on a unit mass of tissue basis or in relation to total ovarian mass. In superovulated, EFA-deficient rats, esters of 18:1, 20:1, 22:5w6, and 22:6w3 were significantly lower per unit mass of tissue but this was due, in all cases except that of 22:6w3, to the increased mass of ovarian tissue; there was no decrease in total esters per ovary weight during superovulation of deficient rats. The pattern and degree of selective changes in ovarian cholesteryl esters during superovulation were different from those previously reported for adrenal esters of stressed rats.     

Essential fatty acid deficiency and adrenal cortical function in vitro.

Adrenocortical cells were prepared from rats maintained on essential fatty acid-deficient diets and control litter mates. Cells from control rats had high concentrations of essential fatty acids in the cholesteryl ester fraction of which approximately 22% was arachidonate. In contrast, cells from EFA-deficient rats had only 2.5% arachidonate in the cholesteryl esters, even though the total esterified cholesterol level was comparable to that of controls. In place of the essential fatty acids, the cholesteryl esters of these cells were rich in 20:3(n--9) and 22:3(n--9). When cells from EFA-deficient rats were incubated with ACTH or dibutyryl cyclic AMP, the output of corticosterone was the same as in controls. Also sterol esters were hydrolyzed to the same extent as in controls despite the unusual composition of the fatty acid esters. The phospholipids in both control and EFA-deficient cells contained high levels of arachidonate but were not hydrolyzed in either type of cell during incubation with ACTH or dibutyryl cyclic AMP. The results indicate that high levels of the prostaglandin precursors, namely linoleate and arachidonate, are not a sine qua non for the steroidogenic action of ACTH or cyclic AMP.     

Effects of dietary n-3 fatty acids on the composition of cholesteryl esters and triglycerides in plasma and liver perfusate of the rat

The effects of polyunsaturated fatty acids of the omega-3 family (PUFA n-3), (addition of fish oil), on the molecular composition of cholesteryl esters and triglycerides in plasma and liver perfusate of rats were studied. Rats fed a diet rich in saturated fatty acids (addition of lard) served as controls. Supplemention with PUFA n-3 not only decreases the plasma concentrations of free cholesterol, cholesteryl esters, and triglycerides, it also significantly alters the plasma composition of cholesteryl esters and triglycerides. Analyses of liver perfusate indicate a decrease in triglycerides secretion by in vitro perfused liver and reciprocal changes in relative contents of cholesteryl esters fractions with C(16) and C(20) acyl chains. This finding may be a result of chain-shortening of long-chain fatty acids probably in peroxisomal beta-oxidative system. Alterations in plasma cholesteryl esters and triglycerides composition of the fish oil group could be affected further by additional factors such as increased plasma cholesterol esterification activity and presence of triglyceride species of intestinal origin.     

Short-term effect of dietary cholesterol on tissue n-6 fatty acids in fat-deficient rats

Weanling female rats raised on a fat-free diet for 8 weeks were then given the same diet supplemented with 0, 0.25, 0.5, or 1% by weight of cholesterol in addition to 10% of safflower oil for 3 days. Fatty acid compositions of cholesteryl esters (CE), triglycerides (TG), and phospholipids (PL) in liver and plasma were examined. Cholesterol feeding increased plasma and liver cholesterol contents and also affected the patterns of n-6 polyunsaturated fatty acids. There were no consistent changes in either plasma and liver TG which contained little 20:3n-6 and 20:4n-6. The levels of 20:3n-6 increased in plasma and liver PL, while proportions of 20:4n-6 decreased in liver and plasma CE. However, the absolute amount of 20:4n-6 in cholesteryl esters increased because of a threefold rise in cholesteryl ester levels. The changes might be attributable to an increased utilization of 20:4n-6 for cholesterol transport and/or an inhibition of delta 5-desaturation of n-6 fatty acids by cholesterol feeding.     

Pregnenolone synthesis from cholesterol and hydroxycholesterols by mitochondria from ovaries following the stimulation of immature rats with pregnant mare's serum gonadotropin and human choriogonadotropin

The rate of pregnenolone synthesis by cytochrome P-450scc was measured in mitochondria isolated from ovaries of immature rats treated with pregnant mare's serum gonadotropin and human choriogonadotropin. Using cholesterol, 25-hydroxycholesterol, 20 alpha-hydroxycholesterol, (22R)-22-hydroxycholesterol and (22R)-20 alpha,22-dihydroxycholesterol as substrates, we have determined that the first hydroxylation of cholesterol, in the 22R position, is rate limiting in pregnenolone synthesis. It proceeds at only 22% of the rate of either of the subsequent two hydroxylations. 25-Hydroxycholesterol proved to be a suitable substrate for determining the maximum rate of pregnenolone synthesis by cytochrome P-450scc in isolated mitochondria. The maximum rate was 13 mol steroid.min-1.mol cytochrome P-450scc-1 and did not change after the follicles in the immature ovary had been stimulated to mature and luteinize with gonadotropin. Using endogenous cholesterol in isolated mitochondria as substrate, the time course of pregnenolone synthesis was the same during the follicular phase as in the luteal stage of gonadotropin-induced development. We conclude that during the artificial induced development of follicles in the immature ovary, the major cause of the increase in the rate of pregnenolone synthesis is the increase in the cytochrome P-450scc content of the mitochondria, rather than changes in the catalytic activity of cytochrome P-450scc or the cholesterol availability to the cytochrome.     

Effect of large doses of the oral contraceptive Enovid on cholesterol metabolism in the rat

Short-term effects of the oral contraceptive drug, Enovid, a mixture of estrogenic and progesterone compounds, have been determined in experiments on male and female rats. Oral administration of large doses for 7 days resulted in marked decreases of cholesteryl esters in plasma accompanied by only slight elevations of hepatic cholesterol content. Cholesteryl esters were also much lower in adrenals and ovaries, organs which are usually responsible for steroid hormone biosynthesis. At the same time, cholesterol-esterifying activity in plasma was substantially increased. Enovid administration was shown also to affect the fatty acid composition of sterol esters remaining in plasma, adrenals, and ovaries. The concentration of linoleate and arachidonate was significantly decreased in plasma sterol esters, whereas the concentrations of arachidonate and docosatetraenoate in adrenals and of docosatetraenoate in ovaries were significantly lowered. All of the changes reported were more pronounced in the female than in the male rat. It is hypothesized that the decreased levels of cholesteryl ester found in the organs investigated, together with the increase in plasma cholesterol-esterifying activity, are probably associated either with changes in cholesterol biosynthesis and(or) transport or with an increase in cholesterol transformation to other steroids and excretion. These possibilities are now under investigation.     

Effect of the stress of unilateral adrenalectomy on the depletion of individual cholesteryl esters in the rat adrenal

Normal rats were subjected to unilateral adrenalectomy and were killed 3 hr later. The concentration and composition of the cholesteryl esters in adrenals removed at operation and after death were compared. The esterified cholesterol concentration was lower in the adrenals obtained 3 hr after surgery. Cholesteryl arachidonate decreased in concentration significantly more than any other ester, followed in order of magnitude by linoleate and oleate. The cholesteryl ester concentration of adrenals removed from sham-operated rats 3 hr after surgery was also greatly reduced. On the basis of comparison with other work on the hydrolysis of cholesteryl esters by adrenal homogenates, it is concluded that the apparent selectivity in depletion of cholesteryl esters is due to differences in their rates of hydrolysis.     

Metabolism of doubly-labeled chylomicron cholesteryl esters in the rat

Chylomicrons labeled in vitro with doubly-labeled cholesteryl esters were injected intravenously into fasted rats, and the tissue distribution and chemical form of each isotope were observed for 24 hr. The use of doubly-labeled cholesteryl esters provided information about the metabolism of both the sterol and the fatty acid moieties. Similar results were obtained with doubly-labeled cholesteryl palmitate, oleate, and linoleate. In each instance, most (80-90%) of the chylomicron cholesteryl ester was removed from the plasma by the liver; small amounts were also taken up by all other tissues examined. There was no hydrolysis during uptake. In the liver the newly absorbed cholesteryl esters underwent slow hydrolysis (60% after 1 hr and 85-90% after 3.5 hr); the rate of reesterification of the liberated cholesterol was still slower. After 24 hr only 20-28% of the labeled cholesterol present in the animal was found in the liver. Labeled fatty acid disappeared from the liver, and was redistributed among other tissues, much more rapidly than the labeled cholesterol. Most of the labeled fatty acid apparently underwent oxidation, since only 15-20% of the injected labeled fatty acid was present in the animal after 24 hr. At this time the three fatty acids were differently distributed between and within the tissues. These differences reflected some known differences of fatty acid concentration and lipid composition in the various tissues.     

Accumulation of oleate-rich cholesteryl esters by acetyl-LDL in macrophages in the presence of an acyl-CoA: cholesterol acyltransferase inhibitor (Sandoz 58-035)

The mechanism through which cholesteryl esters rich in oleic acid accumulate in the cytoplasm was studied. The fatty acid composition of the cholesteryl esters in acetyl-LDL was high in linoleic acid, while that of cholesteryl ester inclusion bodies accumulated in the cytoplasm was high in oleic acid. This compositional change of fatty acids in cholesteryl esters occurred even in the presence of an acyl-CoA: cholesterol acyltransferase (ACAT) inhibitor, Sandoz 58-035. These results suggest that oleate-rich cholesteryl esters accumulated in the cytoplasm, even though the reesterification in microsome was inhibited by an ACAT inhibitor.     

Gender-related effects of chronic ethanol ingestion on rat hepatic acyl-CoA:cholesterol acyltransferase

The influence of chronic ethanol ingestion on hepatic acyl-CoA: cholesterol acyltransferase activity was investigated to determine the relationship between alcohol intake and cholesterol ester accumulation. Rats were given nutritionally complete liquid diets supplemented with 6.3% ethanol or an isocaloric equivalent of dextrin-maltose for 5 weeks. During this period, the hepatic acyl-CoA: cholesterol acyltransferase activity of ethanol-fed male rats remained constant, whereas the same activity in pair-fed controls as well as chow-fed rats exhibited a 30% decrease in activity. Unlike alcohol-fed male rats, the hepatic acyl-CoA: cholesterol acyltransferase activity of female rats decreased by approximately 30% by the fifth week of ethanol ingestion. Despite the fact that the gender of the animals led to disparate levels of acyl-CoA: cholesterol acyltransferase activity in response to ethanol ingestion, similar levels of cholesteryl ester accumulation were observed. The altered levels of acyl-CoA: cholesterol acyltransferase activity caused no significant change in the cholesterol concentration, cholesterol/phospholipid ratio, phospholipid fatty acid composition, or the membrane fluidity of the hepatic microsomes. We conclude that the altered hepatic acyl-CoA: cholesterol acyltransferase activity of ethanol-fed female rats cannot be directly responsible for ethanol-induced accumulation of cholesteryl esters.     

Cholesterol alterations in young dystrophic mice

Cholesterol and cholesteryl ester concentrations and cholesteryl ester fatty acid substituents have been measured during the first 10 weeks of life in tissues of normal and dystrophic mice. In normal Swiss and 129ReJ(+/?) mice the concentrations of both cholesterol and cholesteryl esters remain essentially constant in liver, increase in brain and fall sharply in both thigh (mixed fiber type muscles) and chest muscles (predominantly slow oxidative muscles) over this period. In all cases the concentration of free cholesterol exceeds that of esterified cholesterol. In dystrophic mice, similar patterns are found in brain and liver. In both thigh and chest muscles, however, the developmental pattern is significantly different. After an initial decrease the concentrations of cholesterol and cholesteryl esters increase rapidly with the largest increase occurring in the concentration of cholesteryl esters which by 10 weeks of age exceeds the concentration of cholesterol in chest muscle. During the same period the pattern of esterified fatty acids changes gradually in dystrophic tissues towards an increasing ratio of unsaturated/saturated fatty acids. By 10 weeks of age this ratio is significantly higher in dystrophic tissues than normal in all tissues tested.     

Esterified cholesterol and triglyceride are present in plasma membranes of Chinese hamster ovary cells.

The chemical composition of highly purified plasma membrane preparations from a series of malignant Chinese hamster ovary (CHO) cell lines were undertaken to ascertain if neutral lipid, including cholesteryl ester and triacylglycerol, were present. Triacylglycerols (33-41 nmol/mg total lipid) and cholesteryl ester (226-271 nmol/mg) were measured in the plasma membranes and differences in the chemical composition of these membranes recorded. The most significant difference was a gradual decrease in the level of free cholesterol from wild type (312 +/- 7 nmol/mg total plasma membrane lipid), Pod RII-6 (268 +/- 64 nmol/mg total plasma membrane lipid), Col R-22 (243 +/- 39 nmol/mg total plasma membrane lipid) to EOT (204 +/- 20 nmol/mg total plasma membrane lipid), with a concomitant increase in the degree of saturation of the cholesteryl ester fatty acids, particularly palmitic acid. No statistically significant differences were apparent in the chemical composition of the whole cells in this series. The one-dimensional (1D) 1H-NMR spectra of the four malignant cell lines showed a gradation in intensity of lipid resonances, in the order of wild type, Pod RII-6, Col R-22 and EOT, with EOT having the strongest lipid spectrum. Interestingly, the increase in acyl-chain signal intensities in the 1H-NMR spectra of this series of CHO cells and emergence of signals from cholesterol and/or cholesteryl ester, coincide with alterations in the amount of free cholesterol and the degree of saturation of the fatty-acyl chain of the esterified cholesterol in the plasma membranes. It is our hypothesis that, together, cholesteryl ester and triacylglycerol form domains in the plasma membrane and that when the cholesteryl ester has a largely saturated fatty acid content, the lipids are in isotropic liquid phase and hence visible by NMR.     

Lipid composition of the vascular system during infancy, childhood, and young adulthood

The object of this study was to determine the changes in lipid composition that occur in blood vessels from infancy to young adulthood. Analyses included levels of total cholesterol, total triglyceride, phospholipid, and cholesteryl ester fatty acids, and the distribution of saturated and unsaturated fatty acids. Triglyceride, total and monoenoic fatty acids, and linoleic acid were lower in the ascending, thoracic, and abdominal aorta than in the pulmonary artery and inferior vena cava. Phospholipids and arachidonic acid were higher in aortic segments than in the other two vessels. Aortic lipids showed significant changes with increasing age: total cholesterol and total fatty acids decreased from <1 wk to 5 yr, then increased to 22 yr of age. Triglycerides decreased whereas cholesteryl esters increased from 10 to 22 yr of age. Saturated fatty acids decreased from 1 wk to 10 yr, then remained relatively constant. Linoleic acid (3.7-9.8% of total fatty acids) and arachidonic acid (15.8-21.7%) both increased with age; the increase in cholesteryl linoleate was highly significant. After 10 yr of age, total cholesterol and total fatty acids were significantly higher in abdominal than in ascending and thoracic segments of aorta.     

Sterol accumulation in developing ovaries of the silkworm Bombyx mori in relation to the diapause hormone

Ovary sterol of the silkworm Bombyx mori during pupal-adult development was investigated using ‘diapause’ and ‘non-diapause’ ovaries. It is composed of cholesterol with high relative concentration of more than 80%, β-sitosterol less than 20% and a trace of campesterol. Cholesterol and β-sitosterol contents per pair of ovaries increased strikingly as ovary development proceeded. In sterol concentration per wet weight of ovaries, free forms of cholesterol and β-sitosterol did not increase, or rather inclined to decrease in the former and were almost constant in the latter. However, the ester forms, especially cholesterol ester, increased remarkably during oöcyte maturation. This suggests that this ester may preferably be deposited in oöcyte yolk.Furthermore, the accumulation of sterol esters especially cholesterol ester was higher in ‘diapause’ than ‘non-diapause’ ovaries. It is highly probable that the diapause hormone stimulates the accumulation of cholesterol ester in oöcytes undergoing oögenesis.     

Effects of cholesterol content on the metabolism of protein-free emulsion models of lipoproteins

After intravenous injection, emulsions with compositions similar to chylomicrons behaved metabolically as described for chylomicrons, with faster removals of triacylglycerols than cholesteryl esters from the blood after injection into rats, and with greater uptakes of cholesteryl esters than triacylglycerols by the liver. In contrast, emulsions with a high content of free cholesterol showed equal removal rates from the blood of triacylglycerols and cholesteryl esters; and similar uptakes by the liver. This pattern of metabolism was that expected for a chylomicron core remnant particle. Emulsions poor in cholesteryl ester but rich in free cholesterol showed remnant-like behavior, whereas emulsions rich in cholesteryl ester but poor in free cholesterol were metabolized like nascent chylomicron particles. The amount of free cholesterol appeared to regulate metabolism by affecting the binding of apolipoproteins to the particle surface. Emulsions with a high content of free cholesterol bound less A-I, A-IV and C apolipoproteins, and the relative amount of apolipoprotein E was increased. All of these effects are consistent with the metabolic differences between chylomicrons and remnant particles, suggesting that the amount of free cholesterol plays a regulatory role in chylomicron metabolism.     

Cholesteryl ester turnover in human plasma lipoproteins during cholestyramine and clofibrate therapy

The effects of cholestyramine and of clofibrate on the turnover rates of individual cholesteryl esters in whole human plasma and in each of the three classes of plasma lipoproteins have been studied. Four hyperlipidemic patients (two under treatment with each of the two drugs) were injected intravenously with cholesterol-(14)C, and serial plasma samples were collected after 3-4 hr, 8 hr, 24 hr, and 4-5 days. The plasma samples were separated into three classes of lipoproteins by ultracentrifugation. The cholesteryl esters and free cholesterol were isolated from each sample, and the specific radioactivity of the free and esterified cholesterol was determined. The specific radioactivity of each individual cholesteryl ester was then determined for each sample, by separately measuring the distribution of cholesterol mass and of radioactivity among four different cholesteryl ester groups, namely the saturated, mono-, di-, and tetra-unsaturated esters. In all subjects the plasma cholesteryl esters were metabolically heterogeneous, and could be divided into three pools corresponding to the three classes of plasma lipoproteins. High density lipoprotein (d > 1.063) cholesteryl esters showed the greatest fractional turnover rate, and low density lipoprotein (d 1.019-1.063) cholesteryl esters showed the smallest fractional turnover rate. In each subject the cholesteryl ester composition of the three classes of plasma lipoprotein was almost identical. Within each lipoprotein, and in whole plasma, all the different individual cholesteryl esters were found to turn over at the same fractional rate. In all respects these results were similar to those previously obtained with normal subjects. The results suggest that neither drug has a strongly selective effect on the turnover of one particular cholesteryl ester, or on the turnover or composition of the cholesteryl esters in one particular plasma lipoprotein.     

Hepatic processing of the cholesteryl ester from low density lipoprotein in the rat

Human low density lipoprotein (LDL), radiolabeled in the cholesteryl ester moiety, was injected into estrogen-treated and -untreated rats. The hepatic and extrahepatic distribution and biliary secretion of [3H]cholesteryl esters were determined at various times after injection. In order to follow the intrahepatic metabolism of the cholesteryl esters of LDL in vivo, the liver was subfractioned into parenchymal and Kupffer cells by a low temperature cell isolation procedure. In control rats, the LDL cholesteryl esters were mainly taken up by the Kupffer cells. After uptake, the [3H]cholesteryl esters are rapidly hydrolyzed, followed by release of [3H]cholesterol from the cells to other sites in the body. Up to 24 h after injection of LDL, only 9% of the radioactivity appeared in the bile, whereas after 72 h, this value was 30%. Hepatic and especially the parenchymal cell uptake of [3H]cholesteryl esters from LDL was strongly increased upon 17 alpha-ethinylestradiol treatment (3 days, 5 mg/kg). After rapid hydrolysis of the esters, [3H]cholesterol was both secreted into bile (28% of the injected dose in the first 24 h) as well as stored inside the cells as re-esterified cholesterol ester. It is concluded that uptake of human LDL by the liver in untreated rats is not efficiently coupled to biliary secretion of cholesterol (derivatives), which might be due to the anatomical localization of the principal uptake site, the Kupffer cells. In contrast, uptake of LDL cholesterol ester by liver hepatocytes is tightly coupled to bile excretion. The Kupffer cell uptake of LDL might be necessary in order to convert LDL cholesterol (esters) into a less toxic form. This activity can be functional in animals with low receptor activity on hepatocytes, as observed in untreated rats, or after diet-induced down-regulation of hepatocyte LDL receptors in other animals.     

Sex differences in plasma cholesterol-esterifying activity in rats

Esterification of free cholesterol was investigated after incubation at 37 degrees C of plasma from immature and adult rats of both sexes kept on stock, fat-free, or cholesterol-supplemented diets. According to measurements of the decrease in free cholesterol, plasma from the fat-deficient rats showed the highest cholesterol-esterifying activity. Esterification was higher in the mature female rats than in the mature males on stock or cholesterol-containing diets, although no sex differences were observed in the sexually immature young or in the fat-free animals. There were no sex differences in the fatty acid composition of the plasma sterol esters, phospholipids, and triglycerides in the immature animals, but arachidonic acid increased at the expense of linoleic acid in the sterol ester fraction in the adult female (not, however, in the adult male). In the phospholipid fraction the higher ratio of palmitic to stearic acids in the male was confirmed. There was an increase in linoleic acid in all three plasma lipid fractions of the mature male after cholesterol feeding. It is suggested that cholesterol may inhibit the conversion of linoleate to arachidonate. During the incubation of plasma, there was little change in the distribution of fatty acids except for a decrease in palmitoleate, and increases in C(20) tri- and tetraenoic acids, in the sterol esters of mature female rats on the stock ration and the fat-free diet. These C(20) acids decreased concomitantly in the phospholipid fraction, as the transesterification reaction mechanism proposed by earlier workers would predict.     

Omega-3 fatty acids increase the arachidonic acid content of liver cholesterol ester and plasma triacylglycerol fractions in the rat.

Recent studies have demonstrated that dietary fish oils rich in eicosapentaenoic acid (C20:5,omega 3) lower the content of arachidonic acid and its metabolites in plasma and tissue phospholipids. The present study examined the fatty acid composition of cholesterol ester and triacylglycerol fractions from plasma and livers of rats fed diets enriched with saturated fatty acids (beef tallow), alpha-linolenic acid (linseed oil) or eicosapentaenoic acid (fish oil). Feeding diets containing linseed oil or fish oil for 28 days increased arachidonic acid (C20:4,omega 6) levels in the cholesterol ester fraction of liver and in the triacylglycerol fraction of the plasma lipids. Plasma cholesterol esters were depleted of C20:4,omega 6 after feeding of the diet containing either linseed oil or fish oil. The changes in C20:4,omega 6 content cannot be explained by alterations in cholesterol ester or triacylglycerol pools of plasma and liver. These results suggest that the decrease in phospholipid C20:4,omega 6 content generally observed after fish oil consumption may be partly due to a shift of C20:4,omega 6 from phospholipid to the triacylglycerol and/or cholesterol ester pools in the same tissue. Triacylglycerols and cholesterol esters may therefore play a buffering role in the homeostatic maintenance of tissue phospholipid levels of arachidonic acid.     

Lysosomal acid lipase deficiency in rats: lipid analyses and lipase activities in liver and spleen

We report the biological characterization of an animal model of a genetic lipid storage disease analogous to human Wolman's disease. Affected rats accumulated cholesteryl esters (13.3-fold), free cholesterol (2.8-fold), and triglycerides (5.4-fold) in the liver, as well as cholesteryl esters (2.5-fold) and free cholesterol (1.33-fold) in the spleen. Triglycerides did not accumulate, and the levels actually decreased in the spleen. Analysis of the fatty acid composition of the cholesteryl esters and triglycerides showed high percentages of linoleic acid (18:2) and arachidonic acid (20:4) in both organs, especially in the liver. No accumulation of phospholipids, neutral glycosphingolipids, or gangliosides was found in the affected rats. Acid lipase activity for [14C]triolein, [14C]cholesteryl oleate, and 4-methyl-umbelliferyl oleate was deficient in both the liver and spleen of affected rats. Lipase activity at neutral pH was normal in both liver and spleen. Heterozygous rats showed intermediate utilization of these substrates in both organs at levels between those for affected rats and those for normal controls, although they did not accumulate any lipids. These data suggest that these rats represent an animal counterpart of Wolman's disease in humans.