Physiologic and pharmacologic factors influencing glyceroneogenic contribution to triacylglyceride glycerol measured by mass isotopomer distribution analysis

An imbalance between triacylglycerol synthesis and breakdown is necessary for the development of obesity. The direct precursor for triacylglycerol biosynthesis is alpha-glycerol phosphate, which can have glycolytic and glyceroneogenic origins. We present a technique for determining the relative glyceroneogenic contribution to triacylglyceride glycerol by labeling the glycerol moiety with 2H2O. The number of hydrogen atoms (n) incorporated from H2O into C-H bonds reflects the metabolic source of alpha-glycerol phosphate and can be calculated by combinatorial analysis of the distribution of mass isotopomers in triacylglyceride glycerol. Three physiological settings with potential effects on glyceroneogenesis and glycolysis were studied in rodents. Adipose tissue acylglyceride glycerol in mice fed a low carbohydrate diet had significantly higher values of n than in mice fed a high carbohydrate diet, suggesting an increased contribution from glyceroneogenesis of from 17 to 50% on the low carbohydrate diet. Similarly, mice administered rosiglitazone had a significant relative increase in glyceroneogenesis (from 17 to 53%), indicated by an increase in adipose acylglyceride glycerol n. Fructose infusion in overnight fasted rats rapidly lowered plasma triacylglyceride glycerol n, reflecting a decreased contribution from glyceroneogenesis (from 66 to 34%) presumably because of increased glycolytic input. In conclusion, we demonstrate that the number of C-H atoms derived from cellular H2O in triacylglyceride glycerol is an informative indicator of alpha-glycerol phosphate origin and, ultimately, triacylglycerol metabolism. Under certain physiological conditions, glyceroneogenesis can be up-regulated in adipose (e.g. low carbohydrate diet) or down-regulated in liver (e.g. fructose infusion). Additionally, stimulation of glyceroneogenesis by rosiglitazone in adipose tissue may be an important factor in the antilipolytic actions of thiazolidinediones.     

Comparison of deuterium incorporation and mass isotopomer distribution analysis for measurement of human cholesterol biosynthesis

To compare endogenous cholesterol biosynthesis measured by deuterium incorporation (DI) and mass isotopomer distribution analysis (MIDA), cholesterol fractional and absolute synthetic rates were measured simultaneously by both techniques under identical physiological conditions. Twelve subjects (22 to 39 years of age) underwent a dual stable isotope protocol, involving oral deuterium oxide administration and measurement of incorporation of deuterium into cholesterol coincident with constant infusion of sodium [1-(13)C]acetate and measurement of the mass isotopomer distribution pattern of newly synthesized cholesterol. Synthesis was determined over 24 h with a 7-h feeding period. Both methods yielded similar measurements of fractional cholesterol synthesis (7.8 +/- 2.5% day(-)(1) for DI vs. 6.9 +/- 2.2% day(-)(1) for MIDA). Correlation of fractional synthesis across techniques was strong (r = 0.84, P = 0.0007). Absolute synthesis rates were also not different at 24 h (13.4 +/- 4.3 mg kg(-)(1) day(-)(1) for DI vs. 11.9 +/- 3.6 mg kg(-)(1) day(-)(1) for MIDA, r = 0.79, P < 0.002).We conclude that despite different assumptions and analytical requirements, deuterium incorporation and MIDA yield similar rates of cholesterogenesis in humans when measurements are made over 24 h. The decision as to which method to adopt depends on available clinical and analytical facilities     

Measurement of gluconeogenesis in exercising men by mass isotopomer distribution analysis.

We evaluated the hypothesis that coordinated adjustments in absolute rates of gluconeogenesis (GNG(ab)) and hepatic glycogenolysis (Gly) would maintain euglycemia and match glucose production (GP) to peripheral utilization during rest and exercise. Specifically, we evaluated the extent to which gradations in exercise power output would affect the contribution of GNG(ab) to GP. For these purposes, we employed mass isotopomer distribution analysis (MIDA) and isotope-dilution techniques on eight postabsorptive (PA) endurance-trained men during 90 min of leg cycle ergometry at 45 and 65% peak O(2) consumption (VO(2 peak); moderate and hard intensities, respectively) and the preceding rest period. GP was constant in resting subjects, whereas the fraction from GNG (f(GNG)) increased over time during rest (22.3 +/- 0.9% at 11.25 h PA vs. 25.6 +/- 0.9% at 12.0 h PA, P < 0.05). In the transition from rest to exercise, GP increased in an intensity-dependent manner (rest, 2.0 +/- 0.1; 45%, 4.0 +/- 0.4; 65%, 5.84 +/- 0.64 mg. kg(-1). min(-1), P < 0.05), although glucose rate of disappearance exceeded rate of appearance during the last 30 min of exercise at 65% VO(2 peak). Compared with rest, increases in GP were sustained by 92 and 135% increments in GNG(ab) during moderate- and hard-intensity exercises, respectively. Correspondingly, Gly (calculated as the difference between GP and MIDA-measured GNG(ab)) increased 100 and 203% over rest during the two exercise intensities. During moderate-intensity exercise, f(GNG) was the same as at rest; however, during the harder exercise f(GNG) decreased significantly to account for only 21% of GP. The highest sustained GNG(ab) observed in these trials on PA men was 1.24 +/- 0.3 mg. kg(-1). min(-1). We conclude that, after an overnight fast, 1) absolute GNG rates increased with intensity of effort despite a reduced f(GNG) at 65% VO(2 peak), 2) during exercise Gly is more responsible than GNG(ab) for maintaining GP, and 3) in 12-h fasted men, neither increased Gly or GNG(ab) nor was their combination able to maintain euglycemia during prolonged hard (65% VO(2 peak)) exercise.     

Regulation of cholesterol 7 alpha-hydroxylase by hepatic 7 alpha-hydroxylated bile acid flux and newly synthesized cholesterol supply

We measured hepatic cholesterol 7 alpha-hydroxylase activity, mass, and catalytic efficiency (activity/unit mass) in bile fistula rats infused intraduodenally with taurocholate and its 7 beta-hydroxy epimer, tauroursocholate, with or without mevalonolactone to supply newly synthesized cholesterol. Enzyme activity was measured by an isotope incorporation assay and enzyme mass by densitometric scanning of immunoblots using rabbit anti-rat liver cholesterol 7 alpha-hydroxylase antisera. Cholesterol 7 alpha-hydroxylase activity increased 6-fold, enzyme mass 34%, and catalytic efficiency 5-fold after interruption of the enterohepatic circulation for 48 h. When taurocholate was infused to the bile acid-depleted animals at a rate equivalent to the hepatic bile acid flux (27 mumol/100-g rat/h), cholesterol 7 alpha-hydroxylase activity and enzyme mass declined 60 and 61%, respectively. Tauroursocholate did not significantly decrease cholesterol 7 alpha-hydroxylase activity, mass and catalytic efficiency. The administration of mevalonolactone, which is converted to cholesterol, modestly increased cholesterol 7 alpha-hydroxylase activity and enzyme mass in the bile acid-depleted rats. However, when taurocholate was infused together with mevalonolactone, cholesterol 7 alpha-hydroxylase activity and catalytic efficiency were markedly depressed while enzyme mass did not change as compared with bile acid-depleted rats. These results show that (a) hepatic bile acid depletion increases bile acid synthesis mainly by activating cholesterol 7 alpha-hydroxylase with only a small rise in enzyme mass, (b) replacement with taurocholate for 24 h decreases both cholesterol 7 alpha-hydroxylase activity and mass proportionally, (c) when cholesterol is available (mevalonolactone supplementation), the infusion of taurocholate results in the formation of a catalytically less active cholesterol 7 alpha-hydroxylase, and (d) tauroursocholate, the 7 beta-hydroxy epimer of taurocholate, does not inhibit cholesterol 7 alpha-hydroxylase. Thus, bile acid synthesis is modulated by the catalytic efficiency and mass of cholesterol 7 alpha-hydroxylase. The enterohepatic flux of 7 alpha-hydroxylated bile acids and the formation of hepatic cholesterol apparently control cholesterol 7 alpha-hydroxylase by different mechanisms.     

Evidence for separate pathways of transport of newly synthesized and preformed cholesterol into bile

Marked kinetic differences were observed when hepatic newly synthesized cholesterol and preformed cholesterol were separately radiolabeled and separately traced into bile. Whereas newly synthesized cholesterol was not evenly distributed throughout the liver but was preferentially secreted into bile, preformed cholesterol was in near-complete equilibrium in the whole liver and bile. Furthermore, whereas newly synthesized cholesterol in bile originated from the interior of the hepatocyte, results suggest that biliary preformed cholesterol may be transported directly from the blood through the plasma membrane of the hepatocyte and secreted from the canaliculus without first entering the interior of the cell and mixing with newly synthesized cholesterol.     

Studies of glycogen synthesis and the Krebs cycle by mass isotopomer analysis with [U-13C]glucose in rats

Starved rats were infused intragastrically via indwelling duodenal cannulae with glucose at a rate of 30 mg/min/kg. The infusate contained [U-13C]glucose at an enrichment of 32 or 17%. At the end of the infusion, after 160 min, glucose and lactate were isolated from arterial and portal blood and from liver, and liver glycogen was isolated and hydrolyzed to glucose. The enrichment in glucose and lactate and the isotopomer distribution in glucose of masses from 180 to 186 were determined by gas chromatography-mass spectrometry (GC-MS). From analysis of these data we determined (a) gluconeogenesis proceeds at half the basal rate in the presence of a large infused glucose load, (b) one-quarter of the hepatic pyruvate pool is derived from nonglucose carbon, (c) half of the labeled molecules in liver glycogen are of mass 186 from the infused glucose and half are of masses 181-183, (d) the contribution of the indirect path from pyruvate when corrected for synthesis from unlabeled pyruvate ranges from 55 to 65%, (e) the rate of pyruvate carboxylase averages 90% that of citrate synthase, and (f) the rate of exchange of oxaloacetate with fumarate is about three times the rate of flux in the Krebs cycle (four times in the "forward" direction), and the enrichment in carbon 1 of oxaloacetate was 2.3 times that in carbon 4. In the Appendix a method to obtain the isotopomer distribution of newly formed glucose and glycogen glucose is described. An algorithm to correct for the contribution of natural abundance of 13C and the presence of 12C in commercial [U-13C]glucose is presented. A novel mathematical analysis to obtain the parameters of the Krebs cycle from the isotopomer distribution is developed in the Appendix. Equations to calculate the relative rates of pyruvate carboxylase (y), and the equilibration of oxaloacetate with fumarate from the isotopomer distribution are derived. Mass isotopomer analysis provides a novel and powerful tool for the study of carbohydrate metabolism and the operation of the Krebs cycle.     

Secretion of the newly synthesized cholesterol by rat hepatocytes in primary culture

We used monolayer cultured rat hepatocytes as an experimental model to study the secretion of the newly synthesized cholesterol by the liver. Cellular cholesterol was labeled by exposing cultured hepatocytes to [14C]acetate prior to the study of secretion. Secretion of the newly synthesized cholesterol was measured by extracting cholesterol in the culture medium and assaying for the radioactivity of [14C]cholesterol. We found that: (a) cultured hepatocytes could secrete newly synthesized cholesterol in serum-free medium; (b) secreted [14C]cholesterol was bound to macromolecule(s) and the secretion rate was not affected by cycloheximide for up to 5 h; (c) serum added to the culture medium greatly enhanced hepatic cholesterol secretion; (d) serum high-density lipoproteins were most effective, lipoprotein-deficient serum (d greater than 1.21) less effective in stimulating cholesterol secretion, whereas low-density and very-low-density lipoproteins had little effect; (e) when the serum-free culture medium was fractionated by ultracentrifugation, a major portion of the secreted [14C]cholesterol was found in the high-density lipoprotein fraction; (f) part of the medium [14C]cholesterol also turned up in the high-density lipoprotein fraction when lipoprotein-deficient serum was added as the acceptor; (g) secreted [14C]cholesterol was found only in free form, although some of the cellular [14C]cholesterol was found as esters.     

Channeling of newly synthesized fatty acids to cholesterol esterification limits triglyceride synthesis in SND1-overexpressing hepatoma cells

SND1 is a putative oncoprotein whose molecular function remains unclear. Its overexpression in hepatocellular carcinoma impairs cholesterol homeostasis due to the altered activation of the sterol regulatory element-binding protein (SREBP) 2, which results in the accumulation of cellular cholesteryl esters (CE). In this work, we explored whether high cholesterol synthesis and esterification originates changes in glycerolipid metabolism that might affect cell growth, given that acetyl-coenzyme A is required for cholesterogenesis and fatty acids (FA) are the substrates of acyl-coenzyme A:cholesterol acyltransferase (ACAT). SND1-overexpressing hepatoma cells show low triglyceride (TG) synthesis, but phospholipid biosynthesis or cell growth is not affected. Limited TG synthesis is not due to low acetyl-coenzyme A or NADPH availability. We demonstrate that the main factor limiting TG synthesis is the utilization of FAs for cholesterol esterification. These metabolic adaptations are linked to high Scd1 expression, needed for the de novo production of oleic acid, the main FA used by ACAT. We conclude that high cholesterogenesis due to SND1 overexpression might determine the channeling of FAs to CEs.     

Phosphatidylcholine de novo synthesis and modification are carried out sequentially in HL60 cells: evidence from mass isotopomer distribution analysis

The traditional (parallel) model of molecular species synthesis of phosphatidylcholine is based on the substrate specificity of two glycerolphosphate acyltransferases. Preformed molecular species of diacylglycerols are then converted to phosphatidylcholine. In this investigation, we used [1,2,3,4-(13)C(4)]palmitate as a tracer to determine the turnover rates of diacylglycerols and phosphatidylcholines. In HL60 cells, the fractional turnover rate is 34.1 +/- 16.6%/h for 1,2-dipalmitoylglycerophosphocholine (16:0,16:0-GPC), which accounts for approximately 10% of total diacylglycerol turnover. The turnover rates of other phosphotidylcholines reflect the primary event of 16:0,16:0-GPC turnover. In addition, the distribution of mass isotopomers is used to study the biosynthesis of diacylglycerols and phosphatidylcholines. On the basis of precursor-product enrichments, we propose a sequential model to account for the synthesis of phosphatidylcholine molecular species. In this model, 1,2-dipalmitoylglycerol is the only molecular species used for the synthesis of phosphatidylcholine. This precursor is converted to 1,2-dipalmitoylglycerophosphocholine, which is then deacylated to provide substrates for chain elongation and/or desaturation. These modified acyl substrates are then reacylated back to form other molecular species. This sequential model is consistent with palmitate being the dominant fatty acid product derived from mammalian fatty acid synthase. It has the advantage of protecting cells from acyl modification by exogenous substrates. Furthermore, this sequence generates only inert 1,2-dipalmitoylglycerol instead of the active diacylglycerol molecular species that contain unsaturated fatty acids.     

Fate of exogenous and newly synthesized cholesterol in intestinal cell lines.

1. The current study was undertaken to test the existence of functionally distinct intracellular pools of cholesterol depending on the origin: neosynthesis or exogenous. 2. This was performed on two subpopulations, either differentiated or undifferentiated, of the HT29 cell line. 3. A parallel study was also carried out on Caco-2 cells. 4. First we checked the ability of differentiated HT29 cells to secrete lipids into the medium and found that lipid production was efficient but less so than in Caco-2 cells. 5. In contrast, undifferentiated HT29 cells were unable to secrete lipids into the medium. 6. Then we studied the fate of [14C]cholesterol incorporated into micellar preparations and of [14C]mevalonate in the different models. 7. The data obtained with labelled exogenous cholesterol show that it enters the membrane cholesterol pool as well as, for the differentiated models, the cholesteryl ester pool. 8. Similarly, labelled newly synthesized cholesterol could be used for membrane formation as well as for incorporation into cholesteryl esters. 9. Thus, in HT29 subpopulations as well as in Caco-2 cells, the results suggest the existence of a common pool of cholesterol whatever its origin.     

Activated polyamine catabolism leads to low cholesterol levels by enhancing bile acid synthesis

Transgenic mice with activated polyamine catabolism due to overexpression of spermidine/spermine N¹-acetyltransferase (SSAT) have significantly reduced plasma total cholesterol levels. In our study, we show that low cholesterol levels were attributable to enhanced bile acid synthesis in combination with reduced cholesterol absorption. Hepatic cholesterol 7α-hydroxylase (CYP7A1), the rate-limiting enzyme catalyzing the conversion of cholesterol to bile acids, plays an important role in the removal of excess cholesterol from the body. We suggest that by reducing activity of Akt activated polyamine catabolism increased the stability and activity of peroxisome proliferator-activated receptor γ co-activator 1α, the critical activator of CYP7A1. This is supported by our finding that the treatment with SSAT activator, N ¹,N ¹¹-diethylnorspermine, reduced significantly the amount of phosphorylated (active) Akt in HepG2 cells. In summary, activated-polyamine catabolism is a novel mechanism to regulate bile acid synthesis. Therefore, polyamine catabolism could be a potential therapeutic target to control hepatic CYP7A1 expression.     

Regulation of bile acid synthesis. III. Correlation between biliary bile salt hydrophobicity index and the activities of enzymes regulating cholesterol and bile acid synthesis in the rat

Hepatic bile acid synthesis is thought to be under negative feedback control by bile salts in the enterohepatic circulation, acting at the level of cholesterol 7 alpha-hydroxylase (C7 alpha H), the initial and rate-limiting step in the bile acid biosynthetic pathway. Bile salts also suppress the activity of the rate-limiting enzyme for cholesterol synthesis, 3-hydroxy-3-methylglutaryl-coenzyme A reductase (HMG-CoA-R). The mechanisms of these regulatory effects are poorly understood, and one or both may be indirect. Previous data suggest that the hydrophilic-hydrophobic balance of bile salts, a major determinant of their cholesterol solubilizing properties, also determines their potency as regulators of bile acid and cholesterol synthesis. To further evaluate the relationship between the physicochemical and regulatory properties of bile acids, we altered the composition of the bile salt pool of rats by feeding one or more of seven different bile acids (1% w/w for 14 days). We then determined the mean hydrophilic-hydrophobic balance (hydrophobicity index) of the bile salts in bile, and correlated this with the specific activities of C7 alpha H and HMG-CoA-R, and of acyl-CoA:cholesterol acyltransferase (ACAT), a third hepatic microsomal enzyme which regulates cholesterol esterification. In all instances following bile acid feeding, conjugates of the fed bile acid(s) became the predominant bile salts in bile. Highly significant negative linear correlations (each P less than 0.0001) were found between the hydrophobicity indices of biliary bile salts and the activities of C7 alpha H (r = 0.79) or HMG-CoA-R (r = 0.63). By contrast, no significant correlation could be demonstrated between ACAT activity and the hydrophobicity index of biliary bile salts. The correlation between activities of HMG-CoA-R and C7 alpha H was also highly significant (r = 0.81; P less than 0.0001). No significant correlation existed between ACAT and either HMG-CoA-R or C7 alpha H. Microsomal free cholesterol was not consistently altered by bile acid feeding. Thus, the potency of circulating bile salts as suppressors of the enzymes regulating bile acid and cholesterol synthesis increases with increasing hydrophobicity. The hydrophobic-hydrophilic balance of the bile salt pool may play an important role in the regulation of cholesterol and bile acid synthesis.     

Recent advancements in mass spectrometry–based tools to investigate newly synthesized proteins

Tight regulation of protein translation drives the proteome to undergo changes under influence of extracellular or intracellular signals. Despite mass spectrometry–based proteomics being an excellent method to study differences in protein abundance in complex proteomes, analyzing minute or rapid changes in protein synthesis and abundance remains challenging. Therefore, several dedicated techniques to directly detect and quantify newly synthesized proteins have been developed, notably puromycin-based, bio-orthogonal noncanonical amino acid tagging–based, and stable isotope labeling by amino acids in cell culture–based methods, combined with mass spectrometry. These techniques have enabled the investigation of perturbations, stress, or stimuli on protein synthesis. Improvements of these methods are still necessary to overcome various remaining limitations. Recent improvements include enhanced enrichment approaches and combinations with various stable isotope labeling techniques, which allow for more accurate analysis and comparison between conditions on shorter timeframes and in more challenging systems. Here, we aim to review the current state in this field.     

Role of cholesterol synthesis in regulation of bile acid synthesis and biliary cholesterol secretion in humans

We used lovastatin, a specific inhibitor of HMG-CoA reductase, to study the role of cholesterol synthesis in regulation of both bile acid synthesis, measured by release of 14CO2 from [26-14C]cholesterol, and biliary cholesterol secretion, measured by standard marked perfusion techniques, in humans. Six volunteers were studied in each of four periods: a) control; b) 6-10 hours after a single 40 mg oral dose of lovastatin to study acute effects; c) after 5-6 weeks of lovastatin 40 mg orally twice a day to study steady-state effects; and d) 24 h after cessation of chronic lovastatin. Mean bile acid synthesis fell to 69% of control (P less than 0.01) after single-dose lovastatin and remained at 83% of control after 5-6 weeks on lovastatin (P less than 0.05). After withdrawal of lovastatin, mean bile acid synthesis was 88% of control (NS). Mean biliary cholesterol secretion did not change after single-dose lovastatin (103% of control), but fell to 81% of control during chronic lovastatin treatment (P less than 0.05). After withdrawal of lovastatin, mean cholesterol secretion remained at 80% of control (P less than 0.05). These data suggest that in humans cholesterol synthesis is an immediate regulator of bile acid synthesis. Cholesterol synthesis also regulates biliary cholesterol secretion, but the effect is not immediate and therefore may be indirect.     

Interaction of bile salt monomer and cholesterol in the aqueous phase

The purpose of the present study was to evaluate the possible interaction of bile salt monomer and cholesterol in the intermicellar aqueous phase. Cholesterol and taurocholate monomer concentrations in the intermicellar aqueous phase were determined using 0-20 mM taurocholate solutions saturated with cholesterol. Maximal solubilities of cholesterol in aqueous solutions having various concentrations of taurocholate, especially below its intermicellar monomer concentration (critical micellar concentration), were determined and compared with the intermicellar cholesterol concentration. The intermicellar monomer concentration of taurocholate was constant (6 mM) and independent of taurocholate concentrations. The cholesterol concentration in the intermicellar aqueous phase gradually increased, depending upon taurocholate concentrations, and became constant (1,3 microM) above 10 mM taurocholate. The solubility of cholesterol increased linearly with the taurocholate concentration even below the critical micellar concentration, and was 0.3 microM at 6 mM taurocholate, which was approx. 20-times higher than the aqueous solubility of cholesterol, but a fifth of the maximal intermicellar cholesterol concentration. The results indicate that the higher cholesterol concentration in the intermicellar aqueous phase compared to its aqueous solubility can be primarily ascribed to the interaction of cholesterol with bile salt monomers possibly forming bile salt-cholesterol dimers, and partly to the sustaining forces induced by numerous micelles.     

The induction of lamellar stacking by cholesterol in lecithin-bile salt model systems and human bile studied by synchrotron X-radiation

Small angle X-ray scattering (SAXS) with synchroton radiation was used to investigate interactions among lipid particles in lecithin-bile salt model systems and in native gallbladder biles. In model systems in the absence of cholesterol, isotropic, continuous spectra were found, indicating the absence of periodic structures. In the presence of excess cholesterol, interaction in the form of lamellar stacking was detected by the appearance of discrete diffraction peaks. In the supersaturated cholesterol region of the commonly accepted phase diagram [1], where cholesterol crystals were expected, we found lamellar stacking. The high proportion of cholesterol to bile salts seems to be the common denominator of these models. The lamellar stacking was also found in native unprocessed bile. This effect of cholesterol on lipid structure has not been previously described. Lamellar stacking may contribute to cholesterol solubilization. Its influence on the kinetics of cholesterol crystallization is presently unknown.     

Autoradiographic studies on the protein synthesis and the distribution of the newly synthesized proteins in oligodendroglial cells

The protein-synthesizing activity of oligodendroglial cells in the rat hippocampus as well as the migration of the newly synthesized proteins within these cells were examined by means of light and microscopic autoradiography when injecting 3H-leucine into the lateral ventricle. The autoradiographs of oligodendroglial cells were compared with the autoradiographs on hippocampal pyramidal cells of the same area. According to the quantitative evaluation of the autoradiographs the protein-synthesizing activity of the oligodendrocytes equals or may even exceed that of the pyramidal cells. In oligodendroglial cells most of the newly synthesized proteins migrate very rapidly from the rough-surfaced endoplasmic reticulum into the cisternae of the Golgi apparatus and disappear continuously. The proteins disappear from the Golgi apparatus simultaneously with the decrease in the radioactive concentration of the perikaryon.     

The sites of synthesis and the subsequent migration of newly synthesized protein in retina

Radioautographs of rabbit retinas fixed immediately after a 1 or 2 min exposure in vitro to ³H leucine revealed high rates of protein synthesis in receptor cell inner segments, perikarya of ganglion cells, and cells of the inner nuclear layer. If these brieflly labelled retinas were returned to unlabelled medium for periods of up to 6 hr, the radioautographs revealed a progressive dispersion of the labelled proteins from their sites of synthesis. This was largely completed by $\text{1}\text{1}\text{2}$ hr and appeared, in one instance at least, to involve processes other than simple diffusion. Superimposed on the dispersive phenomenon was a process of concentration of the newly formed proteins at two sites quite distant from their synthesis, that was apparent after $\text{1}\text{1}\text{2}$hr. One of these sites was the receptor cell outer segments, as has been previously described, the other was the outer plexiform layer.