Uptake and metabolism by rat intestine CHARLES
M. MANSBACH
of circulating
II AND REGENIA F. DOWELL of Gastroenterology, The University
Department of Medicine, Division and The Veterans Affairs Medical
Center, Memphis,
Mansbach, Charles M., II, and Regenia F. Dowell. Uptake and metabolism of circulating fatty acids by rat intestine. Am. J. Physiol. 263 (Gastrointest. Liver Physiol. 26): G927G933 ,1992.-The presentstudy wasdesignedto investigate the uptake and metabolism of circulating fatty acids by the intestinal mucosain rats actively absorbingglyceryl trioleate given intraduodenally to determine the plasmafatty acid contribution to mucosaltriacylglycerol. Rats with duodenal, femoral vein, carotid artery, and mesentericlymph duct cannulaswere used. [3H]oleate was constantly infused into the femoral vein while glyceryl trioleate wasinfused into the duodenum (135 pmol/h). After 5 h of infusion, a massand radioactive steady state existed in the plasmaand mucosa.At 6 h of infusion, the plasmaoleate specific activity was sixfold greater than mucosaloleate and 50 times greater than mucosal triacylglycerol oleate; 86% of the mucosaloleate disintegrations/minute were in triacylglycerol. Chylomicron triacylglycerol oleate specific activity was less than that of the mucosa.Furthermore, the percentageof mucosaltriacylglycerol acyl groupscomposedof oleate was greater than the percentageof oleate in mucosalfree fatty acids. The data indicate that fatty acidsare taken up by the mucosaduring active fat absorption and metabolized primarily to triacylglycerolsby the mucosa.The triacylglycerols in the mucosasynthesized from circulating fatty acids are selectedagainst as a precursor of chylomicron triacylglycerol. The results support our previous hypothesis suggestingthat the mucosahasat leasttwo pools of neutral lipid (J. Lipid Res. 23: 1009-1019, 1982) and that steady-state conditions as performed here yield different results from previous work using bolus tracer injection techniques. lipid absorption; lipid pools; neutral lipid metabolism THE INTESTINAL MUCOSAL CELL isconsideredaproces-
sor of dietary lipids, because these important nutrients enter the body through this cell and undergo a series of complex metabolic steps within it. Endogenous fatty acids (FA), however, are also taken up and metabolized by the enterocyte as has been shown in both humans and rats (3,4). These abluminal lipids could be an important source of portal vein FA and could contribute to the large amount of endogenous FA found in the portal vein under conditions in which rats are absorbing large amounts of glyceryl trioleate (TO; 11). Only two previous studies have examined uptake and metabolism of endogenous FA by the intestinal cell (3, 4). These have indicated that the majority of the FA taken up by the enterocyte is metabolized to phospholipids and water-soluble metabolites (3, 4). The proportion of FA incorporated into triacylglycerol (TG) could be increased, however, by the infusion of a lipid emulsion (3). In the rat study, only a small amount of lipid was infused (31 pmol TG equivalents over 30 min), considerably less than the amount used in our previous experiments (135 pmol/h for 6 h; 11). Furthermore, the previous studies were performed after a bolus injection 0193~1857/92
fatty acids
$2.00
of Tennessee, Memphis
38163,
Tennessee 38104
of [3H]palmitate, a method that allows for tracing of the fate of the probe lipid but does not allow for mean steady-state comparisons of specific activities of mucosal lipids to chylomicron or very low-density lipoprotein (VLDL) lipids. The bolus injection technique results in sequential waves of radiolabeled precursors followed by the appearance of radiolabeled products as a function of time after bolus injection (27). The results reported here were performed at the radioactive steady state (8) in which [3H]oleate was constantly infused intravenously while TO was infused intraduodenally. This technique enables a lipid pool to have a constant specific activity so that two major pieces of information become available. The first is determining the extent to which a pool is diluted with exogenous lipids. This is calculated on the basis of the specific activity of a product pool divided by the specific activity of its precursor. Obviously if both are equal, then the entire product pool is derived from a single precursor. Alternatively, if the product pool has a specific activity less than its precursor, it can be assumed that the product pool is diluted by exogenous (nonradioactive) sources. The second piece of information to be generated from steady-state kinetic data is that a product pool can be shown to be derived from a restricted precursor pool, in which case the product pool will have a specific activity that exceeds that of the precursor pool. For example, when large amounts of TO are infused into rats, the precursor TG pool in the mucosa has a specific activity that is one-half that of the infusate, whereas the product pool, lymph TG, has a specific activity that nearly equals that of the infusate (12) . MATERIALSAND METHODS Male Sprague-Dawley rats weighing 295-345 g were used. They were maintained on Purina Rat Chow (Ralston Purina, St. Louis, MO). The day before an experiment, the rats received cannulas (PE-50, Clay Adams, Parsippany, NJ) in the right femoral vein, the carotid artery, the proximal duodenum,and the main mesentericlymph duct (8, 11). After insertion of the cannulas, the abdomen was closed in two layers, the carotid artery cannula waspluggedwith wax, and the rat wasplacedin a restraining cage.The duodenalcannula wasusedto infusethe rat overnight with 0.15 M NaCl, 0.3 M KCl, and 5% glucoseat 3 ml/h. The femoral vein cannulawasinfusedwith 0.15M NaCl at 0.2 ml/h. The following day, the rat’s duodenalinfusion was changedto 30 mM TO, 10 mM taurocholate (Sigma Chemical, St. Louis, MO), 0.15 M NaCl, and 10 mM tris(hydroxymethyl)aminomethane (Tris HCl), pH 7.0, which was infused at 4.5 ml/h for 6 h. The femoral vein infusion was also changedto include [3H]oleate (2.86 t 0.016 X lo6 disintegrations per minute (dpm)/h, New England Nuclear, Boston, MA) complexed to FA-poor bovine serum albumin (Sigma). [3H]oleate
Copyright 0 1992 The American Physiological
Society
G927
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 10, 2019.
G928
UPTAKE
AND
METABOLISM
OF CIRCULATING
was complexed to albumin (181, M:M) without carrier oleate so that the highest possible oleate specific activity could be achieved. The [3H]oleate infusion was continued during the 6 h of intraduodenal TO infusion. Preliminary experiments were conducted in which the sameprocedure was followed, but the infusion time was changedto 5 and 7 h as well as 6 h. At the end of a 6-h infusion (or 5-7 h in preliminary experiments), blood was obtained from the carotid artery cannula. The rat wasgiven an overdoseof pentobarbital sodium,and the intestines were rapidly removed to an iced glassplate. The plasmasand homogenizedmucosa(11) from the proximal half of the intestine were extracted (2) of their lipid content. Extractions of TG and FA were 99 and 97% efficient, respectively. The lipids were separatedby thin-layer chromatography (TLC) usinghexanes:diethylether:methanol:acetic acid (80:20:6:2,vol: vol) asthe mobilephase.The separatedlipids were identified by the cochromatography of authentic standards (Sigma), which were stained by exposing the plate to IZ vapors. The areas of each of the separatedlipids were scraped from the plate, the silica gel placed in a scintillation vial, and toluene-basedscintillant added.Radioactivity wasdetermined using a model 1500 Tri Carb (Packard Instrument, Downers Grove, IL). In someexperiments the partitioning of dpm between the organic and water-soluble phasesof mucosal extracts (2) was determined.In theseexperiments, 1 ml of homogenizedmucosa was extracted (2), and the volumes of the water and organic phaseswere measured;0.5 ml of each phasewas obtained and dried (Vortex-Evaporator, Haake Buchler Instruments, Saddle Brook, NJ) in a counting vial. Scintillant was added (8), and radioactivity was determined. The plasmaand mucosawerealsoextracted by the method of Warner et al. (25) to remove FA. The FA were derivatized with p-bromophenacyl bromide and separatedon a model 110 highperformance liquid chromatograph (Beckman Instruments, Fullerton, CA) usinga Du Pont Zorbax ODS column (Du Pont, Diagnostic and BioresearchSystems,Wilmington, DE) as previously described (11). The mobile phase was 15% acetonitrile:H,O 1:l (vol:vol) and 85% methanol at a flow rate of 1.5 ml/min for 10 min followed by a lo:90 mixture of the same solvents,respectively, for an additional 25 min. The derivatized lipids were detected, and the signal was continuously recorded at 254 nm. The column effluent wascollectedin 1-min samples. The solvent was evaporated to dryness, toluene-basedscintillant added,and the radioactivity of the sampledetermined. The massesof the various FA were calculated using a previously validated method (11) based on comparing “areas under the curve” to authentic standard values. Oleate radioactivity was assayedby collection of the column solvent in l-min samples, drying of them (Vortex-Evaporator), and determination of the radioactivity aspreviously described(11). Lymph was collected in hourly sampleson ice but only the lymph collected between 5 and 6 h of infusion was processed, except in the preliminary experiments when lymph from 4 to 5 h and 6 to 7 h of infusion was processed.The lymph was defibrinated, 1 ml waslayered beneath a density (d) = 1.006 solution composedof 102mM NaBr and 0.1% EDTA, pH 7.8, and centrifuged in a SW 40 rotor using a model L8-M ultracentrifuge (Beckman Instruments) at 17°C. Chylomicrons were floated by centrifugation for 3 x lo6 gemin and were harvested by a tube-slicingmethod. The remaining fluid wasadjustedwith more d = 1.006solution and recentrifuged for 3 x lo8 gomin to float the VLDL. The VLDL fraction wasobtained by slicing the tube. Both lipoprotein fractions were extracted of their lipid content (2; 99% efficiency), and the lipids were separated on thin-layer chromatographic (TLC) plates as described above. The TG spotswere identified using I2 vapors, scrapedfrom the
FATTY
ACIDS
plate, and the radioactivity was determined as above. TG masswasassayedby scrapingthe TG band from separate laneson the TLC plate and eluting it twice with 5-ml portions of chloroform:methanol 2:l (vol:vol) and once with 5 ml chloroform:methanol4:1. Any remaining silica gel was removedby centrifugation, and the solventswere dried under N2 in a lOO-ml round-bottomed flask. Five milliliters of 0.5 N NaOH was added, and the TG was refluxed for 6 h at 80°C. Twenty-five milliliters of 0.1 N HCl wasadded, and the hydrolyzed FA was extracted by addition of 100 ml chloroform:methanol 2:l. The mixture was allowedto sit in a cold room at 4°C overnight in a separatory funnel to separatethe phases.The lower phasewas drawn off and the solvent evaporated using a rotary evaporator (Buchi Instruments, Flavil, Switzerland). The FA wereremoved from the flask to a test tube by three washeswith 3-ml portions of chloroform:methanol 2:l. The solvents were evaporated to dryness under NS, and 100 ~1chloroform was added. The FA were derivatized for high-performance liquid chromatography as describedpreviously (11) except that 1.35 pmol NaOH was added.The TG masswas determined by combining the masses of the individual FAs and dividing by 3. This method agreed within 10% of measuring the glycerol content of TG as described previously (11). The dpm in oleate was assayedby collecting the effluent from the high-performanceliquid chromatographic column in l-ml fractions, drying the solvents, and determining the radioactivity in oleate as describedpreviously (11) The TO utilized in this study was 99% pure TO from Sigma. The dpm were calculated by using the external radioactive sourcesprovided with the model 1500 Tri Carb, which enables conversion of counts per minute to dpm by referenceto a series of quenchedstandards. Statistical evaluation was performed using Student’s unpaired or paired t test where appropriate. Calculations as to relationshipsbetweenspecific activities of precursorsand products were madeby dividing the specific activity of the precursor by its product. For example, the specific activity of plasmaFA oleate would be divided by the specific activity of mucosalFA oleate. RESULTS
To establish that the plasma and intestine
were under
steady-state conditions during intravenous [3H] oleate in-
fusion, preliminary experiments were conducted in which samples of blood were obtained from the carotid artery and mucosa from the intestine at 5, 6, and 7 h of intraduodenal TO infusion. The data (Table 1) show that mucosal TG mass and TG oleate specific activity were nearly equivalent between 5 and 7 h of infusion, confirming previous work using an intraduodenal [YJ]TO infusion of 90 pmol/h (12). The mucosal FA mass and FA oleate specific activity were also at steady-state levels. Finally, plasma FA oleate specific activity was nearly the same between 5 and 7 h of infusion. We have previously
shown that when [3H]T0 was infused intraduodenally at 135 pmol/h (8) a steady state also existed with regard to chylomicron transport. In sum the data presented in Table 1 and the previous data (12) indicate that the mucosa was under steady-state conditions as was the plasma FA oleate specific activity during the course of the experiments. After preliminary experiments had established that radioactive and mass steady states existed for the pertinent
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 10, 2019.
UPTAKE
AND METABOLISM
OF CIRCULATING
FATTY
G929
ACIDS
Table 1. Mucosa and plasma lipids are at steady state at 6 h of infusion Time, h Sample 5
Mucosa Triacylglycerol, pmol 154k12 Triacylglycerol, dpm/pmol* 3,353+324 FFA, pm01 40t6 FFA, dpm/pmol* 24,080+2,667 Plasma FFA, dpm/pmol* 144,772+ 14,747 Data are means =t SE. Each hour represents a set of 4 rats. * Data refer specifically acid.
lipids after 6 h of TO infusion intraduodenally and [3H]oleate intravenously, further experiments were conducted after 6 h of infusion (intraduodenal and intravenous). During the 6 h intraduodenal TO infusion, a total of 810 pmol TO was delivered to the gut, but only 123 t 9 pmol TG, or 15% of the TG infused, was present in the mucosa at the conclusion of the 6-h period. This supports previous data suggesting rapid clearance of TG from the mucosa under these conditions (9). At 6 h of infusion there was 42 & 4 pm01 total FA in the mucosa indicating that most of the absorbed acyl groups were present as TG because only 10% of the total acyl groups (FA + TG-FA) were as FA. Between 5 and 6 h of infusion, lymph flow was 4.8 & 0.6 ml. At the conclusion of the 6 h infusion, 732,250 t 41,505 dpm were present in the mucosal homogenate. When the mucosa was extracted (2), 99 t 0.2% of the dpm were in the organic phase with the remainder in the aqueous phase. This is in contrast to prior observations (3, 4) in which 42% of dpm were found in the aqueous phase. The difference between prior observations and our own is likely due to the bolus injection technique with observations at 2 min used by Gang1 et al. (3, 4) compared with the prolonged infusion method utilized in the present report. The turnover rate of mucosal lipids is expected to be much slower than its water-soluble metabolic products leading to a progressive increase in lipid dpm over time compared with its metabolites. We also found that chemiluminescence occurred (-10% of the total dpm) unless the aqueous phase was completely dried before addition of scintillant . As expected, the intravenously infused E3H]oleate was widely distributed into various lipid classes indicating considerable metabolism of the probe FA (Fig. 1). As shown, the predominant radiolabeled lipid in the carotid artery was TG followed by FA and then cholesteryl ester. Little dpm were distributed to the other lipids analyzed. In the mucosa, the results were quite different; TG comprised 86% of the total dpm present. These data indicate that plasma lipids taken up by the mucosa were extensively metabolized. Cholesteryl ester was not likely taken up by the intestine because it was significantly radiolabeled in the plasma but hardly at all in the mucosa (Fig. 1). The most likely plasma lipid to be taken up by the enterocyte is FA based on its specific activity in the mucosa compared with mucosal TG (see below). The specific activity of FA oleate in the plasma was
7
6
133*13 2,818*290 38t9 25,158f2,136
147&X 2,838*285 31k6 25,428&3,198
139,804&12,782 to oleate
152,459*15,218
or the oleoyl
CAROTID
PLASMA
moiety
specific
ARTERY
OR MUCOSAL
activity.
FFA, free fatty
MUCOSA
LIPID
CLASSES
Fig. 1. Percentage distribution of 3H into various lipids in carotid artery and intestinal mucosa after intravenous administration of [3H]oleak and intraduodenal infusion of glyceryl trioleate (135 pmol/h). Origin, phospholipids; MG, monoacylglycerol; DG, diacylglycerol; FA, fatty acid; TG, triacylglycerol; CE, cholesteryl esters. Data are means 2 SE of 5 experiments.
very high, 139,804 t 12,782 dpm/pmol. The specific activity of mucosal FA oleate at 6 h (25,158 dpm/pmol) was only 18% of that of plasma FA oleate (139,804 dpm/pmol) secondary to the large amount of absorbed nonradioactive oleate from the duodenally infused TO. Of particular interest are the data shown in Fig. 2, which show that the specific activity in mucosal TG oleate is further reduced ninefold (25,158/2,818, Table 1) compared with the specific activity of mucosal FA oleate, indicating that not all the FA oleate in the mucosa is equally utilized for TG synthesis. In this event, the specific activity of the FA oleate and TG oleate would have been equal. One way in which this would occur is the known conservation of monoacylglycerol (MG) for mucosal TG synthesis (12, l3), which in the present experiments would likely be sn-2-MG-oleate. This would mean that one of every three acyl groups of TG could not come from plasma FA, which could maximally account for a 1.5fold reduction (3 acyl groups of which a maximum of 2 could come from plasma) in specific activity. Because the observed specific activity reduction is much greater, the TG oleate in the mucosa must be preferentially derived from other exogenous (unlabeled, luminal) sources. Also shown in Fig. 2 is the specific activity of TG oleate in lymph chylomicrons and VLDL fractions. As indicated in Fig. 2, the specific activity of the mucosal TG oleate
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 10, 2019.
G930
UPTAKE 4000
1
AND METABOLISM
OF CIRCULATING
P*0.0002 I P
M. MANSBACH
of circulating
II AND REGENIA F. DOWELL of Gastroenterology, The University
Department of Medicine, Division and The Veterans Affairs Medical
Center, Memphis,
Mansbach, Charles M., II, and Regenia F. Dowell. Uptake and metabolism of circulating fatty acids by rat intestine. Am. J. Physiol. 263 (Gastrointest. Liver Physiol. 26): G927G933 ,1992.-The presentstudy wasdesignedto investigate the uptake and metabolism of circulating fatty acids by the intestinal mucosain rats actively absorbingglyceryl trioleate given intraduodenally to determine the plasmafatty acid contribution to mucosaltriacylglycerol. Rats with duodenal, femoral vein, carotid artery, and mesentericlymph duct cannulaswere used. [3H]oleate was constantly infused into the femoral vein while glyceryl trioleate wasinfused into the duodenum (135 pmol/h). After 5 h of infusion, a massand radioactive steady state existed in the plasmaand mucosa.At 6 h of infusion, the plasmaoleate specific activity was sixfold greater than mucosaloleate and 50 times greater than mucosal triacylglycerol oleate; 86% of the mucosaloleate disintegrations/minute were in triacylglycerol. Chylomicron triacylglycerol oleate specific activity was less than that of the mucosa.Furthermore, the percentageof mucosaltriacylglycerol acyl groupscomposedof oleate was greater than the percentageof oleate in mucosalfree fatty acids. The data indicate that fatty acidsare taken up by the mucosaduring active fat absorption and metabolized primarily to triacylglycerolsby the mucosa.The triacylglycerols in the mucosasynthesized from circulating fatty acids are selectedagainst as a precursor of chylomicron triacylglycerol. The results support our previous hypothesis suggestingthat the mucosahasat leasttwo pools of neutral lipid (J. Lipid Res. 23: 1009-1019, 1982) and that steady-state conditions as performed here yield different results from previous work using bolus tracer injection techniques. lipid absorption; lipid pools; neutral lipid metabolism THE INTESTINAL MUCOSAL CELL isconsideredaproces-
sor of dietary lipids, because these important nutrients enter the body through this cell and undergo a series of complex metabolic steps within it. Endogenous fatty acids (FA), however, are also taken up and metabolized by the enterocyte as has been shown in both humans and rats (3,4). These abluminal lipids could be an important source of portal vein FA and could contribute to the large amount of endogenous FA found in the portal vein under conditions in which rats are absorbing large amounts of glyceryl trioleate (TO; 11). Only two previous studies have examined uptake and metabolism of endogenous FA by the intestinal cell (3, 4). These have indicated that the majority of the FA taken up by the enterocyte is metabolized to phospholipids and water-soluble metabolites (3, 4). The proportion of FA incorporated into triacylglycerol (TG) could be increased, however, by the infusion of a lipid emulsion (3). In the rat study, only a small amount of lipid was infused (31 pmol TG equivalents over 30 min), considerably less than the amount used in our previous experiments (135 pmol/h for 6 h; 11). Furthermore, the previous studies were performed after a bolus injection 0193~1857/92
fatty acids
$2.00
of Tennessee, Memphis
38163,
Tennessee 38104
of [3H]palmitate, a method that allows for tracing of the fate of the probe lipid but does not allow for mean steady-state comparisons of specific activities of mucosal lipids to chylomicron or very low-density lipoprotein (VLDL) lipids. The bolus injection technique results in sequential waves of radiolabeled precursors followed by the appearance of radiolabeled products as a function of time after bolus injection (27). The results reported here were performed at the radioactive steady state (8) in which [3H]oleate was constantly infused intravenously while TO was infused intraduodenally. This technique enables a lipid pool to have a constant specific activity so that two major pieces of information become available. The first is determining the extent to which a pool is diluted with exogenous lipids. This is calculated on the basis of the specific activity of a product pool divided by the specific activity of its precursor. Obviously if both are equal, then the entire product pool is derived from a single precursor. Alternatively, if the product pool has a specific activity less than its precursor, it can be assumed that the product pool is diluted by exogenous (nonradioactive) sources. The second piece of information to be generated from steady-state kinetic data is that a product pool can be shown to be derived from a restricted precursor pool, in which case the product pool will have a specific activity that exceeds that of the precursor pool. For example, when large amounts of TO are infused into rats, the precursor TG pool in the mucosa has a specific activity that is one-half that of the infusate, whereas the product pool, lymph TG, has a specific activity that nearly equals that of the infusate (12) . MATERIALSAND METHODS Male Sprague-Dawley rats weighing 295-345 g were used. They were maintained on Purina Rat Chow (Ralston Purina, St. Louis, MO). The day before an experiment, the rats received cannulas (PE-50, Clay Adams, Parsippany, NJ) in the right femoral vein, the carotid artery, the proximal duodenum,and the main mesentericlymph duct (8, 11). After insertion of the cannulas, the abdomen was closed in two layers, the carotid artery cannula waspluggedwith wax, and the rat wasplacedin a restraining cage.The duodenalcannula wasusedto infusethe rat overnight with 0.15 M NaCl, 0.3 M KCl, and 5% glucoseat 3 ml/h. The femoral vein cannulawasinfusedwith 0.15M NaCl at 0.2 ml/h. The following day, the rat’s duodenalinfusion was changedto 30 mM TO, 10 mM taurocholate (Sigma Chemical, St. Louis, MO), 0.15 M NaCl, and 10 mM tris(hydroxymethyl)aminomethane (Tris HCl), pH 7.0, which was infused at 4.5 ml/h for 6 h. The femoral vein infusion was also changedto include [3H]oleate (2.86 t 0.016 X lo6 disintegrations per minute (dpm)/h, New England Nuclear, Boston, MA) complexed to FA-poor bovine serum albumin (Sigma). [3H]oleate
Copyright 0 1992 The American Physiological
Society
G927
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 10, 2019.
G928
UPTAKE
AND
METABOLISM
OF CIRCULATING
was complexed to albumin (181, M:M) without carrier oleate so that the highest possible oleate specific activity could be achieved. The [3H]oleate infusion was continued during the 6 h of intraduodenal TO infusion. Preliminary experiments were conducted in which the sameprocedure was followed, but the infusion time was changedto 5 and 7 h as well as 6 h. At the end of a 6-h infusion (or 5-7 h in preliminary experiments), blood was obtained from the carotid artery cannula. The rat wasgiven an overdoseof pentobarbital sodium,and the intestines were rapidly removed to an iced glassplate. The plasmasand homogenizedmucosa(11) from the proximal half of the intestine were extracted (2) of their lipid content. Extractions of TG and FA were 99 and 97% efficient, respectively. The lipids were separatedby thin-layer chromatography (TLC) usinghexanes:diethylether:methanol:acetic acid (80:20:6:2,vol: vol) asthe mobilephase.The separatedlipids were identified by the cochromatography of authentic standards (Sigma), which were stained by exposing the plate to IZ vapors. The areas of each of the separatedlipids were scraped from the plate, the silica gel placed in a scintillation vial, and toluene-basedscintillant added.Radioactivity wasdetermined using a model 1500 Tri Carb (Packard Instrument, Downers Grove, IL). In someexperiments the partitioning of dpm between the organic and water-soluble phasesof mucosal extracts (2) was determined.In theseexperiments, 1 ml of homogenizedmucosa was extracted (2), and the volumes of the water and organic phaseswere measured;0.5 ml of each phasewas obtained and dried (Vortex-Evaporator, Haake Buchler Instruments, Saddle Brook, NJ) in a counting vial. Scintillant was added (8), and radioactivity was determined. The plasmaand mucosawerealsoextracted by the method of Warner et al. (25) to remove FA. The FA were derivatized with p-bromophenacyl bromide and separatedon a model 110 highperformance liquid chromatograph (Beckman Instruments, Fullerton, CA) usinga Du Pont Zorbax ODS column (Du Pont, Diagnostic and BioresearchSystems,Wilmington, DE) as previously described (11). The mobile phase was 15% acetonitrile:H,O 1:l (vol:vol) and 85% methanol at a flow rate of 1.5 ml/min for 10 min followed by a lo:90 mixture of the same solvents,respectively, for an additional 25 min. The derivatized lipids were detected, and the signal was continuously recorded at 254 nm. The column effluent wascollectedin 1-min samples. The solvent was evaporated to dryness, toluene-basedscintillant added,and the radioactivity of the sampledetermined. The massesof the various FA were calculated using a previously validated method (11) based on comparing “areas under the curve” to authentic standard values. Oleate radioactivity was assayedby collection of the column solvent in l-min samples, drying of them (Vortex-Evaporator), and determination of the radioactivity aspreviously described(11). Lymph was collected in hourly sampleson ice but only the lymph collected between 5 and 6 h of infusion was processed, except in the preliminary experiments when lymph from 4 to 5 h and 6 to 7 h of infusion was processed.The lymph was defibrinated, 1 ml waslayered beneath a density (d) = 1.006 solution composedof 102mM NaBr and 0.1% EDTA, pH 7.8, and centrifuged in a SW 40 rotor using a model L8-M ultracentrifuge (Beckman Instruments) at 17°C. Chylomicrons were floated by centrifugation for 3 x lo6 gemin and were harvested by a tube-slicingmethod. The remaining fluid wasadjustedwith more d = 1.006solution and recentrifuged for 3 x lo8 gomin to float the VLDL. The VLDL fraction wasobtained by slicing the tube. Both lipoprotein fractions were extracted of their lipid content (2; 99% efficiency), and the lipids were separated on thin-layer chromatographic (TLC) plates as described above. The TG spotswere identified using I2 vapors, scrapedfrom the
FATTY
ACIDS
plate, and the radioactivity was determined as above. TG masswasassayedby scrapingthe TG band from separate laneson the TLC plate and eluting it twice with 5-ml portions of chloroform:methanol 2:l (vol:vol) and once with 5 ml chloroform:methanol4:1. Any remaining silica gel was removedby centrifugation, and the solventswere dried under N2 in a lOO-ml round-bottomed flask. Five milliliters of 0.5 N NaOH was added, and the TG was refluxed for 6 h at 80°C. Twenty-five milliliters of 0.1 N HCl wasadded, and the hydrolyzed FA was extracted by addition of 100 ml chloroform:methanol 2:l. The mixture was allowedto sit in a cold room at 4°C overnight in a separatory funnel to separatethe phases.The lower phasewas drawn off and the solvent evaporated using a rotary evaporator (Buchi Instruments, Flavil, Switzerland). The FA wereremoved from the flask to a test tube by three washeswith 3-ml portions of chloroform:methanol 2:l. The solvents were evaporated to dryness under NS, and 100 ~1chloroform was added. The FA were derivatized for high-performance liquid chromatography as describedpreviously (11) except that 1.35 pmol NaOH was added.The TG masswas determined by combining the masses of the individual FAs and dividing by 3. This method agreed within 10% of measuring the glycerol content of TG as described previously (11). The dpm in oleate was assayedby collecting the effluent from the high-performanceliquid chromatographic column in l-ml fractions, drying the solvents, and determining the radioactivity in oleate as describedpreviously (11) The TO utilized in this study was 99% pure TO from Sigma. The dpm were calculated by using the external radioactive sourcesprovided with the model 1500 Tri Carb, which enables conversion of counts per minute to dpm by referenceto a series of quenchedstandards. Statistical evaluation was performed using Student’s unpaired or paired t test where appropriate. Calculations as to relationshipsbetweenspecific activities of precursorsand products were madeby dividing the specific activity of the precursor by its product. For example, the specific activity of plasmaFA oleate would be divided by the specific activity of mucosalFA oleate. RESULTS
To establish that the plasma and intestine
were under
steady-state conditions during intravenous [3H] oleate in-
fusion, preliminary experiments were conducted in which samples of blood were obtained from the carotid artery and mucosa from the intestine at 5, 6, and 7 h of intraduodenal TO infusion. The data (Table 1) show that mucosal TG mass and TG oleate specific activity were nearly equivalent between 5 and 7 h of infusion, confirming previous work using an intraduodenal [YJ]TO infusion of 90 pmol/h (12). The mucosal FA mass and FA oleate specific activity were also at steady-state levels. Finally, plasma FA oleate specific activity was nearly the same between 5 and 7 h of infusion. We have previously
shown that when [3H]T0 was infused intraduodenally at 135 pmol/h (8) a steady state also existed with regard to chylomicron transport. In sum the data presented in Table 1 and the previous data (12) indicate that the mucosa was under steady-state conditions as was the plasma FA oleate specific activity during the course of the experiments. After preliminary experiments had established that radioactive and mass steady states existed for the pertinent
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 10, 2019.
UPTAKE
AND METABOLISM
OF CIRCULATING
FATTY
G929
ACIDS
Table 1. Mucosa and plasma lipids are at steady state at 6 h of infusion Time, h Sample 5
Mucosa Triacylglycerol, pmol 154k12 Triacylglycerol, dpm/pmol* 3,353+324 FFA, pm01 40t6 FFA, dpm/pmol* 24,080+2,667 Plasma FFA, dpm/pmol* 144,772+ 14,747 Data are means =t SE. Each hour represents a set of 4 rats. * Data refer specifically acid.
lipids after 6 h of TO infusion intraduodenally and [3H]oleate intravenously, further experiments were conducted after 6 h of infusion (intraduodenal and intravenous). During the 6 h intraduodenal TO infusion, a total of 810 pmol TO was delivered to the gut, but only 123 t 9 pmol TG, or 15% of the TG infused, was present in the mucosa at the conclusion of the 6-h period. This supports previous data suggesting rapid clearance of TG from the mucosa under these conditions (9). At 6 h of infusion there was 42 & 4 pm01 total FA in the mucosa indicating that most of the absorbed acyl groups were present as TG because only 10% of the total acyl groups (FA + TG-FA) were as FA. Between 5 and 6 h of infusion, lymph flow was 4.8 & 0.6 ml. At the conclusion of the 6 h infusion, 732,250 t 41,505 dpm were present in the mucosal homogenate. When the mucosa was extracted (2), 99 t 0.2% of the dpm were in the organic phase with the remainder in the aqueous phase. This is in contrast to prior observations (3, 4) in which 42% of dpm were found in the aqueous phase. The difference between prior observations and our own is likely due to the bolus injection technique with observations at 2 min used by Gang1 et al. (3, 4) compared with the prolonged infusion method utilized in the present report. The turnover rate of mucosal lipids is expected to be much slower than its water-soluble metabolic products leading to a progressive increase in lipid dpm over time compared with its metabolites. We also found that chemiluminescence occurred (-10% of the total dpm) unless the aqueous phase was completely dried before addition of scintillant . As expected, the intravenously infused E3H]oleate was widely distributed into various lipid classes indicating considerable metabolism of the probe FA (Fig. 1). As shown, the predominant radiolabeled lipid in the carotid artery was TG followed by FA and then cholesteryl ester. Little dpm were distributed to the other lipids analyzed. In the mucosa, the results were quite different; TG comprised 86% of the total dpm present. These data indicate that plasma lipids taken up by the mucosa were extensively metabolized. Cholesteryl ester was not likely taken up by the intestine because it was significantly radiolabeled in the plasma but hardly at all in the mucosa (Fig. 1). The most likely plasma lipid to be taken up by the enterocyte is FA based on its specific activity in the mucosa compared with mucosal TG (see below). The specific activity of FA oleate in the plasma was
7
6
133*13 2,818*290 38t9 25,158f2,136
147&X 2,838*285 31k6 25,428&3,198
139,804&12,782 to oleate
152,459*15,218
or the oleoyl
CAROTID
PLASMA
moiety
specific
ARTERY
OR MUCOSAL
activity.
FFA, free fatty
MUCOSA
LIPID
CLASSES
Fig. 1. Percentage distribution of 3H into various lipids in carotid artery and intestinal mucosa after intravenous administration of [3H]oleak and intraduodenal infusion of glyceryl trioleate (135 pmol/h). Origin, phospholipids; MG, monoacylglycerol; DG, diacylglycerol; FA, fatty acid; TG, triacylglycerol; CE, cholesteryl esters. Data are means 2 SE of 5 experiments.
very high, 139,804 t 12,782 dpm/pmol. The specific activity of mucosal FA oleate at 6 h (25,158 dpm/pmol) was only 18% of that of plasma FA oleate (139,804 dpm/pmol) secondary to the large amount of absorbed nonradioactive oleate from the duodenally infused TO. Of particular interest are the data shown in Fig. 2, which show that the specific activity in mucosal TG oleate is further reduced ninefold (25,158/2,818, Table 1) compared with the specific activity of mucosal FA oleate, indicating that not all the FA oleate in the mucosa is equally utilized for TG synthesis. In this event, the specific activity of the FA oleate and TG oleate would have been equal. One way in which this would occur is the known conservation of monoacylglycerol (MG) for mucosal TG synthesis (12, l3), which in the present experiments would likely be sn-2-MG-oleate. This would mean that one of every three acyl groups of TG could not come from plasma FA, which could maximally account for a 1.5fold reduction (3 acyl groups of which a maximum of 2 could come from plasma) in specific activity. Because the observed specific activity reduction is much greater, the TG oleate in the mucosa must be preferentially derived from other exogenous (unlabeled, luminal) sources. Also shown in Fig. 2 is the specific activity of TG oleate in lymph chylomicrons and VLDL fractions. As indicated in Fig. 2, the specific activity of the mucosal TG oleate
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 10, 2019.
G930
UPTAKE 4000
1
AND METABOLISM
OF CIRCULATING
P*0.0002 I P
