- Email: [email protected]
Further studies on the mechanism of inhibition of intestinal chylomicron transport by Pluronic L-81
Biochimica et Biophysica Acta, 1004 (1989) 357-362 Elsevier
357
BBALIP 53180
Further studies on the mechanism of inhibition of intestinal chylomicron transport by Pluronic Lo.81 D a v i d N u t t i n g 1, J e n n i f e r H a l l 2, J a m e s A. B a r r o w m a n 2 a n d P a t r i c k T s o 3 t Department of Physiology, University of Tennessee Memphis, TN (U.S.A.), 2 Department of Medicine, Memorial University of Newfoundland St. John's (Canada) and "~Department of Physiology, Louisiana State University Shreveport, LA (U.S.A.) (Received 21 February 1989)
Key words: Intestinal lipoprotein; Golgi apparatus; Chyiomicron formation; Endoplasmic reticulum, VLDL
This study explored further the hypothesis that intestinal cells have two pathways for producing large ~iaeylglycerol-rich lipoprotein particles. The hydrophobic surfaetant Piuronic L-8| (L-81) inhibits formation of chylomicrons (containing triacylglyceroi synthesized from dietary fatty acids and monoacyiglycerol, through the monoacylglycerol pathway), but not formation of very.low.density iipoproteins. L-81 does not inhibit lymphatic lipid transport during infusion o[ egg phosphatidylcholine, whose fat~ acid is processed through the a-glycerol phosphate pathway and is transposed in lymph in very-low-density iipoproteins. Thus, the first part of this study tested whether L-81 cannot inhibit the a-glycerol phosphate pathway, and thus L-81 can only affect chylomicron lipid secretion. Intestinal lymph fistula rats were infused with a lipid emulsion containing [l-t4C|oleic acid, but no monoacylglycerol, to ensure that the o|eic acid will be channeled to the a-glycerol phosphate pathway. Experimental rats received 1 m g / h of L-81 in their emulsion whereas control rats lacked L-81. Lymphatic triacylglycerol output, measured both chemically and radioactively, was markedly suppressed in the experimental rats as compared to the. controls. Thus, these data indicate that the reason why lipid transport was unaffected by L-81 when egg phosphatidylcholine was infused was not because of the pathway used for the resynthesis of triacylglycerol from phosphafidylchollne. In the second part of this study, we measured the appearance time for chylomicron (in control rats) and for very-low-density lipoprotein (in L-81-treated rats). The appearance time is defined as the time between placement of radioactive fatty acid into the intestinal lumen and ~he appearance of radioactive lipid in the central lacteal. The average appearance time for the control rats was 10.8 rain, which was sisnificantly shorter than the 16.2 rain in the L-81-treated experimental rats. This difference in appearance time further supports the hypothesis that chyiomicron and very-low-density lipoprotein are packaged separately in the enteroeytes and only the formation of chylomicron is inhibited by L-81.
Introduction
Earlier studies by Mahley et al. [1] and Ockner et al. [2] suggested that in the small intestine, chylomicrons and very-low-density lipoproteins (VLDLs) are packaged and secreted separately. Further support for this concept came from Tso et al. [3] using a much different approach. The non-ionic hydrophobic surfactant, Pluronic L-81 (L-81), is a potent inhibitor of intestinal formation and transport of chylomicrons. The digestion, uptake and the re-esterification of the absorbed monoacylglycerol and fatty acid to form triacylglycerol are not affected by the presence of L-81 in the triolein test meal. Interestingly, L-81 does not inhibit
Correspondence: P. Tso, Department of Physiology, L.S.U. Medical Center, 1501 Kings Highway, Shreveport, LA 71130, U.S.A.
the formation and transport of VLDL-sized particles by the small intestine [3]. This conclusion was based on experiments in which we infused egg phosphatidylcholine, with or without L-81 added, intraduodenally in lymph fistula rats and studied lymphatic lipid and lipoprotein output. The lipoproteins transported in the lymph of both groups of animals were predominantly VLDL, and L-81 had no effect on lymphatic lipid output. Thus we concluded that the infusion of egg phosphatidylcholine results mainly in the formation of VLDL, and for this reason, L-81 failed to inhibit intestinal lipid transport in these rats. In the present study, we examined in greater detail the hypothesis that chylomicron and VLDL are assembled separately in the intestine, using two approaches: (1) we tested whether L-81 can inhibit the lymphatic transport of triacyiglycerol derived from the a-glycerol phosphate pathway. To achieve this goal, fatty acid alone was fed
0005-2760/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)
358 together with L-81. The reason for feeding fatty acid only was to force the fatty acid to form triacylglycerol via the a-glycerol phosphate pathway; (2) we also compared the time needed for synthesis and secretion of chyiomieron (in control animals) with that of VLDL (in L-81-treated animals), hypothesizing that if the two lipoprotein particles are packaged separately, the production times may be different. Phosphatidylcholine is digested in the intestine to form lysophosphatidylcholine and free fatty acid before being absorbed [4-6]. The absorbed lysophosphatidylcholine can be used to re-form phosphatidylcholine or is further hydrolyzed to form fatty acid and glycero-3phosphocholine [5,7,8]. Glycero-3-phosphocholine is water-soluble and thus is transported in the portal blood. Due to the lack of 2-monoacylglycerol, the majority of the fatty acid is converted to triacylglycerol in the enterocytes by the glycerol phosphate pathway. In contrast, the digestion products of triacylglycerol, 2-monoacylglycerol and fatty acid, will be absorbed and converted rapidly to form triacylglyceroi by the monoacylglycerol pathway [9,10]. Thus, we hypothesized that the differential inhibitory action by L-81 on intestinal absorption and transport of triacylglycerol and phosphatidylcholine by the rat small intestine could be a result of the pathway utilized to handle the absorbed lipid. To test this hypothesis, we infused oleic acid, with or without I.-81 added, and determined whether L-81 inhibits the lymphatic transport of the triacylglycerol formed by the glycerol phosphate pathway. By infusing only oleic acid, without 2-monoacylglycerol, we ensured that most of the absorbed fatty acid would be processed by the a-glycerol phosphate pathway [11]. Materials and Methods
Animals Male Sprague-Dawley rats (300-350 g) were used for all experiments. The animals were fasted overnight before surgery. Under ether anesthesia, the intestinal lymph duct was calculated with clear vinyl tubing (0.8 mm outside diameter), according to the method described by Bollman et al. [12]. The only modification to the original surgical procedure was that the cannula was secured through the application of a drop of cyanoacrylate 81ue (Krazy Glue inc., Itasca, IL), instead of suture. A silicone tube (1.6 mm outside diameter) tipped inside with a clear vinyl tubing (1 nun outside diameter), was introduced about 2 cm down the duodenum through the fundus of the stomach. The tube was secured in the duodenum through a transmural suture, and the fundal incision was closed by a pursestring suture. Postoperatively, the animals were infused with 3 ml/h of a saline solution (145 mM NaCI) containing KCI (4 mM) and glucose (0.28 M). The animals were allowed to recover for at least 36 h in
restraining cages kept in a warre chamber (approx. 30°C).
Experimental plan To address the two questions raised, two studies were conducted, study A and study B. Study A (Fatty acid metabolic pathways). The control rats were infused for 8 h intraduodenally at 3 ml/h with a lipid test meal containing 40 mM of oleic acid (labeled with [14C]oleic acid, 2 nCi/p~mol oleic acid) and 19 mM sodium taurocholate in phosphate-buffered saline (pH 6.4). The procedure used in the preparation of the lipid test meal has been described in detail previously [13]. The experimental rats received the same lipid test meal, but with I mg L-81/3 ml per h added. Study B (Appearance times). On the day of the experiment, the infusion of glucose-saline in the control rats was replaced for 2 h with 1 $/dl sodium taurocholate in (phosphate-buffered saline). The rats then received 0.5 ml of an emulsion containin 8 13.33 t~mol of oleic acid (labeled with [1-1'C]oleic acid), 6.67/stool 1-monoolein and 9.33 pmol sodium taurocholate in phosphatebuffered saline. The experimental rats were infused intraduodenally with phosphate-buffered saline containing I 8/dl sodium taurocholate and 33 mg/dl L-81 for 2 h before receiving 0.5 ml of a similar radioactive lipid test meal as the control rats, but with 165 pg of L-81 added. Experimental procedure. In study A, lymph was collected between 0 and 2 h, 2 and 4 h, and each subsequent h during the 8 h infusion. Lymph lipid was extracted by the method of Folch et al. [14] and then dissolved in chloroform. Aliquots were taken for radioactivity and triacylglycerol [15] determinations. At the end of the lipid infusion, the animal was anesthetized and killed by exsanguination. The stomach and the colon were excised separately with care to prevent leakage of luminal contents and put into stoppered Erlenmeyer flasks. The samples were then saponified, acidified, and extracted with petroleum ether [16]. A sample of the petroleum ether phase was taken for quantitation of t4C-labeled lipid content by scintillation spectrometry. The small intestine was divided into four equal lensth segments, and the contents of each were rinsed out with two (5 ml) rinses of 10 mM sodium taurocholate. A.!iquots were taken from these washings for determination of radioactivity in a water-soluble scintillant, Aquasolv (Beckman. Fullerton, CA, U.S.A.). Mucosa from each of the four intestinal sesments was scraped off with glass slide, and lipid was extracted by the method of Folch et al. [14]. Aliquots were taken for radioactivity determination, triacylglycerol deteimination [15] and for separation into lipid classes by thinlayer chromatography using the solvent system light petroleum ether/diethyl ether/acetic acid (?5 : 15 : 0.6,
v/v).
359 In study B, lymph was collected every 2 min for the first 60 min. Lymph flow in microliters per minute was calculated from the increased weight of the vials, assuming a density of 1.0 g / m l for lymph. Then scintillant was added to the vials and the radioactivity was measured. The chylomicron or VLDL appearance time was defined as the time taken for radioactive fatty acid infused into the duodenum to the time when it appears in lymph, after taking into account the time lymph spent in the dead space of the ~ m u l a . Materials. [1-t4C]Oleic acid was purchased from Amersham (Arlington Heights, IL, U.S.A.) and was used without further purification. Oleic acid was purchased from Sigma Chemical Co. (St. Louis, MO, U.S.A.) and was found to be more than 9970 pure. Both the egg phosphatidylcholine and sodium taurocholate were purchased from Sigma and were used without further purification. Pluronic surfactant was kindly donated by Dr. Schmolke of BASF Wyandotte, Wyandotte, MI (U.S.A.). All reagents and solvents used were of analytical grade. Statistics Data are presented as means + S.E. Where applicable, the significance of differences between results from various groups of animals were tested by Student's t-test for independent variables |17]; an a-level of 0.05 was considered significant.
Results
.dy A Lymph flow. Five control 0rod five experimental animals were used for this study. The preinfusion lymph flow was similar in both groups of animals ( m e a n 2.2 + 0.1 ml/h, n - 10). Lymph flow increased significantly in both groups of animals and reached a steady output of 4.2 ± 0.2 m l / h ( n - - 1 0 ) drams the 7th and 8th of lipid infusion. This lymph flow rate is comparable to those ob~rved triolein infusion [13,18,19|. Interestingly, I.,-81 ted not inhibit the lipid-induced increase in lymph flow, as the same dose of L-81 given to rats receiving triolein did [13,18,19]. Triacylglycerol output. Triacylglycerol output in lymph was measured both chemically and by the radioactive lipid present in lymph. Fig, 1 shows the triacylglycerol output as determined chemically. The fasting triacylglycerol output in lymph was 5.1-t-0.3 ttmol/h in the control animals and 5.6 d: 0.4 pmol/h in the experimental a n i m a l s , which were not different. Triacylglycerol output increased in the control animals to reach about 30 p m o l / h between hours 2-4 of lipid infusion. This remained relatively unchanged during the rest of the lipid infusion period (29-33 pmol/h). Because the fasting lymph triacylglycerol output was measured in each rat before giving the test meal, the effect of the lipid
¢~-......, oenltut •e
40.
o .m, "1
30.
,.!
o Q _~
20-
o o
B
-?
10
.....
I-*
|
| .r,L
Fasting0
i i i Hours Ot Inlusion
Fig,. 1. Lymphatic triac,jlglycerot output as determined ~ 1151 in ~ | per h. Five animals ' ~ u.~l in m ~ group and the output values are given as means ±,~.E.
load may be expressed as the excess over the fasting value. Thus, the increase in triacylglycerol output as a result of lipid infusion was about 28 F m o l / h during the 7th :rod 8th h. in contrast, the triacylglyce~l output in lymph of L-81-treated rats increased only slightly from a fasting output of 5.6 p m o l / h to 11.4-12.8 p m o l / h during the 7th and the 8th h. This was significantly lower than in controls ( P < 0.001). Thus, L-81 significandy inhibited intestinal lymphatic lipid transport. Fig. 2 s u m r n m the radioactive lipid output in lymph. Since most of the radioactive lipid in lymph is triacylglycerol, it is reasonable to use lymphatic radioactive fipids as a measure of lymphatic triacylglycerol output. Calculated in this way, the lymph triacylglycerol output during the 7th and 8th h was between 24 and 26 pmol/h, which agreed well with t,Ne chemical measurements. In the L-81-treated animals, the radioactivity and chemical data again agreed, and showed a marked inhibition by L-81 of lipid output from the oleic acid
"6 E 40.
_-
-_ C m d m l
=t
0 m
2O-
10.
O
i Hours of Inhudon
Fi& 2. L~mphatic triacyiglycero~ output as measured by radioactivily dcte~aination in ,~moi per 11.Since rnc~ of the radioactive fatty acid is preseat in lymph as tfia~lglycego~ it is reasonable to assume the lymphatic radioactivity output to be equal to lymphatic: radio~tive triacylglyo:xol outpux. Five animals wczc used in each Ipmsp and the output v a l m are expressed as means + S.E.
360 TABLE !
TABLE II
Recooery of [14C]oleic acid from the intestinal lumen
Recovery of [ t 4C]oleic acid in the intestinal mucosa of the four intestinal segments
Values are expressed mean+S.E, of the radioactive oleic acid recovered in each quarter of the small intestinal lumen after 8 h of lipid infusion. The radioactivity is expressed as % of total radioactive oleic acid administered. The stomach and the cecal lumin'd washings contained negligible radioactivity.
Control
Segment I
Segment 2
Segment 3
Segment 4
2.304-0.75
1.104-0.12
0.234-0.15
~.14+0.03
3,604.0,74
6.33:t: 1.4o 2,01:1:1.43 0,15±0.02
The recovery of mucosal radioactive lipid is expressed as % of total dose of radioactive lipid infused. Values are exposed as mean + S.E. The animals were killed at the end of 8 h of lipid infusion and the smaF intestine was divided into four equal quarters. The intestinal mucos~ was harvested by scraping with a glass slide and the lipid was extracted by the method of Folch et al. [14]. Segment 1
(N= 5) Experimental (N.- 5)
Control
2.95 4. 0.42
Segment 2 2.59 :!:0.42
Segment 3
Segment 4
0.07 ± 0.01 0.04 + 0.01
(N= 5)
,,
Experimental
7.26 ~ 1.29 21.12± 2.33
0.90 :!:0.33
0.08 :!:0.01
(N~ 5)
load (7.5 ± 0.6 itmol/h during hours 7 and 8, 19% of the hourly infused dose). Distribution of [14C]olelc acid in the lumen. Table ! shows the recovery of radiolabeled oleic acid in the lumen of both groups of animals. A negligible amount of radioactive oleic acid was recovered in the stomach and the cecal washings and thus these results are not included in Table I. Most of the unabsorbed radioactive lipid was found in the first and second quarcer of the small intestine in both groups of animals. Of the total radioactive oleic acid infused, only 3-4% remained in the lumen of the control animals. In contrast, significantly more of the refused [t4C]oleic acid remained in the small intestinal lumen of the L-81/treated animals (P < 0.001). The increased amount of radioactive oleic acid remaining in the lumen of the L-~l/treated rats, as compared to the controls, probably resulted directly from the impaired lymphatic lipid transport in these animals, Recooery of t~C in the intestinal mucos~ Table 11 summarizes the recovery of labeled oleic acid in the four quarters of the small intestinal mucosa of both groups of rats, Moot of the radioactive lipid was recc~vered in the first and second quarters of the small intestine of both groups of rats. However, there was significantly more radioactive lipid in the intestinal mucosa of segments 1 and 2 in the experimental rats as compared :o the controls, and it is significant for both comparisions
(P <0.001). Table III summarizes the distrt~ttion of radioactive oleic acid among the various lipid classes in the intestinal mucosa of segment 1 in both the controls and the experimental rats. There was no dif,~re:~ce in the distribution of radioactive oleic acid between segments 1 and 2 in either group of rats, and thus only the data on segment 1 are presented. In the controls, mo~t of the radioactive oleic ~.cid was recovered in triacylglycerol (47% of segment I 14C radioactivity), the fatty acid (26%) and the monoacylglycerol plus phespholipid fractions (16%). In the experimental rats, there was significantly more radioactive oleic acid in the triacylglycerol fraction as compared to controls (P < 0.001). This increase in recovery in the triacylglycerol fraction was accompanied by a relative decrease in the fatty acid and the monoacylglycerol + phospholipid fractions. However, when the distribution of label was expressed as percentage of the dose of 14C infused over 8 h, it is clear that only the triacylglycerol changed significantly after L-81 (see values in parentheses in Table III).
Study B In this study, we measured the chylomicron and the VLDL appearance time in the control and the L-81treated rats, respectively. Lymph flow was comparable in both groups. The appearance time was 10.82 + 0.99
TABLE !!1
Distribution of I,C radionctivity among lipid classes in the mucosa of intestinal segment ! Percent distribution of segment 1 mucosal 14C; values are expressed means+ S.E. Five animals were studied in each group. CE, cholesteryl ester; 1"(3, triacylgly,~erol; FA, fatty acid; DG, diacylglyceml; MG, monoacylglycerol; PL, phospholipid. Numbers in parentheses are the percentage of the to~al dose of [t4C]oleic acid fed found in these fractions. Treatment
CE
TG
FA
DG
MG + PL
Control
4.7 + 1.4 (0.14)
47.3 ~ 3.2 (1.39)
25.6 + 0.9 (0.76)
6.7 + 0.8 (0.20)
! 5.7 + 2.4 (0.47)
Experimental
2.8 + 0.6 (0.21)
73.8 + 4.2 (5.36)
11.3 + 2.6 (0.82)
6.3 + 0.9 (0.45)
5.9 + 1.1 (0.42)
361 rain (mean + S.E.) and 16.60 + 0.91 rain for the control and the L-81-treated rats, respectively. The appearance time was significantly longer ( P < 0 . 0 1 ) for VLDLs (L-81-treated) as compared to chylomicrons (control). Discmsion It is generally accepted that there is very little monoacylglycerol stored in the enterocytes. Consequently, when oleic acid is infused without the accompanying monoacylglycerol, most of the absorbed fatty acid is processed by the a-glycerol phosphate pathway. Both the lymphatic triacylglycerol and the radioactivity output data clearly demonstrated ~ at L-81 significantly inhibited the intestinal lymphatic 'ipid transport when oleic acid is infused alone. These Lata would imply that the reason why the lymphatic lipid transport is not inhibited when egg phosphatidylcholine is infused is not because phospholipid-defived fatty acids are metabolized through the a-glycerol phosphate pathway to produce triacylglycerol. Rather, it is probably related to the fact that the fatty acid derived from egg phosphatidylcholine is mainly channeled to the formation of VLDL. Morphologic data seem to indicate that L-81 inhibits the traffic between the endoplasmic reticulum and the Golgi apparatus [20]. It is possible that the VLDLs produced during L-81 treatment do not require further processing in the Golgi apparatus prior to secretion. We have as yet no morphological data concerning whether the VLDL formed during the absorption of egg phosphatidylcholine is first processed in the Golgi apparatus prior to its release by the enterocytes. At least in the rainbow trout, there is morphological evidence that triacylglycerol-rich lipoproteins can be transported directly from the endoplasmic reticulum to the intercellular space [21]. We did not perform any electron microscopy in this study. However based on previous morphological studies from rats infused intraduodenally with triolein plus L-81, there probably were large lipid droplets accumulated in the cytoplasm of the L-81-treated rats [13,20]. When we analysed the distribution of the radioactive fatty acid in the intestinal mucosa, most of the fatty acid was in the triacylglycerol fraction in the L-81treated rats. This confirms our previous finding that L-81 does not interfere with the re-esterification of the absorbed lipid to form triacylglycerol (compare with mucosal lipid distribution in triolein-fed rats, Table II in Refs. 13 and 18). Although the majority of the labeled fatty acid was in the triacylglycerol fraction in the control animals, there was a substantial amount recovered in the free fatty acid and the monoacylglycerol + phospholipid fractions. This higher recovery of labeled fatty acid in the free fatty acid and partial acylglycerol fractions would seem to indicate that the lack of monoacylglycerol does impair the
esterification of fatty acid to form tfiacylglycerol. Consequently, the supply of 2-monoacylglycerol is important in ensuring the rapid re-esterification of fatty acid to form triacylglycerol. The fatty acids in the L-81 animals are reesterified to form triacylglycerol before being stored as large osmiophilic lipid droplets in the cytoplasm. We are currently trying to isolate the large osmiophilic lipid droplets inside the cytoplasm of the L-81-treated rats. Recent biochemical data from our laboratory have demonstrated that the large lipid droplets accumulated within the enterocytes after the intraduodenal infusion of triolein plus L-81 can be cleared rapidly from the mucosa after cessation of infusion of L-81 [22]. Although we do not know the mechanism of how the large lipid droplets are converted to chylomicron, we do know that the mucosal triacylglycerol is not hydrolyzed prior to the transport [22]. This is the opposite to what happens in the liver [23]. We have previously introduced the concept of chylomicron appearance time, wlfich is defined as the time between the placement of radioactive fatty acid in the intestinal lumen to the time when radioactive chylomicron appeared in the central lacteal [24,25]. In this experiment, we studied the chylomicron and the VLDL appearance times in the control and the L-81treated animals. The assumption is that the appearance time we measure in the control animals corresponds to the chylomicron appearance time, whereas that obtained in the L-81-treated animals corresponds to the VLDL appearance time. This is supported by the following observations. In a previous study [26] when the animals were fed fatty acid and monoacylglycerol, we found that the appearance time for radioactive lymph corresponds to the appearance time for chylomicron. That experiment involved ultracentrifugal separation of lipoprotein from the lymph and we found that the first appearance of radioactive fatty acid was associated with the chylomicron particles. In triacylglycerol-fed rats given L-81, previous morphologic studies [18] demonstrated that the majority of the particles transported in lymph were of VLDL size. In this study, the mean chylomicron appearance time for the control animals was 10.8 rain whereas in the experimental animals we obtained a significantly longer appearance time of 16.6 rain; this difference is highly significant, P < 0.01. This data would further support our previous proposal that there are separate pathways for the formation of chylomicron and the VLDL pathways [3]. However, there may be an alternative explanation for our data, which relates ~o the much smaller size of VLDL particles than chy~omicron. If we assume that the intestinal interstitial matrix (consisting mainly of the connective tissue of the lamina propria) behaves like a size exclusion chromatography column, it will take longer for the VLDL to p a s through this sieve relative to the chylomicron pa~ tides. This will be difficult to study and
362 we current have little information on the properties of the small intestinal interstitial matrix. In conclusion, these two experiments further support our proposal that chylomicron and VLDL are packaged separately by enterocytes. However, we want to emphasize that the VLDL produced during L-81 inhibition may be different from the VLDL produced normally during fasting.
Acknowledgements The authors are grateful to Deborah S. Drake, Camille Pool and Reneau Youngblood for their excellent technical help. This work was supported by National Institutes of Health Grant DK 32288. P.T. is the recipient of a Research Career Development Award DK 01575. References I Mahley, R.W., Bennett, B.D., Moore, D.J., Gray, M.E., Thistlethwaite, W. and LeQuire" V.S. (1971) Lab. Invest. 25, 435-444. 20ckner, R.K., Hughes, F.B. and Isselbacher, K.J. (1969) J. Clin. Invest. 48, 2367-2373. 3 Tso, P., Drake, D.S., Black, D. and Sabesin, S.M. (1984) Am. J. Physiol. 247, G599-G610. 4 Arnesjo, B., Nilsson, ~,., Barrowman, J.A. and Borgstrom, B. (1969) Scan& J. Gastroenterol. 4, 653-655. 5 S~w, R.O., Stein, Y. and Stein, 0. (1967) J. Biol. Chem. 242, ~*,919-4924. 6 Nilsson, ~. (1968) Biochim. Biophys. Acta (1968) 137, 240-254. 7 Subbaiah, P.V. and Gaaguly, J. (1970) Bioehem. J. i18, 233-239.
8 Parthasarathy, S., Subbaiah, P.V. and Ganguly, J. (1974) Biochem. J. 140, 503-508. 9 Brindley, D.N. and Hubscher, G. (1965) Biochim. Biophys. Acta 106, 495-509. 10 Johnston, J.M., Rao, G.A. and Lowe, P.A. (1967) Biochim. Biophys. Acta 137, 578-580. 11 Tso, P. (1985) Adv. Lipid Res. 21, 143-186. 12 Bollman, J.L., Cain, J.C., and Grindlay, J.H. (1948) J. Lab. Clin. Mud. 323, 1349-1352. 13 Tso, P., Balint, J.A. and Rodgers, J.B. (1980) Am. J. Physiol. 239, G348-G353. 14 Foleh, J., Lees, M. and Sloane-Stanley, G.H. (1957) J. Biol. Chem. 226, 497-509. 15 Biggs, H.G., Eriekson, J.J. and Morehead, W.R. (1975) Clin. Chem. 21, 437-441. 16 Rodgers, J.B., Fondacaro, J.D. and Kot, J. (1977) J. Lab. Clin. Mud. 89, 14-152. 17 Snedecor, G.W. and Cochran, W.G. (1967) Statistical Methods (6th Edition), Ames, IA; Iowa State University Press. 18 Tso, P., Balint~ J.A., Bishop, M. and Rodgers, J.B. (1981) Am. J. Physiol. 241, G487-G497. 19 Tso, P. and Gollamudi, S.R. (1984) Am. J. Physiol. 247, G32-G36. 20 Sabesin, S.M., Frase, S. and Tso. P. (1985) Gastroenterology 88, 1565 (Abstract). 21 Vernier, J.M. and Sire, M.F. (1986) Tissue Cell 18, 447-460. 22 Halpern, J., Tso, P. and Mansbach, C.M.il. (1988) J. Clin. Invest. 82, 74-81. 23 Mooney, R.A. and Lane, D.M. (198i) J. Biol. Chem. 256, 11724-11733. 24 Tso, P., Pitts, V. and Granger, D.N. (1985) Am. J. Physiol. 249, G21-(328. 25 Tso, P., Barrowman, J.A. and Granger, D.N. (1986) Am. J. Physiol. 250, G497-GS00. 26 Hall, J.L. (1987) Intestinal absorption: Studies of absorption of certain natural lipids and lipophilic xenobiotics. Master of Science Thesis, Memorial University of Newfou,dland.
357
BBALIP 53180
Further studies on the mechanism of inhibition of intestinal chylomicron transport by Pluronic Lo.81 D a v i d N u t t i n g 1, J e n n i f e r H a l l 2, J a m e s A. B a r r o w m a n 2 a n d P a t r i c k T s o 3 t Department of Physiology, University of Tennessee Memphis, TN (U.S.A.), 2 Department of Medicine, Memorial University of Newfoundland St. John's (Canada) and "~Department of Physiology, Louisiana State University Shreveport, LA (U.S.A.) (Received 21 February 1989)
Key words: Intestinal lipoprotein; Golgi apparatus; Chyiomicron formation; Endoplasmic reticulum, VLDL
This study explored further the hypothesis that intestinal cells have two pathways for producing large ~iaeylglycerol-rich lipoprotein particles. The hydrophobic surfaetant Piuronic L-8| (L-81) inhibits formation of chylomicrons (containing triacylglyceroi synthesized from dietary fatty acids and monoacyiglycerol, through the monoacylglycerol pathway), but not formation of very.low.density iipoproteins. L-81 does not inhibit lymphatic lipid transport during infusion o[ egg phosphatidylcholine, whose fat~ acid is processed through the a-glycerol phosphate pathway and is transposed in lymph in very-low-density iipoproteins. Thus, the first part of this study tested whether L-81 cannot inhibit the a-glycerol phosphate pathway, and thus L-81 can only affect chylomicron lipid secretion. Intestinal lymph fistula rats were infused with a lipid emulsion containing [l-t4C|oleic acid, but no monoacylglycerol, to ensure that the o|eic acid will be channeled to the a-glycerol phosphate pathway. Experimental rats received 1 m g / h of L-81 in their emulsion whereas control rats lacked L-81. Lymphatic triacylglycerol output, measured both chemically and radioactively, was markedly suppressed in the experimental rats as compared to the. controls. Thus, these data indicate that the reason why lipid transport was unaffected by L-81 when egg phosphatidylcholine was infused was not because of the pathway used for the resynthesis of triacylglycerol from phosphafidylchollne. In the second part of this study, we measured the appearance time for chylomicron (in control rats) and for very-low-density lipoprotein (in L-81-treated rats). The appearance time is defined as the time between placement of radioactive fatty acid into the intestinal lumen and ~he appearance of radioactive lipid in the central lacteal. The average appearance time for the control rats was 10.8 rain, which was sisnificantly shorter than the 16.2 rain in the L-81-treated experimental rats. This difference in appearance time further supports the hypothesis that chyiomicron and very-low-density lipoprotein are packaged separately in the enteroeytes and only the formation of chylomicron is inhibited by L-81.
Introduction
Earlier studies by Mahley et al. [1] and Ockner et al. [2] suggested that in the small intestine, chylomicrons and very-low-density lipoproteins (VLDLs) are packaged and secreted separately. Further support for this concept came from Tso et al. [3] using a much different approach. The non-ionic hydrophobic surfactant, Pluronic L-81 (L-81), is a potent inhibitor of intestinal formation and transport of chylomicrons. The digestion, uptake and the re-esterification of the absorbed monoacylglycerol and fatty acid to form triacylglycerol are not affected by the presence of L-81 in the triolein test meal. Interestingly, L-81 does not inhibit
Correspondence: P. Tso, Department of Physiology, L.S.U. Medical Center, 1501 Kings Highway, Shreveport, LA 71130, U.S.A.
the formation and transport of VLDL-sized particles by the small intestine [3]. This conclusion was based on experiments in which we infused egg phosphatidylcholine, with or without L-81 added, intraduodenally in lymph fistula rats and studied lymphatic lipid and lipoprotein output. The lipoproteins transported in the lymph of both groups of animals were predominantly VLDL, and L-81 had no effect on lymphatic lipid output. Thus we concluded that the infusion of egg phosphatidylcholine results mainly in the formation of VLDL, and for this reason, L-81 failed to inhibit intestinal lipid transport in these rats. In the present study, we examined in greater detail the hypothesis that chylomicron and VLDL are assembled separately in the intestine, using two approaches: (1) we tested whether L-81 can inhibit the lymphatic transport of triacyiglycerol derived from the a-glycerol phosphate pathway. To achieve this goal, fatty acid alone was fed
0005-2760/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)
358 together with L-81. The reason for feeding fatty acid only was to force the fatty acid to form triacylglycerol via the a-glycerol phosphate pathway; (2) we also compared the time needed for synthesis and secretion of chyiomieron (in control animals) with that of VLDL (in L-81-treated animals), hypothesizing that if the two lipoprotein particles are packaged separately, the production times may be different. Phosphatidylcholine is digested in the intestine to form lysophosphatidylcholine and free fatty acid before being absorbed [4-6]. The absorbed lysophosphatidylcholine can be used to re-form phosphatidylcholine or is further hydrolyzed to form fatty acid and glycero-3phosphocholine [5,7,8]. Glycero-3-phosphocholine is water-soluble and thus is transported in the portal blood. Due to the lack of 2-monoacylglycerol, the majority of the fatty acid is converted to triacylglycerol in the enterocytes by the glycerol phosphate pathway. In contrast, the digestion products of triacylglycerol, 2-monoacylglycerol and fatty acid, will be absorbed and converted rapidly to form triacylglyceroi by the monoacylglycerol pathway [9,10]. Thus, we hypothesized that the differential inhibitory action by L-81 on intestinal absorption and transport of triacylglycerol and phosphatidylcholine by the rat small intestine could be a result of the pathway utilized to handle the absorbed lipid. To test this hypothesis, we infused oleic acid, with or without I.-81 added, and determined whether L-81 inhibits the lymphatic transport of the triacylglycerol formed by the glycerol phosphate pathway. By infusing only oleic acid, without 2-monoacylglycerol, we ensured that most of the absorbed fatty acid would be processed by the a-glycerol phosphate pathway [11]. Materials and Methods
Animals Male Sprague-Dawley rats (300-350 g) were used for all experiments. The animals were fasted overnight before surgery. Under ether anesthesia, the intestinal lymph duct was calculated with clear vinyl tubing (0.8 mm outside diameter), according to the method described by Bollman et al. [12]. The only modification to the original surgical procedure was that the cannula was secured through the application of a drop of cyanoacrylate 81ue (Krazy Glue inc., Itasca, IL), instead of suture. A silicone tube (1.6 mm outside diameter) tipped inside with a clear vinyl tubing (1 nun outside diameter), was introduced about 2 cm down the duodenum through the fundus of the stomach. The tube was secured in the duodenum through a transmural suture, and the fundal incision was closed by a pursestring suture. Postoperatively, the animals were infused with 3 ml/h of a saline solution (145 mM NaCI) containing KCI (4 mM) and glucose (0.28 M). The animals were allowed to recover for at least 36 h in
restraining cages kept in a warre chamber (approx. 30°C).
Experimental plan To address the two questions raised, two studies were conducted, study A and study B. Study A (Fatty acid metabolic pathways). The control rats were infused for 8 h intraduodenally at 3 ml/h with a lipid test meal containing 40 mM of oleic acid (labeled with [14C]oleic acid, 2 nCi/p~mol oleic acid) and 19 mM sodium taurocholate in phosphate-buffered saline (pH 6.4). The procedure used in the preparation of the lipid test meal has been described in detail previously [13]. The experimental rats received the same lipid test meal, but with I mg L-81/3 ml per h added. Study B (Appearance times). On the day of the experiment, the infusion of glucose-saline in the control rats was replaced for 2 h with 1 $/dl sodium taurocholate in (phosphate-buffered saline). The rats then received 0.5 ml of an emulsion containin 8 13.33 t~mol of oleic acid (labeled with [1-1'C]oleic acid), 6.67/stool 1-monoolein and 9.33 pmol sodium taurocholate in phosphatebuffered saline. The experimental rats were infused intraduodenally with phosphate-buffered saline containing I 8/dl sodium taurocholate and 33 mg/dl L-81 for 2 h before receiving 0.5 ml of a similar radioactive lipid test meal as the control rats, but with 165 pg of L-81 added. Experimental procedure. In study A, lymph was collected between 0 and 2 h, 2 and 4 h, and each subsequent h during the 8 h infusion. Lymph lipid was extracted by the method of Folch et al. [14] and then dissolved in chloroform. Aliquots were taken for radioactivity and triacylglycerol [15] determinations. At the end of the lipid infusion, the animal was anesthetized and killed by exsanguination. The stomach and the colon were excised separately with care to prevent leakage of luminal contents and put into stoppered Erlenmeyer flasks. The samples were then saponified, acidified, and extracted with petroleum ether [16]. A sample of the petroleum ether phase was taken for quantitation of t4C-labeled lipid content by scintillation spectrometry. The small intestine was divided into four equal lensth segments, and the contents of each were rinsed out with two (5 ml) rinses of 10 mM sodium taurocholate. A.!iquots were taken from these washings for determination of radioactivity in a water-soluble scintillant, Aquasolv (Beckman. Fullerton, CA, U.S.A.). Mucosa from each of the four intestinal sesments was scraped off with glass slide, and lipid was extracted by the method of Folch et al. [14]. Aliquots were taken for radioactivity determination, triacylglycerol deteimination [15] and for separation into lipid classes by thinlayer chromatography using the solvent system light petroleum ether/diethyl ether/acetic acid (?5 : 15 : 0.6,
v/v).
359 In study B, lymph was collected every 2 min for the first 60 min. Lymph flow in microliters per minute was calculated from the increased weight of the vials, assuming a density of 1.0 g / m l for lymph. Then scintillant was added to the vials and the radioactivity was measured. The chylomicron or VLDL appearance time was defined as the time taken for radioactive fatty acid infused into the duodenum to the time when it appears in lymph, after taking into account the time lymph spent in the dead space of the ~ m u l a . Materials. [1-t4C]Oleic acid was purchased from Amersham (Arlington Heights, IL, U.S.A.) and was used without further purification. Oleic acid was purchased from Sigma Chemical Co. (St. Louis, MO, U.S.A.) and was found to be more than 9970 pure. Both the egg phosphatidylcholine and sodium taurocholate were purchased from Sigma and were used without further purification. Pluronic surfactant was kindly donated by Dr. Schmolke of BASF Wyandotte, Wyandotte, MI (U.S.A.). All reagents and solvents used were of analytical grade. Statistics Data are presented as means + S.E. Where applicable, the significance of differences between results from various groups of animals were tested by Student's t-test for independent variables |17]; an a-level of 0.05 was considered significant.
Results
.dy A Lymph flow. Five control 0rod five experimental animals were used for this study. The preinfusion lymph flow was similar in both groups of animals ( m e a n 2.2 + 0.1 ml/h, n - 10). Lymph flow increased significantly in both groups of animals and reached a steady output of 4.2 ± 0.2 m l / h ( n - - 1 0 ) drams the 7th and 8th of lipid infusion. This lymph flow rate is comparable to those ob~rved triolein infusion [13,18,19|. Interestingly, I.,-81 ted not inhibit the lipid-induced increase in lymph flow, as the same dose of L-81 given to rats receiving triolein did [13,18,19]. Triacylglycerol output. Triacylglycerol output in lymph was measured both chemically and by the radioactive lipid present in lymph. Fig, 1 shows the triacylglycerol output as determined chemically. The fasting triacylglycerol output in lymph was 5.1-t-0.3 ttmol/h in the control animals and 5.6 d: 0.4 pmol/h in the experimental a n i m a l s , which were not different. Triacylglycerol output increased in the control animals to reach about 30 p m o l / h between hours 2-4 of lipid infusion. This remained relatively unchanged during the rest of the lipid infusion period (29-33 pmol/h). Because the fasting lymph triacylglycerol output was measured in each rat before giving the test meal, the effect of the lipid
¢~-......, oenltut •e
40.
o .m, "1
30.
,.!
o Q _~
20-
o o
B
-?
10
.....
I-*
|
| .r,L
Fasting0
i i i Hours Ot Inlusion
Fig,. 1. Lymphatic triac,jlglycerot output as determined ~ 1151 in ~ | per h. Five animals ' ~ u.~l in m ~ group and the output values are given as means ±,~.E.
load may be expressed as the excess over the fasting value. Thus, the increase in triacylglycerol output as a result of lipid infusion was about 28 F m o l / h during the 7th :rod 8th h. in contrast, the triacylglyce~l output in lymph of L-81-treated rats increased only slightly from a fasting output of 5.6 p m o l / h to 11.4-12.8 p m o l / h during the 7th and the 8th h. This was significantly lower than in controls ( P < 0.001). Thus, L-81 significandy inhibited intestinal lymphatic lipid transport. Fig. 2 s u m r n m the radioactive lipid output in lymph. Since most of the radioactive lipid in lymph is triacylglycerol, it is reasonable to use lymphatic radioactive fipids as a measure of lymphatic triacylglycerol output. Calculated in this way, the lymph triacylglycerol output during the 7th and 8th h was between 24 and 26 pmol/h, which agreed well with t,Ne chemical measurements. In the L-81-treated animals, the radioactivity and chemical data again agreed, and showed a marked inhibition by L-81 of lipid output from the oleic acid
"6 E 40.
_-
-_ C m d m l
=t
0 m
2O-
10.
O
i Hours of Inhudon
Fi& 2. L~mphatic triacyiglycero~ output as measured by radioactivily dcte~aination in ,~moi per 11.Since rnc~ of the radioactive fatty acid is preseat in lymph as tfia~lglycego~ it is reasonable to assume the lymphatic radioactivity output to be equal to lymphatic: radio~tive triacylglyo:xol outpux. Five animals wczc used in each Ipmsp and the output v a l m are expressed as means + S.E.
360 TABLE !
TABLE II
Recooery of [14C]oleic acid from the intestinal lumen
Recovery of [ t 4C]oleic acid in the intestinal mucosa of the four intestinal segments
Values are expressed mean+S.E, of the radioactive oleic acid recovered in each quarter of the small intestinal lumen after 8 h of lipid infusion. The radioactivity is expressed as % of total radioactive oleic acid administered. The stomach and the cecal lumin'd washings contained negligible radioactivity.
Control
Segment I
Segment 2
Segment 3
Segment 4
2.304-0.75
1.104-0.12
0.234-0.15
~.14+0.03
3,604.0,74
6.33:t: 1.4o 2,01:1:1.43 0,15±0.02
The recovery of mucosal radioactive lipid is expressed as % of total dose of radioactive lipid infused. Values are exposed as mean + S.E. The animals were killed at the end of 8 h of lipid infusion and the smaF intestine was divided into four equal quarters. The intestinal mucos~ was harvested by scraping with a glass slide and the lipid was extracted by the method of Folch et al. [14]. Segment 1
(N= 5) Experimental (N.- 5)
Control
2.95 4. 0.42
Segment 2 2.59 :!:0.42
Segment 3
Segment 4
0.07 ± 0.01 0.04 + 0.01
(N= 5)
,,
Experimental
7.26 ~ 1.29 21.12± 2.33
0.90 :!:0.33
0.08 :!:0.01
(N~ 5)
load (7.5 ± 0.6 itmol/h during hours 7 and 8, 19% of the hourly infused dose). Distribution of [14C]olelc acid in the lumen. Table ! shows the recovery of radiolabeled oleic acid in the lumen of both groups of animals. A negligible amount of radioactive oleic acid was recovered in the stomach and the cecal washings and thus these results are not included in Table I. Most of the unabsorbed radioactive lipid was found in the first and second quarcer of the small intestine in both groups of animals. Of the total radioactive oleic acid infused, only 3-4% remained in the lumen of the control animals. In contrast, significantly more of the refused [t4C]oleic acid remained in the small intestinal lumen of the L-81/treated animals (P < 0.001). The increased amount of radioactive oleic acid remaining in the lumen of the L-~l/treated rats, as compared to the controls, probably resulted directly from the impaired lymphatic lipid transport in these animals, Recooery of t~C in the intestinal mucos~ Table 11 summarizes the recovery of labeled oleic acid in the four quarters of the small intestinal mucosa of both groups of rats, Moot of the radioactive lipid was recc~vered in the first and second quarters of the small intestine of both groups of rats. However, there was significantly more radioactive lipid in the intestinal mucosa of segments 1 and 2 in the experimental rats as compared :o the controls, and it is significant for both comparisions
(P <0.001). Table III summarizes the distrt~ttion of radioactive oleic acid among the various lipid classes in the intestinal mucosa of segment 1 in both the controls and the experimental rats. There was no dif,~re:~ce in the distribution of radioactive oleic acid between segments 1 and 2 in either group of rats, and thus only the data on segment 1 are presented. In the controls, mo~t of the radioactive oleic ~.cid was recovered in triacylglycerol (47% of segment I 14C radioactivity), the fatty acid (26%) and the monoacylglycerol plus phespholipid fractions (16%). In the experimental rats, there was significantly more radioactive oleic acid in the triacylglycerol fraction as compared to controls (P < 0.001). This increase in recovery in the triacylglycerol fraction was accompanied by a relative decrease in the fatty acid and the monoacylglycerol + phospholipid fractions. However, when the distribution of label was expressed as percentage of the dose of 14C infused over 8 h, it is clear that only the triacylglycerol changed significantly after L-81 (see values in parentheses in Table III).
Study B In this study, we measured the chylomicron and the VLDL appearance time in the control and the L-81treated rats, respectively. Lymph flow was comparable in both groups. The appearance time was 10.82 + 0.99
TABLE !!1
Distribution of I,C radionctivity among lipid classes in the mucosa of intestinal segment ! Percent distribution of segment 1 mucosal 14C; values are expressed means+ S.E. Five animals were studied in each group. CE, cholesteryl ester; 1"(3, triacylgly,~erol; FA, fatty acid; DG, diacylglyceml; MG, monoacylglycerol; PL, phospholipid. Numbers in parentheses are the percentage of the to~al dose of [t4C]oleic acid fed found in these fractions. Treatment
CE
TG
FA
DG
MG + PL
Control
4.7 + 1.4 (0.14)
47.3 ~ 3.2 (1.39)
25.6 + 0.9 (0.76)
6.7 + 0.8 (0.20)
! 5.7 + 2.4 (0.47)
Experimental
2.8 + 0.6 (0.21)
73.8 + 4.2 (5.36)
11.3 + 2.6 (0.82)
6.3 + 0.9 (0.45)
5.9 + 1.1 (0.42)
361 rain (mean + S.E.) and 16.60 + 0.91 rain for the control and the L-81-treated rats, respectively. The appearance time was significantly longer ( P < 0 . 0 1 ) for VLDLs (L-81-treated) as compared to chylomicrons (control). Discmsion It is generally accepted that there is very little monoacylglycerol stored in the enterocytes. Consequently, when oleic acid is infused without the accompanying monoacylglycerol, most of the absorbed fatty acid is processed by the a-glycerol phosphate pathway. Both the lymphatic triacylglycerol and the radioactivity output data clearly demonstrated ~ at L-81 significantly inhibited the intestinal lymphatic 'ipid transport when oleic acid is infused alone. These Lata would imply that the reason why the lymphatic lipid transport is not inhibited when egg phosphatidylcholine is infused is not because phospholipid-defived fatty acids are metabolized through the a-glycerol phosphate pathway to produce triacylglycerol. Rather, it is probably related to the fact that the fatty acid derived from egg phosphatidylcholine is mainly channeled to the formation of VLDL. Morphologic data seem to indicate that L-81 inhibits the traffic between the endoplasmic reticulum and the Golgi apparatus [20]. It is possible that the VLDLs produced during L-81 treatment do not require further processing in the Golgi apparatus prior to secretion. We have as yet no morphological data concerning whether the VLDL formed during the absorption of egg phosphatidylcholine is first processed in the Golgi apparatus prior to its release by the enterocytes. At least in the rainbow trout, there is morphological evidence that triacylglycerol-rich lipoproteins can be transported directly from the endoplasmic reticulum to the intercellular space [21]. We did not perform any electron microscopy in this study. However based on previous morphological studies from rats infused intraduodenally with triolein plus L-81, there probably were large lipid droplets accumulated in the cytoplasm of the L-81-treated rats [13,20]. When we analysed the distribution of the radioactive fatty acid in the intestinal mucosa, most of the fatty acid was in the triacylglycerol fraction in the L-81treated rats. This confirms our previous finding that L-81 does not interfere with the re-esterification of the absorbed lipid to form triacylglycerol (compare with mucosal lipid distribution in triolein-fed rats, Table II in Refs. 13 and 18). Although the majority of the labeled fatty acid was in the triacylglycerol fraction in the control animals, there was a substantial amount recovered in the free fatty acid and the monoacylglycerol + phospholipid fractions. This higher recovery of labeled fatty acid in the free fatty acid and partial acylglycerol fractions would seem to indicate that the lack of monoacylglycerol does impair the
esterification of fatty acid to form tfiacylglycerol. Consequently, the supply of 2-monoacylglycerol is important in ensuring the rapid re-esterification of fatty acid to form triacylglycerol. The fatty acids in the L-81 animals are reesterified to form triacylglycerol before being stored as large osmiophilic lipid droplets in the cytoplasm. We are currently trying to isolate the large osmiophilic lipid droplets inside the cytoplasm of the L-81-treated rats. Recent biochemical data from our laboratory have demonstrated that the large lipid droplets accumulated within the enterocytes after the intraduodenal infusion of triolein plus L-81 can be cleared rapidly from the mucosa after cessation of infusion of L-81 [22]. Although we do not know the mechanism of how the large lipid droplets are converted to chylomicron, we do know that the mucosal triacylglycerol is not hydrolyzed prior to the transport [22]. This is the opposite to what happens in the liver [23]. We have previously introduced the concept of chylomicron appearance time, wlfich is defined as the time between the placement of radioactive fatty acid in the intestinal lumen to the time when radioactive chylomicron appeared in the central lacteal [24,25]. In this experiment, we studied the chylomicron and the VLDL appearance times in the control and the L-81treated animals. The assumption is that the appearance time we measure in the control animals corresponds to the chylomicron appearance time, whereas that obtained in the L-81-treated animals corresponds to the VLDL appearance time. This is supported by the following observations. In a previous study [26] when the animals were fed fatty acid and monoacylglycerol, we found that the appearance time for radioactive lymph corresponds to the appearance time for chylomicron. That experiment involved ultracentrifugal separation of lipoprotein from the lymph and we found that the first appearance of radioactive fatty acid was associated with the chylomicron particles. In triacylglycerol-fed rats given L-81, previous morphologic studies [18] demonstrated that the majority of the particles transported in lymph were of VLDL size. In this study, the mean chylomicron appearance time for the control animals was 10.8 rain whereas in the experimental animals we obtained a significantly longer appearance time of 16.6 rain; this difference is highly significant, P < 0.01. This data would further support our previous proposal that there are separate pathways for the formation of chylomicron and the VLDL pathways [3]. However, there may be an alternative explanation for our data, which relates ~o the much smaller size of VLDL particles than chy~omicron. If we assume that the intestinal interstitial matrix (consisting mainly of the connective tissue of the lamina propria) behaves like a size exclusion chromatography column, it will take longer for the VLDL to p a s through this sieve relative to the chylomicron pa~ tides. This will be difficult to study and
362 we current have little information on the properties of the small intestinal interstitial matrix. In conclusion, these two experiments further support our proposal that chylomicron and VLDL are packaged separately by enterocytes. However, we want to emphasize that the VLDL produced during L-81 inhibition may be different from the VLDL produced normally during fasting.
Acknowledgements The authors are grateful to Deborah S. Drake, Camille Pool and Reneau Youngblood for their excellent technical help. This work was supported by National Institutes of Health Grant DK 32288. P.T. is the recipient of a Research Career Development Award DK 01575. References I Mahley, R.W., Bennett, B.D., Moore, D.J., Gray, M.E., Thistlethwaite, W. and LeQuire" V.S. (1971) Lab. Invest. 25, 435-444. 20ckner, R.K., Hughes, F.B. and Isselbacher, K.J. (1969) J. Clin. Invest. 48, 2367-2373. 3 Tso, P., Drake, D.S., Black, D. and Sabesin, S.M. (1984) Am. J. Physiol. 247, G599-G610. 4 Arnesjo, B., Nilsson, ~,., Barrowman, J.A. and Borgstrom, B. (1969) Scan& J. Gastroenterol. 4, 653-655. 5 S~w, R.O., Stein, Y. and Stein, 0. (1967) J. Biol. Chem. 242, ~*,919-4924. 6 Nilsson, ~. (1968) Biochim. Biophys. Acta (1968) 137, 240-254. 7 Subbaiah, P.V. and Gaaguly, J. (1970) Bioehem. J. i18, 233-239.
8 Parthasarathy, S., Subbaiah, P.V. and Ganguly, J. (1974) Biochem. J. 140, 503-508. 9 Brindley, D.N. and Hubscher, G. (1965) Biochim. Biophys. Acta 106, 495-509. 10 Johnston, J.M., Rao, G.A. and Lowe, P.A. (1967) Biochim. Biophys. Acta 137, 578-580. 11 Tso, P. (1985) Adv. Lipid Res. 21, 143-186. 12 Bollman, J.L., Cain, J.C., and Grindlay, J.H. (1948) J. Lab. Clin. Mud. 323, 1349-1352. 13 Tso, P., Balint, J.A. and Rodgers, J.B. (1980) Am. J. Physiol. 239, G348-G353. 14 Foleh, J., Lees, M. and Sloane-Stanley, G.H. (1957) J. Biol. Chem. 226, 497-509. 15 Biggs, H.G., Eriekson, J.J. and Morehead, W.R. (1975) Clin. Chem. 21, 437-441. 16 Rodgers, J.B., Fondacaro, J.D. and Kot, J. (1977) J. Lab. Clin. Mud. 89, 14-152. 17 Snedecor, G.W. and Cochran, W.G. (1967) Statistical Methods (6th Edition), Ames, IA; Iowa State University Press. 18 Tso, P., Balint~ J.A., Bishop, M. and Rodgers, J.B. (1981) Am. J. Physiol. 241, G487-G497. 19 Tso, P. and Gollamudi, S.R. (1984) Am. J. Physiol. 247, G32-G36. 20 Sabesin, S.M., Frase, S. and Tso. P. (1985) Gastroenterology 88, 1565 (Abstract). 21 Vernier, J.M. and Sire, M.F. (1986) Tissue Cell 18, 447-460. 22 Halpern, J., Tso, P. and Mansbach, C.M.il. (1988) J. Clin. Invest. 82, 74-81. 23 Mooney, R.A. and Lane, D.M. (198i) J. Biol. Chem. 256, 11724-11733. 24 Tso, P., Pitts, V. and Granger, D.N. (1985) Am. J. Physiol. 249, G21-(328. 25 Tso, P., Barrowman, J.A. and Granger, D.N. (1986) Am. J. Physiol. 250, G497-GS00. 26 Hall, J.L. (1987) Intestinal absorption: Studies of absorption of certain natural lipids and lipophilic xenobiotics. Master of Science Thesis, Memorial University of Newfou,dland.