Effects of colchicine on the intestinal transport of endogenous lipid

GASTROENTEROLOGY 1983;84:544-55

Effects of Colchicine on the Intestinal Transport of Endogenous Lipid Ultrastructural, Radiochemical MARGIT

PAVELKA

Biochemical, and Studies in Fasting RQts and

ALFRED

GANGL

Institute of Micromorphology and Electronmicroscopy Hepatology, University of Vienna, Austria

The involvement of microtubules in the transepitheJiaJ transport of exogenous lipid in intestinal absorptive cells has been suggested. Using electronmicroscopic, biochemical, and radiochemical methods, we have studied the effects of the antimicrotubular agent colchicine on the intestinal mucosa and on the intestinal transport of endogenous lipid of rats in the fasting state. After colchicine treatment, the concentration of triglycerides in intestinal mucosa of rats fasted for 24 h doubled, and electron microscopic studies showed a striking accumulation of lipid particles in absorptive epithelial cells of the tips of jejunal viJJi. These findings suggest that colchicine interferes with the intestinal transepithelial transport of endogenous lipoproteins. Additional studies, using an intraduodenal pulse injection of [14C]JinoJeic acid, showed that colchicine does not affect the uptake of fatty acids by intestinal mucosa. However, it had divergent effects on fatty acid esterification, enhancing their incorporation into triglycerides relative to phospholipids, and caused a significant accumulation of endogenous diglycerides, triglycerides, and cholesterol esters within the absorptive intestinal epithelium. Detailed ultrastructural and morphometric studies revealed a decrease of visible microtubules, and a displacement of the smooth and rough endoplasmic reticulum and GoJgi apparatus.

Received March 24, 1980. Accepted October 7, 1982. Address requests for reprints to: Univ. Prof. Dr. A. Gangl, I. Universitatsklinik fiir Gastroenterologie und Hepatologie, Lazarettgasse 14, A-1090 Vienna, Austria. The expert technical help provided by Mrs. B. Klose, Mrs. J. Selbmann, and Mrs. E. Scherzer, the photographic help by Mr. H. Wagner, and the secretarial help by Mrs. Ch. Andra is gratefully acknowledged. 0 1983 by the American Gastroenterological Association 0016-5085/83/030544-12$03.00

and the II. Clinic of Gastroenterology

and

Furthermore, it is shown that after colchicine treatment, microvilli appear at the lateral plasma membrane of intestinal absorptive cells, a change not previously reported to our knowledge. Thus, our study shows that [a) colchicine causes significant changes in enterocyte ultrastructure and (bJ colchitine perturbs the reesterification of absorbed endogenous fatty acids and their secretion in the form of triglyceride-rich lipoproteins from the enterocyte.

An involvement of microtubules in intracellular transport mechanisms has been suggested for various tissues (l-20); evidence for their role in the transport of lipid in intestinal absorptive cells has recently been presented by Glickman et al. (21) and by Arreaza-Plaza et al. (22) who showed that the rate of intestinal lipoprotein secretion during fat absorption was decreased in rats treated with colchicine, a well-documented antimicrotubular agent (23). Furthermore, studies on the regional distribution and content of microtubules in intestinal absorptive cells in relation to the transport of lipid revealed a decrease of assembled microtubules after a fat-containing meal, giving additional evidence for an involvement of microtubules in the transport of lipids (24). As the intestinal mucosa has been shown to secrete lipoproteins also during fasting (25), we have studied the effect of the antimicrotubular drug colchicine on the intestinal mucosa of rats in the fasting state using electronmicroscopic, biochemical, and radiochemical methods. Preliminary reports of these studies have been presented at the i’2nd Meeting of the German Anatomical Society (26), at the 6th International Congress on Electronmicroscopy (27), and at the 26th Colloquium on Protides in the Biological Fluids (28).

March

COLCHICINE AND INTESTINAL LIPID TRANSPORT 545

1983

Methods Animals Female albino rats of 200-250 g body wt, kept on a standard rat pelleted diet, were fasted 24 h before the experiment. Colchicine (Fluka AG, Buchs, Switzerland, biochemical purity >98%) was dissolved in 0.9% NaCl immediately before use and injected intraperitoneally at a dose of 0.5 mg/lOO g body wt. Control animals received either lumicolchicine [prepared by irradiation of colchitine (29)] at a dose of 1.0 mg/lOO g body wt or 0.9% NaCl. The time of application of the drug was 9 AM. Ninety minutes, 2, 3, 4, and 6 h later, animals were killed by decapitation. Blood was collected from the cervical wound. The entire small bowel between the ligament of Treitz and the cecum was excised immediately, flushed with 50 ml of chilled saline, and divided into two halves. Only the proximal half (jejunum) was used for further workup.

Ultrastructural

Investigations

Small pieces of the proximal third of the jejunum were opened longitudinally and covered immediately with 2.5% glutaraldehyde buffered to pH 7.2 in 0.1 M sodium cacodylate or sodium phosphate buffer (osmolarity, 430 mosmol). Subsequently, the tissue was submerged and kept in the same fixative for 6-10 h at 20°C. After postfixation in veronalacetate-buffered 1% 0~0~ for 2 h, specimens were dehydrated in a graded series of ethanol and embedded in Epon. Semithin sections were stained with toluidine blue and scanned for upright villi. Only villi containing longitudinally oriented absorptive cells were chosen for ultrastructural examination. Ultrathin sections were stained with alcoholic uranyl-acetate and alkaline lead citrate and examined on a Zeiss EM 9 (Carl Zeiss, Inc., New York, N.Y.) or on a Philips EM 400 (Philips Electronic Instruments, Inc., Mahwah, N.J.). For morphometric analysis, only absorptive cells of the upper third of the villi were studied and the first 10 cells showing the nucleus were taken for examination. Three portions of each cell were photographed individually at a magnification of 13,000. In general, micrographs from the first portion (apical region) included the terminal web and the superficial half of the apical cytoplasm, but did not include microvilli. The second portion (Golgi region) contained the Golgi complex, the tip of the nucleus, and the cytoplasm between Golgi complex and the lateral plasma membrane. The third portion (basal region) represented the cytoplasm between nucleus and the basal plasma membrane. Cells of the villus tips exhibiting pronounced intracellular accumulations of lipid particles were excluded from the morphometric analysis, because in these cells a clear differentiation between endoplasmic reticulum and Golgi-derived membranes was not possible in every case. Quantitative analysis of the microtubular content was performed using the point-counting stereological technique described by Weibel(30) and used for analysis of the

microtubular content of intestinal epithelial cells by Reaven and Reaven (24). Transparent millimeter grids were placed over each photographic enlargement (X 3) of the electron micrographs and the number of points (P) lying over microtubules relative to the number covering the cytoplasm were recorded. The volume density of the microtubules was calculated from the ratio of Pmicrofubuies/ P cytopiasmand corrected by a multiplication factor of 0.4 (24). For quantitative analysis of the membranes of the endoplasmic reticulum and of elements of the Golgi apparatus, the membrane concentration expressed as the ratio of microns of membrane to square microns of cytoplasm was calculated as described by Loud (31,32). A transparent sheet with equidistant parallel lines (spaced 10 mm apart) was placed over the photographic enlargements (X 3) both vertically and horizontally. The length of each line crossing the cytoplasm excluding nuclei and intercellular spaces was measured. The total cytoplasmic area under investigation was represented by summation of the total length of all lines of the micrograph. The intersections between individual lines and the membranes of the endoplasmic reticulum and Golgi elements were counted and the total membrane concentration, expressed as microns of membrane/square microns of cytoplasm was calculated using the formula presented by Loud (31,32).

Biochemical

Investigations

From the residual two-thirds of the jejunum, the mucosa was extruded on a chilled glass plate by squeezing with a spatula. It was weighed, transferred immediately into a Teflon-glass-homogenizer containing 3 ml methanol per gram mucosal weight, and homogenized. Lipids were extracted from mucosal homogenates and whole serum by the method of Folch et al. (33) and separated into individual lipid classes on precoated thin-layer chromatography plates (silica gel, 0.25 mm, EM reagents, E. Merck, Darmstadt, W. Germany) in petroleum etheridiethyletheriacetic acid, 90:15:1.5, as described earlier (34). The lipid zones were eluted from the silica gel, and these eluates were used for determination of fatty acids (35), triglycerides (36), and radioactivity (see Isotope Studies).

Isotope

Studies

In additional experiments 2 h after intraperitoneal injection of either colchicine or saline (controls) as described previously, several rats were anesthetized with diethylether; laparotomy was performed, the duodenum was exposed, punctured with a steel needle (gauge 211, and 2-3 &i of [l-‘4C]linoleic acid (60 &i/mmol, radiochemical purity >97%, The Radiochemical Centre, Amersham, England) solubilized in 2 ml of 10 mM taurocholate (Calbiochem-Behring Corp., American Hoechst Corp., San Diego, Calif.; >95% pure) were injected directly into the duodenum as a single pulse. The abdominal wound was closed and 2, 10, and 120 min after this intraduodenal injection, rats were decapitated, the blood of the neck wound was collected, and the intestinal mucosa was

GASTROENTEROLOGY Vol. 84, No. 3

546 PAVELKA AND GANGL

isolated from the jejunum. It was homogenized, and the lipids were extracted and separated as described. Aliquots of homogenates (0.2ml) were assayed for radioactivity in liquifluor toluene solution (New England Nuclear, Boston, Mass.) containing 10% Biosolv [Beckman Instruments, Inc., Fullerton, Calif.) in a Beckman liquid scintillation counter (model LS 3150 T). For lipid-soluble extracts, Biosolv was not added. Quenching was corrected for by an automatic external standard, and results were expressed as disintegrations per minute. Calculation

l-l I

controls

ISI

colchicine

2 hours

of Data

To make radioactivity data of different rats comparable, these data were normalized by multiplying each disintegration-per-minute value by the factor 1 &i/&i injected. Results were given as means 2 1 standard error; comparison between experimental and control animals was made by Student’s t-test.

l-

Results Detailed ultrastructural descriptions of the intestinal absorptive cells have been published repeatedly (37-41); our own morphologic findings in fasted control rats do not exhibit marked deviations from these reports. The most conspicuous colchicine-induced ultrastructural changes concern microtubules, Golgi apparatus, lipid particles, and the lateral plasma membrane: they appear most prominently 2-6 h after application of colchicine. In control animals, microtubules measuring 25 nm in diameter are demonstrable in all cytoplasmic regions of jejunal absorptive cells; however, morphometric analysis shows that the microtubular content is significantly higher (p < 0.01) in the supranuclear cytoplasm than in other cellular areas (Figure 1). Frequently, microtubules are situated in the vicinity and in parallel orientation to Golgi cisternae (Figure 2) and to the lateral plasma membrane. In animals treated with colchicine, the number of visible microtubules is significantly reduced (p < 0.001) in all three cytoplasmic portions of jejunal absorptive cells, the decrease being most pronounced in the supranuclear cytoplasm (Figure 1). In jejunal absorptive cells of fasted control animals (saline as well as lumicolchicine), the Golgi apparatus is constantly localized in the supranuclear cytoplasm (Figure 2a). Most of the Golgi stacks exhibit a showing dilated cisternae and polar appearance, vacuoles containing lipid particles (30-130 nm) at one side and narrow cisternae at the other (Figure 2). The presence of lipid particles within Golgi cisternae and vacuoles of fasted control rats indicates secretion of endogenous intestinal lipids, consistent with earlier reports by others (38).

,-

i

cl SN

Figure

cl B

I. Microtubule content of the apical (A), supranuclear (SN), and basal (B) cytoplasmic region of jejunal absorptive cells from the upper third of the villi from 10 control animals and 10 experimental animals treated with colchicine for 2 h (mean C 1 SE).

After application of colchicine, structure as well as position of the Golgi apparatus are altered. Golgi stacks with irregularly dilated cisternae and mostly lacking their typical polarity are not restricted to the supranuclear cytoplasmic regions, but are dispersed throughout the entire cytoplasm (Figure 3 a,b]. In addition, apical and basal cytoplasmic areas contain a various number of vacuoles, the origin of which is not clearly recognizable (Figure 3); possibly, they also represent transformed Golgi components. Quantitatively, the jejunal absorptive cells of colchicinetreated animals show a significant decrease of Golgi cisternae (p < 0.001) and of Golgi vacuoles (p <

March

Figi ure

1983

COLCHICINE

AND

INTESTINAL

LIPID TRANSPORT

547

2. a. Supranuclear and apical cytaplasmic areas of two jejunal absorptive cells from the upper portion of a villus. The Golgi the characteristic polar apparatus (arrow) is typically localized in the supranuclear regions. (X 7400). b. Golgi stack exhibiting organization with narrow cisternae at one side and dilated cisternae containing lipid particles (arrow] at the other side. N = nucleus. Arrowhead = microtubule. (~48,000.)

0.05) in the supranuclear cytoplasm. Furthermore, elements of Golgi stacks appear in the apical cytoplasm (Figure 4). In addition, vacuoles are increased significantly in the supranuclear cytoplasm (p < 0.001) and appear also in the apical and basal cytoplasmic regions (Figure 4). A dislocation of the endoplasmic reticulum was observed in that the membranes of the smooth endoplasmic reticulum are decreased significantly in the apical (p < 0.001) and increased in the basal cytoplasm (p < 0.001) 2 h after colchicine application (Figure 5). In addition, an increase of the membranes of the rough endoplasmic reticulum in the supranuclear cytoplasm 2 h after colchicine treatment (p < 0.05) and a decrease of these membranes in the basal cytoplasm (p < 0.05) was noted (Figure 5).

In rats treated with colchicine, numerous lipid particles ranging between 80 and 4000 nm in diameter are accumulated in the cytoplasm of absorptive cells of the villus tips (Figure 6). Most of these lipid particles are surrounded by a membrane. The enwrapping membranes may partially belong to the smooth endoplasmic reticulum and may be partially derived from changed Golgi stacks. Because of the dilatations of the endoplasmic reticulum and of the vacuolarlike alterations and the dislocation of the Golgi stacks after colchicine treatment in cells of the villus tips, a clear morphologic differentiation between membranes derived from the endoplasmic reticulum and those originating from Golgi components is not possible in every case. Epithelial cells of other villous portions contain lipid particles measur-

548 PAVELKA AND GANGL

GASTROENTEROLOGY Vol.84.No. :I

Figure 3. Colchicine treatment. Apical (a] and basal (b) cytoplasmic areas of jejunal absorptive cells from the medium portion of a villus. Altered Golgi stacks (arrow] and vacuoles (arrowheads], possibly also originating from Golgi components, are located atypically in apical (a] and basal (b) cytoplasmic regions. As compared with the control rats (Figure 2a), membranes of the smooth endoplasmic reticulum are reduced in the apical cytoplasmic region, and those of the rough endoplasmic reticulum are increased in the supranuclear cytoplasmic area (the changes of Golgi components and of the endoplasmic reticulum have been confirmed morphometrically-see Figures 4 and 5). At the lateral plasma membrane, microvilli similar to those of the apical brush border are apparent. (a, x 16,000. b, X20,000.)

ing 50-400 nm in diameter, which are predominantly situated in the dislocated and altered Golgi stacks and vacuoles described previously [Figure 3). These lipid particles, as well as those accumulated in absorptive cells of the villus tips (Figure 6), presumably represent retained lipid. An additional interesting morphologic detail occurring after colchicine treatment concerns the lateral plasma membrane: microvilli resembling those of the apical brush border of intestinal absorptive cells are present also at the lateral cellular face (Figures 3b and 7). This phenomenon, which is never seen in controls, was observed in all 25 animals treated with colchicine for 6 h. Corresponding with the ultrastructural appearance of an increase in number and size of intracellular lipid particles, the biochemical analysis of intes-

tinal mucosa revealed a significantly higher mucosal triglyceride concentration at 1.5, 2, and 3 h after colchicine application compared with controls (Figure 8); 6 h after colchicine injection, the intestinal mucosal triglyceride concentration was still higher than that of the controls, yet not significant statistically. By contrast, the triglyceride concentration of serum was reduced in rats treated with colchicine to about half of that found in controls at 2 and 6 h (Figure 8). Despite this impressive change of the serum triglyceride concentration after colchicine application, the concentration of plasma free fatty acids was unaffected by colchicine when compared with controls (Figure 8). Similarly, the concentration of free fatty acids in intestinal mucosa was equal at 1.5, 2, and 6 h in rats treated with colchicine and in

March 1983

Figure

COLCHICINE

cl

controls

&l

cdchicint

2 hours

fl

cokhicine

3 hours

4. Content of membranes of the Golgi apparatus (Golgi cisternae and Golgi vacuoles] and of vacuoles (the origin of which is not clearly recognizeable) of the apical [A], supranuclear (SN), and basal (B) cytoplasm of jejunal absorptive cells from the upper third of the villi of controls and of animals treated with colchicine for 2 and 3 h. Each bar represents the mean + 1 SE of 10 rats.

controls. Only at 3 h did colchicine-treated rats have a higher mucosal concentration of free fatty acids than control rats (Figure 8). To study the effect of colchicine on the intestinal metabolism of luminal long chain fatty acids more directly, the radioactivity of intestinal mucosa after intraduodenal injection of [‘4C]linoleic acid was determined in rats to which either colchicine or saline had been administered 2 h before isotope injection. In both groups, the mucosal radioactivity was significantly higher at 2. and 10 min than at 120 min (Table 1). Furthermore, 2 min after intraduodenal injection of [*4C]linoleic acid, no difference between colchicine-treated and control rats was ob-

AND

INTESTINAL

LIPID TRANSPORT

549

served, suggesting equal initial uptake,of the fatty acid from the lumen. At 10 min, and more pronounced at 120 min after the intraduodenal pulse injection of [14C]linoleic acid, however, the mucosal radioactivity of rats pretreated with colchicine was significantly higher than that of control rats. This means that, despite equal initial uptake, the “export” of 14C from intestinal mucosa, which presumably starts only a few minutes after the uptake of the [14C]linoleic acid from the lumen, has occurred to a smaller extent at 10 and particularly at 120 min in colchicine-treated rats than in controls (Table 1).

Cl

controls

69

colchicinc

2 hours

: colchicine ICI

3 hours

SER

RER

SER

RER

SER

RER

Content of membranes of the smooth (SEA) and rough (RER) endoplasmic reticulum of the apical (A). supranuclear (SN), and basal (B) cytoplasm of jejunal absorptive cells from the upper third of the villi of controls and of animals treated with colchicine for 2 and 3 h. Each bar represents the mean f 1 SE of 10 rats.

550

PAVELKA AND GANGL

GASTROENTEROLOGY Vol. 84. h’o. 3

re 6. Colchicine treatment. Supranuclear and apical cytoplasmic regions of jejunal absorptive cells from the villus tip filled with numerous lipid particles. Most of them are surrounded by a membrane (arrow). (X 10,000.)

The incorporation of 14C into individual lipid classes of intestinal mucosa is shown in Table 2. It is of particular interest that despite equal initial uptake of 14C, at 2 min after the rapid intraduodenal injection of [i4C]linoleic acid, intestinal mucosa of rats pretreated with colchicine contains significantly less 14C in phospholipids and more 14C in triglycerides when compared with controls.

Discussion Previous studies have emphasized a significant inhibitory effect of colchicine on the lymphatic absorption of exogenous lipids (21,22).The ultrastructural, biochemical, and radiochemical data of our studies provide evidence for a similar inhibitory

effect of colchicine on the intestinal transepithelial transport of endogenous lipids in the fasting state. Consistent with the well-documented impairment of the assembly of microtubules by colchicine in vitro (23) and in agreement with the studies of Reaven and Reaven (24),the intracellular content of visible microtubules of intestinal epithelial cells was significantly reduced 2 h after colchicine application in vivo; at the same time the lipid content of intestinal epithelial cells was increased significantly, thus supporting the hypothesis of a functional significance of microtubules in the intestinal lipoprotein secretion, put forward by several authors (21,22,24). Our isotope studies show that 2 min after the intraduodenal pulse injection of a trace amount of [‘4C]linoleic acid, total intestinal mucosal radioac-

March

Figure

1983

COLCHICINE

AND

INTESTINAL

LIPID TRANSPORT

551

7. Colchicine treatment. Longitudinal (a,b) and transverse (b) sections of microvilli resembling those of the apical brush border but localized at the lateral plasma membrane. The filamentous core continues into the cytoplasm where a network similar to the apical terminal web is apparent. (a, ~43,000. b, ~35,000.)

tivity was equal in colchicine-treated rats and controls. Provided there was no difference in intraluminal fatty acid concentration (derived from biliary phospholipids and shed epithelium) between the two groups of rats studied, intestinal mucosal radioactivity at 2 min would reflect an equal uptake of the fatty acid from the lumen. It is very unlikely that, within the 2 h of the study period of interest, a significant difference in epithelial shedding would have existed, and previous investigators have shown that the effect of colchicine on biliary secretion of phospholipids is small or nonexistent at 2 h (1,3,6). We therefore conclude that the equal amount of mucosal **C at 2 min after the intraduodenal injection of [‘*C]linoleic acid indeed indicates that colchicine did not interfere with the uptake of luminal

fatty acid mass. Given the fact of an equal intestinal uptake of fatty acids in both groups of rats studied, the striking difference in the l*C radioactivity of intestinal mucosal triglycerides and phospholipids between colchicine-treated rats and controls (Table 2, data at 2 min) indicates a colchicine-induced shift in the fatty acid esterification pathways in favor of triglycerides at the expense of phospholipids. Alternatively, if it is assumed that the luminal fatty acid concentration was diminished in the colchicine-treated rats, then luminal fatty acid specific activity would be higher and the observed equality of uptake of ‘*C in the two groups would imply a reduced uptake of fatty acid mass in the colchicine group. In this case, however, the decreased incorpoin the ration of l*C into mucosal phospholipids

552

PAVELKA AND GANGL

GASTROENTEROLOGY Vol.

ii

51

1;s

2P

*

*

30

30

hours

hours

hours

6/.l

6~3

E$J

t @ I

(I

p-qoo5;

t

p-=

005

0

CONTROLS

specific radioactivity, colchicine has had a profound effect on one or more aspects of mucosal fatty acid esterification. Specific activities of mucosal triglycerides and phospholipids were not measured directly in this study. However, using the triglyceride concentrations given in Figure 8 (2 h after colchicine) and the isotope data presented in Table 2 (2 h and 2 min after colchicine), it is possible to estimate specific activities of mucosal triglycerides. Thus, the specific activity of triglycerides is 73,455 dpm/pmol in colchitine-treated rats and 86,046 dpm/pmol in controls. Apparently there is not much difference. Considering the fact that 2 h after colchicine application the mucosal triglyceride concentration had already doubled due to the “exit block,” and the 14C trace was taken up at an equal rate in both colchicine and control rats, the calculated equality of triglyceride specific activities simply reflects the fact that more label must have been incorporated into mucosal triglycerides of colchicine-treated rats than in control rats. Interestingly, in this context, evidence has been presented in vitro for an increased reesterification of free fatty acids in adipose tissue of rats exposed to colchicine (42) and furthermore, colchicine causes a slight increase of the incorporation of [14C]glucose into the glycerol moiety of the triglyceride of adipose tissue of rats (42). Thus, the effect of colchicine on the transepithelial lipid transport does not appear to be a mere exit block involving the assembly and exocytosis of lipoproteins only, as suggested by previous studies (21). Rather, colchicine seems to affect the esterification of absorbed fatty acids to triglycerides and phospholipids as well. More detailed studies of fatty acid esterification rates are needed for the further elucidation of the significance of this effect, however. Table

1. Radioactivity

hours

&j

@I

COLCHICINE

Figure 8. Biochemical lipid analysis of serum and intestinal mucosa of rats 1.5,2,3, and 6 h after the intraperitoneal injection of colchicine (broken line, hatched bars) or of saline (unbroken line, white bars]. Data are means 2 1 SE of at least 4 rats at any given point of time. Statistically significant differences between experimental animals and controls are indicated (*,+).

colchicine-treated group would reflect an even more profound depression of the incorporation of fatty acid mass. Thus, regardless of the assumptions made with respect to luminal fatty acid concentration and

of Intestinal Mucosa After Injection of 1 &i [‘4CJLinoleic

Intraduodenal Acid

I

Y)

84, No. 3

Time after injection of 14C (min] 2 10 120

92 radioactivity of intestinal mucosa (corr. dpm x 10e4 per g wet mucosa) Colchicine

Controls

Significance

7.3

66.9

+ 6.7

NS

94.4 2 3.4

77.6

+ 6.4

44.0

24.4

2 4.7

p < 0.05 p < 0.02

76.1

f

k 4.6

Rats fasted for 24 h were injected intraperitoneally either with colchicine (0.5 mg/lOO g) or with 0.9% saline. Two hours later, laparotomy was performed under ether anesthesia and a trace amount of [‘%]linoleic acid, solubilized in 2 ml of 10 mM taurocholate, was injected directly into the duodenum. Two, 10, and 120 min later, rats were decapitated, the small bowel mucosa was isolated, and its 14C radioactivity was determined (Methods). Data represent the means 2 1 standard error of 5 experimental and 5 control rats at each time point.

COLCHICINE AND INTESTINAL LIPID TRANSPORT

March 1983

Table

2. Incorporation of 14C Into Individual Lipid Classes of Intestinal Mucosa After Intraduodenal Injection of l&i [‘4C]Linoleic Acid

Colchicine” After 2 min 15.6 t 1.6 PL

Control”

Significance of difference

3.5 1.3 30.4 1.2

t k + ?

0.5 0.2 4.1 0.2

28.7 3.7 1.2 16.9 1.0

2 + 2 + 2

3.9 1.1 0.2 3.2 0.2

After 10 min PL 20.2 DG 3.7 FA 0.9 TG 35.0

? 2 2 t

4.6 0.9 0.2 3.8

29.2 3.7 2.2 28.3

r t + 2

2.0 0.8 0.6 1.5

NS NS NS NS

1.9 t 0.2

NS

DC FA TG CE

CE

1.7 t 0.1

After 120 min PL 14.5 ? DG 2.4 * FA 0.6 ” TG 18.3 + CE 0.5 +

1.3 0.3 0.1 2.6 0.0

14.5 0.8 0.3 2.0 0.1

+ 2 + r 2

2.3 0.2 0.1 0.3 0.0

p < 0.02 NS NS p < 0.05 NS

p p p p

< < < <

NS 0.005 0.01 0.001 0.001

Difference as % of control - 46% + 80% -

-

+ + + +

203% 121% 835% 403%

See also Table 1. Lipids were extracted from intestinal mucosa, separated into individual lipid classes by thin-layer chromatography, and the radioactivity was determined in each lipid class (Methods). Each value represents the mean ? I standard error of 5 rats. (PL = phospholipids, DG = diglycerides, FA = fatty acids, TG = triglycerides, CE = cholesterol esters.) a Corrected disintegrations per minute X 1O-4 per gram wet mucosa.

The finding of an increased amount of l*C in triglycerides, cholesterol esters, and diglycerides compared with controls at 120 min after the intraduodenal injection of [14C]linoleic acid in rats pretreated with colchicine indicates that the export of endogenous lipids was held up in the intestinal mucosa, consistent with the concept of an exit block. This has been previously demonstrated for exogenous lipids by others (21,22). Although it is well known that endogenous intestinal very low density lipoproteins contribute to plasma lipoproteins (25,43), the significant reduction of plasma triglycerides in our studies cannot be attributed to an impairment of endogenous intestinal lipoprotein secretion only, as colchicine does of course impair the secretion of hepatic lipoproteins as well (5,6). Contrary to the striking fall of the plasma triglyceride level in colchicine-treated rats, the level of plasma free fatty acids remained virtually unchanged. This finding is consistent with the observation of Schimmel (42) that, in vitro, colchicine does not alter basal, unstimulated free fatty acid release from segments of epididymal adipose tissue from normal male rats, and represents first evidence that

553

lipolysis in peripheral adipose tissue most likely is not dependent on microtubules. Further investigation on this point is needed, however, to establish this hypothesis more definitely. Although we have used the same dosage of colchitine (0.5 mg/100 g body wt) as Glickman in his studies (21), our ultrastructural findings are at variance with Glickman’s general description of a normal ultrastructural appearance of intestinal mucosa in fasting rats given colchicine without subsequent lipid administration (21); particularly striking are our observations of the accumulation of lipid particles, displacements of the Golgi apparatus and of the endoplasmic reticulum, and especially the changes of the lateral plasma membrane. The changes of the Golgi apparatus are in agreement with reports on alterations and displacements of Golgi stacks induced by antimicrotubular agents in various other cells and tissues (5,7-9,12-15,1720,44,45). In intestinal epithelial cells, Reaven and Reaven (24) described an abnormal position of the Goigi apparatus in animals treated with colchicine. Our studies confirm this finding morphometrically and, in addition, reveal structural alterations of the individual Golgi stacks. Colchicine-induced alterations of the plasma membrane similar to those observed at the lateral plasma membrane of intestinal epithelial cells in our study have not been described so far. However, colchicine has been shown to affect the mobility of various membrane components (46) and to cause changes of the site of specific membrane functions (47). The appearance of microvilli at the lateral plasma membrane might suggest an assimilation of membrane properties of the lateral to those of the apical plasma membrane after colchicine treatment. Additional indications for this hypothesis have been provided by studies showing the incorporation of [“Hlfucose in plasma membrane glycoproteins (48) and by cytochemical localization of thiamine pyrophosphate activity (49). The alterations of the lateral plasma membrane are of particular interest with regard to the changes of the Golgi apparatus, since plasma membrane is preformed in the Golgi apparatus (50,51) and both structures are involved in final steps of lipoprotein secretion (52,53). Generally, it is of interest that several subcellular structures, which

are known to be involved in formation and secretion of lipoproteins-smooth and rough endoplasmic reticulum, Golgi apparatus, and lateral plasma membrane-are affected after colchicine application. In agreement with the radiochemical finding of a colchicine-induced shift in the intracellular fatty acid esterification pathways, the dislocation of the smooth endoplasmic reticulum, where esterification is localized, indicates an interference of colchicine

554

PAVELKA AND GANGL

with initial processes of lipoprotein formation. Furthermore, the changes of the smooth and rough endoplasmic reticulum as well as the prominent alterations of the Golgi apparatus make it conceivable that the transport of lipid particles from the channels of the endoplasmic reticulum to the Golgi apparatus (52,533 and the final formation of lipid particles in the Golgi apparatus (52,53) might be disturbed. Finally, the alterations of the lateral plasma membrane must also be considered as a possible cause of impaired lipoprotein secretion. The multiplicity of changes obviates an exact localization of the disturbances in the transepithelial pathway of lipid particles; however, in contrast to former reports (21,22) our morphologic as well as biochemical and radiochemical data give evidence that colchicine interferes with both initial and final steps in formation and secretion of lipid particles. Although the antimicrotubular effect of colchicine provides an attractive hypothetical explanation for the ultrastructural changes as well as for the impairment of the intestinal lipoprotein secretion, other effects of colchicine must be taken into consideration as well; of particular significance in this respect is the demonstration of a direct binding of colchicine to subcellular membranes (54), implying the possibility of direct colchicine-membrane interactions.

References 1.

2.

3.

4.

5.

6.

Gregory DH, Vlahcevic R, Prugh MF, et al. Mechanism of secretion of biliary lipids: role of a microtubular system in hepatocellular transport of biliary lipids in the rat. Gastroenterology 1978;74:93-100. LeMarchand Y, Patzelt C, Assimacopoulos-Jeannet F, et al. Evidence for a role of the microtubular system in the secretion of newly synthesized albumin and other proteins by the liver. J Clin Invest 1974;53:1512-7. Dubin M, Maurice M, Feldmann G, et al. Influence of colchitine and phalloidin on bile secretion and hepatic ultrastructure in the rat. Possible interaction between microtubules and microfilaments. Gastroenterology 1980;79:646-54. Orci L, LeMarchand Y, Singh A, et al. Role of microtubules in lipoprotein secretion of newly synthesized albumin and other proteins by the liver. Nature 1973;244:30-2. Reaven EP, Reaven GM. Evidence that microtubules play a permissive role in hepatocyte very low density lipoprotein secretion. J Cell Biol 1980;84:28-39. Stein 0, Sanger L, Stein Y. Colchicine induced inhibition of lipoproteins and protein secretion into the serum and lack of interference with secretion of biliary phospholipids and cholesterol by rat liver in vivo. J Cell Biol 1974;62:90-103. Moskalewski S, Thyberg J, Friberg V. In vitro influences of colchicine on the Golgi complex in A- and B-cells of guinea pig pancreatic islets. J Ultrastruct Res 1976;54:304-17. Ericson LE. Inhibition of intracellular protein transport in the mouse exocrine pancreas induced by vinblastine. Cell Tissue Res 1980;206:73-81. Patzelt C, Brown D, Jeanrenaud B. Inhibitory effect of colchi-

GASTROENTEROLOGY Vol. 84. No. 3

10.

11.

12.

13.

14.

15.

16. 17.

18.

19.

20.

21.

22.

23. 24.

25.

26.

tine on amylase secretion by rat parotid glands. Possible localization in the Golgi area. J Cell Biol 1977;73:578-93. Nickerson SC, Smith JJ, Keenan TW. Role of microtubules in milk secretions: action of colchicine on microtubules and exocytosis of secretory vesicles in rat mammary epithelial cells. Cell Tissue Res 1980;207:361-76. Chajek T, Stein 0, Stein Y. Interference with the transport of heparin-releasable lipoprotein lipase in the perfused rat heart by colchicine and vinblastine. Biochim Biophys Acta 1975;388:260-7. Ehrlich HP, Ross R, Bornstein P. Effects of antimicrotubular agents on the secretion of collagen. J Cell Biol 1974;62:390405. Moskalewski S, Thyberg J, Lohmander S, et al. Influence of colchicine and vinblastine on the Golgi complex and matrix deposition in chondrocyte aggregates. Exp Cell Res 1975; 95:440-54. Piasek A, Thyberg J. Effect of colchicine on endocytosis and cellular inactivation of horseradish peroxidase in cultured chondrocytes. J Cell Biol 1979;81:426-37. Thyberg J, Nilsson S, Moskalewski S, et al. Effects of colchitine on the Golgi complex and lysosomal system of chondrocytes in monolayer culture. An electron microscopic study. Cytobiologie 1977;15:175-91. Williams JA, Wolff J. Possible role of microtubules in thyroid secretion. Proc Nat1 Acad Sci USA 1970;67:1901-8. Vaquez R. Modifications of the Golgi apparatus in neurons of rat’s supraoptic nucleus, after treatment with colchicine and posterior exposition to cold. An Anat 1976;25:557-78. Thyberg J, Hinek A. Electron microscopic studies on embryonic chick spinal ganglion cells-in vitro effects of antimicrotubular agents of Golgi complex. J Neurocytol 1977;6:27-39. Hinek A, Thyberg J, Friberg V. Electron microscopic studies on embryonic chick spinal ganglion cells: relationship between microtubules and Golgi complex. J Neurocytol 1977;6:13-27. Hoffstein S, Goldstein IM, Weissmann G. Role of microtubule assembly in lysosomal enzyme secretion from human polymorphonuclear leukocytes. A reevaluation. J Cell Biol 1977; 73:242-56. Glickman RM, Perrotto JL, Kirsch K. Intestinal lipoprotein formation: effect of colchicine. Gastroenterology 1976; 70:347-52. Arreaza-Plaza CA, Bosch U, Otayek MA. Lipid transport across the intestinal epithelial cell. Effect of colchicine. Biochim Biophys Acta 1976;431:297-302. Wilson L, Bamburg LR, Mizel SB, et al. Interaction of drugs with microtubule proteins. Fed Proc 1974:33:158-66. Reaven EP, Reaven GM. Distribution and content of microtubules in relation to the transport of lipid. An ultrastructural quantitative study of the absorptive cell of the small intestine. J Cell Biol 1977;75:559-72. Ockner RK, Hughes FF, Isselbacher KH. Very low density lipoproteins in intestinal lymph: origin, composition and role in lipid transport in the fasting state. J Clin Invest 1969; 48:2079-88. Pavelka M, Gang1 A. Die Ultrastruktur der Dunndarmepithelzelle nach Colchicinverabreichung-Untersuchungen im Niichternzustand und wiihrend Fettresorption. Verh Anat Ges

1978;72:687-9. 27. Pavelka M, Stockinger

L. The ultrastructure of rat jejunal enterocytes after colchicine application. In: Microscopic Society of Canada, ed. Proceedings of the Ninth International Congress on Electron Microscopy. Vol 2. Toronto: Imp. Press Lt., 1978:538-9. 28. Gang1 A, Pavelka M, Klose B. Evidence for functional significance of microtubules in intestinal transepithelial lipid trans-

March 1983

29.

30. 31. 32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

port. In: Peeters, ed. Protides of the biolog fluids. Oxford, New York: Pergamon Press, 1979;527-30. Wilson L, Friedkin M. The biochemical events of mitosis. I. Synthesis and properties of colchicine labeled with tritium in its acetyl moiety. Biochemistry 1966;5:2463-8. Weibel ER. Stereological principles for morphometry in electron microscopic cytology. Int Rev Cytol 1969;26:235-302. Loud AB. A method for the quantitative estimation of cytoplasmic structures. J Cell Biol 1962;15:481-7. Loud AB. A quantitative stereological description of the ultrastructure of normal rat liver parenchymal cells. J Cell Biol 1968;37:27-46. Folch J, Lees M, Sloane-Stanley GH. A simple method for the isolation and purification of total lipids from animal tissues. J Biol Chem 1957;226:497-509. Gang1 A, Ockner RK. Intestinal metabolism of plasma free fatty acids: intracellular compartmentation and mechanisms of control. J Clin Invest 1975;55:803-13. Regouw BJM, Cornelissen PJHC, Helder RAP, et al. Specific determination of free fatty acid in plasma. Clin Chim Acta 1971;31:187-95. Eggstein M, Kreutz FH. Eine neue Bestimmung der Neutralfette im Blutserum und Gewebe. I. Mitteilung. Prinzip, Dhrchfiihrung und Besprechung der Methode. Klin Wochenschr 1966;44:262-7. Cardell RR Jr, Badenhausen S, Porter KR. Intestinal triglyceride absorption in the rat: an electron microscopical study. J Cell Biol 1967;34:123-55. Jones AL, Ockner RK. An electron microscopic study of endogenous very low density lipoprotein production in the intestine of rat and man. J Lipid Res 1971;12:580-9. Palay SL, Karlin LJ. An electron microscopic study of the intestinal villus. II. The pathway of fat absorption. J Biophys Biochem Cytol 1959;5:373-84. Pfeiffer CJ, Rowden G, Weibel J. Gastrointestinal ultrastructure. An atlas of scanning and transmission electron micrographs. Stuttgart: Georg Thieme. Strauss EW. Morphological aspects of triglyceride absorption. Handbook of physiology. Vol 3 (sect 6). Washington, DC: American Physiological Society. 1968:1377-406. Schimmel RJ. Inhibition of free fatty acid mobilization by colchicine. J Lipid Res 1974:15:206-10.

COLCHICINE AND INTESTINAL LIPID TRANSPORT

555

43. Cenedella RJ, Crouthamel WC, Mengolf HF. Intestinal versus hepatic contribution to circulating triglyceride levels. Lipids 1974;9:35-42. 44. Robbins E, Gonatas NK. Histochemical and ultrastructural studies on HeLa cell cultures exposed to spindle inhibitors with special reference to the interphase cell. J Histochem Cytochem 1964;12:704-11. 45. Thyberg J, Piasek A, Moskalewski S. Effects of colchicine on the Golgi complex and GERL of cultured rat peritoneal macrophages and epiphyseal chondrocytes. J Cell Sci 1980;45:42-58. 46. Wunderlich F, Miiller R, Speth V. Direct evidence for a colchicine-induced impairment in the mobility of membrane components. Science 1973;182:1136-8. 47. Ukena TE, Berlin RD. Effect of colchicine and vinblastine on the topographical separation of membrane functions. J Exp Med 1972;136:1-7. 48. Pavelka M, Ellinger A. Effect of colchicine on the biosynthesis of plasma membrane glycoproteins in rat intestinal epithelial cells as visualized by radioautography with 3H-fucose (abstr). Eur J Cell Biol 1980;22:259. 49. Pavelka M, Ellinger A. Effect of colchicine on the Golgi apparatus and on GERL of rat jejunal absorptive cells. Ultrastructural localization of thiamine pyrophosphatase and acid phosphatase activity. Em J Cell Biol 1981;24:53-61. 50. Whaley W. The Golgi apparatus. In: Cell Biology Monograph. Vol 2. Vienna, New York: Springer Verlag, 1975. 51. Whaley WG, Dauwalder M. The Golgi apparatus, the plasma membrane, and functional integration. Int Rev Cytol 1979; 58:199-245. 52. Sabesin SM. Ultrastructural aspects of the intracellular assembly, transport and exocytosis of chylomicrons by rat intestinal absorptive cells. In: Rommel K, Goebell H, Bohmer R, eds. Lipid absorption. Biochemical and clinical aspects. Lancaster, England: MTP Press Ltd. St. Leonhard’s House, 113-48. 53. Sabesin SM, Frase S. Electron microscopic studies of the assembly, intracellular transport and secretion of chylomicrons by rat intestine. J Lipid Res 1977;18:496-511. 54. Stadler J, Franke WW. Characterization of the colchicine binding of membrane fractions from rat and mouse liver. J Cell Biol 1974;60:297-303.