- Email: [email protected]
Oxidative Metabolism of Foreign Compounds in Rat Small Intestine: Cellular Localization and Dependence on Dietary Iron
70:1063-1070,1976 Copyright © 1976 by The Williams & Wilkins Co.
Vol. 70, No.6
GASTROENTEROLOGY
Printed in U.S.A.
OXIDATIVE METABOLISM OF FOREIGN COMPOUNDS IN RAT SMALL INTESTINE: CELLULAR LOCALIZATION AND DEPENDENCE ON DIETARY IRON HARALD HOENSCH, M.D.,
C.
H.
Woo,
M.S., STEVEN
B.
RAFFIN, M.D., AND RUDI SCHMID,
M.D.
Department of Medicine, University of California, School of Medicine, San Francisco, California
Oxidative metabolism of foreign compounds was measured in the intestinal mucosa of male rats. Activities of benzpyrene hydroxylase, p-nitroanisole O-demethylase, and NADPH-cytochrome P-450 reductase and cytochrome P-450 content were 3 to 10 times higher in epithelial cells of the upper villus than in mucosal crypt cells. Villous tip cells of the upper small intestine exhibited much higher cytochrome P-450 content and drug-metabolizing enzyme activity than did tip cells oflower intestinal segments. In rats fed commercial.chow diet, cytochrome P-450 content and drug-metabolizing enzyme activity in villous tip cells of duodenal mucosa were higher than in animals fed a semisynthetic diet, but cytochrome b 5 and NADPH-cytochrome P-450 reductase were unaffected. On restriction of dietary iron intake, cytochrome P-450 and oxidative enzyme activity fell sharply, but were completely restored in 24 hr by oral iron supplementation, whereas parenteral iron administration was ineffective. These findings suggest that intestinal drug metabolism is localized primarily in the upper villous cells of the proximal intestinal mucosa, that cytochrome P-450 is synthesized in maturing epithelial cells as they migrate from the crypts to the tip of the mucosal villi, and that this process is dependent critically upon absorption of iron from the intestinal lumen. A large number of drugs, carcinogens, environmental contaminants, and potentially toxic food constituents is metabolized in the body to derivatives that are biologically less active or less toxic but more water-soluble than the parent compounds. 2 Oxidative reactions catalyzed by microsomal enzymes play a prominent role in this process. 2, 3 Many of these microsomal enzymes represent mixed function monooxygenase systems consisting of cytochrome P-450, which serves as a terminal oxidase, and NADPH-cytochrome P-450 reductase, which catalyzes the reduction of the oxidized cytochrome 4 ; in most Received October 14, 1975. Accepted December 4, 1975. This paper was presented in part at the Annual Meeting of the American Gastroenterological Association, San Antonio, Texas, May, 1975. 1 Address reprint requests to: Dr. Rudi Schmid, University of California, San Francisco, HSW 1120, San Francisco, California 94143. This work was supported in part by Research Grants AM 17365 and AM 11275 from the National Institutes of Health and by the Walter C. Pew Fund for Gastrointestinal Research. Dr. Hoensch was a fellow of the Paul Martini Foundation, Frankfurt, Germany. His present address is: Department of Medicine, Gastroenterology Unit, University of Tiibingen, Tiibingen, Germany. The authors wish to thank Dr. Lloyd L. Brandborg for the histological examinations of the small intestine, and Dr. Myron Pollycove for the serum iron determinations. Ms. Lydia Hammaker, M.S., and Drs. D. Montgomery Bissell, Maria A. Correia, and Philip Guzelian provided help and advice in several aspects of this study, and Ms. Gail Persson gave expert editorial assistance.
instances the cytochrome reductase appears to be the rate-limiting enzyme. 5 In many of these oxidative reactions the liver exhibits the highest enzyme activity per unit weight; however, other tissues such as lung, skin, and intestine also have been shown to contain monooxygenase activity.6-11 These organs constitute the portal of entry for a variety of foreign compounds, and the small intestine, in particular, is exposed to many biologically active and potentially harmful substances that are ingested with food. For example, Wattenberg has suggested that the intestinal aryl hydrocarbon hydroxylase system may act as a first line of defense against dietary carcinogens such as polycyclic aromatic hydrocarbons. 12 This and other intestinal oxidative enzyme systems have been shown to be stimulated by ingestion of aromatic hydrocarbons, 13, 14 phenobarbital,15 charcoal-broiled meat, or leafy vegetables. 16. 17 The active agents in vegetables appear to be indoles,18 which are naturally occurring plant constituents. Whereas activity of intestinal drug-metabolizing enzymes can readily be measured in microsomes prepared from whole intestinal mucosa, 19 spectrophotometric determination of cytochrome P-450 in intestinal mucosa was found to be more difficult. 20 This probably was due to several factors, including the relatively low cytochrome content of whole and particularly of distal intestinal mucosa,21, 22 contamination with hemoglobin 21 and presence of cytochrome P_420.20 This technical
1063
1064
HOENSCH ET AL.
problem was minimized in the present investigation by selective use of duodenal mucosa and by segregation of the mucosal cells into individual cell populations of reasonably uniform maturity. Intestinal mucosa is a rapidly proliferating tissue, in which relatively undifferentiated, dividing cells of the basal crypts progressively mature as they migrate to the upper villous stratum. 23 For several enzyme systems, marked differences have been shown between proliferating crypt cells and more mature absorptive cells of the mucosal villus. 24 - 27 Recently it has been documented that drug-metabolizing enzyme activity and cytochrome P-450 content of duodenal mucosa are located primarily in the epithelial cells of the upper villus. 22 This made it possible to use epithelial cells from the villous tip to investigate the effects of dietary factors and particularly of dietary iron on intestinal cytochrome P-450 content and oxidative drug metabolism.
Materials and Methods Animals and diets. All experiments were carried out in adult male rats (300 to 400 g) purchased from Charles River Breeding Laboratories, Inc., Wilmington, Mass., and certified to be "specific pathogen" (Bartonella)-free. The animals were housed in stainless steel cages in a temperature- and light-controlled room, and were protected from contact with pesticides. The following diets were used: commercial rat chow diet (Berkeley Diet, Feed Stuffs Processing Co., San Francisco, Calif.) , containing ground vegetables, vegetable oil, dried yeast, vitamins, and trace minerals, including 20 mg of iron per 100 g of diet; semisynthetic iron-supplemented diet (Nutritional Biochemicals Corp., Cleveland, Ohio), consisting of vitamin test casein, 27%; corn starch, 55%; hydrogenated vegetable oil, 14%; salt mixture without ferric phosphate, 3%; vitamin mixture, 1%; and ferrous sulfate, 8 mg of iron per 100 g of diet (iron content of the semisynthetic diet up to 40 mg of iron per 100 g of diet did not increase the content of intestinal cytochrome P-450); and semisynthetic low iron diet, containing all of the above ingredients except ferrous sulfate, resulting in an overall iron content of less than 0.1 mg per 100 g of diet. In some experiments, semisynthetic diets were supplemented with 0.05% (w/w) 3-methyl-cholanthrene. In one experiment, rats were given intramuscular injections of 0.1 ml of Imferon (Lakeside Laboratories, Milwaukee, Wis.) containing 50 mg of iron per m!. All animals were fed ad libitum, had free access to water, and gained weight at similar rates regardless of the dietary regimen. Unless stated otherwise, the rats were kept on a specific diet for at least 7 days before being killed. At approximately 9:00 AM of the day of experiment the animals were anesthesized with ether and exsanguinated by aortic puncture. All subsequent procedures were conducted at 4 DC. Preparation of intestinal cell fractions. Beginning at the pylorus, consecutive 15-cm segments of small intestine were excised, perfused with ice-cold 1.15% (w/v) KCI, split longitudinally along the mesenteric border, and opened. Villous tip, lower villous, and crypt cells of the intestinal mucosa were separated by a modification of the differential scraping technique. 28 Villous tip cells were removed by light hand scraping with the blade of a metal spatula; slightly increased scraping pressure yielded lower villous cells. Crypt cells were harvested by abrasion of the remaining mucosa. Histological examination of the separated cell fractions indicated that they consisted of reasonably uniform cell
Vol. 70, No.6
populations. The discriminatory effectiveness of the scraping procedure was further evaluated by determining the incorporation of 3H-thymidine or 59Fe into the individual cell fractions in vivo. An intraperitoneal injection of 50 J.l.c of 3H-thymidine (specific activity 6.7 c per mmole, New England Nuclear Corp., Boston, Mass.) was administered to each of 9 rats fed the semisynthetic iron-supplemented diet. Groups of 3 rats were killed 1, 24, and 48 hr later, and the duodenal mucosa was fractionated by scraping. Individual cell fractions from 3 animals were pooled and homogenized in 0.25 M sucrose solution, the· cell nuclei were sedimented by centrifugation at 900 g for 10 min, DNA was precipitated with cold 10% trichloroacetic acid, and the radioactivity of the preparation was measured in a liquid scintillation spectrophotometer (model LS-250, Beckman Instruments, Inc., Fullerton, Calif.). Results were expressed as counts per minute per milligram of nuclear protein. In confirmation of previous reports, 23. 26 after 1 hr most of the label appeared in the crypt cell fraction, whereas after 48 hr, villous tip cells exhibited the highest radioactivity (table 1). Three rats fed the semisynthetic iron-supplemented diet and 3 rats on semisynthetic low iron diet for 2 days were each given an intravenous injection of 50 J.l.c of 59Fe (specific activity 1.71 c per mmole, ICN Isotope and Nuclear Division, Cleveland, Ohio). The radioactive ferrous citrate was preincubated with 0.5 ml of rat serum and injected in a single pulse into a femoral vein. The animals were killed 4 hr later and the radioactivity of homogenized individual mucosal cell fractions was determined in a well-type scintillation detector (model 8275, Nuclear Chicago Corp., Des Plaines, Ill.). Results were expressed as counts per minute per milligram of cellular protein (table 2). As expected from previous autoradiographic observations,29 uptake of radioactive iron was highest in the epithelial crypt cells whereas the villous tip cell fraction contained much less radioactivity (table 2). No difference was found between animals on iron-supplemented and low iron diet. The results of these tracer experiments and the histological findings indicate TABLE
1. Consecutive incorporation of 3H-thymidine into cell fractions
of duodenal mucosa" Time after administration (hr)
Cell fraction
24
48
cpm/mg nuclear protein
Villous tip Lower villous Crypt
174 454 2822
320 808 1497
908 789 372
"Rats were given intraperitoneal injections of 3R-thymidine at time zero and killed after the intervals indicated. TABLE
2. Differential incorporation of '·Fe into separated cell fractions
of duodenal mucosa" Cell fraction
Semisynthetic iron-supplemented diet
Semisynthetic low iron diet
cpm/mg cellular protein
Villous tip Lower villous Crypt
123 ( 70-155) 278 (149-346) 577 (509-635)
130 ( 90-170) 194 (107-280) 500
(442-558)
" Three rats were given intravenous injections of '·Fe and killed 4 hr later. Values represent mean and range of individual rats.
June 1976
that the manual scraping procedure yielded mucosal cell fractions of acceptable functional and morphological uniformity. Preparation of microsomes. Individual cell fractions were weighed and homogenized in 0.1 M potassium phosphate buffer, pH 7.4, using 15 strokes of a Potter-Elvehjem teflon glass homogenizer. The homogenate was then sonicated for 25 sec at 35 watts (model W185 D, Sonifier Cell Disruptor, Heat, Systems-Ultrasonics, Inc., Plainview, N. Y.) and sedimented at 20,000 x g for 10 min in a Sorval supers peed automatic refrigerated centrifuge (model RC2-B, Ivan Sorval, Inc., Norwalk, Conn.). The supernatant was centrifuged at 105,000 x g for 60 min, the microsomes were washed once with 1.15% aqueous KCI to remove contaminating hemoglobin, and the pellet was resuspended in 0.1 M potassium phosphate buffer in a final volume equivalent to 300 mg of wet mucosal tissue per ml. The microsomal suspensions of villous tip, lower villous, and crypt cells contained approximately 3 to 4 mg of protein per ml and the yield of microsomal protein was 10 to 15 mg per g wet weight of tissue for each cell fraction. In additional experiments, equal volumes of microsomal suspensions obtained from tip and crypt cell fractions were mixed and incubated for 90 min at room temperature 2 • and cytochrome P-450 content was measured. Also, microsomal preparations of tip cells were incubated for 90 min at room temperature or at 4 DC with a 105,000 x g supernatant fraction of tip or crypt cell homogenate, and cytochrome P-450 content and benzpyrene hydroxylase activity were then determined. Analytical procedures. Cytochromes P-450 and b. were determined in microsomal suspensions by the method of Omura and Sato,3. benzpyrene hydroxylase activity by the method of Wattenberg et al.I< as modified by Kuntzman et al. 31 and Alvares et al,32 and p-nitroanisole O-demethylase activity by the method of Netter and Seidel. 33 NADPH-cytochrome P-450 reductase was measured by the procedure of Williams et al. 34 Protein concentration was estimated by the procedure of Lowry et al. 3. using human albumin as standard. Serum iron concentration was measured by the method of Stookey."·
Results On fractionation of duodenal mucosa into villous tip, lower villous, and crypt cells, striking differences in cytochrome P-450 content and drug-metabolizing enzyme activity were observed (table 3). Thus, in rats fed standard rat chow diet, cytochrome P-450 content of tip cell fractions was 10 times higher than in crypt cells, and for NADPH-cytochrome P-450 reductase the difference was 3-fold. By contrast, cytochrome b s content of crypt cells was only slightly less than that of villous tip cells. Benzpyrene hydroxylase and p-nitroanisole O-demethTABLE
Cell fraction
ylase activity showed a distribution pattern similar to that of cytochrome P-450 (table 3), with lower villous cell fractions exhibiting enzyme activity intermediate between that of villous tip and crypt cells. These differences in cytochrome P-450 content and enzyme activity between villous tip and crypt cells could not be attributed to the presence of inhibitors or activators, as on incubation of microsomal mixtures prepared from these cell fractions the results obtained were those predictable from the constituent components of the mixture. Moreover, when microsomal preparations of villous tip cells were incubated with 105,000 x g supernatant of crypt cell homogenate, no changes in cytochrome P-450 content and benzpyrene hydroxylase activity were observed. The previously noted occurrence of an absorptive peak at 420 nm in the carbon monoxide-reduced difference spectrum of intestinal microsomes 2o , 21 was confirmed, but its magnitude was markedly reduced by the preparative procedures employed for removal of contaminating hemoglobin. Thus, the animals were thoroughly exsanguinated, residual blood was removed from the small intestine by perfusion of intestinal arteries with cold isotonic saline, and the microsomal pellet was washed with 1.15% M KCI solution. With these precautions, only microsomes of crypt cells exhibited a demonstrable, albeit minor, absorption peak in the 420 nm region . The present findings confirm the previously reported observation that monooxygenase activity is higher in the mucosa of the upper small intestine than in that of more distal segments of the bowel. 14, 37 Thus, in microsomes of villous tip cells, activity of benzpyrene hydroxylase (fig. 1) and content of cytochrome P-450 (fig. 2) progressively decreased from duodenum to ileum. A similar profile of activity was observed with p-nitroanisole O-demethylase and NADPH-cytochrome P-450 reductase. Because of the reported effect of dietary constituents on intestinal drug-metabolizing monooxygenase activity, 12, 16, 18 most. present experiments were performed with semisynthetic iron-supplemented diet of defined composition. Villous tip cells of rats on this diet exhibited distinctly lower cytochrome P-450 content and benzpyrene hydroxylase and p-nitroanisole O-demethylase activity than did those of animals on commercial chow diet (table 4). When the semisynthetic iron-containing diet was supplemented with 3-methylcholanthrene for 4 days, cytochrome P-450 content, benzpyrene hydroxylase, and p-nitroanisole O-demethylase activi-
3. Microsomal hemoproteins and drug metabolism in different cell fractions of duodenal mucosa of rats" Cytochrome P·450
Cytochrome b,
Benzpyrene hydroxylase
0.115 0.070 0.011
± ± ±
0.007 (9) 0.004 (3) 0.002 (3)
0.119 0.088 0.082
p·Nitroanisole O·demethylase
nmoles/mg protein/hr
nmoles/mg protein
Villous tip Lower villous Crypt
1065
DRUG METABOLISM IN SMALL INTESTINE
± ± ±
0.004 (9) 0.007 (3) 0.001 (3)
6.5 3.4 1.4
± ± ±
0.6 (9) 0.1 (3) 0.4 (3)
15.5 ± 1.1 (4) 7.6 ± 0.6 (3) Trace
NADPH-cytochrome P-450 reductase nmol.s/mg protein / min
133.8 74.5 43.1
± ± ±
5.2 (3) 2.6 (3) 7.8 (3)
" Values represent mean ± standard error of mean of individual experiments using mucosal cell fractions pooled from 4 rats on commercial rat chow diet. NADPH-cytochrome P·450 reductase was determined in cell fractions from individual rats. Number of individual experiments are given in parentheses. All values are significantly different from those of villous tip cells at the P < 0.01 level.
1066
Vol. 70, No.6
HOENSCH ET AL.
ties were significantly increased in villous tip cells of duodenal mucosa, but cytochrome b s and NADPH-cytochrome P-450 reductase remained unaffected (table 4). Restriction of oral iron intake resulted in a prompt decrease of cytochrome P-450 content and benzpyrene hydroxylase activity in villous tip cells (table 5, fig. 3). Within 2 days on the semisynthetic low iron diet, both values fell to less than one-half and then leveled off on continued dietary iron restriction (table 5). On the other hand, cytochrome b s did not change significantly over 30
days of iron deprivation. NADPH-cytochrome P-450 reductase activity also was not affected by feeding the low iron diet for 2 days. In rats fed the low iron diet, serum iron concentration fell from 202 ± 7 (mean ± SEM) Jlg per 100 ml to 137 ± 5 Jlg within the first 24 hr and further decreased on iron deprivation for 30 days (table 5). The striking reduction of cytochrome P-450 content and benzpyrene hydroxylase activity of villous tip cells' 0.15
6
Z ii:i
>-
o
5 Z ii:i >-
0
...'"
4
~
3
"0 0V> E 0 v
2
0>
.....E ~
~
G
~
~
'"
CYTOCHROME P·450
'"
BENZPYRENE HYDROXYlASE
0 .1 0
0..
~ ~
oV> o
'"v ~
. ~
.....
0 .05
G
i
'0
E ~
15
30
45
60
75
90
105
120
15
SMAll INTESTINE (em beyond pylorus)
FIG. 1. Distribution of benzpyrene hydroxylase activity in villous tip cells of sequential segments of small intestine of rats fed commercial chow diet. Vertical bars represent standard error of mean of three individual experiments. TABLE 4.
30
45
60
75
90
105
120
SMAll INTESTINE lem beyond pylorus)
FIG. 2. Distribution of cytochrome P-450 content in villous tip cells of sequential segments of small intestine of rats fed commercial chow diet. Vertical bars represent standard error of mean of three individual experiments.
Effects of diet s and feeding of 3-methylcholanthrene on microsomal hemoproteins and drug metabolism in villous tip cells of duodenal mucosa of rats"
Diet
Cytochrome P·450
Benzpyrene hydroxylase
Cytochrome b,
nmoles/ mg protein
0.115 Commercial chow 0.067 Semisynthet ic iron·sup· plemented Semisynthetic iron·sup0.132 plemented with 50 mg 3-methyl-cholanthrene/1oo
nmoles/mg protein/hr
6.5 2.4
±
±
0.004 (9) 0.005 (9)
±
0.004 (4)
47.4
0.119 0.124
±
±
0.007 (9) 0.004 (12)"
±
0.010 (4)
0.097
±
p·Nitroanisole O·demethylase
NADPH·cytochrome P·450 reductase nmoles/ mg protein/ min
±
0.6 (9)" 0.2 (8)"
15.5 ± 1.1 (4)" Trace
133.8 ± 5.2 (3) 115.0 ± 9.4 (3)
±
2.8 (4)"
40.6 ± 1.5 (4)"
108.6 ± 6.6 (4)
g
"Values represent mean ± standard error of mean of individual experiments using villous tip cells pooled from 4 rats. In rats fed 3-methylcholanthrene for 4 days, determinations were made in samples from individual rats. Number of experiments is given in parentbeses. " Indicates that the value differs from the two other values at the P < 0.01 level. TABLE
5. Effect of dietary iron restriction on microsomal hemoproteins and benzpyrene hy droxylase activity in villous tip cells
of duodenal mucosa of rats" Days on low iron diet
Benzpyrene hydroxylase
Cytochrome b,
Cytochrome P·450
nmoles/hr/ mg protein
nmoles /mg protein
o 2 4
30
0.004 (l2) 0.067 0.060 ± 0.011 (3) 0.030 ± 0.003 (6)" 0.031 ± 0.003 (3)" 0.025 ± 0.003 (6)" ±
0.124 0.127 0.131 0.146 0.112
± ± ± ± ±
0.005 0.005 0.008 0.004 0.017
(9) (3) (7) (3) (4)
2.4 ± 0.2 (8) 1.3 ± 0.2 (3)" 0.4 ± 0.1 (3)" 0.7 ± 0.1 (3)" 0.8 ± 0.1 (7)"
Serum iron Ilg/lOO ml
202 137 147 148 101
± ± ± ± ±
7 (16) 5 (l0)" 4 (17)" 4 (9)" 5 (10)"
" Values represent mean ± standard error of mean of individual experiments using villous tip cells pooled from 4 rats. Number of experiments is given in parentheses. b Values differ significantly from those of day 0 at the P < 0.01 level.
June 1976
1067
DRUG METABOLISM IN SMALL INTESTINE
on brief restriction of oral iron intake was completely reversible by dietary iron supplementation for 24 hr (fig. 3). By contrast, parenteral iron adminstration was without effect on intestinal cytochrome P-450, despite a sharp rise in serum iron concentration (fig. 4). In rats on low iron diet for 2 days, intramuscular injection of two or four individual doses of 5 mg of iron each raised the serum iron concentration from 147 ± 4 to 717 ± 22 and 735 ± 30 Jlg per 100 ml, respectively, but failed to alter the cytochrome P-450 content of villous tip cells (fig. 4). Feeding of the low iron diet also depressed the response of intestinal cytochrome P-450 to the stimulatory effect of dietary 3-methyl-cholanthrene. In rats fed 0.05% 3-methylcholanthrene (w/w) in semisynthetic diet for 4 consecutive days, restriction of oral iron intake during the last 48 hr of the experiment reduced cytochrome P-450 content to less than one-half of the value in 3-methylcholanthrene-fed animals on iron-supplemented diet (fig. 5). On the other hand, dietary iron restriction failed to reduce the 3-methylcholanthrenestimulated activity of benzpyrene hydroxylase and pnitroanisole O-demethylase (fig. 5). Activity of NADPHcytochrome P-450 reductase of villous tip cells remained unchanged when rats were given oral 3-methylcholanthrene and fed either the semisynthetic iron-supplemented or low iron diet.
cerned primarily with the metabolism of foreign compounds that are taken up from the intestinal lumen. 9, 1214, 16 It is methodologically difficult, however, to determine in vivo to what extent ingested substances actually are metabolized in the intestine itself rather than being absorbed intact and then undergoing biotransformation on their first pass through the liver.38 Nonetheless, preliminary observations with tetrahydrocannabinol,39 phenacetin,40 and chlorpromazine 41 in humans and rats suggest that intestinal transformation plays a major role in the oxidative metabolism of these drugs. Since many of these lipophilic foreign compounds appear to be absorbed mainly in the proximal small intestine,42 it is not surprising that drug-metabolizing enzyme activity in mucosal homogenates 14 or microsomal preparations is highest in the duodenum and jejunum (fig. 1). The present findings that drug-metabolizing monooxygenase systems and their constituent component cytochrome P-450 are localized primarily in the absorptive cells of the upper mucosal villus (table 3) provide further evidence that ingested foreign compounds may undergo biotransformation at the very site of their absorption. Thus, aryl hydrocarbon hydroxylase and O-demethylase activity and cytochrome P-450 content were much higher in epithelial cells of the villus than in the immature cells of the crypt region. This observation implies that these enzyme systems are formed as the epithelial cells differentiate in the course of their migraDiscussion tion from the mucosal crypts to the tip of the villous Recent observations have indicated that in addition to structure. 23 This process of "enzyme maturation" apwell defined absorptive and transport functions, the pears to be controlled in part by dietary constituents mucosa of the small intestine contains a number of that are absorbed and presumably'serve as substrates for enzyme systems that catalyze metabolic transforma- these enzyme systems. Thus, benzpyrene hydroxylase tions of many foreign compounds. 3, 12, 20, 21, 37 Whereas and p-nitroanisole O-demethylase activity and cytoin this regard intestinal mucosa qualitatively resembles chrome P-450 content of villous tip cells significantly the liver, the intestinal enzymes presumably are con- decreased when semisynthetic iron-supplemented diet
~
~
....w
CYTOCHROME P-450
0.10
...'"...
«
'" '" !::!
«
l: 0 0
'"
0
l: 0 0
...w 0 ...'"...
0.05
l:
'" !::!
'~
2.5
r-----1
2.0
200
1.5
'"
E
-;:-
1.0
~
';;..!
Q
'"E
BENZPYRENE HYDROXYLASE ACTIVITY
l:
137
0
E c:
147
...... ..!
3.0
0.5
2
147
~
0
E c:
0
f--lO~I:~ON~ I - - IRON~I~~PPl'---t
I 0
I 2
I 3 DAYS
I
"
I 5
0
~OWD::~N---l ~ON~I~~PPl---1
I 0
I
2
3 DAYS
"
I 5
FIG. 3. Effects of dietary iron restriction and oral iron supplementation on cytochrome P-450 content and benzpyrene hydroxylase activity in villous tip cells of rat duodenal mucosa. Vertical bars represent standard error of mean of three to five individual experiments using villous tip cells pooled from 4 rats. Numbers indicate average serum iron, p.g per 100 m!.
was substituted for standard rat chow (table 4), but increased strikingly when 3-methylcholanthrene was added to the semisynthetic diet (table 4)_ Previous reports already had indicated that dietary factors 12, 14, 16 or supplementation of diet with certain drugs 9 affect oxidative enzyme systems involved in the metabolism of foreign compounds in whole mtestinal mucosa_ The present findings suggest that these observed alterations in the whole mucosa may have reflected largely changes of enzyme activity in the mucosal villous tip cells. In analogy with observations in the liver,43 it has been assumed that this diet- or drug-mediated stimulation of enzyme activity in the intestinal mucosa represents
~
w
0
137
:;( ~
0
0.04
7?~0
CYTOCHROME P-450
i
'"E ---
,,,,
INTRAMUSCULAR IRON
0.06
'"0..
0
0
735
147 0.02
~ 0
E c
Vol. 70, No.6
HOENSCH ET AL.
1068
Ll_
~-
0 DAYS ON lOW IRON DIET
FIG. 4. Effects of dietary iron restriction and parenteral iron administration on cytochrome P-450 content in villous tip cells of rat duodenal mucosa. Vertical bars represent standard error of mean of three to five individual experiments. For the 3rd day, individual values are given of two pairs of experiments each based on analysis of pooled villous tip cells from 4 rats. Five milligrams of iron as Imferon were given intramuscularly at times indicated by arrows. Numbers indicate average serum iron, J.l.g per 100 m!. Cytochrome
BenzpYfene
p-N itroanisol e
P-4S0
hydroxylase
O-demethylase
0.15
50 w
'" 0..
0
'"0.. ~
<
0.10
~
0
'" i
'" ---!
'"
0
E c
---
""
:<~ 0
30
E
0.05
'"0..
~O
U
i
E
:<~
0
W
0
I-
I-
20
150 -
~
w
0
~
NADPH-cytochrome P-4S0 reductase
50
~
~
w
enzyme induction. 37 Although this appears a likely explanation, mechanistic interpretation of these findings is complicated by the fact that intestinal mucosa is an actively proliferating tissue with highest enzyme activity residing in the villous tip cells, which, while fully differentiated,23 have a very short life span. 44 Because of this rapid cell turnover, it is difficult to procure the stringent experimental evidence that would be required' to establish enzyme induction, and additional studies are necessary to clarify this point. Since cytochrome P-450 content is 10 times higher in villous tip cells than in crypt cells (table 3), it is likely that most of this iron-containing hemoprotein is synthesized or assembled during the migration of these epithelial cells from the mucosal crypts to the villous surface. This raises the question regarding the source of the iron required for the synthesis of this cytochrome in the upper villous stratum. Prevous investigations with autoradiographic techniques have demonstrated that parenterally administered radioiron initially appears predominantly in the mucosal crypt region,29 whereas orally ingested radioiron is taken up preferentially by the absorptive cells of the mucosal villi.45 This has been confirmed in the present study, in that intravenously injected 59Fe 4 hr after administration was concentrated largely in the crypt cell fraction whereas tip cell preparations exhibited a much lower isotope content (table 2). As the radioiron-laden crypt cells differentiate and migrate along the villous structure, the radioiron eventually is carried to the top of the villus. 29 That this source of iron apparently is insufficient for the synthesis of cytochrome P-450 in the ascending epithelial cells is shown by the data in table 5 and figure 3. In rats placed on a virtually iron-free diet, cytochrome P-450 content of villous tip cells fell within 48 hr to less than one-half of control but was completely restored by oral iron supplementation for 24 hr (fig_ 3). Restriction of oral iron for up to 13 days in adult rats is reported not to produce appreciable depletion of body iron stores, nor to lead to demonstrable
I
'" i
I-
0 40
'" 0..
30
U
E '" ---
""
20
-! 0
E c
10
FIG. 5. Cytochrome P-450 content and activity of benzpyrene hydroxylase, p-nitroanisole O-demethylase, and NADPH-cytochrome P-450 reductase in villous tip cells of duodenal mucosa of rats on semisynthetic iron-supplemented diet (shaded column), semisynthetic iron-supplemented diet with added 3-methylcholanthrene (open column), and semisynthetic low iron diet with 3-methylcholanthrene (cross-hatched column). Vertical bars represent standard error of mean of three to five individual experiments.
June 1976
1069
DRUG METABOLISM IN SMALL INTESTINE
changes in plasma iron turnover.46 This also is shown by the present observation that in rats fed the low iron diet for 48 hr, incorporation of parenterally administered radioiron into mucosal crypt cells did not differ significantly from that of control animals (table 2). Whereas the present findings with iron-free diet conceivably could result from accelerated mucosal cell turnover or from destabilization of cytochrome P-450 caused by the iron restriction, there is no supportive evidence for these explanations. More likely is that absorption of iron from the intestinal lumen is essential for the synthesis of cytochrome P-450 in the villous epithelial cells. This is further substantiated by the ineffectiveness of intramuscular iron injections to raise the mucosal cytochrome P-450 content in animals that had been on restricted iron intake for 48 hr (fig. 4). The finding that on the low iron diet cytochrome P-450 fell sharply during the initial 48 hr but then leveled off for the remainder of the 30-day observation period (table 5) may indicate the limited extent to which body iron is available for cytochrome P-450 synthesis in the epithelial villous cells. Since these diet-induced alterations in mucosal cytochrome P-450 content were accompanied by comparable changes in benzpyrene hydroxylase (table 5, fig. 3), but not in NADPH-cytochrome P-450 reductase activity, it is possible that under these experimental conditions, cytochrome P-450 may be rate-controlling for the aryl hydrocarbon hydroxylase reaction. The present findings raise several questions to which answers are not readily available. It previously had been shown that in rats made iron-deficient, cytochrome P -450 and oxidative drug metabolism of the liver are not reduced. 47 , 48 In fact, the activity of several hepatic drug-metabolizing enzymes and of NADPH-cytochrome P-450 reductase tends to rise after induction of iron deficiency. 47 This was found to be associated with increased plasma turnover of aniline and aminopyrine, 47 suggesting that the hepatic metabolism of these compounds may be enhanced in vivo. This response of the liver to iron deficiency is strikingly different from the present findings in the gut. Whereas the effect of chronic iron deficiency on intestinal cytochrome P-450 and drug metabolism was not investigated, it is evident that brief restriction of intestinal iron absorption resulted in a rapid albeit fully reversible fall of mucosal cytochrome P-450 content and monooxygenase activity. It is unclear whether this difference between liver and gut reflects dissimilar regulatory mechanisms or may be related to the slow cell turnover in the liver, permitting more effective conservation of cellular iron. Marked differences also were noted in the intestinal mucosa between cytochrome P-450 and cytochrome b s ; the role of the latter cytochrome in oxidative drug metabolism is still controversial. 49 The content of cytochrome b s in villous tip cells was slightly higher than that in crypt cells (table 3), but the gradient from crypts to villous tips was much less than for cytochrome P-450. This suggests that most of the mucosal cytochrome b s is synthesized in the proliferating crypt cells and is carried within the differentiating epithelial cells to the upper
villous stratum. Moreover, neither dietary iron restriction nor supplementation of the diet with 3-methylcholanthrene affected the mucosal cytochrome b s content (table 4). This difference in behavior between the cytochromes b s and P-450 of the intestinal mucosa is reminiscent of comparable findings in the liver; in developing rat liver, cytochrome b s tends to appear earlier than cytochrome P-450, so, S1 and its turnover in adult rat hepatocytes is considerably slower than that of cytochrome P-450. 2 The observed decrease of mucosal benzpyrene hydroxylase activity during dietary iron restriction (table 5, fig. 3) resembles that found in the fasting animal '2 and suggests that the reduced level of cytochrome P-450 (fig. 3) might preclude an adaptive increase of intestinal oxidative drug metabolism in response to ingestion of stimulatory compounds. That this is not the case is shown by the data in figure 5, which demonstrate that in rats fed the low iron diet and 3-methylcholanthrene, benzpyrene hydroxylase and p-nitroanisole O-demethylase activity markedly rose despite the low cytochrome P-450 level. By contrast, neither low iron diet nor 3-methylcholanthrene affected NADPH-cytochrome P-450 reductase activity (fig. 5). These findings indicate that on stimulation with 3-methyl-cholanthrene, cytochrome P450 could not be rate-controlling for the activity of intestinal benzpyrene hydroxylase and p-nitroanisole O-demethylase. The differences in ratio between intestinal drug-metabolizing enzymes, NADPH-cytochrome P-450 reductase, and cytochrome P-450 in rats fed semisynthetic iron-supplemented diet with 3-methylcholanthrene or semisynthetic low iron diet with 3-methylcholanthrene illustrate the difficulties that are encountered in defining the rate-limiting factors in oxidative drug metabolism. 2 REFERENCES 1. Hoensch HP, Woo CH, Schmid R: Cytochrome P-450 in duodenal mucosa: dependence on dietary (luminal) iron (abstr). Gastroenterology 68:A-55/912, 1975 2. Remmer H: Induction of drug metabolizing enzyme system in the liver. Eur J Clin Pharmacol 5:116-136, 1972 3. Hartiala K: Metabolism of hormones, drugs and other substances by the gut. Physiol Rev 53:496-534, 1973 4. Estabrook RW, Matsubara T, Mason JL, et al: Studies on the molecular function of cytochrome P-450 during drug metabolism. In Microsomes and Drug Oxidations. Edited by RW Estabrook, JR Gillette, and KC Leibman. Baltimore, Williams & Wilkins Company, 1973, p 98-110 5. Gillette JR, Davis DC, Sasame HA: Cytochrome P-450 and its role in drug metabolism. Ann Rev Pharmacol 12:57-84, 1972 6. Bend JR, Hook GE, Gram TE: Characterization of lung microsomes as related to drug metabolism. Drug Metab Dispos 1:358-366, 1973 7. Neal RA: A comparison of the in vitro metabolism of parathion in the lung and liver of the rabbit. Toxicol Appl Pharmacol 23:123-130, 1972 8. Feuer G, Sosa-Lucero JC, Lumb G: Failure of various drugs to induce drug-metabolizing enzymes in extrahepatic tissues of the rat. Toxicol Appl Pharmacol 19:579-589, 1971 9. Lake BG, Hopkins R, Chakraborty J, et al: The influence of some hepatic enzyme inducers and inhibitors on extrahepatic drug
1070
HOENSCH ET AL.
metabolism. In Microsomes and Drug Oxidations. Edited by RW Estabrook, JR Gillette, and KC Leibman. Baltimore, Williams & Wilkins Company, 1973, p 342-349 10. Bickers DR, Eiseman J, Kappas A: Microscope immersion oils: effects of skin application on cutaneous and hepatic drug-metabolizing enzymes. Biochem Pharmacol 24:779-783, 1975 11. Welch RM, Cavallito J, Loh A: Effect of exposure to cigarette smoke on the metabolism of benzo[a)pyrene and acetophenetidin by lung and intestine of rats. Toxicol Appl PharmacoI23:749-758, 1972 12. Wattenberg LW: Dietary modification of intestinal and pulmonary aryl hydrocarbon hydroxylase activity. Toxicol Appl Pharmacol 23:741-748, 1972 13. Pantuck EJ, Hsiao KC, Kaplan SA, et al: Effects of enzyme induction on intestinal phenacetin metabolism in the rat. J Pharmacol Exp Ther 191:45-52, 1974 14. Wattenberg LW, Leong JL, Strand PJ: Benzpyrene hydroxylase activity in the gastrointestinal tract. Cancer Res 22:1120-1125, 1962 15. Lehrmann C, Ullrich V, Rummel W: Phenobarbital inducible drug monooxygenase activity in the small intestine of mice. Naunyn Schmiedebergs Arch Pharmacol 276:89-98, 1973 16. Pantuck EJ, Hsiao KC, Kuntzman R, et al: Intestinal metabolism of phenacetin in the rat: effect of charcoal-broiled beef and rat chow. Science 187:744-746, 1975 17. Hsiao KC, Pantuck EJ, Wattenberg LW, et al: Stimulatory effect of a brussel sprouts or cabbage diet on intestinal drug metabolism in the rat. Fed Proc 34:742, 1975 18. Loub WD, Wattenberg LW, Davis DW: Aryl hydrocarbon hydroxylase induction in rat tissues by naturally occurring in doles of cruciferous plants. J Nat! Cancer Inst 54:985-988, 1975 19. Chhabra RS, Fouts JR: Sex differences in the metabolism of xenobiotics by extrahepatic tissue in rats. Drug Metab Dispos 2:375-379, 1974 20. Chhabra RS, Pohl RJ, Fouts JR: A comparative study of xenobioticmetabolizing enzymes in liver and intestine of various animal species. Drug Metab Dispos 2:443-447, 1974 21. Zampaglione NG, Mannering GJ: Properties of benzpyrene hydroxylase in the liver, intestinal mucosa and adrenal of untreated and 3-methylcholanthrene-treated rats. J Pharmacol Exp Ther 185:676-685, 1973 22. Hoensch HP, Woo CH, Schmid R: Cytochrome P-450 and drug metabolism in intestinal villous and crypt cells of rats: effect of dietary iron. Biochem Biophys Res Commun 65:399-406, 1975 23. Imondi AR, Balis ME, Lipkin M: Changes in enzyme levels accompanying differentiation of intestinal epithelial cells. Exp Cell Res 58:323-330, 1969 24. Nordstrom C, Dahlqvist A: Quantitative distribution of some enzymes along the villi and crypts of human small intestine. Scand J Gastroenterol 8:407-416, 1973 25. Fortin-Magana R, Hurwitz R, Herbst JJ: Intestinal enzymes: indicators of proliferation and differentiation in the jejunum. Science 167: 1627 -1628, 1969 26. Weiser MM: Intestinal epithelial cell surface membrane glycoprotein synthesis. J Bioi Chern 248:2536-2541, 1973 27. Kim YS, Perdomo J, Ochoa P: Regional and cellular localization of glycosyltransferases in rat small intestine. Changes in enzymes with differentiation of intestinal epithelial cells. Biochim Biophys Acta 391:39-50, 1975 28. Dietschy JM, Siperstein MD: Cholesterol synthesis by the gastrointestinal tract: localization and mechanisms of control. J Clin Invest 44:1311-1327, 1965
Vol. 70, No.6
29. Conrad ME, Crosby WH: Intestinal mucosal mechanisms controlling iron absorption. Blood 22:406-415, 1963 30. Omura T, Sato R: The carbon monoxide-binding pigment of liver microsomes. J Bioi Chern 239:2370-2385, 1964 31. Kuntzman R, Mark LC, Brand L, et al: Metabolism of drugs and carcinogens by human liver enzymes. J Pharmacol Exp Ther 152: 151-156, 1966 32. Alvares AP, Schilling G, Garbut A: Studies on the hydroxylation of. 3,4-benzpyrene by hepatic microsomes. Biochem Pharmacol 19:1449-1455, 1970 33. Netter KJ, Seidel G: An adaptively stimulated O-demethylating system in rat liver microsomes and its kinetic properties. J Pharmacol Exp Ther 146:61-65, 1964 34. Williams CH, Kamin H: Microsomal triphosphopyridine nucleotide-cytochrome c reductase of liver. J Bioi Chern 237:587-595, 1962 35. Lowry OH, Rosebrough NJ, Farr AL, et al: Protein measurement with the folin phenol reagent. J Bioi Chern 193:265-275, 1951 36. Stookey LL: Ferrozine-a new spectrophotometric reagent for iron. Anal Chern 42:779-781, 1970 37. Scharf R, Ullrich V: In vitro induction by phenobarbital of drug monooxygenase activity in mouse isolated small intestine. Biochern Pharmacol 23:2127-2137, 1974 38. Gibaldi M, Perrier D: Route of administration and drug disposition. Drug Metab Rev 3:185-199, 1974 39. Greene ML, Saunders DR: Metabolism of tetrahydrocannabinol by the small intestine. Gastroenterology 66:365-372, 1974 40. Pantuck EJ, Kuntzman R, Conney AH: Decreased concentration of phenacetin in plasma of cigarette smokers. Science 175:1248-1250, 1972 41. Curry SH, D'Mello A, Mould GP: Destruction of chlorpromazine during absorption in the rat in vivo and in vitro. Br J Pharmacol 42:403-411, 1971 42. Knoefel PK: Absorption, chap 2, Absorption, Distribution, Transformation and Excretion of Drugs. Edited by PK Knoefel. Springfield, Ill., Charles C Thomas, 1972, p 39-55 43. Conney AH: Pharmacological implications of microsomal enzyme induction. Pharmacol Rev 19:317-366, 1967 44. Messier B, Leblond CP: Cell proliferation and migration as revealed by radioautography after injection of thymidine-H' into male rats and mice. Am J Anat 106:247-265, 1960 45. Bedard YC, Pinkerton PH, Simon GT: Radioautographic observations on iron absorption by the duodenum of mice with iron overload, iron deficiency and x-linked anemia. Blood 42:131-140, 1973 46. Kaufman RM, Pollack S, Crosby WH: Iron-deficient diet: effects in rats and humans. Blood 28:726-737, 1966 47. Becking GC: Influence of dietary iron levels on hepatic drug metabolism in vivo and in vitro in the rat. Biochem Pharmacol 21:1585-1593, 1972 48. Dallman PR, Goodman JR: The effects of iron deficiency on the hepatocyte: a biochemical and ultrastructural study. J Cell Bioi 48:79-90, 1971 49. Lu AYH, West SB, Yore M, et al: Role of cytochrome b. in hydroxylation by a reconstituted cytochrome P-450-containing system. J Bioi Chern 249:6701-6709, 1974 50. Dallner G, Siekevitz P, Palade GE: Biogenesis of endoplasmic reticulum membranes. II. Synthesis of constitutive microsomal enzymes in developing rat hepatocyte. J Cell Bioi 30:97-116, 1966 51. Guenther TM: Studies on induction of drug-metabolizing enzymes in fetal, neonatal and maternal rats. Ph.D. Thesis. University of Minnesota Graduate School, 1975
Vol. 70, No.6
GASTROENTEROLOGY
Printed in U.S.A.
OXIDATIVE METABOLISM OF FOREIGN COMPOUNDS IN RAT SMALL INTESTINE: CELLULAR LOCALIZATION AND DEPENDENCE ON DIETARY IRON HARALD HOENSCH, M.D.,
C.
H.
Woo,
M.S., STEVEN
B.
RAFFIN, M.D., AND RUDI SCHMID,
M.D.
Department of Medicine, University of California, School of Medicine, San Francisco, California
Oxidative metabolism of foreign compounds was measured in the intestinal mucosa of male rats. Activities of benzpyrene hydroxylase, p-nitroanisole O-demethylase, and NADPH-cytochrome P-450 reductase and cytochrome P-450 content were 3 to 10 times higher in epithelial cells of the upper villus than in mucosal crypt cells. Villous tip cells of the upper small intestine exhibited much higher cytochrome P-450 content and drug-metabolizing enzyme activity than did tip cells oflower intestinal segments. In rats fed commercial.chow diet, cytochrome P-450 content and drug-metabolizing enzyme activity in villous tip cells of duodenal mucosa were higher than in animals fed a semisynthetic diet, but cytochrome b 5 and NADPH-cytochrome P-450 reductase were unaffected. On restriction of dietary iron intake, cytochrome P-450 and oxidative enzyme activity fell sharply, but were completely restored in 24 hr by oral iron supplementation, whereas parenteral iron administration was ineffective. These findings suggest that intestinal drug metabolism is localized primarily in the upper villous cells of the proximal intestinal mucosa, that cytochrome P-450 is synthesized in maturing epithelial cells as they migrate from the crypts to the tip of the mucosal villi, and that this process is dependent critically upon absorption of iron from the intestinal lumen. A large number of drugs, carcinogens, environmental contaminants, and potentially toxic food constituents is metabolized in the body to derivatives that are biologically less active or less toxic but more water-soluble than the parent compounds. 2 Oxidative reactions catalyzed by microsomal enzymes play a prominent role in this process. 2, 3 Many of these microsomal enzymes represent mixed function monooxygenase systems consisting of cytochrome P-450, which serves as a terminal oxidase, and NADPH-cytochrome P-450 reductase, which catalyzes the reduction of the oxidized cytochrome 4 ; in most Received October 14, 1975. Accepted December 4, 1975. This paper was presented in part at the Annual Meeting of the American Gastroenterological Association, San Antonio, Texas, May, 1975. 1 Address reprint requests to: Dr. Rudi Schmid, University of California, San Francisco, HSW 1120, San Francisco, California 94143. This work was supported in part by Research Grants AM 17365 and AM 11275 from the National Institutes of Health and by the Walter C. Pew Fund for Gastrointestinal Research. Dr. Hoensch was a fellow of the Paul Martini Foundation, Frankfurt, Germany. His present address is: Department of Medicine, Gastroenterology Unit, University of Tiibingen, Tiibingen, Germany. The authors wish to thank Dr. Lloyd L. Brandborg for the histological examinations of the small intestine, and Dr. Myron Pollycove for the serum iron determinations. Ms. Lydia Hammaker, M.S., and Drs. D. Montgomery Bissell, Maria A. Correia, and Philip Guzelian provided help and advice in several aspects of this study, and Ms. Gail Persson gave expert editorial assistance.
instances the cytochrome reductase appears to be the rate-limiting enzyme. 5 In many of these oxidative reactions the liver exhibits the highest enzyme activity per unit weight; however, other tissues such as lung, skin, and intestine also have been shown to contain monooxygenase activity.6-11 These organs constitute the portal of entry for a variety of foreign compounds, and the small intestine, in particular, is exposed to many biologically active and potentially harmful substances that are ingested with food. For example, Wattenberg has suggested that the intestinal aryl hydrocarbon hydroxylase system may act as a first line of defense against dietary carcinogens such as polycyclic aromatic hydrocarbons. 12 This and other intestinal oxidative enzyme systems have been shown to be stimulated by ingestion of aromatic hydrocarbons, 13, 14 phenobarbital,15 charcoal-broiled meat, or leafy vegetables. 16. 17 The active agents in vegetables appear to be indoles,18 which are naturally occurring plant constituents. Whereas activity of intestinal drug-metabolizing enzymes can readily be measured in microsomes prepared from whole intestinal mucosa, 19 spectrophotometric determination of cytochrome P-450 in intestinal mucosa was found to be more difficult. 20 This probably was due to several factors, including the relatively low cytochrome content of whole and particularly of distal intestinal mucosa,21, 22 contamination with hemoglobin 21 and presence of cytochrome P_420.20 This technical
1063
1064
HOENSCH ET AL.
problem was minimized in the present investigation by selective use of duodenal mucosa and by segregation of the mucosal cells into individual cell populations of reasonably uniform maturity. Intestinal mucosa is a rapidly proliferating tissue, in which relatively undifferentiated, dividing cells of the basal crypts progressively mature as they migrate to the upper villous stratum. 23 For several enzyme systems, marked differences have been shown between proliferating crypt cells and more mature absorptive cells of the mucosal villus. 24 - 27 Recently it has been documented that drug-metabolizing enzyme activity and cytochrome P-450 content of duodenal mucosa are located primarily in the epithelial cells of the upper villus. 22 This made it possible to use epithelial cells from the villous tip to investigate the effects of dietary factors and particularly of dietary iron on intestinal cytochrome P-450 content and oxidative drug metabolism.
Materials and Methods Animals and diets. All experiments were carried out in adult male rats (300 to 400 g) purchased from Charles River Breeding Laboratories, Inc., Wilmington, Mass., and certified to be "specific pathogen" (Bartonella)-free. The animals were housed in stainless steel cages in a temperature- and light-controlled room, and were protected from contact with pesticides. The following diets were used: commercial rat chow diet (Berkeley Diet, Feed Stuffs Processing Co., San Francisco, Calif.) , containing ground vegetables, vegetable oil, dried yeast, vitamins, and trace minerals, including 20 mg of iron per 100 g of diet; semisynthetic iron-supplemented diet (Nutritional Biochemicals Corp., Cleveland, Ohio), consisting of vitamin test casein, 27%; corn starch, 55%; hydrogenated vegetable oil, 14%; salt mixture without ferric phosphate, 3%; vitamin mixture, 1%; and ferrous sulfate, 8 mg of iron per 100 g of diet (iron content of the semisynthetic diet up to 40 mg of iron per 100 g of diet did not increase the content of intestinal cytochrome P-450); and semisynthetic low iron diet, containing all of the above ingredients except ferrous sulfate, resulting in an overall iron content of less than 0.1 mg per 100 g of diet. In some experiments, semisynthetic diets were supplemented with 0.05% (w/w) 3-methyl-cholanthrene. In one experiment, rats were given intramuscular injections of 0.1 ml of Imferon (Lakeside Laboratories, Milwaukee, Wis.) containing 50 mg of iron per m!. All animals were fed ad libitum, had free access to water, and gained weight at similar rates regardless of the dietary regimen. Unless stated otherwise, the rats were kept on a specific diet for at least 7 days before being killed. At approximately 9:00 AM of the day of experiment the animals were anesthesized with ether and exsanguinated by aortic puncture. All subsequent procedures were conducted at 4 DC. Preparation of intestinal cell fractions. Beginning at the pylorus, consecutive 15-cm segments of small intestine were excised, perfused with ice-cold 1.15% (w/v) KCI, split longitudinally along the mesenteric border, and opened. Villous tip, lower villous, and crypt cells of the intestinal mucosa were separated by a modification of the differential scraping technique. 28 Villous tip cells were removed by light hand scraping with the blade of a metal spatula; slightly increased scraping pressure yielded lower villous cells. Crypt cells were harvested by abrasion of the remaining mucosa. Histological examination of the separated cell fractions indicated that they consisted of reasonably uniform cell
Vol. 70, No.6
populations. The discriminatory effectiveness of the scraping procedure was further evaluated by determining the incorporation of 3H-thymidine or 59Fe into the individual cell fractions in vivo. An intraperitoneal injection of 50 J.l.c of 3H-thymidine (specific activity 6.7 c per mmole, New England Nuclear Corp., Boston, Mass.) was administered to each of 9 rats fed the semisynthetic iron-supplemented diet. Groups of 3 rats were killed 1, 24, and 48 hr later, and the duodenal mucosa was fractionated by scraping. Individual cell fractions from 3 animals were pooled and homogenized in 0.25 M sucrose solution, the· cell nuclei were sedimented by centrifugation at 900 g for 10 min, DNA was precipitated with cold 10% trichloroacetic acid, and the radioactivity of the preparation was measured in a liquid scintillation spectrophotometer (model LS-250, Beckman Instruments, Inc., Fullerton, Calif.). Results were expressed as counts per minute per milligram of nuclear protein. In confirmation of previous reports, 23. 26 after 1 hr most of the label appeared in the crypt cell fraction, whereas after 48 hr, villous tip cells exhibited the highest radioactivity (table 1). Three rats fed the semisynthetic iron-supplemented diet and 3 rats on semisynthetic low iron diet for 2 days were each given an intravenous injection of 50 J.l.c of 59Fe (specific activity 1.71 c per mmole, ICN Isotope and Nuclear Division, Cleveland, Ohio). The radioactive ferrous citrate was preincubated with 0.5 ml of rat serum and injected in a single pulse into a femoral vein. The animals were killed 4 hr later and the radioactivity of homogenized individual mucosal cell fractions was determined in a well-type scintillation detector (model 8275, Nuclear Chicago Corp., Des Plaines, Ill.). Results were expressed as counts per minute per milligram of cellular protein (table 2). As expected from previous autoradiographic observations,29 uptake of radioactive iron was highest in the epithelial crypt cells whereas the villous tip cell fraction contained much less radioactivity (table 2). No difference was found between animals on iron-supplemented and low iron diet. The results of these tracer experiments and the histological findings indicate TABLE
1. Consecutive incorporation of 3H-thymidine into cell fractions
of duodenal mucosa" Time after administration (hr)
Cell fraction
24
48
cpm/mg nuclear protein
Villous tip Lower villous Crypt
174 454 2822
320 808 1497
908 789 372
"Rats were given intraperitoneal injections of 3R-thymidine at time zero and killed after the intervals indicated. TABLE
2. Differential incorporation of '·Fe into separated cell fractions
of duodenal mucosa" Cell fraction
Semisynthetic iron-supplemented diet
Semisynthetic low iron diet
cpm/mg cellular protein
Villous tip Lower villous Crypt
123 ( 70-155) 278 (149-346) 577 (509-635)
130 ( 90-170) 194 (107-280) 500
(442-558)
" Three rats were given intravenous injections of '·Fe and killed 4 hr later. Values represent mean and range of individual rats.
June 1976
that the manual scraping procedure yielded mucosal cell fractions of acceptable functional and morphological uniformity. Preparation of microsomes. Individual cell fractions were weighed and homogenized in 0.1 M potassium phosphate buffer, pH 7.4, using 15 strokes of a Potter-Elvehjem teflon glass homogenizer. The homogenate was then sonicated for 25 sec at 35 watts (model W185 D, Sonifier Cell Disruptor, Heat, Systems-Ultrasonics, Inc., Plainview, N. Y.) and sedimented at 20,000 x g for 10 min in a Sorval supers peed automatic refrigerated centrifuge (model RC2-B, Ivan Sorval, Inc., Norwalk, Conn.). The supernatant was centrifuged at 105,000 x g for 60 min, the microsomes were washed once with 1.15% aqueous KCI to remove contaminating hemoglobin, and the pellet was resuspended in 0.1 M potassium phosphate buffer in a final volume equivalent to 300 mg of wet mucosal tissue per ml. The microsomal suspensions of villous tip, lower villous, and crypt cells contained approximately 3 to 4 mg of protein per ml and the yield of microsomal protein was 10 to 15 mg per g wet weight of tissue for each cell fraction. In additional experiments, equal volumes of microsomal suspensions obtained from tip and crypt cell fractions were mixed and incubated for 90 min at room temperature 2 • and cytochrome P-450 content was measured. Also, microsomal preparations of tip cells were incubated for 90 min at room temperature or at 4 DC with a 105,000 x g supernatant fraction of tip or crypt cell homogenate, and cytochrome P-450 content and benzpyrene hydroxylase activity were then determined. Analytical procedures. Cytochromes P-450 and b. were determined in microsomal suspensions by the method of Omura and Sato,3. benzpyrene hydroxylase activity by the method of Wattenberg et al.I< as modified by Kuntzman et al. 31 and Alvares et al,32 and p-nitroanisole O-demethylase activity by the method of Netter and Seidel. 33 NADPH-cytochrome P-450 reductase was measured by the procedure of Williams et al. 34 Protein concentration was estimated by the procedure of Lowry et al. 3. using human albumin as standard. Serum iron concentration was measured by the method of Stookey."·
Results On fractionation of duodenal mucosa into villous tip, lower villous, and crypt cells, striking differences in cytochrome P-450 content and drug-metabolizing enzyme activity were observed (table 3). Thus, in rats fed standard rat chow diet, cytochrome P-450 content of tip cell fractions was 10 times higher than in crypt cells, and for NADPH-cytochrome P-450 reductase the difference was 3-fold. By contrast, cytochrome b s content of crypt cells was only slightly less than that of villous tip cells. Benzpyrene hydroxylase and p-nitroanisole O-demethTABLE
Cell fraction
ylase activity showed a distribution pattern similar to that of cytochrome P-450 (table 3), with lower villous cell fractions exhibiting enzyme activity intermediate between that of villous tip and crypt cells. These differences in cytochrome P-450 content and enzyme activity between villous tip and crypt cells could not be attributed to the presence of inhibitors or activators, as on incubation of microsomal mixtures prepared from these cell fractions the results obtained were those predictable from the constituent components of the mixture. Moreover, when microsomal preparations of villous tip cells were incubated with 105,000 x g supernatant of crypt cell homogenate, no changes in cytochrome P-450 content and benzpyrene hydroxylase activity were observed. The previously noted occurrence of an absorptive peak at 420 nm in the carbon monoxide-reduced difference spectrum of intestinal microsomes 2o , 21 was confirmed, but its magnitude was markedly reduced by the preparative procedures employed for removal of contaminating hemoglobin. Thus, the animals were thoroughly exsanguinated, residual blood was removed from the small intestine by perfusion of intestinal arteries with cold isotonic saline, and the microsomal pellet was washed with 1.15% M KCI solution. With these precautions, only microsomes of crypt cells exhibited a demonstrable, albeit minor, absorption peak in the 420 nm region . The present findings confirm the previously reported observation that monooxygenase activity is higher in the mucosa of the upper small intestine than in that of more distal segments of the bowel. 14, 37 Thus, in microsomes of villous tip cells, activity of benzpyrene hydroxylase (fig. 1) and content of cytochrome P-450 (fig. 2) progressively decreased from duodenum to ileum. A similar profile of activity was observed with p-nitroanisole O-demethylase and NADPH-cytochrome P-450 reductase. Because of the reported effect of dietary constituents on intestinal drug-metabolizing monooxygenase activity, 12, 16, 18 most. present experiments were performed with semisynthetic iron-supplemented diet of defined composition. Villous tip cells of rats on this diet exhibited distinctly lower cytochrome P-450 content and benzpyrene hydroxylase and p-nitroanisole O-demethylase activity than did those of animals on commercial chow diet (table 4). When the semisynthetic iron-containing diet was supplemented with 3-methylcholanthrene for 4 days, cytochrome P-450 content, benzpyrene hydroxylase, and p-nitroanisole O-demethylase activi-
3. Microsomal hemoproteins and drug metabolism in different cell fractions of duodenal mucosa of rats" Cytochrome P·450
Cytochrome b,
Benzpyrene hydroxylase
0.115 0.070 0.011
± ± ±
0.007 (9) 0.004 (3) 0.002 (3)
0.119 0.088 0.082
p·Nitroanisole O·demethylase
nmoles/mg protein/hr
nmoles/mg protein
Villous tip Lower villous Crypt
1065
DRUG METABOLISM IN SMALL INTESTINE
± ± ±
0.004 (9) 0.007 (3) 0.001 (3)
6.5 3.4 1.4
± ± ±
0.6 (9) 0.1 (3) 0.4 (3)
15.5 ± 1.1 (4) 7.6 ± 0.6 (3) Trace
NADPH-cytochrome P-450 reductase nmol.s/mg protein / min
133.8 74.5 43.1
± ± ±
5.2 (3) 2.6 (3) 7.8 (3)
" Values represent mean ± standard error of mean of individual experiments using mucosal cell fractions pooled from 4 rats on commercial rat chow diet. NADPH-cytochrome P·450 reductase was determined in cell fractions from individual rats. Number of individual experiments are given in parentheses. All values are significantly different from those of villous tip cells at the P < 0.01 level.
1066
Vol. 70, No.6
HOENSCH ET AL.
ties were significantly increased in villous tip cells of duodenal mucosa, but cytochrome b s and NADPH-cytochrome P-450 reductase remained unaffected (table 4). Restriction of oral iron intake resulted in a prompt decrease of cytochrome P-450 content and benzpyrene hydroxylase activity in villous tip cells (table 5, fig. 3). Within 2 days on the semisynthetic low iron diet, both values fell to less than one-half and then leveled off on continued dietary iron restriction (table 5). On the other hand, cytochrome b s did not change significantly over 30
days of iron deprivation. NADPH-cytochrome P-450 reductase activity also was not affected by feeding the low iron diet for 2 days. In rats fed the low iron diet, serum iron concentration fell from 202 ± 7 (mean ± SEM) Jlg per 100 ml to 137 ± 5 Jlg within the first 24 hr and further decreased on iron deprivation for 30 days (table 5). The striking reduction of cytochrome P-450 content and benzpyrene hydroxylase activity of villous tip cells' 0.15
6
Z ii:i
>-
o
5 Z ii:i >-
0
...'"
4
~
3
"0 0V> E 0 v
2
0>
.....E ~
~
G
~
~
'"
CYTOCHROME P·450
'"
BENZPYRENE HYDROXYlASE
0 .1 0
0..
~ ~
oV> o
'"v ~
. ~
.....
0 .05
G
i
'0
E ~
15
30
45
60
75
90
105
120
15
SMAll INTESTINE (em beyond pylorus)
FIG. 1. Distribution of benzpyrene hydroxylase activity in villous tip cells of sequential segments of small intestine of rats fed commercial chow diet. Vertical bars represent standard error of mean of three individual experiments. TABLE 4.
30
45
60
75
90
105
120
SMAll INTESTINE lem beyond pylorus)
FIG. 2. Distribution of cytochrome P-450 content in villous tip cells of sequential segments of small intestine of rats fed commercial chow diet. Vertical bars represent standard error of mean of three individual experiments.
Effects of diet s and feeding of 3-methylcholanthrene on microsomal hemoproteins and drug metabolism in villous tip cells of duodenal mucosa of rats"
Diet
Cytochrome P·450
Benzpyrene hydroxylase
Cytochrome b,
nmoles/ mg protein
0.115 Commercial chow 0.067 Semisynthet ic iron·sup· plemented Semisynthetic iron·sup0.132 plemented with 50 mg 3-methyl-cholanthrene/1oo
nmoles/mg protein/hr
6.5 2.4
±
±
0.004 (9) 0.005 (9)
±
0.004 (4)
47.4
0.119 0.124
±
±
0.007 (9) 0.004 (12)"
±
0.010 (4)
0.097
±
p·Nitroanisole O·demethylase
NADPH·cytochrome P·450 reductase nmoles/ mg protein/ min
±
0.6 (9)" 0.2 (8)"
15.5 ± 1.1 (4)" Trace
133.8 ± 5.2 (3) 115.0 ± 9.4 (3)
±
2.8 (4)"
40.6 ± 1.5 (4)"
108.6 ± 6.6 (4)
g
"Values represent mean ± standard error of mean of individual experiments using villous tip cells pooled from 4 rats. In rats fed 3-methylcholanthrene for 4 days, determinations were made in samples from individual rats. Number of experiments is given in parentbeses. " Indicates that the value differs from the two other values at the P < 0.01 level. TABLE
5. Effect of dietary iron restriction on microsomal hemoproteins and benzpyrene hy droxylase activity in villous tip cells
of duodenal mucosa of rats" Days on low iron diet
Benzpyrene hydroxylase
Cytochrome b,
Cytochrome P·450
nmoles/hr/ mg protein
nmoles /mg protein
o 2 4
30
0.004 (l2) 0.067 0.060 ± 0.011 (3) 0.030 ± 0.003 (6)" 0.031 ± 0.003 (3)" 0.025 ± 0.003 (6)" ±
0.124 0.127 0.131 0.146 0.112
± ± ± ± ±
0.005 0.005 0.008 0.004 0.017
(9) (3) (7) (3) (4)
2.4 ± 0.2 (8) 1.3 ± 0.2 (3)" 0.4 ± 0.1 (3)" 0.7 ± 0.1 (3)" 0.8 ± 0.1 (7)"
Serum iron Ilg/lOO ml
202 137 147 148 101
± ± ± ± ±
7 (16) 5 (l0)" 4 (17)" 4 (9)" 5 (10)"
" Values represent mean ± standard error of mean of individual experiments using villous tip cells pooled from 4 rats. Number of experiments is given in parentheses. b Values differ significantly from those of day 0 at the P < 0.01 level.
June 1976
1067
DRUG METABOLISM IN SMALL INTESTINE
on brief restriction of oral iron intake was completely reversible by dietary iron supplementation for 24 hr (fig. 3). By contrast, parenteral iron adminstration was without effect on intestinal cytochrome P-450, despite a sharp rise in serum iron concentration (fig. 4). In rats on low iron diet for 2 days, intramuscular injection of two or four individual doses of 5 mg of iron each raised the serum iron concentration from 147 ± 4 to 717 ± 22 and 735 ± 30 Jlg per 100 ml, respectively, but failed to alter the cytochrome P-450 content of villous tip cells (fig. 4). Feeding of the low iron diet also depressed the response of intestinal cytochrome P-450 to the stimulatory effect of dietary 3-methyl-cholanthrene. In rats fed 0.05% 3-methylcholanthrene (w/w) in semisynthetic diet for 4 consecutive days, restriction of oral iron intake during the last 48 hr of the experiment reduced cytochrome P-450 content to less than one-half of the value in 3-methylcholanthrene-fed animals on iron-supplemented diet (fig. 5). On the other hand, dietary iron restriction failed to reduce the 3-methylcholanthrenestimulated activity of benzpyrene hydroxylase and pnitroanisole O-demethylase (fig. 5). Activity of NADPHcytochrome P-450 reductase of villous tip cells remained unchanged when rats were given oral 3-methylcholanthrene and fed either the semisynthetic iron-supplemented or low iron diet.
cerned primarily with the metabolism of foreign compounds that are taken up from the intestinal lumen. 9, 1214, 16 It is methodologically difficult, however, to determine in vivo to what extent ingested substances actually are metabolized in the intestine itself rather than being absorbed intact and then undergoing biotransformation on their first pass through the liver.38 Nonetheless, preliminary observations with tetrahydrocannabinol,39 phenacetin,40 and chlorpromazine 41 in humans and rats suggest that intestinal transformation plays a major role in the oxidative metabolism of these drugs. Since many of these lipophilic foreign compounds appear to be absorbed mainly in the proximal small intestine,42 it is not surprising that drug-metabolizing enzyme activity in mucosal homogenates 14 or microsomal preparations is highest in the duodenum and jejunum (fig. 1). The present findings that drug-metabolizing monooxygenase systems and their constituent component cytochrome P-450 are localized primarily in the absorptive cells of the upper mucosal villus (table 3) provide further evidence that ingested foreign compounds may undergo biotransformation at the very site of their absorption. Thus, aryl hydrocarbon hydroxylase and O-demethylase activity and cytochrome P-450 content were much higher in epithelial cells of the villus than in the immature cells of the crypt region. This observation implies that these enzyme systems are formed as the epithelial cells differentiate in the course of their migraDiscussion tion from the mucosal crypts to the tip of the villous Recent observations have indicated that in addition to structure. 23 This process of "enzyme maturation" apwell defined absorptive and transport functions, the pears to be controlled in part by dietary constituents mucosa of the small intestine contains a number of that are absorbed and presumably'serve as substrates for enzyme systems that catalyze metabolic transforma- these enzyme systems. Thus, benzpyrene hydroxylase tions of many foreign compounds. 3, 12, 20, 21, 37 Whereas and p-nitroanisole O-demethylase activity and cytoin this regard intestinal mucosa qualitatively resembles chrome P-450 content of villous tip cells significantly the liver, the intestinal enzymes presumably are con- decreased when semisynthetic iron-supplemented diet
~
~
....w
CYTOCHROME P-450
0.10
...'"...
«
'" '" !::!
«
l: 0 0
'"
0
l: 0 0
...w 0 ...'"...
0.05
l:
'" !::!
'~
2.5
r-----1
2.0
200
1.5
'"
E
-;:-
1.0
~
';;..!
Q
'"E
BENZPYRENE HYDROXYLASE ACTIVITY
l:
137
0
E c:
147
...... ..!
3.0
0.5
2
147
~
0
E c:
0
f--lO~I:~ON~ I - - IRON~I~~PPl'---t
I 0
I 2
I 3 DAYS
I
"
I 5
0
~OWD::~N---l ~ON~I~~PPl---1
I 0
I
2
3 DAYS
"
I 5
FIG. 3. Effects of dietary iron restriction and oral iron supplementation on cytochrome P-450 content and benzpyrene hydroxylase activity in villous tip cells of rat duodenal mucosa. Vertical bars represent standard error of mean of three to five individual experiments using villous tip cells pooled from 4 rats. Numbers indicate average serum iron, p.g per 100 m!.
was substituted for standard rat chow (table 4), but increased strikingly when 3-methylcholanthrene was added to the semisynthetic diet (table 4)_ Previous reports already had indicated that dietary factors 12, 14, 16 or supplementation of diet with certain drugs 9 affect oxidative enzyme systems involved in the metabolism of foreign compounds in whole mtestinal mucosa_ The present findings suggest that these observed alterations in the whole mucosa may have reflected largely changes of enzyme activity in the mucosal villous tip cells. In analogy with observations in the liver,43 it has been assumed that this diet- or drug-mediated stimulation of enzyme activity in the intestinal mucosa represents
~
w
0
137
:;( ~
0
0.04
7?~0
CYTOCHROME P-450
i
'"E ---
,,,,
INTRAMUSCULAR IRON
0.06
'"0..
0
0
735
147 0.02
~ 0
E c
Vol. 70, No.6
HOENSCH ET AL.
1068
Ll_
~-
0 DAYS ON lOW IRON DIET
FIG. 4. Effects of dietary iron restriction and parenteral iron administration on cytochrome P-450 content in villous tip cells of rat duodenal mucosa. Vertical bars represent standard error of mean of three to five individual experiments. For the 3rd day, individual values are given of two pairs of experiments each based on analysis of pooled villous tip cells from 4 rats. Five milligrams of iron as Imferon were given intramuscularly at times indicated by arrows. Numbers indicate average serum iron, J.l.g per 100 m!. Cytochrome
BenzpYfene
p-N itroanisol e
P-4S0
hydroxylase
O-demethylase
0.15
50 w
'" 0..
0
'"0.. ~
<
0.10
~
0
'" i
'" ---!
'"
0
E c
---
""
:<~ 0
30
E
0.05
'"0..
~O
U
i
E
:<~
0
W
0
I-
I-
20
150 -
~
w
0
~
NADPH-cytochrome P-4S0 reductase
50
~
~
w
enzyme induction. 37 Although this appears a likely explanation, mechanistic interpretation of these findings is complicated by the fact that intestinal mucosa is an actively proliferating tissue with highest enzyme activity residing in the villous tip cells, which, while fully differentiated,23 have a very short life span. 44 Because of this rapid cell turnover, it is difficult to procure the stringent experimental evidence that would be required' to establish enzyme induction, and additional studies are necessary to clarify this point. Since cytochrome P-450 content is 10 times higher in villous tip cells than in crypt cells (table 3), it is likely that most of this iron-containing hemoprotein is synthesized or assembled during the migration of these epithelial cells from the mucosal crypts to the villous surface. This raises the question regarding the source of the iron required for the synthesis of this cytochrome in the upper villous stratum. Prevous investigations with autoradiographic techniques have demonstrated that parenterally administered radioiron initially appears predominantly in the mucosal crypt region,29 whereas orally ingested radioiron is taken up preferentially by the absorptive cells of the mucosal villi.45 This has been confirmed in the present study, in that intravenously injected 59Fe 4 hr after administration was concentrated largely in the crypt cell fraction whereas tip cell preparations exhibited a much lower isotope content (table 2). As the radioiron-laden crypt cells differentiate and migrate along the villous structure, the radioiron eventually is carried to the top of the villus. 29 That this source of iron apparently is insufficient for the synthesis of cytochrome P-450 in the ascending epithelial cells is shown by the data in table 5 and figure 3. In rats placed on a virtually iron-free diet, cytochrome P-450 content of villous tip cells fell within 48 hr to less than one-half of control but was completely restored by oral iron supplementation for 24 hr (fig_ 3). Restriction of oral iron for up to 13 days in adult rats is reported not to produce appreciable depletion of body iron stores, nor to lead to demonstrable
I
'" i
I-
0 40
'" 0..
30
U
E '" ---
""
20
-! 0
E c
10
FIG. 5. Cytochrome P-450 content and activity of benzpyrene hydroxylase, p-nitroanisole O-demethylase, and NADPH-cytochrome P-450 reductase in villous tip cells of duodenal mucosa of rats on semisynthetic iron-supplemented diet (shaded column), semisynthetic iron-supplemented diet with added 3-methylcholanthrene (open column), and semisynthetic low iron diet with 3-methylcholanthrene (cross-hatched column). Vertical bars represent standard error of mean of three to five individual experiments.
June 1976
1069
DRUG METABOLISM IN SMALL INTESTINE
changes in plasma iron turnover.46 This also is shown by the present observation that in rats fed the low iron diet for 48 hr, incorporation of parenterally administered radioiron into mucosal crypt cells did not differ significantly from that of control animals (table 2). Whereas the present findings with iron-free diet conceivably could result from accelerated mucosal cell turnover or from destabilization of cytochrome P-450 caused by the iron restriction, there is no supportive evidence for these explanations. More likely is that absorption of iron from the intestinal lumen is essential for the synthesis of cytochrome P-450 in the villous epithelial cells. This is further substantiated by the ineffectiveness of intramuscular iron injections to raise the mucosal cytochrome P-450 content in animals that had been on restricted iron intake for 48 hr (fig. 4). The finding that on the low iron diet cytochrome P-450 fell sharply during the initial 48 hr but then leveled off for the remainder of the 30-day observation period (table 5) may indicate the limited extent to which body iron is available for cytochrome P-450 synthesis in the epithelial villous cells. Since these diet-induced alterations in mucosal cytochrome P-450 content were accompanied by comparable changes in benzpyrene hydroxylase (table 5, fig. 3), but not in NADPH-cytochrome P-450 reductase activity, it is possible that under these experimental conditions, cytochrome P-450 may be rate-controlling for the aryl hydrocarbon hydroxylase reaction. The present findings raise several questions to which answers are not readily available. It previously had been shown that in rats made iron-deficient, cytochrome P -450 and oxidative drug metabolism of the liver are not reduced. 47 , 48 In fact, the activity of several hepatic drug-metabolizing enzymes and of NADPH-cytochrome P-450 reductase tends to rise after induction of iron deficiency. 47 This was found to be associated with increased plasma turnover of aniline and aminopyrine, 47 suggesting that the hepatic metabolism of these compounds may be enhanced in vivo. This response of the liver to iron deficiency is strikingly different from the present findings in the gut. Whereas the effect of chronic iron deficiency on intestinal cytochrome P-450 and drug metabolism was not investigated, it is evident that brief restriction of intestinal iron absorption resulted in a rapid albeit fully reversible fall of mucosal cytochrome P-450 content and monooxygenase activity. It is unclear whether this difference between liver and gut reflects dissimilar regulatory mechanisms or may be related to the slow cell turnover in the liver, permitting more effective conservation of cellular iron. Marked differences also were noted in the intestinal mucosa between cytochrome P-450 and cytochrome b s ; the role of the latter cytochrome in oxidative drug metabolism is still controversial. 49 The content of cytochrome b s in villous tip cells was slightly higher than that in crypt cells (table 3), but the gradient from crypts to villous tips was much less than for cytochrome P-450. This suggests that most of the mucosal cytochrome b s is synthesized in the proliferating crypt cells and is carried within the differentiating epithelial cells to the upper
villous stratum. Moreover, neither dietary iron restriction nor supplementation of the diet with 3-methylcholanthrene affected the mucosal cytochrome b s content (table 4). This difference in behavior between the cytochromes b s and P-450 of the intestinal mucosa is reminiscent of comparable findings in the liver; in developing rat liver, cytochrome b s tends to appear earlier than cytochrome P-450, so, S1 and its turnover in adult rat hepatocytes is considerably slower than that of cytochrome P-450. 2 The observed decrease of mucosal benzpyrene hydroxylase activity during dietary iron restriction (table 5, fig. 3) resembles that found in the fasting animal '2 and suggests that the reduced level of cytochrome P-450 (fig. 3) might preclude an adaptive increase of intestinal oxidative drug metabolism in response to ingestion of stimulatory compounds. That this is not the case is shown by the data in figure 5, which demonstrate that in rats fed the low iron diet and 3-methylcholanthrene, benzpyrene hydroxylase and p-nitroanisole O-demethylase activity markedly rose despite the low cytochrome P-450 level. By contrast, neither low iron diet nor 3-methylcholanthrene affected NADPH-cytochrome P-450 reductase activity (fig. 5). These findings indicate that on stimulation with 3-methyl-cholanthrene, cytochrome P450 could not be rate-controlling for the activity of intestinal benzpyrene hydroxylase and p-nitroanisole O-demethylase. The differences in ratio between intestinal drug-metabolizing enzymes, NADPH-cytochrome P-450 reductase, and cytochrome P-450 in rats fed semisynthetic iron-supplemented diet with 3-methylcholanthrene or semisynthetic low iron diet with 3-methylcholanthrene illustrate the difficulties that are encountered in defining the rate-limiting factors in oxidative drug metabolism. 2 REFERENCES 1. Hoensch HP, Woo CH, Schmid R: Cytochrome P-450 in duodenal mucosa: dependence on dietary (luminal) iron (abstr). Gastroenterology 68:A-55/912, 1975 2. Remmer H: Induction of drug metabolizing enzyme system in the liver. Eur J Clin Pharmacol 5:116-136, 1972 3. Hartiala K: Metabolism of hormones, drugs and other substances by the gut. Physiol Rev 53:496-534, 1973 4. Estabrook RW, Matsubara T, Mason JL, et al: Studies on the molecular function of cytochrome P-450 during drug metabolism. In Microsomes and Drug Oxidations. Edited by RW Estabrook, JR Gillette, and KC Leibman. Baltimore, Williams & Wilkins Company, 1973, p 98-110 5. Gillette JR, Davis DC, Sasame HA: Cytochrome P-450 and its role in drug metabolism. Ann Rev Pharmacol 12:57-84, 1972 6. Bend JR, Hook GE, Gram TE: Characterization of lung microsomes as related to drug metabolism. Drug Metab Dispos 1:358-366, 1973 7. Neal RA: A comparison of the in vitro metabolism of parathion in the lung and liver of the rabbit. Toxicol Appl Pharmacol 23:123-130, 1972 8. Feuer G, Sosa-Lucero JC, Lumb G: Failure of various drugs to induce drug-metabolizing enzymes in extrahepatic tissues of the rat. Toxicol Appl Pharmacol 19:579-589, 1971 9. Lake BG, Hopkins R, Chakraborty J, et al: The influence of some hepatic enzyme inducers and inhibitors on extrahepatic drug
1070
HOENSCH ET AL.
metabolism. In Microsomes and Drug Oxidations. Edited by RW Estabrook, JR Gillette, and KC Leibman. Baltimore, Williams & Wilkins Company, 1973, p 342-349 10. Bickers DR, Eiseman J, Kappas A: Microscope immersion oils: effects of skin application on cutaneous and hepatic drug-metabolizing enzymes. Biochem Pharmacol 24:779-783, 1975 11. Welch RM, Cavallito J, Loh A: Effect of exposure to cigarette smoke on the metabolism of benzo[a)pyrene and acetophenetidin by lung and intestine of rats. Toxicol Appl PharmacoI23:749-758, 1972 12. Wattenberg LW: Dietary modification of intestinal and pulmonary aryl hydrocarbon hydroxylase activity. Toxicol Appl Pharmacol 23:741-748, 1972 13. Pantuck EJ, Hsiao KC, Kaplan SA, et al: Effects of enzyme induction on intestinal phenacetin metabolism in the rat. J Pharmacol Exp Ther 191:45-52, 1974 14. Wattenberg LW, Leong JL, Strand PJ: Benzpyrene hydroxylase activity in the gastrointestinal tract. Cancer Res 22:1120-1125, 1962 15. Lehrmann C, Ullrich V, Rummel W: Phenobarbital inducible drug monooxygenase activity in the small intestine of mice. Naunyn Schmiedebergs Arch Pharmacol 276:89-98, 1973 16. Pantuck EJ, Hsiao KC, Kuntzman R, et al: Intestinal metabolism of phenacetin in the rat: effect of charcoal-broiled beef and rat chow. Science 187:744-746, 1975 17. Hsiao KC, Pantuck EJ, Wattenberg LW, et al: Stimulatory effect of a brussel sprouts or cabbage diet on intestinal drug metabolism in the rat. Fed Proc 34:742, 1975 18. Loub WD, Wattenberg LW, Davis DW: Aryl hydrocarbon hydroxylase induction in rat tissues by naturally occurring in doles of cruciferous plants. J Nat! Cancer Inst 54:985-988, 1975 19. Chhabra RS, Fouts JR: Sex differences in the metabolism of xenobiotics by extrahepatic tissue in rats. Drug Metab Dispos 2:375-379, 1974 20. Chhabra RS, Pohl RJ, Fouts JR: A comparative study of xenobioticmetabolizing enzymes in liver and intestine of various animal species. Drug Metab Dispos 2:443-447, 1974 21. Zampaglione NG, Mannering GJ: Properties of benzpyrene hydroxylase in the liver, intestinal mucosa and adrenal of untreated and 3-methylcholanthrene-treated rats. J Pharmacol Exp Ther 185:676-685, 1973 22. Hoensch HP, Woo CH, Schmid R: Cytochrome P-450 and drug metabolism in intestinal villous and crypt cells of rats: effect of dietary iron. Biochem Biophys Res Commun 65:399-406, 1975 23. Imondi AR, Balis ME, Lipkin M: Changes in enzyme levels accompanying differentiation of intestinal epithelial cells. Exp Cell Res 58:323-330, 1969 24. Nordstrom C, Dahlqvist A: Quantitative distribution of some enzymes along the villi and crypts of human small intestine. Scand J Gastroenterol 8:407-416, 1973 25. Fortin-Magana R, Hurwitz R, Herbst JJ: Intestinal enzymes: indicators of proliferation and differentiation in the jejunum. Science 167: 1627 -1628, 1969 26. Weiser MM: Intestinal epithelial cell surface membrane glycoprotein synthesis. J Bioi Chern 248:2536-2541, 1973 27. Kim YS, Perdomo J, Ochoa P: Regional and cellular localization of glycosyltransferases in rat small intestine. Changes in enzymes with differentiation of intestinal epithelial cells. Biochim Biophys Acta 391:39-50, 1975 28. Dietschy JM, Siperstein MD: Cholesterol synthesis by the gastrointestinal tract: localization and mechanisms of control. J Clin Invest 44:1311-1327, 1965
Vol. 70, No.6
29. Conrad ME, Crosby WH: Intestinal mucosal mechanisms controlling iron absorption. Blood 22:406-415, 1963 30. Omura T, Sato R: The carbon monoxide-binding pigment of liver microsomes. J Bioi Chern 239:2370-2385, 1964 31. Kuntzman R, Mark LC, Brand L, et al: Metabolism of drugs and carcinogens by human liver enzymes. J Pharmacol Exp Ther 152: 151-156, 1966 32. Alvares AP, Schilling G, Garbut A: Studies on the hydroxylation of. 3,4-benzpyrene by hepatic microsomes. Biochem Pharmacol 19:1449-1455, 1970 33. Netter KJ, Seidel G: An adaptively stimulated O-demethylating system in rat liver microsomes and its kinetic properties. J Pharmacol Exp Ther 146:61-65, 1964 34. Williams CH, Kamin H: Microsomal triphosphopyridine nucleotide-cytochrome c reductase of liver. J Bioi Chern 237:587-595, 1962 35. Lowry OH, Rosebrough NJ, Farr AL, et al: Protein measurement with the folin phenol reagent. J Bioi Chern 193:265-275, 1951 36. Stookey LL: Ferrozine-a new spectrophotometric reagent for iron. Anal Chern 42:779-781, 1970 37. Scharf R, Ullrich V: In vitro induction by phenobarbital of drug monooxygenase activity in mouse isolated small intestine. Biochern Pharmacol 23:2127-2137, 1974 38. Gibaldi M, Perrier D: Route of administration and drug disposition. Drug Metab Rev 3:185-199, 1974 39. Greene ML, Saunders DR: Metabolism of tetrahydrocannabinol by the small intestine. Gastroenterology 66:365-372, 1974 40. Pantuck EJ, Kuntzman R, Conney AH: Decreased concentration of phenacetin in plasma of cigarette smokers. Science 175:1248-1250, 1972 41. Curry SH, D'Mello A, Mould GP: Destruction of chlorpromazine during absorption in the rat in vivo and in vitro. Br J Pharmacol 42:403-411, 1971 42. Knoefel PK: Absorption, chap 2, Absorption, Distribution, Transformation and Excretion of Drugs. Edited by PK Knoefel. Springfield, Ill., Charles C Thomas, 1972, p 39-55 43. Conney AH: Pharmacological implications of microsomal enzyme induction. Pharmacol Rev 19:317-366, 1967 44. Messier B, Leblond CP: Cell proliferation and migration as revealed by radioautography after injection of thymidine-H' into male rats and mice. Am J Anat 106:247-265, 1960 45. Bedard YC, Pinkerton PH, Simon GT: Radioautographic observations on iron absorption by the duodenum of mice with iron overload, iron deficiency and x-linked anemia. Blood 42:131-140, 1973 46. Kaufman RM, Pollack S, Crosby WH: Iron-deficient diet: effects in rats and humans. Blood 28:726-737, 1966 47. Becking GC: Influence of dietary iron levels on hepatic drug metabolism in vivo and in vitro in the rat. Biochem Pharmacol 21:1585-1593, 1972 48. Dallman PR, Goodman JR: The effects of iron deficiency on the hepatocyte: a biochemical and ultrastructural study. J Cell Bioi 48:79-90, 1971 49. Lu AYH, West SB, Yore M, et al: Role of cytochrome b. in hydroxylation by a reconstituted cytochrome P-450-containing system. J Bioi Chern 249:6701-6709, 1974 50. Dallner G, Siekevitz P, Palade GE: Biogenesis of endoplasmic reticulum membranes. II. Synthesis of constitutive microsomal enzymes in developing rat hepatocyte. J Cell Bioi 30:97-116, 1966 51. Guenther TM: Studies on induction of drug-metabolizing enzymes in fetal, neonatal and maternal rats. Ph.D. Thesis. University of Minnesota Graduate School, 1975