Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

The epithelial cells of the human small intestine, the enterocytes, provide the first site for metabolism of orally ingested xenobiotics, including therapeutic drugs. The major catalysts for this metabolism are probably the gene products of the cytochrome P-450 (CYP)¹ gene superfamily (Nelson et al., 1996). As an example, small intestinal metabolism of midazolam by CYP3A4 contributes almost half of the overall first pass metabolism of this drug, and similar extents of metabolism may apply generally for all orally ingested, high turnover, CYP3A4 substrates (Paine et al., 1996). Small intestinal CYP metabolism has the potential to substantially affect the toxicity and therapeutic efficacy of orally administered xenobiotics by diminishing or blocking systemic uptake (Watkins, 1997). Small intestinal biotransformed xenobiotics, which probably will have altered activities, can be absorbed systemically, excreted back into the lumen, or be covalently bound to enterocyte macromolecules and be sloughed off with the short-lived enterocytes.

The capacity of xenobiotics to undergo metabolism in enterocytes, with alteration of putative toxicity or therapeutic efficacy, is based on the forms of CYP expressed in the enterocytes and on whether the xenobiotics are substrates for the expressed CYPs.

Despite the potential for small intestinal CYPs to play major roles in xenobiotic toxicity and pharmacology, there is a relative paucity of knowledge of CYP expression in human enterocytes. It has been clearly established that CYP3A4 is the major form expressed in the human small intestine, that it is inducible by rifampin (Kolars et al., 1992), and that its inhibition by constituents of grapefruit juice significantly increases the uptake of drugs such as felodipine (Lown et al., 1997), cyclosporine A (Ducharme et al., 1993), saquinavir (Kupferschmidt et al., 1998), and ethinylestradiol (Weber et al., 1996). Furthermore, the identity of the intestinally and hepatically expressed forms of human CYP3A4 has been established (Lown et al., 1998). It has also been proposed that small intestinal CYP3A4 acts in concert with p-glycoprotein to block systemic uptake of xenobiotics (Wacher et al., 1998).

However, apart from CYP3A4 there are very limited data available on other human small intestinal CYPs and some of the available data are contradictory. Thus for CYP1A1, which is the predominant form expressed in rat small intestine (Zhang et al., 1996), its protein has been reported to be undetectable in human small intestine (Windmill et al., 1997) and, in contrast, detectable at low levels in intestinal biopsies (Lown et al., 1997). Two studies have reported that CYP1A1 is inducible in human small intestine by omeprazole in six volunteers (Kashfi et al., 1995) and by omeprazole and smoking in larger populations (Buchthal et al., 1995). CYP2D6 protein has been detected at very low levels in human small intestinal biopsies (Lown et al., 1997) and its activity toward 1-bufuralol has also been detected (Prueksaritanont et al., 1995).

Although it has been established that CYP3A5 is the major CYP3A expressed in human colon (Gervot et al., 1996), it is not clear to what extent it is expressed in the small intestine. CYP3A5 mRNA has been detected in the small intestines of all of 19 patients tested (Kivisto et al., 1996) and its protein has been reported to be detectable in 70% of 20 patients (Lown et al., 1994). However, some of the same authors have reported the absence of CYP3A5 mRNA in human small intestine (Kolars et al., 1992).

In this paper we have characterized CYP expression in human enterocytes to provide a basis for assessment of the potential of human small intestine to modify the activity of drugs and other xenobiotics. To achieve this objective, a method for the reproducible isolation of enterocytes from small intestinal villi was developed.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. Peroxidase-conjugated rabbit anti-goat IgG, peroxidase-conjugated rabbit anti-mouse IgG and peroxidase-conjugated goat anti-rabbit IgG were purchased from Sigma Chemical Co. (St. Louis, MO). The bicinchoninic acid protein assay kit was obtained from Pierce Chemical Co. (Rockford, IL). The enhanced chemiluminescence kit was purchased from Amersham (Arlington Heights, IL). The TRI reagent was purchased from Molecular Research Center, Inc. (Cincinnati, OH). The reverse transcriptase-polymerase chain reaction (RT-PCR) kit and other PCR reagents were from Perkin-Elmer (Branchburg, NJ).

Isolation of Human Small Intestinal Epithelial Cells and Preparation of Microsomes. Human small intestines were obtained from the International Institute for the Advancement of Medicine (IIAM; Scranton, PA) through the Organ Procurement Organization (OPO). All specimens were collected from brain-dead donors with informed consent for research from next of kin. Samples were protected from ischemic injury by flushing with ice-cold University of Wisconsin solution immediately after vascular clamping and resection. Those intestines derived from organ donors whose pancreas was used for transplantation had a short length of the intestine removed and were thus devoid of a portion of the duodenum. After procurement all specimens were placed in a preservative ice-cold University of Wisconsin solution and shipped on ice. Details of the donors are provided in Table 1. Cold ischemic time, before cell harvest, was less than 24 h. All intestines were in good condition and tested negative for HIV and for hepatitis B and C viruses. Upon arrival, the intestine was cut into 2-foot segments from the proximal end (unless otherwise indicated), each segment was filled with solution A (1.5 mM KCl, 96 mM NaCl, 27 mM sodium citrate dihydrate, 8 mM KH₂PO₄, 5.6 mM Na₂HPO₄, 40 µg/ml phenylmethylsulfonyl fluoride, pH 7.4), clamped, and gently agitated at 4°C for 5 to 10 min. The solution was drained and the enterocytes were then eluted from the small intestine sections with solution B (PBS, pH 7.2, containing 1.5 mM EDTA, 3 U/ml heparin, 40 µg/ml phenylmethylsulfonyl fluoride, and 0.5 mM dithiothreitol) as described previously for isolating rat enterocytes (Fasco et al., 1993). The elution process was repeated three to four times to collect eluted cells. Microsomes of small intestinal epithelial cells were prepared as described previously (Fasco et al., 1993). Microsomes were stored at 80°C before use. Microsomal protein concentrations were determined using bicinchoninic acid with bovine serum albumin as a standard, according to the method of the manufacturer (Pierce Chemical Co., Rockford, IL).

Histology. Histologic evaluations were undertaken to establish the source of eluted intestinal epithelial cells. Sections (1 cm) of small intestine were collected at the distal end of the first 2-foot segment. Sections were collected after 0, 1, 2, 3, 4, 5, and 6 15-min periods of shaking of the small intestine containing elution buffer B at 4°C. In some cases, buffer B was modified to contain 5 mM EDTA. All sections were fixed in neutral-buffered formalin, stained with hematoxylin and eosin and examined microscopically.

RNA Isolation and RT-PCR. Total RNA was prepared from human small intestinal epithelial cells eluted from the first segment of the intestines, according to the method of Chomczynski (1993) with the use of TRI Reagent. RNA concentration and purity were determined spectrally, and the integrity of the RNA samples was assessed by ethidium bromide staining after agarose gel electrophoresis. RT-PCR was performed using 0.5 µg of total RNA from small intestine as described recently (Zhang et al., 1998), except that the RNA samples were first treated with DNase I at 37°C for 30 min (Huang et al., 1996). PCR primers used for amplifying cDNA sequences were designed using DNASIS software (Hitachi Software Engineering America, San Bruno, CA) and synthesized by the molecular genetics core facility at this institution. The sequences of these primers are as described previously (Huang et al., 1996) and are shown in Table 2.

Immunoblot Analysis. Microsomal proteins were separated by SDS-polyacrylamide gel electrophoresis as described previously (Laemmli, 1970) in 10% polyacrylamide gels. Proteins were electrophoretically transferred to nitrocellulose sheets (Towbin et al., 1979), which were then treated with 5% nonfat dry milk in 20 mM Tris-HCl (pH 7.4), containing 0.5 M NaCl and 0.05% Tween-20 (TBST) for 1 h at room temperature, incubated with a primary antibody in TBST containing 2.5% milk for an additional 1 h, washed with TBST, and then incubated with a secondary antibody at 1/10,000 dilution in TBST containing 2.5% milk. Polyclonal antibodies to human CYP3A4, CYP3A5, CYP1A1/2, CYP1B1, CYP2E1, and rat CYP2C6, and monoclonal antibodies to human CYP2D6 and CYP2E1 were all purchased from Gentest Co. (Woburn, MA). The secondary antibody was peroxidase-labeled rabbit anti-goat IgG, goat anti-rabbit IgG, or rabbit anti-mouse IgG and was detected with an enhanced chemiluminescence kit.

Other Methods and Materials. Spectral determination of total CYP was performed according to published procedures (Omura and Sato, 1964). Formaldehyde formed from the N-demethylation of erythromycin was measured using the method of Nash (1953).

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

The progression of elution of epithelial cells (enterocytes) from human small intestine as a consequence of exposure to an EDTA-containing buffer at 4°C is shown in Fig. 1. The photomicrographs represent the intestine after 0, 15, 30, and 45 min of gentle shaking in elution buffer, solution B. Before incubation in this buffer the villous epithelial cells were generally slightly detached from the lamina propria, possibly as a consequence of mechanical forces during transportation of the intestine. A similar detachment was not noted with rat small intestine (Fasco et al., 1993). Although all villous epithelial cells were eluted from the intestine by three 15-min incubations in elution buffer, crypt cells remained intact. An increase in the EDTA concentration in the elution buffer to 5 mM only produced minimal (<10%) removal of crypt cells.

View larger version (144K): [in this window] [in a new window]   Fig. 1.   Photomicrographs of human small intestinal tissue (2 ft from proximal end). A, before incubation in elution buffer showing columnar epithelium of villi and crypts of Lieberkuhn. Surface epithelium is intact but is detached from the lamina propria of the villi. B, after 15-min incubation in elution buffer at 4°C, columnar epithelium is still intact but detached from laminar propria. Crypt cells are intact. C, after 30-min incubation in elution buffer, epithelium of villi almost completely absent. Crypt epithelium intact. D, after 45-min incubation in elution buffer, epithelium of villi completely removed but crypt epithelia essentially intact. LP, lamina propria; VE, villous epithelium; CEA, crypt epithelium attached; VEF, villous epithelium-free. Stain, hematoxylin and eosin; magnification, 300×.

To examine which forms of CYP are expressed in human small intestine, RT-PCR experiments were conducted to detect CYP m-RNAs in total RNA from small intestinal epithelial cells eluted from the first segment of the intestine. PCR products were analyzed by agarose gel electrophoresis and the results are summarized in Table 3. Use of primers specific for CYP1A1, CPY2C, CYP2D6, CYP2E1, CYP3A4, and CYP3A5 resulted in bands of the predicted sizes in all four individual intestinal samples tested. However, CYP1A1 and CYP2E1 signals were very weak. CYP1B1 mRNA was detected in some human small intestinal epithelial cells (two of four). In contrast, mRNAs for CYP1A2, CYP2A6, CYP2A7, CYP2B6, CYP2F1, CYP3A7, and CYP4B1 were undetectable by ethidium bromide staining of PCR products.

CYP protein expression in human enterocyte microsomes was examined by immunoblot analysis for those forms with detected mRNA expression and the results are summarized in Table 3. The antiserum against rat CYP2C6, which cross-reacts with human CYP2C8, 2C9, and 2C19, gave a signal in all samples from the 10 intestines tested, but the antiserum against CYP1A1 detected a very weak signal and only in two of the eight human samples examined. CYP1B1, CYP2D6, and CYP2E1 were not detected. The antiserum against CYP3A4 gave a strong signal with all 10 human samples, which is consistent with previous reports that CYP3A is the major CYP form in human small intestine. However, this antibody also cross-reacts with CYP3A5, which cannot be separated from CYP3A4 under the conditions used. To distinguish CYP3A5 from CYP3A4, a specific antibody to human CYP3A5 was used to detect CYP3A5 protein in human small intestinal microsomes. No CYP3A5 was detected in any of the 10 human samples tested. However, when protein loading was increased from 10 to 15 µg/lane and the gel was overexposed, one of the 10 intestinal samples yielded a weak band corresponding to CYP3A5. Figure 2 shows a representative immunoblot that demonstrates the ready detection of CYP3A4 protein and the lack of detection of CYP3A5 protein.

View larger version (11K): [in this window] [in a new window]   Fig. 2.   Immunoblot analysis of CYP3A4 and CYP3A5 expression in human small intestinal epithelial cells. Microsomal samples from human small intestinal epithelial cells (lane 1) or from baculovirus-expressed CYP3A4 or CYP3A5 (lane 2) were submitted to SDS-polyacrylamide gel electrophoresis, transferred to a nitrocellulose membrane, and examined immunochemically with goat antiserum to CYP3A4 (A) or rabbit antiserum to CYP3A5 (B) as described in Experimental Procedures. Antibodies were used at 1:500 dilution for anti-CYP3A4 and 1:1000 dilution for anti-CYP3A5. Microsomal proteins from human small intestine were applied at 10 µg in each lane. Baculovirus-expressed CYP3A4 or CYP3A5 was loaded at 1 µg each.

To examine CYP distribution in different parts of the small intestine, total CYP content and CYP3A activity were measured in segments along the entire length of small intestines. Maximal CYP content varied between 0.06 and 0.18 nmol/mg and maximal CYP3A4 erythromycin N-demethylase activity varied between 0.30 and 0.76 nmol/min/mg microsomal protein for the 10 intestines under investigation. As shown in Fig. 3 and Table 4 for a representative intestine (number 9), although total microsomal protein yield decreased dramatically from the duodenum to the ileum, CYP content, as a function of microsomal protein content, increased slightly in progressing from the duodenum to the jejunum and then decreased toward the ileum. CYP3A activity generally correlated with the total CYP distribution pattern. To confirm the distribution of CYP3A4 and CYP2C along the length of the small intestine, immunoblot analysis was performed with polyclonal antibodies to the two forms as shown for a representative intestine (number 7; Fig. 4). The results show that the contents of both CYP forms decreased toward the distal end of the small intestine, but the levels of each form varied differently as a function of position along the intestine.

View larger version (14K): [in this window] [in a new window]   Fig. 3.   Distribution of CYP content (), CYP3A activity (), and microsome protein content () along the length of a representative human small intestine. The entire length of human small intestine was cut into alternating 2-foot and 1-foot segments from the proximal end with segment 1 being nearest to the proximal end and segment 6 nearest to the distal end. Analyses were undertaken with the 2-foot segments only. CYP contents were determined spectrally. CYP3A activity was assayed by measuring erythromycin N-demethylation using the method of Nash (1953).

View larger version (17K): [in this window] [in a new window]   Fig. 4.   Immunoblot analysis of the distribution of CYP3A4 (A) and CYP2C (B) along the length of a representative human small intestine. Microsomes from each segment of human small intestine were prepared as described in the legend to Fig. 3. Microsomes were submitted to immunoblot analysis with antisera to human CYP3A4 and rat CYP2C6 as described in the legend to Fig. 2. Anti-CYP2C6 was used at 1:500 dilution.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

The method of EDTA-mediated elution of human enterocytes, applied in these studies to the human small intestine for the first time, offers advantages over mechanical scraping of the mucosa (Paine et al., 1997); villous and crypt cells can be separated for investigation individually and the more gentle nature of the isolation procedure is less likely to damage the cells. Human and rat (Fasco et al., 1993) villous enterocytes exhibited similar susceptibility to release and elution from the small intestine by EDTA. However, in contrast to the rat small intestine where 1.5 mM EDTA effectively removes crypt cells after an extended period of incubation, 5.0 mM EDTA only removed up to 10% of crypt cells from human small intestine after a similar incubation period. A separation of human villous and crypt cells can thus be achieved by first eluting the villous cells with EDTA and then releasing the residual crypt cells mechanically by scraping.

The current studies support the well established observation that CYP3A4 is the predominant CYP form in the human small intestine (Kolars et al., 1992) and in addition indicate that only a very limited number of other CYPs are expressed in this organ compared with the number expressed in the liver. Thus, based on RT-PCR and Western immunoblot analyses of the 10 human small intestines derived from Caucasian males and females, only CYP3A4, CYP2C, and CYP1A1 (the latter being weakly expressed and only in two of the eight donor intestines investigated) were detectable of the 14 forms for which we tested. However, it was not possible to definitively resolve whether these expressed forms were constitutive or induced, because complete donor histories of exposure to all potential inducers are not available. Our detection of variable CYP1A1 protein in some of the intestinal microsomes is consistent with conflicting reports of its expression (Windmill et al., 1997; Lown et al., 1997) and, based on reports of CYP1A1 inducibility in human small intestine (Kashfi et al., 1995; Buchthal et al., 1995), the detected CYP1A1 is likely to be induced rather than constitutive. A recent report indicates that CYP1A1 was detected in 3 of 18 human small intestinal preparations using an anti-rat CYP1A1/2 antibody (Paine et al., 1999). Although expression of CYP2D6 in human small intestine has been reported (Prueksaritanont et al., 1995; Lown et al., 1997), we were unable to detect this protein in any of the 10 intestinal enterocyte preparations, despite our detection of its mRNA. Our detection of CYP2C in human small intestine confirms a similar observation in an earlier study (DeWaziers et al., 1989) and it is possible that substrates for these forms could undergo first-pass intestinal metabolism.

In the 10 intestines investigated in this study, CYP3A4 protein was strongly expressed in all regions of all samples, whereas CYP3A5 protein was not detected (with the exception of a very weak band in one case) in any of the samples. A recent study of 20 enterocyte preparations detected a band on immunoelectrophoresis "presumed" to be CYP3A5 in four of the samples, but positive identification of CYP3A5 was not made (Paine et al., 1997). Based on our studies it is apparent that CYP3A5 protein expression, at least in the Caucasian small intestine, is at best a rare event. The contrast between this observation and the reported predominant expression of CYP3A5 in the human colon (Gervot et al., 1996) raises interesting questions concerning the reason for the difference, the consequences of the difference, and the underlying regulatory differences in these contiguous organs.

The observed decreases in enterocyte and enterocyte microsome yields, as a function of distance along the small intestine from the duodenum to the ileum, is consistent with an earlier report (Paine et al., 1997) and with reported data on the rat small intestine (Fasco et al., 1993). Both total CYP concentrations and CYP3A4 activities, however, although generally following the trend, do increase slightly along the length of the duodenum before decreasing significantly toward the ileum. Our Western immunoblot data also indicate that concentrations of CYP3A4 and CYP2C, although both decreasing dramatically in the distal small intestine, do not decrease exactly in concert with one another.

The maximal spectrally determined CYP content of the small intestines varied over a range of 0.06 to 0.18 nmol/mg, which compares well with a previous report of a CYP content range of 0.03 to 0.21 nmol/mg (Paine et al., 1997). The similarity in the levels of CYP between these studies suggests that the cold ischemic time of <24 h achieved during delivery of intestines in our study compared with a time of <3 h for the previously reported study (Paine et al., 1997) did not compromise the integrity of the total CYP content of the small intestines.

In conclusion, this study has confirmed the predominant role of CYP3A4 among the CYPs expressed in the human small intestine and has shown that CYP2C is also expressed in this organ. CYP3A5 plays little, if any, role in human small intestinal metabolism. The majority of CYP activity resides in the proximal region of the small intestine.