![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Cell and Molecular Pharmacology and Experimental Therapeutics (T.W., U.K.W.), and Department of Cell Biology and Anatomy (D.S.), Medical University of South Carolina, Charleston, South Carolina; and The MatTek Corporation, Ashland, Massachusetts (M.K.)
(Received October 19, 2005; Accepted December 14, 2005)
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
; Romano et al., 1999; Besarati Nia et al., 2000). Tobacco smoke components, including polycyclic aromatic hydrocarbons (PAHs), N-nitrosamines, and aromatic amines, are all thought to be potent carcinogens. Metabolic activation of tobacco-associated PAHs like benzo[a]pyrene (BaP) to reactive metabolites and the DNA adducts formed have been postulated to be central to the carcinogenic process of PAH-induced cancers (Hecht, 2002). However, how the carcinogens arrive at the critical cells from which the oral cancers develop, i.e., the basal cells of the highly stratified oral epithelium, is not clearly understood, mainly because of lack of a suitable biological model to study this question.
Epidemiological studies have demonstrated a protective role of fruits and vegetables in oral cancers, presumably mediated by their content of polyphenols, particularly flavonoids (Block et al., 1992; Levi et al., 1998; Sakagami et al., 1999). However, how these potentially protective dietary compounds may get to the critically important basal cells in the complex stratified oral epithelium is not well understood, mainly, again, because of lack of a suitable biological model. Limited studies in this area using epidermis and oral mucosa from pigs (Squier and Lesch, 1988; Du et al., 2000) and humans (Bánóczy et al., 2003) have suggested that small organic molecules indeed can penetrate these highly complex stratified cell layers. However, the mechanisms involved are poorly understood.
In a recent study, we examined the bioactivation of BaP and its binding to cellular DNA in a human oral epithelial cell line cultured as monolayers, as well as the protective role of several polyphenols (Wen and Walle, 2005). This study explored the molecular mechanisms of BaP bioactivation and its inhibition by polyphenols. Highly specific for oral cells, BaP induced CYP1B1 mRNA and protein expression and, to a much lower extent, CYP1A1 expression. Selected polyphenols blocked this induction. It now became important to determine whether these molecular events may occur in the physiologically more relevant stratified oral epithelium, using a novel bioengineered human tissue construct closely resembling the normal gingival tissue. The main hypothesis under test was that both carcinogens and potentially protective polyphenols can reach the basal cells of this complex multilayer tissue.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
BaP was used as the model carcinogen, and the radiolabeled forms of quercetin, resveratrol, and ellagic acid, common dietary polyphenols present, for instance, in onions and apples, grapes and peanuts, and berries, respectively, were used in the transport studies. For the chemoprevention studies, the choice of polyphenols in this study was mainly based on their abilities to affect CYP1A1/1A2/1B1 activities (Wen et al., 2005).
Permeability Experiments. EpiOral tissues grown in inserts were equilibrated in a 37°C humidified incubator (5% CO2) with 0.3 ml of assay medium in 24-well plates for 1 h; i.e., only the basolateral side of the tissue was exposed to medium. Assay medium (0.5 ml) containing the various radiolabeled test compounds (quercetin, resveratrol, ellagic acid, mannitol, and BaP) was added to the apical side of the tissues. At specified intervals, each tissue insert was moved to another well containing 0.3 ml of assay medium. Basolateral and apical solutions were assayed for total radioactivity. The tissues were digested with 0.5 M NaOH and assayed for radioactivity.
BaP DNA Binding. For measurements of DNA binding, the generally preferred method is the 32P-postlabeling assay (Reddy, 2000). However, because BaP has been well established to bind covalently to DNA after enzymatic bioactivation (Xue and Warshawsky, 2005), we used the simpler radiolabeling assay, which has been modified and refined (Shibutani et al., 1999; Walle et al., 2003). EpiOral tissues were incubated with 1 µM [3H]BaP alone for specified times or for 24 h together with 25 µM quercetin, resveratrol, chrysin, or 5,7-dimethoxyflavone. The tissue inserts were washed, and the tissues peeled off the membranes and homogenized with a Polytron homogenizer in 1 ml of 1% SDS/10 mM EDTA/20 mM Tris-HCl (pH 7.4). The homogenate was then incubated at 37°C for 30 min with 0.2 mg of RNase A and 0.5 mg of proteinase K (Shibutani et al., 1999). After extraction with phenol/chloroform/isoamyl alcohol (four times) and precipitation of DNA with sodium acetate and ethanol, the pellets were washed twice with ethanol and dissolved in water. The DNA purity (260/280 nm ratio 
1.7) and concentration were determined by UV, and [3H]BaP bound to DNA was determined by scintillation counting.
Expression of CYP1 Isoforms after BaP Treatment. EpiOral tissues were treated for 24 h with 1 µM BaP or dimethyl sulfoxide in medium. The tissues were transferred to microtubes and homogenized in lysis buffer. RNA was isolated using an RNeasy Mini kit (QIAGEN, Valencia, CA). Typical yields were 10 to 15 µg/tissue. RNA (1 µg) was reverse-transcribed using an Omniscript RT kit (QIAGEN). CYP1B1 expression was then determined using semiquantitative PCR with normalization to ß-actin, essentially as described previously (Port et al., 2004). The PCR products mixed with loading buffer were separated on a 1.5% agarose gel with ethidium bromide, the gel was photographed under UV light, and the bands were quantified (Fluor-S Multi-Imager; Bio-Rad, Hercules, CA).
Tissue Localization of [14C]Quercetin. [14C]Quercetin (25 µM) in culture medium was added on the apical side of EpiOral tissues. After 1, 15, or 30 min, or 2, 4, or 8 h, the tissues were rinsed and fixed with 4% paraformaldehyde. Rinsed tissues on membrane supports were cut from the plastic cell culture inserts and embedded in OCT media. Frozen sections (12 µm thick) were cut on a cryostat. The sections were mounted on positively charged histology slides, air-dried, and stored at –20°C.
The sections were examined on a compound Leica DMLB microscope (Leica Microsystems GmbH, Wetzlar, Germany) in phase contrast and epifluorescence mode. A 4',6-diamidino-2-phenylindole filter set was used (excitation band pass 340–380 nm, emission long pass 430 nm). Fluorescence images obtained with a 3CCD Hamamatsu camera (Hamamatsu Photonics, Hamamatsu City, Japan) were superimposed on the phase-contrast images, and depth of penetration into the multilayered tissue construct (in micrometers) was measured based on the drop in fluorescence intensity (Marti et al., 1999).
To measure the penetration of total radioactivity (quercetin + metabolites and degradation products), slides with mounted sections were exposed for 48 h at 4°C to Kodak autoradiographic film (Eastman Kodak, Rochester, NY). Pictures of the developed film were taken on the same microscope and matched to phase contrast and fluorescence images of the same sections. The depth of radiolabel penetration was determined in the same fashion as for the fluorescence.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
|
When comparing the tissue uptake at 3.5 h with the transport, the compounds studied behaved quite differently (Table 1). As expected for a paracellular transport marker, mannitol showed minimal tissue uptake. In contrast, the tissue uptake of resveratrol and quercetin was substantial. This was also true for ellagic acid, which had very low transepithelial flux. Not unexpectedly, the highly lipophilic BaP demonstrated very large tissue accumulation with little transepithelial transport.
DNA Binding of BaP and Inhibition by Polyphenols. The carcinogenic process is thought to involve metabolic activation of BaP and binding to DNA. To examine this process in the EpiOral tissue, we first determined the accumulation of BaP over time (Fig. 1A). The accumulation was very slow, with maximum tissue levels reached at 24 h.
|
Induction of BaP-Activating Cytochrome P450 Isoforms. Previous studies have indicated that CYP1A1 and CYP1B1 can oxidize BaP, eventually leading to covalent binding of BaP to cellular DNA (Kim et al., 1998; Hecht, 1999; Nebert et al., 2004). When EpiOral tissue was treated with 1 µM BaP as compared with dimethyl sulfoxide for 24 h, CYP1B1 mRNA was clearly induced (Fig. 2A). CYP1A1 and 1A2 mRNAs were less induced. A semiquantitative comparison is given in Fig. 2B. Thus, CYP1B1 is the isoform most effectively induced in the epithelial tissue. Attempts to measure protein expression by Western blotting, as previously done in SCC-9 cell monolayers (Wen and Walle, 2005), failed because of very low expression levels.
|
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
). The higher value for resveratrol may also be due to transcellular absorption, which is highly efficient for this lipid-soluble polyphenol, at least in cell monolayers (Kaldas et al., 2003). However, this does not exclude the possible involvement of specific transporters for these molecules. The low value for ellagic acid is consistent with the previous finding of a basolateral transport barrier for this polyphenol (Whitley et al., 2003). All three polyphenols showed considerable cell uptake, consistent with their relatively lipophilic nature, in contrast to mannitol. The very low transepithelial transport for BaP is most likely due to the extensive cellular uptake of this highly lipophilic compound. The cell uptake of BaP is thus about 10-fold higher than that for the polyphenols (Table 1).
With the focus of our study on the carcinogen BaP, we observed that its cellular uptake progressed very slowly in the oral tissue construct, with a maximum uptake achieved in about 24 h (Fig. 1A). Also, DNA binding of BaP occurred in the tissue. Statistically significant binding occurred after 6 h of exposure to BaP (Fig. 1B), which further increased after 24 h of exposure. Still, the DNA binding of BaP in the tissue was low compared with that of monolayers of human oral epithelial cells (Wen and Walle, 2005), probably because of the lower bioavailability of BaP and/or lower bioactivating enzyme levels.
The findings in Fig. 2 clearly indicated that the enzymes required for bioactivation of BaP were expressed in the oral tissue construct. Because both CYP1A1 and 1B1 are capable of metabolizing BaP (Kim et al., 1998; Hecht, 1999; Nebert et al., 2004), it was important to identify which isoforms were expressed and whether they were inducible by BaP in this tissue in a manner similar to that in human oral epithelial SCC-9 cells grown as monolayers (Wen and Walle, 2005). Interestingly, both CYP1A1 and 1B1 mRNA were expressed in uninduced tissue. As shown in Fig. 2, a 24-h exposure to a low (1 µM) concentration of BaP induced all CYP1 mRNAs, but primarily CYP1B1 and CYP1A1, with a very modest expression of CYP1A2. This is similar to recent observations in SCC-9 cells (Wen and Walle, 2005). The relatively higher expression level of CYP1B1 in SCC-9 cells compared with the tissue construct in this study may be due to differences between tongue (SCC-9) and gingival (EpiOral) cells or cancer (SCC-9) versus normal (EpiOral) cells. The present study in the oral tissue construct is probably more physiologically relevant than the SCC-9 cell monolayer study.
When the tissue was coincubated for 24 h with BaP and four different polyphenols, all proposed in various studies to be chemopreventive, there was a significant reduction in the DNA binding (Fig. 1C). Possible mechanisms for this effect include direct inhibition of CYP1A1/1B1 at the protein level, inhibition of CYP1A1/1B1 transcription (Wen and Walle, 2005), or induction of bioinactivating enzymes. Mechanistic questions may be more difficult to resolve in this tissue construct compared with the monolayer culture because of its greater complexity; however, use of the differentiated tissue may make the results more applicable to native oral tissue function.
The extent and site of penetration of BaP as well as the polyphenols in the oral tissue construct with respect to time of exposure could not be determined from these experiments. We therefore used microscopy coupled with the fluorescence of the polyphenol quercetin (Fig. 3A), as well as microautoradiography, when using [14C]quercetin in our incubation experiments (Fig. 3B). After 2 h, the native fluorescence of quercetin had penetrated about one third of the oral epithelial tissue construct. Conversely, at that time, the radioactivity had penetrated the entire tissue layer (Fig. 3C), i.e., reached all the way down to the basal cell layer from which the oral cancer develops. The reasons for this discrepancy can only be speculated on. Quercetin has previously been shown to be oxidized intracellularly, presumably by hydrogen peroxide and peroxidases, and to bind covalently to proteins (Walle et al., 2003). When quercetin is oxidized, multiple reactions occur, including disruption of the C-ring, most likely resulting in loss of the fluorescence (Nordström, 1968; Nishinaga and Matsuura, 1973; Nishinaga et al., 1979). Figure 3 gives further strong evidence that once quercetin enters cells, it is effectively altered by cellular oxidation. Figure 3 also shows that quercetin metabolites/degradation products reach the basal layer of the oral epithelium. Previous studies (Boulton et al., 1999) have indicated that degradation products of quercetin may have beneficial health effects.
The time for maximum induction of CYP1A1/1B1 by BaP was not determined in these experiments. It is clear that the present study only provides a limited view of transport as well as metabolic processes in this highly complex oral epithelial tissue. Paracellular versus transcellular transport has not been clearly addressed. Also, the role of possible transporters in this tissue, as well as metabolic enzymes other than CYP1A1/1B1/1A2, is unknown.
It should be emphasized that for some of the polyphenols, the concentrations used in this study can easily be achieved in the diet. For example, resveratrol concentrations can reach 35 µM in red wine and 8 µM in peanuts (Sanders et al., 2000), and total quercetin concentrations in cooked onions can reach 1000 µM (Walle et al., 2000). Other polyphenols are present in the diet at lower levels.
In summary, in a bioengineered construct of the multilayered stratified human oral epithelial tissue, the data generated is consistent with the notion that smoking-derived carcinogens, e.g., BaP, are bioactivated by BaP-induced CYP1B1/1A1 to bind to DNA and to initiate the carcinogenic process. We have also shown that multiple polyphenols believed to be cancer-chemopreventive agents inhibit this DNA binding in this tissue. Using microscopic techniques, we generated evidence that these compounds indeed will penetrate this complex tissue and thus reach the highly essential basal cell layer. It can be concluded that bioengineered human oral tissue can serve as a novel tool in studies pursuing improved prevention/treatment of oral cancer. It is also clear that this complex, multilayered tissue will require more in depth studies to better understand both transport, bioactivation, and chemoprevention mechanisms.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
This work was presented in part at the Experimental Biology 2005 meeting in San Diego CA, April 2–6, 2005: Walle T, Walle UK, Sedmera D, and Klausner M (2005) Transport of dietary polyphenols through the multilayer oral epithelium: implications for prevention of oral cancer. FASEB J 19:A543.
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
ABBREVIATIONS: PAH, polycyclic aromatic hydrocarbon; BaP, benzo[a]pyrene; RT-PCR, reverse transcription-polymerase chain reaction.
Address correspondence to: Dr. Thomas Walle, Department of Cell and Molecular Pharmacology and Experimental Therapeutics, Medical University of South Carolina, 173 Ashley Avenue, P.O. Box 250505, Charleston, SC 29425. E-mail: wallet{at}musc.edu
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. Y. Kim, J.-Y. Chung, J.-E. Park, S. G. Lee, Y.-J. Kim, M.-S. Cha, M. S. Han, H.-J. Lee, Y. H. Yoo, and J.-M. Kim Benzo[a]pyrene Induces Apoptosis in RL95-2 Human Endometrial Cancer Cells by Cytochrome P450 1A1 Activation Endocrinology, October 1, 2007; 148(10): 5112 - 5122. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
) and autoradiography (
).