Advan. Enzyme Regul., Vol. 41, pp. 159–175, 2001 # 2001 Elsevier Science Ltd. All rights reserved Printed in Great Britain 0065-2571/01/$ - see front matter
REGULATION OF RETINOIC ACID METABOLISM LUONG LUU, HEATHER RAMSHAW, ALI TAHAYATO, ANDREW STUART, GLENVILLE JONES, JAY WHITE and MARTIN PETKOVICH* Cancer Research Laboratories, Department of Biochemistry and Pathology, 355 Botterell Hall, Queen’s University, Kingston, Ont., Canada K7L 3N6 INTRODUCTION
Retinoic acid (RA), the most active form of vitamin A, is an important regulator of pattern formation during embryonic development and is necessary for the maintenance of epithelial tissues in the adult (1–4). RA acts through the regulation of gene expression mediated by speciﬁc nuclear receptors. The activity of RA in speciﬁc tissues is also controlled by regulating its availability, balancing the rate of RA synthesis with that of destruction. The consequences of RA excess or deﬁciency can be severe, particularly during embryogenesis, where local control over the expression of RA synthetic and catabolic enzymes is tightly controlled. Therapeutically, RA has been shown to be highly eﬀective in the treatment of skin disorders (5–7) and has promising anticarcinogenic and antitumor properties (2, 4, 8). It is important to consider the eﬀects of exogenous RA treatment on normal metabolic processes regulating RA levels; however, it is only recently that the molecular tools to do so have been made available through the discovery of genes encoding RA synthesizing and catabolizing enzymes. This paper will focus on one of these enzymes, P450RAI (CYP26), a cytochrome P450 that plays an important role in regulating RA levels in developing tissues during embryogenesis and in adult epithelia. The characterization, activity and expression of this enzyme will be discussed in context with its role in development. This overview of recent research will establish that (1) P450RAI is a highly conserved component of retinoid signaling which speciﬁcally metabolizes the all-trans isomer of RA, (2) P450RAI expression and activity are strongly induced by exogenous RA forming an autoregulatory feedback loop to control RA levels, and (3) the expression of P450RAI in embryos and in the adult is consistent with it having a protective role, preventing undue exposure of sensitive cells and tissues to RA.
*Corresponding author. 159
L. LUU et al.
Retinoic Acid Signaling The activity of RA in a given tissue is determined at a number of levels. First, the eﬀects of RA on gene expression are principally mediated by two families of retinoid nuclear receptors comprised of three subtypes each, retinoic acid receptors (RARs a, b, and g) and retinoid X receptors (RXRs a, b, and g) (9). RARs and RXRs commonly participate together in the form of heterodimers to regulate gene expression. Most tissues, especially during embryonic development, express one or more RAR and RXR subtypes in diﬀerent combinations possibly giving rise to diﬀerent responses to RA in diﬀerent tissues (10). Second, RARs activate target genes at speciﬁc short target sequences known as RA response elements (RAREs) (11). Typically, an RARE is comprised of two direct repeats of the motif, 50 PuGTTCA-30 separated by a 5 bp spacer; however, various polymorphic forms of RAREs have been characterized, having 1, 2 as well as 5 base pair spacers (12,13). Several studies suggest that the form of the RAREs may preferentially bind diﬀerent heterodimeric RAR/RXR pairs (14). On a DR5 element, the heterodimer is conﬁgured such that the RXR is in the 50 position, whereas on a DR1 element, the opposite orientation appears to be more stable. However, the RAR/RXR-DR1 conﬁguration appears to be transcriptionally inactive (12,13). In contrast, RXRs can form homodimers on DR1, which are transcriptionally active in the presence of 9-cis RA. Third, at least two isoforms of RA, all-trans, and 9-cis RA, are ligands for these receptors; RARs are activated by both isoforms while RXRs appear to be exclusively activated by 9-cis RA. It is not clear at present how interconversion between the two forms is controlled; however, the balance of all-trans RA and 9 cis-RA may be important for RA activity. On a fourth level of control, the distribution of RA appears to be an important determinant in the patterned regulation of RA responsive genes, especially in developing tissues. Tight spatial and temporal control over RA synthesis and catabolism may therefore be critical in establishing regional distribution patterns of RA. Retinoic Acid Synthesis Control of RA tissue distribution is established by balancing expression of RA synthesizing and RA catabolizing enzymes. There are a number of retinoid binding components that function in vitamin A metabolism and storage pathways (15). Retinyl esters and -carotene are ingested and converted to all-trans retinol in the intestine, which is then reconverted to retinyl esters for storage, mainly in the liver (16). Demand for retinol results in the release of retinol bound to plasma retinol-binding protein from the liver. Retinol bound to RBP is eﬃciently taken up by many extrahepatic tissues including eye, skin, adipose tissue, kidney, testes, lung and bone marrow, all important target tissues (17). Conversion of retinol to the active
AUTOREGULATION OF RA METABOLISM
forms of RA occurs in many diﬀerent tissues; however, the exact biochemical mechanisms have not been ﬁrmly established (18). The eﬃcient conversion of retinol to retinaldehyde and retinaldehyde to retinoic acid by the retinol and retinaldehyde dehydrogenases, respectively, is facilitated by the presence of cellular retinol-binding proteins (CRBPs) (18, 19). Several retinaldehyde dehydrogenases have already been implicated in the irreversible conversion of retinaldehyde to the active RA (20). RALDH-2 is thought to be a key enzyme in the localized production of RA, especially during development since it exhibits expression patterns consistent with those of a retinoid-responsive LacZ reporter transgene. Furthermore, RALDH-2 knock-out mice have severe developmental defects and die at mid-gestation (21, 22). RA Metabolism The irreversible conversion of retinaldehyde to RA in tissues where RALDH-2 is expressed creates a situation where RA is committed to either activate receptors to regulate RA responsive genes or is catabolized to inactive forms by the RA metabolic machinery and eliminated. RA catabolism thus governs tissue sensitivity to RA. The metabolism of RA is thought to be initiated by hydroxylation either at the C4-, or C18-position of the -ionone ring of RA (17–19, 23). The C4-hydroxylation step is mediated by cytochrome P450 activity, evidenced by the ability of broadspectrum P450 inhibitors such as ketoconazole and liarozole to block 4hydroxylation (24–28). In certain tissues, including testis, skin and lung and in numerous cell lines, such as NIH3T3 ﬁbroblasts, HL60 myelomonocytic leukemic cells, F9 and P19 murine embryonal carcinoma cells, MCF7 human breast cancer cells and HeLa human cervical cancer cells, RA metabolism can be induced by RA pretreatment (27, 29–31). P450RAI (CYP26) is a cytochrome P450 enzyme which speciﬁcally metabolizes RA and is likely responsible for much of the RA inducible RA metabolism observed in the earlier studies described above. We ﬁrst cloned and characterized from zebraﬁsh, cDNAs encoding a cytochrome P450dependent enzyme (P450RAI) which is induced by RA and metabolizes RA to more polar derivates including 4-hydroxy retinoic acid (4-OH RA), 18hydroxy retinoic acid (18-OH RA) and 4-oxo retinoic acid (4-oxo RA) (32). The identiﬁcation of P450RAI gene is an important step in our understanding of RA signaling but its presence has been known since Roberts et al. (1979) ﬁrst postulated that the catabolism of RA was mediated by a P450 enzyme (29, 30). More recently, we and others (33, 34), have isolated cDNAs which encode the full-length human and mouse P450RAI orthologs whose expression, like that of the ﬁsh cytochrome, is highly inducible by RA. Homologs have also been isolated from human, mouse, chick and xenopus all exhibiting a high degree of sequence conservation (35–37). There is
L. LUU et al.
extensive identity between the human and ﬁsh P450RAI genes which overall is 68% at the amino acid level (over 90% between mouse and human). In this paper, we have examined the regulation of expression of this enzyme by its substrate, RA.
MATERIALS AND METHODS
Chemicals. All-trans-retinoic acid (RA) and 9-cis-retinoic acid (9-cis-RA) were purchased from Sigma (St. Louis, MO). TRIzol reagent for RNA preparation was purchased from Gibco BRL (Burlington, ON). QuickHyb solution for northern blot hybridization and the Prime It II random primer kit were purchased from Stratagene (La Jolla, CA). Hybond-C nitrocellulose membrane for northern blot analysis was purchased from Amersham Canada Limited (Oakville, ON). For preparation of 32P radiolabeled probes, (a 32P) dATP was purchased from NEN Research Products (Boston, MA). All other chemicals used were of electrophoresis or HPLC grade and were obtained from a variety of commercial sources. Cell culture and manipulation. Cell culture materials were purchased from Gibco BRL (Burlington, ON). MCF-7 cells were purchased from ATCC (Rockville, MD). Cells were maintained at 378C in a humidiﬁed atmosphere of 5% carbon dioxide in air and kept in the dark during experimentation. MCF-7 cells were grown in plastic culture dishes (diameter 100 mm) as monolayers and cultured in 10 ml minimal essential media (MEM) supplemented with 10 % fetal bovine serum, insulin (10 ng/ml), sodium pyruvate (500 nM), l-glutamine (2 mM), non-essential amino acids (100 nM), penicillin (5 mg/ml), streptomycin (5 mg/ml), fungizone (200 ng/ ml), and gentamycin (10 mg/ml). MCF-7 cells were grown to 80 or 90% conﬂuence prior to treatment. Cells were treated with RA or vehicle (0.1% (v/v) DMSO) alone prior to extraction of total RNA. Total RNA was extracted using TRIzol reagent as outlined by the manufacturer and redissolved in DEPC-treated distilled and deionized water. For removal of RA in washout experiments in MCF-7 cells, the media were removed by aspiration, and cells in the ﬂask were rinsed three times with 1 PBS solution supplemented with 1% fetal bovine serum. Analysis of P450RAI expression. P450RAI expression was determined by Northern blotting as described elsewhere (38). Brieﬂy, 20 mg of total RNA were electrophoresed in a 1% agarose gel containing 0.66 M formaldehyde. RNA was transferred to Hybond-C nitrocellulose membrane by capillary action using 10 SSC buﬀer. Radiolabelled P450RAI and GAPDH cDNA
AUTOREGULATION OF RA METABOLISM
probes were generated by random priming of the respective cDNAs using the Prime-It II kit as outlined by the manufacturer. Prehybridization and hybridization were carried out at 688C in a circulating water bath. Addition of radiolabelled probe and membrane washing following hybridization were carried out as described in the QuickHyb Instruction Manual. Hybridization was detected by autoradiography (Kodak X-OMAT-AR or BIOMAX MR) at 708C, typically for 17 hr. Following hybridization to radiolabelled P450RAI cDNA probe, membranes were stripped using two 30 min washes in heated stripping solution (0.1 SSC, 1% SDS) and then air-dried and hybridized to radiolabeled GAPDH cDNA probe as an internal standard for RNA content. Metabolism analysis. For washout experiments, cells were treated with RA to a ﬁnal concentration of 1 mM or with an equal vol of vehicle alone (DMSO). Cells were washed three times with 1 PBS solution as described above. Following removal of RA, 10 ml MEM was added to cells (MCF-7 cells). At the indicated time points following washout, cells were treated with 575 pM [11,12-3H] RA (a concentration shown not to signiﬁcantly induce RA metabolic activity or P450RAI) for 4 hr, then acidiﬁed with 0.1% acetic acid. Parallel cultures were used to determine P450RAI levels by Northern blot analysis. Aqueous- and lipid-soluble metabolites were separated using a total lipid extraction of the medium as described (39). Aqueous-soluble [11,12-3H] RA metabolites were measured by b-scintillation counting of aliquots of the aqueous-soluble extract.
Concentration-dependent Induction of P450RAI mRNA in MCF-7 Cells Following Treatment with RA RA-inducible RA metabolism was previously described in the human breast epithelial adenocarcinoma-derived MCF-7 cell line (27), and we have shown that P450RAI mRNA was strongly induced in these cells following RA treatment (35). As shown by Northern blot analysis in Fig. 1, when MCF-7 cells were treated with various concentrations of RA for 12 h, both all-trans-RA and 9-cis-RA strongly induced the expression of P450RAI mRNA. All-trans-RA appeared to be more eﬀective at inducing expression of P450RAI mRNA than 9-cis-RA, particularly at concentrations below 1 mM. The 4-oxo- and 4-OH-RA metabolites, two of the major lipid soluble products of P450RAI activity (32), were both found to be poor inducers of P450RAI mRNA (data not shown). A number of non-retinoids were also tested for ability to induce P450RAI message in MCF-7 cells including: 1a, 25-dihydroxy vitamin D3 (1 mM), diethylstilbestrol (4 mM), dexamethasone (up to 1 mM), and the classical
L. LUU et al.
FIG. 1. The levels of P450RAI mRNA induced are dependent on the concentration of RA used for treatment. MCF-7 cells were treated with the indicated concentrations of RA isomer for 12 hr prior to extraction of total RNA and Northern blot analysis of P450RAI and GAPDH mRNAs.
P450 inducers 3-methyl cholanthrene (up to 5 mM) and phenobarbital (9 mM). However, no P450RAI mRNA was detected by Northern blot analysis following treatment with these compounds (data not shown). Time Course of P450RAI mRNA Induction in MCF-7 Cells We were interested in determining the duration of P450RAI mRNA expression in MCF-7 cells following treatment with RA. Fig. 2 shows the results of Northern blot analysis of total RNA extracted from MCF-7 cells at various times following a single treatment with 1 mM all-trans-RA. At 1 h following RA treatment, low levels of P450RAI mRNA are detectable (lane 2), showing that this mRNA was rapidly upregulated in these cells. The levels of P450RAI mRNA were highest between 6 and 12 h of treatment (lanes 6 and 8), but rapidly declined by 24 h (lane 10) and were barely detectable by 48 h (lane 14). RA Dependence of P450RAI mRNA Expression MCF-7 Cells The transient expression of P450RAI mRNA suggested to us that P450RAI expression in MCF-7 cells might be dependent on the continuous presence of RA and that P450RAI protein, once produced, would curtail
AUTOREGULATION OF RA METABOLISM
FIG. 2. Time course of P450RAI mRNA induction by RA in MCF-7 cells. MCF-7 cells were treated with 1 mM RA (+) or DMSO () for the indicated time points prior to Northern blot analysis of total RNA preparations as carried out in Fig. 1.
further induction of P450RAI mRNA. To examine this hypothesis, we ﬁrst performed a washout experiment whereby, following a designated time of exposure to RA, cells were rinsed with PBS solution to remove RA. Fig. 3 shows the results of Northern blot analysis of two sets of MCF-7 cells cultured in parallel and analyzed for P450RAI expression at various time points after treatment with 1 mM all-trans-RA. In one set of cells, RA was removed by washout 12 h after initial treatment (lanes 5–8). The level of P450RAI mRNA declined dramatically within 12 h following removal of RA (cf. lanes 5 and 6). This rapid decline in P450RAI mRNA upon removal of RA is consistent with P450RAI expression being dependent on the continuous presence of RA. To further examine the RA dependence of P450RAI expression, a time course experiment was performed in a manner similar to that in Fig. 2 above except that cells were treated with a second dose of 1 mM all-trans-RA at the 48 h time point. Fig. 4 shows that, after P450RAI mRNA expression has subsided following the ﬁrst RA treatment, the cells are capable of mounting a full induction of P450RAI mRNA in response to RA. RA Metabolic Activity in MCF-7 Cells We next examined the ability of RA-treated MCF-7 cells to metabolize RA. We chose to follow the metabolic activity of MCF-7 cells at various
L. LUU et al.
FIG. 3. Dependence of P450RAI mRNA expression on continuous presence of RA in MCF-7 cells. Two sets of MCF-7 cells cultured in parallel were treated with 1 mM RA for 12 h. RA was then removed in one set of cells (lanes 5–8) but not in the other (lanes 1–4). P450RAI and GAPDH mRNAs in total RNA preparations were analyzed as in Fig. 1.
FIG. 4. Time course of P450RAI mRNA induction in MCF-7 cells following a second treatment with RA. 48 h after initial treatment with 1 mM RA, a second dose of 1 mM RA was added. Total RNA was prepared at the indicated time points after the ﬁrst and second treatments. P450RAI and GAPDH mRNAs in total RNA preparations were analyzed as in Fig. 1.
AUTOREGULATION OF RA METABOLISM
FIG. 5. Correlation of P450RAI mRNA expression and RA-inducible RA catabolic activity in MCF-7 cells. MCF-7 cells cultured in parallel were treated with 1 mM RA for 17 h prior to removal of RA by rinsing with PBS. (A) top panel; Northern blot analysis of P450RAI and GAPDH mRNAs in total RNA preparations obtained at the indicated time points following washout. (B) bottom panel; Measurement of conversion of labeled RA to aqueous-soluble counts.
time points during a RA washout experiment to determine if the decay in mRNA levels following RA removal paralleled the RA metabolic activity. Cells used for Northern blot analysis and metabolic studies were treated in parallel with 1 mM RA for 17 h and then rinsed with PBS and cultured for the remaining time period without RA. Fig. 5 shows that, after 17 h of treatment with RA, cells expressed high levels of P450RAI mRNA (panel A, lane 1). At 12 h post-washout, levels of the mRNA sharply declined (lanes 2), and decreased further at the 24 and 48 hr time points (lanes 3 and 4). Cells cultured in parallel for metabolic studies were incubated with radiolabelled all-trans-RA at the corresponding time points for a 4 h window. A low concentration of RA (575 pM) was used in the metabolic studies because this concentration of RA has been shown not to induce P450RAI mRNA (Fig. 1, lane 5) or metabolic activity in MCF-7 cells.
L. LUU et al.
Following treatment and incubations, the organic and aqueous layers were separated and analyzed. In Fig. 5B, radioactivity remaining in the aqueous layer is shown. The highest level of conversion of substrate into aqueoussoluble products in MCF-7 cells occurred at the time of washout, corresponding to the time point in which highest levels of P450RAI mRNA are seen (Fig. 5A, lane 1). Several of the polar metabolites coelute with 4oxo-RA and 4-OH-RA standards on HPLC (data not shown), consistent with our previous characterizations of P450RAI activity (35, 37). Furthermore, as P450RAI mRNA levels declined in the post-washout cultures (Fig. 5A, lanes 3 and 4), corresponding decreases in the conversion of substrate into aqueous-soluble forms were evident (Fig. 5B).
An Autoregulatory Feedback Loop Controlling RA Levels MCF7 has previously been shown to have RA-inducible RA metabolism (27). We have demonstrated that P450RAI may be responsible for this RAinducible RA metabolism. Our results also suggest that P450RAI is part of a feedback loop involved in the autoregulation of RA levels such that when normal physiological levels of RA are exceeded, induction of P450RAI acts to normalize RA levels (Fig. 6). In RA washout experiments, a strong correlation was demonstrated between the patterns of P450RAI mRNA expression and RA metabolic activity in response to RA treatment. Furthermore, the duration of P450RAI mRNA expression in response to RA treatment in MCF-7 cells was prolonged by co-treatment with an inhibitor of RA metabolism, suggesting that the rapid decline in levels of the mRNA was related to the clearance of RA. The rapid neutralization of RA in MCF-7 cells and the T47D breast epithelial carcinoma cells (32) suggest that tumor responsiveness to RA may be limited by inducible RA metabolic activity and that the prognostic value of P450RAI expression for RA treatment should be evaluated. We have also previously demonstrated that in mouse embryonal carcinoma cells, once P450RAI has been induced by RA, expression remains high long after RA has been removed from the media, suggesting constitutive expression of this gene (37). The mechanisms accounting for these diﬀerences in P450RAI regulation are unclear. We have also previously observed that certain cell types, such as the non-small cell lung carcinoma cell line SK-LC6 and the embryonic kidney cell line HEK293, express P450RAI mRNA constitutively (35). Thus, there are at least three modes of P450RAI regulation. Recently, we have identiﬁed a functional, canonical retinoic acid response element in the promoter of the P450RAI gene (40). It is possible that the induction of P450RAI expression is, to a
AUTOREGULATION OF RA METABOLISM
FIG. 6. Autoregulatory feedback-loop to control RA levels. The inducibility of P450RAI expression in certain cell lines and tissues deﬁnes a mechanism whereby high physiological levels of RA can induce P450RAI expression through transcriptional regulation of the P450RAI gene. The increased expression of P450RAI expression will act to normalize RA levels. Once RA levels are normalized, P450RAI expression falls oﬀ in accordance with residual RA levels.
large extent, transcriptionally regulated. In support of this, F9 cells lacking the retinoic acid receptor, RAR g and retinoid X receptor, RXR a, neither exhibit induced P450RAI expression nor increased RA metabolism (37). Coordination of RA Synthesis and Metabolism We have also observed inducible expression of P450RAI in mouse embryos suggesting that an autoregulatory mechanism similar to that seen for MCF7 cells may limit exposure to RA-sensitive tissues during development (41). Since RAs morphogen-like properties were described almost two decades ago, several attempts to deﬁne gradients of RA in developing tissues has been met with limited success. Two important approaches gave strong indications that RA levels in tissues were tightly regulated: one approach was to directly analyze RA present in the anterior and posterior portions of a collected pool of 5000 chick wing buds establishing that there was an A-P graded RA distribution (42); a second
L. LUU et al.
approach utilized transgenic mice harboring a RAR-RARE directed reporter gene to indirectly identify restricted regions of the embryo possibly synthesizing or sequestering active retinoids (43, 44). The recent cloning of RALDH-2 and P450RAI has provided a way to visualize by whole-mount in-situ hybridization both synthesis and degradation of RA in developing tissues. Subdivisions between RA-synthesizing and RA-degrading regions can be seen in early chick embryos where RALDH-2 expression is localized to presomitic and lateral plate mesoderm whereas P450RAI is expressed in presumptive mid- and forebrain (45). At later stages of development, the complementary patterns of expression of these enzymes can also be observed in the development of the anterior and posterior neural tube (41), in the developing limb bud and in the eye (46). At early stages during the development of the mouse retina, P450RAI is expressed in a narrow horizontal strip forming a boundary between two dehydrogenases capable of converting retinaldehyde to retinoic acid, RALDH-2 ventrally and ALDH-1 dorsally. This creates three zones of RA activity that subdivides the retina into territories, which portend further eye development (46). RA Resistance and Cancer What has become clear from developmental studies is that the function of P450RAI may be two-fold; (i) to act together with RALDH-2 at the tissue level to deﬁne regional patterns of RA distribution which may initiate and establish pattern formation, and (ii) to act at the cellular level to restrict access of RA to the transcriptional machinery. It is the latter function that may be an important factor determining the eﬀectiveness of RA in a clinical setting and in the case of cancer is clearly a potential cause of RA resistance. RA and Cancer Early studies of retinol deﬁciency indicated a correlation between vitamin A depletion and a higher incidence of cancer and increased susceptibility to chemical carcinogenesis (47). Several animal models have been used to demonstrate the eﬀectiveness of retinoids in suppressing carcinogenesis in a variety of tissues including skin, mammary epithelia, oral cavity, aerodigestive tract, liver, bladder and prostate (48). These studies have led to the preventive use of retinoids to treat premalignant lesions including actinic keratosis and oral leukoplakia, as well as in the prevention of secondary tumors of the head and neck and the recurrence of non small cell lung carcinomas and basal cell carcinomas (8, 49). The most dramatic therapeutic use of RA has been in the treatment of acute promyelocytic leukemia (APL) (50, 51). Studies over the past several years indicate that a high proportion of patients with acute promyelocytic leukemia (APL) achieve complete remission after a short period of treatment with all-trans RA. Unfortunately, this high rate of remission is in most cases brief. Following
AUTOREGULATION OF RA METABOLISM
relapse, patients are clinically resistant to further treatment with RA (52–54). The nature of this resistance is unknown. Retinoic Acid Resistance Cells normally sensitive to RA can become resistant through diﬀerent mechanisms. Defects in a number of the components of RA signaling have been implicated either directly or indirectly in reducing or suppressing RA response. For example, defects in the structure or expression of RARs have been observed in several diﬀerent types of resistant cell lines: RA resistant HL-60 myeloid leukemia cells have been found with a point mutation in RARa and could be restored to RA sensitivity by wild-type RARa complementation (55, 56); aberrant RARb expression and RA inducibility have been documented in lung tumor cell lines (57–60); and, decreased levels of RARg expression have been observed in a number of independently derived RA-resistant teratocarcinoma cell lines (61). Defects in factors other than receptors have also been implicated in RA resistance. In a RA-resistant line of HL-60, P-glycoprotein, a multidrug resistance gene, has been implicated in lowering RA sensitivity since the inhibitor verapamil and RA in combination could increase the responsiveness of the cells to RA (62). Clinical studies suggest that the expression of factors involved in RA metabolism may play a role in the acquired resistance to RA seen in acute promyelocytic leukemia (APL). In a limited study, CRABPII expression was elevated in cells derived from RA-resistant patients (54). Furthermore, RAinduced RA metabolism has been demonstrated in the APL-derived cell line NB4 correlating with expression of P450RAI (data not shown). Whether these observations are related is unclear but are consistent with increased RA metabolism in patients following RA therapy (63). There is strong evidence that RA metabolism due to P450RAI expression may play an important role in cellular RA resistance: (i) It can be demonstrated in culture that cells containing a RA-reporter gene cotransfected with P450RAI require 100-fold higher concentrations of RA to achieve the same level of reporter gene activity seen in cells containing the RA-reporter gene alone (34), (ii) Similarly, F9 teratocarcinoma cells constitutively expressing P450RAI are hyposensitive to the diﬀerentiating eﬀects of RA (34), (iii) embryonic tissues expressing P450RAI have no detectable RA activity, whereas such activity can readily be detected in adjacent tissue where P450RAI is not expressed (41) and (iv) the teratogenic eﬀects of RA in xenopus embryos can be prevented by overexpression of xenopus P450RAI/CYP26 (36). Clinical studies also support the possibility that P450RAI presents a barrier to optimal therapeutic activity of RA. For example, leukemic cells taken from APL patients exhibiting clinical resistance to RA have been shown to be sensitive to the diﬀerentiating action of RA when grown in vitro
L. LUU et al.
(50, 51). This suggests that pharmacokinetic mechanisms may account for the acquired resistance to RA. This possibility is supported by studies showing that peak plasma concentrations of RA were much higher in patients after initial administration than in patients treated following relapse. This decrease in peak plasma RA concentration was accompanied by a 10-fold increase in urinary 4-oxo-retinoic acid concentration. In addition, ketoconazole, a broad-spectrum inhibitor of cytochrome P450 function, was shown to modulate RA pharmacokinetics in vivo (50, 51). It is therefore likely that RA increases the rate of its own metabolism, which in turn results in the inability to sustain eﬀective therapeutic doses of RA. P450RAI may be involved since it is the only known RA-inducible cytochrome P450 for which RA is a substrate and it is expressed both in the liver and in leukemic cells following RA treatment. This presents two types of barriers to therapeutic doses of RA; ﬁrst, liver P450RAI can reduce systemic levels of RA; second, induced expression of P450RAI in the leukemic cells themselves would reduce their sensitivity to RA. Thus, factors that directly limit the ability of cells to respond to RA such as receptor defects or limit the availability of RA to cells such as increased RA metabolism, may predispose individuals to cancerous lesions as well as reduce the eﬀectiveness of RA therapy. SUMMARY
RA metabolism is a critical mechanism in the local control of RA signaling. In developmental processes requiring RA, the balance between RA synthesis and RA catabolism is tightly controlled by the enzymes RALDH-2 and P450RAI, respectively. Also in the adult, the expression of these enzymes in tissues undergoing cell proliferation and diﬀerentiation, such as skin or tumor tissues, may have a direct inﬂuence on how they respond to endogenous or therapeutic doses of retinoids, in particular since it can be clearly demonstrated that in some cells and tissues, P450RAI/ CYP26 expression is regulated by RA. The identiﬁcation and characterization of P450RAI provides an important tool to examine the role that metabolism plays in directing RA signaling and a possible target for the design of novel therapeutics to enhance current retinoid therapies. ACKNOWLEDGEMENTS
Supported by grants from the Medical Research Council of Canada and the National Cancer Institute of Canada (to M.P.).
REFERENCES L. GUDAS, M. SPORN and A. ROBERTS, Cellular biology and biochemistry of the retinoids, in The Retinoids. (M. SPORN, A. ROBERTS and D. S. GOODMAN eds.), pp. 443–520, Raven Press, New York, (1994).
AUTOREGULATION OF RA METABOLISM 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22.
23. 24. 25.
R. M. LOTAN, Squamous diﬀerentiation and retinoids (Review), Cancer Treatment & Res. 74, 43–72 (1995). G. M. MORRISS-KAY, Embryonic development and pattern formation, FASEB J. 10, 961–968 (1996). R. LOTAN, Retinoids in cancer chemoprevention, FASEB J. 10, 1031–1039 (1996). C. CAMISA, Treatment of severe psoriasis with systemic drugs, Dermatol. Nursing 7, 107–118 (1995). G. J. FISHER and J. J. VOORHEES, Molecular mechanisms of retinoid actions in skin, FASEB J. 10, 1002–1013 (1996). D. B. WINDHORST, The use of isotretinoin in disorders of keratinization, J. Am. Acad. Derm. 6, 708–709 (1982). W. HONG, Retinoids and human cancer, in The Retinoids. (M. SPORN, A. ROBERTS and D. S. GOODMAN eds.), pp. 597–630, Raven Press, New York, (1994). P. CHAMBON, A decade of molecular biology of retinoic acid receptors, FASEB J. 10, 940–954 (1996). V. GIGUERE, Retinoic acid receptors and cellular binding proteins: complex interplay in retinoid signalling, Endocrine Rev. 15, 61–79 (1994). M. PETKOVICH, Regulation of gene expression by vitamin A: the role of nuclear retinoic acid receptors, Ann. Rev. Nutr. 12, 443–471 (1992). M. LEID, P. KASTNER and P. CHAMBON, Multiplicity generates diversity in the retinoic acid signaling pathways, Trends Biochem. Sci. 17, 427–433 (1992). F. RASTINEJAD, T. PERLMANN, R. M. EVANS and P. B. SIGLER, Structural determinants of nuclear receptor assembly on DNA direct repeats, Nature 375, 203–211 (1995). H. GRONEMEYER and V. LAUDET, Retinoic acid receptors, in Protein Proﬁle, Transcription Factors 3: Nuclear Receptors, Vol. 2 pp. 1188–1189 (1995). J. L. NAPOLI, Retinoic acid synthesis from beta-carotene in vitro, Methods Enzymol. 214, 193–202 (1993). J. L. NAPOLI, Biosynthesis and metabolism of retinoic acid: roles of CRBP and CRABP in retinoic acid: roles of CRBP and CRABP in retinoic acid homeostasis, J. Nutr. 123, 362–366 (1993). W. BLANER, Retinol and retinoic acid metabolism, in The Retinoids. (M. SPORN, A. ROBERTS, D. S. GOODMAN, eds.), pp. 229–255, Raven Press, New York, (1994). J. L. NAPOLI, Retinoic acid biosynthesis and metabolism, FASEB J. 10, 993–1001 (1996). J. L. NAPOLI, M. H. BOERMAN, X. CHAI, Y. ZHAI and P. D. FIORELLA, Enzymes and binding proteins aﬀecting retinoic acid concentrations, J. Steroid Biochem. Mol. Biol. 53, 497–502 (1995). G. DUESTER, Retinoids and the alcohol dehydrogenase gene family, Exs. 71, 279–290 (1994). P. MCCAFFERY and U. C. DRAGER, Retinoic acid synthesizing enzymes in the embryonic and adult vertebrate, Adv. Exper. Med. Biol. 372, 173–183 (1995). K. NIEDERREITHER, P. MCCAFFERY, U. C. DRAGER, P. CHAMBON and P. DOLLE, Restricted expression and retinoic acid-induced downregulation of the retinaldehyde dehydrogenase type 2 (RALDH-2) gene during mouse development, Mech. Devel. 62, 67–78 (1997). F. FORMELLI, A. BARUAAND and J. OLSON, Bioactivities of N-[4-hydroxyphenyl) retinimide and retinoyl B-glucuronide, FASEB J. 10, 1014–1024 (1996). J. P. VAN WAUWE, M. -C. COENE, J. GOOSSENS, G. VAN NIJEN, W. COOLS and W. LAUWERS, Ketoconazole inhibits the in vitro and in vivo metabolism of all-transretinoic acid, J. Pharmacol. Exper. Therapeut. 245, 718–722 (1988). J. P. VAN WAUWE, G. VAN NYEN, M. -C. COENE, P. STOPPIE, W. COOLS, J. GOOSSENS, P. BORGHGRAEF and P. A. J. JANSSEN, Liarozole, an inhibitor of retinoic acid metabolism, exerts retinoid-mimetic eﬀects in vivo, J. Pharmacol. Exper. Therapeut. 261, 773–779 (1992). J. P. VAN WAUWE, M. -C. COENE, J. GOOSSENS, W. COOLS and J. MONBALIU, Eﬀects of cytochrome P-450 inhibitors on the in vivo metabolism of all-trans-retinoic acid in rats, J. Pharmacol. Exper. Therapeut. 252, 365–369 (1990).
L. LUU et al.
27. W. WOUTERS, A. DILLEN, M. -C. COENE and W. COOLS, Eﬀects of liarozole, a new antitumoral compound, on retinoic acid-induced inhibition of cell growth and on retinoic acid metabolism in MCF-7 human breast cancer cells, Cancer Res. 52, 2841–2846 (1992). 28. J. B. WILLIAMS and J. L. NAPOLI, Inhibition of retinoic acid metabolism by imidazole antimycotics in F9 embryonal carcinoma cells, Biochem. Pharmacol. 36, 1386–1388 (1987). 29. C. A. FROLIK, A. B. ROBERTS, T. E. TAVELA, P. P. ROLLER, D. L. NEWTON and M. B. SPORN, Isolation and identiﬁcation of 4-hydroxy- and 4-oxoretinoic acid, in vitro metabolites of all-trans-retinoic acid in hamster trachea liver, Biochemistry 18, 2092–2097 (1979). 30. A. B. ROBERTS, C. A. FROLIK, M. D. NICHOLS and M. B. SPORN, Retinoiddependent induction of the in vivo and in vitro metabolism of retinoic acid in tissues of the vitamin A-deﬁcient hamster, J. Biol. Chem. 254, 6303–6309 (1979). 31. E. A. DUELL, A. ASTROM, C. E. GRIFFITHS, P. CHAMBON and J. J. VOORHEES, Human skin levels of retinoic acid and cytochrome p-450-derived 4-hydroxyretinoic acid after topical application of retinoic acid in vivo compared to concentrations required to stimulate retinoic acid receptor-mediated transcription in vitro, J. Clin. Invest. 90, 1269–1274 (1992). 32. J. WHITE, Y. GUO, K. BAETZ, B. BECKETT-JONES, J. BONASORO, K. HSU, J. DILWORTH, G. JONES and M. PETKOVICH, Identiﬁcation of the retinoic acidinducible all trans retinoic acid 4-hydroxylase, J. Biol. Chem. 271, 29922–29927 (1996). 33. W. J. RAY, G. BAIN, M. YAO and D. I. GOTTLIEB, CYP26, a novel mammalian cytochrome P450, is induced by retinoic acid and deﬁnes a new family, J. Biol. Chem. 272, 18702–18708 (1997). 34. H. FUJII, T. SATO, S. KANEKO, O. GOTOH, Y. FUJII-KURIYAMA, K. OSAWA, S. KATO and H. HAMADA, Metabolic inactivation of retinoic acid by a novel P450 diﬀerentially expressed in developing mouse embryos, EMBO Journal 16, 4163–4173 (1997). 35. J. A. WHITE, B. BECKETT-JONES, Y. D. GUO, F. J. DILWORTH, J. BONASORO, G. JONES and M. PETKOVICH, cDNA cloning of human retinoic acid-metabolizing enzyme (hP450RAI) identiﬁes a novel family of cytochromes P450, J. Biol. Chem. 272, 18538–18541 (1997). 36. T. HOLLERMANN, Y. CHEN, H. GRUNZ and T. PIELER, Regionalized metabolic activity establishes boundaries of retinoic acid signaling, EMBO J. 17, 7361–7372 (1998). 37. S. S. ABU-ABED, B. R. BECKETT, H. CHIBA, J. V. CHITHALEN, G. JONES, D. METZGER, P. CHAMBON and M. PETKOVICH, Mouse P450RAI (CYP26) expression and retinoic acid-inducible retinoic acid metabolism in F9 cells are regulated by retinoic acid receptor gamma and retinoid X receptor alpha, J. Biol. Chem. 273, 2409–2415 (1998). 38. J. A. WHITE, M. B. BOFFA, B. JONES and M. PETKOVICH, A zebraﬁsh retinoic acid receptor expressed in the regenerating caudal ﬁn, Development 120, 1861–1872 (1994). 39. E. G. BLIGH and W. J. DYER, A rapid method of total lipid extraction and puriﬁcation, Canad. J. Biochem. 37, 911–917 (1957). 40. O. LOUDIG, C. BABICHUK, J. WHITE, S. ABU-ABED, C. MUELLER and M. PETKOVICH, Retinoic acid inducibility of P450RAI (CYP26) promoter involves a highly conserved retinoic acid response element and a Sp1/Sp3 site, Mol. Endo. 14, 1483–1497 (2000). 41. A. IULIANELLA, B. BECKETT, M. PETKOVICH and D. LOHNES, A molecular basis for retinoic acid-induced axial truncation, Dev. Biol. 205, 33–48 (1999). 42. C. THALLER and G. EICHELE, Identiﬁcation and spatial distribution of retinoids in the developing chick limb bud, Nature 327, 625–628 (1987). 43. J. ROSSANT, R. ZIRNGBL, D. CADO, M. SHAGO and V. GIGUERE, Expression of a retinoic acid responsive element-hsplacZ transgene deﬁnes speciﬁc domains of transcriptional activity during mouse embryogenesis, Genes Dev. 5, 1333–1344 (1991). 44. C. MENDELSOHN, E. RUBERTE, M. LEMEUR, G. MORRISS-KAY and P. CHAMBON, Developmental analysis of the retinoic acid-inducible RAR-2 promoter in transgenic animals, Development 113, 723–734 (1991). 45. E. SWINDELL, C. THALLER, S. SOCKANATHAN, M. PETKOVICH, T. JESSELL and G. EICHELE, Complementary domains of retinoic acid production and degradation in the early chick embryo, Dev. Biol. 216, 282–296 (1999).
AUTOREGULATION OF RA METABOLISM
46. P. MCCAFFERY, E. WAGNER, J. O’NEIL, M. PETKOVICH and U. C. DRAGER, Dorsal and ventral retinal territories deﬁned by retinoic acid synthesis, break-down and nuclear receptor expression, Mech. Devel. 85, 203–214 (1999). 47. F. CHYTIL, Retinoic acid: biochemistry, toxicology, pharmacology, and therapeutic use, Pharmacol. Rev. 36, 93–99 (1984). 48. R. C. MOON, R. G. MEHTA and K. V. N. RAO, Retinoids and cancer in experimental animals, in The Retinoids. (M. SPORN, A. ROBERTS and D. S. GOODMAN eds.), pp. 573–595, Raven Press, New York, (1994). 49. S. M. LIPPMAN, R. A. HEYMAN, J. M. KURIE, S. E. BENNER and W. K. HONG, Retinoids and chemoprevention: clinical and basic studies, J. Cell. Biochem. Supplement 22, 1–10 (1995). 50. J. R. F. MUINDI, S. R. FRANKEL, C. HUSELTON, F. DEGRAZIA, W. GARLAND, C. W. YOUNG and R. P. WARRELL Jr., Clinical pharmacology of oral all-trans retinoic acid in patients with acute promyelocytic leukemia, Cancer Res. 52, 2138–2142 (1992). 51. J. R. MUINDI, C. W. YOUNG and R. J. WARRELL, Clinical pharmacology of all-trans retinoic acid, Leukemia 8, s16–s21 (1994). 52. R. P. WARRELL Jr., Applications for retinoids in cancer therapy, Semin. Hematol. 31, 1– 13 (1994). 53. R. P. WARRELL Jr., P. MASLAK, A. EARDLEY, G. HELLER, W. J. MILLER and S. R. FRANKEL, Treatment of acute promyelocytic leukemia with all-trans retinoic acid: an update of the New York experience, Leukemia 8, 929–933 (1994). 54. C. CHOMIENNE, R. FENAUX and L. DEGOS, Retinoid diﬀerentiation therapy in promyelocytic leukemia, FASEB J. 10(9), 1025–1030. 55. K. A. ROBERTSON, B. EMAMI and S. J. COLLINS, Retinoic acid-resistant HL-60R cells harbor a point mutation in the retinoic acid receptor ligand-binding domain that confers dominant negative activity, Blood 80, 1885–1889 (1992). 56. R. MOROSETTI, F. GRIGNANI, C. LIBERATORE, P. G. PELICCI, G. J. SCHILLER, M. KIZAKI, C. R. BARTRAM, C. W. MILLER and H. P. KOEFFLER, Infrequent alterations of the RAR alpha gene in acute myelogenous leukemias, retinoic acid-resistant acute promyelocytic leukemias, myelodysplastic syndromes, and cell lines, Blood 87, 4399–4403 (1996). 57. X. ZHANG, Y. LIU, M. LEE and M. PFAHL, A speciﬁc defect in the retinoic acid response associated with human lung cancer cell lines, Cancer Res. 54, 5663–5669 (1994). 58. J. BERARD, L. GABOURY, M. LANDERS, R. Y. DE, B. HOULE, R. KOTHARY and W. E. BRADLEY, Hyperplasia and tumours in lung, breast and other tissues in mice carrying a RAR beta 4-like transgene, EMBO J. 13, 5570–5580 (1994). 59. B. HOULE, M. PELLETIER, J. WU, C. GOODYER and W. E. BRADLEY, Fetal isoform of human retinoic acid receptor beta expressed in small cell lung cancer lines, Cancer Res. 54, 365–369 (1994). 60. B. HOULE, E. C. ROCHETTE and W. E. BRADLEY, Tumor-suppressive eﬀect of the retinoic acid receptor beta in human epidermoid lung cancer cells, Proc. Natl. Acad. Sci. U.S.A. 90, 985–989 (1993). 61. M. M. MOASSER, A. DEBLASIO and E. DMITROVSKY, Response and resistance to retinoic acid are mediated through the retinoic acid nuclear receptor gamma in human teratocarcinomas, Oncogene 9, 833–840 (1994). 62. M. KIZAKI, H. UENO, Y. YAMAZOE, M. SHIMADA, N. TAKAYAMA, A. MUTO, H. MATSUSHITA, H. NAKAJIMA, M. MORIKAWA, H. P. KOEFFLER and Y. IKEDA, Mechanisms of retinoid resistance in leukemic cells: possible role of cytochrome P450 and P-glycoprotein, Blood 87, 725–733 (1996). 63. M. LAZZARINO, A. CORSO, M. B. REGAZZI, I. IACONA and C. BERNASCONI, Modulation of all-trans retinoic acid pharmacokinetics in acute promyelocytic leukemia by prolonged interferon therapy, Brit. J. Haematol. 90, 928–930 (1995).