UDP-glucuronic acid (UDPGA), trans-TAM, trans-4-OH-TAM (98% pure), trans-4-OH-TAM:cis-4-OH-TAM (70:30% ratio), alamethicin, β-glucuronidase, anticalnexin antibody and bovine serum albumin were purchased from Sigma-Aldrich (St Louis, MO, USA). HPLC-grade ammonium acetate and acetonitrile were purchased from Fisher Scientific (Pittsburgh, PA, USA) and were used after filtration. DMEM, Dulbecco's phosphate-buffered saline (minus calcium chloride and magnesium chloride), fetal bovine serum, penicillin-streptomycin and geneticin (G418) were purchased from Gibco (Grand Island, NY, USA).

The Platinum^® Pfx DNA polymerase and the pcDNA3.1/V5-His-TOPO mammalian expression vector were obtained from Invitrogen (Carlsbad, CA, USA), while the restriction enzymes DpnI and StuI were purchased from New England Biolabs (Beverly, MA, USA). The BCA protein assay kit was purchased from Pierce (Rockford, IL, USA) while the QIAEX^® II gel extraction kit was purchased from Qiagen (Valencia, CA, USA). The human UGT1A western blotting kit and anti-UGT1A antibody were purchased from Gentest (Woburn, MA, USA). All other chemicals used were purchased from Fisher Scientific (Pittsburgh, PA, USA) unless otherwise specified.

A description of the normal human liver tissue specimens used for these studies was presented previously 36. Briefly, tissues were quick-frozen at -70°C within 2 hours postsurgery. Liver microsomes were prepared through differential centrifugation as previously described 37 and were stored (10–20 mg protein/ml) at -70°C. Microsomal protein concentrations were measured using the BCA assay (Pierce). All protocols involving the analysis of tissue specimens were approved by the institutional review board at the Penn State College of Medicine and in accordance with assurances filed with and approved by the United States Department of Health and Human Services.

The UGT1A4 wild-type cDNA was synthesized by RT-PCR using normal human liver total RNA and was inserted into the pcDNA3.1/V5-His-TOPO plasmid. The sense and antisense primers used for RT-PCR were UGT1A4s1 (5'-ACAGTCAGCTGTCGGTGGC-3', corresponding to -29 to -11 relative to the UGT1A4 translation start site; GenBank accession number NM007120) and UGT1A4as1 (5'-ATTTTACCTTATTTCCCACCC-3', corresponding to +1611 to +1631 relative to the UGT1A4 translation start site), respectively.

Incubations were performed in a GeneAmp 9700 thermocycler (Applied Biosystems, Foster City, CA, USA) as follows: one cycle at 94°C for 2 minutes, 41 cycles at 94°C for 30 seconds, at 55°C for 30 seconds and at 72°C for 2 minutes, followed by a final cycle of 7 minutes at 72°C. The PCR product (1662 base pairs) was purified after electrophoresis in 1.5% agarose using the QIAEX^® II gel extraction kit (Qiagen), and was subsequently subcloned into the pcDNA3.1/V5-His-TOPO mammalian expression vector using standard methodologies.

Confirmation of insert orientation was performed by restriction enzyme digestion, and UGT1A4 sequences were confirmed by dideoxy sequencing of the entire PCR-amplified UGT1A4 cDNA product (performed at the Molecular Biology Core Facility at Penn State University College of Medicine) using two vector primers (T7 and BGH; IDT, Coralville, IA, USA) and one internal sense primer (UGT1A4s2 – 5'-GAAGGAATTTGATCGCGTTAC-3', corresponding to nucleotides +258 to +278 relative to the UGT1A4 translation start site). The cloned UGT1A4 insert was compared with that described in GenBank and was confirmed to be 100% homologous to the wild-type UGT1A4^24Pro/48Leu sequence (accession number NM007120).

The UGT1A4^24Thr/48Leu and UGT1A4^24Pro/48Val variants were generated by site-directed mutagenesis of the pcDNA3.1/V5-His-TOPO plasmid expressing wild-type UGT1A4^24Pro/48Leu. The primers used to change codon 24 from Pro to Thr were UGT1A4c24F (5'-CAGTGTCCAGACCTGGGCTGAGAGTG-3') and UGT1A4c24R (5'-CACTCTCAGCCCAGGTCTGGACACTG-3'), both primers corresponding to nucleotides +60 to +85 relative to the UGT1A4 translation start site (mutated base indicated in bold). The primers used to change codon 48 from Leu to Val were UGT1A4c48F (5'-CTCAGCATGCGGGAGGCCGTGCGGGAGCTCCATGC-3') and UGT1A4c48R (5'-GCATGGAGCTCCCGCACGGCCTCCCGCATGCTGAG-3'), both primers corresponding to nucleotides +124 to +158 relative to the UGT1A4 translation start site.

The products were amplified by PCR using 5 U/ml Pfx polymerase, 1 × Pfx buffer, 2 × enhancer, 10 μM dNTPs, 500 ng template, 1 mM MgSO₄ and 20 μM of each primer in a BioRad MyCycler (Hercules, CA, USA) with an initial denaturation at 95°C for 2 minutes, followed by 25 cycles at 95°C for 30 seconds, at 65°C for 30 seconds and at 68°C for 18 minutes. Following amplification, 20 U Dpn Irestriction enzyme was added to each reaction and incubated for 1 hour at 37°C to specifically digest the wild-type template DNA. The remaining PCR product was then transformed into competent DH5α Escherichia coli, individual colonies were isolated, and subsequent plasmid DNA minipreps were screened for the UGT1A4^24Thr/48Leu and UGT1A4^24Pro/48Val variants using HPY188III or StuI, respectively. The UGT1A4^24Thr/48Leu and UGT1A4^24Pro/48Val cDNA sequences were confirmed by direct dideoxy sequencing using the same primers used for wild-type UGT1A4 analysis described earlier.

Human embryonic kidney fibroblast HEK293 cell lines overexpressing wild-type or variant UGT1A4 were generated by standard electroporation techniques in the Bio-Rad GenePulser Xcell using 10 μg pcDNA3.1/V5-His-TOPO/UGT1A4 plasmid DNA with 5 × 10⁶ HEK293 cells (in 0.5 ml) in serum-free DMEM media, with electroporation at 250 V and 1000 μF. Following transfection, HK293 cells were grown in 5% CO₂ to 80% confluence in DMEM supplemented with 4.5 mM glucose, 10 mM HEPES, 10% fetal bovine serum, 100 U/ml penicillin, 100 μg/ml streptomycin and geneticin (700 μg/ml medium) for the selection of geneticin-resistant clones, with selection medium changed every 3–4 days.

Individual UGT1A4-overexpressing cell colonies were selected and monitored for UGT1A4 expression via western blotting analysis (described in the next section). Cell homogenates were prepared by resuspending pelleted cells in Trisbuffered saline (25 mM Tris base, 138 mM NaCl, 2.7 mM KCl, pH 7.4) and subjecting them to three rounds of freeze-thaw prior to gentle homogenization. Microsomal fractions were prepared by centrifugation of whole cell homogenate at 10,500 rpm (9,000 × g) for 30 minutes at 4°C; the supernatant was collected and subsequently centrifuged at 33,500 rpm (105,000 × g) for 60 minutes at 4°C in a SW-55 Ti rotor (Beckman, Palo Alto, CA, USA). Pellets were collected by suspension in Tris-buffered saline (25 mM Tris base, 138 mM NaCl, 2.7 mM KCl, pH 7.4) and were stored at -80°C in 100 μl aliquots. Total microsomal protein concentrations were measured using the BCA protein assay (Pierce).

Levels of UGT1A4 expression in UGT-overexpressing cell lines were measured by western blot analysis using the anti-UGT1A antibody in a 1:5,000 dilution as per the manufacturer's instructions (Gentest), while calnexin protein levels were assayed using a 1:5,000 dilution of the monoclonal anticalnexin antibody (after stripping the UGT1A4 antibody of the same western blot using standard techniques). UGT1A4 was detected by chemiluminescence using the SuperSignal West Dura Extended Duration Substrate (Pierce Biotechnology, Inc., Rockford, IL, USA). Secondary antibodies supplied with the Dura ECL kit (anti-rabbit and anti-mouse) were used at 1:3,000. UGT1A4 levels were quantified against a known amount of human UGT1A protein (200–300 ng, supplied in the western blotting kit; Gentest) by densitometric analysis of X-ray film exposures (1 s–1 min exposures) of western blots using a GS-800 densitometer with Quantity One software (Bio-Rad).

UGT1A4 protein levels were calculated against known titrated quantities of UGT1A standard, with quantification made relative to the levels of calnexin observed in each lane (quantified by densitometric analysis of western blots as already described). Determinations of aglycone-glucuronide formation in UGT1A4-overexpressing cell lines were calculated relative to the levels of UGT1A4 expression in the respective cell lines. X-ray film bands were always below densitometer saturation levels as indicated by the densitometer software. Densitometric results were always consistent irrespective of the exposure time. Western blot analysis and subsequent densitometric analysis were performed in triplicate on three separate occasions, using the same UGT1A4-containing cell homogenates used for activity assays, with relative UGT1A4 protein levels expressed as the mean of these experiments.

Glucuronidation activities of microsomes from human UGT1A4-overexpressing HEK293 cells toward trans-TAM, trans-4-OH-TAM and cis-4-OH-TAM were performed after an initial incubation of microsomal protein (5–25 μg) with alamethicin (50 μg/mg protein) for 15 minutes in an ice bath. Glucuronidation reactions were then performed in a final reaction volume of 100–500 μl at 37°C in 50 mM Tris–HCl (pH 7.4), 10 mM MgCl₂, 4 mM UDPGA and 1–6 μM trans-TAM or trans-4-OH-TAM, and 1–15 μM cis-4-OH-TAM. Reactions were terminated by the addition of 95–495 μl cold methanol on ice. Five microliters of propranolol (2.5 μg/ml) was added to the final reactions, the mixtures were centrifuged for 10 minutes at 4°C at 16,100 × g, the supernatants were collected and evaporated, and the resulting dried sample was resuspended in 200 μl of 50% methanol. Assays with human liver microsomes were identical except that 10 μg microsomal protein was used in glucuronidation assays.

Samples (100 μl) were analyzed for glucuronidated TAM or for TAM metabolites by HPLC using a Beckman Coulter System Gold 126 Solvent Module HPLC system (Fullerton, CA, USA) equipped with an automatic injector (model 508) and a UV detector operated at 254 nm (model 168). HPLC was performed using a 3 μ Luna C₁₈ analytical column (4.6 mm × 150 mm; Phenomenex, Torrance, CA, USA) in series with a 5 μ Aquasil C₁₈ guard column (10 mm × 4 mm; Thermo Hypersil-Keystone, Bellefonte, PA, USA). The gradient elution conditions for assays with trans-4-OH-TAM or cis-4-OH-TAM were as follows: starting with 75% buffer A (100 mM ammonium acetate, pH 5.0) and 25% acetonitrile (5 min), a subsequent linear gradient to 75% acetonitrile (25% buffer A) over 25 min was performed and then maintained at 75% acetonitrile for 10 minutes. The flow rate was 0.5 ml/min. For assays with TAM, the linear gradient was from 70% buffer A (30% acetonitrile) to 90% acetonitrile (10% buffer A) over 30 minutes.

The amount of N⁺-glucuronide formed was calculated based on the ratio of the peak area of the N⁺-glucuronide versus that observed for the internal standard, propranolol. Tamoxifen quaternary ammonium glucuronide (TAM-N⁺-glucuronide), the 4-hydroxytamoxifen quaternary ammonium glucuronides (trans-4-OH-TAM-N⁺-glucuronide and cis-4-OH-TAM-N⁺-glucuronide), and trans-4-OH-TAM-O-glucuronide and cis-4-OH-TAM-O-glucuronide were confirmed by 1 M NaOH hydrolysis and sensitivity to β-glucuronidase as previously described 34. As controls, glucuronidation assays were regularly performed using human liver microsomes (as a positive control for glucuronidation activity) and untransfected HK293 cell homogenate protein (as a negative control for glucuronidation activity) as previously described 3438. Experiments were always performed in triplicate in independent assays.

The predicted trans-4-OH-TAM-N⁺-glucuronide and cis-4-OH-TAM-N⁺-glucuronide were collected after separation by HPLC as already described and were identified by liquid chromatography–mass spectrometry (LC–MS) using a Shimadzu LC-MS 2010 EV system (Shimadzu, Tokyo, Japan). The trans-4-OH-TAM-N⁺-glucuronide and cis-4-OH-TAM-N⁺-glucuronide were loaded onto a Shimadzu reverse-phase column (Shimadzu C18, 4.6 mm × 50 mm) and were analyzed at a flow rate of 0.2 ml/min by applying a linear mobile phase gradient from 10% to 80% (v/v) methanol/H₂O over 30 minutes. An electrospray voltage of 1.5 kV was applied using a positive mode.

The Student t test (two-sided) was used for comparing the rates and the kinetic values of glucuronide formation for the UGT1A4^24Pro/48Leu, UGT1A4^24Thr/48Leu and UGT1A4^24Pro/48Val variant isoforms against the different substrates examined in this study. Kinetic constants were determined using Graphpad Prism4 software (GraphPad Software, San Diego, CA, USA).

While N⁺-glucuronide was shown to be the only glucuronide formed with TAM in human liver microsomes and UGT1A4-overexpressing baculosomes 30, 4-OH-TAM O-glucuronides and N⁺-glucuronides were formed by human liver microsomes. As shown in Figure 1, a HPLC assay was developed to efficiently separate TAM and TAM metabolites from their glucuronide conjugates.

Using propranolol as an internal standard (Figure 1a), the cis and trans isomers of 4-OH-TAM eluted at retention times of 27.1 minutes and 27.6 minutes, respectively, as shown by HPLC of a 70:30 mix of trans-4-OH-TAM:cis-4-OH-TAM (Sigma) (Figure 1b), of 98% pure trans-4-OH-TAM (Sigma) (Figure 1c), and of cis-4-OH-TAM purified from the 70:30 trans-4-OH-TAM:cis-4-OH-TAM mix (Sigma) (Figure 1d). In assays with human liver microsomes, two 4-OH-TAM glucuronide peaks with retention times of 18.2 minutes (peak 4) and 23.0 minutes (peak 5) were observed in assays with trans-4-OH-TAM (Figure 1e), and two 4-OH-TAM glucuronide peaks with retention times of 18.6 minutes (peak 6) and 22.5 minutes (peak 7) were observed in assays with cis-4-OH-TAM (Figure 1f). Peaks 4–7 were sensitive to treatment with β-glucuronidase in glucuronidation assays with either trans-4-OH-TAM (Figure 1g) or cis-4-OH-TAM (Figure 1h). Only peaks 5 and 7 were sensitive to treatment with alkali after glucuronidation assays with either trans-4-OH-TAM (Figure 1i) or cis-4-OH-TAM (Figure 1j).

This pattern is similar to that observed for N-glucuronide versus O-glucuronide deconjugation for other compounds 32 – suggesting that peaks 4 and 6 corresponded to trans-TAM-4-O-glucuronide and cis-TAM-4-O-glucuronide, and that peaks 5 and 7 corresponded to trans-4-OH-TAM-N⁺-glucuronide and cis-4-OH-TAM-N⁺-glucuronide in assays with either trans-4-OH-TAM or cis-4-OH-TAM, respectively. The products corresponding to predicted trans-OH-TAM-N⁺-glucuronide and cis-4-OH-TAM-N⁺-glucuronide were analyzed using LC with electrospray MS detection (Figure 2). The mass spectrum for the trans-4-OH-TAM-N⁺-glucuronide showed a clear [M⁺] ion at m/z 563.80 (calculated molecular weight, 563.29); a virtually identical pattern was observed for cis-4-OH-TAM-N⁺-glucuronide, with a clear [M⁺] ion at m/z 563.70 (data not shown). These data suggest that the glucuronides derived from both isomers were in fact monoglucuronides.

In assays with microsomes prepared from wild-type UGT1A4^24Pro/48Leu-overexpressing cells, significant glucuronidating activities were observed against both trans-4-OH-TAM and cis-4-OH-TAM (Figure 3). Single glucuronide peaks were observed in assays with either substrate (Figure 3a,b), with these peaks corresponding to the retention times of peaks 5 and 7 in glucuronidation assays with human liver microsomes (see Figure 1). These peaks were alkali sensitive (Figure 3c,d) and were sensitive to treatment with β-glucuronidase (results not shown), suggesting that these glucuronides corresponded to 4-OH-TAM-N⁺-glucuronide. Predicted UGT1A4-induced monoglucuronides were confirmed by LC–MS for both trans-4-OH-TAM and cis-4-OH-TAM (results not shown). No TAM-4-O-glucuronide formation was observed in assays with either the trans or cis isomer of 4-OH-TAM with UGT1A4-overexpressing cell microsomes, and no glucuronidation activity was observed for untransfected HK293 cell homogenates for any substrate examined in this study (results not shown).

As shown in a concentration curve for both isoforms of 4-OH-TAM (Figure 4), the rate of UGT1A4-catalyzed glucuronide formation was fully saturated at approximately 10–15 μM for trans-4-OH-TAM and cis-4-OH-TAM in our assay conditions. Using a concentration range of 4-OH-TAM (1–10 μM for trans-4-OH-TAM and 1–15 μM for cis-4-OH-TAM) and a specific incubation time (30 min), the apparent K_m values and the V_max/K_m ratios for UGT1A4-induced glucuronidation of trans-4-OH-TAM and of cis-4-OH-TAM were 2.2 ± 0.4 μM and 29.3 ± 2.7 μl/min/μg, and 2.1 ± 0.4 μM and 2.0 ± 0.3 μl/min/μg, respectively (performed in three independent experiments for both isomers).

To determine whether the Pro>Thr amino acid change at codon 24 or the Leu>Val amino acid change at codon 48 affected UGT1A4 enzyme activity against TAM or 4-OH-TAM isomers, stable HK293 cell lines overexpressing either the UGT1A4^24Thr/48Leu or the UGT1A4^24Pro/48Val variant isoforms were created by site-directed mutagenesis using the pcDNA3.1/V5-His-TOPO plasmid expressing wild-type UGT1A4^24Pro/48Leu as a template. Semiquantitative western blot analysis showed high levels of UGT1A4 protein in microsomes prepared from wild-type UGT1A4^24Pro/48Leu-overexpressing and UGT1A4^24Thr/48Leu-overexpressing or UGT1A4^24Pro/48Val-overexpressing HK293 cell lines (Figure 5). The UGT1A4 expression levels were normalized to the levels of the endoplasmic reticulum-specific protein, calnexin, in each lane as measured by densitometry and were determined against varying amounts of human UGT1A protein (200–300 ng; also measured by densitometry). The results demonstrated that the level of expression of UGT1A4 in the UGT1A4^24Thr/48Leu-overexpressing and UGT1A4^24Pro/48Val-overexpressing cell lines was respectively 0.73-fold and 0.81-fold that observed in the wild-type UGT1A4-overexpressing cell line.

As shown in Table 1, kinetic analysis demonstrated that higher glucuronidation activities were observed for UGT1A4^24Pro/48Val-overexpressing microsomes as compared with microsomes from wild-type UGT1A4^24Pro/48Leu-overexpressing cells against TAM and against both the trans and cis isomers of 4-OH-TAM. A significantly (P ≤ 0.02) lower K_m value (~1.6-fold to 1.8-fold) was observed for both 4-OH-TAM isomers, while a near-significant (P = 0.053) decrease in K_m was observed for TAM – for the UGT1A4^24Pro/48Val variant compared with wild-type UGT1A4. The V_max/K_m ratio for the UGT1A4^24Pro/48Val variant was significantly (P ≤ 0.005) higher than that observed for the wild-type UGT1A4 isoform after normalization for UGT1A4 expression by western blotting for both the trans and cis isomers of 4-OH-TAM. No significant effect on enzyme kinetics was observed for the UGT1A4^24Thr/48Leu variant against either isomer of 4-OH-TAM or against TAM itself.