Tris base, glucose, HEPES and TNF-α were purchased from Sigma Chemical (St. Louis, MO). Hanks' balanced salt solution (HBSS) and Dulbecco's modified Eagle's medium (DMEM), Trizol, Lipofectamine™ 2000, Superscript III reverse transcriptase and the 1 kb DNA ladder were obtained from Invitrogen (Carlsbad, CA). Dual-Luciferase Reporter assay system, pGL3 basic vector, pRL-TK plasmid, GoTaq^R Green Master Mix and EMSA kit were purchased from Promega (Madison, WI). QuickChange Site-Directed Mutagenesis kit was obtained from Stratagene (La Jolla, CA). The nuclear extraction kit was purchased from Active Motif (Carlsbad, CA). Recombinant human glucocorticoid receptor protein (RP-500) was obtained from Affinity Bioreagents (Golden, CO). Antibodies for p65 or p50 subunit of NF-κB, c-jun and c-fos were purchased from Santa Cruz Biotechnology (Santa Cruz, CA).

Genomic DNA was isolated from the human erythroleukemia cell line K562 and approximately 3 kb of the cd38 promoter was amplified by PCR using the following primers: 3181F 5'-TGATGCCTCCTGTTGGGGGTCTA-3' and 3181R 5'-CGGGAAAGCGCTTGGTGGTG-3' (GenBank Acc. No. DQ091293). The reverse primer (3181R) was phosphorylated using T4 polynucleotide kinase and PCR was performed under the following conditions: 94°C for 3 min denaturing, then 30 cycles of 94°C for 50 s, 59.6°C for 50 s, 72°C for 90 s, and a final extension at 72°C for 10 min to yield a 3240 bp fragment. A truncated 1.8 kb promoter was also amplified employing the same PCR program with annealing at 60°C using the primer pairs 1378F 5'-GCATGCATATGTTCATTGTAGCACTAT-3' and 3181R 5'-CGGGAAAGCGCTTGGTGGTG-3' which was phosphorylated using T4 polynucleotide kinase. The resulting 3 kb and 1.8 kb PCR fragments were gel purified, cloned into pCR 3.1 Uni vector (Invitrogen) and the reverse orientation was confirmed by sequencing at the Advanced Genetic Analysis Center, University of Minnesota. The 3 kb and 1.8 kb (truncated) positive clones were digested with HindIII/EcoRV and ligated into SmaI/HindIII digested pGL3 basic vector (Promega, WI, USA). This enabled cloning of the larger and truncated promoter fragments in the forward orientation to drive the expression of the luciferase reporter gene. The 3 kb and the truncated cd38 promoter fragments in the pGL3 basic vector were confirmed by nucleotide sequence analysis. To mutate the putative NF-κB and AP-1 binding sites, primers for mutated NF-κB and AP-1 binding sites were designed (Table 2). Putative binding sites are underlined and mutated sequences are shown in bold font. Mutations of the putative NF-κB or AP-1 binding sites in the promoter constructs were performed by the QuickChange Site-Directed Mutagenesis Method (Strategene, La Jolla, CA) using Pfu Turbo polymerase. Template DNAs were digested with the methylation-dependent restriction enzyme DpnI. Bacteria were then transformed with DpnI-digested DNA, and the cloned mutated constructs were confirmed by sequencing.

The GeneQuest module of Lasergene 6.0 program from DNASTAR was used to identify the potential transcription factor binding sites in the cd38 promoter. The 3 kb sequence of the cd38 promoter was analyzed using GeneQuest for the potential transcription factor binding sites using tfd.dat file. Analysis revealed six AP-1 binding sites, one NF-κB binding site and four GRE binding sites within the cd38 promoter. The putative transcription factor binding sites on the cd38 promoter are shown in Table 1.

Human airway smooth muscle (HASM) isolated from the trachealis muscle and propagated as described previously 910. were used in this study. The cells were plated at a density of 1.0 × 10⁴ cells/cm² and were cultured in DMEM supplemented with 10% FBS, 2 mM L-glutamine, 100 U/ml of penicillin, 0.1 mg/ml of streptomycin, and 0.25 μg/ml of amphotericin B. HASM cells were transfected as described below, then 24 hrs following transfection they were growth-arrested by maintaining them for at least 24 hrs in arresting medium containing no serum, but in the presence of transferrin and insulin prior to TNF-α (50 ng/ml) or dexamethasone (1 μM) treatment and measurement of luciferase reporter activity.

Transient transfections were performed with Lipofectamine™ 2000 according to the manufacturer's instructions. Cells (0.5–1 × 10⁵) in 500 μl of growth medium without antibiotics were plated one day before transfection. For the transfection, 0.8 μg of the vector DNA and 2 μl of Lipofectamine™ 2000 in 50 μl of Opti-MEM^® were mixed gently and incubated for 5 min at room temperature. Diluted DNA and lipofectamine were mixed and incubated for 20 min at room temperature to form complexes which were added to each well, and incubated at 37°C for 6 hrs. The cells were growth-arrested 24 hrs following transfection before exposing to TNF-α and dexamethasone. The cells were collected for luciferase reporter activity (described below).

Reporter gene assays were performed 24 hrs after transfection. Cell lysates were subjected to the Dual-Luciferase Reporter assay system and luciferase activities were measured with a luminometer (Lumat LB9507; Berthold). Cells were washed twice with phosphate-buffered saline (PBS) with no calcium and magnesium, and covered (0.1 ml/well) with Passive Lysis Buffer (Promega). The cells were then scraped, the lysate transferred to microcentrifuge tubes, which was mixed by vortexing for 15 s, then passed a few times through a needle and used for the reporter assay. A 20 μl aliquot of the lysate was mixed with 100 μl of luciferase assay reagent and placed in a luminometer to measure the firefly luciferase activity. The fluorescence was quenched by the addition of the Stop and Glo buffer and Renilla luciferase activity was measured after a 2 second delay. Firefly luciferase activities were normalized to Renilla luciferase activity to account for transfection efficiency. Samples were analyzed in triplicate and the experiment was repeated at least twice.

Nuclear extracts were prepared from growth-arrested HASM cells at confluence. The media were aspirated and washed with ice-cold PBS containing phosphatase inhibitors and the cells were scraped in 3 ml of the same buffer. The cells were pelletted by centrifugation at 1000 × g for 5 minutes and the supernatant discarded. The cells were resuspended in 500 μl 1× hypotonic buffer by pipetting several times, transferred to a chilled microcentrifuge tube and incubated for 15 mins on ice. Detergent (25 μl) was added, vortexed for 10 sec and pelleted by centrifugation at 14,000 × g for 30 sec at 4°C. The supernatant was removed and the nuclear pellet was resuspended in 50 μl of complete lysis buffer and vortexed for 10 sec. The mixture was incubated on ice for 30 min, vortexed briefly and pelleted at 14,000 × g for 10 min at 4°C. The supernatant (nuclear fraction) was aliquoted, protein content measured and stored at -80°C until use.

The protein concentration of the nuclear extract was quantitated using the Bradford protein assay (Bio-Rad, Hercules, CA). EMSA was performed as described earlier 910. The double-stranded oligonucleotides containing the consensus binding sites for NF-κB, AP-1, GRE and the putative cd38 binding sites (as shown in Table 3) were labeled with [γ-³²P]ATP (3,000 Ci/mmol at 10 mCi/ml) by T4 Polynucleotide Kinase (Promega, Madison, WI). Nuclear extracts (5 μg) from HASM cells or 1 μg of recombinant human GR protein were incubated for 30 min at room temperature with 0.4 pmol of double-stranded³²P-labeled oligonucleotide containing the consensus binding sites in a total volume of 10 μl in a buffer containing 20% glycerol, 5 mM MgCl₂, 2.5 mM EDTA, 2.5 mM DTT, 250 mM NaCl, 50 mM Tris-HCl (pH 7.5), and 0.25 mg/ml poly (dI-dC). After 30 min at room temperature, samples were separated on a nonreducing 4% polyacyrlamide gel using 0.5 M TBE buffer. The gels were dried and autoradiography carried out with intensifying screens at -70°C. To confirm specificity of the EMSA, competition assays were performed with a 100-fold excess of unlabeled NF-κB or AP-1 probe, or the SP-1 probe as a nonspecific competitor. Gel super shift assays were performed to confirm the specificity of the EMSA using anti-p65 or -p50 subunit of NF-κB, and anti-c-jun and anti-c-fos antibodies.

HASM cells isolated from three different donors were used in the experiments. The experiments involving EMSA and transient transfections of the constructs were repeated three times. The samples were compared by one-way ANOVA with Bonferroni's test for multiple comparisons. GraphPad PRISM statistical software program was used for statistical analyses and significance established at P value of ≤ 0.05.

To investigate the transcriptional regulation of CD38 expression in HASM cells, we cloned a putative 3 kb promoter fragment (GenBank Acc. No. DQ091293) from K562 cells into the pGL3 basic vector. The cd38 promoter sequence was examined for the presence of typical consensus elements using the GeneQuest module of Lasergene 6.0 program from DNASTAR. We identified six AP-1, one NF-κB, and four GRE motifs which are shown in Table 1. Using the electrophoretic mobility shift assay (EMSA), we examined whether transcription factors from HASM nuclear extracts or recombinant human GR proteins can bind to these putative binding sites following exposure of cells to TNF-α and dexamethasone. Oligonucleotides were synthesized from putative NF-κB, AP-1 and GRE binding sites (Table 3). The specificity of the EMSA was confirmed by competition experiments using unlabeled oligonucleotide sequences and gel supershift assays using specific antibodies. TNF-α increased the specific binding of nuclear proteins to consensus (Figure 1) as well as putative cd38 NF-κB sites (Figure 1), which was effectively competed with excess unlabeled consensus or putative sequences (Figure 1). EMSA also demonstrated that TNF-α increased the specific binding of nuclear proteins to the AP-1 consensus oligonucleotide sequence (Figure 2) and the putative cd38 AP-1 sites 1, 4 and 6 (referred to as AP1–1, AP1–4 and AP1–6 respectively), with the strongest binding to AP1–4 (Figure 2). Strong competition for binding to the consensus AP-1 sequence was observed with excess unlabeled AP1–4 sequence (Figure 3). AP-1 binding to the putative AP1–4 was confirmed by a gel supershift assay with anti-c-fos antibodies (Figure 3).

Glucocorticoid receptor (GR) binding to consensus GRE and putative GREs from cd38 sequences were performed using nuclear extract obtained from dexamethasone-treated HASM cells. Dexamethasone increased the binding of nuclear proteins to putative cd38 GRE sites 1, 3 (slight increase) and 4, but not to the GRE site 2 (Figure 4). This binding was inhibited with the respective excess unlabeled oligonucleotide sequences. To examine the direct binding of GR to putative GRE sites, we performed EMSA with recombinant human GR protein. There was binding of recombinant GR to labeled oligonucleotide putative cd38 GRE sites 1, 3 and 4 (Figure 5) as well as consensus GRE sequence (Figure 6). The binding of GR to the putative cd38 GRE sites 1, 3 and 4 was inhibited by excess unlabeled oligonucleotide sequences (Figure 5). Furthermore, the GR binding to the labeled consensus GRE sequence was inhibited by excess unlabeled cd38 putative GRE1, but not by the other putative GRE sequences (Figure 6) as well as by GRE-TAT, a GRE site from tyrosine aminotransferase gene (Figures 6). There was no binding of GR to an irrelevant sequence, as shown by a lack of binding to CREB binding sites (Figure 6). The specificity of GR binding to the consensus GRE sequence was further substantiated by gel supershift with an anti-GR antibody. The EMSA with HASM nuclear extract and putative GRE sites showed several binding complexes (Figure 4), which is not unexpected since GR is known to interact with many co-activators in the nucleus 1516.

The EMSA studies revealed that TNF-α increased the binding of nuclear proteins to the putative NF-κB site, and to some of the putative AP-1 sites in the cd38 promoter. Likewise, dexamethasone increased the binding of nuclear proteins selectively to some of the putative cd38 GREs. To investigate whether TNF-α modulates cd38 promoter activity directly, HASM cells were transiently transfected with a cd38 promoter-driven luciferase reporter construct. In the initial studies, we used the 3 kb promoter (Figure 7) and a truncated 1.8 kb promoter that lacks the NF-κB site and the AP1–4 site that exhibited very strong binding following TNF-α treatment. HASM cells were transiently transfected with the promoter constructs and luciferase activity was determined following exposure to TNF-α. TNF-α caused an increase in luciferase activity of the 3 kb promoter, but not the truncated 1.8 kb promoter, and dexamethasone decreased the TNF-α-induced activation of the 3 kb promoter (Figure 8)). To determine the transcription factor binding sites within the 3 kb promoter that are involved in the regulation of CD38 expression, HASM cells were transfected with site directed mutated constructs. For these studies, cd38 promoter luciferase constructs mutated at the NF-κB site or the AP1–4 site, or at both of these sites were used. Following exposure to TNF-α, luciferase activity was abolished in the promoter constructs with mutations of either the NF-κB or the AP1–4 sites, or mutation in both the sites (Figure 8). The EMSA results and the decreased activation of the promoter with mutations (that lack the NF-κB and the dominant AP1–4 binding sites) confirm a functional role for NF-κB and AP1–4 in the transcriptional regulation of CD38. Glucocorticoid regulation also involves binding to cd38 GREs and inhibition of NF-κB- and AP-1-dependent transcription.