Parathyroid hormone-related protein (PTHrP) was originally isolated from specific cancers as the primary cause of humoral hypercalcemia of malignancy, a paraneoplastic syndrome occurring in humans with a wide variety of malignancies (1). PTHrP also has been reported to be overexpressed by many types of neoplasms not associated with hypercalcemia (2). PTHrP is a polypeptide hormone with structural similarities to parathyroid hormone (PTH) (3)(4). Amino-terminal fragments of PTHrP exert PTH-like actions in bone and kidney by binding to a common receptor for PTH/PTHrP (PTH1 receptor), producing hypercalcemia (5)(6)(7)(8). High expression of PTHrP by cancer cells also has been proposed to play a role in the progression of breast cancer metastasis to bone (9)(10)(11). The human PTHrP gene is composed of nine exons (Fig. 1A ). Products of exons 5 and 6 are present in all PTHrP transcripts and encode for the prepro region and the majority of the mature peptide. Alternative splicing of the 3′ end produces three PTHrP isoforms 139-, 173-, and 141-amino acids in length. Transcriptional regulation of the PTHrP gene is achieved by three distinct promoters located at the 5′ end and identified as P1, P2, and P3, respectively. Alternative promoter usage has been evaluated previously by reverse transcription (RT)-PCR based on 5′ alternative splicing, and previous studies showed that P3-initiated transcripts were detectable in most tumors, whereas transcripts initiated by either P1 or P2 were present in only a subset of tumors (12)(13)(14)(15)(16). Genomic structure of human PTHrP (A) calibration curve generated by the LightCycler from PTHrP common region calibrators (B), and fluorescence curves for each of the calibrators (C). (A), PTHrP exons are represented by boxes (1–9). The promoter regions are identified above the map (P1, P2, and P3). P1 and P3 are TATA promoters, whereas P2 is a GC-rich promoter. The sites of alternative splicing and the three possible isoforms (139 aa, 141 aa, and 173 aa) are indicated below the map. In panels B and C, individual calibrators ranging from 5 × 102 to 5 × 108 copies are indicated by □ (B) and solid lines (C). We describe a novel real-time RT-PCR assay for the specific quantification and characterization of PTHrP mRNA expression, alternative promoter usage, and 3′ splicing in a variety of cancer cell lines as well as healthy and neoplastic human lung tissues. Neoplastic and adjacent healthy tissue samples from 15 human patients with a previous diagnosis of lung carcinoma were provided by the Division of Tissue Procurement at the James Cancer Hospital and Solove Research Institute at The Ohio State University. For the studies in this investigation, the four sample pairs with the greatest expression of PTHrP mRNA in lung carcinoma were chosen. The samples obtained were snap-frozen and stored at −80 °C. MT-2 (human adult T-cell leukemia/lymphoma) cells were cultured in suspension with RPMI 1640 supplemented with 100 mL/L fetal bovine serum. 786-O (human renal cell adenocarcinoma) and BEN (human squamous lung carcinoma) cells were cultured in monolayers to 70% confluence in RPMI 1640 (786-O) and DMEM/medium 199 (BEN) containing 100 mL/L fetal bovine serum. Total RNA from HaCaT cells (immortalized human keratinocytes) was generously provided by Dr. J. Foley (Indiana University, Bloomington, IN). Total RNA was extracted by homogenization of either 50–100 mg of frozen tissue or 5 × 106 cells with TRIzol reagent (Invitrogen) followed by standard chloroform extraction and isopropanol precipitation. Total RNA was incubated for 30 min with DNase I (10 U/μL; Roche) followed by a second total RNA isolation with use of TRIzol reagent. RNA was measured spectrophotometrically by the absorbance at 260 nm, and purity was determined by the ratio of the absorbance at 260/280 (A260/280). Reverse transcription reactions were performed in triplicate for each sample, according to the manufacturer’s protocol, with 2.5 μg of total RNA, oligo(dT)12–18, and 50 U of Superscript II reverse transcriptase (Invitrogen). Negative controls were prepared under the same conditions, but without addition of reverse transcriptase. Completed reverse transcription reactions were brought to a final volume of 50 μL by the addition of RNase and DNase-free doubly distilled water. cDNA calibrators were prepared by PCR amplification of BEN cell cDNA with designed or previously published primers (17)(18) as follows: PTHRP P1-initiated transcript (obrf 15.93-obrf 15.89), PTHRP P1/P2-initiated transcript (obrf 15.95-obrf 15.89), PTHRP P3-initiated transcript (P3 fw-obrf 15.89), PTHRP common region transcript (obrf 15.84-obrf 15.89), PTHRP 139-aa (obrf 15.84-obrf 15.90), PTHrP 141-aa (obrf 15.84-obrf 15.92), PTHrP 173-aa (obrf 15.84-obrf 15.91), and β2-microglobulin (B2M.fw-B2M.rv). The resulting cDNAs had unique bands by agarose gel electrophoresis and were purified by gel extraction (QIAquick gel extraction set; Qiagen). A second PCR amplification was performed with previously purified cDNA, and the concentration of DNA in purified PCR reactions was determined by spectrophotometry (A260 nm), and the copy number was determined according to the respective double-stranded DNA molecular weights. Calibrators in concentrations from 2.5 × 102 to 2.5 × 108 copies/μL were prepared by serial dilution. cDNA calibrators were cloned with use of the TOPO TA Cloning® Kit (Invitrogen), and their sequences were confirmed with an Applied Biosystems automated 3700 DNA Analyzer. Quantitative PCR assays that measured alternatively spliced PTHrP and β2-microglobulin gene expression were performed in duplicate; reactions contained 2 μL of cDNA sample or calibrator, 2.5 mM MgCl2, and 0.5 μM each of the PTHrP or the β2-microglobulin primers in a LightCycler apparatus (Roche Diagnostics) with use of QuantiTect SYBR Green PCR reaction mixture (Qiagen), according to the manufacturer’s protocol. Primers were designed using software Vector NTI 7.0 (Informax) based on sequences retrieved from the National Center for Biotechnology Information Nucleotides Database (http://www.ncbi.nlm.nih.gov). The following primer pairs were used for the specific amplification of each splice variant: PTHrP common region: All.fw (5′-GTCTCAGCCGCCGCCTCAA-3′) and All.rv (5′-GGAAGAATCGTCGCCGTAAA-3′); exon 5/6, 93-bp amplicon P1-initiated transcript: P1.fw (5′-CAGCCAGAAGAGCAGAGAGAA-3′) and P1.rv (5′-GCGAGTTGAAAACCGAGCG-3′); exon 1/3, 89- and 182-bp amplicons P1/P2-initiated transcript: P1/P2.fw (5′-GAAGCAACCAGCCCACCAGA-3′) and P1/P2.rv (5′-TGAGACCCTCCACCGAGC-3′); exon 3/5, 137-bp amplicon P3-initiated transcript: P3.fw (5′-GGAGAAAGCACAGTTGGAGTAGC-3′) and P3.rv (5′-TCTTTTGAGGCGGCGGCTGA-3′); exon 4/6, 167-bp amplicon 139-aa transcript: 139.fw (5′-TCTCAGCCGCCGCCTCAAAA-3′) and 139.rv (5′-AGAGAAGCCTGTTACCGT-3′); exon 5/7, 435-bp amplicon 141-aa transcript: 141/173.fw (5′-ACTCGCTCTGCCTGGTTA-3′) and 141.rv (5′-CAATGCCTCCGTGAATCG-3′); exon 6/9, 125-bp amplicon 173-aa transcript: 141/173.fw and 173.rv (5′-GTTCTTCTGTTGTTTTCCTT-3′); exon 6/8, 150-bp amplicon Target cDNA was amplified with the following conditions: 35 cycles of denaturation at 94 °C for 20 s, annealing at 55–60 °C for 20 s, and extension at 72 °C for 10–25 s. After PCR, we constructed a melting curve by increasing the temperature from 65 to 95 °C and plotting the first negative derivative (−dF/dT) of the fluorescence vs temperature to determine the melting temperature of the PCR products (Fig. 1B ). To ensure that the correct product was amplified in the reaction, we separated all samples by electrophoresis on a 2% agarose gel. The fluorescence of individual samples was measured by the LightCycler at the end of every cycle. The LightCycler software algorithm (Ver. 3.5), which uses the second derivative maximum method for quantification, automatically determined the crossing point (Cp) for the individual samples, including the calibrators. Calibration curves were constructed by plotting the Cp vs the logarithm of the number of copies for each calibrator. The numbers of copies in unknown samples were established by comparing their Cps with the calibration curve. P2-initiated copy numbers in samples were determined by subtracting the P1-initiated from the P1/P2-initiated copy numbers because the 3′ end of the P1-initiated transcripts measured by PCR was included in the P1/P2-initiated amplicon. Data were normalized by use of the ratio of the target cDNA concentration to β2-microglobulin to correct for differences in RNA quantity between samples. The specificity of each real-time PCR assay was assured by the primer design, and melting curves of the PCR products in each assay for the different PTHrP splice variants were characterized by one sharp peak indicating the correct melting temperature, which was 81–89 °C for the double-stranded amplicons. The final PCR products were the expected size, and the intensities of the bands correlated with the quantitative fluorescence obtained by the LightCycler when resolved by gel electrophoresis. No fluorescence or bands were observed in samples devoid of template cDNA or in negative-control samples prepared by the omission of reverse transcription. The sensitivity of the PCR assay was determined with the use of cDNA calibrators, with the limit of detection of the real-time PCR assay being 5 × 102 copies; Samples with Cp values below the limit of detection were considered nondetectable. Calibration curves were log-linear over the quantification range with correlation coefficients (r2) ≥0.99 and slopes ranging from −3.47 (β2-microglobulin) to −4.05 (PTHrP 139-aa; Fig. 1B ). The intraassay variability had a mean (SD) difference in the Cp between duplicates of 0.28 (0.03) cycles, with a CV of 1.18% and the interassay variability had a mean SD of 0.31 (0.04) cycles with a CV of 1.48% when the Cp value means were compared for the three runs of each transcript. When we evaluated the different transcripts for each of the four cell lines tested, the intra- and interassay variability was comparable. BEN and HaCaT cells showed the highest total PTHrP mRNA expression as determined by the copy number of common region transcripts, whereas the lowest expression was observed in MT-2 cells (Table 1 ). P3 was the most commonly used promoter in PTHrP gene transcription in all cell lines examined, representing 55–86% of the total of PTHrP common region transcripts. The copy number of P2-initiated transcripts was generally intermediate between that of P1- and P3-initiated transcripts, and its proportion ranged from 7% to 28% of the PTHrP common region transcript. P1-initiated transcripts were not detectable (MT-2) or were expressed in very low concentrations in all cell lines except the HaCaT cell line. PTHrP 139-aa copy number was the most abundant of the 3′ splice variants, representing 50–93% of PTHrP common region transcripts in all of the cell lines examined. The copy number of 141-aa was intermediate, ranging from 7% to 23% of the total, whereas 173-aa splice variants were the lowest. The number of PTHrP common region transcripts was 4–130 copies/103 copies of β2-microglobulin in lung carcinoma samples and 0.2–0.6 copies/103 copies of β2-microglobulin in adjacent healthy lung tissue samples. There was an overall average 131-fold increase in total PTHrP mRNA expression in lung carcinomas compared with adjacent tissue (Table 1 ). The mean ratio of promoter-initiated transcripts (P1:P2:P3) was 21:39:40 in carcinoma samples compared with 3:74:23 for healthy tissue. The mean copy numbers for all three promoter-initiated transcripts were significantly higher in carcinoma samples compared with the corresponding healthy tissue samples. Moreover, P3- and P2-initiated transcripts had the highest increase in mean absolute copy numbers in carcinoma compared with healthy tissue samples. PTHrP 139-aa and 141-aa were the most abundant of the 3′ splice variants and were expressed in comparable amounts in both carcinoma and noncarcinoma samples. The absolute copy numbers of these transcripts were significantly greater in carcinoma samples than in healthy tissue samples and were increased an average of 125- and 109-fold compared with healthy tissue samples. The 173-aa splice variant was the least expressed. Expression of PTHrP splice variants in four cell lines and four patients with lung carcinoma. Means (SD) were calculated from results of three experiments and are given as copy numbers of target amplicons/1 × 103 copy numbers of β2-microglobulin. Relative amounts are amounts of the individual splice variants relative to the PTHrP common region copy numbers. Expression of PTHrP splice variants in four cell lines and four patients with lung carcinoma. Means (SD) were calculated from results of three experiments and are given as copy numbers of target amplicons/1 × 103 copy numbers of β2-microglobulin. Relative amounts are amounts of the individual splice variants relative to the PTHrP common region copy numbers. Previous studies have examined the differential usage of the three PTHrP promoters in healthy tissue and tumor samples from individual patients based on semiquantitative RT-PCR (16), Northern blots (19), RNase protection assay(20), or quantitative competitive PCR (21), but no clear consensus on tissue-specific patterns of promoter usage has been reached. Our results are in accordance with previous studies that measured alternative promoter usage in various cell lines and in lung carcinoma (16)(17). Furthermore, several reports on the regulation of individual promoters in different cancers have identified some of the transcription factors regulating PTHrP expression and have highlighted the importance of the P3 promoter in gene transactivation (20)(22)(23)(24)(25). The results given in previous reports analyzing the prevalence of all three isoforms in healthy and neoplastic tissues by use of nonquantitative RT-PCR or Northern blots are comparable to our results. These studies showed a high prevalence of the 1–139 and 1–141 isoforms in most healthy tissues and tumors tested, whereas the 1–173 isoform was present more consistently in breast and prostate carcinoma compared with other neoplasms (17)(19)(26). Interestingly, increased expression of the 1–139 isoform in breast cancer cells has been associated with a higher prevalence of bone metastasis in vivo (11). PTHrP mRNA isoforms have a short (1–139 and 1–141 isoforms) to long (1–173 isoform) half-life ranging from 30 min to >4 h, depending on the 3′ splice variant transcribed (27)(28)(29)(30). We have recently reported that transforming growth factor-β stabilized 1–141 mRNA, possibly through cis-acting elements in the terminal coding region rather than the 3′ untranslated region (30). In conclusion, we found that PTHrP P3-initiated transcripts were the most abundant of the transcripts arising from alternative promoter usage in all cell lines and most of the lung carcinomas examined, whereas the PTHrP 139-aa splice variant was predominant in all cell lines and was coexpressed in comparable amounts with the 141-aa splice variant in lung carcinomas. Overall, our findings were in agreement with previous reports examining PTHrP alternative splicing in cancers. This novel quantitative real-time RT-PCR assay for the measurement of PTHrP mRNA expression, alternative promoter usage, and alternative 3′ splice variants in healthy and cancerous tissues offers numerous advantages over previously described methods and could be a useful tool for the study of PTHrP gene transcription, posttranscriptional regulation, and mRNA stability. This work was supported by grants from the National Cancer Institute (CA77911) and National Center for Research Resources (RR00168 to T.J.R.) and a V-Foundation Translational Award (to C.P.). C.P. is a Leukemia and Lymphoma Society Scholar. V.R. was the recipient of a Glenn Barber fellowship from the College of Veterinary Medicine at The Ohio State University. We thank Andrea L. Levine and Daniel Lima for excellent technical assistance.
No takes yet. Share an insight, caveat, or question.
Richard et al. (2003) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: