Generation, Characterization, and Use of Monoclonal Antibodies against Parathyroid Hypertensive Factor

Parathyroid hypertensive factor (PHF) may be useful as a diagnostic marker of salt-sensitive, low-renin hypertension. PHF was discovered in the plasma of salt-sensitive hypertensive humans (1) and was also found to be increased in spontaneously hypertensive rats, DOCA-salt-hypertensive rats, and Dahl-salt-sensitive rats, but not in two-kidney one-clip rats or in Dahl-salt-insensitive rats (2)(3). Studies on the mechanism of PHF action indicate that this substance acts directly on vascular smooth muscle cells to enhance Ca2+ influx (4), likely associated with depolarization of the plasma membrane by inhibition of voltage-gated K+ channels (5). Together these actions will sensitize vascular tissues to other vasoconstrictors, such as norepinephrine and angiotensin II (6). PHF may therefore be a causative factor in the development of hypertension in some individuals. It was found that PHF-positive (salt-sensitive) patients respond best to calcium channel blockers and diuretics and that PHF-negative patients respond better to angiotensin-converting enzyme inhibitors and beta blockers (7). Recently, an enzyme immunoassay for detection of PHF in human plasma has been reported that uses anti-PHF oligoclonal antibodies (8). The present study describes the further development, characterization, and clinical application of monoclonal antibodies (MAbs) against PHF. Male BALB/c mice were immunized with partially purified PHF (9) prepared from medium harvested from cultured parathyroid glands of spontaneously hypertensive rats, as was described previously (10). Splenocytes were harvested and fused with SP/0-2 myeloma cells by standard procedures (11). The recloning procedure involved limiting-dilution conditions in 96-well plates (0.3 cells/well). The bovine serum albumin (BSA) conjugate of PHF was synthesized by standard procedures (12) and used in a direct ELISA for anti-PHF antibodies. Microtiter plates were coated with BSA-PHF (100 μL/well; 1:2000 dilution). After the plates were washed with phosphate-buffered saline (PBS) containing 5 mL/L Tween 20, supernatants of hybridoma cell cultures were added (100 μL/well) and incubated for 2 h at 37 °C. The bound antibodies were detected with peroxidase-conjugated goat anti-mouse polyvalent immunoglobulins (100 μL/well; 1:2000 dilution), with 3,3′,5,5′-tetramethylbenzidine as the substrate. Two cell lines, C40D21 B12 P4 and C40D21 B22 P6 (designated MAb B12 and MAb B22, respectively), as well as their parent cell line, C40D21 B2 (designated MAb B2), had 100% recloning efficiency and the best anti-PHF activity. These were chosen for the large-scale production of MAbs in ascites. All antibodies were shown to be of the IgM isotype. Hybridoma cell lines (1–5 × 105 cells/mL) were injected into pristane-primed BALB/c mice. Anti-PHF antibodies were purified from ascites on Superdex G-200 and eluted with PBS. All three cell lines were tested by bioassay for their ability to inactivate PHF mixed with antibody before injection into normotensive rats. The blood pressure bioassay for PHF (13) was adapted for bioassay of anti-PHF antibodies. MAbs produced by all three cell lines inactivated PHF mixed with antibody in concentrations of 300–400 mg/L. PHF was measured in plasma samples by competitive ELISA as described previously (8). Briefly, microtiter plates were coated with 2 mg/L purified anti-PHF antibodies (100 μL/well) in phosphate buffer. The plates were washed, and triplicates of controls, unknown samples, or each PHF calibrator in plasma (50 μL/well; 1:20 dilution with PBS) were added, followed by PHF-horseradish peroxidase conjugate (50 μL/well; 1:2000 dilution with PBS containing 10 g/L BSA). Bound PHF-horseradish peroxidase was measured with 3,3′,5,5′-tetramethylbenzidine as the peroxidase substrate. The calibration curves for the ELISAs using any of the three MAbs were linear in the range 0.03–1 unit/mL when plotted on a log-linear basis. The limit of detection for PHF (the smallest single value that could be distinguished from zero) was calculated to be 0.02 units/mL (the mean + 2 SD from 20 determinations of the zero calibrator). The precision of the ELISA using MAb B2 was estimated with three different plasma pools of samples containing PHF. The intraassay imprecision (as CV), determined from the mean of triplicates measured 18 times in the same assay, was 9.7%. The interassay imprecision, determined from the mean of the average of triplicates for eight separate runs by three operators, was 14%. For accuracy studies, human plasma samples with three known concentrations of PHF were used. Recoveries, determined by comparing the expected concentration vs the measured concentration, were 87–110%. Three human samples were diluted with a pool of normal human plasma, and recoveries ranged from 84% to 108%. All three MAbs reacted equally with either rat or human PHF. Further studies were carried out to determine whether other circulating substances might cross-react with the anti-PHF MAbs (8). The list of substances tested included many common circulating hormones, including parathyroid hormone and its fragments, and endothelin, two products of the parathyroid gland (8). MAbs B12 and B22 showed minimal (0.5–2.8%) cross-reactivity with six substances: secretin, spermidine, 5-hydroxytryptamine, bradykinin, estrogen, and calcitonin gene-related peptide. There was no detectable cross-reactivity for any of the compounds with MAb B2. All subsequent studies were carried out with MAb B2. Patients were randomly selected for PHF analysis from volunteer patients routinely seen in Family Practice clinics in Edmonton, Alberta. The criteria for selection were that patients were not taking any medications, including antihypertensive medications, and had no known medical condition that could interfere with the PHF analysis. Blood was collected into heparin-containing tubes. All patient samples were collected in accordance with University of Alberta Ethical Guidelines. For the normotensive patients [mean arterial pressure (MAP) <95 mmHg; n = 47; 19 males and 28 females], the mean (SE) age was 40 (2) years, and the mean (SE) MAP was 86.3 (0.9) mmHg. For the hypertensive group (MAP ≥95 mmHg; n = 27; 14 males and 13 females), the mean (SE) age was 49 (3) years, and the mean (SE) MAP was 104.4 (1.5) mmHg. Analysis of fresh plasma samples yielded PHF values of 0.042–0.673 units/mL for all patients. PHF was detectable in all individuals, which suggests that PHF is usually present in human plasma. PHF measured with the ELISA was compared with PHF measured with the traditional blood pressure bioassay in 10 samples (Fig. 1A ). Results indicated that there is a significant linear relationship between results obtained with the two assay methods [slope, 4.103; y-intercept, −2.987 units/mL; R = 0.6409; P = 0.0458 (significant)]. Attempts to correlate PHF concentrations with patients’ blood pressures initially yielded inconclusive results. However, when the patients were split into normotensive and hypertensive groups, we were able to draw several important conclusions. Results for normotensive individuals (Fig. 1B ) indicate that the PHF concentration is not correlated with blood pressure [slope, 0.00119 ± 0.00233; y-intercept, 0.0664 ± 0.202 units/mL; R = 0.0488; P = 0.611 (not significant); n = 47]. The mean (SD) concentration of circulating PHF measured with the ELISA was 0.169 (0.103) units/mL. In contrast, in hypertensive individuals, the mean (SD) PHF concentration was 0.212 (0.158) units/mL, and mean arterial pressure was significantly correlated with PHF concentrations [Fig. 1C ; slope, 0.008014 ± 0.003734; y-intercept, −0.6246 ± 0.3908 units/mL; R = 0.394, P = 0.0417 (significant); n = 27]. Correlation between plasma PHF measurements obtained with the blood pressure bioassay and ELISA (A) and between blood pressure and PHF as measured with the ELISA (B and C). (A), the PHF bioassay values are the change in mean arterial pressure in the bioassay rat 50 min after intravenous injection of 1 mL of the filtered (0.45 μm) plasma sample, and each point is the mean for at least four measurements; the error bars represent the SE. (B and C), correlation of blood pressure and PHF concentrations measured with the ELISA in normotensive (B) and hypertensive (C) individuals. Our goal was to develop MAbs with high affinities to both the rat and human PHF antigens. The antibodies produced were selective for PHF, as demonstrated in the bioassay and cross-reactivity studies. Results generated with the PHF ELISA to detect PHF in biological samples correlated well with results obtained for PHF with the bioassay. The ELISA offers a distinct advantage over the bioassay in that it is faster, less expensive, and quantitative, making it a more reliable system. Although the antigen used in this ELISA was not highly purified and the structure has not been completely elucidated, the strong correlation between this ELISA and the bioassay indicate that this assay may be useful in studies that will further our understanding of this substance. Highly pure antigen, as well as improved understanding of the etiology of PHF-related disease, will be required for this assay to be used as a routine clinical tool. In the normotensive group, PHF was not correlated with blood pressure, but in the hypertensive group PHF was correlated with blood pressure. The number of hypertensive samples was much lower than the number of normotensive samples because of the restriction placed on sample selection. In most cases, once people are diagnosed with high blood pressure, therapy is initiated. Blood pressures in the hypertensive group were therefore in the lower range of hypertension. In both groups, there were some values that fell outside the expected correlation. This observation is not unexpected because our previous studies have demonstrated that PHF may occur in some patients with other conditions, such as type 2 diabetes mellitus (14). PHF expression appears to precede increases in systemic blood pressure, and because hypertension is a heterogeneous disease, not all hypertensive patients express PHF. Creation of an ELISA for PHF in biological samples may facilitate future research into the biology and pathology of PHF in humans. Future clinical studies with larger sample sizes will be required to clearly define the relationship of PHF to hypertension. Availability of a test with which clinicians may subclassify hypertension may ultimately allow for selection of the most appropriate therapeutic regimen without requiring the empirical approach that is currently used. We would like to acknowledge the technical contributions of Lynda Tang and Lei Zhang. This work was supported by University-Industry Grant UI-13521 from the Medical Research Council of Canada (to C.G.B.).

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.