TY - JOUR
AU - Valdes,, Roland
AB - Abstract Background: Ouabain-like factor (OLF) and its newly discovered reduced species, dihydroouabain-like factor (Dh-OLF), are mammalian cardenolides whose structural and functional characteristics are similar to the plant-derived compounds ouabain and dihydroouabain. These endogenous compounds are believed to be produced by the adrenals and to constitute part of an hormonal axis that may regulate the catalytic activity of the α-subunit of Na+,K+-ATPase. We developed antibodies sufficiently specific to distinguish between OLF and Dh-OLF, and in this study demonstrate the selective secretion of OLF and Dh-OLF from human H295R-1 adrenocortical cells in culture. Methods: We used reversed-phase HPLC, inhibition of Na+,K+-ATPase catalytic activity, and two enzyme immunoassays developed with antibodies specific to ouabain and dihydroouabain to purify and characterize the secretion of these two compounds by human adrenal cells in culture. Purified antisera had high titers (1 × 106 for ouabain and 8 ×105 for dihydroouabain) and were specific to their corresponding antigens. Results: Human H295R-1 cells grown in serum-free medium secreted 0.18 ± 0.03 pmol of OLF and 0.39 ± 0.04 pmol of Dh-OLF per 106 cells in 24 h. Both OLF and Dh-OLF inhibited the ouabain-sensitive catalytic activity of the sodium pump (0.03 μmol/L OLF inhibited 29% of the catalytic activity; 0.07 μmol/L Dh-OLF inhibited 17%). Stimulation of the cell culture by dibutryl cAMP increased the secretion of Dh-OLF 50% over control (unstimulated), whereas the secretion of OLF did not increase significantly. Conclusions: OLF and Dh-OLF are secreted by human adrenal cells, and antibodies specific to these two compounds can be developed, using the plant-derived counterparts as antigens. The secretion of Dh-OLF is responsive to a cAMP-dependent stimulation mechanism, whereas OLF is not. Our data suggest that either the secretory or biosynthetic pathways for production of these two compounds by human adrenal cells may have different control mechanisms or that they may be linked via a precursor–product relationship. Compounds that resemble plant-derived cardiac glycosides have been found in human plasma (1), urine (2), mammalian adrenal glands (3)(4)(5)(6), hypothalamus (7)(8), and mammalian adrenal cells in culture (9)(10)(11). The endogenous mammalian-derived cardiac glycosides, often referred to as “mammalian cardenolides”, inhibit the catalytic activity of the α-subunit of Na+,K+-ATPase (12). A mechanism for direct hormonal control of the sodium pump may play an important role in understanding pathophysiologic conditions linked to control of blood volume and homeostasis (13), as well as sodium-pump-related mechanisms underlying kidney (14), cardiac (15) and neurologic disorders (16). Evidence is now mounting that suggests that these mammalian-derived compounds have hormone-like properties (12)(17). Two main classes of mammalian cardenolides have been identified: the cardiac glycoside-like factors, such as digoxin-like factors (18) and ouabain-like factors (OLFs)2 (1)(19), and the bufodienolide-like factors, such as bufalin-like factors (20)(21) and proscillaridin-like factors (7)(22). Although the chemical structures of the mammalian-derived digoxin-like factors have not been completely elucidated, several reports do provide compelling evidence on the structure of OLFs (23)(24). OLFs have an oxidized lactone ring attached to the steroid nucleus (see Fig. 1) similar to their plant-derived counterpart, ouabain (12). When saturated with two protons, this lactone moiety significantly reduces the affinity for binding to the α-subunit of Na+,K+-ATPase (25)(26). Importantly, we recently isolated from bovine adrenocortical tissue a dihydroouabain-like factor, we termed Dh-OLF, that is intrinsically less biologically active than OLF and is characterized by having a chemically reduced lactone ring similar to that of dihydroouabain (4). These findings underscore the importance of understanding the natural biosynthesis of these newly discovered chemically reduced forms of the mammalian cardenolides (27). Figure 1. Open in new tabDownload slide Conjugation strategy and structures of ouabain and dihydroouabain. Conjugation of plant-derived ouabain and dihydroouabain to the proteins KLH and BSA is through the rhamnose residue, which exposes the lactone ring for immunogenic response. The only difference between these two cardiac glycosides is the lactone ring. In ouabain this ring is oxidized, whereas in dihydroouabain it is reduced. We previously reported that commercially available dihydroouabain has two isomers (20). These two components, dho-A and dho-B, differ in the orientation of the proton on the C-20 position. Antigenic response was therefore directed against both components. Both cardiac glycosides, ouabain and dihydroouabain, consist of a four-ring steroid nucleus, a rhamnose sugar attached at the C-3 position of the steroid, and a lactone ring attached at the C-17 position. Ouabain/dihydroouabain-KLH conjugates were used to immunize rabbits for antisera production. Ouabain/dihydroouabain-BSA conjugates were used to coat microtiter plates as the plate-bound competitor in the aforementioned EIA. Figure 1. Open in new tabDownload slide Conjugation strategy and structures of ouabain and dihydroouabain. Conjugation of plant-derived ouabain and dihydroouabain to the proteins KLH and BSA is through the rhamnose residue, which exposes the lactone ring for immunogenic response. The only difference between these two cardiac glycosides is the lactone ring. In ouabain this ring is oxidized, whereas in dihydroouabain it is reduced. We previously reported that commercially available dihydroouabain has two isomers (20). These two components, dho-A and dho-B, differ in the orientation of the proton on the C-20 position. Antigenic response was therefore directed against both components. Both cardiac glycosides, ouabain and dihydroouabain, consist of a four-ring steroid nucleus, a rhamnose sugar attached at the C-3 position of the steroid, and a lactone ring attached at the C-17 position. Ouabain/dihydroouabain-KLH conjugates were used to immunize rabbits for antisera production. Ouabain/dihydroouabain-BSA conjugates were used to coat microtiter plates as the plate-bound competitor in the aforementioned EIA. The steroid-like nature of the endogenous cardenolides (21)(28) and their increased concentrations in adrenal cortical tissues (5)(29) suggest that the steroidogenic pathway is likely involved in their biosynthesis. One approach for elucidating the biosynthetic pathway of a natural steroid-like compound is to selectively stimulate steroid synthesis in adrenal cells and subsequently characterize the secretion of the compounds (11). Hence, the availability of well-characterized cell culture model systems is important for undertaking this work. Although several animal adrenal cells have been shown to secrete OLFs (9)(10), to date no reports have documented the secretion of both OLF and Dh-OLF from any human cell line. In addition to a well-defined cell culture system, the availability of immunoassays sufficiently specific for distinguishing the two forms (unsaturated and reduced) of OLF molecules would provide a valuable tool for characterizing the biosynthesis and physiologic role of these compounds. In this study we report the development of two specific enzyme immunoassays (EIAs) for detection and quantification of OLF and Dh-OLF. Using a combination of reversed-phase HPLC, inhibition of Na+,K+-ATPase catalytic activity, and two immunoassays developed with antibodies specific to ouabain and to dihydroouabain, we characterized the secretion of these two compounds from human adrenal cells (H295R-1) in culture. Our data indicate that although both OLF and Dh-OLF are secreted by human adrenal cells, when the cells are challenged with cAMP, the secretion of Dh-OLF is favored over that of OLF. These results suggest a cAMP-dependent pathway for production of these compounds with possible differential regulation. A preliminary account of portions of this work has been presented (24). Materials and Methods materials All chemicals used were reagent grade. Chromatography-grade CH3CN was obtained from Aldrich Co.;5-sulfosalicylic acid, CaCO3, ouabain, dihydroouabain, porcine cerebral cortex, all reagents for catalytic inhibition of the sodium pump [ATP, ammonium molybdate, Tween 80, and bovine serum albumin (BSA)] as well as goat anti-rabbit horseradish peroxidase conjugate, phosphate-buffered saline (PBS)-Tween (pH 7.4), and antibodies to progesterone were purchased from Sigma. 3,3′,5,5′-Tetramethylbenzidine (TMB) soluble reagent and TMB stop buffer were purchased from SKYTEK Laboratories. Human cells in culture were disrupted by sonication in a Microson ultrasonic cell disruptor (Heat Systems Inc.). For the EIA assays, microtiter plates (Immulon 2) were purchased from Dynex Technologies, Inc. Ultraviolet absorbance was measured on an SLT Rainbow Microplate Reader, and the data were analyzed by WinSelect Data analysis Software (Microsoft Corp). For sodium pump inhibition assays, disposable nonsterile 96-well flat-bottomed polystyrene microtiter plates were purchased from Corning, and the absorbance was measured on a Multiskan MCC/340 Microplate Reader II (DuPont). methods Ouabain and dihydroouabain conjugates. Ouabain and dihydroouabain were each conjugated to both keyhole limpet hemocyanin (KLH) and BSA by HTI Bio-products Inc. to our specifications: KLH and BSA were conjugated through the rhamnose sugar moieties of ouabain and dihydroouabain (see Fig. 1) to increase the antigenic response in rabbits. The ouabain-BSA and dihydroouabain-BSA conjugates were used to coat the microtiter plate wells in the development of both EIAs as detailed below. Ouabain and dihydroouabain antisera. Production of ouabain and dihydroouabain antisera from rabbits was performed by HTI Bio-products according to our specifications and schedule of injection/boosting. Briefly, KLH-conjugated ouabain and dihydroouabain were each emulsified in Freund’s complete adjuvant and used in intradermal immunization of rabbits. Three weeks later, the animals were boosted for the first time, followed by a total of four boosts, each 2 weeks apart. Blood samples from the rabbits were drawn 3 weeks after the primary injection and every 2 weeks after subsequent booster injections to test antiserum titers. Animals were bled 14 days after the final boost, and serum was harvested and stored at −80 °C for further use. Immunoglobulins were further purified by ammonium sulfate precipitation as described elsewhere with some modifications (30). Briefly, serum was clarified by centrifugation at 10 000g at 4 °C. The supernatant was decanted and stirred slowly in a beaker. A saturated solution of (NH4)2SO4 was slowly added until precipitate started forming and until an equal volume of saturated (NH4)2SO4 was added. The precipitate was allowed to form for an additional 4–5 h. The contents of the beaker were then centrifuged at 10 000g at 4 °C for 30 min, and the precipitate was dissolved in PBS (one-half of the original volume of serum). The IgG concentration was measured spectrophotometrically at 280 nm. Finally, the solution was dialyzed for 12 h at 4 °C against 10 L of PBS, with several changes; sodium azide to a final concentration of 10 mmol/L was added, and serum IgG extracts (anti-ouabain and anti-dihydroouabain IgG) were stored at 4 °C for further purification. The resulting extracts were further purified by affinity purification on a protein A column according to the manufacturer’s recommendations (Pierce). A protein A-immobilized gel slurry was packed into a column, degassed, and equilibrated; dilute samples (2:1 in binding buffer) of the extracts were passed through each of the columns. Each column was then washed eight times with binding buffer, followed by elution of the bound IgGs with elution buffer (50 mmol/L glycine-HCl buffer, pH 2.5–2.8). The eluates were collected in 1-mL fractions, the pH of the fractions was immediately increased to neutral by the addition of 100 μL of 1 mol/L Tris-Cl, pH 7.6, and the absorbance spectra of the proteins were monitored at 280 nm. Ouabain and dihydroouabain EIAs. Ouabain-like and dihydroouabain-like immunoreactivities were measured by EIAs using antisera containing polyclonal antibodies to ouabain and dihydroouabain, respectively. Both EIAs were based on the competitive binding of free cardiac glycosides (ouabain/dihydroouabain) or free mammalian cardenolides (OLF/Dh-OLF) in the sample and bound cardiac glycoside to a constant amount of the respective antibodies in the EIAs. The presence of cardiac glycosides or mammalian cardenolides in the microtiter-plate wells decreases antibody binding to the cardiac glycoside-coated plate. This in turn causes a reduction in the binding of the goat anti-rabbit horseradish peroxidase conjugate to the primary antibody and, therefore, a decrease in the signal obtained from the conversion of the substrate by the conjugated enzyme. The signal intensity is inversely proportional to the amount of cardiac glycoside or mammalian cardenolide in the well. The procedure for the EIA is as follows: Microtiter plate wells were coated for a minimum of 18 h at 4 °C with ∼0.5 μg/well of BSA-conjugated cardiac glycoside diluted in carbonate–bicarbonate coating buffer containing 15 mmol/L Na2CO3, 35 mmol/L NaHCO3, and 3.1 mmol/L NaN3 in water (pH 9.6). After coating, the plate was washed with 0.5 mL/L Tween 20 in PBS and then blocked with a 10 g/L BSA solution in PBS for 1 h at 37 °C. After washing, the calibrators and samples were added, followed by the addition of the appropriate antibody, and the plate was incubated at room temperature for 1 h. After another washing step, goat anti-rabbit horseradish peroxidase conjugate was added and allowed to bind to the primary antibody for additional 2 h at room temperature. Finally, the plate was washed, and 100 μL of TMB soluble reagent was added to each well. Color development was monitored at 650 nm for a maximum of 30 min, after which the reaction was stopped with 100 μL of TMB stop buffer and the plate was read at 450 nm. The readings were blanked and adjusted for nonspecific binding. We used the plant-derived ouabain and dihydroouabain as calibrators in the respective immunoassays. Therefore, all concentrations and amounts of measured OLF and Dh-OLF refer to the respective immunoequivalences to their plant-derived counterparts. Cell culture growth conditions and stimulation of steroidogenesis. Human adrenal cells (H295R-1) were a gift from Dr. William E. Rainey (University of Texas Southwestern Medical Center, Dallas, TX). The cells were grown and maintained in DMEM/F12 medium containing 25 mL/L NuSerum type I (Collaborative Biomed Products), 10 g/L ITS culture supplements, and antibiotics in 75-cm2 flasks at 37 °C under a humid atmosphere of 5% CO2–95% air, as described previously (31). Tissue culture flasks were seeded with H295R-1 cells and grown to ∼80% confluent density at 37 °C in 5% CO2 (∼1 × 106 cells). The cells were then washed twice with PBS, and serum-free DMEM/F-12 medium was added to the cells. The incubation was continued for an additional 24 h, after which the medium was removed and stored at −80 °C for isolation of OLF and Dh-OLF. Steroidogenesis was stimulated by replacing the medium with fresh serum-free culture medium containing 1 mmol/L dibutyryl cAMP [(Bu)2cAMP] and incubating for an additional 24 h, after which OLF, Dh-OLF, and progesterone in the cell medium were measured. Progesterone was measured directly in the cell medium by RIA as described previously (32). The data are expressed as μg progesterone/L of medium, and the mean values ± SD were determined from triplicate determinations for each sample. To isolate OLF and Dh-OLF, the culture medium was extracted by solid-phase extraction followed by HPLC purification, and the final purified substance was reconstituted in plasma before measurement by EIA. The amount of each compound measured after 24 h (OLF, Dh-OLF, and progesterone) in serum-free medium in the absence of (Bu)2cAMP was set as 100%, and the amount of each compound (OLF, Dh-OLF, and progesterone) secreted after 24 h in (Bu)2cAMP-supplemented medium was set relative to 100%. For all experiments above, at the end of the incubation, medium was collected and stored frozen at −80 °C before analysis. Approximately 1 × 106 cells/flask were harvested by scraping with a rubber policeman and centrifuged; the cell pellet was washed twice with PBS and placed in distilled, deionized H2O (ddH2O) for sonication. After sonication three times (30 s each time), disrupted cells were centrifuged, and the released contents in ddH2O were collected and frozen at −80 °C for further use. The collected cell medium and cell contents in water were further purified by solid-phase extraction and HPLC with two different chromatographic gradients to isolate and purify the endogenous compounds as described below. Isolation and purification of OLF and Dh-OLF. We isolated and purified OLF and Dh-OLF from human adrenocortical cells as described previously (4). We diluted the cell extract (or cell contents in the case of cell pellet analysis) with two volumes of ddH2O, precipitated the denatured proteins with 10 g/L 5-sulfosalicylic acid (90 °C for 10 min), and immediately added 1 mol/L Tris-Cl (pH 8.0) to a final concentration of 10 mmol/L to neutralize the pH. We then centrifuged the extract and decanted and filtered the supernatant. Initial purification was performed with reversed-phase C18 Sep-Pak extraction cartridges. We then eluted the compounds of interest with two volumes of 200 mL/L CH3CN in ddH2O. The CH3CN was evaporated, and the samples were reconstituted in ddH2O and filtered. Chromatographic separation of OLF and Dh-OLF involved two HPLC chromatographic steps. HPLC was performed with a C18 reversed-phase μBondapak column (3.9 × 300 mm; 10-μm particle size) connected to a Waters 600E system controller and a Waters 996 photodiode array detector. HPLC fractions were collected with a Waters Fraction Collector (Millipore Corp.) and evaporated with a RC 10.22 Centrifugal Vacuum Concentrator connected to a RCT 60 Refrigerated Trap (both from Jouan). The first HPLC step involved a linear gradient of 20–80% CH3CN in water over 30 min, and eighty 0.5-mL (1 min) fractions were collected, dried, and reconstituted in ddH2O. The fractions eluting at 4–6 min into the linear gradient were measured by ouabain and dihydroouabain EIA before any further analysis. These fractions from the linear gradient were then rechromatographed using an isocratic mobile phase of 100 mL/L CH3CN in ddH2O for 40 min to separate OLF (29 min) and Dh-OLF (27.5 min) from each other, as described previously (4). All fractions of interest were resuspended in plasma and then analyzed by the appropriate EIA. Na+,K+-ATPase catalytic activity inhibition assay. This assay was used to measure the effect of OLF and Dh-OLF on phosphate release in the hydrolysis of ATP, based on the method of Chan and Swaminathan (33) with minor modifications (25), as we detailed previously, to increase the sensitivity of the assay for samples of small volume. Briefly, to 20 μL of sample/calibrator (containing the desired concentration of calibrator), equal volumes of porcine cerebrocortical Na+,K+-ATPase solution (1 U/L) and ATP (10 mmol/L) were added and incubated at 37 °C for 10 and 15 min, respectively. Sample, calibrator, and ATP solutions were all resuspended in Tris-Cl buffer (pH 7.8). We then added molybdate solution (150 μL), and color development was allowed to proceed for a maximum of 30 min before measurement at 340 nm. The color intensity is proportional to ATP breakdown and, therefore, Na+,K+-ATPase activity. All samples were corrected for background, averaged, and normalized to ouabain-sensitive Na+,K+-ATPase activity (100% inhibition at 1 mmol/L ouabain). The percentage of inhibition of Na+,K+-ATPase activity represents the proportion of ouabain-sensitive ATPase activity that is inhibitable by either OLF or Dh-OLF. Results development of antisera and immunoassays We used commercially purchased ouabain and dihydroouabain without further purification for the production of antibodies in rabbits. The resulting titers of purified antisera were 1 ×106 for ouabain antiserum and 8 ×105 for dihydroouabain antiserum. Titers are expressed as the reciprocals of the serum dilutions that produced an absorbance at 492 nm (A492 nm) of 0.2 in an ELISA with BSA-conjugated antigen on the solid phase. Except for the chemical saturation of the lactone ring, ouabain and dihydroouabain are structurally identical. To distinguish between these two compounds and to render them immunogenic at or near the lactone ring position, each glycoside was conjugated to KLH and BSA through the rhamnose sugar moiety. The rhamnose ring was opened, and the two aldehyde moieties created were coupled to NH2 groups on KLH and BSA (Fig. 1). We tested these antisera for specificity toward structurally related compounds. As shown in Table 1, antisera raised against dihydroouabain demonstrated cross-reactivities of only 1.4% with ouabain and <0.05% with its aglycone, ouabagenin. Furthermore, the antisera did not detect (<0.01%) digoxin or its aglycone, digoxigenin, and only slightly detected dihydrodigoxin (<0.05% cross-reactivity). The antisera raised to ouabain demonstrated cross-reactivities of 18% with dihydroouabain and 10% with ouabagenin. Thus, the antiserum developed to measure Dh-OLF seem more specific than the one developed to measure OLF for structural differences in the lactone ring. Shown in Fig. 2 are typical calibration curves for OLF (Fig. 2A) and Dh-OLF (Fig. 2B) for assays using the antiserum indicated and the plant-derived ouabain or dihydroouabain, respectively, as the calibrators. The within-run imprecision (CV) for the OLF immunoassay was 7.5%, 6.4%, and 7.1% for ouabain concentrations of 0.34, 1.02, and 3.40 nmol/L, respectively. For the Dh-OLF assay, the within-run imprecision (CV) was 3.8%, 12%, and 6.5% at comparable concentrations and for six determinations each. The lower limit of detection, defined as 2 SD above the mean of the zero calibrator, was 72 pmol/L for the OLF and 60 pmol/L for the Dh-OLF assay. Figure 2. Open in new tabDownload slide Calibration curves of EIAs for ouabain (A) and dihydroouabain (B). %C/C0 (y axis) is calculated by dividing the signal obtained from a sample containing cardiac glycoside by that of a sample containing no cardiac glycoside and multiplying by 100%. The mean of three experiments, each done in duplicate, is plotted, and the SE is represented by error bars. Amounts of ouabain and dihydroouabain are represented by molar concentrations (M). Figure 2. Open in new tabDownload slide Calibration curves of EIAs for ouabain (A) and dihydroouabain (B). %C/C0 (y axis) is calculated by dividing the signal obtained from a sample containing cardiac glycoside by that of a sample containing no cardiac glycoside and multiplying by 100%. The mean of three experiments, each done in duplicate, is plotted, and the SE is represented by error bars. Amounts of ouabain and dihydroouabain are represented by molar concentrations (M). Table 1. Cross-reactivities of ouabain and dihydroouabain antibodies.1 Substance . Cross-reactivity, % . . . Anti-ouabain . Anti-dihydroouabain . Dihydroouabain 18 100 Ouabain 100 1.40 Ouabagenin 10 <0.05 Digoxin 0.66 <0.01 Dihydrodigoxin 6.70 <0.05 Digoxigenin 3.30 <0.01 Cholesterol 0.03 0.70 Bufalin 10 0.07 Substance . Cross-reactivity, % . . . Anti-ouabain . Anti-dihydroouabain . Dihydroouabain 18 100 Ouabain 100 1.40 Ouabagenin 10 <0.05 Digoxin 0.66 <0.01 Dihydrodigoxin 6.70 <0.05 Digoxigenin 3.30 <0.01 Cholesterol 0.03 0.70 Bufalin 10 0.07 1 Cross-reactivity is expressed as the molar concentration of the calibrator (ouabain or dihydroouabain) at 50% c/c0 over the molar concentration of the compound at 50% c/c0 multiplied by 100%. Ouabain cross-reactivity was measured by ouabain EIA, and dihydroouabain cross-reactivity was measured by dihydroouabain EIA. Cross-reactivity for the following compounds was <0.05% to ouabain and dihydroouabain: pregnenolone, corticosterone, progesterone, deoxycorticosterone, aldosterone, cortisone, hydroxyprogesterone, uracil. Open in new tab Table 1. Cross-reactivities of ouabain and dihydroouabain antibodies.1 Substance . Cross-reactivity, % . . . Anti-ouabain . Anti-dihydroouabain . Dihydroouabain 18 100 Ouabain 100 1.40 Ouabagenin 10 <0.05 Digoxin 0.66 <0.01 Dihydrodigoxin 6.70 <0.05 Digoxigenin 3.30 <0.01 Cholesterol 0.03 0.70 Bufalin 10 0.07 Substance . Cross-reactivity, % . . . Anti-ouabain . Anti-dihydroouabain . Dihydroouabain 18 100 Ouabain 100 1.40 Ouabagenin 10 <0.05 Digoxin 0.66 <0.01 Dihydrodigoxin 6.70 <0.05 Digoxigenin 3.30 <0.01 Cholesterol 0.03 0.70 Bufalin 10 0.07 1 Cross-reactivity is expressed as the molar concentration of the calibrator (ouabain or dihydroouabain) at 50% c/c0 over the molar concentration of the compound at 50% c/c0 multiplied by 100%. Ouabain cross-reactivity was measured by ouabain EIA, and dihydroouabain cross-reactivity was measured by dihydroouabain EIA. Cross-reactivity for the following compounds was <0.05% to ouabain and dihydroouabain: pregnenolone, corticosterone, progesterone, deoxycorticosterone, aldosterone, cortisone, hydroxyprogesterone, uracil. Open in new tab secretion of Dh-OLF and OLF from human adrenal cell cultures Human adrenal cells (H295R-1) grown in serum-free medium for 24 h produced 0.18 ± 0.03 pmol of ouabain-equivalent OLF and 0.39 ± 0.04 pmol of dihydroouabain-equivalent Dh-OLF per 106 cells (n = 3 separate experiments;Table 2). We used previously described purification techniques (4) to isolate and purify OLF and Dh-OLF simultaneously from the medium of human adrenal cell cultures. In the isocratic HPLC system described above, ouabain coeluted with OLF (29) and Dh-OLF coeluted with one of the newly discovered isomers of plant-derived dihydroouabain (dho-B; Fig. 3) (25). Under these conditions (100 mL/L CH3CN), OLF eluted at 29 min, whereas Dh-OLF eluted at 27.5 min, and each was detected by use of their respective EIAs. Consistent with our previous findings for human plasma and bovine adrenal cortex (4), we did not detect any immunoreactivity with the dihydroouabain EIA at the elution position (23.5 min; Fig. 4A) corresponding to the other recently discovered isomer of the plant-derived dihydroouabain (dho-A; Fig. 3) (25). Figure 3. Open in new tabDownload slide HPLC elution profile of cardiac glycosides. HPLC elution profile (100 mL/L CH3CN; isocratic mode) of micromolar concentrations of the plant-derived cardenolides dihydroouabain [2.5 μmol/L dho-A and 3.0 μmol/L dho-B, estimated by molar absorptivity at 196 nm, as detailed previously (20)] and ouabain (5.0 μmol/L, estimated by weight). Absorbance was monitored at 196 nm, which is the absorbance maximum of dihydroouabain (dho-A and dho-B). Figure 3. Open in new tabDownload slide HPLC elution profile of cardiac glycosides. HPLC elution profile (100 mL/L CH3CN; isocratic mode) of micromolar concentrations of the plant-derived cardenolides dihydroouabain [2.5 μmol/L dho-A and 3.0 μmol/L dho-B, estimated by molar absorptivity at 196 nm, as detailed previously (20)] and ouabain (5.0 μmol/L, estimated by weight). Absorbance was monitored at 196 nm, which is the absorbance maximum of dihydroouabain (dho-A and dho-B). Figure 4. Open in new tabDownload slide Separation of OLF and Dh-OLF by HPLC measured by EIA and corresponding biological activity. (A), HPLC elution profile of mammalian cardenolides and detection by EIA. OLF (eluting at 29 min) and Dh-OLF (eluting at 27.5 min) were purified by use of isocratic 100 mL/L CH3CN in water. OLF (486 pg of ouabain equivalents) and Dh-OLF (981 pg of dihydroouabain equivalents) were detected by the ouabain and dihydroouabain EIAs, respectively. (B), inhibition of the sodium pump catalytic activity by OLF and Dh-OLF. OLF (0.03 μmol/L) and Dh-OLF (0.07 μmol/L), which eluted at 29 and 27.5 min, respectively, in the above gradient and were identified by immunoreactivity, inhibited 29% and 17% of ouabain-sensitive sodium pump catalytic activity, respectively. Neighboring fractions (10–40 min) were processed similarly and assayed for immunoreactivity with the two antibodies and for inhibition of the catalytic activity of the sodium pump. No other immunoreactive fractions were detected, and no fractions inhibited the sodium pump catalytic activity. Figure 4. Open in new tabDownload slide Separation of OLF and Dh-OLF by HPLC measured by EIA and corresponding biological activity. (A), HPLC elution profile of mammalian cardenolides and detection by EIA. OLF (eluting at 29 min) and Dh-OLF (eluting at 27.5 min) were purified by use of isocratic 100 mL/L CH3CN in water. OLF (486 pg of ouabain equivalents) and Dh-OLF (981 pg of dihydroouabain equivalents) were detected by the ouabain and dihydroouabain EIAs, respectively. (B), inhibition of the sodium pump catalytic activity by OLF and Dh-OLF. OLF (0.03 μmol/L) and Dh-OLF (0.07 μmol/L), which eluted at 29 and 27.5 min, respectively, in the above gradient and were identified by immunoreactivity, inhibited 29% and 17% of ouabain-sensitive sodium pump catalytic activity, respectively. Neighboring fractions (10–40 min) were processed similarly and assayed for immunoreactivity with the two antibodies and for inhibition of the catalytic activity of the sodium pump. No other immunoreactive fractions were detected, and no fractions inhibited the sodium pump catalytic activity. Table 2. Production of OLF and Dh-OLF by H295R-1 cells.1 Mammalian factor . Concentration in medium, pmol/106 cells . . Concentration in cell pellet, pmol/106 cells . . . 12 h . 24 h . 12 h . 24 h . OLF <0.0423 0.18 ± 0.034 <0.0423 <0.0434 Dh-OLF <0.0323 0.39 ± 0.044 <0.0423 <0.0334 Mammalian factor . Concentration in medium, pmol/106 cells . . Concentration in cell pellet, pmol/106 cells . . . 12 h . 24 h . 12 h . 24 h . OLF <0.0423 0.18 ± 0.034 <0.0423 <0.0434 Dh-OLF <0.0323 0.39 ± 0.044 <0.0423 <0.0334 1 OLF and Dh-OLF amounts were measured by the OLF and Dh-OLF EIAs, respectively. Unity molar immunoreactivity of OLF and Dh-OLF with ouabain and dihydroouabain antibodies, respectively, was assumed as described in our previous work (4) to calculate moles of hormone-like factor secreted into the medium and retained in the cell pellet. For each compound, the amount measured was after 24 h of cell growth in serum-free medium and in the absence of (Bu)2cAMP. 2 Each measurement was performed in duplicate, and the average of two experiments was calculated. 3 The detection limits per 106 cells were calculated from the detection limits of the respective OLF and Dh-OLF EIAs. 4 Each measurement was performed in duplicate, and the average of three experiments was calculated. Open in new tab Table 2. Production of OLF and Dh-OLF by H295R-1 cells.1 Mammalian factor . Concentration in medium, pmol/106 cells . . Concentration in cell pellet, pmol/106 cells . . . 12 h . 24 h . 12 h . 24 h . OLF <0.0423 0.18 ± 0.034 <0.0423 <0.0434 Dh-OLF <0.0323 0.39 ± 0.044 <0.0423 <0.0334 Mammalian factor . Concentration in medium, pmol/106 cells . . Concentration in cell pellet, pmol/106 cells . . . 12 h . 24 h . 12 h . 24 h . OLF <0.0423 0.18 ± 0.034 <0.0423 <0.0434 Dh-OLF <0.0323 0.39 ± 0.044 <0.0423 <0.0334 1 OLF and Dh-OLF amounts were measured by the OLF and Dh-OLF EIAs, respectively. Unity molar immunoreactivity of OLF and Dh-OLF with ouabain and dihydroouabain antibodies, respectively, was assumed as described in our previous work (4) to calculate moles of hormone-like factor secreted into the medium and retained in the cell pellet. For each compound, the amount measured was after 24 h of cell growth in serum-free medium and in the absence of (Bu)2cAMP. 2 Each measurement was performed in duplicate, and the average of two experiments was calculated. 3 The detection limits per 106 cells were calculated from the detection limits of the respective OLF and Dh-OLF EIAs. 4 Each measurement was performed in duplicate, and the average of three experiments was calculated. Open in new tab Inhibition of Na+,K+-ATPase by Dh-OLF and OLF Both OLF and Dh-OLF isolated from H295R-1 cells inhibited the catalytic activity of porcine cerebrocortical Na+, K+-ATPase in vitro. Conditioned medium from five culture flasks, each containing ∼106 cells, was pooled and processed as described to extract OLF and Dh-OLF. Purified OLF (fraction eluting at 29 min) and Dh-OLF (fraction eluting at 27.5 min) were each resuspended in ddH2O to measure ouabain-sensitive sodium pump inhibition or in plasma to measure immunoreactivity. Neighboring fractions (10–40 min) were treated similarly, and both immunoreactivity and sodium pump inhibition data demonstrated that only the fractions eluting at 29 and 27.5 min had measurable immunoreactivities to ouabain or dihydroouabain antisera (Fig. 4A) in addition to inhibiting sodium pump catalytic activity (Fig. 4B). Specifically, 29% of the ouabain-sensitive catalytic activity of the sodium pump was inhibited at 0.03 μmol/L OLF and 17% was inhibited at 0.07 μmol/L Dh-OLF (Fig. 4B) stimulation of OLF and Dh-OLF secretion by (Bu)2cAMP Cells were grown for a total of 48 h, of which growth in serum-free medium supplemented with (Bu)2cAMP was for 24 h. At the end of the incubation, we collected the conditioned medium and extracted the compounds of interest. Stimulation of steroidogenesis by (Bu)2cAMP, as expected, was demonstrated by an increase in progesterone secretion into the medium (Fig. 5A). Progesterone, as well as other common steroids, is readily separable from the OLF or Dh-OLF in our HPLC isolation procedures (unpublished data). We observed that 24 h after (Bu)2cAMP treatment, the secretion of Dh-OLF into the medium was increased by 50% compared with unstimulated cells (Fig. 5B). Although there seemed to be a slight increase in OLF secretion in (Bu)2cAMP-stimulated cells, this increase was not statistically significant (Fig. 5B). We also observed that progesterone, OLF, and Dh-OLF (Table 2) were found mainly in the conditioned medium, i.e., when collected and processed, the cell pellets did not contain measurable amounts of these compounds. Additionally, in a parallel experiment we collected and processed the conditioned medium and cell pellet after 12 h without stimulation (cells of both experiments originated from the same passage) and were unable to measure detectable quantities of OLF and Dh-OLF (Table 2). This may be attributable to limitations in the sensitivities of the assays and, therefore, does not exclude possible secretion of OLF and Dh-OLF after 12 h. Finally, 100 mL of unconditioned DMEM/F12 medium containing 25 mL/L NuSerum type 1 was processed and injected into the HPLC as above to measure OLF and Dh-OLF. No measurable quantities of mammalian cardenolides were detected in either the cell pellet equivalent portion or the NuSerum-containing unconditioned medium. Figure 5. Open in new tabDownload slide Secretion of progesterone, OLF, and Dh-OLF by human adrenocortical cells (H295R-1) after stimulation by (Bu)2cAMP. (A), (Bu)2cAMP-stimulated progesterone secretion. Progesterone secretion into the medium was measured after 24 h of growth in serum-free medium with (stimulated) and without (unstimulated) added (Bu)2cAMP. Two samples of medium were withdrawn (50 and 25 μL) and analyzed by RIA. Duplicate measurements were taken, and the mean was calculated from both samples. (B), (Bu)2cAMP-stimulated Dh-OLF secretion into the medium. OLF and Dh-OLF secreted into the medium were measured by the ouabain and dihydroouabain EIAs, respectively, after cell growth in serum-free medium with and without the addition of (Bu)2cAMP. Duplicate measurements were taken, and the mean of three experiments (n = 3) was calculated. The SE is represented by error bars. Figure 5. Open in new tabDownload slide Secretion of progesterone, OLF, and Dh-OLF by human adrenocortical cells (H295R-1) after stimulation by (Bu)2cAMP. (A), (Bu)2cAMP-stimulated progesterone secretion. Progesterone secretion into the medium was measured after 24 h of growth in serum-free medium with (stimulated) and without (unstimulated) added (Bu)2cAMP. Two samples of medium were withdrawn (50 and 25 μL) and analyzed by RIA. Duplicate measurements were taken, and the mean was calculated from both samples. (B), (Bu)2cAMP-stimulated Dh-OLF secretion into the medium. OLF and Dh-OLF secreted into the medium were measured by the ouabain and dihydroouabain EIAs, respectively, after cell growth in serum-free medium with and without the addition of (Bu)2cAMP. Duplicate measurements were taken, and the mean of three experiments (n = 3) was calculated. The SE is represented by error bars. Discussion The objective of this study was to investigate the production of OLF and Dh-OLF by the human adrenal cell line H295R-1 and to develop immunoassays sufficiently specific to distinguish between OLF and Dh-OLF. We succeeded in developing two competitive EIAs and used them in combination with chromatographic separation and inhibition of Na+,K+-ATPase to demonstrate the ability of a human adrenocortical cell line to produce both of these compounds de novo. These cells secrete substantially more Dh-OLF than OLF, with the former being cAMP-dependent. Using plant-derived ouabain and dihydroouabain as immunogens, we developed antisera to detect OLF and Dh-OLF and confirmed the presence of OLF and Dh-OLF by use of EIAs in combination with several chromatographic separation steps. The antisera were sensitive to the only structural difference between the two compounds, i.e., an oxidized or reduced lactone ring. The selectivity was accomplished using the approach originally reported by Butler et al. (28) to couple cardenolide compounds to KLH/BSA through their rhamnose sugar moieties (Fig. 1). This technique maximally exposes the lactone ring portion of the cardenolides, thus giving the structural specificity needed to distinguish between the oxidized (OLF) and reduced (Dh-OLF) lactone rings (Fig. 6). The antisera directed against the saturated lactone ring (dihydroouabain) seems more specific to the structural difference on the lactone ring than antisera raised against the oxidized ring (ouabain), e.g., the cross-reactivity of antibodies to dihydroouabain was <2% with ouabain, whereas the cross-reactivity of antibodies to ouabain was 18% with dihydroouabain (Table 1). Although the antigenic response was designed to maximize specificity to the lactone rings as epitopes, some selectivity for the sugar portion was evident, as reported by others (34) for antibodies developed against ouabain and digoxin. Figure 6. Open in new tabDownload slide Lactone ring-dependent specificity of antibodies to ouabain and dihydroouabain. Ouabain-specific antibodies had 100% cross-reactivity with ouabain, but only 18% with the structurally almost identical dihydroouabain (left). Dihydroouabain-specific antibodies reacted 100% to dihydroouabain but reacted with only 1% of the ouabain exposed to the dihydroouabain antiserum. At the top are the structures of the oxidized (left) and reduced (right) lactone rings of ouabain and dihydroouabain, respectively. Figure 6. Open in new tabDownload slide Lactone ring-dependent specificity of antibodies to ouabain and dihydroouabain. Ouabain-specific antibodies had 100% cross-reactivity with ouabain, but only 18% with the structurally almost identical dihydroouabain (left). Dihydroouabain-specific antibodies reacted 100% to dihydroouabain but reacted with only 1% of the ouabain exposed to the dihydroouabain antiserum. At the top are the structures of the oxidized (left) and reduced (right) lactone rings of ouabain and dihydroouabain, respectively. Because we separated the oxidized and reduced species chromatographically before analysis by immunoassay or by inhibition of Na+,K+-ATPase, the issue of antibody cross-reactivity becomes less critical in our study. However, this is an important consideration in development of assays that would measure these compounds directly in a nonpurified medium such as serum or other biological fluids. The concentration working ranges and analytical sensitivities of both EIAs (Fig. 2) were adequate to measure the amounts of OLF and Dh-OLF secreted by the cell cultures after processing and purification as in our experiments. However, for applications involving direct measurements of these compounds from serum or other biological fluids, further refinement of the two assays will be necessary to obtain clinically relevant reproducibility and specificity. In addition, the lower limits of detection for the two EIAs (72 pmol/L for OLF and 60 pmol/L for Dh-OLF) seem adequate for direct measurements of these compounds in serum based on the wide OLF concentration range (10–800 pmol/L) described by Vakkuri et al. (34) and on the OLF (380 ± 42 pmol/L) and Dh-OLF (5000 ± 460 pmol/L) concentrations in human serum we reported in the blood of mammals (4). It is important to note that in this study we calibrated the two EIAs using the plant-derived compounds and then used immunoequivalence to assess the concentrations of both OLF and Dh-OLF. Previous work, based on independent estimates of molar absorptivity by spectrophotometry, suggests there is a one-to-one correspondence in the reactivity of antibodies raised to ouabain or dihydroouabain toward OLF or Dh-OLF, respectively (4). We therefore believe that measurements in this present study reasonably approximate the amounts of these compounds secreted by the cell cultures. In addition to the selectivity of the antibodies developed, we were able to authenticate the identity of OLF and Dh-OLF from H295R-1 cells by showing that after multiple separation steps, both endogenous compounds have HPLC elution profiles (Fig. 4A) identical to those of ouabain and dihydroouabain, respectively (Fig. 3). We raised the dihydroouabain antiserum against the mixture of both dihydroouabain components, dho-A and dho-B (see Fig. 1). It is interesting to note that in secretions from our human cell cultures, the dihydroouabain antiserum was able to detect only a Dh-OLF component corresponding to the plant-derived dho-B, and not dho-A (Fig. 4A). This is consistent with our previous findings; we were unable to detect any immunoreactivity in the position at which dho-A secreted by bovine adrenal cells elutes in a mobile phase of 100 mL/L CH3CN in water (23.5 min; see Fig. 3) (4). Possible explanations include that the detection limit of the assay was too high to detect a Dh-OLF compound at the dho-A elution position or that humans simply do not secrete the dho-A form of the Dh-OLF compound. Both OLF and Dh-OLF isolated from H295R-1 cells inhibited the catalytic activity of porcine cerebrocortical Na+,K+-ATPase in vitro (Fig. 4B). In a previous study using OLF and Dh-OLF extracted from bovine adrenal cortex and Na+,K+-ATPase preparations from porcine cerebral cortex, we used inhibition curves over a wide range of concentrations to show that OLF is more potent than Dh-OLF in inhibiting the catalytic activity of the sodium pump (25). These relative potencies for inhibition of Na+,K+-ATPase reflect the relative inhibitory potencies of the plant-derived counterparts;ouabain (0.66 ± 0.095 μmol/L) has a lower IC50 than dihydroouabain-B (1.63 ± 0.12 μmol/L) (25). Although we did not isolate enough material in the present study to obtain complete inhibition curves, our data do show a similar pattern of inhibitory potency between OLF and Dh-OLF. For example, chromatographic isolates of OLF and Dh-OLF inhibited the catalytic activity of the sodium pump (29% inhibition by 0.03 μmol/L ouabain immunoequivalents for OLF and 17% inhibition by 0.07 μmol/L dihydroouabain immunoequivalents for Dh-OLF; Fig. 4B). These results are within 10% of the potencies we reported previously for the inhibitory activities of OLF and Dh-OLF isolated from bovine adrenal glands (4). Human adrenal H295R-1 cells grown in serum-free medium produce both OLF and Dh-OLF (Table 2). Compared with the culture incubation medium, the cell pellets had undetectable amounts of either OLF or Dh-OLF. In addition, 100 mL of unconditioned NuSerum medium processed as described earlier contained no detectable OLF and Dh-OLF immunoreactivity in our EIAs for OLF and Dh-OLF, respectively. In a previous study by Doris et al. (10), the authors showed that ouabain can accumulate in adrenocortical cells and then be released into the cell culture medium. To ensure that we are measuring OLF and Dh-OLF produced by the human adrenal cells and not “recycled” mammalian cardenolides absorbed during incubation, we demonstrated that OLF and Dh-OLF were not detectable in the cell pellets before their secretion into the medium. We are therefore confident that our measured OLF and Dh-OLF are indeed produced de novo. In a recent study we demonstrated the isolation and purification of a dihydroouabain-like compound (Dh-OLF) from bovine adrenal tissues and human serum (4). Subsequently, we found that Y-1 murine adrenal cells in culture produced 43 pg (0.074 pmol) of OLF and 300 pg (0.5 pmol) of Dh-OLF/106 cells (24). This present study now indicates that human adrenal cells produce comparable amounts of these compounds de novo: 103 pg (0.18 ± 0.03 pmol; n = 3) of OLF and 230 pg (0.39 ±0.04 pmol; n = 3) of Dh-OLF/106 cells. Although more detailed studies are needed to define the relative amounts of the OLFs secreted between species, it is notable that Dh-OLF seems to be consistently present at higher amounts than OLF as is dihydrodigoxin-like immunoreactive factor compared with digoxin-like immunoreactive factor (4)(24). It is interesting to note that a recent study from our laboratory demonstrated the incorporation of radiolabeled carbon into the structure of OLF and Dh-OLF by adrenal cells in culture (35). This supports the concept of de novo synthesis of OLF and Dh-OLF by adrenal cells. The discovery of Dh-OLF is recent (24), and little information is available on its biosynthesis. The present study indicates that cAMP stimulates the secretion of Dh-OLF but not OLF in human adrenal cells. Cells grown for 24 h in serum-free medium supplemented with (Bu)2cAMP secreted proportionately more Dh-OLF than OLF. Cells in which steroidogenesis was stimulated by (Bu)2cAMP secreted more progesterone into the medium, as expected (Fig. 5A). Additionally, to demonstrate the viability of the adrenocortical cells, we measured steroid production, i.e., progesterone (Fig. 5A) and aldosterone [data not shown and Refs. (31)(36)]. To date, those working to characterize the secretion of the mammalian cardenolides by cell cultures have used the oxidized compounds, e.g., OLF or the digitalis-like factors, as the secretory products. If the structural difference between OLF and Dh-OLF is limited to the chemical state of the lactone ring, as is currently suspected, both compounds are likely to have similar biosynthetic origins for the better part of their structures. In most studies, the biosynthesis of these kinds of endogenous factors has been investigated by stimulating steroid synthesis in cultured adrenal cells and observing changes in the secretion of the endogenous glycosides/cardenolides. Several stimuli that increase steroidogenesis have been shown to stimulate the secretion of OLF in mammalian adrenocortical tissues or cultures (9)(37). In 1998, Hinson et al. (3) showed increased secretion of a ouabain-like compound (OLC) by perfused rat adrenal glands after stimulation of steroidogenesis by adrenocorticotropic hormone (ACTH). In another study, stimulation of OLC synthesis occurred 10 min after administration of ACTH and indicated that OLC is acutely regulated in rat adrenals (6). Previous work has shown that ACTH and angiotensin II stimulate secretion of OLF as well as aldosterone and cortisol from bovine adrenals (38)(39). Similarly, other work has shown that the use of both progesterone and pregnenolone as precursors in the biosynthetic pathway noticeably increased the secretion of OLF in adrenal cells, but that cortisol as a precursor only slightly increased OLF secretion into the medium (40). Additional evidence on the origin and biosynthesis of endogenous ouabain in adrenals has been provided by studies of Hamlyn et al. (29) that show secretion of endogenous ouabain from adrenal glands after the medium was supplemented with pregnenolone. Their results also suggested that although pregnenolone seems to be a substrate for both endogenous ouabain and aldosterone biosynthesis, a divergence in the biosynthetic pathway of these steroids seems to occur before progesterone synthesis. However, Perrin et al. (40) have reported that progesterone also increased the biosynthesis of OLC and that pregnenolone, via progesterone, may represent a precursor in the biosynthesis of OLC in adrenal glands. Their argument was supported by results showing that both pregnenolone and progesterone increase OLC secretion in primary cultures of bovine adrenal cells. The possibility of a second-messenger mechanism being involved in regulating secretion of these compounds has also been considered. A study by Shah et al. (9) seemed to indicate that neither cAMP or protein kinase C is involved in the signaling pathway of endogenous ouabain synthesis. Therefore, although it has been shown that ACTH stimulates endogenous ouabain secretion, preliminary evidence suggests that cAMP does not seem to serve as the cellular mediator. This has led to the conclusion that the mechanisms regulating secretion of aldosterone and endogenous ouabain are distinct. Hamlyn et al. (29) showed that (Bu)2cAMP increased the production of aldosterone in adrenal cells while not significantly increasing secretion and production of endogenous ouabain. In the present study, we investigated the role of the second messenger cAMP in stimulating Dh-OLF synthesis in human adrenal cells by supplementing cell culture medium with (Bu)2cAMP, a membrane-penetrable form of cAMP. We were able to stimulate steroidogenesis by use of (Bu)2cAMP, measuring an increase in progesterone secretion. We also showed that the concentration of OLF was unaltered, consistent with the observation by Shah et al. (9). Importantly, the secretion of Dh-OLF increased by 50% after stimulation with (Bu)2cAMP, providing preliminary evidence that the secretion of this newly discovered chemically reduced form of OLF may be controlled by cAMP-mediated stimulation. This evidence also points to a possible product–precursor relationship between OLF and Dh-OLF in human cells. In conclusion, this report suggests that OLF and Dh-OLF are synthesized de novo by a human adrenal cell culture, that immunoassays sufficiently specific to distinguish between OLF and Dh-OLF can be developed, that mammalian cells produce only one isomer of Dh-OLF in lieu of the two plant-derived isomers of dihydroouabain, and that the secretion of Dh-OLF is more sensitive to cAMP stimulation than is OLF. Our findings suggest the possibility of a product–precursor relationship between these two compounds, and this warrants additional investigation. The availability of a human cell line that secretes both OLF and Dh-OLF now provides a working model for characterizing the biosynthesis and secretory regulation of these two compounds in humans. 1 " Current address: Department of Psychiatry, School of Medicine, University of Louisville, Louisville, KY 40202. 2 " Nonstandard abbreviations: OLF, ouabain-like factor; Dh-OLF, dihydroouabain-like factor; EIA, enzyme immunoassay; BSA, bovine serum albumin; PBS, phosphate-buffered saline; TMB, 3,3′,5,5′-tetramethylbenzidine;KLH, keyhole limpet hemocyanin; (Bu)2cAMP, dibutryl cAMP; ddH2O, distilled, deionized water; OLC, ouabain-like compound; and ACTH, adrenocorticotropic hormone. We are deeply grateful to Rebecca Combs for invaluable technical assistance. This work was supported in part by NIH Grants HL R01-36172 and HL R01-59404 (to R.V.) and National Science Foundation Grant OSR-9452895 (to R.V.). References 1 Hamlyn JM, Blaustein MP, Bova S, DuCharme DW, Harris DW, Mandel F, et al. Identification and characterization of a ouabain-like compound from human plasma. Proc Natl Acad Sci U S A 1991 ; 88 : 6259 -6263. Crossref Search ADS PubMed 2 De Angelis C, Riscazzi M, Salvini R, Piccoli A, Ferri C, Santucci A. Isolation and characterization of a digoxin-like immunoreactive substance from human urine by affinity chromatography. Clin Chem 1997 ; 43 : 1416 -1420. Crossref Search ADS PubMed 3 Hinson JP, Harwood S, Dawnay AB. Release of ouabain-like compound (OLC) from the intact perfused rat adrenal gland. Endocr Res 1998 ; 24 : 721 -724. Crossref Search ADS PubMed 4 Qazzaz HM, El-Masri MA, Valdes R, Jr. Secretion of a lactone-hydrogenated ouabain-like effector of sodium, potassium-adenosine triphosphatase activity by adrenal cells. Endocrinology 2000 ; 141 : 3200 -3209. Crossref Search ADS PubMed 5 Ludens JH, Clark MA, Robinson FG, DuCharme DW. Rat adrenal cortex is a source of a circulating ouabainlike compound. Hypertension 1992 ; 19 : 721 -724. Crossref Search ADS PubMed 6 Beck M, Szalay KS, Nagy GM, Toth M, de Chate R. Production of ouabain by rat adrenocortical cells. Endocr Res 1996 ; 22 : 845 -849. Crossref Search ADS PubMed 7 Li S, Eim C, Kirch U, Lang RE, Schoner W. Bovine adrenals and hypothalamus are a major source of proscillaridin A- and ouabain-immunoreactivities. Life Sci 1998 ; 62 : 1023 -1033. Crossref Search ADS PubMed 8 Tymiak A, Norman J, Bolgar M, DiDonato G, Lee H, Parker W, et al. Physiochemical characterization of a ouabain isomer isolated from bovine hypothalamus. Proc Natl Acad Sci U S A 1993 ; 90 : 8189 -8193. Crossref Search ADS PubMed 9 Shah JR, Laredo J, Hamilton BP, Hamlyn JM. Different signaling pathways mediate stimulated secretion of endogenous ouabain and aldosterone from bovine adrenocortical cells. Hypertension 1998 ; 31 : 463 -468. Crossref Search ADS PubMed 10 Doris PA, Hayward-Lester A, Bourne D, Stocco DM. Ouabain production by cultured adrenal cells. Endocrinology 1996 ; 137 : 533 -539. Crossref Search ADS PubMed 11 Laredo J, Hamilton BP, Hamlyn JM. Ouabain is secreted bovine adrenocortical cells. Endocrinology 1994 ; 135 : 794 -797. Crossref Search ADS PubMed 12 Doris PA. Regulation of Na,K-ATPase by endogenous ouabain-like materials. Exp Biol Med 1994 ; 205 : 202 -212. Crossref Search ADS 13 Doris PA. Endogenous inhibitors of the Na,K pump. Miner Electrolyte Metab 1996 ; 22 : 303 -310. PubMed 14 Hollenberg NK, Graves SW. Endogenous sodium pump inhibition: current status therapeutic opportunities. Prog Drug Res 1996 ; 46 : 9 -42. PubMed 15 Rose AM, Valdes R, Jr. Understanding the sodium pump and its relevance to disease. Clin Chem 1994 ; 40 : 1674 -1685. Crossref Search ADS PubMed 16 Li R, El-Mallakh RS, Harrison L, Changaris DG, Levy RS. Lithium prevents ouabain-induced behavioral changes. Mol Chem Neuropathol 1997 ; 31 : 65 -72. Crossref Search ADS PubMed 17 Ferrandi M, Manunta P. Ouabain-like factor: is this the natriuretic hormone?. Curr Opin Nephrol Hypertens 2000 ; 9 : 165 -171. Crossref Search ADS PubMed 18 Ludnes JH, Clark MA, DuCharme DW, Harris DW, Lutzke BS, Mandel F, et al. Purification of an endogenous digitalislike factor from human plasma for structural analysis. Hypertension 1991 ; 17 : 923 -929. Crossref Search ADS PubMed 19 Goto A, Yamada K. Ouabain-like factor. Curr Opin Nephrol Hypertens 1998 ; 7 : 189 -196. Crossref Search ADS PubMed 20 Bagrov AY, Fedorova OV, Austin-Lane JL, Dmitrieva RI, Anderson DE. Endogenous marinobufagenin-like immunoreactive factor and Na,K-ATPase inhibition during voluntary hypoventilation. Hypertension 1995 ; 26 : 781 -788. Crossref Search ADS PubMed 21 Dmitrieva RI, Bagrov AY, Lalli E, Sassone-Corsi P, Stocco DM, Doris PA. Mammalian bufadienolide is synthesized from cholesterol in the adrenal cortex by a pathway that is independent of cholesterol side-chain cleavage. Hypertension 2000 ; 36 : 442 -448. Crossref Search ADS PubMed 22 Schoner W, Kirch U, Sich B, Antonolovich R. Purification and some properties of an inhibitor of the sodium pump from bovine adrenals cross-reacting with proscillaridin A antibodies. Biol Chem Hoppe Seyler 1993 ; 374 : SP-181 . 23 Mathews WR, DuCharme DW, Hamlyn JM, Harris DW, Mandel F, Clark MA, et al. Mass spectral characterization of an endogenous digitalislike factor from human plasma. Hypertension 1991 ; 17 : 930 -935. Crossref Search ADS PubMed 24 El-Masri MA, Qazzaz HM, Clark BJ, Valdes R, Jr. Production of ouabain-like factor (OLF) and it’s chemically reduced form dihydroouabain-like factor (dihydro-OLF) by human adrenal cells [Abstract]. FASEB J 2000 ; 14 : A1536 . 25 Qazzaz MA, El-Masri MA, Stolowich NJ, Valdes R, Jr. Two biologically active isomers of dihydroouabain isolated from a commercial preparation. Biochim Biophys Acta 1999 ; 1472 : 486 -497. Crossref Search ADS PubMed 26 Flash H, Heinz N. Correlation between inhibition of (Na+,K+)-membrane-ATPase and positive inotropic activity of cardenolides in isolated papillary muscles of guinea pig. Naunyn Schmiedebergs Arch Pharmacol 1978 ; 304 : 37 -44. Crossref Search ADS PubMed 27 Feng J, Lingrel JB. Analysis of amino acid residues in the H5–H6 transmembrane and extracellular domains of the Na,K-ATPase α subunit identifies threonine 797 as a determinant of ouabain sensitivity. Biochemistry 1994 ; 33 : 4218 -4224. Crossref Search ADS PubMed 28 Butler VP, Jr, Tse-Eng D, Lindenbaum J, Kalman SM, Preibisz JJ, Rund DG, et al. The development and application of a radioimmunoassay for dihydrodigoxin, a digoxin metabolite. J Pharmacol Exp Ther 1982 ; 221 : 123 -131. PubMed 29 Hamlyn JM, Lu Z, Manunta P, Ludens JH, Kimura K, Shah JR, et al. Observation on the nature, biosynthesis, secretion, and significance of endogenous ouabain. Clin Exp Hypertens 1998 ; 20 : 523 -533. Crossref Search ADS PubMed 30 Fernandez-Botran F Vetvicka V eds. Methods in cellular immunology 1995 : 203 -204 CRC Press Boca Raton, FL. . 31 Clark BJ, Pezzi V, Stocco DM, Rainey WE. The steroidogenic acute regulatory protein is induced by angiotensin II and K+ in H295R human adrenocortical cells. Mol Cell Endocrinol 1995 ; 115 : 215 -219. Crossref Search ADS PubMed 32 Resko JA, Norman GD, Niswender GD, Spies HG. The relationship between progestins and gonadotropins during the late luteal phase of the menstrual cycle in rhesus monkeys. Endocrinology 1974 ; 94 : 128 -135. Crossref Search ADS PubMed 33 Chan ELP, Swaminathan R. A rapid assay for the measurement of Na+, K+-ATPase inhibitors. Clin Biochem 1992 ; 25 : 15 -19. Crossref Search ADS PubMed 34 Vakkuri OS, Arnason SS, Pouta A, Vuolteenaho O. Radioimmunoassay of plasma ouabain in healthy and pregnant individuals. J Endocrinol 2000 ; 165 : 669 -677. Crossref Search ADS PubMed 35 Qazzaz HM, Clements JM, Clark BJ, Valdes R Jr. Production of 14C-labelled mammalian ouabain-like factors (OLF) using mouse adrenal Y-1. J Clin Ligand Assay 2002;in press.. 36 Clark BJ, Combs R. Angiotensin II and cyclic adenosine 3′,5′-monophosphate induce human steroidogenic acute regulatory protein transcription through a common steroidogenic factor-1 element. Endocrinology 1999 ; 140 : 4390 -4398. Crossref Search ADS PubMed 37 Shah JR, Laredo J, Hamilton BP, Hamlyn JM. Effects of angiotensin II on sodium potassium pumps, endogenous ouabain and aldosterone in bovine zona glomerulosa cells. Hypertension 1999 ; 33 : 373 -377. Crossref Search ADS PubMed 38 Pena C, Arnaiz de Lores GR. Differential properties between an endogenous brain Na,K-ATPase inhibitor and ouabain. Neurochem Res 1997 ; 22 : 379 -383. Crossref Search ADS PubMed 39 Li M, Wong KS, Martin A, Withworth A. Adrenocorticotrophin-induced hypertension in rats. Am J Hypertens 1994 ; 7 : 59 -68. Crossref Search ADS PubMed 40 Perrin A, Brasmes B, Chambaz EM, Defaye G. Bovine adrenocortical cells in culture synthesize an ouabain-like compound. Mol Cell Endocrinol 1997 ; 126 : 7 -15. Crossref Search ADS PubMed © 2002 The American Association for Clinical Chemistry This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Human Adrenal Cells in Culture Produce Both Ouabain-like and Dihydroouabain-like Factors
JF - Clinical Chemistry
DO - 10.1093/clinchem/48.10.1720
DA - 2002-10-01
UR - https://www.deepdyve.com/lp/oxford-university-press/human-adrenal-cells-in-culture-produce-both-ouabain-like-and-oZVdcqhdhq
SP - 1720
VL - 48
IS - 10
DP - DeepDyve
ER -