Development of an Adrenocorticotropin-Responsive Human Adrenocortical Carcinoma Cell Line

Abstract

Context: The molecular mechanisms regulating adrenal steroidogenesis continue to be defined. The only current human adrenocortical cell line is the NCI-H295 and its substrains. One of the strains, H295R, has retained the ability to respond to angiotensin II (Ang II); however, it lacks ACTH responsiveness. An ACTH-responsive human adrenocortical model would add significantly to studies directed at defining the molecular control of corticosteroid biosynthesis.

Objective: The objective of the study was to develop a human adrenal cell line that retained both Ang II- and ACTH-regulated corticosteroid production.

Design: Human adrenocortical carcinoma (HAC) cells were isolated from an adrenal tumor removed from a girl presenting with virilization and hypertension. Clonal populations of cells were established and characterized. HAC cells were treated with ACTH, Ang II, and forskolin, followed by examination of steroidogenic enzyme mRNA expression using quantitative real-time PCR and steroid production.

Results: HAC clone 15 (HAC15) cells responded to treatment with ACTH, Ang II, and forskolin, with increased cortisol and aldosterone production. ACTH, Ang II, and forskolin also increased expression of mRNA, encoding all enzymes needed for cortisol and aldosterone biosynthesis, namely steroidogenic acute regulatory protein, cholesterol side-chain cleavage, cytochrome P450 17α-hydroxylase-17, 20-lyase, 3β-hydroxysteroid dehydrogenase type II, 21-hydroxylase, 11β-hydroxylase, and 11β-aldosterone synthase. In addition, the cells expressed mRNA for ACTH receptor (MC2R) and Ang II receptor. MC2R protein was also expressed in HAC15 cells.

Conclusion: The current study describes the development and characterization of an ACTH- and Ang II-responsive human adrenal cell line. The HAC15 cell line should provide an important model system for defining the molecular mechanisms regulating aldosterone and cortisol production.

The development and characterization of an ACTH- and angiotensin II-responsive human adrenal cell line, human adrenocortical carcinoma clone 15 (HAC15), is described.

Defining the molecular mechanisms and cellular signaling pathways that regulate the production of adrenocortical steroids is often limited by the availability of appropriate model systems. One method of modeling the adrenal is through the use of cell lines that retain adrenocortical differentiated functions (1). The list of permanent cell lines established from human adrenal tumors is remarkably short, consisting of the NCI-H295 and its substrains (1,2). The NCI-H295, H295R, and H295A retain several differentiated functions, which make them useful for defining the cellular mechanisms regulating steroid production (1,2,3,4,5,6,7,8,9,10,11,12,13,14,15). However, the NCI-H295 cells have little or no ACTH response. Thus, there is currently no human adrenal cell model that responds to a sustained ACTH stimulation.

In this report, we describe the establishment of a permanent human adrenocortical carcinoma (HAC) cell line. These cells continue to secrete glucocorticoids and mineralocorticoids under the control of ACTH and angiotensin II (Ang II).

Materials and Methods

HAC cell culture establishment

Primary cultures of HAC cells were isolated after surgical removal of an adrenocortical carcinoma from an 11-month-old female with hypertension and hirsutism. The use of tumor tissue was approved by the Institutional Review Board of the Medical College of Georgia (Augusta, GA). Primary carcinoma cells were isolated after the tumor was minced into small pieces and incubated in Dulbecco’s Modified Eagle/F12 (Invitrogen, Carlsbad, CA) containing 0.1% collagenase (Roche, Indianapolis, IN). Digestion and mechanical dispersion were carried out for 1.5 h at 37 C, with a final digestion of the dispersed cells in medium containing 0.01% deoxyribonuclease I. After isolation, cells were frozen in DME/F12 medium, 50% Nu Serum (BD Biosciences, Franklin Lakes, NJ), and 10% dimethyl sulfoxide (Sigma, St. Louis, MO). The frozen aliquots were later suspended in growth media consisting of DME/F12 medium supplemented with 10% cosmic calf serum (HyClone, Logan, UT), antibiotics and 1% insulin/transferrin/selenium Premix (BD Biosciences) and plated at cloning density. After 3 wk, clones were isolated for characterization. The most ACTH-responsive steroidogenic clone, HAC clone 15 (HAC15), was used for this study. H295R cells, grown under similar conditions were used for comparison studies. For experiments, cells were plated onto 12 dishes (400,000 cells/well) in growth medium. Cells were treated in low serum medium (0.1% cosmic calf serum) with Ang II (10 nm; Sigma), forskolin (10 μm; Sigma), or ACTH (10 nm; Organon, Bedford, OH).

Analysis of steroids

Experimental media from HAC15 cells was analyzed for aldosterone and cortisol levels using specific immunoassays (Siemens Diagnostics, Tarrytown, NY, and Alpco Diagnostics, Salem, NH, respectively). Steroid data were normalized to milligrams total cell protein.

Real-time RT-PCR (qPCR)

Total cellular RNA was extracted using RNeasy minikits (QIAGEN, Valencia, CA). Two micrograms of total RNA were reverse transcribed using the high-capacity cDNA archive kit (Applied Biosystems, Foster City, CA). cDNA of the specific mRNA encoding enzymes and receptors were examined using probe sets obtained from TaqMan gene expression assays (Applied Biosystems). Quantitative normalization of cDNA for each gene was performed using the expression of 18s rRNA.

Protein immunoblot analysis

Cells were lysed and 40 μg of lysate was applied to polyacrylamide gel electrophoresis using 12% bis-Tris NuPage gels (Invitrogen). Membranes were incubated overnight at 4 C, with antibody against MC2R (ACTH receptor) (Chemicon International, Temecula, CA), at a 1:1000 dilution. Membranes were incubated with horseradish peroxidase-conjugated secondary antibodies (Santa Cruz, Santa Cruz, CA). Immunoreactive bands were visualized using the ECL plus Western detection system (GE Healthcare, Piscataway, NJ) and a GBOX image analyzer system (Syngene, Frederick, MD).

Statistical analysis

All experiments were repeated a minimum of three times, and within each experiment, variables were performed in triplicate. Statistical comparisons were analyzed using one-way ANOVA. Significance was accepted at the 0.05 level of probability.

Results

The effects of Ang II, ACTH, and forskolin on HAC15 cell production of aldosterone and cortisol were examined after treatment for 24 h. Treatment with ACTH and Ang II significantly increased HAC15 cortisol production by more than 2-fold (Fig. 1A). HAC15 cells also increased aldosterone production by 3-fold in response to ACTH and 8-fold in response to Ang II (Fig. 1A). Bypassing the ACTH receptor by treating with forskolin (10 μm) significantly increased both aldosterone production (2.5-fold) and cortisol production (3-fold) in HAC15 (Fig. 1A).

Figure 1.

A, Steroidogenic responsiveness of HAC15 cells to agonist treatment. Cell lines were treated for 24 h with Ang II (10 nm), ACTH (10 nm), or forskolin (10 μm). Cortisol and aldosterone concentrations were determined by enzyme immunoassay and RIA, respectively. Values represent the mean ± se of six independent experiments. ***, P < 0.001 for agonist-stimulated steroidogenesis vs. basal. B, HAC15 cell expression of MC2R and AT1R receptors. Protein immunoblot analysis was used to examine expression of MC2R in cell lysates of H295R and HAC15. Forty micrograms of protein were loaded per lane. Comparison of MC2R and AT1R mRNA levels in HAC15 to H295R was done after qPCR, with the H295R cells used as a calibrator. Values represent the mean ± se of six independent RNA samples. ***, P < 0.001.

Due to the ACTH response observed for HAC cells, we examined expression of MC2R mRNA and protein. Protein immunoblot analysis was used to examine MC2R in cell lysates of H295R and HAC15 (Fig. 1B). MC2R expression was evident in HAC15, resulting in an unambiguous band of approximately 38 kDa. In contrast, MC2R expression was undetectable in the H295R cells. MC2R and type 1 Ang II receptor (AT1R) mRNA levels were also compared between the HAC15 and H295R using qPCR. HAC15 mRNA for MC2R was elevated 9.25-fold compared with the H295R cell line. AT1R mRNA levels were similar in both cell lines (Fig. 1B).

To further assess adrenal differentiated function in HAC15 cells, the expression of mRNA encoding enzymes involved in the production of aldosterone and cortisol were examined (Fig. 2). Treatment with ACTH increased mRNA expression of all the enzymes needed for aldosterone and cortisol production, namely steroidogenic acute regulatory protein, cholesterol side-chain cleavage, 17α-hydroxylase-17, 20-lyase, 21-hydroxylase, 3β-hydroxysteroid dehydrogenase type II, 11β-hydroxylase, and 11β-aldosterone synthase (CYP11B2) mRNA compared with untreated cells (Fig. 2). Treatment with Ang II and forskolin also caused an increase in mRNA expression of all the enzymes needed for aldosterone and cortisol production.

Figure 2.

HAC15 cell expression and regulation of genes encoding enzymes involved in steroid hormone biosynthesis. HAC15 cells were treated with or without Ang II (10 nm), ACTH (10 nm), or forskolin (10 μm) for 24 h. mRNA level for each enzyme was determined using qPCR. Values represent the mean ± se of six independent experiments. Nomenclature for the steroidogenic enzymes are as follows: StAR, Steroidogenic acute regulatory protein; CYP11A1, cholesterol side-chain cleavage; CYP17, 17α-hydroxylase-17, 20-lyase; HSD3B2, 3β-hydroxysteroid dehydrogenase type II; CYP21, 21-hydroxylase; CYP11B1, 11β-hydroxylase. ***, P < 0.001; **, P < 0.01; *, P < 0.05.

Discussion

We previously developed a human adrenocortical cell model, H295R, which retains Ang II and forskolin responses (1). These cells have been widely used to study steroid hormone production; however, they are not a good model for ACTH-regulated steroidogenesis. Herein we isolated a clonal population of adrenocortical carcinoma cells that retained Ang II, ACTH, and forskolin responsiveness. These cells will provide an alternative to the H295 substrains and should act as a better model for ACTH response.

Adrenal cell lines have proven useful for defining basic mechanisms involved in adrenal function. There have been a number of rodent adrenal cell lines developed using either adrenal tumors or transgenic mice with adrenal-targeted oncogenes (1). The Y-1 mouse adrenal cell line is the most widely used for studying acute regulation of steroidogenic enzymes (1,16). However, the Y-1 cells have lost the expression of 21 hydroxylase and therefore cannot produce corticosteroids. Two other mouse adrenocortical cell lines, ATC1 and ATC7-L, were developed using genetically targeted oncogenesis (17). Both cell lines retain a zona fasciculata phenotype, which is manifested at multiple levels, including corticosterone secretion, ACTH responsiveness, and expression profiles of steroidogenic enzymes. Whereas each of these cell lines are useful for many adrenal research projects, mouse adrenal cell lines cannot replicate the ability of human adrenal cells to produce cortisol or adrenal androgens. Therefore, human adrenal cell lines are needed for many studies directed at understanding the human adrenal.

The NCI-H295 and its substrains are currently the only human adrenocortical cell models available (1,2,3,4,5,6,7,8,9,10,11,12,13,14,15). Two strains of H295 cells, designated H295R and H295A, have been selected to grow as a monolayer (1). These cells have been used to study steroidogenic enzyme expression and steroid hormone production. In addition, the cells have been used to examine the molecular mechanisms regulating transcription of CYP11A1, CYP11B1, CYP11B2, CYP17, CYP21, and HSD3B2 (7,8,9,10,11,12,13,14). The H295R cells are also responsive to Ang II and potassium through the production of aldosterone and the expression of CYP11B2 (4,7). Thus, these cells have been widely used to better understand the molecular mechanisms regulating human adrenal cell steroidogenesis.

The successful development of the original NCI-H295 cells from an adrenocortical carcinoma led us to use a similar approach to develop the HAC15 cells. Our findings indicate that the HAC15 cells have similar steroidogenic profiles compared with the H295R. These findings include the production of all the corticosteroids. Cortisol was produced at significantly higher levels than aldosterone. However, this may be the due to low expression of CYP11B2 (aldosterone synthase), which requires Ang II treatment to increase its expression. The HAC15 cells also consistently produce significant amounts of the adrenal androgen, dehydroepiandrostenedione (data not shown). Further studies will be needed to determine whether chronic hormonal treatment (Ang II or ACTH) can influence the HAC15 phenotype to be either glomerulosa, fasciculata or reticularis like.

Similar to the NCI-H295 and its substrains, the HAC15 cells expressed mRNA encoding all enzymes needed for production of aldosterone, cortisol, and dehydroepiandrostenedione. All of these transcripts were responsive to both Ang II and ACTH. HAC15 cells were also responsive to extracellular potassium through an increase in aldosterone production and CYP11B2 expression (data not shown). The major difference between the HAC15 and H295R appears to be their ACTH response.

MC2R is the well-documented receptor to which ACTH binds and regulates different cellular responses (18). Whereas the H295R cells have a number of limitations, one of the most obvious is the lack of a sustained ACTH response. The lack of responsiveness in H295R cells is not clearly understood. The cells have mRNA expression for MC2R, which is regulated by Ang II and cAMP signaling systems (19). In addition, the H295R cells respond acutely to ACTH with an activation of the ERK 1/2 phosphorylation (20). However, our direct comparison of H295R and HAC15 cells demonstrated significantly higher levels of MC2R mRNA and protein expression in the HAC15 cells. These data suggest that the differences between the HAC15 and H295R cells are, at least in part, due to differences in MC2R expression.

In summary, we have developed an adrenal cell line, HAC15, that exhibits hormonal responses, steroidogenesis, and expression of steroid-metabolizing enzymes. This cell line represents only the second human adrenocortical cell line available. The ability of HAC15 cells to respond to Ang II, potassium, and ACTH make them the first adrenal cell line capable of responding to the three main adrenocortical physiologic regulators.