Differential Expression of a Stress-modulating Gene,BRE,in the Adrenal Gland,in Adrenal Neoplasia,and in Abnormal Adrenal Tissues

Abstract

Genes that modulate the action of hormones and cytokines play a critical role in stress response, survival, and in growth and differentiation of cells. Many of these biological response modifiers are responsible for various pathological conditions, including inflammation, infection, cachexia, aging, genetic disorders, and cancer. We have previously identified a new gene, BRE, that is responsive to DNA damage and retinoic acid. Using multiple-tissue dot-blotting and Northern blotting, BRE was recently found to be strongly expressed in adrenal cortex and medulla, in testis, and in pancreas, whereas low expression was found in the thyroid, thymus, small intestine and stomach. In situ hybridization and immunohistochemical staining indicated that BRE was strongly expressed in the zona glomerulosa of the adrenal cortex, which synthesizes and secretes the mineralocorticoid hormones. It is also highly expressed in the glial and neuronal cells of the brain and in the round spermatids, Sertoli cells, and Ledig cells of the testis, all of which are associated with steroid hormones and/or TNF synthesis. However, BRE expression was downregulated in human adrenal adenoma and pheochromocytoma, whereas its expression was enhanced in abnormal adrenal tissues of rats chronically treated with nitrate or nitrite. These data, taken together, indicate that the expression of BRE is apparently associated with steroids and/or TNF production and the regulation of endocrine functions. BRE may play an important role in the endocrine and immune system, such as the cytokine–endocrine interaction of the adrenal gland.

Keywords

BRE adrenal steroid stress response neoplasia

Steroid hormones play important roles in stress responses and in the maintenance of homeostasis of the organism. Steroids are synthesized mainly in specialized steroidogenic cells in the adrenal gland, in the ovary and placenta of females, and in the testis of males. The adrenal glucocorticoids and mineralocorticoids are important for carbohydrate metabolism, stress management, and salt balance (Neville and O'Hare 1979). On the other hand, the sex steroids, estrogens and androgens, are essential for reproduction and for maintaining secondary sex characteristics (Davidoff et al. 1993). An additional class of steroids known as the neuroactive or neural steroids are synthesized by the central nervous system and other organs and appear to have specific functions in neurological action or signaling (Mayerhofer et al. 1992; Davidoff et al. 1993; Middendorff et al. 1993). Although a great deal has been learned about the functions of steroid hormones, much remains to be elucidated about the regulation, synthesis, and transport of steroids and their precursors. Several proteins that regulate steroid synthesis were previously found. Among them, the steroidogenic acute regulatory (StAR) protein plays a critical role in steroidogenesis by enhancing the delivery of the substrate cholesterol from the outer mitochondrial membrane to the cholesterol side-chain cleavage enzyme (P450scc) (Stocco 2000). Another candidate, the sterol carrier protein-2 (SCP-2), also known as the nonspecific lipid transfer protein (nsL-TP), may function to maintain sterol movement within the cell to the mitochondria in support of steroidogenesis by affecting the utilization of peroxisome-derived cholesterol (Seedorf et al. 2000). However, the details of mobilizing cholesterol to the mitochondrial membrane and their regulation are still not well understood.

On the other hand, cytokines such as tumor necrosis factor (TNF) also play important roles in many normal and disease states, including sepsis, cachexia, anorexia, neural and muscle damage, degeneration, atrophy, and cancer (Friedberg et al. 1995; Darnay and Aggarwal 1997). These biological mediators modulate apoptosis (Adam et al. 1996; Castellino and Chao 1996), cell growth, homeostasis, differentiation, and many other important physiological functions. TNF acts by binding to the TNF receptor (TNFR), which then recruits a number of adaptor molecules and initiates a cascade of events leading to apoptosis (Adam et al. 1996; Castellino and Chao 1996). TNFR may initiate a survival pathway by activating the transcriptional factor NF-κB.

We were interested in genes responsive to DNA damage and oncogenesis (Li et al. 1995; Gu et al. 1998; Tian et al. 1999; Zhang et al. 2000), and we have previously published on identification of a new human gene that is downregulated after treatment of cells with DNA-damaging agents such as UV light and 4-nitroquinoline-1-oxide and with the differentiating agent retinoic acid (Li et al. 1995). This gene is expressed in brain and reproductive organs and hence it was named BRE (EMBL/GenBank Data Libraries accession No. L38616). The BRE gene encodes a 1.9-kb mRNA and gives rise to a protein of 383 amino acid residues with a molecular weight of 44 kD. Recently, an interaction between BRE and the TNFR-1 was found using the two-hybrid yeast system and by co-immunoprecipitation (Gu et al. 1998). Moreover, BRE interacts with the juxtamembrane (JM) region of TNFR1 but has no affinity for the p75-TNFR, Fas, or p75-NGFR of the TNFR family. Interestingly, overexpression of BRE could modulate the transcriptional factor NF-κB induced by TNF (Gu et al. 1998). However, BRE may be involved in regulating other unknown mediators in addition to TNF. To elucidate the function of this novel BRE gene, we analyzed the expression of BRE in various normal and abnormal tissues of human and rat using dot-blotting, Northern blotting, in situ hybridization (ISH), and immunohistochemistry (IHC).

Materials and Methods

Sample Preparation

Samples of human tissues were obtained from the Department of Pathology at Prince of Wales Hospital (Shatin, Hong Kong), and the pathology of the samples was identified by one of us (FMM Lai). Rat samples were from male Sprague–Dawley rats treated with or without 50 mg/liter of sodium nitrate or sodium nitrite in their drinking water for a period of 4 weeks to induce accumulation of lipid droplets in their adrenal glands (Parker and Parker 1998; Panesar 1999; Panesar and Chan in press). The animals were sacrificed at the end of the regimen and tissue samples were fixed immediately in 4% paraformaldehyde for 24 hr and then embedded in paraffin. The CUHK institutional guidelines for animal research were followed in handling animals.

Plasmid Preparation

The full-length BRE gene was amplified from a recombinant phage of λgt11 as template (Li et al. 1995) by using the upstream primer F1 (5′GGagatctGGGTACTGTGGGGAAAAACAC3′), and the downstream primer R2 (5′TAgtcgacGAACAGCGAGGGGCATTTA3′); an extra Sal I site was introduced after the stop codon of BRE. The 1.5-kb PCR product of BRE was extracted and first inserted into a compatible plasmid, pGEM-T (Promega Lab; Madison, WI). It was then subcloned to the mammalian expression-vector pCDNA3 (Invitrogen Lab; Carlsbad, CA) with different orientations to produce the sense p3BRE(+) or the antisense p3BRE(–) plasmid, both of which were expressed under the control of the CMV promoter.

Dot-blotting and Northern Blotting Analysis of mRNA

Dot-blotting and Northern blotting membranes were purchased from a commercial source (Clontech; Palo Alto, CA). The RNA master blot was used as a screening tool for the expression level of BRE mRNA in different human tissues. The multiple-tissue Northern blot was used to confirm the dot-blotting results. All hybridization steps were processed as described (Ausubel et al. 1997) and as recommended by the manufacturer. Briefly, the 1.1-kb BRE cDNA fragment was cut from the p3BRE(+) vector, which was used as template for the labeling of BRE probe with [α-³²P]-dCTP, using the Random Primers DNA Labeling Systems (Gibco/BRL; Rockville, MD). After prehybridization with the reagent provided by the manufacturer at 65C for 2 hr, the membranes were incubated with fresh hybridization buffer containing the labeled BRE probe and denatured–sonicated salmon sperm DNA and were hybridized overnight at 65C. The membranes were then washed several times with washing solution and exposed to X-ray films. To re-use the membranes, the labeled probe was stripped by washing the membranes twice for 20 min each with 0.5% SDS at 95C and rehybridized with the control gene β-actin.

Preparation of BRE-specific Transcript Probes for ISH

To synthesize the labeled transcripts for RNA hybridization, the plasmid DNA of BRE was treated with restriction enzymes that created either a 5′ overhang or a 3′ overhang of the linearized vector, which were used as templates to synthesize the sense or antisense RNA probes. A digoxigenin (DIG)-labeled antisense RNA probe was obtained using an Nco I-digested BRE template, SP6 RNA polymerase, and with a DIG-RNA labeling kit from Boehringer–Mannheim (Mannheim, Germany) as described (Ausubel et al. 1997). Similarly, the sense RNA probe was prepared as a negative control in experiments by using a Not I-cut template, T7 RNA polymerase, and with the same labeling kit. The DIG-labeled BRE RNA probes were purified by ethanol precipitation. A spot test with the DIG-labeled reaction was performed to estimate the yield of the labeling.

In Situ Hybridization

The method was essentially the same as described by us recently for the PML gene (Zhang et al. 2000). Tissue sections 5 μm thick were prepared from the fixed samples with a disposable knife and attached to the APES-coated slides. The sections were dried overnight at 37C and slides were dewaxed extensively by treating with xylene and a graded series of ethanol solutions. The sections were fixed with 4% paraformaldehyde for 1 hr and then rinsed three times with PBS for 5 min each. After treatment with 0.3% Triton X-100/PBS for 15 min, the samples were digested with 20 μg/ml proteinase K at 37C for 30 min. The sections were rinsed three times for 5 min each with PBS and then immersed in hot 70% deionized formamide/2 × SSC at 70C for 5 min, and for the BRE cDNA probe tissues were incubated at 95C for 5 min. After dehydration in a graded series of cold ethanol solutions and air-drying, the slides were hybridized with 1 ng/μl BRE RNA probe at 42C overnight in 50 μl hybridization-buffer (4 × SSC, 50% deionized formamide, 10% dextran sulfate, 1 × Denhardt's solution and 1 μg/μl of sonicated and denatured salmon sperm DNA). RNase inhibitor at 0.4 U/μl was also added to prevent RNA degradation. After hybridization, the slides were washed with 2 × SSC twice for 5 min each. Samples were further washed once in 60% formamide/0.2 × SSC for 15 min, three times in 0.2 × SSC for 5 min each, three times in Buffer I (100 mM Tris-HCl, pH 8.0, 150 mM NaCl) for 5 min each, and once in 0.5% blocking solution/Buffer I for 30 min at room temperature. The slides were then incubated for 60 min with alkaline phosphatase-conjugated anti-DIG antibody (1:200 in blocking solution). After rinsing of the sections three times for 5 min with PBS and then 5 min with Buffer II (100 mM Tris-HCl, pH 8.0, 100 mM NaCl, and 50 mM MgCl₂), the alkaline phosphatase reaction was developed with the NBT/BCIP color reagent. To confirm the specificity of the ISH, the sense BRE probe was used as mRNA negative control.

Preparation of BRE Antibody

To generate BRE antibody, a recombinant polypeptide of BRE from codon 38 to 226 was expressed in E. coli using the pRSET-B vector. The 30-kD BRE recombinant protein was purified through an Ni⁺ column from the bacterial lysate. The antibody was produced in rabbits and subsequently purified through affinity chromatography of the recombinant BRE protein on a fee-for-service basis by Research Genetics Laboratory (Huntsville, AL). The titer of the antibody was determined to be positive at over 200, 000-fold dilution with an ELISA assay. The antibody was also characterized by Western blotting and immunohistochemical staining, indicating specific interaction with the BRE protein expressed in HeLa and 293 cells after transfection with the BRE cDNA.

Immunohistochemical Staining

Rat or human tissues were fixed in 10% formalin in PBS for 20 hr before dehydration in graded ethanol and embedding in Paraplast. Tissue sections of 5 μm were cut and mounted on silanized slides. After dewaxing with xylene and rehydration with an alcohol gradient, the slides were boiled in 10 mM sodium citrate buffer, pH 6.8, for 10 min in a microwave oven. The samples were then treated with 3% hydrogen peroxide in PBS for 10 min to quench endogenous peroxidase activity. After blocking with 5% rabbit serum for 20 min to reduce the nonspecific binding, the sections were incubated for 1 hr with rabbit anti-BRE antibody diluted at 1:800, then for 20 min with biotinylated goat anti-rabbit IgG (Zymed; San Francisco, CA), and then for 20 min with HRP–streptavidin (Zymed). The reactivity of the BRE antibody was visualized by incubating the slides with DAB–hydrogen peroxide solution from a DAB detection kit (Zymed). Finally, the sections were counterstained with hematoxylin, dehydrated, and mounted.

Results

Differential Expression of BRE in Various Human Tissues Using the Dot-blotting Hybridization Assay

To rapidly determine the expression of this new gene, BRE, in different tissues, we used a commercially available mRNA dot-blot membrane, the RNA MasterBlot from Clontech, to examine BRE expression in a large number of tissues. As shown in Figure 1, 50 different human tissue samples were analyzed. The expression of BRE was strongest in adrenal gland, followed by the heart, pituitary gland, spleen, skeletal muscle, occipital lobe, and putamen. Moderate expression was found in the whole brain, testis, and ovary, and low expression was found in all fetal tissues, appendix, cerebellum, thymus, peripheral leukocytes, trachea, and placenta.

Differential Expression of BRE in Various Normal Human Tissues Using the Northern Blotting Analysis

To rule out nonspecific hybridization in the above dot-blotting assay, Northern blotting was then performed on some of the tissue mRNA, particularly that of the adrenal gland. To avoid differences created in sample preparation, a commercially available Northern blot membrane, a Multiple-Tissue Northern Blot, was purchased and analyzed (Clontech). As shown in Figure 2, Northern blotting analysis of eight different human tissues indicated that only one major species of mRNA at 1.9 kb was hybridized to the BRE probe (Figure 2A), similar to the results published previously by us (Li et al 1995). This BRE mRNA was highly expressed in the adrenal medulla and cortex, testis, and pancreas, whereas low expression was found in the thyroid, thymus, small intestine, and stomach. By scanning with a densitometer, the values of expression for BRE from these tissues were 10.2-, 9.1-, 6.0-, 6.3-, 1.6-, 1.2-, 1-, and 1.5-fold, respectively, for these samples, using the value of the lowest (small intestine) as 1-fold. The control β-actin mRNA was readily detected on the same membrane and the expression levels were similar for all tissues except for the pancreas, which apparently contained a larger amount of mRNA (Figure 2B). However, these data, in general, confirmed most of the findings of the dot-blotting analysis as well as our previously published data.

Figure 1

Differential expression of BRE mRNA in various human tissues using the dot-blotting hybridization assay. BRE expression was analyzed in a dot-blot membrane filter from Clontech using labeled BRE-cDNA probe as described in Materials and Methods. (A) The names and positions of the tissue mRNA spotted on the membrane. (B) Hybridization results.

Figure 2

Differential expression of BRE mRNA in various human tissues using the Northern blotting analysis. Northern blotting of BRE was performed with an mRNA membrane filter for different human organs (Clontech). Hybridization and washing were performed according to the manufacturer's instructions. (A) Results of the BRE probe. (B) β-Actin control.

ISH of BRE mRNA and IHC Staining of BRE in Various Rat Organs

To elucidate the cell types or specific regions of the various tissues that expressed BRE, ISH was performed. Because the human BRE gene shares more than 90% homology with the BRE cDNA fragments from rodent, as shown by the NCBI-Genbank data search, rat organs were used that are readily available. Figures 3A and 3B show the ISH of the human BRE in the adrenal gland of the rat with the antisense probe. Strong expression of BRE mRNA (dark purple staining) was found in the cytoplasm of cells at the zona glomerulosa (ZG) and moderate expression in the zona fasciculata (ZF) of the adrenal cortex (c). The adrenal medulla (m) also showed considerable expression. In contrast, the control hybridization with the BRE sense probe in the same adrenal tissue showed no staining (Figure 3C). The hybridization was confirmed by IHC staining with the BRE antibody (Figures 3D and 3E), which also showed strong reaction in the ZG (dark brown staining). BRE protein was localized mostly in the cytoplasm and in the perinuclear regions of cells. However, treatment with pre-immune serum produced no detectable reaction, indicating that the staining was specific (Figure 3F). In the rat brain, ISH with BRE antisense probe gave positive results in the glial cells (Figure 3G, arrowhead). In the rat testis, the immature spermatocytes and the Sertoli cells stained strongly with the BRE antisense probe (Figure 3H, arrow), while the Leydig cells were weakly positive (Figure 3H, arrowhead). In contrast, only the bile duct cells in rat liver showed strongly positive signal (Figure 3I, arrowhead), while the hepatocytes showed weak hybridization (Figure 3I, arrow). The IHC stainings for BRE (dark brown) of the latter three tissues are shown in Figures 3J-3L respectively. In brain, the cytoplasm of both the glial cells and the pyramidal neuronal cells showed positivity. In testis, the interstitial Leydig cells, which are the androgen-producing cells, showed strong positive signal for BRE protein (Figure 3K, arrowhead). In liver, the hepatocytes showed weak but detectable BRE expression (Figure 3L).

Figure 3

ISH of BRE mRNA and IHC staining of BRE protein in various normal tissues of the rat. (A,B) Adrenal gland of rat hybridized with the antisense BRE probe (dark purple staining). (C) The same adrenal gland hybridized with the sense BRE probe as control. The adrenal cortex (C) and medulla (M), zona glomerulosa (ZG), and zona fasciculata (ZF) are shown. (D,E) The corresponding adrenal tissue immunohistochemically stained with the BRE antibody (dark brown staining). (F) The same adrenal tissue immunohistochemically stained with the pre-immune serum as control. (G–I) ISH with the BRE antisense probe (dark purple staining) on brain, testis, and liver of the rat, respectively. (J–L) The corresponding IHC staining with anti-BRE (dark brown staining) on brain, testis, and liver, respectively. Bars: A,C,D,F = 350 μm; B,E = 70 μm; G–L = 35 μm.

Figure 4

Altered expression of BRE mRNA in human adrenal tumors and abnormal adrenal tissues of rats treated with nitrate or nitrite. Antisense BRE labeled with digoxigenin (dark purple staining) was hybridized to slides of adrenal tissues of human and rat that had previously been fixed in paraformaldehyde and embedded in paraffin. (A,D) Non-tumorous adrenal gland of human. (B,E) Human adrenal adenoma. (C,F) Human adrenal pheochromocytoma (T, tumor; arrows show the cells expressing BRE). (G) Adrenal gland from normal untreated control rat. (H) Adrenal gland from sodium nitrate-treated rat. (I) Adrenal gland from sodium nitrite-treated rat. Bars: A-C = 350 μm; D–I = 70 μm.

Decreased Expression of BRE mRNA in Adrenal Tumors

To determine if BRE expression may be altered in neoplastic tissues of the adrenal gland, human adrenal adenomas and pheochromocytomas were analyzed. Figure 4A and 4D show the ISH of non-tumorous human adrenal gland, which showed strong expression of BRE mRNA. Adrenal adenomas were tumors of the adrenal cortex, which were from patients with primary hyperaldosteronism (Conn's syndrome). These neoplasias showed decreased or undetectable expression of the BRE mRNA in the tumor cells (T) as compared to the adjacent non-tumorous cells (Figures 4B and 4E, arrowhead) in all of the samples examined (n = 5). Similarly, in pheochromocytomas, tumors of the adrenal medulla, there was decreased or undetectable expression of BRE mRNA in all of the samples (n = 4) analyzed (Figures 4C and 4F).

Enhanced Expression of BRE mRNA in Adrenal Glands of Rats Treated with Nitrate or Nitrite

It has been previously documented that inorganic nitrate inhibits gonadotropin-stimulated steroidogenesis in mouse Leydig tumor cells (Panesar 1999; Panesar 2000). In addition, it was recently found that rats treated with nitrate and nitrite have decreased levels of corticosterone and testosterone and the adrenal glands contain a large number of cells with large lipid droplets in their cytoplasm (Panesar and Chan in press). The cause of the latter is most likely impairment of steroidogenesis by nitric oxide from nitrite and nitrate, with accumulation of a large amount of cholesterol, a situation akin to lipoidal congenital adrenal hyperplasia (White et al. 1987). The adrenal glands of rats fed 50 mg/liter nitrate or nitrite in their drinking water for 4 weeks showed lipid-laden cells in the zona fasciculata and zona reticulata of the adrenal cortex. To determine if BRE may be involved in these abnormal conditions, the expression of BRE was examined in adrenal glands of rats treated with nitrate or nitrite. As shown in Figures 4G–4I, the numbers of lipid droplets in 100-μm² section of adrenal cortices were quantified. In five randomly picked regions of adrenal cortices in microscopic slides, the average numbers of lipid droplets were 10 ± 4, 68 ± 7, and 59 ± 10 for control, nitrate-, and nitrite-treated rats, respectively. Untreated control rats showed moderate to high levels of BRE mRNA in the adrenal cortex (Figure 4G). However, the adrenal glands of rats treated with nitrate and nitrite showed two- to fourfold increased expression of BRE mRNA in the zona glomerulosa and the lipid-laden zona fasciculata, (Figures 4H and 4I).

Discussion

Although BRE had previously been found to bind to TNFR-1, its precise role in vivo is still unknown. Here we found increased expression of BRE mRNA and protein in several steroid hormone-producing organs, including the adrenal gland, testis, ovary, and brain. ISH and immunohistochemical staining identified the cell types with strong BRE signals, including the cells in the zona glomerulosa of the adrenal gland. In contrast, BRE mRNA and protein in zona fasiculata and zona reticulata was much reduced. Cells at the zona glomerulosa of the adrenal synthesize and secrete mineralocorticoid hormones, such as aldosterone (Neville and O'Hare 1979), as well as the cytokine TNF (Nussdorfer and Mazzocchi 1998; Judd 1998). Although the mRNA of BRE was found to be high in the glial cells and low in the pyramidal neuronal cells of the brain, BRE protein was strongly expressed in both cell types. Similarly, mRNA of BRE was high in the germ cells but low in the interstitial Leydig cells of the testis, while BRE protein was found in both cell types. This may indicate that the BRE mRNA is less stable, and the BRE protein more stable, in certain cell types. In addition, it was previously found that BRE had limited homology (amino acid residues 138–197) to a tyrosine-specific transport protein of E. coli, ecotyr (Li et al. 1995). We have recently found that BRE also contains motifs of N-terminal amphipathic amino acids (codons 1–36) composed of two α-helical segments, which were previously postulated to represent putative phospholipid interaction sites and mitochondria targeting. It also contains peroxisome localization signals (SRL-XXX-SRL) consisting of two tripeptides of SRL at codons 130 and 136 as well as a two-step for peptides (RXØXXS) at codons 131–136 for mitochondrial protein. These motifs were apparently similar to those of the sterol carrier protein-2 (SCP2) (Seedorf et al. 2000), and the steroidogenic acute regulatory protein (StAR) for steroid transport and metabolism (Stocco 2000). In addition, BRE protein was localized in the cytoplasm and the perinuclear regions of cells, indicating that it is not a nuclear protein and functions in the cytoplasm. BRE may act to transport or regulate organic compounds or the precursors of steroid hormones through the cytoplasm. Therefore, BRE, a stress-responsive gene, may play a role in modulating the effect or the production of steroid hormones in the steroid-producing organs. Aldosterone and corticosterone secretion in the adrenal gland is regulated by the anterior pituitary hormones angiotensin II and ACTH, respectively, which are inhibited by TNF-α (Judd 1998; Nussdorfer and Mazzocchi 1998). Many stimuli, including stress, promote the secretion of glucocorticoids, which then adjust the body metabolism accordingly. In addition, the zona fasciculata is also the site of secretion of androgenic sex hormones.

However, the majority of adrenal adenoma represents Conn's syndrome, with hyperaldosteronism and low plasma renin (Neville and O'Hare 1979). This condition is associated with hyperplasia of the zona glomerulosa, but BRE expression was decreased. Because Conn's syndrome is independent of ACTH, the decreased BRE expression is apparently not a feedback inhibition by the hormone. The lack of expression of BRE in adrenal adenomas and pheochromocytomas is, however, a clue suggesting that it might play a role in suppressing tumorigenesis or the transformed phenotypes. On the other hand, we observed enhanced expression of BRE in the zona glomerulosa and zona fasciculata in the adrenal glands of rats treated with nitrate and nitrite. BRE was strongly expressed in cells of the zona fasciculata adjacent to the lipid droplets, which represented defective synthesis of the steroid hormone and were the site of cholesterol accumulation (Parker and Parker 1998; Panesar 1999). These cells are known to be defective in the synthesis of steroid hormones and accumulate large quantities of cholesterol as visible lipid vesicles (Panesar and Chan in press; Szabó et al. 1996). Therefore, the cells are under stress or are under the influence of steroid hormones, TNF, and other inflammatory cytokines, in which BRE could then play a role.

It is interesting to note that several cytokines, including TNF, are known to stimulate the hypothalamo–pituitary corticotropin-releasing hormone (CRH/ACTH system), thereby evoking a secretory response by the adrenal cortex (de Almeida and Magalhães 1998; Judd 1998; Nussdorfer and Mazzocchi 1998). TNF and interelukin (IL)-1, −2, −6, and INF-γ are synthesized in the adrenal gland by parenchymal cells and macrophages. The release of TNF is regulated by the main agonists of steroid hormone secretion, including ACTH and angiotensin-II, and bacterial endotoxins. TNF and IL-1 directly inhibit aldosterone secretion by zona glomerulosa cells, whereas IL-6 enhances it. In addition, TNF depresses glucocorticoid synthesis. There is evidence that local immune–endocrine interactions may play an important role in modulating adrenal responses to inflammation and to immune challenges and stresses. Here, we showed that the expression of BRE is in the zona glomerulosa, which is the exclusive site for synthesis and release of TNF. In this context, BRE may play an important role in regulating the cytokine–endocrine interaction in the adrenal gland.

Because the glial cells are the supportive cells in the brain, they are rapidly activated in response to injury, inflammation, neurodegeneration, and infection. The microglial cells, which are the resident macrophages, produce and release TNF (Benveniste and Benos 1995; Soliven and Szuchet 1995). The glial cells respond to TNF, leading to destruction of myelin and oligodendrocytes and to the remyelination and repair. In addition, glial cells can easily develop into the major malignancy of the brain, glioblastoma. Therefore, BRE may be important in the cellular response to stress and xenobiotics for the glial cells. In testis, BRE mRNA expression was found in immature germ cells, such as the round spermatids, and in the Sertoli cells, while the BRE protein was very high in the androgen-producing Leydig cells. Again, TNF-α is secreted by the round spermatids in the seminiferous tubules and by macrophages in the interstitial tissues (Delfino and Walker 1998; Yan et al. 1999). TNF has two types of action in the testis: one is to inhibit LH-induced testosterone production in Leydig cells and the second one is to inhibit FSH-induced inhibin production in Sertoli cells (De et al. 1993; Del Punta et al. 1996). In addition, TNF produced by germ cells enhances the expression of growth factors in Sertoli cells and the production of energetic metabolites important for germ cells, such as lactate via the induction of lactate dehydrogenase isozyme 5 or A (Nehar et al 1997; Delfino and Walker 1998; Yan et al., 1999). TNF therefore plays a key role in the communication between Sertoli and germ cells, which is important for the interaction of the immune system and spermatogenesis.

In view of the complexity and the diverse roles of steroids, TNF, and other signaling molecules in cell survival, apoptosis, growth, and differentiation, it is not surprising that BRE is expressed differently in different tissues and cell types and that it may have differential effects on cells. We are now generating transgenic and knockout models for analyzing the in vivo effect of BRE, and therefore, our results may provide clues to the organs and the phenotypes to be identified.

Footnotes

Acknowledgements

Supported in part by CUHK-UGC grant 2040519, RGC-earmark grant CUHK 4279/97M, and a grant from National Science Council, Taiwan, awarded to JYH Chan, and by a UGC grant awarded to NS Panesar.

References

Adam

Adam-Klages

Kronke

(1996) Identification of p55 tumor necrosis factor receptor-associated proteins that couple to signaling pathways not initiated by the death domain. J Inflammation 47:1–66

Ausubel

Brent

Kingston

Moore

Seidman

Smith

Struhl

, eds. (1997) Current Protocol in Molecular Biology. New York, John Wiley & Sons

Benveniste

Benos

(1995) TNF-alpha- and IFN-gamma-mediated signal transduction pathways: effects on glial cell gene expression and function. FASEB J 9:1577–1584

Castellino

Chao

(1996) Trans-signaling by cytokine and growth factor receptors. Cytokine Growth Factor 7:297–302

Darnay

Aggarwal

(1997) Early events in TNF signaling: a story of associations and dissociations. J Leukocyte Biol 61:559–566

Davidoff

Schulze

Middendorff

Holstein

(1993) The Leydig cell of the human testis—a new member of the diffuse neuroendocrine system. Cell Tissue Res 271:429–439

Chen

Pace

Hunt

Terranova

Enders

(1993) Expression of tumor necrosis factor-alpha in mouse spermatogenic cells. Endocrinology 133:389–396

de Almeida

Magalhães

(1998) Age-related changes in the inner zone of the adrenal cortex of the rat—a morphologic and biochemical study. Mech Ageing Dev 105:1–18

Delfino

Walker

WJH

(1998) Stage-specific nuclear expression of NF-kB in mammalian testis. Mol Endocrinology 12:1696–1707

10.

Del Punta

Charreau

Pignataro

(1996) Nitric oxide inhibits Leydig cell steroidogenesis. Endocrinology 137:5337–5343

11.

Friedberg

Walker

Siede

, eds. (1995) DNA Repair and Mutagenesis. Washington, DC, ASM Press

12.

Castellino

Chan

JYH

Chao

(1998) BRE: a modulator of TNF-alpha action. FASEB J 12:1101–1108

13.

Judd

(1998) Cytokine expression in the rat adrenal cortex. Horm Metab Res 30:404–410

14.

Yoo

Becker

Ali-Osman

Chan

JYH

(1995) Identification of a brain- and reproductive-organs-specific gene responsive to DNA damage and retinoic acid. Biochem Biophys Res Commun 206:764–774

15.

Mayerhofer

Seidl

Lahr

Bitter–Suermann

Christoph

Barthels

(1992) Leydig cells express neural cell adhesion molecules in vivo and in vitro. Biol Reprod 47:656–664

16.

Middendorff

Davidoff

Holstein

(1993) Neuroendocrine marker substances in human Leydig cells—changes by disturbances of testicular function. Andrologia 25:257–262

17.

Nehar

Mauduit

Boussouar

Benahmed

(1997) Tumor necrosis factor-alpha-stimulated lactate production is linked to lactate dehydrogenase A expression and activity increase in porcine cultured Sertoli cells. Endocrinology 138:1964–1971

18.

Neville

O'Hare

(1979). Aspects of Structure, Function, and Pathology. James

VHT

, ed. The Adrenal Gland. New York, Raven Press, 1–66

19.

Nussdorfer

Mazzocchi

(1998) Immune-endocrine interactions in the mammalian adrenal gland: facts and hypotheses. Int Rev Cytol 183:143–184

20.

Panesar

(1999) Role of chloride and inhibitory action of inorganic nitrate on gonadotropin-stimulated steroidogenesis in mouse Leydig tumor cells. Metabolism 48:693–700

21.

Panesar

(2000) Is steroid deficiency the cause of tolerance in nitrate therapy? Med Hypotheses 55:310–313

22.

Panesar

Chan

(in press) Decreased steroid hormone synthesis from inorganic nitrite and nitrate: studies in vitro and in vivo. Toxicol Appl Pharmacol

23.

Parker

(1998) Nitrate therapy for stable angina pectoris. N Engl J Med 338:520–531

24.

Seedorf

Ellinghaus

Roch Nofer

(2000) Sterol carrier protein-2. Biochim Biophys Acta 1486:45–54

25.

Soliven

Szuchet

(1995) Signal transduction pathways in oligodendrocytes: role of tumor necrosis factor-alpha. Int J Dev Neurosci 13:351–367

26.

Stocco

(2000) The role of the StAR protein in steroidogenesis: challenges for the future. J Endocrinol 164:247–253

27.

Szabó

Tóth

Szalay

(1996) Viscosity of rat adrenocortical lipids in different functional states: morphological characteristics. J Steroid Biochem Mol Biol 58:329–335

28.

Tian

Chan

JYH

Pang

JCS

Chen

TSS

Leung

(1999) Altered expression of the suppressors PML and p53 in glioblastoma cells with the antisense-EGF-receptor. Br J Cancer 81:994–1001

29.

White

New

Dupont

(1987) Congenital adrenal hyperplasia. N Engl J Med 316:1519–1524

30.

Yan

Linderborg

Suominen

Toppari

(1999) Stage-specific regulation of stem cell factor gene expression in the rat seminiferous epithelium. Endocrinology 140:1499–1504

31.

Zhang

Chin

Chow

LTC

Chan

ASK

Yim

APC

Leung

Mok

TSK

Chang

Johnson

Chan

JYH

(2000) Lack of expression for the suppressor PML in human small cell lung carcinoma. Int J Cancer 85:599–605