Abstract
The human adrenal cortex secretes aldosterone and cortisol as major corticosteroids. For their production, CYP11B2 and CYP11B1 catalyze the last steps in the syntheses of aldosterone and cortisol, respectively. In our previous study, CYP11B2 was the first successfully purified from rat adrenals and human clinical samples and then was proved to be aldosterone synthase. We demonstrated the immunohistochemistry for CYP11B2 of both rats and humans and applied it clinically to visualize the functional histology of aldosterone-producing adenoma (APA) causing primary aldosteronism (PA). We discovered aldosterone-producing cell clusters (APCCs) and possible APCC-to-APA transitional lesions (pAATLs) and further visualized aldosterone-producing lesions for rare forms of PA including familial hyperaldosteronism type 3 and novel non-familial juvenile PA. Here we review the history of our research on aldosterone-producing lesions.
Introduction
In addition to aldosterone-producing adenoma (APA), there are newly discovered aldosterone-producing lesions such as aldosterone-producing cell clusters (APCCs) and lesions that suggest the transition from APCC to APA (which we have called possible APCC-to-APA transitional lesions: pAATLs). It has been difficult to detect them without using specific methods visualizing aldosterone-producing cells. The terms are based on our observations of adrenal tissues, and, by using the terms, we intend to facilitate dissemination of our thoughts in the literature. To understand these newly discovered lesions, the latest knowledge regarding the genetics of aldosterone production is also helpful. In this review, we will thus explain our findings on the latest pathology of aldosterone production along with its genetic background.
Synthetic Pathways of Aldosterone and Cortisol
All steroid hormones including sex steroids are synthesized from cholesterol. In the zona glomerulosa (ZG) of the adrenal gland, cholesterol side-chain cleavage enzyme (CYP11A1), 3β-hydroxysteroid dehydrogenase (HSD3B2), steroid 21-hydroxylase (CYP21A2), and aldosterone synthase (CYP11B2) work sequentially to synthesize aldosterone (Fig. 1). In the zona fasciculata (ZF), CYP11A1, steroid 17α-hydroxylase (CYP17A1), HSD3B2, CYP21A2, and steroid 11β-hydroxylase (CYP11B1) work to synthesize cortisol. CYP11A1, CYP11B1, and CYP11B2 are localized in the mitochondria, while HSD3B2, CYP17A1, and CYP21A2 are localized in the endoplasmic reticulum (ER). The cellular synthetic processes of aldosterone and cortisol are illustrated in Fig. 2.
Immunohistochemical Staining of CYP11B2 and CYP11B1
Antibodies against CYP11B2 and CYP11B1
CYP11B2 was purified from rat adrenals in 1989 and from human APA samples in 1991 in our previous works, and we demonstrated that CYP11B2 was aldosterone synthase [1, 2]. In these studies, we generated the species-specific polyclonal antibodies that distinguished CYP11B2 from CYP11B1. In 1992, we used immunohistochemical staining of CYP11B2 and CYP11B1 to clearly visualize the functions of the ZG and ZF in rats [3, 4]. However, the functional zonation in humans was not visualized until our findings were published in 2010 [5].
Histology of normal adrenal cortex
The ZG was first described by the German pathologist Julius Arnold [6] in 1866. This zone is characterized by spherical cell populations comprised of small cells. Recently, such cell populations were referred to as rosettes [7]. Using the polyclonal antibodies specific to human CYP11B2 and those specific to CYP11B1 in 2010 we visualized the aldosterone-producing cells in the human normal adrenal cortex [5]. As is the case with rat adrenal cortex, CYP11B2 was detected in the ZG and CYP11B1 was detected in the ZF (Fig. 3A, Monoclonal mouse anti-human CYP11B2 antibody and monoclonal rat anti-human CYP11B1 antibody in this figure were developed by others [8]). The expression of CYP17A1 and HSD3B2 was consistent with the steroidogenic abilities of these zones (Fig. 3B and C, respectively). Interestingly, subcapsular cell clusters (0.2–1 mm) expressing CYP11B2 are detected in most of the adult adrenal cortices (Fig. 3D and E). In the clusters, cells underneath the capsules are morphologically ZG-like, and cells inside (medullary side) appear to be ZF-like. Therefore, it is difficult to detect such clusters on tissue sections stained with hematoxylin and eosin (H&E). These masses express HSD3B2 (Fig. 3F) but do not express CYP17A1 (Fig. 3G), being consistent with the notion that the cells synthesize aldosterone. We termed these masses aldosterone-producing cell clusters (APCCs).
The immunohistochemical staining of CYP11B2 and CYP11B1 is also useful for the diagnosis of APA and CPA. It was found that APA has CYP11B2-positive cells but also CYP11B1-positive cells and double-negative cells (Fig. 4A and B) [5]. Adrenocortical adenomas generally consist of small (ZG-like) cells and large lipid-rich (ZF-like) cells. However, there is no correlation between the morphology of these cells and the expression of either CYP11B2 or CYP11B1. In the adjacent adrenal cortex of APA, very few CYP11B2-positive cells are detected in the morphological ZG presumably because of the low blood angiotensin II levels in patients with APA (Fig. 4C and D). By contrast, APCCs are found in the non-tumor adrenals of most of patients with APA (Fig. 4E and F). This suggests that APCCs produce aldosterone autonomously, that is, their aldosterone production is independent from the renin-angiotensin system.
In CPA, the lesion responsible for Cushing’s syndrome, CYP11B2-expressing cells are not detected at all, as it consists exclusively of CYP11B1-positive cells (CPA in Fig. 4G and H). CPA produces cortisol, which is consistent with its expression of HSD3B2 (Fig. 4I) and CYP17A1 (no figure). Due to the negative feedback mechanism, low levels of the blood ACTH cause the non-tumor portions to be significantly atrophic. Such portions express very low levels of CYP11B1 and HSD3B2 (the NT [non-tumor] portion in Fig. 4H), suggesting that these portions do not efficiently produce cortisol. ACTH is also involved in the production of aldosterone in the ZG under stress [9]. Interestingly, the non-tumor portions of CPA often have APCCs, where the expression levels of CYP11B2 and HSD3B2 are high [5]. These observations suggest that even in patients whose blood ACTH levels are low, such as those with Cushing’s syndrome, APCCs exist independently of ACTH and produce aldosterone autonomously.
Juvenile adrenal cortex
As described above, the adult adrenal cortex has APCCs in most cases and an adenoma occasionally, while the juvenile adrenal cortex exhibits a clear three-layer structure (Fig. 5A and B) [10]. Our previous study using rat adrenals indicated the presence of a cell layer devoid of CYP11B2 nor CYP11B1 between the ZG and ZF [3]. Similarly, such an undifferentiated cell layer was observed in the human juvenile adrenal cortex (Fig. 5A). The layer structure in childhood remodels themselves to form APCCs with age increasing in number and size [10, 11].
Gene Abnormalities Involved in Pathological Production of Aldosterone
Abnormalities in the ion channel/pump genes were discovered in APA [12-15]. The first mutation reported in 2011 was in the inwardly rectifying potassium channel subfamily J member 5 (KCNJ5) gene [15]. The KCNJ5 gene encodes G protein-activated inward rectifier potassium channel 4 (GIRK4). Both germline and somatic mutations of the gene impair the ion selectivity of GIRK4 and cause the influx of Na+ into cells, leading to the depolarization of the cell membrane and the elevation of the intracellular calcium concentration. As a result, the transcription of CYP11B2 is promoted, and aldosterone is overproduced. Following the report on the KCNJ5 mutation, somatic mutations in the ATP1A1, ATP2B3, and CACNA1D genes were identified in APAs by similar analyses using exome sequencing [12-14]. All these APA-associated mutations are thought to influence intracellular ions and homeostasis and cause the overproduction of aldosterone. Germline mutations in the KCNJ5 gene cause familial aldosteronism type 3, which exhibits an autosomal dominant pattern of inheritance [15], and germline mutations in the CACNA1D gene cause PA with seizures and neurologic abnormalities (PASNA) syndrome characterized by comorbid central nervous system symptoms [14].
Our studies on the ion channel/pump genes have shown that most of APCCs have a mutation of ATP1A1, ATP2B3, or CACNA1D genes similar to those in APA but not a mutation of the KCNJ5 gene [16]. These results are consistent with the hypothesis that APCCs produce aldosterone autonomously and that APCCs are the origin of APA carrying a mutation in ATP1A1, ATP2B3, or CACNA1D genes.
Classification and Lesions of PA
Classification of PA
PA is broadly classified into unilateral PA (unilateral APA and rare unilateral non-tumor PA), bilateral PA (idiopathic hyperaldosteronism [IHA] and rare bilateral APA [17]), familial aldosteronism, and other forms of PA. Unilaterality and bilaterality are distinguished by adrenal venous sampling. Familial aldosteronism is further classified into types 1–3 (FH-1 to FH-3), PASNA syndrome, and others. FH-1 is caused by the formation of a CYP11B1/B2 chimeric gene due to unequal crossing over between the CYP11B1 gene and the CYP11B2 gene [18] (Fig. 6). Close relatives of a patient with FH-2 frequently develop either unilateral APA or bilateral adrenal hyperplasia. FH-3 is caused by germline mutations in the KCNJ5 gene.
The histology of APA is described in the preceding section, and, as the word “idiopathic” in its name indicates, the histology of IHA is not well known because adrenalectomy is usually not performed. In the following section, we discuss histopathologies of non-tumor case of unilateral PA, IHA, FH-3, and a rare non-familial juvenile case of PA.
Histology of unilateral non-tumor PA
We performed CYP11B2/CYP11B1 staining for adrenals from patients who were clinically diagnosed as unilateral PA and were pathologically diagnosed as adrenal hyperplasia (Fig. 7A and B) [19]. Immunohistochemical staining for CYP11B2 revealed an aldosterone-producing lesion consisting of a subcapsular APCC-like portion and an inner APA-like portion (Fig. 7C and D). The APCC-like and the APA-like portions showed histological and functional features similar to those of APCC and APA, respectively: The APCC-like portion consisted of ZG-like and ZF-like cells, and the APA-like portions consisted of both CYP11B2-positive and CYP11B1-positive cells. We collected tissues separately from the two portions (Fig. 7E and F, respectively) and analyzed them for the APA-associated mutations reported for the genes encoding the ion pump/channels. It was found that the APCC-like portion had none of the mutations, but the APA-portion had a novel mutation in the ATP1A1 gene.
Using another case of unilateral non-tumor PA, we analyzed two lesions consisting of an APCC-like and an APA-like portions and obtained the following findings [19]. One is that an APCC-like and an inner APA-like portions from a single lesion both had the same ATP1A1 gene mutation as that already reported for APAs, suggesting that a single cell carrying the mutation could develop to both portions. The other is that the APA-like portion, but not the APCC-like portion in the same lesion, had a KCNJ5 gene mutation identical to one of the mutations already reported for APAs, suggesting a possibility that when some cells acquire the APA-associated KCNJ5 gene mutation in APCC, they develop to APA. From these findings, we previously defined an aldosterone-producing lesion consisting of an APCC-like and an APA-like portions as a newly identified transitional lesion and therefore used the tentative term possible APCC-to-APA transitional lesion (pAATL) [19].
Histology of bilateral PA
Recently, we reported that the cure rate of PA by unilateral adrenalectomy was higher in patients with a lesion size ≥5 mm (mostly APAs, Fig. 8A–D) than those with that <5 mm (small APA, APCC, and pAATL, Fig. 8E–H) [20]. Of note, our in-house monoclonal mouse anti-human CYP11B2 antibody was utilized (CYP11B2-4E11-1G7 [21]) in the study. The results suggest that although the lateralization indexes (lateralized ratio and contralateralized ratio) from adrenal venous sampling of the patients with a lesion size <5 mm satisfied the criteria for unilateral adrenalectomy, the contralateral adrenal glands secreted excessive levels of aldosterone. Thus, the PA conditions which were not cured or improved by surgery are in part characterized by IHA. It is possible that small lesions develop in the contralateral adrenal gland. On the basis of the treatment outcome and the biological analyses, we presume that the small APA, APCC, and pAATL are aldosterone-producing lesions of IHA.
Histology of FH-3
Both adrenal glands of patients with FH-3 markedly enlarge and histologically lose their layered structure. Immunohistochemical examination indicated that CYP11B2-positive and CYP11B1-positive cells intermingle irregularly throughout the adrenal glands [22, 23] (Fig. 9).
Histology of other forms of PA
It is supposed that there are other pathologies of PA in addition to those described above. For example, we describe a patient whose both adrenals had a genetic mosaicism with a somatic mutation of the KCNJ5 gene [24]. The patient was diagnosed as having PA at eight years of age. Her PA was severe and resistant to pharmacotherapy. Using super-selective adrenal venous sampling [25], aldosterone production in the lower portion of the right adrenal was found to be lower compared to the other parts of both glands. Left adrenalectomy and right partial adrenalectomy were performed, preserving the lower portion of the right adrenal gland. In the excised left and right adrenal glands, the normal parts (green arrows in Fig. 10A) and yellow-colored hyperplasic or nodular parts (* in Fig. 10A) coexisted. The normal parts had a morphologically-clear three-layer structure (Fig. 10B), while the others were histologically hyperplasic (Fig. 10C). CYP11B2-positive cells were hardly detected in the ZG (Fig. 10D), while most cells in the hyperplasic parts were CYP11B2-positive (Fig. 10E). In the normal parts, no KCNJ5 gene mutation was detected (29 of Fig. 10F), while in the hyperplasic parts, a mutation of the KCNJ5 gene was detected not only in CYP11B2-positive hyperplasic parts (30 of Fig. 10G) but also in CYP11B1-positive regions in the hyperplasic parts (31 of Fig. 10G). Some hyperplasic parts were APA-like (Fig. 10H and I), and the mutation of the KCNJ5 gene was also detected in these parts. Both adrenals exhibited very similar histologies to each other. The identical somatic mutation of the KCNJ5 gene was detected from both adrenals (Fig. 10J and K), but it was not present in blood cells from the patient and her parents. These findings suggested that the somatic mutation of the KCNJ5 gene probably occurred during the prenatal period and that both adrenals developed from the mutated and unmutated cells. This patient was thus reported to have non-familial juvenile PA [24].
Delineation of Aldosterone-producing Tissues by Matrix-assisted Laser Desorption Ionization Imaging Mass Spectrometry (MALDI-imaging)
There is a question whether distribution of CYP11B2-expressing cells corresponds to that of aldosterone molecule on adrenal tissue sections. To solve the question we conducted MALDI-imaging analysis, which detect steroids including aldosterone on tissues [26-28]. Using this method, we reported that APCC and pAATL had aldosterone at high concentrations (Cases 1–2 and Case 5 in Fig. 11, respectively) and that some portions in APA gave no signal of aldosterone despite being positive for CYP11B2 (Case 8 in Fig. 11). CYP11B2 is required for aldosterone synthesis, but it is possible that aldosterone is not always synthesized in CYP11B2-positive cells due to lacking of steroid precursors, electron donation, or oxygen supply.
Concluding Remarks
The histological evaluation of aldosterone-producing lesions includes (1) conventional morphological observation by H&E staining; (2) immunohistochemical detection of the aldosterone-synthesizing enzyme CYP11B2; and (3) imaging mass spectrometry for detection of aldosterone molecule. Based on our studies on aldosterone production using human samples for over a decade, the latter two are critical for detecting aldosterone-producing lesions with sufficient specificity.
To describe our findings and hypotheses, we have used the term ‘APCC’ for CYP11B2-positive cell populations that attach to the capsule. Recently, it has been suggested that the term aldosterone-producing (micro)nodule (APN) should be used instead of APCC [29]. However, there is biological rationale for using ‘APCC,’ but not ‘nodule.’ (i) The term ‘nodule’ refers to a bump or swelling in morphology. (ii) Instead, APCC exhibits no neoplastic feature and is indistinguishable from cells in the normal ZG and ZF. (iii) The term ‘adrenocortical (micro)nodule’ was once used mistakenly for aldosterone-producing lesion based on morphological findings and positive HSD3B staining [30], although these ‘adrenocortical (micro)nodules’ were often CYP11B2-negative. From these points of view, the term ‘nodule’ could lead to confusion and misunderstanding to describe aldosterone production.
From the overview of our findings, we presume that APCC gives an important clue to elucidate the pathology of PA. Furthermore, our observation that APCCs occur in most of human adrenal glands with age let us speculate that APCC potentially associates with essential hypertension. Recently, we [31] and others [32] provided deeper molecular biological insights on APCC using latest technologies including single cell RNA sequencing and genome-wide association study. We hope that this review will be useful for future pathological diagnosis and for the research on PA and other related conditions.
Disclosures
The authors declare that they have nothing to disclose.
References
- 1 Ogishima T, Mitani F, Ishimura Y (1989) Isolation of aldosterone synthase cytochrome P-450 from zona glomerulosa mitochondria of rat adrenal cortex. J Biol Chem 264: 10935–10938.
- 2 Ogishima T, Shibata H, Shimada H, Mitani F, Suzuki H, et al. (1991) Aldosterone synthase cytochrome P-450 expressed in the adrenals of patients with primary aldosteronism. J Biol Chem 266: 10731–10734.
- 3 Mitani F, Suzuki H, Hata J, Ogishima T, Shimada H, et al. (1994) A novel cell layer without corticosteroid-synthesizing enzymes in rat adrenal cortex: histochemical detection and possible physiological role. Endocrinology 135: 431–438.
- 4 Ogishima T, Suzuki H, Hata J, Mitani F, Ishimura Y (1992) Zone-specific expression of aldosterone synthase cytochrome P-450 and cytochrome P-45011 beta in rat adrenal cortex: histochemical basis for the functional zonation. Endocrinology 130: 2971–2977.
- 5 Nishimoto K, Nakagawa K, Li D, Kosaka T, Oya M, et al. (2010) Adrenocortical zonation in humans under normal and pathological conditions. J Clin Endocrinol Metab 95: 2296–2305.
- 6 Arnold J (1866) Ein Beitrag zu der feineren Structur und dem Chemismus der Nebennieren. Virchows Arch 35: 64–107.
- 7 Leng S, Pignatti E, Khetani RS, Shah MS, Xu S, et al. (2020) β-Catenin and FGFR2 regulate postnatal rosette-based adrenocortical morphogenesis. Nat Commun 11: 1680.
- 8 Gomez-Sanchez CE, Qi X, Velarde-Miranda C, Plonczynski MW, Parker CR, et al. (2014) Development of monoclonal antibodies against human CYP11B1 and CYP11B2. Mol Cell Endocrinol 383: 111–117.
- 9 Hattangady NG, Olala LO, Bollag WB, Rainey WE (2012) Acute and chronic regulation of aldosterone production. Mol Cell Endocrinol 350: 151–162.
- 10 Nishimoto K, Seki T, Hayashi Y, Mikami S, Al-Eyd G, et al. (2016) Human adrenocortical remodeling leading to aldosterone-producing cell cluster generation. Int J Endocrinol 2016: 7834356.
- 11 Nanba K, Vaidya A, Williams GH, Zheng I, Else T, et al. (2017) Age-related autonomous aldosteronism. Circulation 136: 347–355.
- 12 Beuschlein F, Boulkroun S, Osswald A, Wieland T, Nielsen HN, et al. (2013) Somatic mutations in ATP1A1 and ATP2B3 lead to aldosterone-producing adenomas and secondary hypertension. Nat Genet 45: 440–444, 444e1–2.
- 13 Azizan EAB, Poulsen H, Tuluc P, Zhou J, Clausen MV, et al. (2013) Somatic mutations in ATP1A1 and CACNA1D underlie a common subtype of adrenal hypertension. Nat Genet 45: 1055–1060.
- 14 Scholl UI, Goh G, Stölting G, de Oliveira RC, Choi M, et al. (2013) Somatic and germline CACNA1D calcium channel mutations in aldosterone-producing adenomas and primary aldosteronism. Nat Genet 45: 1050–1054.
- 15 Choi M, Scholl UI, Yue P, Björklund P, Zhao B, et al. (2011) K+ channel mutations in adrenal aldosterone-producing adenomas and hereditary hypertension. Science 331: 768–772.
- 16 Nishimoto K, Tomlins SA, Kuick R, Cani AK, Giordano TJ, et al. (2015) Aldosterone-stimulating somatic gene mutations are common in normal adrenal glands. Proc Natl Acad Sci U S A 112: E4591–E4599.
- 17 Kitamoto T, Kitamoto KK, Omura M, Takiguchi T, Tsurutani Y, et al. (2020) Precise mapping of intra-adrenal aldosterone activities provides a novel surgical strategy for primary aldosteronism. Hypertension 76: 976–984.
- 18 Lifton RP, Dluhy RG, Powers M, Rich GM, Cook S, et al. (1992) A chimaeric 11 beta-hydroxylase/aldosterone synthase gene causes glucocorticoid-remediable aldosteronism and human hypertension. Nature 355: 262–265.
- 19 Nishimoto K, Seki T, Kurihara I, Yokota K, Omura M, et al. (2016) Case report: nodule development from subcapsular aldosterone-producing cell clusters causes hyperaldosteronism. J Clin Endocrinol Metab 101: 6–9.
- 20 Nishimoto K, Umakoshi H, Seki T, Yasuda M, Araki R, et al. (2021) Diverse pathological lesions of primary aldosteronism and their clinical significance. Hypertens Res 44: 498–507.
- 21 Hayashi T, Zhang Z, Al-Eyd G, Sasaki A, Yasuda M, et al. (2019) Expression of aldosterone synthase CYP11B2 was inversely correlated with longevity. J Steroid Biochem Mol Biol 191: 105361.
- 22 Gomez-Sanchez CE, Qi X, Gomez-Sanchez EP, Sasano H, Bohlen MO, et al. (2017) Disordered zonal and cellular CYP11B2 enzyme expression in familial hyperaldosteronism type 3. Mol Cell Endocrinol 439: 74–80.
- 23 Takizawa N, Tanaka S, Nishimoto K, Sugiura Y, Suematsu M, et al. (2021) Familial hyperaldosteronism type 3 with a rapidly growing adrenal tumor: an in situ aldosterone imaging study. Curr Issues Mol Biol 44: 128–138.
- 24 Tamura A, Nishimoto K, Seki T, Matsuzawa Y, Saito J, et al. (2017) Somatic KCNJ5 mutation occurring early in adrenal development may cause a novel form of juvenile primary aldosteronism. Mol Cell Endocrinol 441: 134–139.
- 25 Makita K, Nishimoto K, Kiriyama-Kitamoto K, Karashima S, Seki T, et al. (2017) A novel method: super-selective adrenal venous sampling. J Vis Exp 127: 55716.
- 26 Sugiura Y, Takeo E, Shimma S, Yokota M, Higashi T, et al. (2018) Aldosterone and 18-oxocortisol coaccumulation in aldosterone-producing lesions. Hypertension 72: 1345–1354.
- 27 Takeo E, Sugiura Y, Uemura T, Nishimoto K, Yasuda M, et al. (2019) Tandem mass spectrometry imaging reveals distinct accumulation patterns of steroid structural isomers in human adrenal glands. Anal Chem 91: 8918–8925.
- 28 Zhang Z, Sugiura Y, Mune T, Nishiyama M, Terada Y, et al. (2019) Immunohistochemistry for aldosterone synthase CYP11B2 and matrix-assisted laser desorption ionization imaging mass spectrometry for in-situ aldosterone detection. Curr Opin Nephrol Hypertens 28: 105–112.
- 29 Williams TA, Gomez-Sanchez CE, Rainey WE, Giordano TJ, Lam AK, et al. (2021) International histopathology consensus for unilateral primary aldosteronism. J Clin Endocrinol Metab 106: 42–54.
- 30 Sasano H (1994) Localization of steroidogenic enzymes in adrenal cortex and its disorders. Endocr J 41: 471–482.
- 31 Iwahashi N, Umakoshi H, Seki T, Gomez-Sanchez CE, Mukai K, et al. (2022) Characterization of aldosterone-producing cell cluster (APCC) at single-cell resolution. J Clin Endocrinol Metab 107: 2439–2448.
- 32 Le Floch E, Cosentino T, Larsen CK, Beuschlein F, Reincke M, et al. (2022) Identification of risk loci for primary aldosteronism in genome-wide association studies. Nat Commun 13: 5198.
