in Figure 1D–E, respectively) were composed of compact cells and lipid-rich cells but did not fulfill the criteria for adrenocortical carcinoma (Weiss score: 2; Ki-67 proliferation index: 4.0%), leading to the diagnosis of an adrenal adenoma. The adjacent non-tumor portion lost typical adrenocortical zonation in hematoxylin and eosin staining and revealed many cells that contained lipid vacuoles (Figure 1F, Supplementary Figures S3 and S4). CYP11B2 immunohistochemistry confirmed that the tumors were APAs, as CYP11B2 was expressed throughout the tumors (\* and # in Figure 1G) [5]. The cortex of the adjacent non-tumor portion had many CYP11B2-positive cells with irregular arrangement, similar to the adrenal cortices of the previously removed left adrenal gland (Supplementary Figure S5) and as in a previously reported FH3 case [8].

#### *4.2. Production of Aldosterone in FH3 Adrenal*

We performed MALDI-IMS, using SolariX attached with Fourier transform ion cyclotron resonance mass spectrometry (Bruker Daltonics, Billerica, MA, USA), to demonstrate in situ aldosterone production throughout the adrenal, as we previously reported [9]. Aldosterone and cortisone, which share identical mass-to-charge ratio values (m/z), were identified mainly in the subcapsular areas of non-tumor adrenal gland, but not in tumors (Figure 1H), irrespective of strong CYP11B2 expression throughout the non-tumor adrenal gland and adrenal tumors (Figure 1G). The hybrid steroid 18-oxo-cortisol, a steroid marker of aldosterone-producing cells [9], was similarly detected in the subcapsular area only (Figure 1I). To determine the CYP11B2 mRNA levels in various areas of the adrenal, including the APA and adjacent adrenal cortices to the APA, we performed quantitative real-time polymerase chain reaction (qRT-PCR), as previously reported [4,7,11,19]. We confirmed that the CYP11B2 expression levels were not significantly different between the tumor and non-tumor portions (T1 – T3 and N1 – N3, respectively) (∆CT in Supplementary Table S1, *p* = 0.303, Student's *t*-test). In addition, there was no significant difference in the expression level of CYP1B2 between the case, i.e., the average of T1–T3 and N1–N3, and unrelated archived cases of sporadic APA (*n* = 16, Supplementary Figure S6, data of the archived APA cases are shown in Supplementary Table S1). Immunohistochemistry of KCNJ5 was performed in frozen sections obtained from a normal adrenal gland of a renal cell carcinoma patient (left in Supplementary Figure S7A) and FB10 (right), as previously reported [19]. KCNJ5 was detected only in the subcapsular area of the normal adrenal tissue, as previously reported (Supplementary Figure S7B) [19]; whereas, in this case, it was found throughout the adrenal cortex and tumors (Supplementary Figure S7C,D), suggesting that the *KCNJ5* mutation induced KCNJ5 and CYP11B2 co-expression throughout the adrenal cortex and tumors in the patient. Irrespective of high levels of CYP11B2 and KCNJ5, the tumors produced much lower levels of aldosterone and 18oxoF than the non-tumor portion, as shown in Figure 1H,I.

Whole exome sequencing confirmed a germline mutation of *KCNJ5* in T1, N1, and the patient's blood (Materials and Methods). Several somatic mutations were identified in the larger tumor, T1, but not in N1 and blood, which included β-catenin (*CTNNB1*, c.134C > A, p.S45Y), *ADAM17*, *CENPE*, *COL12A1*, *LETM2*, *ALG10B*, and *SRCAP*. Among these mutations, the *CTNNB1* mutation presumably caused rapid tumor growth. Immunohistochemistry confirmed nuclear CTNNB1 expression in the tumor but not in the nontumor portions, suggesting activation of CTNNB1 in the tumor (Supplementary Figure S8). Microarray analyses of T1–T3 (RNA#89 – 91, respectively) and N1–N3 (RNA#86 – 88, respectively) were performed, as previously reported [7], and confirmed that genes of the cell proliferation pathway were upregulated in T1 and T3 (\* in Supplementary Table S2). Except for hydroxysteroid 17-β-dehydrogenase 14 (HSD17B14), which is not associated with aldosterone synthesis, expression of steroidogenic enzymes did not differ between the aldosterone-negative tumors (T1–T3) and non-tumor portions (N1–N3), suggesting that aldosterone production in the non-tumor subcapsular area was controlled by other factor(s) than the steroidogenic enzymes, including CYP11B2.

### *4.3. Cellular Progression in Non-Tumor and Tumor Portions of the Case*

*Curr. Issues Mol. Biol.* **2021**, *1*, FOR PEER REVIEW 7

The non-tumor adrenal gland was hyperplastic (Figure 2A,B) and harbored mitotic cells (yellow arrowhead in Figure 2C). To assess the cell cycle progression status of the adrenal cells of the patient, we compared the Ki-67 index [20] between the non-tumor portions, the larger tumor (\* in Figures 1 and 2), smaller tumor (# in Figure 1), and archived sporadic APAs and their adjacent adrenal sections ("adjacent") in cases APA #8, 10–17, and 20–25 (*n* = 15) in Supplementary Table S1. It is noteworthy that APA #9 was removed from the analysis because the case harbored a non-APA tumor (sample name: KS-APA\_9\_T2, Supplementary Table S1). Non-tumor portions were analyzed using two parts each from FFPE blocks #4, 10, and 14 (*n* = 6, Supplementary Figure S4). The larger tumor was analyzed using two parts each from FFPE block #9, 10, and 14 (*n* = 6). The smaller tumor was analyzed using four parts from FFPE block #8 (*n* = 4). Upon comparing these five groups (Kruskal– Wallis one-way analysis of variance on ranks, followed by post hoc comparison with Dunn's methods), we found that the larger tumor (3.20 [interquartile range: 2.84–3.70] unit) had a higher index than the APAs (0.47 [0.39–0.77] unit, *p* = 0.001) and their adjacent adrenal tissue (0.45 [0.36–0.70] unit, *p* < 0.001) (Figure 3). Interestingly, the non-tumor portion of the patient showed a higher Ki-67 index (1.09 [0.94–1.48] unit) than the adjacent adrenal tissue in sporadic APAs (*p* = 0.047). These results suggest that chronic stimulation from the mutated *KCNJ5* channel and/or high aldosterone concentration around the cells might be associated with increased cell cycle progression and/or second hit mutations in genes, including *CTNNB1*. The non-tumor adrenal gland was hyperplastic (Figure 2A,B) and harbored mitotic cells (yellow arrowhead in Figure 2C). To assess the cell cycle progression status of the adrenal cells of the patient, we compared the Ki-67 index [20] between the non-tumor portions, the larger tumor (\* in Figures 1 and 2), smaller tumor (# in Figure 1), and archived sporadic APAs and their adjacent adrenal sections ("adjacent") in cases APA #8, 10–17, and 20–25 (*n* = 15) in Supplementary Table S1. It is noteworthy that APA #9 was removed from the analysis because the case harbored a non-APA tumor (sample name: KS-APA\_9\_T2, Supplementary Table S1). Non-tumor portions were analyzed using two parts each from FFPE blocks #4, 10, and 14 (*n* = 6, Supplementary Figure S4). The larger tumor was analyzed using two parts each from FFPE block #9, 10, and 14 (*n* = 6). The smaller tumor was analyzed using four parts from FFPE block #8 (*n* = 4). Upon comparing these five groups (Kruskal–Wallis one-way analysis of variance on ranks, followed by post hoc comparison with Dunn's methods), we found that the larger tumor (3.20 [interquartile range: 2.84–3.70] unit) had a higher index than the APAs (0.47 [0.39–0.77] unit, *p* = 0.001) and their adjacent adrenal tissue (0.45 [0.36–0.70] unit, *p* < 0.001) (Figure 3). Interestingly, the non-tumor portion of the patient showed a higher Ki-67 index (1.09 [0.94–1.48] unit) than the adjacent adrenal tissue in sporadic APAs (*p* = 0.047). These results suggest that chronic stimulation from the mutated *KCNJ5* channel and/or high aldosterone concentration around the cells might be associated with increased cell cycle progression and/or second hit mutations in genes, including *CTNNB1*.

**Figure 2.** Adrenal histology of the case. The non-tumor adrenal portions were hyperplastic (panels (**A**) and (**B**)) and harbored mitotic cells (yellow arrowhead in panel (**C**)). \*: the larger tumor (\* in Figure 1). **Figure 2.** Adrenal histology of the case. The non-tumor adrenal portions were hyperplastic (panels (**A**) and (**B**)) and harbored mitotic cells (yellow arrowhead in panel (**C**)). \*: the larger tumor (\* in Figure 1).

**Figure 3.** Comparison of Ki-67 index among APA cases (*n* = 15 each) and index case. \* *p* < 0.05.

### **5. Discussion**

This is a case of FH3 with unusual tumor development and histological/molecular findings. The patient was initially diagnosed with PA at the age of 6 years, and her adrenals were removed at the age of 15 years (left adrenal, to alleviate hyperaldosteronism) and at the age of 27 years (right adrenal, due to an enlarging tumor on CT). Clinical data of this patient from birth to 14 years old, including those of kidney biopsy at 10 years old, were described in a preceding article [10], and those from the age of 14 years are provided in Supplementary Table S3. Comprehensive pathological and molecular analyses of the removed right adrenal and blood resulted revealed FH3. The findings can be summarized as follows: (i) an abnormal cellular arrangement of CYP11B2-positive cells throughout the adrenal gland, similar to observed in a previously reported FH3 case [8]; (ii) limited localization of aldosterone production, primarily in the non-tumorous sections, with widespread and very strong CYP11B2 expression in the tumor areas; (iii) rapid tumor growth, which may represent an early stage of adrenocortical carcinoma, caused presumably by second hit mutation of *CTNNB1* (p.S45Y) in *KCNJ5* mutated cells of the larger tumor. The presence of a β-catenin mutation causing constitutive activation of adrenal cell growth has been shown to induce adrenal hyperplasia and adrenal cancer development in mice [21].

There were several peculiar aspects in this case. The pathological findings were remarkably similar to those of previously reported cases [22,23], but with the development of adrenal tumors. The *KCNJ5* mutation, absent in the parents of the patient, represents a de novo mutation. In this case, the larger adrenal tumor also had a second mutation of *CTNNB1,* which led to accelerated cellular growth. Furthermore, even though the patient underwent a bilateral adrenalectomy, serum aldosterone was still detectable and within the normal range, suggesting the presence of extra-adrenal adrenocortical cells harboring the *KCNJ5* mutation, causing aldosterone production [24].

While CYP11B2 staining has been used as a marker for aldosterone production, in the current case, in situ imaging of aldosterone revealed that aldosterone was only found in the non-tumorous areas of the distorted adrenal, which is an additional peculiar finding. Microarray and qRT-PCR studies revealed that all steroidogenic enzymes responsible for aldosterone synthesis were present in sufficient quantities to lead to the increased production of aldosterone in both the tumorous and non-tumorous areas. However, aldosterone production did not occur in the tumorous areas. It has been reported that *KCNJ5* mutations result in depolarization of the adrenal cells with stimulation of calcium mobilization and calmodulin phosphorylation, resulting in transcriptional induction of steroidogenic enzymes [1]. The fact that the tumors exhibited increased levels of steroidogenic enzymes suggests that there is an additional deficiency (or deficiencies) in the steps required for aldosterone biosynthesis in this case. For the synthesis of aldosterone, in addition to transcription of steroidogenic enzymes, cholesterol transporters to mitochondria, such as steroidogenic acute regulatory protein (StAR), are also needed. In addition, StAR must be phosphorylated to exert its action. We speculate that phosphorylation of StAR was repressed, as microarray analysis revealed that StAR expression was not downregulated (Supplementary Table S2).

Although some authors have reported that a treatment with mineralocorticoid receptor antagonist controls hyperaldosteronism well in FH3, most cases in FH3 needed bilateral adrenalectomy for management of severe hyperaldosteronism [1,23]. We had held off a decision of bilateral adrenalectomy because we preferred to avoid lifelong glucocorticoid replacement therapy and the risk of adrenal crisis from bilateral adrenalectomy, if possible. However, that led to the chronic kidney disease and, ultimately, maintenance dialysis at an early age. Notably, this case suggested that chronic stimulation from *KCNJ5* mutation and/or severe hyperaldosteronism contribute to tumorigenic transformation. Therefore, we believe that FH3 patients should undergo bilateral adrenalectomy as soon as possible before irreversible organ damages occur, if mineralocorticoid receptor antagonists cannot control patients' blood pressure.

#### **6. Conclusions**

This is an interesting case of a de novo germline mutation of the *KCNJ5* gene that resulted in severe hyperaldosteronism with serious target organ damage, resulting in endstage renal disease. Steroidogenic and molecular studies of the adrenal gland demonstrated discrepancies that need further investigation.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/ 10.3390/cimb44010010/s1, Figure S1: The sections after formaldehyde fixation, Figure S2: Pedigree of the index patient and the results of Sanger sequencing, Figure S3: The original microscopic imaging data, Figure S4: HE and IHC of FFPE tissue sections, Figure S5: HE and IHC of FFPE tissue sections of previously resected adrenal gland, Figure S6: CYP11B2 expression, Figure S7: *KCNJ5* immunostaining, Figure S8: CTNNB1 immunostaining, Table S1: Clinical data, qPCR data, and *KCNJ5* mutation analysis of the index case, the parents of the index case, and other APA cases, Table S2: Microarray data, Supplementary Table S3: clinical course from 15 years old.

**Author Contributions:** Conceptualization, N.T., S.T., K.N., T.S., C.E.G.-S. and T.M.; formal analysis, N.T., S.T., K.N., Y.S., M.S, Y.M., K.M. and K.O.; investigation, N.T., S.T., K.N., Y.S., M.S, C.O., H.O., Y.M., K.M., T.S., K.O., C.E.G.-S. and T.M.; resources, N.T., S.T., K.N., Y.S., M.S, C.O., H.O., Y.M., K.M., K.O., C.E.G.-S. and T.M.; data curation, N.T., S.T., K.N., Y.S., M.S., C.O., Y.M., K.M. and K.O.; writing—original draft preparation, N.T., S.T., K.N., Y.S., C.O., K.M. and K.O.; writing—review and editing, N.T., S.T., K.N., Y.S., M.S., C.O., H.O., K.M., T.S., K.O., C.E.G.-S. and T.M.; visualization, N.T., S.T., K.N., Y.S., C.O., K.M. and K.O.; supervision, K.N., M.S, T.S. and T.M.; project administration, K.N., M.S, T.S. and T.M.; funding acquisition, K.N., M.S. and C.E.G.-S. All authors have read and agreed to the published version of the manuscript.

**Funding:** K.N. was supported by JSPS KAKENHI, grant number 18K09205. The infrastructure of imaging metabolomics was partly supported by a grant from Shimadzu Corporation. Please note, this corporation did not contribute to the study design and data analyses; thus, there are no conflicts of interests to declare. K.N. was also partly supported by a grant from the Ministry of Health, Labor, and Welfare, Japan (20FC1020). Dr. Celso E. Gomez-Sanchez was supported by the R01 HL144847 grant, from the National Heart, Lung, and Blood Institute, the 1U54GM115428 grant from the National Institute of General Medical Sciences, and the BX004681 grant from the Department of Veteran Affairs.

**Institutional Review Board Statement:** The study was conducted according to the guidelines of the Declaration of Helsinki. In addition, comprehensive molecular analyses were performed under approval of the institutional review boards of Saitama Medical University International Medical Center (approval #18-308), Keio University School of Medicine (#20090018), and Kansai Medical University (#2016902).

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author. The data are not publicly available due to privacy and ethical concerns.

**Acknowledgments:** The authors are grateful to the patient and her parents for their cooperation and participation in this study. The authors would like to thank the Department of Pathology and Internal Medicine II at Kansai Medical University Hospital for their contribution to the diagnosis, and Masakazu Kohda (Intractable Disease Research Center, Graduate School of Medicine, Juntendo University, Tokyo, Japan) for the contribution to analyze the sequence data. This study was also supported in part by Shimadzu Corporation, Kyoto, Japan.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

