Fig 2: Salt-Induced CYP3A5*3 Protein Expression, Enzyme Activity and Mineralocorticoid Receptor Signaling in HEK293 Cells. (a) Confluent HEK293 cells were exposed for 48 h to (A) control media or media supplemented with (B) 10 mM KCl, (C) 50 mM KCl, (D) 100 mM KCl, or (E) 100 mM NaCl. Two protein bands attributed to CYP3A5 were detected via western blot, including the full-length reference protein (CYP3A5WT) at 56 kilodaltons (kDa) and a truncated 30 kDa splice variant form (CYP3A5sv). The housekeeping gene GAPDH, detected at 37 kDa, was used to normalize sample loading. (b) Total CYP3A5 protein and the ratio of CYP3A5WT to CYP3A5SV protein normalized to GAPDH expression was determined using densiometric analysis in ImageJ. (c) CYP3A enzyme activity was monitored in HEK293 cells under normal conditions and with elevated KCl (12 mM; 48 h.) and PMOs (1 µM; 48 h.), as measured fluorometrically by BFC metabolism (see “Methods”). KCl supplementation significantly induced enzyme activity (p < 0.01; t-test) in HEK293 cells (KCl) compared to control cells (CON). The G4 disrupting PMO (G4) and the CYP3A5 start-site inhibitor (AUG) PMO both significantly prevented KCl induction of BFC metabolism (p < 0.01 compared to KCl treated control) and were not different from control cells. The scrambled (SCR) PMO control did not inhibit KCl induction of CYP3A activity (ANOVA p value < 0.01. Post test p < 0.01 = **, and p < 0.005 = ***). (d) LC MS/MS analysis of HEK293 cell media demonstrated KCl enrichment (12 mM, 24 h) rapidly stimulated the conversion of cortisol to 6ß-hydroxycortisol. An ~ sixfold change in the metabolic ratio was observed for HEK293 cells grown in normal media (1.93 × 10-3, solid black) compared to 12 mM KCL-supplemented media (1.19 × 10-2, solid gray). (e) Conditioned media from HEK293 cells (+ /- KCl (12 mM); 48 h.) were added to a luciferase-based cell reporter system for mineralocorticoid receptor (MR) transactivation (IB00501-32; Indigo Biosciences). MR activity was significantly elevated > two-fold (p < 0.01; t-test) in HEK293 media treated with 12 mM KCl. The ligand-specific sensitivity of the MR activity assay was validated using a standard curve for aldosterone (1 nM sensitivity = ~ 100,000 RLU).

Fig 3: Schematic Diagram of CYP3A5's Dual Role in the Regulation of Hypertension. Environmental factors (e.g., diet, chemical exposure, and stress) and homeostatic signaling mechanisms (i.e., Renin–Angiotensin–Aldosterone System (RAAS)) can alter renal and adrenal gland function to promote hypertension in humans. There is increasing evidence that renal CYP3A5 activity plays a dual role in modulating vitamin D- and glucocorticoid-hormone signaling cascades that drive HBP pathology. First, CYP3A5 activity can convert stress-related glucocorticoids (corticosterone and cortisol) to mineralocorticoids via 6ß-hydroxylation, which enhances MR signaling and triggers sodium retention (via several transporter related mechanisms) and increased ROS production. This activity also silences glucocorticoid signaling (via ACTH and NFkB) and increases inflammation. Second, CYP3A5 can hijack endogenous vitamin D hormone signaling cascades by shunting 25-hydroxyvitamin D3 metabolism from CYP27B1 (1a-hydroxylase) towards a 4ß-hydroxylation pathway that limits 1,25-dihydroxyvitamin D3 signaling via the VDR, reducing renin production that suppresses systemic RAAS activity. Thus, CYP3A5 appears to play a pathological endogenous role in many human polymorphs by regulating positive feedback loops that drives the accumulation of stress hormones and sodium in the renal proximal tubule cells. Sodium cationic-stabilization of G4-like elements in CYP3A5 intron 3 may exacerbate this hypertensive signaling arc in CYP3A5 *1/*3 and *3/*3 polymorphs, who may conditionally express the gene. This salt-sensitive positive feedback loop may mask any adaptive effects of the 6986A/G polymorphism (rs776746) by promoting stress- and dietary-sensitive expression of CYP3A5, complicating any GWAS studies focused on the *1/*3 or *3/*3 genotypes.

Fig 4: G4 Structures Regulate CYP3A5 Expression and Represent Novel Targets for ASO Modulation. (a) Schematic diagram of the CYP3A5 gene showing the location of a putative G4 motif in Intron 3. Cation-specific stabilization of native G4 structure shifts the splice acceptor back to the reference exon 4 splice acceptor, creating an in frame CYP3A5*3/*3 transcript that bypasses pseudo-exon introduction and suppression of wild-type protein translation via PTC or NMD-related mechanisms. Salt-sensitive splice correction of *3 transcripts (shown in Fig. 2a) can be abrogated using the G4 disrupting PMO (Table 1). (b) Computational analysis of the CYP3A5 gene (see Supplemental Fig. 2) and RNA/DNA gel shift assays (shown here), confirmed the existence of a G4-like structure in the CYP3A5 intron 3 region. Control lanes 1 and 2 depict the mobility of the 151 nucleobase, CYP3A5 intron 3 “C-strand” DNA, in KCl-enriched running buffer [100 mM KCl]. The C-strand does not display faster migration typical of a G4 structure in the presence of KCL, and a G4 disrupting PMO does not alter C-strand mobility (lane 2). In contrast, the 151 nucleobase “G-strand” DNA demonstrated enhanced mobility bands reflecting both inter- and intra-strand G4 formation in KCl-enriched running buffer (lane 3). Addition of the G4 disrupting PMO diminishes the enhanced mobility bands at low concentration (1:1 molar ratio; lane 4) and ablates the enhanced mobility bands at high concentration (5:1 molar ratio; lane 5). (c) A 151 nucleobase RNA strand also forms a putative G4-structure in KCl-enriched running buffer (Lane 1). Addition of increasing amounts of G4 disrupting PMO to the G4 RNA (Lane 2 – 1:1 PMO:RNA ratio, Lane 3 – 3:1 PMO:RNA ratio, and Lane 4 – 5:1 PMO:RNA ratio) shifts the RNA structure to the slower mobility form. (d) qPCR analysis of CYP3A5 mRNA from HEK293 cells exposed to no PMO and normal media (HEK control), scramble PMO (1 µM, 48 h.) and KCl (12 mM) (HEK + Scr + KCl), G4 disrupting PMO only (1 µM, 48 h.) (HEK + G4 Disrupt), and G4 disrupting PMO (1 µM, 48 h.) and KCl (12 mM) (HEK + G4 Disrupt + KCl). Significant statistical differences were monitored by ANOVA; p value < 0.01. Post test p < 0.01 = **. (e) (Top) A schematic illustrating the location of the EJC in relation to the cryptic splice acceptor in intron 3 and the G4 structure. PMOs designed to skip exon 4 in CYP3A5 are also shown (-17 = E4SA-17, -9 = E4SA-9, -3 = E4SA-3, and + 94 = E4SA + 94). (Bottom) Endpoint PCR results for RNA expression in HEK293 cells treated with three PMOs (E4SA-3, E4SA-9, or a scramble control (Scr); 1 µM; 48 h.) and 100 mM NaCl are shown. The Scr. oligomer did not induce stable transcripts, while exon 4 skipping PMOs (E4SA -3 and -9) induced transcripts that both include and exclude exon 4. The addition of high salt media diminished the cell’s ability to express transcripts including exon 4. Cationic-specific stabilization of the transcript may have improved the ability of exon skipping PMOs to properly bind the transcript near the EJC assembly zone and promote targeted exon 4 exclusion. Molecular drawings were generated using The PyMOL Molecular Graphics System, Version 1.3, Schrödinger, LLC.

Fig 5: The CYP3A5*3 SNP Regulates Gene Splicing via a Salt-sensitive Mechanism that can be Modulated with ASO in Renal Cell Culture Models. (a) Schematic for the CYP3A5*3/*3 polymorphism found in HEK293 cells. The CYP3A5*3-SNP creates a cryptic splice acceptor within intron 3 that results in the aberrant splicing of a pseudo-exon containing a PTC that targets the mRNA transcript for degradation by NMD. Both the cryptic splice acceptor (CAG/TA) and the exon 4 splice acceptor (CAG/AA) are preceded by canonical U2 snRNP branch point cis-acting elements. (b) Endpoint PCR results monitoring exon 4 inclusion in mature CYP3A5 mRNA transcripts demonstrate the role of the *3-SNP in altering splicing and transcript stability. Wild-type CYP3A5 mRNA transcripts encoding exon 4 are virtually undetectable (no band) in HEK293 cell lysates under normal conditions, however when the *3-SNP site is masked by interactions with an antisense oligomer (CYP3A5*3 PMO; 3 µM; 48 h) transcript stability is restored, leading to enhanced detection of exon 4 inclusion via endpoint PCR. (c) Quantitative PCR (qRT-PCR; SYBR green) results demonstrating a 1.5-fold increase in total CYP3A5 mRNA when HEK293 cells were treated with the CYP3A5*3 PMO (1 µM; 48 h). (d) Quantification of the formation of 1-OH-midazolam from midazolam by LC/MS/MS after treatment with either 3A5*3 PMO or AUG PMO for 6 days in primary human PTECs. CYP3A5 activity was increased in PTECs treated with 3A5*3 PMO but not AUG PMO (**p < 0.01; t-test). Efficiency of restoration of CYP3A5 activity by the 3A5*3 PMO was donor-dependent (Donor 3; See Supplemental Table 1), and was not related to cellular toxicity, as PMO treatments did not alter normal PTEC morphology (see Supplemental Fig. 9).