Potential Transcriptional Biomarkers to Guide Glucocorticoid Replacement in Autoimmune Addison's Disease

Åse Bjorvatn Sævik; Anette B. Wolff; Sigridur Björnsdottir; Katerina Simunkova; Martha Schei Hynne; David William Peter Dolan; Eirik Bratland; Per M Knappskog; Paal Methlie; Siri Carlsen; Magnus Isaksson; Sophie Bensing; Olle Kämpe; Eystein S Husebye; Kristian Løvås; Marianne Øksnes

Disclosures

J Endo Soc. 2021;5(3) 

In This Article

Results

Step 1: Gene Expression in Response to 100 mg Hydrocortisone

Infusion of 100 mg HC to medication-fasting AAD patients had a profound short-term effect on the expression FKBP5, MMP9, and DSIPI; increased gene expression levels were evident after 2h and had further increased at 4h. At 6h, the expression levels stabilized at an upregulated level compared to before the HC infusion (Figure 1). Log2 fold change was 0.429 (P < 0.001, FDR < 0.001) for FKBP5, 0.560 (P < 0.001, FDR < 0.002) for MMP9, and 0.274 (P < 0.001, FDR < 0.002) for DSIPI (S1).

Figure 1.

Gene expression of FKBP5 (a), MMP9 (b), and DSIPI (c) before and up to 6 hours after intravenous infusion of 100 mg hydrocortisone.

Individual serum cortisol and plasma ACTH levels are listed in Table 2. At 2h, there was a strong positive correlation between serum cortisol levels and the gene expression levels of DSIPI (r = 0.709, P < 0.022) and MMP9 (r = 0.673, P < 0.033), but not for FKBP5 (r = 0.442, P < 0.200). No significant associations were found between ACTH and gene expression levels (data not shown).

Step 2: Gene Expression in Normal Versus Very Low Morning Cortisol Levels

Eight of the 93 investigated genes were significantly different in normal compared to very low cortisol levels (Figure 2). DDIT4 (z = −3.05, P < 0.002), CEBPB (z = −2.84, P < 0.005), DSIPI (z = −2.82, P < 0.007), FKBP5 (z = −2.79, P < 0.005), and VDR (z = −2.67, P < 0.008) revealed increased expression, while ADARB1 (z = −3.84, P < 0.001), POU2F1 (z = −2.79, P < 0.005), and ARID5B (z = −2.64, P < 0.008) were decreased. The effect size was largest for ADARB1 (r = 0.52), followed by DDIT4 (r = 0.42), CEBPB (r = 0.39), FKBP5 (r = 0.38), POU2F1 (r = 0.38), DSIPI (r = 0.37), VDR (r = 0.36), and ARID5B (r = 0.36). Fold change was greatest for DDIT4 (md = 2.32 [0.34, 10.00]) and FKBP5 (md = 2.03 [0.43, 6.28]) (Figure 3).

Figure 2.

Genes with significant change in expression levels in very low cortisol (VLC) compared with normal cortisol (NC) levels. The black circles and black squares mark the median values for VLC and NC, respectively, and the whiskers mark the range.

Figure 3.

Gene expression fold change in normal cortisol relative to very low cortisol. The black circles mark the median fold change and the whiskers mark the range.

As expected, median serum cortisol was significantly higher and plasma ACTH significantly lower at normal cortisol exposure compared with very low cortisol (cortisol; 277 nmol/L [7,467] vs 3 nmol/L [0.2, 32], P < 0.0001, ACTH; 7.3 pmol/L [1.1, 276] vs. 123 pmol/L [1.1, 278], P < 0.0001, respectively). At normal cortisol levels, serum cortisol correlated negatively with plasma ACTH (r = −0.435, P < 0.030) and positively with DDIT4, FKBP5, and DSIPI (r = 0.450, P < 0.021; r = 0.419, P < 0.033; and r = 0.400, P < 0.043, respectively). At very low cortisol levels, no correlations were found for serum cortisol and gene expression levels, but plasma ACTH negatively correlated with gene expression levels of DDIT4 and FKBP5 (r = −0.411, P < 0.033 and r = −0.399, P < 0.039, respectively).

Expression levels of DSIPI, FKBP5, and DDIT4 strongly correlated with each other at both normal and very low cortisol levels (DSIPI and FKBP5: r = 0.705, P < 0.001 and r = 0.706, P < 0.001; DSIPI and DDIT4: r = 0.695, P < 0.001 and r = 0. 702, P < 0.001; FKBP5 and DDIT4: r = 0.573, P < 0.001 and r = 0.692 P < 0.002, respectively). Other significant correlations were found between CEBPB and DSIPI (r = 0.564, P < 0.001 and r = 0. 584, P < 0.002) and CEBPB and FKBP5 (r = 0.522, P < 0.005 and r = 0.521, P < 0.005) at normal and very low cortisol levels, respectively.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....