Transcriptome analysis reveals differentially expressed transcripts in rat adrenal zona glomerulosa and zona fasciculata

Abstract

In mammals, aldosterone is produced in the zona glomerulosa (zG), the outermost layer of the adrenal cortex, whereas glucocorticoids are produced in adjacent zona fasciculata (zF). However, the cellular mechanisms controlling the zonal development and the differential hormone production (i.e. functional zonation) are poorly understood. To explore the mechanisms, we defined zone-specific transcripts in this study. Eleven-week-old male rats were used and adrenal tissues were collected from zG and zF using laser-capture microdissection. RNA was isolated, biotin labeled, amplified, and hybridized to Illumina microarray chips. The microarray data were compared by fold change calculations. In zG, 235 transcripts showed more than a 2-fold up-regulation compared to zF with statistical significance. Similarly, 231 transcripts showed up-regulation in zF. The microarray findings were validated using quantitative RT-PCR and immunohistochemical staining on selected transcripts, including Cyp11b2 (zG/zF: 214.2x), Rgs4 (68.4x), Smoc2 (49.3x), and Mia1 (43.1x) in zG as well as Ddah1 (zF/zG 16.2x), Cidea (15.5x), Frzb (9.5x), and Hsd11b2 (8.3x) in zF. The lists of transcripts obtained in the current study will be an invaluable tool for the elucidation of cellular mechanisms leading to zG and zF functional zonation.

Fig. 1.

Adrenal sections before and after LCM A, Hematoxylin-eosin staining of rat adrenal gland section showing capsule (c), zG (g), zF (f), zR (r), and medulla (m). B, Adrenal section after the capture of zG. A cresyl violet-stained section that had undergone the laser-capture was later stained with hematoxylin and eosin for clear presentation. C, Adrenal section after the capture of zF. A, B, and C were serial sections. Scale bar, 100 μm.

Fig. 2.

Microarray scatter plot showing pooled data from four zG and four zF samples. Each spot indicates a unique probe with a total of 22,523 independent transcripts based on normalized microarray data. Dots outside the parallel lines represent transcripts with greater than 2-fold differences in expression.

Fig. 3.

Heat map representation of microarray analysis for the 25 transcripts with the highest differential expression in zG vs. zF. Arbitrary signal intensity acquired from microarray analysis is represented by color (see color bar). Fold change was calculated as average of normalized zG signal intensities divided by that of normalized zF signal intensities. Predicted transcripts are indicated by an asterisk.

Fig. 4.

qPCR analysis for the transcripts with the highest differential expression in zG vs. zF. A, Fold change in each transcript. Cyp11b2 and Smoc2 were not detectable (ND) in all (rats 1–4) zF samples, preventing calculation of their fold differences. Rgs4 was detectable only in two zF samples (rats 2 and 3); therefore, the fold changes of the transcript were calculated from two pairs of zG/zF data. Results are shown as mean ± sem based on four independent experiments with duplicate wells in each experiment. B and C, qPCR curves illustrating mean ΔRn against PCR cycle number for transcripts of ND in zF (Cyp11b2 and Smoc2). ΔRn represents the normalized reporter signal (Rn) minus the baseline signal.

Fig. 5.

Immunohistochemistry for the four most highly up-regulated transcripts in zG. Each panel shows a representative image of immunohistochemistry for CYP11B2 (A), RGS4 (B), SMOC2 (C), and MIA1 (D). These figures show clear histological zonation of zG (g), zF (f), zona reticularis (r), and medulla (m) according to nuclear staining using hematoxylin. Arrowheads in D indicate a subset of fasciculata cells that are stained with MIA1. Scale bar, 100 μm.

Fig. 6.

Heat map representation of microarray analysis for the 25 transcripts with the highest differential expression in zF vs. zG. Arbitrary signal intensity acquired from microarray analysis is represented by color (see color bar). Fold change was calculated as average of normalized zF signal intensities divided by that of normalized zG signal intensities. Predicted transcripts are indicated by an asterisk. Hypothetical transcripts are indicated by two asterisks.

Fig. 7.

qPCR analysis for four transcripts with differential expression in zF vs. zG. Ddah1, Cidea, Frzb, and Hsd11b2 were selected based on the availability of antibodies for immunohistochemistry. A, Fold change in each transcript (zF vs. zG). Ddah1 was not detectable (ND) in any of the zG samples (rats 1–4). Hsd11b2 was detectable in only three zG samples (rats 1, 3, and 4); therefore, the fold change was calculated from three pairs of zF/zG data. B, qPCR curves illustrating ΔRn against PCR cycle number for transcripts of ND in all zG samples (Ddah1). ΔRn represents the normalized reporter signal (Rn) minus the baseline signal.

Fig. 8.

Immunohistochemistry for highly up-regulated transcripts in zF. Each panel shows a representative image of immunohistochemistry for DDAH1 (A), CIDEA (B), FRZB (C), and 11βHSD2 (D). These figures show clear histological zonation of zG (g), zF (f), zR (r), and medulla (m) according to nuclear staining using hematoxylin. Scale bar, 100 μm.