Varying expression of four genes sharing a common regulatory sequence may differentiate rheumatoid arthritis from ageing effects on the CD4(+) lymphocytes

Abstract

The CD28 gene is similarly down-regulated in CD4(+) lymphocytes from both healthy elderly people and patients with rheumatoid arthritis (RA) because of impaired protein-binding activity of the 'α' sequence in its promoter region. Other genes important for the CD4(+) cell function may share that sequence and may be similarly regulated and affected. We searched GenBank for possible 'α' homologues and then compared transcriptional activities of the respective genes in the CD4(+) cells of young and older healthy individuals and those with RA by real-time PCR. We show here that genes encoding one of the zinc finger proteins (ZNF334), the 'aging hormone' Klotho, the retinoid acid receptor β2 (RARβ2) and the T-cell adapter protein GRAP-2, contain sequences with various (exceeding 70%) degrees of homology to the 'α' sequence near their promoters. These genes are transcribed in human CD4(+) lymphocytes; the expressions of RARβ2, KLOTHO and ZNF334 are significantly decreased in a correlated manner in the cells of patients with RA compared with those of healthy individuals. In RA patients, the extremely reduced expression of ZNF334 does not depend on the individual's age, apparently constituting a disease-related phenomenon; whereas that of RARβ2 and KLOTHO occurs mostly in the cells of relatively younger patients, making them similar to the lymphocytes of healthy elderly in this aspect.

Figure 1

Transcriptional activities of KLOTHO, RARβ2, ZNF334 and GRAP-2 genes in the CD4⁺ cells vary depending on age and healthy/rheumatoid arthritis (RA) status. The genes’ activity was assessed by real-time PCR relative to that of the β-actin gene. Comparison of the expression levels between healthy individuals and those with RA is shown in the left column (a, c, e, g; medians, 25th and 75th percentiles, and the minimum and maximum values obtained from 14 paired experiments; statistical significance was assessed by Mann–Whitney’s U-test). Comparison of the expression levels between young and older healthy individuals and those with RA is shown in the right column (b, d, f, h; bars depict means ± SD; unpaired Student’s t-test used for statistical comparison).

Figure 2

Transcriptional activities of KLOTHO, RARβ2, ZNF334 and GRAP-2 genes in the CD4⁺ lymphocytes are variably correlated. The genes’ activity in the lymphocytes isolated as described in the Materials and methods was assessed by real-time PCR relative to that of the β-actin gene. Correlation coefficients for each combinatorial pair [KLOTHO vs. ZNF334 (a), GRAP-2 vs. ZNF334 (b), RARb2 vs. ZNF334 (c), RARb2 vs. KLOTHO (d), GRAP-2 vs. KLOTHO (e) and GRAP-2 vs. RARb2 (f)] and their statistical powers are shown in the inserts.

Figure 3

‘α’ homologous sites in the ZNF334, RARβ2, KLOTHO and GRAP-2 genes bind with distinctly similar pattern of CD4⁺ T-cell nuclear proteins. For the detailed description of the electrophoretic moblity shift assay experiment see the Materials and methods. ³²P-tagged oligonucleotide probes for ZNF334, KLOTHO, GRAP-2 (but not RARβ2) bind proteins similar to those bound by the ‘α’ sequence (filled solid arrowhead). All four sequences share the ability to bind another nuclear protein complex (open solid arrowhead) from lymphocyte nuclear lysates, whereas RARβ2 and ZNF334 sequences preferentially bind yet another complex of higher molecular weight (dashed arrowhead). Data from one out of three representative experiments yielding similar results are shown.