Nrf expression has been known to be associated
Nrf2 expression has been known to be associated with better outcome for BCa , and it regulates transcription of multiple genes when it gets translocated into the nucleus after its release from inhibitory Keap1 [28,29]. Promoter of cxcl13, but not cxcr5, possesses putative Nrf2 binding sites. We have demonstrated that Nrf2 significantly abrogates the positive inducing effect of RelA for cxcl13 transcription in vitro (Fig. 2). However, we did not witness any significant effect of Nrf2 for cxcr5 transcription (Fig. 2). In “RelA-high” samples, CXCL13 mRNA expression is negatively correlated with Nrf2 expression (Fig. 3A). This is important because high Nrf2 could uncouple co-expression of CXCL13 with CXCR5 in “RelA-high” conditions. Literature suggests that Midostaurin (PKC412) signaling negatively regulates Nrf2 expression and a significant proportion of breast tumors are ER positive . In our study population, we observed higher Nrf2 expression in ER negative samples (Supplementary Fig. S6). Correspondingly, ER positive samples with lower Nrf2 are more prone to co-express CXCL13 and CXCR5.
Interestingly, among “RelA-high Nrf2-low” samples, we also identified tumors where high CXCL13 does not accompanied by high CXCR5, suggesting a possible negative regulation of cxcr5 transcription. Cxcr5 promoter has a CpG-island and we observed lack of CXCR5 expression in “RelA-high” tumors where cxcr5 promoter is methylated, showing a high negative correlation (Fig. 3C). Notably, RelA-induced BCa cells not only produce significantly increased CXCL13, but also secrete it in functionally active form (Fig. 4A–D). Moreover, RelA-induced BCa cells express CXCR5 on the cell surface (Fig. 2D and E). We have also identified regions in cxcl13 promoter for RelA and Nrf2 binding (Fig. 4E and F), and confirmed that both RelA and Nrf2 bind to cxcl13 promoter in the chromosome (Fig. 5A and B). Additionally, occupancy of Nrf2 on cxcl13 promoter remained unaltered upon RelA overexpression as compared to the Vector control, which is evident from the unaltered signals for Amplicon 2 and Amplicon 5 in the Nrf2-immunoprecipitates. Similarly, simultaneous overexpression of Nrf2 with RelA did not alter occupancy of RelA on cxcl13 promoter as compared to the RelA-overexpressed cells, reflected by the unaltered Amplicon 1, Amplicon 3 and Amplicon 6 signals in the RelA-immunoprecipitates (Fig. 5B). These observations suggested that binding of RelA and Nrf2 within the cxcl13 promoter is independent of each other.
The following are the supplementary data related to this article.