IL-8 and CXCR1 Remodel the Vascular Niche to Promote Hematopoietic Stem and Progenitor Cell Engraftment

The microenvironment is an important regulator of hematopoietic stem and progenitor cell (HSC/HSPC) engraftment during development and in recipients of hematopoietic stem cell transplantation (HSCT). Factors secreted by the hematopoietic microenvironment that promote HSC/HSPC engraftment in the developing zebrafish may therefore be therapeutic targets for enhancing HSC engraftment in patients undergoing HSCT. We previously described a novel behavior we called endothelial cuddling in which sinuosoidal endothelial cells of the niche make intimate interactions with stem cells. To find candidate extracellular factors regulating this behavior, gene expression profiling was performed on sorted zebrafish endothelial cells. Gene set enrichment analysis showed that expression of chemokines and TNF family members was significantly enriched in all endothelial cells. The leading edge gene sets included 16 chemokines and chemokine receptors. Thirteen of these genes were used as candidates in a gain-of-function screen to test whether overexpression was sufficient to stimulate the hematopoietic niche in favor of HSC engraftment. High level, global gene expression was induced at 36 and 48 hours post fertilization (hpf) using a heat shock-inducible system. One gene, CXCR1, enhanced HSC/HSPC engraftment when globally overexpressed (p=0.03, N=63). CXCR1 is a specific receptor for the chemokine IL-8/CXCL8 in higher vertebrates. Zebrafish IL-8 was used in similar gain of function experiments and was also sufficient to enhance HSC/HSPC engraftment (p=0.003, N=41). CXCR2 is a promiscuous chemokine receptor for IL-8, Gro-α and Gro-β and did not enhance HSC/HSPC engraftment in this system. To further characterize the effects of CXCR1 on HSC engraftment, it was overexpressed in transgenic zebrafish carrying a stem-cell specific reporter gene, Runx1:mCherry. HSC engraftment in the CHT was enhanced when CXCR1 expression was induced beginning at 36 hpf (3.0 +/- 2.0 vs 7.4 +/- 2.6 HSC per CHT) or 48 hpf (4.3 +/- 1.1 vs 9.4 +/- 3.6 HSC per CHT). Inhibition of CXCR1 signaling from 48 to 72 hpf using the selective CXCR1/2 antagonist, SB225002, decreased HSC engraftment in Runx1:mCherry animals (1.2 +/- 0.39 vs 0.4 +/- 0.2 HSC per CHT, p=0.03). We next hypothesized that overexpression of CXCR1 might also have effects on the endothelial cell niche itself. Using FLK1(VEGFR2):mCherry reporter zebrafish and 3-dimensional reconstruction of the CHT, we found that global overexpression of CXCR1 increased the volume of the endothelial cell niche (2.0 +/- 0.09 x 10⁶ vs 2.4 +/- 0.1 x 10⁶ μm³, p=0.005) while treatment with SB225002 reduced its volume (6.3 +/- 0.3 x 10⁵ vs 4.9 +/- 0.5 x 10⁵ µm³, p=0.04). Finally, we asked if CHT remodeling would still be enhanced if CXCR1 were constitutively expressed only within the endothelial cell niche. FLK1:CXCR1; FLK1:mCherry double transgenic animals had significantly increased CHT volume when compared with FLK1:mCherry single transgenic animals (1.1 +/- 0.05 x 10⁶ vs 1.3 +/- 0.06 x 10⁶ um³, p=0.02). These findings suggest a model whereby HSC/HSPCs actively participate in the remodeling of the endothelial niche via CXCR1/IL-8 in order to promote their own engraftment. Further, they suggest that CXCR1/IL-8 is a potential therapeutic target for enhancing HSC/HSPC engraftment in patients undergoing HSCT.