**4.1. HOXA3**

molecular events favoring the implantation of the embryo. In this stage, the role of several intercellular mediators has been implicated, which include specific cytokines, growth factors, adhesion molecules, lipid mediators, steroid hormones, and Hox transcription factors [25]. Like in endometrial tissue, *HOXA10* also plays a role during embryo implantation as it has been shown that despite the fact that *HOXA10*-deficient mice (*HOXA10−/−*) exhibited normal ovulation cycle, the implantation did not occur. Interestingly, implantation was restored when embryos from *HOXA10−/−* were transferred to wild-type mice; however, wild-type embryos were not implanted in *HOXA10−/−* female mice [18], suggesting that *HOXA10* is required to have an adequate implantation environment. Moreover, *HOXA10−/−* and *HOXA11−/−* mice also exhibit poor implantation due to insufficient development of stromal glandular tissue and decidua during pregnancy [26]. In humans, the expression of both *HOXA10* and *HOXA11* genes rises gradually during the proliferative phase of the menstrual cycle, showing a peak of expression in mid-cycle, when implantation typically occurs [13, 27]. Interestingly, this peak of expression was not observed in women with endometriosis or in mice with induced endometriosis [13, 27], suggesting that HoxA10 and HoxA11 peaks require a healthy endometrium to support and continue with the implantation process. Several studies have shown that Hox10 not only promotes implantation directly but also inhibits detrimental factors such as empty spiracles homeobox 2 (EMX2), P300/ CBP-associated factor (P/CAF), and gamma-aminobutyric acid (GABA). Studies by Taylor and colleagues demonstrated that HoxA10 repressed EMX2 expression, which in turn inhibited the proliferation of endometrial cells [28], suggesting that HoxA10 is a pro-proliferative and proimplantation factor in these cells. Zhu and colleagues demonstrated that HoxA10 repressed the promoter activity of P/CAF, which impairs endometrial receptivity and embryo implantation by downregulating integrin β3 [29]. Recent studies have also shown that HoxA10 decreased mRNA levels and protein translocation of GABA receptor [30], which plays a role in the generation of uterine contractions and labor [31]. Thus, the quiescent uterus is required for adequate implantation and embryo development, along with reduced expression or activity of GABA receptor.

32 Endothelial Dysfunction - Old Concepts and New Challenges

Hox genes are highly expressed in hematopoietic stem cells (HSC) and immature progenitor cells [32]; however, this expression is gradually decreased upon cell differentiation. Moreover, overexpression of genes from the *HOXA* cluster impairs B and T lymphocyte differentiation, affects erythropoiesis, and reduces stem cell bone marrow homing, favoring the induction of myeloproliferative disorders and leukemias [33]. In fact, overexpression of *HOXA1*, *HOXA4*, and *HOXA6* genes has been shown to favor the generation of permanent cell lines [34]. Studies by Wang et al. showed increased proliferation and higher self-growth and self-renewal of hematopoietic stem progenitor cells (HSC) (Line 9 and Line H1) when HoxA6 was overexpressed compared to normal conditions [34]. The authors observed that overexpression of this gene sustained HSC self-renewal and multipotency by promoting mature erythroid lineage

Another gene involved in this process is *HOXA5*. Overexpression of HoxA5 in HSC isolated from umbilical cord blood, bone marrow [35], or mice [36] promotes a significant shift toward myeloid differentiation in relation to erythroid differentiation when compared to respective control cells [35, 36]. Then, the authors evaluated genes affected by HoxA5, and they observed downregulation of several genes involved in cell proliferation, differentiation, and metabolism [35, 36].

**3.3. Hematopoiesis**

cells and partial apoptosis of erythroid progenitors.

The *HOXA3* gene is required for modeling the anterior body plan during embryogenesis, but they can also play a role in promoting angiogenesis [46, 47]. It has been shown that activation of *HOXA3* favors the migration of endothelial cells and keratinocytes, associated with increased expression of urokinase-type plasminogen activator receptor (uPAR) in

(EphB4) by *HOXA9*, since previous reports [52, 53] showed that Eph receptors were homeobox protein potential targets. Then, they decided to study EphB4 since it was specifically associated with angiogenesis and cell migration processes [54, 55]. After elegant experimentation and analysis, they conclude that HoxA9 regulated endothelial cell migration and tube formation by promoting the expression of EphB4. Later in 2012, Zhang and colleagues established that HoxA9 was essential for postnatal neovascularization in vivo. In addition, they found that HoxA9 was able to regulate the expression of endothelial genes such as endothelial nitric oxide synthase (eNOS), vascular endothelial growth factor receptor 2 (VEGFR2), and VE-cadherin in vitro in mature endothelial cells exposed to "shear stress" [56]. Furthermore, the *HOXA9−/−* mouse model showed a reduced number of circulating endothelial progenitor cells (EPCs) as well as reduced overall postnatal neovascularization after ischemia compared to wild-type mice. Altogether,

Hox Genes in Adult Tissues and Their Role in Endothelial Cell Differentiation and Angiogenesis

http://dx.doi.org/10.5772/intechopen.76745

35

these results demonstrated that HoxA9 is critical for postnatal neovascularization [57].

The central function of the placenta is to allow the formation of a vascular labyrinth, a juxtaposed series of finely branched blood vessels and trophoblast that regulate the exchange of nutrients and residues while maintaining the separation of maternal and fetal blood supplies. The study by Shaut et al. showed a morphological alteration in the labyrinth endothelial cells, branching of the vessels, and in the integrity of the vessels when *HOXA13* was dysfunctional [58, 59]. These findings suggest that *HOXA13* regulates a number of genes in the vascular endothelium required for vessel adhesion and branching, providing a functional explanation of the mean gestational lethality exhibited by *HOXA13* mutant mice. The same authors identified that EphA6 and EphA7 were direct transcriptional targets of *HOXA13* in the genital tubercle vascular endothelia [59]. Altogether, these findings provide a new genetic pathway to consider when placental pathologies or placental evolutionary ontogeny are characterized. Evidence for this coordination is observed in the labyrinth endothelium, where the genes required for cell adhesion and vascular branching are affected concomitantly by the loss of *HOXA13* function, including Neuropilin-1, Enpp2, Lyve1, Caveolin-1, Foxf1, and Tie2, resulting in reduced levels of provascular factors required for the vascular development of the labyrinth [58].

Besides HoxA genes, the HoxB and HoxD loci have also been involved in endothelial and angiogenesis regulation processes [60]. HUVECs, for example, express several genes from these loci [7], and it has been shown that some of these genes inhibit in vitro proliferation of HUVECs, whereas others have been associated with increased capillary morphogenesis and vasculogenesis [61].

Previous studies have revealed an overlap between HoxA1 and HoxB1 functions during the specification of the rhombomeres, a transiently divided segment of the developing neural tube, from which neural crest cells emerge. It has been demonstrated that both HoxA1 and HoxB1 functions are required for the heart development [62, 63]. HoxB1−/− embryos were previously described as embryos with normal pharyngeal arch arteries and cardiac neural crestderived tissue remodeling [64]. However, more recently, Roux et al. observed one HoxB1 mutant embryo with an aortic arch artery defect, which is characteristic of a developmental failure of the left pharyngeal arch arteries (PAA) [65]. These data suggest that *HOXB1* is

**4.3. HOXA13**

**4.4. HOXB1**

**Figure 2.** HOX genes modulate the expression of crucial target genes to promote the differentiation of mature endothelial cells. Hox genes promote the differentiation of endothelial progenitor cells, which exhibit an immature phenotype (CD70+ CD34+ Oct-4+ ), into mature endothelial cells that express endothelial nitric oxide synthase (eNOS), vascular endothelial growth factor receptor 2 (VEGFR2 or KDR), CD31, von Willebrand factor (vWF), and the lectin-type oxidized LDL receptor 1 (LOX-1). To promote this phenotype, some Hox genes upregulate crucial genes such as fetal liver kinase 1 (Flk1), angiopoietin 2 (ANG2), ephrin type-B receptor 4 (EphB4), and FI3K receptor, whereas other Hox genes downregulate other factors such as hypoxia-induced factor type 1α (HIF1α), cyclooxygenase-2 (cox-2), ephrin type-a receptor 1 (EphwA1), and VEGFR2.

both in vitro and in vivo studies using mice [46, 47]. Similar results were demonstrated by Hansen et al. who confirmed that *HOXA3* is a potent inducer of angiogenesis in vivo and also promotes direct keratinocyte migration [48]. These results suggest that *HOXA3* potentiates two key processes involved in efficient wound repair: angiogenesis and reepithelialization [46, 48]. Gene transfer studies of *HOXA3* suggest that this gene also functions as a potent inducer of wound repair in genetically modified diabetic animals. A single application of protein HoxA3 resulted in complete healing of wounds after 42 days, while wounds treated with the control plasmid without *HOXA3* (β gal) required 77 days for complete tissue repair. In addition, it was demonstrated that secreted protein HoxA3 or HoxA5, coming from respective genes and derived from composite skin constructs, exhibits decreased expression of CCL-2 and CxCL-12 inflammatory mediators, which play a key role in the attraction of monocytes, macrophages, and other wound immune cells [48]. Thus, reduced recruitment of leukocytes mediated by *HOXA3* may contribute to the prolonged integrity and viability of the composite skin constructs expressing *HOXA3*, by reducing inflammation during wound healing process. Taken together, the combined actions of HoxA3 on endothelial cells and keratinocytes lead to increased angiogenesis, normal epidermal differentiation, reduced expression of inflammatory mediators, and reduced graft contraction. These effects suggest that HoxA3 may have therapeutic benefits in wound repair by improving the integrity of composite skin grafts.

#### **4.2. HOXA9**

The *HOXA9* gene code for two different proteins, HA-9A and HA-9B isoform A (HA-9A) and HoxA9 protein isoform B (HA-9B) [49] that share a common homeodomain [15]. The expression of HA-9A has been observed exclusively during fetal development, whereas the HA-9B has been found not only in fetal but also in adult organism and specifically in endothelial cells [49, 50].

In 2004, Bruhl et al. showed that *HOXA9* was able to regulate angiogenesis [51]. These authors using human umbilical vein endothelial cells (HUVECs) with sense/antisense oligonucleotides or siRNA for this gene observed that *HOXA9* expression was essential for endothelial cell migration and tube formation. Also, they evaluated the regulation of ephrin type-B (Eph) receptor B4 (EphB4) by *HOXA9*, since previous reports [52, 53] showed that Eph receptors were homeobox protein potential targets. Then, they decided to study EphB4 since it was specifically associated with angiogenesis and cell migration processes [54, 55]. After elegant experimentation and analysis, they conclude that HoxA9 regulated endothelial cell migration and tube formation by promoting the expression of EphB4. Later in 2012, Zhang and colleagues established that HoxA9 was essential for postnatal neovascularization in vivo. In addition, they found that HoxA9 was able to regulate the expression of endothelial genes such as endothelial nitric oxide synthase (eNOS), vascular endothelial growth factor receptor 2 (VEGFR2), and VE-cadherin in vitro in mature endothelial cells exposed to "shear stress" [56]. Furthermore, the *HOXA9−/−* mouse model showed a reduced number of circulating endothelial progenitor cells (EPCs) as well as reduced overall postnatal neovascularization after ischemia compared to wild-type mice. Altogether, these results demonstrated that HoxA9 is critical for postnatal neovascularization [57].
