How to cite this article: López-Romero R, Marrero-Rodríguez D, Romero-Morelos P, Villegas V, Valdivia A, Arreola H, Huerta-Padilla V, Salcedo M. The role of developmental HOX genes in cervical cancer. Rev Med Inst Mex Seguro Soc. 2015;53 Supl 2:S188-93.
Received: October 22nd 2014
Accepted: May 15th 2015
Ricardo López-Romero,a Daniel Marrero-Rodríguez,a Pablo Romero-Morelos,a Vanessa Villegas,a Alejandra Valdivia,a Hugo Arreola,a Víctor Huerta-Padilla,a Mauricio Salcedoa
aUnidad de Investigación Médica en Enfermedades Oncológicas, Hospital de Oncología, Centro Médico Nacional Siglo XXI, Instituto Mexicano del Seguro Social, Distrito Federal, México
Communication with: Ricardo López-Romero
Telephone: (55) 5627 6900, extensión 22705
Cervical cancer (CC) is a multifactorial disease associated to genetic, environmental and epigenetic factors, being the infection by human papillomavirus the main etiologic agent. Additionally, the alteration in the expression of transcription factors has been considered of importance for the development of this tumor. HOX genes encode a group of transcription factors involved in cellular proliferation and differentiation processes during the development of embryonic structures in vertebrates; their aberrant expression is associated with tumorigenesis and metastasis. A range of evidence suggests a role for HOX genes in the development of cervical neoplastic cell. Studies in CC cell lines, primary tumors and premalignant lesions have suggested the involvement of HOXA1, HOXC5, C6, C8 and C10, HOXD9 and HOXD13 in the process of cervical carcinogenesis. Also, the de novo expression of genes HOXB2, B4, B13 and HOXC11-C13 appears to be involved in the process of malignant transformation of cervical epithelial cell. These data would allow to open a field in search of new molecular markers in cervical cancer and the development of new therapeutic strategies for this malignancy.
Keywords: Homeobox genes, Uterine cervical neoplasms.
Currently, cancer is one of the most common diseases worldwide, and one of the leading causes of death in the population, being the second leading cause of death in our country alone.1 Cancer is not caused by a single cause, because multiple factors operate in its generation; this is why it is said to be a multifactorial disease, which makes it a serious public health problem (international and national) that can affect any person at any stage of life.
Cancer is the result of the accumulation of multiple alterations in genes that regulate cell growth, cell division, proliferation, etc., bringing other morphological and metabolic alterations in cells that present it.
Specifically, cervical cancer (CC) and its precursor lesions are a problem that affects the female population and that is becoming ever more common in women who started their sexual life at less than 18 years old, who are mostly characterized by having multiple sexual partners, indiscriminate use of hormonal contraceptives, inadequate hygiene habits, multiple pregnancies, drug abuse, sexually transmitted infections- mainly by the human papillomavirus (HPV)- and zero routine gynecological examination.
CC probably starts with precancerous lesions, also known as cervical intraepithelial neoplasia, and the human papilloma virus or HPV is considered the main etiological agent of premalignant lesions and invasive tumors of the human cervix.2 However, the development of cervical cancer is a multifactorial process that cannot be explained simply by the single infection with specific types of HPV. Events such as alteration in the expression of transcription factors could constitute additional steps in cervical carcinogenesis.
In the fruit fly (Drosophila melanogaster) there is a group of genes called homeotic complex or HOM-C, containing gene groups Antennapedia and Bithorax.3,4 These homeotic genes are involved in the development of individual segments along the anterior-posterior axis of the fly embryo. Antennapedia genes control the identity of the segments of the head and thorax of the embryo, while Bithorax genes control segments of the thorax and abdomen (Figure 1). Homeotic genes of fruit flies have counterparts in mammals and are known as HOX genes.6
Figure 1 Genomic organization and similarity of HOM complex of Drosophila and HOX groups of humans. Schematic representation of Drosophila homeotic complex (HOM-C), the four human HOX clusters, and a hypothetical ancestral homeotic complex are shown with their possible phylogenetic relationships. Each column of genes indicates the correspondence, based on the homology of the homeobox sequence, between HOM complex and the four HOX groups of mammals. The numbers 1 to 13 indicate the paralogous genes identified so far. The numbers in parentheses indicate the chromosomes on which the human HOX groups are located. Each gene is represented by a colored box. Expression domains of the HOM/HOX genes are outlined in the fly and in the CNS of a human fetus (extrapolated from data in mice). Each color in the boxes describes the expression domain of each HOX group in the anterior-posterior axis (taken and modified from Mark et al.).5 Lab = labial; pb = proboscipedia; Df = Deformed; Scr = Sex combs reduced; Antp = antennapedia; U = Ultrabithorax; Abd-A = abdominal-A; Abd-B =Abdominal-B
These genes play a critical role during embryonic development; they constitute a family of transcription factors that control the pattern of anterior-posterior formation and participate in cellular proliferation and differentiation processes.4,7 HOX genes contain a sequence of 183 nucleotide pairs called homeobox, which encodes a protein of 61 amino acids, called homeodomain.
The homeodomain of HOX genes is able to bind to specific DNA sequences in its target genes8 and regulate their expression. The homeoproteins (those containing homeodomain) can therefore act as transcription factors.
Since its initial description, 39 HOX genes have been identified in mammals and classified into four groups labeled A, B, C, and D9 (Figure 1). These genes are located in humans in chromosomal regions 7p15.3, 17q21.3, 12q13.3, and 2q31, respectively.10,11 Each HOX group contains between 9 and 11 aligned genes in 13 paralogous groups according to the similarity their nucleotide and amino acid sequences and their position in the chromosome (Figure 1). The order of genes within each HOX complex is essentially the same as in the Drosophila HOM complex, suggesting that the four full groups of vertebrates originated by duplication of one primary complex, which has preserved its basic organization.12,6
In humans (HOX) and mice (Hox), these genes are involved in controlling the differentiation of structures along the rostro-caudal axis of the embryo, and their expression is regulated under the rules of space-time collinearity, matching their position along the chromosome; that is, the genes located at the 3’ end of the chromosome are expressed early in the more anterior structures of the embryo body, whereas those genes located in the 5' end of the chromosome are expressed late in the more posterior regions of the embryo (Figure 1).4 Thus, 5’ genes (paralogs 9-13) are involved in the differentiation of the urogenital structures of the lumbosacral region.
Besides embryonic structures, HOX genes are expressed in various adult tissues, where they are apparently involved in a variety of biological pathways including homeostasis, cell differentiation, and maintenance of organ function;13 however, this role has not been fully defined.
HOX genes have also been involved in neoplastic processes; in fact, the aberrant expression of HOX is typically associated with oncogenesis and may vary according to histologic type and stage of progression of neoplasia, including metastasis. Different cancer types show different changes in the expression of HOX, whether by increased expression when this is usually low in normal tissues, or as a de novo expression when it does not exist in healthy tissues.14 Numerous studies have shown deregulated expression of HOX genes in cancers in different human tissues, including lung, prostate, breast, colon, bladder, and thyroid,15 as well as ovarian cancer. In summary, it is suggested that some HOX genes can be key at specific stages of cellular differentiation or maintenance of differentiated cells, and that deregulation of these may be important in cellular transformation or tumor progression.16
The development and maintenance of cell identity are vital to the function of adult tissues and key to this is the establishment of a stable cell transcriptional state. Proper regionalization of the female reproductive tract to give identity to the cervix, uterus, and vagina uses a fine homeostasis between signaling molecules and transcription factors; HOX genes play a crucial role in this regard.
As mentioned above, the arrangement of the various HOX genes on the chromosome determines the spatiotemporal expression that these have along segments of the embryo of organisms, so that, in the case of genitourinary structures, the HOX members involved are those located at the 5' end, that is, those that have late expression and expression in the caudal region. In this regard, paralogous genes 9 to 13 (Abdominal-B or Abd-B type) are those involved in the development of structures that give rise to the end organs. According to Taylor et al.,17 a coordinated and conserved axis is presented from which the reproductive system of the human and mouse is developed, in which HOXA cluster expression is involved. According to the authors, the HOXA9 gene is expressed in the oviduct, HOXA10 in the uterus, HOXA11 is expressed in the cervix, while the expression of HOXA13 is restricted to the more posterior structures, thus its expression is weak in the cervix but strong in the vagina. HOX expression occurs primarily in developing structures, in which differentiation is taking place; and a similarity may occur in adult genital tissues, in which cellular and differentiation changes take place during the course of the hormonal cycle. In fact, it has been suggested that expression of HOXA10, HOXA11, and HOXA13 may have an important role in the plasticity of reproductive tissues during the menstrual cycle.17 Apparently HOXA1018,19 and HOXA1120 are important in the development of the endometrium and myometrium under sexual steroid hormone regulation21,18,20 and are essential for implantation in mice. Tests with deletions directed at Hoxa1022 and Hoxa1123 have been shown to induce infertility in mice.
Additionally, HOXA13 mutations have been reported in women with hand-foot-genital syndrome (HFGS), in which a partially or completely divided uterus is developed due to incomplete fusion of the Mullerian tubes.24
Embryonic tissues follow a collinear pattern of HOX expression in time and space; however, tissues with cancer lose this characteristic, and may present expression of genes of the 5' end (paralogs 9-13), the 3' end (paralogs 1-4), or the middle (paralogs 5-8).
Most studies have focused on determining the expression of HOX genes in invasive cancer and their comparison with that of healthy cervical epithelium (Table I). From this it has been suggested that individual members of the four HOX clusters may be involved in transforming a normal cervical cell into a neoplastic cervical cell. In this regard, some reports have shown expression of HOXD9,25 HOXC5 and C8,26 HOXC10 and D1327 in neoplastic cervical cells in culture but not in normal cells. Shim et al.28 reported by cDNA microarray analysis in cervical cancer and cervical epithelium cells that HOXA1 may be involved in cervical tumorigenesis.
|Table I Description of HOX genes reported in human cervical cancer|
|HOX Gene||Tissue or cell line||Alteration||Detection||Reference|
|A1||Squamous cell carcinoma||Overexpression||Expression microarrays||28|
|C5, C8||SiHa cells||Expression de novo||RT-PCR||26|
|D9||HeLa cells||Expression de novo||RT-PCR||25|
|A1, B2, B4, C5,
|CC cell lines||Expression de novo||RT-PCR||27|
|C6, C10||Invasive carcinoma||Overexpression||Expression microarrays||29|
|B2, B4, B13||Invasive carcinoma||Expression de novo||RT-PCR||31|
|B13, C9, C11,
C12, C13, D9, D10
|Invasive carcinoma||Expression de novo||RT-PCR||30|
|C10||Invasive carcinoma||Expression de novo||Expression microarrays||34|
|B4||HISL, Invasive carcinoma||Expression de novo||Mass spectrometry/IHC||32|
|B7||CC cell lines||Deregulation||RT-PCRq /WB||35|
|HSIL = high-grade squamous intraepithelial lesion; CC = cervical cancer; RT-PCRq = quantitative RT-PCR; IHQ = immunohistochemistry; WB = western blot|
Through analysis of expression by RT-PCR, Li et al.25 showed that HOXD9 is differentially expressed in cervical cancer cells but not in normal cervical cells, and they suggest that HOXD9 could be involved in the pathogenesis of cervical cancer.
Alami et al.26 performed expression assays by RT-PCR and reported that out of 39 HOX genes, only HOXC5 and C8 are expressed in transformed SiHa cells but not in primary cervical keratinocytes in culture; the authors extended their analysis to two other cell lines of transformed keratinocytes of independent origin, Eil-8 and 18-11S3 cells, and found the same results, so they suggest that HOXC5 and C8 could be involved in the process that directs transformation of cervical keratinocytes.
The HOXC cluster has been reported as one of the most active in cervical cancer, because overexpression or de novo expression of some of its members have been shown in invasive tumor tissues. Thus, apparently an overexpression of HOXC6 and C10 occurs in invasive cervical carcinomas compared to normal cervical keratinocytes,29 while de novo expression of HOXC9, C11-C13 has been reported in invasive cervical cancer.30 Similarly, HOXB cluster members seem to be present only in invasive cervical carcinomas but not in healthy epithelium, which has led to the suggestion that some of these members may be related to the process of malignant transformation of cervical epithelial cells, as has been reported for HOXB2, B4, and B13.27,31
Despite this, little is known about the role of these genes in the neoplastic process of the human cervical cell; however, some data have shown and suggested specific functions in this process. Barba et al.32 demonstrated by mass spectrometric analysis and IHC that HOXB4 was differentially observed in cervical cancer relative to normal tissue. The HOXB4 protein was immunodetected in the nucleus of epithelial cells of invasive tumors and preinvasive lesions, whereas it was absent from normal cervical epithelium. These findings suggest that HOXB4 could be a protein linked to the neoplastic state of the human cervical epithelium and could eventually be a marker of undifferentiated cells.
Experimental data have shown that HOXB7 could be one of the orchestrators of angiogenesis in the invasion process in CC. Using different experimental strategies, it was determined that miRNA-196b is a transcriptional regulator of HOXB7, which in turn induces expression of VEGF (vascular endothelial growth factor). Abatement of HOXB7 expression by siRNA, and its regulation through miR-196b, resulted in reduced cell growth, clonicity, migration, and in vitro invasion, as well as reduced angiogenesis (one of the hallmarks of cancer) and in vivo tumor cell proliferation. The deregulation of this pathway was significantly associated with poor disease-free survival in patients with CC treated with chemoradiation. Thus, the authors suggest that the miR-196b~HOXB7~VEGF route plays an important role in the progression of cervical cancer and that this pathway could be targeted for the development of new therapeutic strategies for future management of this neoplasm.
Mouse gene mgl-1, analogous to the tumor suppressor "lethal l(2) giant larvae (l(2)gl)" of D. melanogaste, is a target gene of Hoxc833 and HOXC8 and is actively expressed in neoplastic cells of CC;26 however, a similar function for HOXC8 in human CC has not been demonstrated yet. If there are tumor suppressor genes regulated by this or other HOX genes in the human cervix, the loss of these tumor suppressors in cancer tissue, probably due to their inactivation by HOX genes, could be an important event in the development of cervical cancer.
Through various molecular approaches in vitro and in vivo, Zhai et al.34 have shown that HOXC10 is one of the important genes in the process of cell invasion for a preinvasive cervical lesion to progress to invasive carcinoma. The absence of HOXC10 in the healthy epithelium, mild expression in preinvasive lesions, and strong expression in invasive carcinomas, as well as in Matrigel assays, support the authors’ position that HOXC10 is one of the leaders of the invasion process in the CC.
Evidence shows that other pathways related to cancer and in particular cervical cancer are changes in homeotic HOX genes. The fine gene regulation of some HOX by microRNAs shows the complexity of the tumor cell. Thus, the molecular basis of cancer is increasingly changing largely due to the new findings of genes that are potentially related to the process of cellular transformation. Thus, HOX genes could be considered as cancer genes.
The human cervix is a tissue that presents tissue plasticity during the hormonal cycle and apparently some Abd-B-type members of HOX genes are involved in the development and maintenance of adult urogenital structures. An alteration in the regulation of various HOX genes could be partly responsible for the processes that drive the transformation of a healthy cervical cell into a neoplastic cervical cell, so researching this in CC and determining its target genes are of great interest for a better understanding of the pathophysiology and so to eventually open a new field for the development of therapeutic strategies for future management of this neoplasm considered a worldwide public health problem.
We appreciate the support given by the Red de Investigación en Virus del Papiloma Humano of the Coordinación de Investigación en Salud of the Instituto Mexicano del Seguro Social for the creation of this work, especially to doctors Fabio Salamanca, Israel Grijalva, Maria Elena Furuya, Maria Elena Galvan, and Eduardo Almeida, for their ceaseless work and their crucial support.
Conflict of interest statement: The authors have completed and submitted the form translated into Spanish for the declaration of potential conflicts of interest of the International Committee of Medical Journal Editors, and none were reported in relation to this article.