Reconstruction from serial sections: A tool for developmental biology. Application to Hox genes expression in chicken wing buds

Abstract Homeobox genes (HOX) encode transcription factors that are frequently deregulated in leukemias. Our previous findings described that HOX gene expression differs among genetically characterized subtypes of pediatric AML with PML-RARa+ patients having the lowest overall HOX gene expression. We observed that HOX gene expression positively correlated with expression of histone 3 lysine 27 (H3K27) demethylases JMJD3 and UTX and negatively with DNA methyltransferase DNMT3b. Interestingly, it has been shown that JMJD3 is a direct target of PML-RARa protein (Martens, JH et al, 2010, Cancer Cell). These findings led us to postulate the hypothesis that reduced levels of HOX genes in PML-RARa+ AML can be caused by the suppressed expression of histone demethylases, such as JMJD3 and UTX, resulting in increased H3K27 methylation and transcription inhibition. We chose PML-RARa+ NB4 cell line to study the role of PML-RARa fusion gene in the regulation of HOX gene expression. To inhibit the effect of PML-RARa we used all-trans retinoic acid (ATRA; 1 uM, 10 uM) which was described to release the block caused by this fusion protein. Expression of particular HOX genes (e.g., HOXA1, HOXA3, HOXA5, HOXA7) together with that of JMJD3 and UTX assessed by qPCR was significantly elevated after ATRA treatment, while gene expression of DNMT3b was decreased. To test whether the reduction in HOX gene expression is directly related to the levels of JMJD3 and UTX, we cultured NB4 cells with a specific inhibitor of these histone demethylases, GSK-J4 (1 uM, 10 uM), in combination with ATRA. This co-treatment led to inhibition of JMJD3 and UTX proteins, followed by significant reduction of HOX genes expression (e.g., HOXA1, HOXA3, HOXA5, HOXA7). This result supports our hypothesis that HOX genes expression is directly related to JMJD3/UTX activity. To determine the effect of ATRA and GSK-J4 on histone marks we have isolated histones by acid extraction and detected the levels of histones by western blot in NB4 ATRA or GSK-J4/ATRA treated cells. We observed that the level of repressive histone methylation mark (trimethylated H3K27; H3K27me3) was decreased after ATRA treatment (activation of JMJD3/UTX) and increased after GSK-J4/ATRA co-treatment (inhibition of JMJD3/UTX). The opposite effect was observed in active histone methylation marks where di- and tri-methylated H3K4 (H3K4me2, H3K4me3) increased after ATRA treatment and decreased after GSK-J4/ATRA co-treatment. H3K9 dimethylated (another repressive histone methylation mark) levels did not change. Next, to investigate the histone code directly in particular HOX genes regions we performed chromatin immunoprecipitation (ChIP) assays. We studied the presence of H3K27me3 and H3K4me2 in 5´UTR genomic region of particular HOX genes (HOXA1, HOXA2, HOXA3, HOXA5, HOXA7) in cells treated with ATRA alone or in the combination with GSK-J4. Preliminary results showed reduction in repressive marks (H3K27me3) upon ATRA treatment, whereas addition of GSK-J4 prevented this decrease. Accordingly, we observed that ATRA/GSK-J4 co-treatment reduced active histone mark H3K4me2. To evaluate the role of DNA methylation in observed expression changes after ATRA treatment we performed bisulfite sequencing of particular promoter sites of HOX genes (e.g., HOXA7, HOXA5). Although we detected decreased DNMT3b gene expression after ATRA treatment there was no change in DNA methylation of CpGs in studied regions. Our results demonstrate that changes in chromatin activity correspond with changes in HOX gene expression. Moreover, ChIP data show direct binding of the modified histones and HOX 5´UTR sites. Our data implicate histone demethylases in regulation of HOX gene expression in PML-RARa+ leukemic blasts. DNA methylation in these particular HOX genes is not involved in the regulation. Elucidating the mechanism of regulation of HOX genes expression can help to understand their role in the leukemogenic process. Supported by GACR P304/12/2214 and GAUK 568213. Disclosures No relevant conflicts of interest to declare.

Download Full-text

Comparative study of some skin and lung fibroblast genes expression of rats of different age

V N Karazin National University Series “Biology” ◽

10.26565/2075-5457-2018-31-15 ◽

2018 ◽

Keyword(s):

Comparative Study ◽

Hox Genes ◽

Protein Product ◽

Fibroblast Growth ◽

Fibroblast Growth Factor 8 ◽

Expression Of Genes ◽

Age Related ◽

Genes Expression ◽

Old Rats ◽

Tissue Characteristics

Comparative study of the expression of three different groups of genes and their protein products amount in the culture of skin fibroblasts from Wistar rats of different ages (2 weeks, 1, 12 and 24 months) was carried out. The traits of similarities and differences in age dynamics for vimentin, vinculin, decorin have been found. These three genes, the products of which participate in intracellular interactions (vimentin) and interactions of cytoskeleton proteins with components of the extracellular matrix, are characterized by an increase in the expression with age both in the skin and in the lungs. They are expressed much stronger in the skin than in the lungs. The amount of their products fluctuates without any single direction. The most active is the expression of vinculin, both in the skin and in the lungs; the amount of the product is also the maximal for it. The least effective is the expression of the vimentin gene in the lungs of two-week-old rats. For the gene expression of fibroblast growth factors 1, 2 and 8, significant differences have been found in their changes in ontogenesis. The first two of them, whose products stimulate the synthesis of one of the most common and important forms of collagen 1, are minimally expressed in old animals, both in the skin and in the lungs. The gene of fibroblast growth factor 8 is expressed in both tissues significantly weaker than the genes of factors 1 and 2. The direction of age-related expression of factor 8 is opposite to that which is inherent for the expression of genes 1 and 2. As for the protein product, its amount is maximal in 1 month, and in the skin significantly increased in the second half of ontogenesis. Some homeobox genes, the HOX genes (2, 4, 5, 6, 7), have been studied. They are most important for the early stages of ontogenesis due to their influence on organogenesis, especially in the embryonic period. In general, both their expression and the product amount decrease, especially in old animals. HOX 5 is most expressed among these genes in the lungs and in the skin. The results obtained are discussed in connection with the functional and tissue characteristics of the studied genes and their products.

Download Full-text

Hox Genes Expression

Encyclopedia of Neuroscience ◽

10.1016/b978-008045046-9.01067-6 ◽

2009 ◽

pp. 1221-1231 ◽

Cited By ~ 2

Author(s):

M. Parrish ◽

C. Nolte ◽

R. Krumlauf

Keyword(s):

Hox Genes ◽

Genes Expression

Download Full-text

HOTAIR, HOXC13, and HOXC10 Are Overexpressed in Gastric Cancer

10.21203/rs.3.rs-834508/v1 ◽

2021 ◽

Author(s):

Mahvash Habibi ◽

Reza Fakhari ◽

Mina Zamani ◽

Hamid Galehdari

Keyword(s):

Gastric Cancer ◽

Significant Relationship ◽

Perineural Invasion ◽

Hox Genes ◽

Cancer Tissue ◽

Tissue Samples ◽

Qrt Pcr ◽

Genes Expression ◽

Carcinoma Metastasis ◽

Positive Correlations

Abstract Gastric cancer is concerned as the second leading cause of tumor-associated death worldwide after lung tumors and is specified as one of the most common malignant cancers. Given the increasing importance of HOX genes in gastric cancer research, this study was aimed at exploring the expression profile of HOTAIR, HOXC13, HOXC10, HOXC13-AS, and HOXC‑AS3 in gastric cancer. To achieve this goal, 30 pairs of tumor and normal margin samples were assessed for these genes expression analyses via qRT-PCR. result we found that, the HOTAIR, HOXC13, and HOXC10 expression was significantly increased in cancer tissue samples in comparison with adjacent normal tissue samples, P<0.01. Moreover, there were significant positive correlations between the expressions of genes studied: HOXC‑AS3 and HOXC10 (r=0.52, P<0.003), HOXC13-AS and HOXC13 (r=0.57, P<0.001), HOTAIR and HOXC‑AS3 (r=0.39, P<0.03), HOTAIR and HOXC13-AS (r=0.36, P<0.05). The HOXC13 and HOXC10 expression exhibited a significant relationship with both distant metastasis and perineural invasion (metastasis: P<0.014, r=0.443 and P<0.0041, r=0.51 perineural invasion: P<0.025, r=0.41 and P<0.0017, r=0.55). The HOXC13-AS displayed a significant relationship with the sex factor so that in females the expression of this gene was higher than males (P<0.030, r=0.4). Our findings suggest that the expression of HOXC genes and HOTAIR lncRNA increased through gastric tumor progression and thus they possibly participate in malignant transformation and gastric carcinoma metastasis.

Download Full-text

Activation of Hottip lncrna Perturbs HSC Function Leading to AML like Disease in Mice

Blood ◽

10.1182/blood-2018-99-110218 ◽

2018 ◽

Vol 132 (Supplement 1) ◽

pp. 3877-3877

Author(s):

Huacheng Luo ◽

Ganqian Zhu ◽

Jie Zha ◽

Bowen Yan ◽

Ying Guo ◽

...

Keyword(s):

Gene Expression ◽

Hox Genes ◽

Chromatin Domain ◽

Active Chromatin ◽

Self Renewal ◽

Hematopoietic Stem ◽

Chromatin Signature ◽

Genes Expression ◽

Hoxa Genes

Abstract Several HOX loci associated long noncoding RNAs (lncRNAs) have been shown to regulate transcription of HOX genes through influencing epigenetic landscape. Especially, the posterior HOXA domain associated lncRNA HOTTIP acts as an epigenetic regulator that recruits WDR5/MLL complex to coordinate active chromatin modifications and HOXA genes expression in the development of animal digits. Despite HOX genes, especially HOXA genes, are highly expressed in many acute myeloid leukemia (AML) patients, it remains largely unknown whether and how HOTTIP lncRNA regulates hematopoietic stem cell (HSC) function and contributes to leukemogenesis. We showed previously that disruption of the CTCF boundary located between HOXA7 and HOXA9 genes (CBS7/9) resulted in reduced lncRNA HOTTIP and HOXA genes expression in MLL rearranged AML suggesting that HOTTIP may play a role in ectopic expression of the posterior HOXA gene. We employed a pooled CRISPR-Cas9 KO library to specifically screen lncRNAs in four HOX gene loci and identify HOTTIP as acritical regulator in controlling oncogenic HOX chromatin signature and associated gene expression patterns in AML by collaborating with posterior HOXA chromatin boundary. HOTTIP is upregulated in AML patients with MLL-rearrangement or NPM1 mutation. AML patients with high HOTTIP expression exhibits a significantly shortened survival compared to low HOTTIP expressing patients. To test whether HOTTIP acts to coordinate posterior chromatin domain and HOXA genes activation in AML, we manipulated HOTTIP lncRNA expression levels in the MLL-AF9 rearranged MOLM13 by loss-of-function KO and gain-of function rescue, as well as carried out genome wide chromatin and transcriptomic analysis to intterrogate the role of HOTTIP in control of AML specific posterior HOXA chromatin domain. We found that knock-out of HOTTIP lncRNA led to a loss of active chromatin structure and invasion of repressive H3K27me3 mark over the posterior HOXA domain. HOTTIP KO attenuated progression of AML in the transplanted AML mouse model resembling the effect of CBS7/9 boundary disruption, while transcriptional activation of HOTTIP lncRNA in the CBS7/9 boundary-disrupted AML cells restored HOXA locus chromatin signature and gene expression as well as reversed the CBS7/9-mediated anti-leukemic effects. To further determine the role of HOTTIP lncRNA in regulating HSC function and leukemogenesis, we generated transgenic mice that expresses Hottip lncRNA under the control of the hematopoietic specific Vav1 enhancer and promoter. The Hottip transgenic (Tg) mice exhibited increased WBC and neutrophil counts and developed splenomegaly indicating that enforced expression of Hottip lncRNA resulted in perturbation of hematopoiesis. Furthermore, overexpression of Hottip lncRNA in mice bone marrow hematopoietic compartment strongly perturbed hematopoietic stem and progenitor cell (HSC/HPC) function by altering self-renewal and differentiation property of HSC/HPCs through affecting homeotic gene associated oncogenic transcription program. Approximately 20% of Hottip lncRNA transgenic mice developed abnormal hematopoietic phenotypes resembling AML-like disease. RNA-seq and ATAC-seq analysis indicated that overexpression of Hottip enhanced promoter chromatin accessibility and stimulates transcription of genes and pathways involved in HSC function and leukemogenesis, including WNT signaling, hematopoietic cell lineage, cell cycle, Hoxa9, Hoxa13, and Meis1, Runx1, and Twist1 genes. Thus, Hottip lncRNA overexpression acts as an oncogenic event to promote HSC self-renewal and HPC proliferation by reprograming leukemic associated chromatin signature and transcription programs. Disclosures No relevant conflicts of interest to declare.

Download Full-text

PS989 ABNORMAL DNA METHYLATION MODIFIES HOX GENES EXPRESSION IN BONE MARROW MESENCHYMAL STROMAL CELLS OF MYELODYSPLASIAS AND DE NOVO ACUTE MYELOID LEUKEMIAS

HemaSphere ◽

10.1097/01.hs9.0000562252.48946.d8 ◽

2019 ◽

Vol 3 (S1) ◽

pp. 444

Author(s):

B. Roux ◽

F. Picou ◽

C. Vignon ◽

C. Debeissat ◽

N. Gallay ◽

...

Keyword(s):

Dna Methylation ◽

Bone Marrow ◽

Mesenchymal Stromal Cells ◽

Stromal Cells ◽

Hox Genes ◽

De Novo ◽

Genes Expression ◽

Myeloid Leukemias ◽

Acute Myeloid

Download Full-text

CALM-AF10 Positive T-ALLs Show a Pattern of Expression Similar to MLL-Translocated Acute Leukemias.

Blood ◽

10.1182/blood.v104.11.1108.1108 ◽

2004 ◽

Vol 104 (11) ◽

pp. 1108-1108

Author(s):

Eric Delabesse ◽

Wim A. Dik ◽

Wajih Brahim ◽

Charlene Braun ◽

Vahid Asnafi ◽

...

Keyword(s):

Hox Genes ◽

Fusion Gene ◽

Insertion Site ◽

Homeobox Gene ◽

Chromosome 11 ◽

Acute Leukemias ◽

Bmi1 Expression ◽

Independent Analysis ◽

Genes Expression ◽

Therapeutic Protocols

Abstract The t(10;11) translocation is recurrent in T-ALL and AML. The AF10 gene on chromosome 10 is rearranged either with MLL or CALM located on chromosome 11. CALM-AF10 fusion gene is found in T-ALLs in immature (IM) and TCRγδ-expressing (TCRGD+) T-ALLs. We compared 6 CALM-AF10+ T-ALLs cases (4 IM, 2 TCRGD+) to 17 CALM-AF10 negative T-ALLs cases (14 IM, 3 TCRGD+) using Affymetrix U133A microarrays. 44 genes were significantly overexpressed in CALM-AF10+ T-ALLs, the most significant being HOXA9, a homeobox gene overexpressed in MLL-translocated acute leukemias (MLL-t AL), BMI1, a polycomb family member whose function in regulation of HOX genes expression is opposite to Trithorax genes (whose MLL belongs), SOX4, a frequent insertion site in retroviral-induced leukemogenesis, SFRS6 and COMMD3 (p≤0.001). Only two other HOX genes, HOXA5 and HOXA10, were significantly increased. 89 genes were significantly underexpressed in CALM-AF10+ T-ALLs, the most significant being GGH, ARL6IP4, NBS1, OGFR and TUBB (p≤0.001). An independent analysis of the expression of HOXA5, HOXA9, HOXA10 and BMI1 genes was done by quantitative RT-PCR in 10 CALM-AF10+ T-ALLs and 27 CALM-AF10 negative T-ALLs. These were compared to 19 MLL-translocated acute leukemias (2 MLL-AF10, 5 MLL-AF4, 3 MLL-AF6, 5 MLL-AF9, 3 MLL-ELL and 1 MLL-ENL), since HOXA9 overexpression had been previously associated with MLL-t AL. HOXA5, HOXA9 and HOXA10 expression were higher in CALM-AF10+ T-ALLs than in CALM-AF10 negative T-ALLs (p<0.001), confirming the microarray results. HOXA5 and HOXA9 expressions in CALM-AF10+ T-ALLs were similar to those detected in MLL-t AL and lower for HOXA10 in CALM-AF10+ T-ALLs as compared to the values of MLL-t AL (p=0.008). BMI1 expression in CALM-AF10+ T-ALLs was significantly higher than in CALM-AF10 negative T-ALLs and MLL-t AL (p<0.001). Additionally, MEIS1 expression was determined as this gene was associated in MLL-t AL with the overexpression of HOXA9. As for BMI1, MEIS1 expression was significantly higher in CALM-AF10+ T-ALLs compared to CALM-AF10 negative T-ALLs and MLL-t AL (p<0.001 and p=0.019, respectively). In summary, we demonstrated here the association between CALM-AF10 in T-ALLs and overexpression of HOXA5, HOXA9, HOXA10, BMI1 and MEIS1 genes. Overexpression of BMI1 is restricted to CALM-AF10+ T-ALLs. Although no obvious similarities are apparent between MLL and CALM proteins, the activation of HOXA and MEIS1 genes represent a highly recurrent pattern of expression in CALM-AF10+ T-ALLs and MLL-t AL. Consequently, the leukemias resulting in the activation of HOXA9 (MLL-t AL, CALM-AF10+ AL, NUP98-HOXA9 AML) should be seen as an independent group of acute leukemias and may benefit from common therapeutic protocols.

Download Full-text

HOX Genes Expression Pattern Is Related to Phenotypical and Genotypical Subgroups but Not to Treatment Response in Pediatric ALL.

Blood ◽

10.1182/blood.v114.22.1288.1288 ◽

2009 ◽

Vol 114 (22) ◽

pp. 1288-1288

Author(s):

Julia Starkova ◽

Blanka Vicenova ◽

Roman Krejci ◽

Harry A. Drabkin ◽

Jan Trka

Keyword(s):

Gene Expression ◽

Expression Pattern ◽

Hox Genes ◽

Conflicts Of Interest ◽

Fusion Gene ◽

Expression Patterns ◽

Hox Gene ◽

Age Related ◽

Genes Expression ◽

Significant Difference

Abstract Abstract 1288 Poster Board I-310 Homeodomain (HOX) genes encode transcription factors important for embryonic development. They are involved in normal hemopoiesis regulation and likely also in leukemogenesis as a result of translocations and other aberrations present in leukemias. In previous work Drabkin et al. demonstrated that HOX gene expression patterns differentiate major cytogenetic groups in acute myeloid leukemias. In this study we focused on HOX gene expression in pediatric acute lymphoblastic leukemias (ALL). We were interested if certain HOX genes or expression pattern could distinguish subpopulations of ALL. We analyzed the expression pattern of 21 HOX genes from HOXA and HOXB clusters and non-cluster HOX genes, CDX1 and CDX2 using qRT-PCR approach. We looked at 54 patients chosen according to phenotypic (T-ALL, BCP-ALL), prognostic (PGR – prednisone good responders, PPR – prednisone poor responders) and genotypic (BCR/ABL, MLL/AF4, TEL/AML1, hyperdiploid) characteristics. Overall analysis comparing all studied groups showed that HOXA7 (Kruskal-Wallis test p=0.000045), HOXA3 (p=0.000098), HOXB3 (p=0.00015), HOXA4 (p=0.000619) and HOXB4 (p=0.001925) genes were differently expressed among groups. Wilcoxon signed-rank test, a non-parametric statistical analysis comparing two groups against each other, showed that HOXA3, A4 and B3 distinguish BCP-ALL (w/o fusion gene) and T-ALL. Interestingly, particular HOX genes expression showed significant difference among the groups: HOXA7 gene is significantly downregulated in hyperdiploid ALL (p=0.03) compared to all other subgroups. Furthermore, HOXB7 gene is specifically upregulated in TEL/AML-positive patients (p=0.0048 vs BCP-ALL w/o fusion gene) and CDX2 is downregulated in BCR/ABL-positive patients (p=0.001 vs hyperdiploid; p=0.006 vs TEL/AML1; p=0.03 vs MLL/AF4). Suprisingly, TEL/AML1-positive patients have similar expression of HOXA1-A4 as T-ALL patients. HOX genes expression pattern seemed to differ in MLL/AF4-positive patients according to the age at diagnosis. Three patients younger than 2 months at presentation clustered together in clear contrast to the MLL/AF4-positive patient diagnosed at the age of 13 years with secALL who presented with very low overall expression of all HOX genes. Next, we looked for diversity and similarity between groups. We determined how many HOX genes were expressed differently (p<0.05) and similarly (p=1.0) between particular ALL subtypes. The most outlying couples were T-ALL vs PPR (11 genes differently expressed), T-ALL vs PGR (9 genes) and T-ALL vs TEL/AML1 (6 genes). In contrast, the closest groups were BCR/ABL vs PPR, MLL/AF4 vs T-ALL and MLL/AF4 vs PPR. Our data demonstrate that BCP-ALL (w/o known fusion gene) can be distinguished from T-ALL by the HOX gene expression (in particular HOXA3, HOXB3, HOXA4). Like in AML, expression pattern differs also among the major cytogenetical subgroups of ALL. On the other hand, within the BCP-ALL subgroup, no expression difference was found between patients with good (PGR) and poor (PPR) response to the initial steroid therapy which is known to be an excellent predictor of outcome. HOX genes of interest emerged from our analysis: low expression of HOXA7 in hyperdiploid ALL, highly expressed HOXB7 in TEL/AML1-positive ALL and specifically downregulated CDX2 in BCR/ABL-positive ALL. Age-related differences in expression in MLL/AF4-positive ALL seem to link the expression pattern rather with the relative maturity of the cell undergoing (pre)malignant transformation than with the specific changes caused by the leukemogenesis itself. This hypothesis must be tested in comparison to the HOX genes expression in sorted subtypes of normal T and B precursors. This work was supported by MSM0021620813, IGA NR/9526 and GACR 301/08/P532. Disclosures No relevant conflicts of interest to declare.

Download Full-text