- Email: [email protected]
Expression of DHX32 in lymphoid tissues
Experimental and Molecular Pathology 79 (2005) 219 – 223 www.elsevier.com/locate/yexmp
Expression of DHX32 in lymphoid tissues Zaman Alli, Michael Ho, Mohamed Abdelhaleem * Division of Haematopathology, Department of Paediatric Laboratory Medicine, Room 3691, Atrium, Hospital for Sick Children, 555 University Avenue, Toronto, ON, Canada M5G 1X8 Received 11 July 2005 Available online 21 September 2005
Abstract DHX32 is a novel putative RNA helicase identified based on its downregulation in acute lymphoblastic leukemia. DHX32 gene has 12 exons, alternative usage of exons 1 and 2 results in the expression of two transcripts that differ in their 5V untranslated region (UTR), consistent with the involvement of two different promoters. In addition, exon 5 skipping results in the production of a full-length and alternatively spliced transcripts. We used transcript-specific primers in reverse transcriptase (RT)-PCR to show that the thymus and spleen express all of DHX32 transcripts. We also used immunohistochemistry to study the expression of DHX32 protein. The expression pattern of DHX32 protein in normal lymphoid tissues is variable but suggests association with state of lymphocyte activation and/or differentiation. Dysregulated expression of DHX32 is noted in some types of lymphoma compared to their normal counterpart. D 2005 Elsevier Inc. All rights reserved. Keywords: RNA helicase; DHX32; Lymphoma; Gene expression
Introduction RNA helicases are enzymes that utilize the energy derived from nucleotide triphosphate (NTP) hydrolysis to modulate the structure of RNA molecules (Tanner and Linder, 2001). Helicases participate in all biological steps that involve RNA, including transcription, splicing, transport, translation and decay (de LaCruz et al., 1999). Structurally, they are classified into families based on the sequence of the helicase domains, which consist of seven to eight conserved motifs (Luking et al., 1998). We have reported the two largest human RNA helicase families, namely, DDX and DHX families (Abdelhaleem et al., 2003). Two general features have been recognized for RNA helicases, including dysregulation of the expression in various types of cancer (Abdelhaleem, 2004a,b) and involvement in differentiation (Abdelhaleem, 2005). We have identified DDX32 as a novel putative RNA helicase with unique structural features and found it to be downregulated in acute lymphoblastic leukemia (Abdelhaleem, 2002). The gene designation was changed to DHX32 according to the new nomenclature of RNA helicases to reflect its homology to the
* Corresponding author. Fax: +1 416 8136257. E-mail address: [email protected] (M. Abdelhaleem). 0014-4800/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.yexmp.2005.07.002
DHX family (Abdelhaleem et al., 2003). Although the exact biological function of DHX32 is yet to be determined, its structure suggests a role in RNA metabolism (Abdelhaleem, 2002). In this report, we examined the expression of DHX32 in normal and neoplastic lymphoid tissues. Materials and methods Reverse transcription (RT)-PCR The expression of various DHX32 transcripts was determined by RT-PCR of total RNA by using the following primers: primers for exon 1 transcript: forward primer 5V-GTG CCC CGA CGT CGC CGA C-3V and reverse primer 5VGCT GGT AGT GGA TGGA AAG-3V, corresponding to positions 111 – 129 and 798 – 816, respectively of Genbank entry BC06471, and resulting in a 687bp product. Primers for exon 2 transcript: forward 5V-CGG AAC CCA CCA AAC TTT AAA CAC CAG CCT-3V and reverse 5V-CCA TCG CTG GAA TCC AGG GAT TCA GGA AAA-3V, corresponding to positions 57 – 86 and 542 – 571, respectively, of Genbank entry NM_018180, resulting in a 514-bp product. Primers for full-length and D exon 5 transcripts: forward 5V-CGT GCC TGT CAT AGA AGT-3V and reverse 5V-GTC ATG TCT TTG GAG GCA-3V, corresponding to positions 978 – 995 and 1509 – 1526 of Genbank entry AF427340 resulting in a 548- and 305-bp products for full-length and D exon 5 transcripts, respectively. Tissue RT-PCR expression was performed at The Center for Applied Genomic (TCAG) at the Hospital for Sick Children Research Institute. The PCR conditions were as follows: an initial 95-C for 5 min followed by 35 cycles of 95-C for 30 s, 48-C for 30 s and 72-C for 3 min and a final 72-C for 10 min.
220
Z. Alli et al. / Experimental and Molecular Pathology 79 (2005) 219 – 223
Fig. 1. Schematic representation of DHX32 gene and its transcripts. (A) DHX32 gene has 12 exons that span 60 kbp (Genbank entry GI:15391100). Exon 1 occurs in a genomic sequence identified to have a CpG island indicative of a potential promoter, the distal promoter. The genomic sequence 5V of exon 2 is predicted to have a TATA-containing promoter, the proximal promoter. (B) DHX32 transcripts: two transcripts, exon 2 and exon 1 transcripts that differ at the 5V end result from the alternative usage of either the entire of exon 2 or exon 1 instead of the 5V end of exon 2, respectively. The start ATG codon is not affected by the difference at the 5V end. An additional transcript in which exon 5 is deleted is identified; however, the 5V end of this transcript has not been characterized.
Immunohistochemistry The protocol for studying the expression of DHX32 in lymphoid tissues from anonymous archival human cases was approved by the Hospital for Sick Children Research Ethics Board. The following tissues were used for analysis: thymus, tonsil, spleen, follicular lymphoma (3 cases), mantle cell lymphoma (3 cases), large B-cell lymphoma (4 cases), Burkitt lymphoma (4 cases) and nodular sclerosis Hodgkin disease (6 cases). Five micrometers of formalinfixed, paraffin-embedded tissue sections was mounted on positive charged microscope slides. Tissue sections where baked over night at 60-C, dewaxed in xylene and hydrated to distilled water through decreasing concentrations of alcohol. Immunohistochemical procedure for DHX32 was performed on the NEXESi auto-immuno stainer (Ventana Medical Systems, Tuscon, Arizona, USA) by using anti-DHX32 rabbit oligoclonal antibody at a dilution of 1:400. The antibody was raised against the 15 amino acids at the C-terminus of DHX32 by Sigma Genosys. Control antibody was the preimmune serum used at the same dilution. Immunodetection was carried out using an ABC system employing the Ventana DAB (3-3V-Diaminobenzidine) Detection System (Cat# 250-001). All tissue sections were treated with Ventana Protease I (Cat# 2502018) for 16 min and blocked for endogenous biotin with the Ventana Endogenous Biotin Blocking Kit (Cat#250-050) as part of the automated staining protocol. The slides were counterstained with hematoxylin.
whereas exon 2 transcript is represented by Genbank entry NM_018180 (Fig. 1). In addition, we have cloned an alternatively spliced message from HL-60 acute myeloid leukemia cells in which exon 5 is removed (D exon 5 transcript) (Genbank entry AF427341) (Abdelhaleem, 2002). However, the 5V end of D exon 5 transcript has not been fully characterized yet (Fig. 1). In order to determine the extent of the expression of these DHX32 transcripts in normal lymphoid tissues, we designed transcript-specific primers and used them in RT-PCR. As shown in Fig. 2, exon 1 and exon 2 transcripts are expressed in both the thymus and spleen, whereas bone marrow only expresses exon 2 transcript. Sequence analysis by using the CpG plot program (http://www.ebi.ac.uk/emboss/cpgplot) shows that exon 1 occurs in a CpG island, suggesting the presence of GC-rich promoter. Also, the 2-kb genomic
Results and discussion DHX32 has two transcripts that differ in the 5V untranslated region (UTR) and two alternatively spliced transcripts We have previously shown that DHX32 has 11 exons (Abdelhaleem, 2002). However, by using BLAST program to align the DHX32 genomic sequence (GI:15391100) against human EST database (NCBI), we identified an additional exon located approximately 15 kb upstream of the 11 exons that we reported earlier. Thus, DHX32 gene consists of 12 exons that span 60 kb (Fig. 1). Alternative usage of exon 1 and the first 243 bp of exon 2 results in the production of two transcripts that differ at the 5V UTR without affecting the start ATG codon. Exon 1 transcript is represented by Genbank entry BC068471,
Fig. 2. Expression of DHX32 transcripts in lymphoid tissues: RT-PCR was performed on total RNA isolated from thymus, spleen and bone marrow. -ve RT are control samples without RT step to rule out any possible genomic contamination. Exon 1 and exon 2 reactions use transcript specific primers to detect their respective transcripts. In DHX32, a set of primers that flank exon 5 is used to detect full-length transcript (upper band) and the transcript that lack exon 5 (lower band). The reaction for G6PD is shown as a loading control.
Z. Alli et al. / Experimental and Molecular Pathology 79 (2005) 219 – 223
sequence 5V of exon 2 is strongly predicted to have a TATAcontaining promoter according to the Promoter Scan program (http://thr.cit.nih.gov/molbio/proscan/). These findings suggest that the expression of the two transcripts is regulated by two different promoters, proximal and distal for exon 2 and exon 1 transcripts, respectively (Fig. 1). The use of alternative promoters is an established mechanism to regulate tissuespecific and/or developmental stage-specific gene expression (Landry et al., 2003). Our RT-PCR was performed on total RNA isolated from tissues, it is currently unknown which particular cell type (lymphoid versus non-lymphoid) or lymphocyte subset expresses either exon 2 and exon 1 transcript. Such analysis requires the use of purified lymphoid subpopulations. Also, cloning and characterization of the proximal and distal DHX32 promoters and determination of the transcription factors involved in regulating their activity will shed more light on the mechanisms regulating the expression of exon 1 and exon 2 DHX32 transcripts. In addition to the transcripts that differ at the 5V UTR, all the three tissues examined express full-length and D exon 5 transcripts of the DHX32 gene (Fig. 2). The D exon 5 message is expressed at a lower level compared to fulllength message. The protein product of the alternatively spliced transcript is predicted to lack 81 amino acids in an area structurally suggested to be involved in RNA interaction (Abdelhaleem, 2002). Therefore, the deletion might result in a change of the specificity of RNA binding of DHX32. It is possible that an additional regulation of DHX32 expression/function occurs at the level of regulating alternative splicing.
221
Variable expression of DHX32 in lymphoid tissues In order to determine DHX32 distribution within normal lymphoid tissue, we studied its expression at the protein level with immunohistochemistry. DHX32 was not uniformly expressed in lymphoid tissues (Fig. 3). For example, in lymphoid follicles, follicle center lymphocytes in the germinal center express more DHX32 compared to lymphocytes in the mantle zone. In the spleen, splenic white pulp lymphocytes express very little DHX32 compared to red pulp. Finally, medullary thymocytes express higher levels of DHX32 compared to cortical thymocytes (Fig. 3). These findings suggest variable levels of DHX32 protein expression in normal lymphoid tissues. The level of expression does not appear to depend on the lymphocyte subset. However, the location of the lymphocyte in lymphoid organs appears to be more relevant. The location of lymphocyte can represent varied state of differentiation (mature lymphocytes in thymic medulla compared to cortex) and/or activation (activated lymphocytes in the germinal center compared to mantle zone). Therefore, the expression pattern is consistent with the notion that DHX32 levels depend on the state of lymphocyte activation and/or differentiation. Lymphocyte transition from the resting to activated state is associated with increased transcription of several genes (Samelson, 2002). The newly transcribed genes have to pass through processing steps including splicing, transport and translation before their proteins are expressed. RNA helicases are expected to participate in all these processes (de LaCruz et al., 1999). Additionally, there are several examples of the involvement of RNA helicases in differenti-
Fig. 3. Expression of DHX32 in normal lymphoid tissues. (A) Immunohistochemistry shows no staining of lymphoid follicle with preimmune serum as a negative control. (B) Lymphoid follicle shows strong staining with anti-DHX32 antibody in the follicle center lymphocytes (FC) compared to those in the mantle zone (MZ). (C) Thymus shows strong staining the thymic medulla (M) with the arrow pointing to Hassel’s corpuscle compared to thymic cortex (X). (D) Spleen shows strong staining in the red pulp (R) compared to white pulp (W), the arrowhead points to central arteriole.
222
Z. Alli et al. / Experimental and Molecular Pathology 79 (2005) 219 – 223
Fig. 4. Expression of DHX32 in lymphoma. Immunohistochemistry was used to study the expression of DHX32 by using anti-DHX32 antibody at a dilution of 1 in 400. No staining was detected with preimmune serum. Panels A and B are cases of follicular lymphoma showing low levels of DHX32 expression. Panels C and D are cases of mantle cell lymphoma showing strong staining with DHX32 antibody in the pseudo follicular growth of malignant cells. Panels E and F are cases of large cell lymphoma showing strong cytoplasmic expression of DHX32 in malignant cells. Panels G and H are cases of Burkitt lymphoma showing low levels of expression of DHX32. Panels J and K are cases of nodular sclerosis Hodgkin disease showing cytoplasmic expression of DHX32 in Reed – Sternberg cells.
Z. Alli et al. / Experimental and Molecular Pathology 79 (2005) 219 – 223
ation including DDX3 and DDX25 in spermatogenesis and DDX5 in organ/tissue differentiation (reviewed in Abdelhaleem, 2005). The ability of RNA helicases to modulate the structure and thus availability of RNA molecules for processing is the likely mechanism by which they contribute to differentiation. Detailed analysis of the DHX32 expression in purified lymphocyte populations and/or cell line representing different differentiation and activation states is required to examine these possibilities. We have shown that DHX32 is downregulated in lymphoblastic lymphoma, which has the same cell of origin as acute lymphoblastic leukemia (Abdelhaleem et al., 2005). Here, we studied the expression of DHX32 in other types of lymphoma. As shown in Fig. 4, low level of DHX32 expression was noted in follicular lymphoma, as opposed to the strong expression of its normal counterpart, follicle center lymphocytes (Jaffe et al., 2001). Conversely, mantle cell lymphoma strongly expressed DHX32 compared to its normal counterpart lymphocytes in the mantle zone. Strong expression of DHX32 was noted in large B-cell lymphoma cases. Lymphocytes in Burkitt lymphoma showed varied but generally low level of DHX32 expression. Finally, Reed Sternberg cells in cases of nodular sclerosis Hodgkin disease strongly expressed DHX32. The expression pattern of DHX32 in neoplastic lymphoid tissues, in particular, follicular and mantle cell lymphomas, does not follow the pattern of expression in their normal counterparts. Although the results do not support a general role for DHX32 either as a tumor suppressor or as an oncogene for lymphoma, they raise the possibility of lymphoma-associated dysregulation of expression. Further studies of much larger case series are required for confirmation. Nonetheless, several RNA helicases have been shown to be dysregulated in cancer (reviewed in Abdelhaleem, 2004a,b). The essential roles of this class of enzymes in RNA processing and, hence, final protein
223
expression suggest that their dysregulation can contribute to carcinogenesis. Acknowledgments This work has been supported in part by the Hospital for Sick Children Foundation. The authors acknowledge the technical support of Sherilyn Bell from The Center of Applied Genomics at the Hospital for Sick Children. References Abdelhaleem, M., 2002. The novel helicase homologue DDX32 is downregulated in acute lymphoblastic leukemia. Leuk. Res. 26, 945 – 954. Abdelhaleem, M., Maltais, L., Wain, H., 2003. The human DDX and DHX families of putative RNA helicases. Genomics 81, 618 – 622. Abdelhaleem, M., 2004a. Do RNA helicases have a role in cancer? BBA Rev. Cancer 1704, 37 – 46. Abdelhaleem, M., 2004b. Over-expression of RNA helicases in cancer. Anticancer Res. 24, 3951 – 3954. Abdelhaleem, M., 2005. RNA helicase: regulators of differentiation. Clin. Biochem. 38, 499 – 503. Abdelhaleem, M., Sun, T., Ho, M., 2005. DHX32 expression suggests a role in lymphocyte differentiation. Anticancer Res. 25, 2645 – 2648. de LaCruz, J., Kressler, C.J., Linder, P., 1999. Unwinding RNA in Saccharomyces cerevisiae: DEAD-box proteins and related families. Trends Biochem. Sci. 24, 192 – 198. Jaffe, E.S., Harris, N.L., Stein, H., Vardiman, J.W., 2001. Tumours of haematopoietic and lymphoid tissues. WHO Classification of Tumors. IARC Press, Lyon. Landry, J.-R., Mager, D.L., Wilhelm, B.T., 2003. Complex controls: the role of alternative promoters in mammalian genomes. Trends Genet. 19, 640 – 648. Luking, A., Stahl, U., Schmidt, U., 1998. The protein family of RNA helicases. Crit. Rev. Biochem. Mol. Biol. 33, 259 – 296. Samelson, L.E., 2002. Signal transduction mediated by T cell antigen receptor: the role of adaptor proteins. Ann. Rev. Immunol. 20, 371 – 394. Tanner, N.K., Linder, P., 2001. DExD/H Box RNA helicases: from generic motors to specific dissociation functions. Mol. Cell 8, 251 – 262.
Expression of DHX32 in lymphoid tissues Zaman Alli, Michael Ho, Mohamed Abdelhaleem * Division of Haematopathology, Department of Paediatric Laboratory Medicine, Room 3691, Atrium, Hospital for Sick Children, 555 University Avenue, Toronto, ON, Canada M5G 1X8 Received 11 July 2005 Available online 21 September 2005
Abstract DHX32 is a novel putative RNA helicase identified based on its downregulation in acute lymphoblastic leukemia. DHX32 gene has 12 exons, alternative usage of exons 1 and 2 results in the expression of two transcripts that differ in their 5V untranslated region (UTR), consistent with the involvement of two different promoters. In addition, exon 5 skipping results in the production of a full-length and alternatively spliced transcripts. We used transcript-specific primers in reverse transcriptase (RT)-PCR to show that the thymus and spleen express all of DHX32 transcripts. We also used immunohistochemistry to study the expression of DHX32 protein. The expression pattern of DHX32 protein in normal lymphoid tissues is variable but suggests association with state of lymphocyte activation and/or differentiation. Dysregulated expression of DHX32 is noted in some types of lymphoma compared to their normal counterpart. D 2005 Elsevier Inc. All rights reserved. Keywords: RNA helicase; DHX32; Lymphoma; Gene expression
Introduction RNA helicases are enzymes that utilize the energy derived from nucleotide triphosphate (NTP) hydrolysis to modulate the structure of RNA molecules (Tanner and Linder, 2001). Helicases participate in all biological steps that involve RNA, including transcription, splicing, transport, translation and decay (de LaCruz et al., 1999). Structurally, they are classified into families based on the sequence of the helicase domains, which consist of seven to eight conserved motifs (Luking et al., 1998). We have reported the two largest human RNA helicase families, namely, DDX and DHX families (Abdelhaleem et al., 2003). Two general features have been recognized for RNA helicases, including dysregulation of the expression in various types of cancer (Abdelhaleem, 2004a,b) and involvement in differentiation (Abdelhaleem, 2005). We have identified DDX32 as a novel putative RNA helicase with unique structural features and found it to be downregulated in acute lymphoblastic leukemia (Abdelhaleem, 2002). The gene designation was changed to DHX32 according to the new nomenclature of RNA helicases to reflect its homology to the
* Corresponding author. Fax: +1 416 8136257. E-mail address: [email protected] (M. Abdelhaleem). 0014-4800/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.yexmp.2005.07.002
DHX family (Abdelhaleem et al., 2003). Although the exact biological function of DHX32 is yet to be determined, its structure suggests a role in RNA metabolism (Abdelhaleem, 2002). In this report, we examined the expression of DHX32 in normal and neoplastic lymphoid tissues. Materials and methods Reverse transcription (RT)-PCR The expression of various DHX32 transcripts was determined by RT-PCR of total RNA by using the following primers: primers for exon 1 transcript: forward primer 5V-GTG CCC CGA CGT CGC CGA C-3V and reverse primer 5VGCT GGT AGT GGA TGGA AAG-3V, corresponding to positions 111 – 129 and 798 – 816, respectively of Genbank entry BC06471, and resulting in a 687bp product. Primers for exon 2 transcript: forward 5V-CGG AAC CCA CCA AAC TTT AAA CAC CAG CCT-3V and reverse 5V-CCA TCG CTG GAA TCC AGG GAT TCA GGA AAA-3V, corresponding to positions 57 – 86 and 542 – 571, respectively, of Genbank entry NM_018180, resulting in a 514-bp product. Primers for full-length and D exon 5 transcripts: forward 5V-CGT GCC TGT CAT AGA AGT-3V and reverse 5V-GTC ATG TCT TTG GAG GCA-3V, corresponding to positions 978 – 995 and 1509 – 1526 of Genbank entry AF427340 resulting in a 548- and 305-bp products for full-length and D exon 5 transcripts, respectively. Tissue RT-PCR expression was performed at The Center for Applied Genomic (TCAG) at the Hospital for Sick Children Research Institute. The PCR conditions were as follows: an initial 95-C for 5 min followed by 35 cycles of 95-C for 30 s, 48-C for 30 s and 72-C for 3 min and a final 72-C for 10 min.
220
Z. Alli et al. / Experimental and Molecular Pathology 79 (2005) 219 – 223
Fig. 1. Schematic representation of DHX32 gene and its transcripts. (A) DHX32 gene has 12 exons that span 60 kbp (Genbank entry GI:15391100). Exon 1 occurs in a genomic sequence identified to have a CpG island indicative of a potential promoter, the distal promoter. The genomic sequence 5V of exon 2 is predicted to have a TATA-containing promoter, the proximal promoter. (B) DHX32 transcripts: two transcripts, exon 2 and exon 1 transcripts that differ at the 5V end result from the alternative usage of either the entire of exon 2 or exon 1 instead of the 5V end of exon 2, respectively. The start ATG codon is not affected by the difference at the 5V end. An additional transcript in which exon 5 is deleted is identified; however, the 5V end of this transcript has not been characterized.
Immunohistochemistry The protocol for studying the expression of DHX32 in lymphoid tissues from anonymous archival human cases was approved by the Hospital for Sick Children Research Ethics Board. The following tissues were used for analysis: thymus, tonsil, spleen, follicular lymphoma (3 cases), mantle cell lymphoma (3 cases), large B-cell lymphoma (4 cases), Burkitt lymphoma (4 cases) and nodular sclerosis Hodgkin disease (6 cases). Five micrometers of formalinfixed, paraffin-embedded tissue sections was mounted on positive charged microscope slides. Tissue sections where baked over night at 60-C, dewaxed in xylene and hydrated to distilled water through decreasing concentrations of alcohol. Immunohistochemical procedure for DHX32 was performed on the NEXESi auto-immuno stainer (Ventana Medical Systems, Tuscon, Arizona, USA) by using anti-DHX32 rabbit oligoclonal antibody at a dilution of 1:400. The antibody was raised against the 15 amino acids at the C-terminus of DHX32 by Sigma Genosys. Control antibody was the preimmune serum used at the same dilution. Immunodetection was carried out using an ABC system employing the Ventana DAB (3-3V-Diaminobenzidine) Detection System (Cat# 250-001). All tissue sections were treated with Ventana Protease I (Cat# 2502018) for 16 min and blocked for endogenous biotin with the Ventana Endogenous Biotin Blocking Kit (Cat#250-050) as part of the automated staining protocol. The slides were counterstained with hematoxylin.
whereas exon 2 transcript is represented by Genbank entry NM_018180 (Fig. 1). In addition, we have cloned an alternatively spliced message from HL-60 acute myeloid leukemia cells in which exon 5 is removed (D exon 5 transcript) (Genbank entry AF427341) (Abdelhaleem, 2002). However, the 5V end of D exon 5 transcript has not been fully characterized yet (Fig. 1). In order to determine the extent of the expression of these DHX32 transcripts in normal lymphoid tissues, we designed transcript-specific primers and used them in RT-PCR. As shown in Fig. 2, exon 1 and exon 2 transcripts are expressed in both the thymus and spleen, whereas bone marrow only expresses exon 2 transcript. Sequence analysis by using the CpG plot program (http://www.ebi.ac.uk/emboss/cpgplot) shows that exon 1 occurs in a CpG island, suggesting the presence of GC-rich promoter. Also, the 2-kb genomic
Results and discussion DHX32 has two transcripts that differ in the 5V untranslated region (UTR) and two alternatively spliced transcripts We have previously shown that DHX32 has 11 exons (Abdelhaleem, 2002). However, by using BLAST program to align the DHX32 genomic sequence (GI:15391100) against human EST database (NCBI), we identified an additional exon located approximately 15 kb upstream of the 11 exons that we reported earlier. Thus, DHX32 gene consists of 12 exons that span 60 kb (Fig. 1). Alternative usage of exon 1 and the first 243 bp of exon 2 results in the production of two transcripts that differ at the 5V UTR without affecting the start ATG codon. Exon 1 transcript is represented by Genbank entry BC068471,
Fig. 2. Expression of DHX32 transcripts in lymphoid tissues: RT-PCR was performed on total RNA isolated from thymus, spleen and bone marrow. -ve RT are control samples without RT step to rule out any possible genomic contamination. Exon 1 and exon 2 reactions use transcript specific primers to detect their respective transcripts. In DHX32, a set of primers that flank exon 5 is used to detect full-length transcript (upper band) and the transcript that lack exon 5 (lower band). The reaction for G6PD is shown as a loading control.
Z. Alli et al. / Experimental and Molecular Pathology 79 (2005) 219 – 223
sequence 5V of exon 2 is strongly predicted to have a TATAcontaining promoter according to the Promoter Scan program (http://thr.cit.nih.gov/molbio/proscan/). These findings suggest that the expression of the two transcripts is regulated by two different promoters, proximal and distal for exon 2 and exon 1 transcripts, respectively (Fig. 1). The use of alternative promoters is an established mechanism to regulate tissuespecific and/or developmental stage-specific gene expression (Landry et al., 2003). Our RT-PCR was performed on total RNA isolated from tissues, it is currently unknown which particular cell type (lymphoid versus non-lymphoid) or lymphocyte subset expresses either exon 2 and exon 1 transcript. Such analysis requires the use of purified lymphoid subpopulations. Also, cloning and characterization of the proximal and distal DHX32 promoters and determination of the transcription factors involved in regulating their activity will shed more light on the mechanisms regulating the expression of exon 1 and exon 2 DHX32 transcripts. In addition to the transcripts that differ at the 5V UTR, all the three tissues examined express full-length and D exon 5 transcripts of the DHX32 gene (Fig. 2). The D exon 5 message is expressed at a lower level compared to fulllength message. The protein product of the alternatively spliced transcript is predicted to lack 81 amino acids in an area structurally suggested to be involved in RNA interaction (Abdelhaleem, 2002). Therefore, the deletion might result in a change of the specificity of RNA binding of DHX32. It is possible that an additional regulation of DHX32 expression/function occurs at the level of regulating alternative splicing.
221
Variable expression of DHX32 in lymphoid tissues In order to determine DHX32 distribution within normal lymphoid tissue, we studied its expression at the protein level with immunohistochemistry. DHX32 was not uniformly expressed in lymphoid tissues (Fig. 3). For example, in lymphoid follicles, follicle center lymphocytes in the germinal center express more DHX32 compared to lymphocytes in the mantle zone. In the spleen, splenic white pulp lymphocytes express very little DHX32 compared to red pulp. Finally, medullary thymocytes express higher levels of DHX32 compared to cortical thymocytes (Fig. 3). These findings suggest variable levels of DHX32 protein expression in normal lymphoid tissues. The level of expression does not appear to depend on the lymphocyte subset. However, the location of the lymphocyte in lymphoid organs appears to be more relevant. The location of lymphocyte can represent varied state of differentiation (mature lymphocytes in thymic medulla compared to cortex) and/or activation (activated lymphocytes in the germinal center compared to mantle zone). Therefore, the expression pattern is consistent with the notion that DHX32 levels depend on the state of lymphocyte activation and/or differentiation. Lymphocyte transition from the resting to activated state is associated with increased transcription of several genes (Samelson, 2002). The newly transcribed genes have to pass through processing steps including splicing, transport and translation before their proteins are expressed. RNA helicases are expected to participate in all these processes (de LaCruz et al., 1999). Additionally, there are several examples of the involvement of RNA helicases in differenti-
Fig. 3. Expression of DHX32 in normal lymphoid tissues. (A) Immunohistochemistry shows no staining of lymphoid follicle with preimmune serum as a negative control. (B) Lymphoid follicle shows strong staining with anti-DHX32 antibody in the follicle center lymphocytes (FC) compared to those in the mantle zone (MZ). (C) Thymus shows strong staining the thymic medulla (M) with the arrow pointing to Hassel’s corpuscle compared to thymic cortex (X). (D) Spleen shows strong staining in the red pulp (R) compared to white pulp (W), the arrowhead points to central arteriole.
222
Z. Alli et al. / Experimental and Molecular Pathology 79 (2005) 219 – 223
Fig. 4. Expression of DHX32 in lymphoma. Immunohistochemistry was used to study the expression of DHX32 by using anti-DHX32 antibody at a dilution of 1 in 400. No staining was detected with preimmune serum. Panels A and B are cases of follicular lymphoma showing low levels of DHX32 expression. Panels C and D are cases of mantle cell lymphoma showing strong staining with DHX32 antibody in the pseudo follicular growth of malignant cells. Panels E and F are cases of large cell lymphoma showing strong cytoplasmic expression of DHX32 in malignant cells. Panels G and H are cases of Burkitt lymphoma showing low levels of expression of DHX32. Panels J and K are cases of nodular sclerosis Hodgkin disease showing cytoplasmic expression of DHX32 in Reed – Sternberg cells.
Z. Alli et al. / Experimental and Molecular Pathology 79 (2005) 219 – 223
ation including DDX3 and DDX25 in spermatogenesis and DDX5 in organ/tissue differentiation (reviewed in Abdelhaleem, 2005). The ability of RNA helicases to modulate the structure and thus availability of RNA molecules for processing is the likely mechanism by which they contribute to differentiation. Detailed analysis of the DHX32 expression in purified lymphocyte populations and/or cell line representing different differentiation and activation states is required to examine these possibilities. We have shown that DHX32 is downregulated in lymphoblastic lymphoma, which has the same cell of origin as acute lymphoblastic leukemia (Abdelhaleem et al., 2005). Here, we studied the expression of DHX32 in other types of lymphoma. As shown in Fig. 4, low level of DHX32 expression was noted in follicular lymphoma, as opposed to the strong expression of its normal counterpart, follicle center lymphocytes (Jaffe et al., 2001). Conversely, mantle cell lymphoma strongly expressed DHX32 compared to its normal counterpart lymphocytes in the mantle zone. Strong expression of DHX32 was noted in large B-cell lymphoma cases. Lymphocytes in Burkitt lymphoma showed varied but generally low level of DHX32 expression. Finally, Reed Sternberg cells in cases of nodular sclerosis Hodgkin disease strongly expressed DHX32. The expression pattern of DHX32 in neoplastic lymphoid tissues, in particular, follicular and mantle cell lymphomas, does not follow the pattern of expression in their normal counterparts. Although the results do not support a general role for DHX32 either as a tumor suppressor or as an oncogene for lymphoma, they raise the possibility of lymphoma-associated dysregulation of expression. Further studies of much larger case series are required for confirmation. Nonetheless, several RNA helicases have been shown to be dysregulated in cancer (reviewed in Abdelhaleem, 2004a,b). The essential roles of this class of enzymes in RNA processing and, hence, final protein
223
expression suggest that their dysregulation can contribute to carcinogenesis. Acknowledgments This work has been supported in part by the Hospital for Sick Children Foundation. The authors acknowledge the technical support of Sherilyn Bell from The Center of Applied Genomics at the Hospital for Sick Children. References Abdelhaleem, M., 2002. The novel helicase homologue DDX32 is downregulated in acute lymphoblastic leukemia. Leuk. Res. 26, 945 – 954. Abdelhaleem, M., Maltais, L., Wain, H., 2003. The human DDX and DHX families of putative RNA helicases. Genomics 81, 618 – 622. Abdelhaleem, M., 2004a. Do RNA helicases have a role in cancer? BBA Rev. Cancer 1704, 37 – 46. Abdelhaleem, M., 2004b. Over-expression of RNA helicases in cancer. Anticancer Res. 24, 3951 – 3954. Abdelhaleem, M., 2005. RNA helicase: regulators of differentiation. Clin. Biochem. 38, 499 – 503. Abdelhaleem, M., Sun, T., Ho, M., 2005. DHX32 expression suggests a role in lymphocyte differentiation. Anticancer Res. 25, 2645 – 2648. de LaCruz, J., Kressler, C.J., Linder, P., 1999. Unwinding RNA in Saccharomyces cerevisiae: DEAD-box proteins and related families. Trends Biochem. Sci. 24, 192 – 198. Jaffe, E.S., Harris, N.L., Stein, H., Vardiman, J.W., 2001. Tumours of haematopoietic and lymphoid tissues. WHO Classification of Tumors. IARC Press, Lyon. Landry, J.-R., Mager, D.L., Wilhelm, B.T., 2003. Complex controls: the role of alternative promoters in mammalian genomes. Trends Genet. 19, 640 – 648. Luking, A., Stahl, U., Schmidt, U., 1998. The protein family of RNA helicases. Crit. Rev. Biochem. Mol. Biol. 33, 259 – 296. Samelson, L.E., 2002. Signal transduction mediated by T cell antigen receptor: the role of adaptor proteins. Ann. Rev. Immunol. 20, 371 – 394. Tanner, N.K., Linder, P., 2001. DExD/H Box RNA helicases: from generic motors to specific dissociation functions. Mol. Cell 8, 251 – 262.