Limits...
Functional characterization and identification of mouse Rad51d splice variants.

Gruver AM, Yard BD, McInnes C, Rajesh C, Pittman DL - BMC Mol. Biol. (2009)

Bottom Line: In addition, the linker region (residues 54 through 77) of RAD51D was identified as a region that potentially mediates binding with XRCC2.These expression studies also led to the identification of two additional Rad51d ubiquitously expressed transcripts, one deleted for both exon 9 and 10 and one deleted for only exon 10.These results suggest Rad51d alternative splice variants potentially modulate mechanisms of HR by sequestering either RAD51C or XRCC2.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Pharmaceutical and Biomedical Sciences, South Carolina College of Pharmacy, University of South Carolina Campus, Columbia, SC 29208, USA. gruvera@ccf.org

ABSTRACT

Background: The homologous recombination (HR) pathway is vital for maintaining genomic integrity through the restoration of double-stranded breaks and interstrand crosslinks. The RAD51 paralogs (RAD51B, RAD51C, RAD51D, XRCC2, XRCC3) are essential for this process in vertebrates, and the RAD51D paralog is unique in that it participates in both HR repair and telomere maintenance. RAD51D is also known to directly interact with the RAD51C and XRCC2 proteins. Rad51d splice variants have been reported in mouse and human tissues, supportive of a role for alternative splicing in HR regulation. The present study evaluated the interaction of the Rad51d splice isoform products with RAD51C and XRCC2 and their expression patterns.

Results: Yeast-2-hybrid analysis was used to determine that the Mus musculus Rad51d splice variant product RAD51DDelta7b (deleted for residues 219 through 223) was capable of interacting with both RAD51C and XRCC2 and that RAD51D+int3 interacted with XRCC2. In addition, the linker region (residues 54 through 77) of RAD51D was identified as a region that potentially mediates binding with XRCC2. Cellular localization, detected by EGFP fusion proteins, demonstrated that each of the splice variant products tested was distributed throughout the cell similar to the full-length protein. However, none of the splice variants were capable of restoring resistance of Rad51d-deficient cell lines to mitomycin C. RT-PCR expression analysis revealed that Rad51dDelta3 (deleted for exon 3) and Rad51dDelta5 (deleted for exon 5)transcripts display tissue specific expression patterns with Rad51dDelta3 being detected in each tissue except ovary and Rad51dDelta5 not detected in mammary gland and testis. These expression studies also led to the identification of two additional Rad51d ubiquitously expressed transcripts, one deleted for both exon 9 and 10 and one deleted for only exon 10.

Conclusion: These results suggest Rad51d alternative splice variants potentially modulate mechanisms of HR by sequestering either RAD51C or XRCC2.

Show MeSH

Related in: MedlinePlus

Domain mapping of Mus musculus RAD51D. (A-B) Yeast two-hybrid analysis of RAD51DΔ3 and RAD51D+int3 are directly compared with the original deletion constructs of RAD51D. (C) Diagram of the RAD51DΔ3 and RAD5D+int3 alternative splice constructs compared with the amino (4–77) and carboxy-terminal (77–329) regions of RAD51D. The black box represents the predicted linker region of RAD51D. (D) The region of the protein that allows interaction with XRCC2 is illustrated in bold type, while sequence required for its interaction with RAD51C is presented in italics (note a small area of overlap between residues 223–233). The underlined 24 amino acid region (residues 54–77) appears to be critical for determining the specificity of the interaction between RAD51D isoforms and XRCC2. (E) Homology model of RAD51D from the Pyrococcus furiosus RAD51 crystal structure. The yellow highlighted region (left) represents the linker region and the yellow area (right) represents the 24 amino acid region proposed to determine XRCC2 specificity. Abbreviations: 51D; RAD51D-FL, 51C; RAD51C, X2; XRCC2, 4–77; residues 4–77 of the amino-terminal domain of RAD51D, 77–329; residues 77–329 of the carboxy-terminal domain of RAD51D, pGAD; pGADT7 vector, pGBK; pGBKT7 vector.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2667185&req=5

Figure 3: Domain mapping of Mus musculus RAD51D. (A-B) Yeast two-hybrid analysis of RAD51DΔ3 and RAD51D+int3 are directly compared with the original deletion constructs of RAD51D. (C) Diagram of the RAD51DΔ3 and RAD5D+int3 alternative splice constructs compared with the amino (4–77) and carboxy-terminal (77–329) regions of RAD51D. The black box represents the predicted linker region of RAD51D. (D) The region of the protein that allows interaction with XRCC2 is illustrated in bold type, while sequence required for its interaction with RAD51C is presented in italics (note a small area of overlap between residues 223–233). The underlined 24 amino acid region (residues 54–77) appears to be critical for determining the specificity of the interaction between RAD51D isoforms and XRCC2. (E) Homology model of RAD51D from the Pyrococcus furiosus RAD51 crystal structure. The yellow highlighted region (left) represents the linker region and the yellow area (right) represents the 24 amino acid region proposed to determine XRCC2 specificity. Abbreviations: 51D; RAD51D-FL, 51C; RAD51C, X2; XRCC2, 4–77; residues 4–77 of the amino-terminal domain of RAD51D, 77–329; residues 77–329 of the carboxy-terminal domain of RAD51D, pGAD; pGADT7 vector, pGBK; pGBKT7 vector.

Mentions: The observation that RAD51D isoforms may selectively interact with RAD51C and XRCC2 allowed further domain mapping of RAD51 paralog complexes. Miller et al. reported that the amino-terminal domain of RAD51D (residues 4–77) interacts with XRCC2, and the carboxy-terminal region (residues 77–328) is sufficient for interaction with RAD51C [29]. Binding of RAD51DΔ8 (residues 4–233) and RAD51D+int3 (residues 4–109) to XRCC2 support this observation (Figure 2). However, the lack of association between RAD51DΔ8 and RAD51C suggests a more narrow region of the carboxy-terminal domain than previously reported, consisting of amino acids 234–329, is required for the interaction between RAD51D and RAD51C. Interestingly, the yeast two-hybrid analyses displayed in Figure 2 also suggest that residues 54–77 within the amino-terminal region determine whether RAD51D interacts with XRCC2 (shown in black, Figure 3C). To confirm these observations, the original RAD51D deletion constructs used in the study by Miller et al were tested against the RAD51D isoforms. If amino acids 54–77 are present, as with the RAD51D (4–77) construct, the peptide interacts with XRCC2 (Figure 3). When these residues are missing, as in the RAD51DΔ3 isoform, there is an absence of association with XRCC2 and some interaction with RAD51C. This span of 24 amino acids (54–77) is nearly identical to the "linker region" (residues 60–78) proposed from the structure modeling of human RAD51D from the Pyrococcus furiosus RAD51 crystal structure [29].


Functional characterization and identification of mouse Rad51d splice variants.

Gruver AM, Yard BD, McInnes C, Rajesh C, Pittman DL - BMC Mol. Biol. (2009)

Domain mapping of Mus musculus RAD51D. (A-B) Yeast two-hybrid analysis of RAD51DΔ3 and RAD51D+int3 are directly compared with the original deletion constructs of RAD51D. (C) Diagram of the RAD51DΔ3 and RAD5D+int3 alternative splice constructs compared with the amino (4–77) and carboxy-terminal (77–329) regions of RAD51D. The black box represents the predicted linker region of RAD51D. (D) The region of the protein that allows interaction with XRCC2 is illustrated in bold type, while sequence required for its interaction with RAD51C is presented in italics (note a small area of overlap between residues 223–233). The underlined 24 amino acid region (residues 54–77) appears to be critical for determining the specificity of the interaction between RAD51D isoforms and XRCC2. (E) Homology model of RAD51D from the Pyrococcus furiosus RAD51 crystal structure. The yellow highlighted region (left) represents the linker region and the yellow area (right) represents the 24 amino acid region proposed to determine XRCC2 specificity. Abbreviations: 51D; RAD51D-FL, 51C; RAD51C, X2; XRCC2, 4–77; residues 4–77 of the amino-terminal domain of RAD51D, 77–329; residues 77–329 of the carboxy-terminal domain of RAD51D, pGAD; pGADT7 vector, pGBK; pGBKT7 vector.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2667185&req=5

Figure 3: Domain mapping of Mus musculus RAD51D. (A-B) Yeast two-hybrid analysis of RAD51DΔ3 and RAD51D+int3 are directly compared with the original deletion constructs of RAD51D. (C) Diagram of the RAD51DΔ3 and RAD5D+int3 alternative splice constructs compared with the amino (4–77) and carboxy-terminal (77–329) regions of RAD51D. The black box represents the predicted linker region of RAD51D. (D) The region of the protein that allows interaction with XRCC2 is illustrated in bold type, while sequence required for its interaction with RAD51C is presented in italics (note a small area of overlap between residues 223–233). The underlined 24 amino acid region (residues 54–77) appears to be critical for determining the specificity of the interaction between RAD51D isoforms and XRCC2. (E) Homology model of RAD51D from the Pyrococcus furiosus RAD51 crystal structure. The yellow highlighted region (left) represents the linker region and the yellow area (right) represents the 24 amino acid region proposed to determine XRCC2 specificity. Abbreviations: 51D; RAD51D-FL, 51C; RAD51C, X2; XRCC2, 4–77; residues 4–77 of the amino-terminal domain of RAD51D, 77–329; residues 77–329 of the carboxy-terminal domain of RAD51D, pGAD; pGADT7 vector, pGBK; pGBKT7 vector.
Mentions: The observation that RAD51D isoforms may selectively interact with RAD51C and XRCC2 allowed further domain mapping of RAD51 paralog complexes. Miller et al. reported that the amino-terminal domain of RAD51D (residues 4–77) interacts with XRCC2, and the carboxy-terminal region (residues 77–328) is sufficient for interaction with RAD51C [29]. Binding of RAD51DΔ8 (residues 4–233) and RAD51D+int3 (residues 4–109) to XRCC2 support this observation (Figure 2). However, the lack of association between RAD51DΔ8 and RAD51C suggests a more narrow region of the carboxy-terminal domain than previously reported, consisting of amino acids 234–329, is required for the interaction between RAD51D and RAD51C. Interestingly, the yeast two-hybrid analyses displayed in Figure 2 also suggest that residues 54–77 within the amino-terminal region determine whether RAD51D interacts with XRCC2 (shown in black, Figure 3C). To confirm these observations, the original RAD51D deletion constructs used in the study by Miller et al were tested against the RAD51D isoforms. If amino acids 54–77 are present, as with the RAD51D (4–77) construct, the peptide interacts with XRCC2 (Figure 3). When these residues are missing, as in the RAD51DΔ3 isoform, there is an absence of association with XRCC2 and some interaction with RAD51C. This span of 24 amino acids (54–77) is nearly identical to the "linker region" (residues 60–78) proposed from the structure modeling of human RAD51D from the Pyrococcus furiosus RAD51 crystal structure [29].

Bottom Line: In addition, the linker region (residues 54 through 77) of RAD51D was identified as a region that potentially mediates binding with XRCC2.These expression studies also led to the identification of two additional Rad51d ubiquitously expressed transcripts, one deleted for both exon 9 and 10 and one deleted for only exon 10.These results suggest Rad51d alternative splice variants potentially modulate mechanisms of HR by sequestering either RAD51C or XRCC2.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Pharmaceutical and Biomedical Sciences, South Carolina College of Pharmacy, University of South Carolina Campus, Columbia, SC 29208, USA. gruvera@ccf.org

ABSTRACT

Background: The homologous recombination (HR) pathway is vital for maintaining genomic integrity through the restoration of double-stranded breaks and interstrand crosslinks. The RAD51 paralogs (RAD51B, RAD51C, RAD51D, XRCC2, XRCC3) are essential for this process in vertebrates, and the RAD51D paralog is unique in that it participates in both HR repair and telomere maintenance. RAD51D is also known to directly interact with the RAD51C and XRCC2 proteins. Rad51d splice variants have been reported in mouse and human tissues, supportive of a role for alternative splicing in HR regulation. The present study evaluated the interaction of the Rad51d splice isoform products with RAD51C and XRCC2 and their expression patterns.

Results: Yeast-2-hybrid analysis was used to determine that the Mus musculus Rad51d splice variant product RAD51DDelta7b (deleted for residues 219 through 223) was capable of interacting with both RAD51C and XRCC2 and that RAD51D+int3 interacted with XRCC2. In addition, the linker region (residues 54 through 77) of RAD51D was identified as a region that potentially mediates binding with XRCC2. Cellular localization, detected by EGFP fusion proteins, demonstrated that each of the splice variant products tested was distributed throughout the cell similar to the full-length protein. However, none of the splice variants were capable of restoring resistance of Rad51d-deficient cell lines to mitomycin C. RT-PCR expression analysis revealed that Rad51dDelta3 (deleted for exon 3) and Rad51dDelta5 (deleted for exon 5)transcripts display tissue specific expression patterns with Rad51dDelta3 being detected in each tissue except ovary and Rad51dDelta5 not detected in mammary gland and testis. These expression studies also led to the identification of two additional Rad51d ubiquitously expressed transcripts, one deleted for both exon 9 and 10 and one deleted for only exon 10.

Conclusion: These results suggest Rad51d alternative splice variants potentially modulate mechanisms of HR by sequestering either RAD51C or XRCC2.

Show MeSH
Related in: MedlinePlus