| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 280, Issue 23, 22375-22384, June 10, 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Interacting Protein 38*
||
From the
Department of Biochemistry and Molecular Biology, New York Medical College, Valhalla, New York 10595 and the ¶Department of Medicine, New York Medical College, Valhalla, New York 10595
Received for publication, December 27, 2004 , and in revised form, March 31, 2005.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
interacting protein 38 (PDIP38) was identified as a human DNA polymerase (pol) interacting protein through a direct interaction with p50, the small subunit of human pol . PDIP38 was also found to interact with proliferating cell nuclear antigen, which suggested that it might play a role in vivo in the processes of DNA replication and DNA repair in the nucleus. In order to characterize further this novel protein, we have examined its subcellular localization by the use of immunochemical and cellular fractionation techniques. These studies show that PDIP38 is a novel mitochondrial protein and is localized mainly to the mitochondria. PDIP38 was shown to possess a functional mitochondrial targeting sequence that is located within the first 35 N-terminal amino acid residues. The mature PDIP38 protein is about 50 amino acid residues smaller than the full-length precursor PDIP38 protein, consistent with it being processed by cleavage of the mitochondrial targeting sequence during entry into the mitochondria. His-tagged mature PDIP38 inhibited pol activity in vitro and interacted with human papillomavirus 16 E7 oncoprotein, suggesting that PDIP38 might play a role in the pol -mediated viral DNA replication. Although the localization of PDIP38 to the mitochondria suggests that it serves functions within the mitochondria, we cannot eliminate the possibility that it may be involved in pol -mediated DNA replication or DNA repair under certain conditions such as viral infection. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
, , and 
, are involved in DNA replication (1–3). Although it is clear that the DNA polymerase /primase is involved in initiating both leading and lagging strand synthesis, the precise roles of DNA polymerase (pol)1 and DNA polymerase remain to be fully defined. A mutant without the catalytic domain of the large subunit of pol in Schizosaccharomyces pombe was found to be viable (4), and pol and pol were sufficient to replicate both the leading and lagging strands of double-stranded DNA in an SV40 DNA replication system (5). These earlier studies pointed to the fact that pol was likely to be the major polymerase for the elongation of both leading and lagging strands of chromosomal DNA in eukaryotic cells. This was supported by recent studies that showed that there was inefficient elongation of both leading and lagging DNA strands in Xenopus egg extracts depleted of pol (6). pol was suggested to play a role mainly in leading strand DNA synthesis (6). pol , working together with Fen1 and proliferating cell nuclear antigen (PCNA), is required for Okazaki fragment maturation, and the 3'–5'-exonuclease activity of pol is important for this process (7–9). pol is also involved in telomerase-mediated telomere addition in vivo (10) and participates in several DNA repair pathways (11, 12). The ability of pol to participate in multiple aspects of DNA replication/repair raises the question of how it is regulated. The association of pol with different proteins or protein complexes in vivo and post-translational modifications of the pol complex could be important mechanisms (13–16).
Mammalian pol was originally characterized as a tightly associated heterodimer of the catalytic subunit, p125, with a small subunit, p50 (17). More recently, we and others (17–21) have identified two additional subunits of mammalian pol , the p68 and p12 subunits, so that mammalian pol is now regarded as a heterotetramer of four subunits. Recently, we identified a novel protein, PDIP38, which interacted with the small subunit p50 of human pol , as well as with PCNA (22). PDIP38 joins a growing number of proteins that have been reported to have the ability to interact with the small p50 subunit of pol as well as with PCNA. These include p125 itself (18, 23, 24), p68 (25, 26), WRN (27, 28), and PDIP1 (29). PDIP1 could enhance PCNA-dependent pol activity in vitro (29), and WRN increased the rate of nucleotide incorporation by pol in the absence of PCNA (30). The function of PDIP38 in regard to its interaction with pol is unknown. To understand better the physiological function of PDIP38, we examined its cellular localization in detail. In this study, we demonstrate by both optical and biochemical methods that PDIP38 is a mitochondrial protein. We further identified a bona fide mitochondrial targeting sequence (MTS) located at the N-terminal region of PDIP38. The first N-terminal 35 amino acids of PDIP38 are necessary and sufficient for its targeting to the mitochondria. By using a series of N-terminal deleted PDIP38 fusion proteins, we also demonstrated that the cleavage site is most likely located 50 amino acid residues from the N terminus of the PDIP38 precursor protein. We found that PDIP38 could inhibit the PCNA-dependent pol activity and could interact with HPV 16 E7 oncoprotein. These results indicated that PDIP38 might play a role in the pol -mediated viral DNA replication.
|
EXPERIMENTAL PROCEDURES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
Construction of PDIP38 Fusion Constructs and His-p21—Standard low cycle PCR techniques were used to generate the C-terminal enhanced green fluorescent protein (EGFP)-tagged PDIP38 full-length and deletion mutants using pEGFPN3 expression vector (Clontech). The PCR products were digested with HindIII and BamHI and subcloned into the pEGFPN3 (Clontech) expression vector between the HindIII and BamHI sites to generate fusion proteins with EGFP attached to the C termini of the inserted sequences. To generate PDIP38 tagged at its C terminus with the FLAG peptide sequence (PDIP38-C-FLAG), we annealed two oligonucleotides (5'-GATCCGGCGGGGACTACAAAGACGATGACGACAAGTAAGC-3' and 5'-GGCCGCTTACTTGTCGTCATCGTCTTTGTAGTCCCCGCCG-3'), and we inserted these between the BamHI and NotI sites of PDIP38-C-EGFP. The EGFP tag was replaced by FLAG peptide (DYKDDDK) to generate the PDIP38-C-FLAG. The series of C-terminal FLAG-tagged PDIP38 deletion mutants was obtained by inserting HindIII and BamHI fragments of the respective PDIP38-C-EGFP deletion mutants into the HindIII and BamHI sites of PDIP38-C-FLAG. Low cycle PCR was used to generate the N-terminal EGFP-tagged PDIP38-(EGFP-N-PDIP38) full-length fusion protein by using pEGFPC1 expression vector (Clontech). The oligonucleotide pairs were 5'-ACGAATCCGGAGCAGCCTGTACAGCCCGGC-3' and 5'-TACGTAAGCTTCTACCAGTGAAGGCCTGAGGG-3'. The PCR product was digested with BspEI and HindIII and subcloned in-frame into pEGFPC1 (Clontech) expression vector between the BspEI and HindIII sites. Low cycle PCR was used to generate the N-terminal His6-tagged PDIP38 full-length fusion protein. The primer pairs for pQE32-PDIP38-(2–368), N-His-PDIP38 were 5'-AGCTGGATCCCCGCAGCCTGTACAGCCCGG-3' and 5'-TACGTAAGCTTCTACCAGTGAAGGCCTGAGGG-3'. The PCR product was digested with BamHI and HindIII and subcloned in-frame into the pQE32 (Qiagen, CA) expression vector between BamHI and HindIII sites. Low cycle PCR was used to generate C-terminal His6-tagged PDIP38. The primer pairs for full-length pQE70-PDIP38-(1–368), PDIP38-C-His were 5'-ATGCAGCATGCGAGCAGCCTGTACAGCCCGGC-3' and 5'-ACGTAGGATCCCCAGTGAAGGCCTGAGGGTGGTG-3'. The primer pairs for pQE70-PDIP38-(51–368), PDIP38-(51–368)-C-His were 5'-ATGCAGCATGCTCTCGTCCCGAAACCGACC-3' and 5'-ACGTAGGATCCCCAGTGAAGGCCTGAGGGTGGTG-3'. The PCR product was digested with SphI and BamHI and subcloned in-frame into the PQE70 (Qiagen, CA) expression vector between the SphI and BamHI sites to create PDIP-38-(51–368)-C-His. In vitro translation constructs were generated by using pCDNA3.1(+) expression vector (Invitrogen). Low cycle PCR was used to generate the untagged wild type PDIP38 in pCDNA3.1(+) using pQE32 PDIP38 full-length as the template. The oligonucleotide pairs were 5'-ATGCGAGGATCCGCCACCATGGCAGCCTGTACAGCCCGG-3' and 5'-GTTCTGAGGTCATTACTGG-3'. The PCR product was digested with BamHI and inserted into the BamHI and EcoRV sites of pCDNA3.1(+) to create pCDNA3.1(+) PDIP38. To generate the vector expressing FLAG-tagged at its N terminus (pCDNA3.1(+) N-FLAG-PDIP38), we first inserted a DNA sequence encoding FLAG peptide (M-DYKDDDDKGG) between the HindIII and BamHI sites by annealing 5'-AGCTTGCCACCATGGACTACAAAGACGATGACGACAAGGGCGG-3' and 5'-GATCCCGCCCTTGTCGTCATCGTCTTTGTAGTCCATGGTGGCA-3' to create pCDNA3.1(+)FLAG. The pQE32 PDIP38 full-length was digested with BamHI and PvuII, and the DNA fragment containing full-length PDIP38 was inserted into the BamHI and EcoRV sites of pCDNA3.1(+)FLAG to create pCDNA3.1(+)N-FLAG-PDIP38. To generate the pCDNA3.1(+)PDIP38-C-FLAG tag construct, the HindIII/NotI fragment containing PDIP38 from the PDIP38-C-FLAG construct was inserted into the HindIII/NotI sites of pCDNA3.1(+). Low cycle PCR was used to generate the N-terminal His6-tagged P21 full-length fusion protein. The primer pairs for pQE32-P21-(2–164), His-p21 were 5'-AGCTGGATCCCCTCAGAACCGGCTGGGGATG-3' and 5'-TACGTAAGCTTTTAGGGCTTCCTCTTGGAG-3'. The PCR product was digested with BamHI and HindIII and subcloned in-frame into the pQE32 (Qiagen) expression vector between BamHI and HindIII sites. All constructs were sequenced (Davis Sequencing).
Protein Purifications—GST fusion or His-tagged proteins were expressed in Escherichia coli BL21DE3 (pLys) and purified by using either glutathione beads (Amersham Biosciences) or Ni-NTA-agarose (Qiagen). His-p21 and PDIP38-(51–368)-C-His were further purified using Superdex 200 gel filtration chromatography (Amersham Biosciences).
Cell Culture and Transfection—293T, MCF7, and HeLa cells were obtained from ATCC and maintained in Dulbecco's modified Eagle's medium with 10% fetal calf serum. Transient transfection of PDIP38-EGFP constructs and PDIP38-FLAG constructs were performed using Lipofectamine 2000 (Invitrogen) using the supplied protocols.
Immunocytochemistry—HeLa cells were grown in 4-well slides and fixed with 100% methanol and further treated with 0.1% Triton X-100 in PBS. Endogenous PDIP38 was immunostained with an anti-PDIP38 rabbit polyclonal antibody raised against the C-terminal 20 amino acid residues of PDIP38 (H2N-PPFSLESNKDEKTPPSGLHW-COOH) at 1:100 dilution in 2% BSA/PBS for 1 h at room temperature and washed three times with PBS. Endogenous human mitochondrial heat shock protein 70 (mhsp70) was immunostained with anti-mhsp70 (Affinity Bioreagents) mouse monoclonal antibody at 1:200 dilution, washed three times with PBS, and then incubated with Rhodamine Red-X-conjugated anti-rabbit IgG and fluorescein isothiocyanate (green)-conjugated anti-mouse IgG (Jackson ImmunoResearch) at room temperature for 1 h in the dark. For ectopically expressed EGFP fusion proteins, HeLa cells were transfected with the indicated pEGFP-N3 or pEGFP-C1 constructs, and 24 h after transfection, Hoechst 33258 (Sigma) was added at 10 µM and MitoTracker Red CMXRos (M-7512, Molecular Probes) at 20 nM for an additional hour, and the cells were fixed using 100% methanol. Fluorescence microscopy was performed on a Nikon microscope, and images were captured with a digital camera (Optronics) using Optronics MagFire, Image-ProPlus software.
Subcellular Fractionation—Mitochondria were isolated from HeLa cells using a mitochondria isolation kit from Sigma. The cells were also fractionated with the NE-PER kit (Pierce). The proteins were run on SDS-PAGE, transferred to nitrocellulose membranes, and blotted with the following antibodies: mouse monoclonal antibody anti-p50 (17D2 and 13D5) (26); rabbit polyclonal antibody anti-p68 (26); rabbit polyclonal antibody anti-p12 (26); rabbit polyclonal antibody anti-PDIP38-(22); and mouse monoclonal antibody anti-mhsp70 (Affinity Biore-agents, Golden, CO). The following antibodies were obtained from Santa Cruz Biotechnology: mouse monoclonal antibody anti-PCNA (PC10); mouse monoclonal antibody anti-GFP (B-2); goat polyclonal antibody anti-lamin B (C-20); goat polyclonal antibody anti-I
B (C-21); and rabbit polyclonal antibody anti-Tom20 (FL-145).
In Vitro Translation—pCDNA3.1(+)N-FLAG-PDIP38, pCDNA3.1-(+)PDIP38-C-FLAG, and pCDNA3.1(+)PDIP38 were used to generate their respective 35S-labeled proteins using the TNT-coupled reticulocyte lysate in vitro translation kit (Promega, NJ).
GST Pull Down Assay—GST pull down assays using purified proteins or total HeLa cell lysate were performed as described previously (22).
DNA Polymerase Assay—Sparsely primed poly(dA)/oligo(dT) assay was performed as described previously (17). The standard reactions for the poly(dA)/oligo(dT) assay contained 0.25 OD units/ml poly(dA)/oligo(dT) (20:1), 200 µg/ml BSA, 5% glycerol, 10 mM MgCl2, 25 mM HEPES, pH 6.0, 100 cpm/pmol [3H]TTP, and 2 ng of HeLa pol in the absence of PCNA or the presence of 100 ng (138 nM monomer) of PCNA in a total volume of 25 µl. Reaction mixtures were incubated for 30 min at 37 °C and were terminated by spotting onto DE81 papers, which were then washed four times with 0.3 M ammonium formate, pH 7.8, once with 95% ethanol, and counted. For the inhibition assay using His-p21 or PDIP38-(51–368)-C-His, the indicated amount of His-p21 or PDIP38-(51–368)-C-His was preincubated with either 100 ng of PCNA or TGEED buffer and 2 ng of HeLa pol on ice for 30 min before the assay mixtures were added.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
(22). The interaction of PDIP38 with pol was confirmed by a number of experimental approaches, including pull down and overlay experiments. This interaction suggested that PDIP38 might be a nuclear protein that participated in some aspect of DNA replication or repair. To understand better the likely physiological role(s) of PDIP38, we investigated its intracellular localization in cultured cells. The anti-PDIP38 antibody that was used (22) specifically recognized one major band migrating with apparent molecular size slightly larger than 32.5 kDa when total cell lysates of HeLa, 293T, MCF7, or 3T3 cells were immunoblotted (Fig. 1A). When immunofluorescence microscopy was performed by staining the HeLa cells with anti-PDIP38 antibody, the results were unexpected in that a cytoplasmic localization of PDIP38 (Fig. 1B, a) was observed, with a near-complete exclusion from the nuclear region stained with Hoechst (Fig. 1B, b and c). Similar staining patterns were observed when 293T or MCF7 cells were used (data not shown). More importantly, the punctated cytosolic staining pattern that was observed is similar to that typically obtained when mitochondria are immunostained (32).
|
The N-terminal region of human PDIP38 is rich in positively charged amino acids and is predicted to form an -helix, characteristics that are typical of MTS sequences. The co-localization of PDIP38 with a mitochondrial marker, as well as the examination of the N terminus for the presence of a MTS, was therefore investigated by examining the localization of PDIP38-EGFP fusion proteins. We constructed plasmids expressing full-length PDIP38 fusion proteins with EGFP fused to its C terminus (PDIP38-C-EGFP) or to its N terminus (EGFP-N-PDIP38). HeLa cells were transfected with these plasmids as well as the parent pEGFPN3 plasmid, and the live fluorescence images were captured by microscopy. In HeLa cells overexpressing PDIP38-C-EGFP (Fig. 2A, a), the PDIP38 fusion protein exhibited a punctated intracellular localization, whereas EGFP alone (Fig. 2A, b) as well as EGFP-N-PDIP38 (Fig. 2A, c) exhibited a diffused localization. HeLa cells were then transfected with PDIP38-C-EGFP or pEGFPN3 plasmids; the nuclei were stained with Hoechst 33258 and the mitochondria with MitoTracker Red CMXRos. PDIP38 with EGFP fused to its C terminus was clearly co-localized with MitoTracker (Fig. 2C, a–d), whereas EGFP alone was not (Fig. 2D, a–d). The observation is consistent with the presence of an MTS at the N terminus, because an EGFP tag attached to the N terminus of PDIP38 would mask the MTS, which has to be localized at the N terminus in order to function. Examination of the sizes of the PDIP38 EGFP fusion proteins by Western blotting of cell lysates (Fig. 2B) showed that PDIP38 tagged at its C terminus is smaller than the N-terminally tagged PDIP38 on SDS-PAGE, a finding that is consistent with the expected cleavage of an N-terminal MTS after entry into the mitochondria (Fig. 2B).
The First N-terminal 35 Amino Acids Are Sufficient to Mediate the Mitochondrial Localization of PDIP38 —In order to dissect the contribution of the N-terminal region of PDIP38 to mitochondrial localization, we constructed a series of N-terminal or C-terminal deletions of PDIP38 with C-terminal EGFP tags. The localization patterns of some of these EGFP fusion proteins are shown in Fig. 3 and summarized in Table I. Deletion of the first N-terminal 30 amino acids abolished the mitochondrial localization of the EGFP fusion protein (Fig. 3A), showing that the first 30 amino acids are necessary for mitochondrial localization of PDIP38. All C-terminal deletions of PDIP38 with a C-terminal EGFP tag exhibited a mitochondrial localization (Fig. 3, B and C, and Table I). The presence of a functional MTS in the N terminus of PDIP38 was confirmed by fusion of the first 35 amino acids of human PDIP38 to the N terminus of EGFP. This fusion resulted in the localization of EGFP to the mitochondria (Fig. 3B). Thus, the first N-terminal 35 amino acids of PDIP38 are unequivocally shown to function as an MTS.
|
5-kDa N-terminal Region Containing the MTS—The MTS sequences of most nuclearly encoded mitochondrial proteins are processed by mitochondrial proteases during their import into the mitochondria, resulting in mature proteins that are smaller than their precursors. In our original studies, PDIP38 behaved as a protein of 38 kDa when Western-blotted in cell extracts, whereas the predicted molecular mass based on the sequence encoded by its cDNA was 42 kDa (22). These conflicting observations can now be explained as the likely consequence of mitochondrial processing. Analysis of the protein sequence by MITOPROTII (ihg.gsf.de/ihg/mitoprot.html, see Ref. 33) predicted a cleavage site between residues 58 and 59, whereas Program TargetP (www.cbs.dtu.dk/services/TargetP/) (34, 35) predicted a cleavage site between residues 50 and 51. Both predictions may be correct as many mitochondrial proteins are processed in a two-step process by MPP, followed by a second cleavage of eight residues by the MIP.
In order to obtain more information on the location of the cleavage site, we examined the sizes of full-length EGFP-tagged PDIP38 and a series of either N-terminal or C-terminal deletion mutants of PDIP38 with a C-terminally tagged EGFP when expressed in HeLa cells. As already shown (Fig. 2A), full-length PDIP38 tagged at its N terminus is blocked from entry into the mitochondria and exhibits a higher molecular weight than the corresponding C-terminally tagged full-length PDIP38, which is imported into the mitochondria (Fig. 2B). Examination of deletion mutants at the N terminus that were tagged with EGFP at the C terminus by expression in HeLa cells showed that these migrated at the same positions as their calculated molecular masses (data not shown), consistent with a failure to be imported into the mitochondria. The expression products of C-terminal PDIP38 deletions with a C-terminal EGFP tag, on the other hand, all migrated at positions smaller by 5–6 kDa than their predicted molecular masses (data not shown).
|
|
5–6 kDa indicates that the proteolytical cleavage of precursor PDIP38 removes between 50 and 60 residues of the N terminus. The consensus sequence for the MPP and MIP cleavage sites is shown in Fig. 4C (36–38). The region between 50 and 60 amino acids contains an XRXXS (RRHLS, residues 48–52) motif that is one of the substrate recognition sites (R-2 motif) for MPP with an expected cleavage between amino acid residues 50 and 51 (36–38). The sequence also appears to be one that could be further cleaved by MIP between residues 58 and 59 (R-10 motif).
The cleavage site of PDIP38 in vivo was mapped by construction of a series of plasmids for the expression of N-terminal deletions of PDIP38 carrying a C-terminal FLAG tag. These deletions were then expressed in 293T cells. The sizes of these fusion proteins were compared with that of the full-length PDIP38 fusion protein by Western blotting with a mouse monoclonal anti-FLAG antibody (Fig. 4D). The overexpressed PDIP38-C-terminal FLAG-tagged deletion mutants lacking the first 30 N-terminal amino acids yielded a fusion protein that was larger than the processed FLAG-tagged PDIP38-(compare Fig. 4D, lanes 2–3). This is consistent with our immunofluorescence data that shows that the first 30 amino acids are required for the import of PDIP38 into mitochondria (Fig. 3A), so that this mutant would not enter the mitochondria and would not be processed. Thus, the cleavage site must be greater than 30 amino acid residues from the N terminus.
The migration of processed PDIP38 C-terminal FLAG-tagged fusion protein was compared with a series of stepwise deletions at the N terminus that would not be processed and whose sizes can be expected to reflect their amino acid sequence. The migration of processed PDIP38 is similar to that of the deletion mutant lacking the first 50 residues and is clearly bracketed between the deletion mutants lacking the first 40 and 60 amino acids, respectively (Fig. 4D, lanes 2 and 4–6). Thus, our data indicate that the cleavage site most likely is located at about 50 amino acids from the N terminus of the PDIP38 precursor protein. Given the correspondence of the N-terminal region with the consensus for MPP cleavage, our data are consistent with the cleavage of PDIP38 by MPP between residues 50 and 51 (Fig. 4C) and also suggest that MIP is not involved in the processing of PDIP38.
|
The localization of PDIP38 to the mitochondria was also demonstrated by cell fractionation. Mitochondria were isolated from HeLa cells by the use of a mitochondria isolation kit (Sigma). The presence of mhsp70 (Fig. 5B, a) and the absence of PCNA (Fig. 5B, c) indicated the purity of the isolated mitochondrial pellet (Fig. 5B, lane 5). PDIP38 (Fig. 5B, d) is clearly present in the mitochondrial pellet. Furthermore, both mhsp70 and PDIP38 are not subject to proteinase K degradation of the mitochondrial fraction (Fig. 5B, lane 7), as would be expected if PDIP38 was located within the intact mitochondria. When Triton X-100 was added to disrupt the mitochondrial membranes, treatment of the isolated mitochondrial pellet with proteinase K led to the degradation of both PDIP38 and mhsp70 (Fig. 5B, lane 8). Another mitochondria marker protein, Tom20, which binds to the outer membrane through its N-terminal domain, was readily digested by proteinase K even in the absence of Triton X-100 (Fig. 5B, b, lane 7). Taken together, these results show that endogenous PDIP38 is present inside the mitochondria. The large subunit p125 (Fig. 5B, e), small subunit p50 (Fig. 5B, f), third subunit p68, and fourth subunit p12 (data not shown) of pol were not present in the mitochondrial fraction as determined by Western blotting.
Is PDIP38 Also Present in the Nucleus?—Our fractionation results using the mitochondria isolation kit (Sigma) indicated that a significant amount of PDIP38 was also present in the nuclear fraction (Fig. 5B, lane 3, d). This is in contrast to our immunofluorescence results, which showed that both endogenous and overexpressed C-terminally tagged EGFP PDIP38 were almost exclusively present in the cytoplasm. Because we could also detect mhsp70 (Fig. 5B, lane 3, a) and IB- (Fig. 5B, lane 3, h) in the same nuclear fraction, it is possible that the nuclear fraction of HeLa cells isolated using this method was contaminated with unbroken cells or with mitochondrial proteins. We next fractionated the cells using the NE-PER kit from Pierce. We monitored our fractionation procedure using nuclear (lamin B) and cytoplasmic (IB-, mhsp70) marker proteins. The exclusive presence of lamin B in the nuclear fraction (Fig. 6A) and IB- in the cytoplasm fraction (Fig. 6B) indicated that fractionation provided a clear separation of nuclear and cytoplasmic fractions. We observed that a small amount, about 5% of total PDIP38 (Fig. 6C), was nevertheless still present in the nuclear fraction. There was also a very minor fraction of mhsp70 in the nuclear fraction. Thus, it is possible that the PDIP38 found in the nuclear fraction was because of incomplete removal of the mitochondria. Nevertheless, our data do not exclude the possibility that PDIP38 may be present in the nucleus, but if it were, it is only a minor fraction (5%) of the total pool of PDIP38.
|
|
Activity in Vitro—C-terminal His-tagged PDIP38 lacking the first N-terminal 50 amino acids (to mimic the endogenously mature protein) was overexpressed in E. coli and purified by Ni-NTA-agarose and Superdex 200 gel filtration column (Fig. 7A). PDIP38-(51–368)-C-His eluted at a position with a molecular mass of 34 kDa, as determined by comparison to protein standards on Superdex 200 using a semilog plot of Mr versus elution volume. This is similar to the calculated molecular mass of PDIP38-C-His of 34 kDa. We first determined if the N-terminally truncated PDIP38 was able to interact with p50 in vitro. GST-p50 was used in a pull down assay and was able to pull down the PDIP38 mutant that had been truncated by removal of the N-terminal 50 amino acid residues (Fig. 7B). This shows that mature PDIP38 retains the ability to interact with the p50 subunit of pol . We next assessed the effects of mature PDIP38 on HeLa pol activity. His-tagged p21 was used as positive control in our assay (39). His-tagged p21 at the concentration of 800 nM inhibited up to 80% of the pol activity in the presence of PCNA (138 nM monomer). Under the same assay conditions, we consistently observed that His-tagged PDIP38 at the concentration of 1.6 µM inhibited the PCNA-dependent pol activity by about 50% (Fig. 7C). This observation is consistent with our earlier finding that PDIP38 was able to interact with PCNA in vitro (22). We also assessed if PDIP38 has any effect on the pol activity in the absence of PCNA, and we found there was no appreciable effect observed (data not shown).
PDIP38 Interacts with HPV 16 E7—pol is well known to replicate several viral DNAs both in vitro and in vivo (40–42). It has been reported that HPV 18 E6/E7 expression significantly up-regulates the transcription of both p68 and p50 subunits of pol by 6.03- and 5.23-fold, respectively (43). HPV E6 and E7 proteins are two viral proteins expressed in HPV-associated human cancers (44). E7 binds to retinoblastoma protein and is involved in the transformation of cells by the HPV. We determined that GST-tagged HPV 16 E7 was able to interact with pol as well as PDIP38 in total HeLa cell lysates. Our results indicated that GST-HPV 16 E7 oncoprotein could specifically pull down PDIP38 as well as the p125 subunit of pol from total HeLa cell lysates (Fig. 8A). Moreover, both GST-p50 and GST-HPV 16 E7 could specifically pull down purified mature PDIP38 (Fig. 8B), suggesting that PDIP38 could serve as a bridge protein connecting HPV E7 to the pol complex.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
as the bait. Further investigation by a number of methods confirmed the interaction between PDIP38 and p50, and this interaction could take place when p50 was part of the pol complex (22). Furthermore, an interaction of PDIP38 and the pol complex in cell lysates was readily demonstrated in cell extracts. In addition, PDIP38 also interacted with PCNA, which functions as a DNA sliding clamp that binds pol (22). These strong interactions suggested that PDIP38 might be a nuclear protein involved in some aspect of DNA metabolism involving pol . In this study, we show rigorously that PDIP38 is primarily a mitochondrial protein. We have unequivocally demonstrated that PDIP38 possesses a functional MTS at its N terminus. In addition, our studies show that the cleavage site for the mitochondrial processing of PDIP38 is located at 50 residues from the N terminus. This agrees with the presence of a predicted cleavage site for mitochondrial processing protease between residues 50 and 51 (MPP, R-2 motif; see Refs. 36–38). MTSs with MPP cleavage sites conforming to a 10-residue consensus sequence are often cleaved by a second protease, MIP, by the removal of an octapeptide at the N terminus. Our data indicate that PDIP38 is not further processed by MIP. Although the sequence of PDIP38 conforms very well with the 10-residue consensus (Fig. 4C), a conserved Thr, Ser, or Gly at the +4 position from the MPP site is replaced by a Arg. This might account for the failure for PDIP38 to be cleaved by MIP. The exact cleavage site of mature PDIP38 in vivo needs to be determined by Edman N-terminal sequencing of purified endogenous PDIP38.
|
that can be demonstrated in vitro, because this might be involved in signaling of mitochondrial damage to the nucleus. We could not detect any subunits (p125, p50, p68, or p12) of human pol in the mitochondrial pellet; therefore, the role of PDIP38 within the mitochondria is not likely mediated through an interaction with pol . Another possibility is that PDIP38 might be involved in mitochondrial DNA replication or DNA repair, through interaction with mitochondrial DNA replication proteins such as DNA polymerase 
. PCNA, which is generally considered to be a nuclear protein and whose functions are in the nucleus, has been reported to be able to stimulate mitochondrial polymerase activity (45). A more recent study (46) showed that monoclonal antibodies against PCNA specifically recognized a single band of 30 kDa in the mitochondria from the filamentous fungus and yeast. These studies suggest that PDIP38 might function in the mitochondria through its interaction with PCNA, even though the level of PCNA in mammalian mitochondria might be too low to be detected by Western blotting. Our results indicate that the association of PDIP38 with human pol might occur only when triggered by certain cellular events, the most likely of which are conditions that favor mitochondrial damage; this might include conditions that result in activation of DNA damage response or DNA repair.
There are several precedents for proteins that are involved in DNA replication and/or DNA repair to be localized to both the mitochondria and the nucleus. For example, human APE2 protein was shown to localize both to the mitochondria and the nucleus, and like PDIP38, the nuclear human APE2 protein also interacts with PCNA (47). Other proteins that were shown to be localized both in the mitochondria and the nucleus are human 8-oxoguanine DNA glycosylase OGG1 (48), Ku80 (49), human DNA topoisomerase III (50), human uracil-DNA glycosylase (51, 52), and yeast DNA ligase I (53). In most cases, differential initiation sites or alternative splicing of pre-mRNA accounted for the differential intracellular localization of these proteins. In this context, it may be noted that Northern blots of total RNA from both HeLa and MCF7 cells indicated that there are two transcripts for PDIP38 with a major band of 2.0 kb and a minor band of 1.2 kb (22). Thus, there remains a possibility that other isoforms of PDIP38 that lack the N-terminal region of MTS may be localized to the nucleus. In addition, it is noted that the antibody we used was directed to the C-terminal 20 residues of PDIP38 and might not recognize all of the PDIP38 forms that may be expressed. Further investigation is needed to explore these possibilities.
The mitochondrial localization of human PDIP38 does raise the question of whether the interactions we reported between PDIP38 and p50 subunit of pol as well as PCNA occur under physiological conditions (22), given that both pol and PCNA mainly function in the nucleus. Although we cannot completely rule out the possibility that the interactions between PDIP38 and pol that are observed experimentally in cell extracts are due to the disruption of the mitochondria, we favor the idea that PDIP38 does contribute directly to certain DNA metabolic processes mediated by pol . Mature PDIP38 was capable of interacting with the p50 subunit of pol (Fig. 7B) and inhibited pol activity using a PCNA-dependent assay (Fig. 7C).
Another possibility is that PDIP38 might participate in viral DNA replication mediated by pol . Most viruses use cellular components to replicate their genomes in the nucleus (54). Initial studies in our laboratory show that GST-tagged HPV (human papillomavirus) 16 E7 oncoprotein can pull down PDIP38 as well as p125 of pol from total HeLa cell lysates (Fig. 8A). In addition, in vitro experiments show that PDIP38 interacts with both p50 and HPV 16 E7 (Fig. 8B), suggesting that PDIP38 may be able to act as a bridge protein between HPV E7 and the pol complex. Raj et al. (55) reported that HPV E1^E4 protein could associate with mitochondria and lead to the perturbation of the cytokeratin network, disrupt the normal mitochondrial distribution, and reduce the mitochondrial membrane potential. It is possible that PDIP38 would move from the mitochondria to the nucleus after HPV infection, possibly through either structural links between mitochondrial and nuclear membranes or release from mitochondria after the reduction of the mitochondrial membrane potential (56). HPV 16 E7 was reported to interact with a number of cellular proteins including hTid1, a human homolog of the Drosophila tumor suppressor lethal (2) tumorous imaginal discs, to allow the virus to replicate in differentiating epithelial cells (44). hTid1 is largely localized to the mitochondrial matrix and encodes two alternatively spliced isoforms that have opposing effects on apoptosis (57). The mouse mTid1 gene encodes three alternatively spliced isoforms that are localized to the cytosol, mitochondria, or nucleus depending on the cell type (58). hTid-1 was also reported to interact with herpes simplex virus, type 1 (HSV-1), UL9 protein and to play a role in HSV-1 DNA replication (59). It is not clear how hTid1 translocates from the mitochondria to the other cellular compartments during viral infection. Another finding that links PDIP38 with pol and viral replication is a report that PDIP38, p12, and p50 subunits of pol were also able to interact with the Epstein-Barr virus replication protein BBLF2/3 (60), which supports a possible role of PDIP38 in pol -mediated viral replication.
In summary, we have localized human PDIP38 to the mitochondria and demonstrated that it possesses a functional mitochondrial targeting sequence. We also demonstrate that mature PDIP38 is most likely processed by removal of the first N-terminal 50 amino acids during its entry into the mitochondria. Our findings suggest that PDIP38 may possess functions in the mitochondria that may not be related to its ability to associate with pol and PCNA. In this regard, P12, the fourth subunit of pol , was recently suggested to possess additional, cell proliferation-unrelated functions (61). Our results raise the possibility that PDIP38 may have functions that include interaction with the nuclear DNA replication system or in the linkage between pol and viral DNA replication.
|
FOOTNOTES |
|---|
Both authors contributed equally to this work.
|| To whom correspondence should be addressed: Dept. of Biochemistry and Molecular Biology, New York Medical College, Valhalla, NY 10595. Tel.: 914-594-4070; Fax: 914-594-4058; E-mail: Marietta_Lee{at}nymc.edu.
1 The abbreviations used are: pol, polymerase; PDIP, polymerase interacting protein; EGFP, enhanced green fluorescent protein; PCNA, proliferating cell nuclear antigen; MTS, mitochondrial targeting sequence; mhsp70, mitochondrial heat shock protein 70; MPP, mitochondrial processing peptidase; MIP, mitochondrial intermediate peptidase; HPV, human papillomavirus; GST, glutathione S-transferase; Ni-NTA, nickel nitrilotriacetic acid; PBS, phosphate-buffered saline.
|
ACKNOWLEDGMENTS |
|---|
. We also thank Dr. K. Munger at Harvard Medical School for providing us the GST-HPV 16 E7 construct. |
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
30-
