This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Chang, P.-J.
Right arrow Articles by Miller, G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chang, P.-J.
Right arrow Articles by Miller, G.

 Previous Article  |  Next Article 

Journal of Virology, October 2004, p. 10657-10673, Vol. 78, No. 19
0022-538X/04/$08.00+0     DOI: 10.1128/JVI.78.19.10657-10673.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.

Autoregulation of DNA Binding and Protein Stability of Kaposi's Sarcoma-Associated Herpesvirus ORF50 Protein

Pey-Jium Chang1 and George Miller1,2,3*

Departments of Molecular Biophysics and Biochemistry,1 Pediatrics,2 Epidemiology and Public Health, Yale University School of Medicine, New Haven, Connecticut3

Received 18 March 2004/ Accepted 25 May 2004


arrow
ABSTRACT
 
A transcriptional activator encoded in open reading frame 50 (ORF50) of Kaposi's sarcoma-associated herpesvirus (KSHV) initiates the viral lytic cycle. ORF50 protein activates downstream KSHV target genes by at least two mechanisms: direct recognition of response elements in promoter DNA and interaction with cellular proteins bound to promoter DNA. We have identified a multifunctional regulatory region, present in amino acids (aa) 520 to 535 of ORF50 protein, that controls DNA binding and protein stability. Deletion of aa 521 to 534 or mutation of a basic motif (KKRK) in this regulatory region dramatically enhances DNA binding by ORF50 protein, as shown by electrophoretic mobility shift, DNA affinity chromatography, and chromatin immunoprecipitation assays. Deletion of the regulatory region and mutations in the KKRK motif also lead to abundant expression of an electrophoretic mobility variant, ORF50B, which appears to be a form of ORF50 protein that is decreased in posttranslational modification. Enhanced DNA binding and enhanced expression of ORF50B are independent phenomena. The regulatory region likely inhibits DNA binding through interactions with the DNA binding domain in aa 1 to 390 and destabilizes ORF50B through interactions with a domain located in aa 590 to 650. Mutants in the KKRK motif that are enhanced in DNA binding are nonetheless impaired in activating direct targets, such as polyadenylated nuclear RNA, and indirect targets, such as ORF50 itself. The identification of an autoregulatory region emphasizes that the many functions of ORF50 protein must be subject to exquisite control to achieve optimal KSHV lytic-cycle gene expression.


arrow
INTRODUCTION
 
Kaposi's sarcoma (KS)-associated herpesvirus (KSHV), or human herpesvirus 8, a gammaherpesvirus that is closely related to Epstein-Barr virus (EBV), herpesvirus saimiri, rhesus monkey rhadinovirus, and murine gammaherpesvirus 68 (1, 6, 37, 40, 48), has been implicated in the etiology and pathogenesis of KS, primary effusion lymphoma (PEL), and multicentric Castleman's disease (2, 4, 6, 45). Like all herpesviruses, KSHV manifests two distinct phases of its life cycle, latency and lytic replication. Most KS tumor cells are latently infected and exhibit limited viral gene expression. The viral gene products expressed during latency include latency-associated nuclear antigen (LANA), v-cyclin, v-FLIP, kaposin, and vIRF-2 or LANA-2 (3, 11, 24, 34, 36). Although latent infection plays a role in viral persistence and tumorigenesis, lytic reactivation is also important for KS development. Antibodies to lytically expressed proteins are significantly elevated in the sera of patients developing KS (12). Treatment with ganciclovir, a drug that is active against KSHV lytic but not latent replication, markedly reduces the incidence of KS development in AIDS patients (31). Since lytic replication allows the virus to spread from cell to cell and among individuals, it is important to understand control of lytic-cycle gene expression for providing insights into KSHV pathogenesis.

The switch between latency and lytic-cycle gene expression of KSHV is initiated by a single immediate-early protein encoded by open reading frame 50 (ORF50) of the viral genome (30, 46). Ectopic expression of ORF50 protein is necessary and sufficient to disrupt viral latency and activate lytic replication to completion in KSHV-infected PEL cell lines (13, 30, 46). No significant homology between the ORF50 protein and cellular proteins has been identified. However, ORF50 protein is related to immediate-early transcriptional activator proteins of other gammaherpesviruses, such as ORF50A protein in herpesvirus saimiri, ORF50 protein in rhesus monkey rhadinovirus, and Rta encoded by the BRLF1 gene of EBV (10, 19, 30, 46, 52).

The ORF50 protein is a 691-amino-acid (aa) nuclear polypeptide that is extensively phosphorylated and may be subject to other posttranslational modifications (29, 30). Transient transfection experiments have demonstrated that ORF50 protein activates the promoters of a number of viral genes that function in lytic replication, including its own gene, polyadenylated nuclear (PAN) RNA, K12 (kaposin), ORF57, K8 (K-bZIP), K9 (vIRF), ORF21 (thymidine kinase), K5, ORF6 (single-stranded DNA binding protein), K14 (vOX-2), ORF74 (vGPCR), and K2 (vIL6) (5, 7-9, 13, 18, 22, 28, 29, 38, 42, 44, 53). Although the molecular means by which ORF50 protein activates target genes during lytic replication are not fully understood, emerging evidence suggests that ORF50 protein controls promoters of its targets by several distinct mechanisms. One compelling piece of evidence in support of this hypothesis is that most ORF50 response elements identified in target promoters do not share conserved DNA sequences (5, 28, 43, 47).

Activation of one class of targets, such as the PAN and K12 genes, by ORF50 protein operates mainly through a direct DNA binding mechanism. In our previous studies, we found that ORF50 protein (aa 1 to 490) expressed in human cells bound similar ORF50 response elements found in the promoters of PAN and K12. Mutation of the response elements showed that direct DNA binding correlated with transcriptional activation (5). The specific interaction between ORF50 protein and the ORF50 response elements in the PAN and K12 promoters has also been demonstrated using ORF50 protein purified from Escherichia coli or in vitro translated ORF50 protein (42, 44, 50). The vIL6 gene also appears to be activated by ORF50 protein through a direct DNA binding mechanism. However, the conserved element in the PAN and K12 promoters is not present in the promoter of the vIL6 gene (8). Although purified ORF50 protein expressed in E. coli or insect cells bound a 12-base palindrome sequence found in the ORF57 and K8 promoters, the interaction between ORF50 protein and ORF57/K8 elements was weak and was only detected under limited conditions (28, 39, 43, 50).

Direct DNA binding by ORF50 protein to its response elements is not the only mechanism for transactivation. Activation of another class of target promoters is modulated through indirect access of ORF50 protein to DNA via protein-protein interaction. Recently, Liang et al. found that activation of ORF57 and ORF6 promoters by ORF50 protein was dependent on intact RBP-J{kappa} binding sites in these promoters and the expression of RBP-J{kappa} protein, the target of the Notch signaling pathway (25). Since ORF50 protein directly interacts with RBP-J{kappa} in vitro and in vivo, it was suggested that ORF50 protein binding to target DNA was mediated through RBP-J{kappa} protein. The diversity in DNA sequences of ORF50 response elements from different promoters implies that other cellular or viral proteins may serve as mediators for interacting with ORF50 protein. Several cellular proteins have been shown to bind DNA in target promoters of ORF50 protein. These include OCT1 and C/EBP{alpha} in the ORF50 promoter (38, 51), Sp1 in the thymidine kinase promoter (53), C/EBP{alpha} in the K8 promoter (50), and unknown proteins in the K9 promoter (47). Whether all these cellular proteins mediate gene activation by direct contact with ORF50 protein remains to be determined.

Interaction between ORF50 protein and cellular proteins regulates the function of ORF50 protein. ORF50 protein recruits CREB binding protein, the SWI/SNF chromatin remodeling complex, and the TRAP/Mediator coactivator to viral promoters through direct interactions (14, 15). An ORF50 binding cellular cofactor protein, MGC2663, can enhance the capacity of ORF50 protein to activate viral promoters in reporterassays (49). The association of ORF50 protein with histone deacetylase-1, poly(ADP-ribose) polymerase 1 (PARP1), and the Ste20-like kinase hKFC has been proposed to negatively regulate ORF50 function (15, 16).

In this report, we identify a multifunctional regulatory region of the ORF50 protein that is likely to play an important role in controlling the distinct direct and indirect modes of action of ORF50 protein and probably impinges on the multitude of interactions between ORF50 protein and cellular proteins. The existence of this autoregulating region further emphasizes the complexity of ORF50's many distinct activities in KSHV lytic-cycle gene activation and suggests that elaborate controls are required to achieve highly regulated lytic viral gene expression.


arrow
MATERIALS AND METHODS
 
Cell cultures and transfections. BJAB cells are derived from an EBV-negative B-cell lymphoma (32). HKB5/B5 is an EBV-negative cell line formed by fusion of HH514-16 cells with 293T cells (kindly provided by M.-S. Cho, Bayer Corporation, Berkeley, Calif.). HH-B2 cells are derived from a PEL infected with KSHV (13). All cells were grown in RPMI 1640 medium plus fetal calf serum. BJAB and HH-B2 cells were transfected by electroporation, and HKB5/B5 cells were transfected using the DMRIE-C reagent (Invitrogen) (5).

Plasmid construction. pRTS, pRTS/ORF50 (gRta), pPANp(–91 to –58)/E4CAT, and pK12p(–105 to –72)/E4CAT have been described previously (5, 46). To make ORF50 deletion mutants, various regions of the ORF50 gene were amplified by PCR and cloned into pRTS with XbaI and BglII sites. The series of ORF50 deletions fused to VP16 was generated by cloning a DNA fragment encoding the VP16 activation domain (AD) and the simian virus 40 T-antigen nuclear localization signal from pVP16AD (CLONTECH) downstream of portions of the ORF50 gene. pCMV-FLAG-ORF50 (kindly provided by Ren Sun, University of California at Los Angeles) contains DNA sequences encoding FLAG peptide upstream of ORF50 cDNA (44). Single or double amino acid point mutations were introduced into the ORF50 gene in pRTS/ORF50 or in pCMV-FLAG-ORF50 using the QuikChange site-directed mutagenesis kit (Stratagene).

CAT assays. BJAB cells (1.2 x 107) were transfected with 5 µg of activator (ORF50 derivatives) or vector DNA and 5 µg of reporter DNA. Chloramphenicol acetyltransferase (CAT) activity was determined as described previously (41). Activation was calculated as the percent acetylation of chloramphenicol in the presence of the activator divided by the percent acetylation in the presence of the vector.

Protein isolation for electrophoretic mobility shift assay (EMSA) and Western blot analysis. Extracts of total cell proteins were prepared by the method of Mosser et al. (33). Briefly, 107 transfected cells were suspended in 150 µl of lysis buffer (0.42 M NaCl, 20 mM HEPES [pH 7.5], 25% glycerol, 1.5 mM MgCl2, 0.2 mM EDTA, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, 2 µg of aprotinin per ml). Lysates were spun at 90,000 rpm at 4°C for 15 min in a benchtop ultracentrifuge; supernatants were aliquoted and stored at –80°C. Protein concentrations were determined by the Bradford method.

EMSAs. Annealed double-stranded oligonucleotides derived from the PAN promoter (nucleotides –91 to –58) were end-labeled with 32P using T4 polynucleotide kinase. Binding reaction mixtures contained 15 µg of cell protein extract in a solution containing 10 mM HEPES (pH 7.5), 50 mM NaCl, 2 mM MgCl2, 2.5 µM ZnSO4, 0.5 mM EDTA, 1 mM dithiothreitol, 15% glycerol, and 1 µg of poly(dI-dC) in a total volume of 20 µl. After incubation for 5 min at room temperature, labeled DNA was added. Unlabeled consensus YY1 oligonucleotides (100-fold molar excess) were added to the initial reaction mixture to remove the YY1/PANp complex. The DNA sequence of the YY1 cold competitor is 5'CGCTCCGCGGCCATCTTGGCGGCTGGT. For supershift assays, antisera were added 10 min after the addition of the probe. After a further 10-min incubation at room temperature, the reactions were loaded onto a 0.5x Tris-borate-EDTA native 4% polyacrylamide gel. After electrophoresis at 200 V for 2 h, gels were transferred to Whatman 3MM paper, dried, and exposed to autoradiography film.

Western blot analysis. Cell protein extracts were mixed with 3x sodium dodecyl sulfate (SDS) gel loading buffer and boiled for 5 min before loading on SDS-8 or 10% polyacrylamide gels. After immunoblotting was performed as described previously (5), immunoreactive bands were detected with 125I-labeled protein A. Antibodies to VP16 (sc-7545; Santa Cruz), YY1 (sc-7341; Santa Cruz), and FLAG (M2; Sigma) were obtained commercially. Anti-ORF50 rabbit serum was generated by using ORF50 protein (aa 333 to 691) expressed in E. coli as an immunogen.

Northern blot analysis. Total cellular RNAs from 1.2 x 107 transfected cells were prepared with an RNeasy kit (QIAGEN), fractionated on 1% formaldehyde-agarose gels, and transferred to nylon membranes (Hybond-N+; Amersham Pharmacia Biotech). All probes were labeled by the random-primed method (46). For ORF50 mRNA and PAN RNA detection, DNA fragments corresponding to nucleotides 72599 to 74626 and 28972 to 29546 of the KSHV genome (37), respectively, were used. Hybridization was carried out in a solution containing 6x SSC (1x SSC is 0.15 M NaCl plus 0.015 M sodium citrate), 5x Denhardt's solution, 0.5% SDS, and 100 µg of salmon DNA per ml at 65°C overnight. Membranes were washed in 2x SSC-0.5% SDS once for 10 min and in 0.1x SSC-0.5% SDS three times for 25 min at 65°C.

DNA affinity chromatography. To prepare a PANp oligonucleotide affinity column, complementary oligonucleotides representing nucleotides –91 to –58 of the PAN promoter were annealed, 5'-phosphorylated, and ligated as described by Kadonaga and Tjian (23). The DNA oligomers were covalently attached to CNBr-activated Sepharose 4B (Pharmacia). Protein extracts were prepared from 2.4 x 107 transfected HKB5/B5 cells. The transfected cells were resuspended in 450 µl of the same lysis buffer used for preparing cell extracts for EMSA. Cell lysates (400 µl) were diluted by adding 7.4 volumes (2,960 µl) of buffer D (10 mM HEPES [pH 7.5], 2 mM MgCl2, 2.5 µM ZnSO4, 0.5 mM EDTA, 1 mM dithiothreitol) to reduce the NaCl concentration to 50 mM. The diluted protein extracts were incubated with 200 µg of poly(dI-dC) for 5 min at room temperature and loaded on the PANp affinity column. Proteins were eluted stepwise with 500 µl of buffer D containing 0.1 to 2 M NaCl. Forty microliters of each eluted fraction was used for detection of ORF50 and YY1 proteins by immunoblotting, and 4 µl of each eluted fraction was examined by EMSA.

Chromatin immunoprecipitation (ChIP) assay. HH-B2 cells (107) were transfected by electroporation with 10 µg of plasmids expressing ORF50 protein or ORF50 mutants. Twenty hours after transfection, cultured cells were incubated with 1% formaldehyde for 10 min at 37°C. The cells were harvested, washed once with phosphate-buffered saline containing 1 mM phenylmethylsulfonyl fluoride, 1 µg of aprotinin per ml, and 1 µg of pepstatin A per ml, and resuspended in 1.5 ml of lysis buffer (1% SDS, 10 mM EDTA, 50 mM Tris-Cl [pH 8.1]). Cell lysates were divided into 200-µl aliquots and sonicated. After a 10-min centrifugation at 12,000 x g at 4°C, 200 µl of the supernatant from the sonicated cell mixtures was diluted with 1.8 ml of dilution buffer (0.01% SDS, 1.1% Triton X-100, 1.2 mM EDTA, 16.7 mM Tris-Cl [pH 8.1], 167 mM NaCl). Before the addition of antibodies, the cell mixtures were precleared with 60 µl of 50% protein G-agarose slurry containing 20 µg of salmon sperm DNA and 1 mg of bovine serum albumin per ml for 1 h at 4°C. For each immunoprecipitation, 10 µg of anti-FLAG antibody, 6 µg of anti-YY1 antibody, or 6 µg of anti-VP16 antibody was added into 2-ml aliquots of cell lysate and incubated at 4°C overnight. The immune complexes were precipitated by adding 60 µl of 50% protein G-agarose and washed five times using 1 ml of the buffers listed below: (i) 0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-Cl (pH 8.1), and 150 mM NaCl; (ii) 0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-Cl (pH 8.1), and 500 mM NaCl; (iii) 0.25 M LiCl, 1% NP-40, 1% sodium deoxycholate, 1 mM EDTA, and 10 mM Tris-Cl (pH 8.1); (iv) 10 mM Tris-Cl (pH 8) and 1 mM EDTA; (v) 10 mM Tris-Cl (pH 8) and 1 mM EDTA. For eluting chromatin complexes from antibodies, the pelleted protein G complex was eluted twice using 200 µl of elution buffer (1% SDS, 0.1 M NaHCO3) for 15 min at room temperature. The eluted chromatin complexes (400 µl) were treated with 20 µl of 5 M NaCl at 65°C for 4 h to reverse protein-DNA cross-links. DNA was recovered by phenol-chloroform extraction and ethanol precipitation. Fifteen percent of the precipitated DNA was used as a template in PCR. The primers for amplification of the PAN promoter region were pan-a (5'GGTGACCCAACATAGTGATTCGG) and pan-b (5'CAGTGCTAAACTGACTCAAGCTG). The primers for PCR detection of the K12 promoter were k12-a (5'AGGGGGGTCGGTCTCCCCTCTTC) and k12-b (5'CCGCCACTCCAGCCGTGTTAGCC).


arrow
RESULTS
 
The C-terminal region of ORF50 protein inhibits DNA binding activity in vitro. A previous study showed that ORF50(1-490)+VP, a protein formed by fusion of ORF50 aa 1 to 490 with the activation domain of herpes simplex viral protein 16, expressed in transfected cells binds the ORF50 response elements upstream of the PAN and K12 genes (5). Extensive mutagenesis of both ORF50 response elements showed a strong concordance of DNA binding in vitro with transcriptional activation in vivo. Constructs encoding ORF50(1-290)+VP, ORF50(1-250)+VP, ORF50(16-490)+VP, and ORF50(91-490)+VP did not bind DNA in vitro and were not able to activate PANp(–91 to –58)/E4CAT or K12p(–105 to –72)/E4CAT reporters in cells. Thus, direct DNA binding of ORF50 protein to the promoters of the PAN and K12 genes was essential for the transcriptional activation of both genes (Fig. 1).



View larger version (14K):
[in this window]
[in a new window]
  		FIG. 1. The C-terminal region of ORF50 protein inhibits DNA binding activity. (Left) Diagram of wild-type ORF50 and ORF50 deletion mutants used in EMSAs and reporter assays. Putative domains are indicated as follows: NLS, nuclear localization signal; LZ, leucine zipper. The VP16 ADs (black boxes) in ORF50 deletions also served as tags for recognition by VP16-specific antibody in EMSAs. Extracts of HKB5/B5 cells transfected with plasmids expressing full-length, deletion, and point (R160K) mutants of ORF50 fused to VP16 were prepared 48 h after transfection. The probe used for EMSA was the PAN promoter (–91 to –58). Transcriptional activation of the PANp(–91 to –58)/E4CAT and K12p(–105 to –72)/E4CAT reporters by ORF50 and ORF50 deletion mutants fused to VP16 was analyzed in BJAB cells. The fold activation was calculated by comparing the CAT activity of ORF50 construct-transfected cells to that of vector-transfected cells. M, mutation at R160.

A previous study reported that ORF50(1-390)+VP did not bind DNA. Subsequently, we found a single-point mutation in this clone, which changed amino acid 160 from arginine (AGG) to lysine (AAG). While this mutant protein did not bind DNA, correction of the R160K mutation in ORF50(1-390)+VP restored DNA binding activity in EMSA and transactivation function (Fig. 1; see below). This result suggested that residue R160 in the DNA binding domain of ORF50 protein played an important role in recognition of ORF50 response element DNA. Furthermore, the result mapped a minimal DNA binding domain of ORF50 protein expressed in mammalian cells to aa 1 to 390.

In EMSA experiments, by using extracts of transfected cells, the association between ORF50(1-490)+VP or ORF50(1-390)+VP and ORF50 response elements of the PAN and K12 promoters was easily detected (Fig. 2B and C and 3B). However, using the same conditions we were unable to detect a specific interaction between full-length ORF50 protein, with or without a VP16 tag, and PAN promoter DNA (Fig. 2B and 3B). This result raised the possibility that the C-terminal region of ORF50 protein, aa 491 to 691, played an inhibitory role with respect to ORF50 DNA binding activity.



View larger version (50K):
[in this window]
[in a new window]
  		FIG. 2. A region that inhibits the DNA binding activity of ORF50 is located between aa 520 and 535. (A) Summary of DNA binding and transactivation function of C-terminal deletions of ORF50 fused to VP16. The transactivation function of ORF50 deletions on reporter PANp(–91 to –58)/E4CAT was analyzed in BJAB cells. The fold activation was obtained as described in the legend to Fig. 1. NLS, nuclear localization signal; LZ, leucine zipper. (B and C) (Top panels) HKB5/B5 cells transfected with the indicated plasmids were harvested at 48 h posttransfection, and cell extracts were prepared for EMSAs. The probe for EMSAs was PANp(–91 to –58). Antibody to VP16 was used for supershift tests. The ORF50-specific complexes and the supershifted complexes (SS) are shown. (Bottom panels) Protein expression of ORF50 deletions in cell extracts detected by antibody to VP16. pRTS is an empty vector control.



View larger version (75K):
[in this window]
[in a new window]
  		FIG. 3. Identification of the DNA binding inhibitory region using C-terminal deletions of ORF50 protein without a VP16 fusion. (A) Summary of the DNA binding activity of wild-type ORF50 and the C-terminal-deletion mutants. NLS, nuclear localization signal; LZ, leucine zipper. (B to D) EMSAs were performed using the same conditions described in the legend to Fig. 2. Anti-peptide antibody to ORF50(230-250) was used to remove the ORF50/DNA complex. Annealed double-stranded consensus YY1 oligonucleotides (YY1 oligos) served as a cold competitor for removing the cellular YY1 complex.

To determine whether an inhibitory region in the C-terminal 200 aa regulated DNA binding activity, we made a series of C-terminal-truncated ORF50 proteins fused to the VP16 AD (Fig. 2A). The VP16 AD in the ORF50 constructs served as a tag for protein recognition in Western blot analysis and in EMSA and also maintained the transactivation function of ORF50/VP16 chimeras from which the natural AD of ORF50 protein had been removed. The full-length ORF50 that was fused to the VP16 AD, ORF50(1-691)+VP, did not bind PANp(–91 to –58), although it manifested a very strong transactivation function in cells. Likewise, constructs ORF50(1-563)+VP and ORF50(1-535)+VP still exhibited transactivation function in reporter assays, but they could not bind PANp in our EMSA conditions. However, constructs ORF50(1-520)+VP and ORF50(1-505)+VP did show DNA binding activity in EMSA and also retained transactivation function. These experiments suggested that a DNA binding inhibitory region was located in aa 520 to 535 of ORF50 protein. The restoration of DNA binding ability in the mutants ORF50(1-520)+VP, ORF50(1-505)+VP, ORF50(1-490)+VP, and ORF50(1-390)+VP was unrelated to their protein expression level; all ORF50 mutants were expressed similarly in cell extracts (Fig. 2B and 2C, bottom panels).

To rule out the possibility that the VP16 AD in these constructs was affecting ORF50 DNA binding ability, we made the same deletion constructs shown in Fig. 2A without the VP16 AD (Fig. 3). Full-length ORF50, ORF50(1-563), and ORF50(1-535) did not bind PANp(–91 to –58) DNA. However, ORF50(1-520), ORF50(1-505), ORF50(1-490), and ORF50(1-390) revealed strong DNA binding ability in EMSA. Thus, the VP16 AD did not influence DNA binding.

Internal deletion of the inhibitory region in the full-length ORF50 enhances ORF50 DNA binding activity and selectively increases the level of a variant ORF50 protein. C-terminal truncations might remove more than one inhibitory region. To determine whether removal of aa 520 to 535 by itself restored DNA binding and to confirm the importance of this inhibitory region in controlling ORF50 DNA binding activity in the context of the full-length protein, we made two internal-deletion constructs, {Delta}506-534 and {Delta}521-534, which removed 29 and 14 amino acids, respectively (Fig. 4A). Unlike full-length ORF50 protein, both internal-deletion proteins expressed in HKB5/B5 cells bound DNA in EMSA (Fig. 4B). Antibody to ORF50 removed the complexes ORF50({Delta}506-534)/PANp and ORF50({Delta}521-534)/PANp.



View larger version (39K):
[in this window]
[in a new window]
  		FIG. 4. Internal deletion of the inhibitory region restores ORF50 DNA binding activity and results in abundant expression of ORF50 form B. (A) Schematic diagram of ORF50 showing location and amino acid sequence of the inhibitory region. The amino acid residues deleted in two ORF50 mutants are indicated by brackets ({Delta}506-534 and {Delta}521-534). NLS, nuclear localization signal; LZ, leucine zipper. (B) An EMSA using PANp(–91 to –58) as a probe showed that ORF50 deletion mutants {Delta}506-534 and {Delta}521-534 restore DNA binding activity. Antibody to ORF50(230-250) was used to remove specific ORF50 complexes. (C) Protein expression by ORF50 deletion mutants. ORF50 expression in cell extracts prepared for EMSA was analyzed by immunoblotting with ORF50 polyclonal antibody. The abundant unidentified form of ORF50 produced by mutants {Delta}506-534 and {Delta}521-534 was named ORF50B. The major form expressed by wild-type ORF50 was named ORF50A.

A surprising result was encountered when the same cell extracts that had been used for the EMSA shown in Fig. 4B were used to detect ORF50 protein expression by immunoblotting (Fig. 4C). Both internal-deletion mutant ORF50 constructs expressed two protein variants of different electrophoretic mobilities and abundances: a slower, less abundant form of approximately 105 kDa (form A) and a more abundant, faster form of approximately 90 kDa (form B). In extracts of cells transfected with wild-type ORF50, form A (approximately 110 kDa) was predominant, and form B (about 90 kDa) was only detected after prolonged exposure of the autoradiograph. This result suggested that the region of ORF50 protein encompassing aa 520 to 535 controlled two phenomena: DNA binding and protein stability. Further experiments attempted to determine whether these two phenomena were or were not linked.

Mutation of a basic motif (KKRK) in the inhibitory region is sufficient to restore ORF50 DNA binding activity and support abundant expression of ORF50B. We first explored the hypothesis that the enhanced DNA binding and stability of ORF50B observed when the inhibitory region was deleted could be accounted for by removal of a region of the protein that was posttranscriptionally modified by phosphorylation or by O-linked glycosylation on serines. There are three serines, S521, S525, and S526, in the inhibitory region. We mutated each serine in the inhibitory region to alanine and tested DNA binding activity and ORF50 protein expression in cells (Fig. 5A). The double mutant ORF50(S525A/S526A) (Fig. 5) and S521A (data not shown) behaved like the wild type. Neither mutant bound DNA, and both mutants' electrophoretic mobility showed a predominance of ORF50A. This result made it unlikely that the failure of DNA binding by ORF50 was the result of posttranscriptional modification of serines in the inhibitory region.



View larger version (46K):
[in this window]
[in a new window]
  		FIG. 5. Effect of point mutations in the inhibitory region of ORF50. (A) Diagram of amino acid substitutions created in the inhibitory region of ORF50. Numbers represent amino acid residues. One single-point (S to A) and three double-point (SS to AA, KK to EE, and RK to DE) mutations are shown. NLS, nuclear localization signal; LZ, leucine zipper. (B) Mutations in the basic motif KKRK of the inhibitory region restore ORF50 DNA binding activity. Antibody to ORF50(230-250) was used to remove the ORF50/DNA complex. SS, supershifted complex. (C) Expression of mutant ORF50 proteins in HKB5/B5 cells. Transfected cell extracts were analyzed for ORF50 expression by immunoblotting with ORF50 polyclonal antibody. (D) Northern blot analysis of ORF50 mRNA in HKB5/B5 cells. The indicated plasmids were transfected into HKB5/B5 cells, and total RNA from transfected cells was isolated 48 h after transfection. A specific ORF50 probe was used to detect ORF50 mRNA. Hybridization with H1 RNA of RNase P probe served as a loading control.

Six of 14 amino acids of the inhibitory region are basic; therefore, by making basic-to-acidic substitutions we tested the hypothesis that these basic amino acids contributed to inhibition of DNA binding. The behavior of two mutants, ORF50(K527E/K528E) (KK/EE) and ORF50(R529D/K530E) (RK/DE), is illustrated in Fig. 5B and C. Mutation of these basic residues markedly enhanced ORF50 DNA binding. The specific interaction between the KK/EE and RK/DE mutants and PANp was confirmed by a supershift assay with antibody to ORF50 (Fig. 5B). Moreover, both KK/EE and RK/DE mutants expressed abundant ORF50B (Fig. 5C). These experiments suggested that the basic motif in the region between aa 521 and 534 was important for regulating ORF50 DNA binding activity and for abundant expression of ORF50B.

The possibility that abundant ORF50B expression was the result of enhanced ORF50 transcription or increased ORF50 mRNA stability was examined by Northern blot analysis (Fig. 5D). ORF50 mRNA levels expressed from the transfected plasmids were similar in mutants of the basic motif, in the serine mutant ORF50(S525A/S526A), and in the wild type. Thus, enhanced expression of ORF50B is controlled at a posttranscriptional level.

The R160A mutation is dominant over mutations in the basic regulatory motif of ORF50 protein. The DNA binding domain of ORF50 protein is located in aa 1 to 390 (Fig. 1 and 2). A mutation at position R160 in the context of ORF50(1-390), which does not contain the inhibitory domain, abolishes DNA binding. The next experiment attempted to determine whether the basic motif mutations, which enhanced the DNA binding of wild-type ORF50 protein, also restored DNA binding by R160 mutants. Introduction of an R160A mutation into full-length ORF50(KK/EE) and ORF50(RK/DE) abolished DNA binding of the KK/EE and RK/DE substituted mutants (Fig. 6A). However, the R160A mutation did not eliminate abundant expression of ORF50B of the KK/EE or RK/DE mutants (Fig. 6B). This experiment provided two tentative conclusions. First, gain-of-function mutations in the inhibitory region require normal activity of the DNA binding domain. Second, enhanced stability of ORF50B is not sufficient to enhance DNA binding.



View larger version (46K):
[in this window]
[in a new window]
  		FIG. 6. R160A, a point mutation in the DNA binding domain of ORF50, is dominant over mutations in the basic inhibitory domain. (A) EMSA performed using extracts prepared from HKB5/B5 cells transfected with the indicated plasmids. Substitutions of alanine for arginine at aa 160 in ORF50(SS/AA), ORF50(KK/EE), and ORF50(RK/DE) are represented as SS/AA (R160A), KK/EE (R160A), and RK/DE (R160A), respectively. All ORF50 mutants with a R160A substitution were unable to bind DNA in the EMSA. Antibody to ORF50(230-250) was used for supershifts (SS). (B) Protein expression of ORF50 mutants. ORF50 expression was detected by Western blot analysis of the same cell extracts used for EMSA.

Enhanced DNA binding by ORF50 basic motif mutants is independent of the expression level of ORF50B protein. The next series of experiments explored further whether the two phenomena of enhanced DNA binding and enhanced ORF50B expression observed with the KK/EE mutants were linked. The KK/EE mutation was introduced into full-length and C-terminal truncations of ORF50 protein that still maintained the inhibitory region. Whenever the KK/EE mutation was present, the proteins were enhanced in DNA binding activity (Fig. 7A and B). Furthermore, the KK/EE mutation enhanced protein expression levels in constructs ORF50(1-691), ORF50(1-680), ORF50(1-665), and ORF50(1-650). However, expression of constructs containing ORF50(1-590), with or without a VP16 tag, was not enhanced by the KK/EE mutation (Fig. 7B). Nonetheless, ORF50(1-590)KK/EE was markedly enhanced in DNA binding. We inferred from these experiments that enhancement of DNA binding activity was independent of the ORF50 protein expression level. Enhanced ORF50 protein expression required the region located between aa 590 and 650, but this region was not required for enhanced DNA binding.



View larger version (58K):
[in this window]
[in a new window]
  		FIG. 7. DNA binding activity of ORF50 restored by mutations in the DNA binding inhibitory domain is independent of ORF50 expression level. (A) (Left) A series of C-terminal-deletion mutants of ORF50 with or without the KK/EE mutation were analyzed for DNA binding capacity by EMSA. (Right) ORF50 expression in each transfected cell extract was analyzed as shown. (B) Comparison of DNA binding and protein expression by ORF50(1-590) and its counterpart containing the KK/EE mutation with or without VP16 tags. (Left) The DNA binding capacity of ORF50(1-590) mutants was analyzed by EMSA. Antibody to VP16 was used to supershift the ORF50/DNA complexes in the EMSA. (Right) ORF50 polyclonal antibody and actin monoclonal antibody were used for Western blot analysis to detect ORF50 protein and actin in cell extracts.

Mutations in the basic motif of the inhibitory region stabilize ORF50B protein tagged with a FLAG epitope at the N or C terminus. The next experiment attempted to determine whether ORF50B represented a degraded protein that resulted from cleavage from the N terminus or the C terminus of the full-length protein. Since ORF50B (90 kDa) is always approximately 20 kDa smaller than ORF50A (110 kDa), the cleavage should take place at a very specific location. Accordingly, the electrophoretic mobilities of form A and form B were examined on ORF50 proteins that were tagged with FLAG at the N and C termini (Fig. 8). Whenever the KK/EE mutation was installed into ORF50 protein without a FLAG tag or with a C-terminal or N-terminal tag, ORF50B was stabilized (Fig. 8A and data not shown). Based on immunoblotting analysis using antibody to FLAG (Fig. 8B and C), ORF50B still contains an intact C terminus and an intact N terminus because it can be recognized with anti-FLAG antibody. Therefore, ORF50B more likely represents a stabilized, unmodified form of ORF50 protein, rather than a cleavage product.



View larger version (29K):
[in this window]
[in a new window]
  		FIG. 8. ORF50B contains intact N and C termini. Western blots (WB) of ORF50 and ORF50(KK/EE) expression in HKB5/B5 cells were performed with antibodies to ORF50 (A) and to FLAG (B and C). The constructs were tagged with FLAG at the C terminus (A and B) or the N terminus (C).

The FLAG-tagged versions of ORF50 protein were also used to analyze the contributions of mutations of individual amino acids in the basic motif to enhanced DNA binding and stability of ORF50B (Fig. 9). All single basic-to-acidic mutations in the KKRK motif (Fig. 5A) enhanced DNA binding, although single mutations were not as effective as double mutants (Fig. 9A and B). Three of the four single mutations also enhanced expression of ORF50B. However, K527E enhanced DNA binding but was not sufficient to support abundant expression of ORF50B (Fig. 9A).



View larger version (58K):
[in this window]
[in a new window]
  		FIG. 9. Effect of single-point mutations in the basic motif of the DNA binding inhibitory region. Wild-type and mutated ORF50 protein containing an N-terminal FLAG tag were expressed in HKB5/B5 cells under control of the CMV promoter. Antibody to the FLAG epitope was used for supershifts (SS) in EMSAs (left panel) or for Western blot analysis (right panel). Mutants K527E/K528E (A) and R529D/K530E (B) contain a single amino acid substitution in the basic motif (KKRK) of ORF50 protein. ORF50(R160A) is a mutant deficient in DNA binding.

Determination of ORF50 DNA binding activity by DNA affinity chromatography. The next experiments attempted to examine the possibility that the lack of DNA binding activity of the full-length ORF50 protein and the restoration of binding by acidic mutations in the KKRK motif were dependent on the conditions of the EMSA. Therefore, DNA affinity chromatography was used as an alternative method to analyze the DNA binding activity of ORF50 protein derivatives. An oligomer of ORF50 response elements from the PAN promoter was used to create an oligonucleotide affinity column (Fig. 10A). To determine whether wild-type or C-terminal-truncated ORF50 proteins bound PANp in the DNA affinity column, two separate transfected cell extracts containing wild-type ORF50 protein or ORF50(1-490)+VP protein were loaded onto the same PANp oligonucleotide affinity column. Protein fractions were gradually eluted with a NaCl gradient and analyzed by immunoblotting with antibodies to ORF50 and VP16. Wild-type ORF50 protein was only detected in the flowthrough fraction. However, ORF50(1-490)+VP could be detected in the flowthrough fraction and in fractions eluted from the PANp affinity column by 0.4 and 0.5 M NaCl (Fig. 10B). This result confirmed the finding of the EMSA that removal of the C-terminal region of ORF50 protein enhanced DNA binding activity in vitro. Protein fractions eluted from the affinity column were also examined for DNA binding by EMSA. Only fractions containing ORF50(1-490)+VP protein displayed specific binding activity (Fig. 10C). Interestingly, YY1 protein could be coeluted with ORF50(1-490)+VP from the PANp oligonucleotide affinity column (Fig. 10C and D).



View larger version (66K):
[in this window]
[in a new window]
  		FIG. 10. Determination of ORF50 DNA binding activity by oligonucleotide affinity chromatography. (A) DNA sequence of the ORF50 responsive element from PANp(–91 to –58) used in affinity chromatography. The boxed nucleotides represent the ORF50 and YY1 binding sequences. (B and C) Extracts of HKB5/B5 cells transfected with pCMV-Flag-ORF50 or with pRTS/ORF50(1-490)+VP were loaded on the same PANp oligonucleotide affinity column and eluted with a NaCl gradient. The fractions were analyzed by immunoblotting with antibodies to VP16 and ORF50 (B). The same fractions were analyzed by EMSA (C). (D and E) Extracts of cells transfected with pCMV-FLAG-ORF50(KK/EE) were applied to a PANp affinity column and eluted. (D) Immunoblot with antibodies to FLAG and YY1. (E) EMSA with the eluted fractions. FT, flowthrough; WT, wild type.

Next, we examined whether the KK/EE mutation enhanced binding of ORF50 protein in DNA affinity chromatography. Extracts of cells transfected with pCMV-FLAG-ORF50(KK/EE) were applied to a PANp oligonucleotide affinity column and eluted by a NaCl gradient. The fractions were probed with antibodies to FLAG and YY1. ORF50(KK/EE) and YY1 proteins were identified in column fractions eluted with 0.4 to 0.7 M NaCl (Fig. 10D). These fractions demonstrated specific binding activity of ORF50(KK/EE) and YY1 proteins in EMSA (Fig. 10E).

ORF50(KK/EE) has stronger DNA binding activity than wild-type ORF50 in vivo. ChIP was carried out in HH-B2 cells to determine whether the enhanced DNA binding activity of ORF50(KK/EE) observed in vitro was also seen in KSHV-infected cells. Chromatin extracts of HH-B2 cells transfected with plasmids expressing wild-type FLAG-ORF50, FLAG-ORF50(KK/EE), and FLAG-ORF50(R160A) were immunoprecipitated with antibodies to FLAG, YY1, or VP16. The association between proteins and DNA was detected by PCR with primers specific for the promoter regions of PAN and K12. As observed in vitro, ORF50(KK/EE) showed stronger DNA binding in vivo on the PAN and K12 promoters than wild-type ORF50 (Fig. 11, compare lanes 2 and 3 of each panel).



View larger version (47K):
[in this window]
[in a new window]
  		FIG. 11. ORF50 (KK/EE) binds DNA in vivo more strongly than wild-type ORF50. ChIP assays were performed with extracts of KSHV-infected HH-B2 cells transfected with a vector (lanes 1, 5, 9, and 13), pCMV-Flag-ORF50 (lanes 2, 6, 10, and 14), pCMV-Flag-ORF50(KK/EE) (lanes 3, 7, 11, and 15), and pCMV-Flag-ORF50(R160A) (lanes 4, 8, 12, and 16). Twenty hours after transfection, HH-B2 cell extracts were precipitated with antibody to FLAG (lanes 1 to 4), YY1 (lanes 5 to 8), or VP16 (lanes 9 to 12). Associations between proteins and promoters were detected by PCR using primers specific for the promoter regions of PAN and K12. Lanes: 13 to 16, input material; 17, negative-control PCR without added template; 18, positive-control PCR with plasmids containing either PAN or K12 promoters as templates.

ORF50(KK/EE) is defective in activating endogenous ORF50 mRNA. To compare the biological phenotypes of wild-type ORF50 and the ORF50(KK/EE) mutant in KSHV-infected cells, we analyzed their protein expression level and their function in activating PAN RNA expression in HH-B2 cells. Expression of ORF50 and the ORF50(KK/EE) mutant was analyzed by immunoblotting with an ORF50 antibody at different times after transfection of HH-B2 cells (Fig. 12A). Only form A was expressed by wild-type ORF50, and form B was not detectable. Both forms of ORF50 were expressed in extracts of cells transfected with pRTS/ORF50(KK/EE), and form B was predominant (Fig. 12A). Optimal expression of ORF50B was seen 12 h after transfection. In HH-B2 extracts harvested at this time, ORF50(KK/EE) also demonstrated enhanced DNA binding activity in EMSA (Fig. 12B). However, when activation of PAN RNA expression by wild-type ORF50 and the ORF50(KK/EE) mutant was compared, wild-type ORF50 manifested a stronger transactivation function than the ORF50(KK/EE) mutant (Fig. 12C and 13B).



View larger version (84K):
[in this window]
[in a new window]
  		FIG. 12. Comparison of the capacity of wild-type ORF50 and the ORF50(KK/EE) mutant to activate PAN expression in HH-B2 cells. (A) Expression of ORF50 and ORF50(KK/EE) was analyzed by immunoblotting with an ORF50 polyclonal antibody at different times after transfection of HH-B2 cells with 10 µg of ORF50 or ORF50(KK/EE) expression plasmid. (B) Extracts of HH-B2 cells harvested 12 h after transfection were used in EMSA with PANp as a probe. Antibodies to ORF50(230-250) and Sp1 were used for the supershift assay. (C) Activation of PAN RNA by ORF50 and ORF50(KK/EE). Total RNA was prepared from transfected HH-B2 cells at different time points, and expression of PAN RNA was analyzed by Northern blot hybridization. Hybridization with H1 RNA of RNase P served as a loading control.



View larger version (51K):
[in this window]
[in a new window]
  		FIG. 13. ORF50(KK/EE) is defective in activating endogenous ORF50 mRNA. Total RNA from HH-B2 cells transfected with increasing amounts of pRTS/ORF50 and pRTS/ORF50(KK/EE) expression plasmids was prepared 16 h after transfection. RNA samples were analyzed with Northern blots using probes specific to ORF50 (A) and PAN RNA (B). The ORF50 mRNA encoded by the transfected plasmid (P) is about 2.4 kb, and the endogenous (E) ORF50 mRNA induced by transfected ORF50 protein is 3.6 kb.

Since ORF50 can autoactivate its own promoter, we speculated that the diminished transactivation function of ORF50(KK/EE) on PAN RNA expression might be due to the loss of autostimulation function in the KK/EE mutant. To test this possibility, we transfected different amounts of plasmids encoding wild-type ORF50 and ORF50(KK/EE) into HH-B2 cells. The expression of ORF50 RNA and PAN RNA in transfected cells was analyzed by Northern blotting (Fig. 13). Wild-type ORF50 activated the endogenous ORF50 mRNA. However, ORF50(KK/EE) did not activate the endogenous ORF50 promoter (Fig. 13A). The activation of PAN RNA expression by ORF50(KK/EE) was also reduced compared to that by wild-type ORF50 (Fig. 13B). Thus, enhanced DNA binding by the ORF50(KK/EE) mutant is not associated with enhanced activation of a direct target of ORF50 protein (i.e., PAN) and eliminates activation of an indirect target (i.e., ORF50 itself).


arrow
DISCUSSION
 
This report identifies an important regulatory region of the KSHV ORF50 protein. This region, located between aa 520 and 535, contains a basic tetrapeptide (KKRK) motif that plays a role in regulating DNA binding and stability of the ORF50 protein. Deletion of the regulatory region or mutation of the KKRK motif to acidic amino acids markedly augments DNA binding in vitro and in vivo. However, ORF50 basic motif mutants that are enhanced in DNA binding appear to be partially defective in activating downstream targets in the virus, both those, such as the ORF50 promoter itself, that are the result of an indirect mechanism of action (38, 51) and also those, such as the promoters of the PAN and K12 genes, that appear to be driven by a direct DNA binding mode of action (5, 42, 44). Deletion of the regulatory region or mutations in the KKRK motif also enhance stability of a smaller, presumably undermodified form of the ORF50 protein called ORF50B. However, the phenomenon of enhanced DNA binding observed in ORF50 proteins that have been mutated in the basic motif does not appear to be the result of enhanced stability of ORF50B. The basic motif in this regulatory region of ORF50 protein appears to influence two distinct and separate regions of ORF50 protein. It modulates the DNA binding activity encoded in the N-terminal 390 aa of ORF50. Mutations in the basic motif that stabilize ORF50B appear to require aa 590 to 650 of the protein (Fig. 14A). Thus, the regulatory domain is multifunctional.



View larger version (19K):
[in this window]
[in a new window]
  		FIG. 14. Models for the multifunctional regulatory motif of ORF50 protein. (A) The basic regulatory motif (KKRK) contained within aa 520 to 535 inhibits DNA binding through interaction with the DNA binding domain in aa 1 to 390 and destabilizes ORF50B through interaction with a domain located in aa 590 to 650. (B) The inhibition of ORF50 DNA binding activity by the KKRK motif may be mediated through two possible mechanisms: intrinsic intramolecular inhibition and extrinsic inhibition through binding by a repressor protein.

What are the possible mechanisms and biological functions of the DNA binding inhibitory domain? There are two distinct mechanistic possibilities: first, in an intrinsic model, the inhibitory domain may fold in such a way that it sterically hinders the interaction of the DNA binding domain with DNA; second, in an extrinsic model, the regulatory region might bind another cellular protein that inhibits DNA binding, again perhaps by an interaction with the DNA binding domain of ORF50 (Fig. 14B). If another cellular protein is required, it may not be possible to recapitulate the phenomenon of autoinhibition using ORF50 protein that has been purified from prokaryotic cells. If autoinhibition is intrinsic, it should be demonstrable with purified protein.

ORF50 protein acts on at least two very different types of target promoters. Many promoters are activated by an indirect mechanism. In these promoters ORF50 protein interacts with a cellular protein bound to the promoter, e.g., OCT1 and C/EBP{alpha} in the ORF50 promoter (38, 51), Sp1 in the viral thymidine kinase promoter (ORF21) (53), C/EBP{alpha} in the K8 promoter (50), and RBP-J{kappa} in the ORF57 promoter (25). ORF50 protein may or may not contact DNA on these promoters. ORF50 only directly activates a few genes, such as PAN and K12, by binding to related response elements in their promoters (5, 42, 44). This duality of mechanism requires that the DNA binding activity of ORF50 protein be tightly regulated. It appears that the default position of full-length ORF50 protein, e.g., in uninfected cells, is in a non-DNA binding conformation. Further events, perhaps occurring within the life cycle of KSHV, may modulate the activity of ORF50 to change some of it into a DNA binding form. This change will likely be brought about by new protein partners or by posttranslational modifications, such as phosphorylation. Several other mammalian transcription factors, including Ets-1, p53, androgen receptor, and IRF-3 contain autoinhibitory domains that block DNA binding (17, 20, 21, 26, 27). Other transcription factors contain autoinhibitory domains that block the activation function of the proteins (35). In all instances the autoinhibition can be reversed, either by new protein-protein interactions or by covalent modifications.

Why are ORF50 mutants that possess enhanced DNA binding defective in activating viral target genes in vivo? We initially excluded the possibility that the DNA binding phenotypes were artifacts of the conditions of the DNA binding assays. Full-length ORF50 protein fails to bind DNA in vitro in EMSAs and DNA oligonucleotide affinity chromatography, whereas the KK/EE basic motif mutants bind DNA in both assays (Fig. 5 and 10). Moreover, wild-type ORF50 protein does not detectably bind PANp or K12p in vivo in the ChIP assay, whereas binding of the KKEE mutant is readily detected (Fig. 11). It was perhaps not surprising that ORF50 proteins with mutants in the regulatory region were inactive or markedly deficient in stimulating expression of an indirect target such as ORF50. These mutants may affect critical protein-protein contacts that are required for indirect activation. However, it was surprising to us that the KK/EE mutant of ORF50 was relatively reduced at activating a direct target, such as PAN RNA (Fig. 12C and 13B). We have considered the following possible explanations for the loss of biological activity. (i) The KK/EE gain-of-function mutations may change the on/off rate of ORF50 protein on the PAN promoter; mutations that enhance or stabilize binding may be detrimental. (ii) Even a direct target such as PANp may require a combination of DNA binding and non-DNA binding mechanisms; by impairing the non-DNA binding mode of action, the KK/EE mutation would be decreased in efficacy. (iii) While enhancing DNA binding the KK/EE mutation might decrease an essential protein-protein interaction with a coactivator or a component of the basal transcription machinery. (iv) The abundant form, ORF50B, may play an inhibitory role with respect to gene activation.

The basic motif in the regulatory domain separately influences DNA binding and ORF50B protein stability. When we first encountered the abundant ORF50B protein expressed from constructs with internal deletions of the regulatory region (Fig. 4), it seemed reasonable to assume that the enhanced DNA binding phenotype and enhanced stability of ORF50B were linked. However, the following lines of experimental evidence showed that the two phenomena were separable. (i) Some constructs lacking the DNA binding inhibitory domain, i.e., ORF50(1-520)+VP, and others possessing the inhibitory domain, i.e., ORF50(1-535)+VP, are equally stable (Fig. 2C). (ii) ORF50(1-590), which does not bind DNA, and ORF50(1-590)KK/EE, which does bind DNA, are equally stable (Fig. 7B). (iii) ORF50(K527E) is enhanced in DNA binding, but form ORF50B is not stabilized in this mutant. (iv) ORF50(R160A) with the KK/EE mutation expresses an abundant form of ORF50B but does not bind to DNA (Fig. 6). (v) Most importantly, both ORF50A and ORF50B of the KK/EE mutant bind to a PANp oligonucleotide affinity column (Fig. 10D).

What is ORF50B? We considered three possibilities: (i) it is a stable degraded form of ORF50 protein, (ii) it is a stable specific cleavage product of ORF50, or (iii) it is a stable form of ORF50 protein that lacks some or all posttranslational modification. Since ORF50B expressed from a full-length KK/EE mutated ORF50 gene is always approximately 20 kDa smaller than ORF50A (Fig. 4C, 5C, 6B, 8, 9, and 10) it is obviously not a randomly degraded form of ORF50. By appending a FLAG epitope tag to either end of the full-length ORF50 protein containing the stabilizing KK/EE mutation (Fig. 8), it was possible to show that both the N-terminal and C-terminal ends of ORF50B protein are intact. Thus, ORF50B does not produce a cleavage product from either end. Although the experiments do not exclude a specific internal cleavage and ligation of ORF50 protein, this would be unusual. At present, the most likely scenario to us is that ORF50B is an unmodified or partially modified form of ORF50 that is protected from proteolytic degradation.

How does the basic motif destabilize ORF50B? In overexposed autoradiographs ORF50B is seen at low abundance when wild-type ORF50 protein is expressed (e.g., Fig. 8A and 9A and B). In mutants of the ORF50 regulatory domain, there is a marked increase in the stability of ORF50B and relatively little or no increase in the stability of ORF50A. These findings prompt the hypothesis that the basic motif in the regulatory region specifically destabilizes an unmodified or partially modified form of ORF50. The fully modified form, i.e., ORF50A, is somehow protected from the destabilizing effects of the regulatory region. The stabilizing effects of basic motif KK/EE mutations on ORF50B protein are seen in constructs containing aa 650 to 691 but not in constructs containing aa 1 to 590 (Fig. 7 and data not shown). These data are compatible with the hypothesis that a motif encompassed by aa 590 to 650 interacts with the basic motif in the regulatory domain to destabilize ORF50B. A hypothesis for further exploration is that a ubiquitin-conjugating enzyme or ligase docks on aa 590 to 650 and modifies the KKRK motif. If the KKRK motif is mutated, the presence of the ligase on its docking site stabilizes the protein. The fully modified ORF50A is not subject to this regulation.

In conclusion, we have described an intricate form of autoregulation of the ORF50 protein that influences DNA binding and protein stability. This autoregulation is likely to represent an essential strategy to allow ORF50 proteins to carry out multiple functions by distinct mechanisms in control of lytic gene expression of KSHV.


arrow
ACKNOWLEDGMENTS
 
This work was supported by grants CA70036 and CA16038 from NIH to G.M.

We thank Ayman El-Guindy and Jill Countryman for helpful discussions and Ren Sun (University of California at Los Angeles) for the FLAG-ORF50 expression plasmid.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: 333 Cedar St., Room 420, LSOG, New Haven, CT 06520. Phone: (203) 785-4758. Fax: (203) 785-6961. E-mail: George.Miller{at}Yale.edu. Back


arrow
REFERENCES
 
    1
  1. Alexander, L., L. Denekamp, A. Knapp, M. R. Auerbach, B. Damania, and R. C. Desrosiers. 2000. The primary sequence of rhesus monkey rhadinovirus isolate 26-95: sequence similarities to Kaposi's sarcoma-associated herpesvirus and rhesus monkey rhadinovirus isolate 17577. J. Virol. 74:3388-3398.[Abstract/Free Full Text]
    2. 2
  2. Boshoff, C., T. F. Schulz, M. M. Kennedy, A. K. Graham, C. Fisher, A. Thomas, J. O. McGee, R. A. Weiss, and J. J. O'Leary. 1995. Kaposi's sarcoma-associated herpesvirus infects endothelial and spindle cells. Nat. Med. 1:1274-1278.[CrossRef][Medline]
    3. 3
  3. Burysek, L., and P. M. Pitha. 2001. Latently expressed human herpesvirus 8-encoded interferon regulatory factor 2 inhibits double-stranded RNA-activated protein kinase. J. Virol. 75:2345-2352.[Abstract/Free Full Text]
    4. 4
  4. Cesarman, E., Y. Chang, P. S. Moore, J. W. Said, and D. M. Knowles. 1995. Kaposi's sarcoma-associated herpesvirus-like DNA sequences in AIDS-related body-cavity-based lymphomas. N. Engl. J. Med. 332:1186-1191.[Abstract/Free Full Text]
    5. 5
  5. Chang, P. J., D. Shedd, L. Gradoville, M. S. Cho, L. W. Chen, J. Chang, and G. Miller. 2002. Open reading frame 50 protein of Kaposi's sarcoma-associated herpesvirus directly activates the viral PAN and K12 genes by binding to related response elements. J. Virol. 76:3168-3178.[Abstract/Free Full Text]
    6. 6
  6. Chang, Y., E. Cesarman, M. S. Pessin, F. Lee, J. Culpepper, D. M. Knowles, and P. S. Moore. 1994. Identification of herpesvirus-like DNA sequences in AIDS-associated Kaposi's sarcoma. Science 266:1865-1869.[Abstract/Free Full Text]
    7. 7
  7. Chen, J., K. Ueda, S. Sakakibara, T. Okuno, and K. Yamanishi. 2000. Transcriptional regulation of the Kaposi's sarcoma-associated herpesvirus viral interferon regulatory factor gene. J. Virol. 74:8623-8634.[Abstract/Free Full Text]
    8. 8
  8. Deng, H., M. J. Song, J. T. Chu, and R. Sun. 2002. Transcriptional regulation of the interleukin-6 gene of human herpesvirus 8 (Kaposi's sarcoma-associated herpesvirus). J. Virol. 76:8252-8264.[Abstract/Free Full Text]
    9. 9
  9. Deng, H., A. Young, and R. Sun. 2000. Auto-activation of the rta gene of human herpesvirus-8/Kaposi's sarcoma-associated herpesvirus. J. Gen. Virol. 81:3043-3048.[Abstract/Free Full Text]
    10. 10
  10. DeWire, S. M., M. A. McVoy, and B. Damania. 2002. Kinetics of expression of rhesus monkey rhadinovirus (RRV) and identification and characterization of a polycistronic transcript encoding the RRV Orf50/Rta, RRV R8, and R8.1 genes. J. Virol. 76:9819-9831.[Abstract/Free Full Text]
    11. 11
  11. Dittmer, D., M. Lagunoff, R. Renne, K. Staskus, A. Haase, and D. Ganem. 1998. A cluster of latently expressed genes in Kaposi's sarcoma-associated herpesvirus. J. Virol. 72:8309-8315.[Abstract/Free Full Text]
    12. 12
  12. Goudsmit, J., N. Renwick, N. H. Dukers, R. A. Coutinho, S. Heisterkamp, M. Bakker, T. F. Schulz, M. Cornelissen, and G. J. Weverling. 2000. Human herpesvirus 8 infections in the Amsterdam Cohort Studies (1984-1997): analysis of seroconversions to ORF65 and ORF73. Proc. Natl. Acad. Sci. USA 97:4838-4843.[Abstract/Free Full Text]
    13. 13
  13. Gradoville, L., J. Gerlach, E. Grogan, D. Shedd, S. Nikiforow, C. Metroka, and G. Miller. 2000. Kaposi's sarcoma-associated herpesvirus open reading frame 50/Rta protein activates the entire viral lytic cycle in the HH-B2 primary effusion lymphoma cell line. J. Virol. 74:6207-6212.[Abstract/Free Full Text]
    14. 14
  14. Gwack, Y., H. J. Baek, H. Nakamura, S. H. Lee, M. Meisterernst, R. G. Roeder, and J. U. Jung. 2003. Principal role of TRAP/Mediator and SWI/SNF complexes in Kaposi's sarcoma-associated herpesvirus RTA-mediated lytic reactivation. Mol. Cell. Biol. 23:2055-2067.[Abstract/Free Full Text]
    15. 15
  15. Gwack, Y., H. Byun, S. Hwang, C. Lim, and J. Choe. 2001. CREB-binding protein and histone deacetylase regulate the transcriptional activity of Kaposi's sarcoma-associated herpesvirus open reading frame 50. J. Virol. 75:1909-1917.[Abstract/Free Full Text]
    16. 16
  16. Gwack, Y., H. Nakamura, S. H. Lee, J. Souvlis, J. T. Yustein, S. Gygi, H. J. Kung, and J. U. Jung. 2003. Poly(ADP-ribose) polymerase 1 and Ste20-like kinase hKFC act as transcriptional repressors for gamma-2 herpesvirus lytic replication. Mol. Cell. Biol. 23:8282-8294.[Abstract/Free Full Text]
    17. 17
  17. Hagman, J., and R. Grosschedl. 1992. An inhibitory carboxyl-terminal domain in Ets-1 and Ets-2 mediates differential binding of ETS family factors to promoter sequences of the mb-1 gene. Proc. Natl. Acad. Sci. USA 89:8889-8893.[Abstract/Free Full Text]
    18. 18
  18. Haque, M., J. Chen, K. Ueda, Y. Mori, K. Nakano, Y. Hirata, S. Kanamori, Y. Uchiyama, R. Inagi, T. Okuno, and K. Yamanishi. 2000. Identification and analysis of the K5 gene of Kaposi's sarcoma-associated herpesvirus. J. Virol. 74:2867-2875.[Abstract/Free Full Text]
    19. 19
  19. Hardwick, J. M., P. M. Lieberman, and S. D. Hayward. 1988. A new Epstein-Barr virus transactivator, R, induces expression of a cytoplasmic early antigen. J. Virol. 62:2274-2284.[Abstract/Free Full Text]
    20. 20
  20. Hupp, T. R., A. Sparks, and D. P. Lane. 1995. Small peptides activate the latent sequence-specific DNA binding function of p53. Cell 83:237-245.[CrossRef][Medline]
    21. 21
  21. Jayaraman, J., and C. Prives. 1995. Activation of p53 sequence-specific DNA binding by short single strands of DNA requires the p53 C-terminus. Cell 81:1021-1029.[CrossRef][Medline]
    22. 22
  22. Jeong, J., J. Papin, and D. Dittmer. 2001. Differential regulation of the overlapping Kaposi's sarcoma-associated herpesvirus vGCR (orf74) and LANA (orf73) promoters. J. Virol. 75:1798-1807.[Abstract/Free Full Text]
    23. 23
  23. Kadonaga, J. T., and R. Tjian. 1986. Affinity purification of sequence-specific DNA binding proteins. Proc. Natl. Acad. Sci. USA 83:5889-5893.[Abstract/Free Full Text]
    24. 24
  24. Kedes, D. H., M. Lagunoff, R. Renne, and D. Ganem. 1997. Identification of the gene encoding the major latency-associated nuclear antigen of the Kaposi's sarcoma-associated herpesvirus. J. Clin. Investig. 100:2606-2610.[Medline]
    25. 25
  25. Liang, Y., J. Chang, S. J. Lynch, D. M. Lukac, and D. Ganem. 2002. The lytic switch protein of KSHV activates gene expression via functional interaction with RBP-Jkappa (CSL), the target of the Notch signaling pathway. Genes Dev. 16:1977-1989.[Abstract/Free Full Text]
    26. 26
  26. Lin, R., Y. Mamane, and J. Hiscott. 1999. Structural and functional analysis of interferon regulatory factor 3: localization of the transactivation and autoinhibitory domains. Mol. Cell. Biol. 19:2465-2474.[Abstract/Free Full Text]
    27. 27
  27. Liu, G. Z., H. Wang, and Z. Wang. 2003. Identification of a highly conserved domain in the androgen receptor that suppresses the DNA-binding domain-DNA interactions. J. Biol. Chem. 278:14956-14960.[Abstract/Free Full Text]
    28. 28
  28. Lukac, D. M., L. Garibyan, J. R. Kirshner, D. Palmeri, and D. Ganem. 2001. DNA binding by Kaposi's sarcoma-associated herpesvirus lytic switch protein is necessary for transcriptional activation of two viral delayed early promoters. J. Virol. 75:6786-6799.[Abstract/Free Full Text]
    29. 29
  29. Lukac, D. M., J. R. Kirshner, and D. Ganem. 1999. Transcriptional activation by the product of open reading frame 50 of Kaposi's sarcoma-associated herpesvirus is required for lytic viral reactivation in B cells. J. Virol. 73:9348-9361.[Abstract/Free Full Text]
    30. 30
  30. Lukac, D. M., R. Renne, J. R. Kirshner, and D. Ganem. 1998. Reactivation of Kaposi's sarcoma-associated herpesvirus infection from latency by expression of the ORF 50 transactivator, a homolog of the EBV R protein. Virology 252:304-312.[CrossRef][Medline]
    31. 31
  31. Martin, D. F., B. D. Kuppermann, R. A. Wolitz, A. G. Palestine, H. Li, C. A. Robinson, et al. 1999. Oral ganciclovir for patients with cytomegalovirus retinitis treated with a ganciclovir implant. N. Engl. J. Med. 340:1063-1070.[Abstract/Free Full Text]
    32. 32
  32. Menezes, J., W. Leibold, G. Klein, and G. Clements. 1975. Establishment and characterization of an Epstein-Barr virus (EBV)-negative lymphoblastoid B cell line (BJA-B) from an exceptional, EBV-genome-negative African Burkitt's lymphoma. Biomedicine 22:276-284.[Medline]
    33. 33
  33. Mosser, D. D., N. G. Theodorakis, and R. I. Morimoto. 1988. Coordinate changes in heat shock element-binding activity and HSP70 gene transcription rates in human cells. Mol. Cell. Biol. 8:4736-4744.[Abstract/Free Full Text]
    34. 34
  34. Muralidhar, S., A. M. Pumfery, M. Hassani, M. R. Sadaie, M. Kishishita, J. N. Brady, J. Doniger, P. Medveczky, and L. J. Rosenthal. 1998. Identification of kaposin (open reading frame K12) as a human herpesvirus 8 (Kaposi's sarcoma-associated herpesvirus) transforming gene. J. Virol. 72:4980-4988.[Abstract/Free Full Text]
    35. 35
  35. Pufall, M. A., and B. J. Graves. 2002. Autoinhibitory domains: modular effectors of cellular regulation. Annu. Rev. Cell Dev. Biol. 18:421-462.[CrossRef][Medline]
    36. 36
  36. Rainbow, L., G. M. Platt, G. R. Simpson, R. Sarid, S. J. Gao, H. Stoiber, C. S. Herrington, P. S. Moore, and T. F. Schulz. 1997. The 222- to 234-kilodalton latent nuclear protein (LNA) of Kaposi's sarcoma-associated herpesvirus (human herpesvirus 8) is encoded by orf73 and is a component of the latency-associated nuclear antigen. J. Virol. 71:5915-5921.[Abstract]
    37. 37
  37. Russo, J. J., R. A. Bohenzky, M. C. Chien, J. Chen, M. Yan, D. Maddalena, J. P. Parry, D. Peruzzi, I. S. Edelman, Y. Chang, and P. S. Moore. 1996. Nucleotide sequence of the Kaposi sarcoma-associated herpesvirus (HHV8). Proc. Natl. Acad. Sci. USA 93:14862-14867.[Abstract/Free Full Text]
    38. 38
  38. Sakakibara, S., K. Ueda, J. Chen, T. Okuno, and K. Yamanishi. 2001. Octamer-binding sequence is a key element for the autoregulation of Kaposi's sarcoma-associated herpesvirus ORF50/Lyta gene expression. J. Virol. 75:6894-6900.[Abstract/Free Full Text]
    39. 39
  39. Seaman, W. T., and E. B. Quinlivan. 2003. Lytic switch protein (ORF50) response element in the Kaposi's sarcoma-associated herpesvirus K8 promoter is located within but does not require a palindromic structure. Virology 310:72-84.[CrossRef][Medline]
    40. 40
  40. Searles, R. P., E. P. Bergquam, M. K. Axthelm, and S. W. Wong. 1999. Sequence and genomic analysis of a Rhesus macaque rhadinovirus with similarity to Kaposi's sarcoma-associated herpesvirus/human herpesvirus 8. J. Virol. 73:3040-3053.[Abstract/Free Full Text]
    41. 41
  41. Serio, T. R., J. L. Kolman, and G. Miller. 1997. Late gene expression from the Epstein-Barr virus BcLF1 and BFRF3 promoters does not require DNA replication in cis. J. Virol. 71:8726-8734.[Abstract]
    42. 42
  42. Song, M. J., H. J. Brown, T.-T. Wu, and R. Sun. 2001. Transcription activation of polyadenylated nuclear RNA by Rta in human herpesvirus 8/Kaposi's sarcoma-associated herpesvirus. J. Virol. 75:3129-3140.[Abstract/Free Full Text]
    43. 43
  43. Song, M. J., H. Deng, and R. Sun. 2003. Comparative study of regulation of RTA-responsive genes in Kaposi's sarcoma-associated herpesvirus/human herpesvirus 8. J. Virol. 77:9451-9462.[Abstract/Free Full Text]
    44. 44
  44. Song, M. J., X. Li, H. J. Brown, and R. Sun. 2002. Characterization of interactions between RTA and the promoter of polyadenylated nuclear RNA in Kaposi's sarcoma-associated herpesvirus/human herpesvirus 8. J. Virol. 76:5000-5013.[Abstract/Free Full Text]
    45. 45
  45. Soulier, J., L. Grollet, E. Oksenhendler, P. Cacoub, D. Cazals-Hatem, P. Babinet, M. F. d'Agay, J. P. Clauvel, M. Raphael, L. Degos, et al. 1995. Kaposi's sarcoma-associated herpesvirus-like DNA sequences in multicentric Castleman's disease. Blood 86:1276-1280.[Abstract/Free Full Text]
    46. 46
  46. Sun, R., S. F. Lin, L. Gradoville, Y. Yuan, F. Zhu, and G. Miller. 1998. A viral gene that activates lytic cycle expression of Kaposi's sarcoma-associated herpesvirus. Proc. Natl. Acad. Sci. USA 95:10866-10871.[Abstract/Free Full Text]
    47. 47
  47. Ueda, K., K. Ishikawa, K. Nishimura, S. Sakakibara, E. Do, and K. Yamanishi. 2002. Kaposi's sarcoma-associated herpesvirus (human herpesvirus 8) replication and transcription factor activates the K9 (vIRF) gene through two distinct cis elements by a non-DNA-binding mechanism. J. Virol. 76:12044-12054.[Abstract/Free Full Text]
    48. 48
  48. Virgin, H. W., IV, P. Latreille, P. Wamsley, K. Hallsworth, K. E. Weck, A. J. Dal Canto, and S. H. Speck. 1997. Complete sequence and genomic analysis of murine gammaherpesvirus 68. J. Virol. 71:5894-5904.[Abstract]
    49. 49
  49. Wang, S., S. Liu, M. H. Wu, Y. Geng, and C. Wood. 2001. Identification of a cellular protein that interacts and synergizes with the RTA (ORF50) protein of Kaposi's sarcoma-associated herpesvirus in transcriptional activation. J. Virol. 75:11961-11973.[Abstract/Free Full Text]
    50. 50
  50. Wang, S. E., F. Y. Wu, M. Fujimuro, J. Zong, S. D. Hayward, and G. S. Hayward. 2003. Role of CCAAT/enhancer-binding protein alpha (C/EBP{alpha}) in activation of the Kaposi's sarcoma-associated herpesvirus (KSHV) lytic-cycle replication-associated protein (RAP) promoter in cooperation with the KSHV replication and transcription activator (RTA) and RAP. J. Virol. 77:600-623.
    51. 51
  51. Wang, S. E., F. Y. Wu, Y. Yu, and G. S. Hayward. 2003. CCAAT/enhancer-binding protein-{alpha} is induced during the early stages of Kaposi's sarcoma-associated herpesvirus (KSHV) lytic cycle reactivation and together with the KSHV replication and transcription activator (RTA) cooperatively stimulates the viral RTA, MTA, and PAN promoters. J. Virol. 77:9590-9612.[Abstract/Free Full Text]
    52. 52
  52. Whitehouse, A., A. J. Stevenson, M. Cooper, and D. M. Meredith. 1997. Identification of a cis-acting element within the herpesvirus saimiri ORF 6 promoter that is responsive to the HVS.R transactivator. J. Gen. Virol. 78:1411-1415.[Abstract]
    53. 53
  53. Zhang, L., J. Chiu, and J. C. Lin. 1998. Activation of human herpesvirus 8 (HHV-8) thymidine kinase (TK) TATAA-less promoter by HHV-8 ORF50 gene product is SP1 dependent. DNA Cell Biol. 17:735-742.[Medline]

Journal of Virology, October 2004, p. 10657-10673, Vol. 78, No. 19
0022-538X/04/$08.00+0     DOI: 10.1128/JVI.78.19.10657-10673.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:

  • Lefort, S., Flamand, L. (2009). Kaposi's Sarcoma-Associated Herpesvirus K-bZIP Protein Is Necessary for Lytic Viral Gene Expression, DNA Replication, and Virion Production in Primary Effusion Lymphoma Cell Lines. J. Virol. 83: 5869-5880 [Abstract] [Full Text]  
  • Bu, W., Palmeri, D., Krishnan, R., Marin, R., Aris, V. M., Soteropoulos, P., Lukac, D. M. (2008). Identification of Direct Transcriptional Targets of the Kaposi's Sarcoma-Associated Herpesvirus Rta Lytic Switch Protein by Conditional Nuclear Localization. J. Virol. 82: 10709-10723 [Abstract] [Full Text]  
  • Chang, P.-J., Shedd, D., Miller, G. (2008). A Mobile Functional Region of Kaposi's Sarcoma-Associated Herpesvirus ORF50 Protein Independently Regulates DNA Binding and Protein Abundance. J. Virol. 82: 9700-9716 [Abstract] [Full Text]  
  • Bu, W., Carroll, K. D., Palmeri, D., Lukac, D. M. (2007). Kaposi's Sarcoma-Associated Herpesvirus/Human Herpesvirus 8 ORF50/Rta Lytic Switch Protein Functions as a Tetramer. J. Virol. 81: 5788-5806 [Abstract] [Full Text]  
  • Chen, L.-W., Chang, P.-J., Delecluse, H.-J., Miller, G. (2005). Marked Variation in Response of Consensus Binding Elements for the Rta Protein of Epstein-Barr Virus. J. Virol. 79: 9635-9650 [Abstract] [Full Text]  
  • Chang, P.-J., Shedd, D., Miller, G. (2005). Two Subclasses of Kaposi's Sarcoma-Associated Herpesvirus Lytic Cycle Promoters Distinguished by Open Reading Frame 50 Mutant Proteins That Are Deficient in Binding to DNA. J. Virol. 79: 8750-8763 [Abstract] [Full Text]  

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Chang, P.-J.
Right arrow Articles by Miller, G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chang, P.-J.
Right arrow Articles by Miller, G.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»