当前位置: 首页 > 医学版 > 期刊论文 > 基础医学 > 病菌学杂志 > 2005年 > 第11期 > 正文
编号:11202630
vLIP, a Viral Lipase Homologue, Is a Virulence Fac
     Department of Microbiology and Immunology, College of Veterinary Medicine, Cornell University, Ithaca, New York 14853

    Department of Biological Chemistry, School of Medicine, University of California at Davis, UC Davis Cancer Center, Sacramento, California 95817

    Viral Oncogenesis Group, Institute for Animal Health, Compton, Berkshire, United Kingdom

    ABSTRACT

    The genome of Marek's disease virus (MDV) has been predicted to encode a secreted glycoprotein, vLIP, which bears significant homology to the /? hydrolase fold of pancreatic lipases. Here it is demonstrated that MDV vLIP mRNA is produced via splicing and that vLIP is a late gene, due to its sensitivity to inhibition of DNA replication. While vLIP was found to conserve several residues essential to hydrolase activity, an unfavorable asparagine substitution is present at the lipase catalytic triad acid position. Consistent with structural predictions, purified recombinant vLIP did not show detectable activity on traditional phospholipid or triacylglyceride substrates. Two different vLIP mutant viruses, one bearing a 173-amino-acid deletion in the lipase homologous domain, the other having an alanine point mutant at the serine nucleophile position, caused a significantly lower incidence of Marek's disease in chickens and resulted in enhanced survival relative to two independently produced vLIP revertants or parental virus. These data provide the first evidence that vLIP enhances the replication and pathogenic potential of MDV. Furthermore, while vLIP may not serve as a traditional lipase enzyme, the data indicate that the serine nucleophile position is nonetheless essential in vivo for the viral functions of vLIP. Therefore, it is suggested that this particular example of lipase homology may represent the repurposing of an /? hydrolase fold toward a nonenzymatic role, possibly in lipid bonding.

    INTRODUCTION

    Marek's disease virus (MDV) is a widespread alphaherpesvirus of poultry that is capable of causing a fatal lymphoproliferative disease within 2 to 3 weeks of infection. The virus is highly cell associated and undergoes a major burst of replication in B lymphocytes during primary infection (17, 19). Latency occurs in CD4+ T cells, and characteristic disease symptoms occur when a subset of latently infected T cells become transformed and invade nervous tissues and various organs, often resulting in paralysis and death in affected birds (17, 19). The virus is shed from the feather follicle, and inhalation of contaminated feather dander is the likely route of infection (18). Economic losses have been largely controlled by vaccination (19). However, vaccination per se does not prevent superinfection, and emergent MDV strains have been known to cause Marek's disease (MD) in vaccinated chickens (83, 84). The genome of MDV-1, as well as those of both of the other members of the Marek's disease-like genus of herpesviruses, has been sequenced (3, 41, 46, 49, 78), and the contributions of some of the specific genes involved in the pathogenesis of MDV are beginning to be addressed (26, 53, 63, 65).

    Amino acid homology to lipases was originally detected in an open reading frame (ORF) of the MDV genome during the examination of sequence data from the unique long (UL) region of the GA strain (49). The ORF in question, originally designated LORF2, was hence termed "viral lipase," or vLIP (49). The vLIP reading frame is conserved among all three members of the Marek's disease-like virus genus (3, 41, 46, 49, 78), and homologues are also found in certain avian adenoviruses (20, 23, 37, 59, 80). The complete vLIP ORF is 756 amino acids in length, and significant homology to pancreatic lipases is found in a stretch of approximately 141 amino acids that span positions 229 to 369. This 141-amino-acid region of vLIP shows 41% similarity (23% identity) to a wasp phospholipase A1 (Polistes annularis) and 42% similarity (26% identity) to a rodent pancreatic lipase (Myocastor coypus). Importantly, this region corresponds to the core of the pancreatic lipase /? hydrolase fold, containing the Ser-Asp-His catalytic triad, which confers enzyme activity in this class of enzymes (70). Moreover, the serine nucleophile position of the vLIP catalytic triad is found in the context of a characteristic GXSXG motif (70). In solved structures of pancreatic lipases, this catalytic serine is located in a sharp turn in the primary amino acid chain between a ?-strand and an -helix (12, 71, 81, 82).

    Lipases are water-soluble enzymes that catalyze the hydrolysis of ester bonds in water-insoluble, "lipid" substrates (27, 30, 75) and perform critical roles in a wide array of biological processes ranging from routine metabolism of dietary triacylglycerides to mediation of cell signaling and immune activity (21, 29, 40, 55, 73, 77). The pancreatic lipase (PL) gene family represents a subset of the lipolytic enzymes that build their catalytic mechanism upon a variation of the /? hydrolase fold, a protein fold consisting of interleaving -helices and ?-strands that has been found to mediate a myriad of diverse enzyme activities in nature (35, 61, 70). Members of the PL gene family include human pancreatic lipase (HPL), pancreatic lipase-related proteins 1 and 2 (PLRP1 and PLRP2), hepatic lipase, endothelial lipase, and several wasp venom phospholipases (21). With the notable exception of the wasp venom phospholipases, which act exclusively on phospholipid substrates, PL gene family members hydrolyze the A1 and A3 positions of triacylglyceride substrates and show varying degrees of activity on the A1 position of phospholipid substrates (21). There are other /? hydrolase fold lipases that are not members of the PL gene family; these include the gastric lipases and several fungal lipases (35, 56, 70). Moreover, there are several important cellular lipase activities, including phospholipases A2, C, and D, which are not mediated by /? hydrolases.

    Although no herpesviruses outside the Marek's disease-like genus have been reported to encode proteins with homology to lipases, it is worth noting that there has been some precedent for lipases encoded by DNA viruses. A phospholipase A2 activity of a parvovirus capsid protein is reportedly essential for viral infectivity (33). The p37 glycoprotein of vaccinia virus bears homology to phospholipase D enzymes, and bona fide lipase activities have been demonstrated in vitro (8). The genomes of at least two entomopoxviruses encode proteins homologous to fungal /? hydrolase fold lipases (2, 9). Furthermore, some DNA viruses have been demonstrated to manipulate host lipid metabolism. For example, orthopoxviruses affect both arachidonic acid and polyphosphoinositide metabolism in infected cells (60, 62). More recently, human cytomegalovirus (HCMV) has been reported to carry a host cell-derived phospholipase A2 activity which is required for infectivity (5).

    An MDV mutant isolated in cell culture was found to harbor a retrovirus long terminal repeat disrupting what is now known to be the vLIP ORF (39). This long terminal repeat insertion provided the first hint that vLIP is not essential for replication in vitro. Here we provide the first data to address the role of vLIP in vivo and also provide basic characterization of the vLIP transcript and gene product. We demonstrate that vLIP is expressed as a secreted glycoprotein from a spliced RNA, lacks detectable lipase activity in vitro, and is completely dispensable for virus replication in vitro but plays an important role in the pathogenesis of MD in the chicken.

    MATERIALS AND METHODS

    Sequence analysis and secondary-structure prediction. BLAST and PSI-BLAST analyses of vLIP protein sequences were performed using National Center for Biotechnology Information servers (6, 7). Prediction of vLIP protein structure was performed using the GenTHREADER program via the PSIPRED server (45). For the prediction of signal peptides and signal peptide cleavage sites, the SignalP prediction server (version 1.1) was used (57, 58). Routine amino acid alignments were performed using MacVector 7.2.2 software (Accelrys, San Diego, CA) using default settings.

    Virus and cells. The MDCC MSB-1 tumor cell line harboring latent MDV (4) was maintained in RPMI 1640 medium supplemented with 10% fetal bovine serum (FBS) at 37°C. Lytic virus replication was induced by the addition of sodium butryate to a final concentration of 3 mM. In some experiments, 0.5 mM foscarnet (trisodium phosphonoformate hexahydrate [PFA]; Sigma-Aldrich, St. Louis, MO) was used at the time of induction to determine kinetic classes of transcripts. For the animal study and for growth of lytic MDV in culture, bacterial artificial chromosome (BAC) clones derived from the RB-1B strain of MDV (pRB-1B) were reconstituted by standard calcium phosphate transfection (72) and were passaged no more than three times for in vivo studies and no more than five times for in vitro studies. Transfection and propagation of pRB-1B-based parental, mutant, and revertant MDVs were performed on secondary chicken embryo cells (CEC), which were cultivated in a 1:1 mixture of EMEM and HMEM supplemented with 2 to 10% FBS (72).

    Trichoplusia ni insect cells, used for the expression of recombinant vLIP, were grown in 250-ml to 1-liter polycarbonate Erlenmeyer flasks (Nalgene) at 140 rpm in shaker culture at 28°C in ExCell 405 medium (JRH Biosciences, Lenexa, KS). Spodoptera frugiperda (Sf9) insect cells were grown in a similar manner to the T. ni cells, except that ExCell 420 medium (JRH Biosciences) was used.

    Cloning and characterization of vLIP mRNA. Total RNA was isolated from 1 x 108 MSB-1 cells using Trizol (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. For cDNA synthesis, 20 μg of total RNA was used for first-strand synthesis primed with the SS-dT oligonucleotide (Table 1). Reverse transcription (RT) was carried out using avian myeloblastosis virus reverse transcriptase (Roche, Indianapolis, IN) at 42°C for 1 h using the conditions specified by the manufacturer. Upon completion of RT, reactions were extracted against phenol-chloroform and passed over Chromaspin TE 200 columns (Stratagene Inc., La Jolla, CA).

    Primers (200 nM each) used in rapid amplification of 3' cDNA ends were SS and sp.LIP5'Cpo (Table 1). Thirty PCR cycles (93°C for 45 s, 56°C for 60 s, and 68°C for 300 s) were performed, using Expand High-Fidelity Taq polymerase (Roche) at 2.3 U (0.7 μl) per 50-μl reaction volume. PCR products were cloned into the pCR2.1-TopoTA vector (Invitrogen). The LIP.RC(AS) and sp.LIP5'CPO primers were used to screen bacterial colonies for plasmids with spliced vLIP cDNA inserts (Table 1). DNA sequencing was performed using vector-specific primers, along with several vLIP-specific primers designed from MDV genomic DNA sequences (Davis Sequencing LLC, Davis, CA). All oligonucleotides were custom synthesized (Integrated DNA Technologies, Coralville, IA).

    For Northern blot analysis, 10 μg of total RNA was denatured for 15 min at 65°C, loaded onto 1% agarose-6% formaldehyde gels, and run at 75 V in 50 mM morpholinepropanesulfonic acid (MOPS) buffer (pH 7.2). Resolved RNAs were transferred overnight onto Hybond-N membranes (Amersham Biosciences Corp., Piscataway, NJ) by the capillary method in 20x SSC (1x SSC is 0.15 M NaCl plus 0.015 M sodium citrate, pH 7.0) (68). Transferred RNA was UV cross-linked using the Stratalinker apparatus (Stratagene). A 32P-labeled 376-bp single-stranded RNA probe antisense to vLIP mRNA was transcribed in vitro from a pSP72 (Promega, Madison, WI)-based plasmid using the MaxiScript T7 kit (Ambion) after linearization of the construct. As a negative control, a sense single-stranded RNA probe corresponding to residues coding for amino acids 158 to 255 of vLIP was prepared and hybridized in parallel to a duplicate membrane. Unincorporated radioactivity was removed using NICK columns (Amersham Biosciences). Membranes were hybridized overnight at 68°C in UltraHyb solution (Ambion), and 1.5 x 106 cpm of probe per ml of hybridization solution was used in each experiment. Blots were washed twice for 5 min each at room temperature in 2x SSC-0.1% sodium dodecyl sulfate (SDS) and twice for 15 min each at 68°C in 0.1x SSC-0.1% SDS; then they were exposed at –80°C to BioMax MR film (Kodak, Rochester, NY) in the presence of an intensifying screen. The Millenium RNA ladder (Ambion), in conjunction with a 32P-labeled Millenium probe (Ambion), was used to determine RNA sizes.

    Construction of plasmids. The QuikChange kit (Stratagene) was used to introduce an S307A change at the predicted nucleophile site of vLIP. According to the manufacturer's instructions, the LipMutS and LipMutAS primers (Table 1) were used to introduce a single-base-pair change from thymidine to guanine, changing a TCC serine codon to a GCC alanine codon. For baculovirus expression, wild-type (wt) vLIP cDNA or S307A vLIP cDNA, starting at residue Val31 (a predicted signal peptide comprises residues 1 to 30) was PCR adapted with BamHI and XhoI sites at the 5' and 3' ends, using primers BamHILipS and XhoILipAS (Table 1), and then inserted into pBAC11 (Novagen, Madison, WI). This cloning strategy placed the N terminus of vLIP cDNA in frame with the gp64 signal peptide from pBAC11 and placed the C terminus of vLIP in frame with a polyhistidine tag also encoded by the vector. All clones were verified by DNA sequencing.

    To incorporate a FLAG epitope tag (DYKDDDDK) at the carboxy terminus of vLIP, overlap extension PCR was used. Separate PCRs were performed with the following pairs of primers: LIPFLAGS-LIPFLAGS2 and LIPFLAGR-LIPF2 (Table 1). The resulting PCR products were purified by agarose gel electrophoresis and combined in an overlap extension PCR performed using the external primers (LIPFLAGS2 and LIPF2). The PCR product was then cloned into pCR2.1-TopoTA (Invitrogen) and sequenced (Cornell BioResource Center). A sequence-verified clone was digested with BsiWI and ClaI to release a FLAG-tagged fragment of the vLIP cDNA. The BsiWI-ClaI fragment was then ligated into pEco4.268, yielding pEco4.268_vLIP (see Fig. 5). The pEco4.268 plasmid was originally constructed by inserting the EcoRI 4,268-bp subfragment of the Md5 genome derived from the sn5 cosmid (65) (a gift of Sanjay Reddy) into pUC18 (Fig. 5).

    In order to construct a shuttle vector to incorporate an in-frame deletion of amino acids 256 to 428 of the vLIP open reading frame, a 519-bp fragment was released from the pEco4.268 plasmid using MluI and SpeI, and overhangs were filled in using T4 DNA polymerase (NEB). Finally, the plasmid was closed using T4 DNA ligase (NEB) and transformed into Escherichia coli. Individual plasmids bearing the deletion were sequenced, and a clone bearing an in-frame deletion was obtained and digested with the BamHI and PshAI enzymes (NEB). The 3,307-bp BamHI-PshAI fragment was inserted into pST76K-SR (1), a vector for performing RecA-based shuttle mutagenesis, which had been opened with BamHI and SmaI. To prepare a similar shuttle vector that would be used to restore the wt vLIP gene to the deletion mutant BAC, pEco4.268 was digested with BamHI and PshAI, and the 3,826-bp fragment was ligated into pST76K-SR as described above. In order to prepare a shuttle vector to transfer the S307A mutation into the pRB-1B BAC, a pCR2.1-vLIP cDNA clone bearing the S307A mutation was digested with MluI and SpeI, and the 516-bp MluI-SpeI fragment was inserted into the pEco4.268 plasmid. After confirmation of the presence of the S307A mutation by DNA sequencing, the 3,826-bp BamHI-PshAI fragment was transferred into pST76K-SR. Finally, to prepare a shuttle vector for incorporation of a FLAG-tagged wt vLIP into the pRB-1B BAC, the pEco4.268_vLIP plasmid was opened with BamHI and PshAI, and the 3,850-bp fragment was transferred to pST76K-SR (Fig. 5).

    Shuttle mutagenesis and BAC DNA preparation. The pST76K-SR plasmid (a gift of Martin Messerle), which has been previously described (1, 11), contains a temperature-sensitive origin of replication, as well as genes encoding the bacterial recombinase RecA, the negative selection marker SacB, and kanamycin resistance. The pST76K-SR shuttle constructs described above were electroporated into DH10B E. coli harboring the pRB-1B BAC, plated on Luria Bertani (LB) agar plates containing kanamycin (50 μg/ml) and chloramphenicol (30 μg/ml), and incubated at 42°C. For the construction of a vLIP deletion mutant, minipreparations of BAC DNA from cointegrates were prepared from overnight cultures grown at 42°C, and DNA was analyzed by restriction enzyme analysis to verify cointegration of the shuttle construct. Verified cointegrates were then grown in LB broth at 37°C for 6 h in the presence of chloramphenicol but without kanamycin, so as to allow for the loss of the shuttle vector through a second RecA-mediated homologous recombination. Cultures were then streaked out on LB chloramphenicol plates containing 10% (vol/vol) sucrose to allow for negative selection by SacB. Colonies on these plates were then replica plated on kanamycin and chloramphenicol plates. Colonies that grew only on the chloramphenicol plates were screened by PCR for the presence of the deletion and then were verified by restriction enzyme analysis. To repair the vLIP deletion mutant, pST76K-SR shuttle vectors bearing the wt vLIP or the FLAG-tagged version were electroporated into DH10B E. coli harboring the pRB-1B vLIPMluI-SpelI BAC. Shuttle mutagenesis was performed as described above, with the exception that the PCR screening procedures were adjusted accordingly. Large-scale purification of BAC DNA was performed using a QIAGEN Maxi-prep kit, as per the manufacturer's instructions for purification of very low copy number plasmids and cosmids.

    Baculovirus expression and protein purification. Recombinant plasmid pBAC11-vLIP WT or pBAC11-vLIP S307A was cotransfected with BacVector3000 DNA into Sf9 insect cells using reagents from the BacVector kit (Novagen). Recombinant viruses were plaque purified twice and verified for expression by Western blotting with antibodies against vLIP.

    For the purification of polyhistidine-tagged vLIP from supernatants of baculovirus-infected T. ni insect cells (multiplicity of infection, 1.0), TALON resin (Clontech, La Jolla, CA) was used. Briefly, supernatants were adjusted to a final concentration of 50 mM sodium phosphate (pH 7.0), 300 mM NaCl, 5 mM imizadole, and 10% glycerol and were mixed gently on a magnetic stirrer with preequilibrated TALON resin (Clontech, La Jolla, CA) for 20 min. The TALON beads were then spun down at 700 x g, washed twice with 20 volumes of TALON phosphate buffer, loaded onto a 10-ml TALON column, and washed with 5 volumes of TALON phosphate buffer. Recombinant proteins were eluted with TALON elution buffer containing 150 mM imizadole and dialyzed against HEPES-buffered saline overnight at 4°C to remove imizadole.

    Immunoprecipitation and Western blot analysis. Protein concentrations were measured using the bicinchoninic acid reagent (Pierce Biotechnology Inc., Rockford, IL). Routinely, vLIP proteins were analyzed by SDS-8% polyacrylamide gel electrophoresis (PAGE), run in a Miniprotean III device (Bio-Rad Laboratories Inc., Hercules CA). For Western blotting, resolved proteins were transferred to Immobilon-P polyvinylidene difluoride (PVDF) membranes (Millipore, Billerica, MA) using a Bio-Rad semidry transfer apparatus. PVDF membranes were blocked using 5% powdered milk in 1x Tris-buffered saline-Tween 20 (5% PM-TBST). Affinity-purified polyclonal rabbit antibodies raised against the peptide TKTTNENGHEKDSKD (amino acids 108 to 122 of vLIP) were used at a 1:1,000 dilution in 5% PM-TBST. The anti-FLAG antibody M2 (Stratagene) was purchased to detect epitope-tagged vLIP expressed from chicken cells. Horseradish peroxidase-conjugated donkey anti-rabbit immunoglobulin G (Amersham Biosciences) was employed as a secondary antibody at a 1:2,000 dilution in 5% PM-TBST, and horseradish peroxidase-conjugated goat anti-mouse immunoglobulin G (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) was used at 1:5,000 in 5% PM-TBST. ECL (Amersham Biosciences) chemiluminescent detection reagents were used according to the manufacturer's instructions to image bound secondary antibodies, and blots were exposed to BioMax MR film (Kodak). Digitized images were processed with Adobe Photoshop 7.0 software for Macintosh.

    Immunoprecipitations were carried out overnight at 4°C in the presence of Complete Mini protease inhibitor (Roche) using 15 ml of cell culture supernatants of either pRB-1B- or pRB-1B-vLIP-infected CEC. Debris was removed by centrifugation at 3,000 x g for 10 min, and 40 μl of EZview ANTI-FLAG M2 affinity gel (Sigma-Aldrich) was added. Beads were washed three times in TBST and eluted by incubation with 3x FLAG peptide (Sigma-Aldrich) as per the manufacturer's instructions. Concentration of cell culture supernatants for Western blot analysis was achieved using 20-kDa MWCO iCON devices (Pierce).

    For deglycoslylation reactions, peptide N-glycosidase F enzyme (PNGase F) (NEB) was used according to the manufacturer's instructions. Briefly, purified vLIP protein was incubated first for 10 min at 100°C in 1x denaturation buffer (NEB) and then in the presence of PNGase F enzyme for 1 h at 37°C in 1x G7 buffer (NEB) containing NP-40.

    Phospholipase assays. Phospholipase assays on [14C]dioleoylphosphatidylcholine ([1,2-14C]oleate) (New England Nuclear, Boston, MA) were performed as follows. Unlabeled dioleoylphosphatidylcholine (160 μg; Avanti Polar Lipids, Alabaster, AL) was combined with 40 μg of unlabeled phosphatidic acid and 5 μCi of [14C]dioleoylphosphatidylcholine, and organic storage solvents were dried off under a stream of N2. Four hundred microliters of lipase assay buffer (100 mM Tris-HCl, pH 8.0, 150 mM NaCl, 2 mM CaCl2) was added to the dry lipid film. Lipids were sonicated, and 40 μl of lipid micelles was transferred to microcentrifuge tubes on ice. Assays were initiated by adding 10 μl containing 10 ng of either recombinant purified wt vLIP protein, S307A vLIP protein, or chicken K60 chemokine (baculovirus expressed) as a negative control to the 40 μl of assay solution. Bee venom phospholipase A2 (Sigma-Aldrich) (10 ng) was used as a positive control. Reaction mixtures were incubated at 37°C for 1 h, and reactions were stopped by the addition of 187.5 μl of MeOH:CHCl3 (3:1). Then 62.5 μl of 0.5 M HCl and 62.5 μl of CHCl3 were added, and reaction products were vortexed. Centrifugation at 2,300 x g was applied to separate phases, and 100 μl of the organic phase was removed, dried under a stream of N2, and resuspended in 10 μl of developing solvent (90:10:10, CHCl3:MeOH:CH3COOH) (8). Lipids were then loaded onto activated silica gel 60A TLC plates (Sigma-Aldrich) and developed in a preequilibrated TLC chamber at room temperature. After the solvent front reached within 1.5 cm of the top of the plate, the plate was removed, allowed to air dry, and then exposed to BioMax MR film (Kodak).

    Triacylglycerol lipase assays. Lipase assays on the radiolabeled triacylglyceride substrate [3H]triolein were performed essentially as described for the measurement of lipoprotein lipase activity by following the protocol of Briquet-Laugier et al. (16) using 12.8 μCi of freshly purified glycerol tri[9,10(n)-3H]oleate (Amersham). Radioactive lipid was mixed with 75 μl of a 200-mg/ml solution of cold triolein (Nu-Chek-Prep, Inc. Elysian, MN) and 10 μl of a 100-mg/ml solution of lecithin, and the mixture was dried under a stream of N2. Then 2.2 ml of 0.2 M Tris-Cl, pH 8.0, and 0.1 ml of 10% Pentex bovine serum albumin (Serologicals, Norcross, GA) were added to the dry lipid film and sonicated. Then 0.45 ml of 10% bovine serum albumin and 0.25 ml of heat-inactivated rat serum (a source of apolipoprotein-CII) were added to the substrate and mixed. Assays were performed in silica glass test tubes (Fisher Scientific, Pittsburgh, PA) and were initiated by adding recombinant protein to 100 μl of the substrate. Samples were incubated for 60 min at 37°C, and reactions were stopped by the addition of 3.25 ml of MeOH:CHCl3:heptane (1.41:1.25:1, vol/vol/vol). Liberated fatty acids were extracted into the aqueous phase by the addition of 1 ml of 0.1 M potassium carbonate-borate buffer (pH 10.5) followed by vigorous vortexing for 15 seconds. The tubes were then centrifuged at 500 x g for 10 min at room temperature to separate the organic and aqueous phases. Finally, 1 ml of the aqueous phase was removed from each tube, mixed with 10 ml of scintillation fluid, and measured for radioactive counts in a scintillation counter.

    Animal experiments. Specific-pathogen-free P2a line chickens (1 day old) were infected by intramuscular injection with 1,000 PFU of various viruses. Mock-infected birds received intramuscular injection of uninfected CEC in culture medium supplemented with 10% FBS. Experiment 1 included five groups of 15 birds infected with viruses reconstituted from pRB-1B, as well as an additional mock-infected group. The specific viruses tested were parental pRB-1B, pRB-1B vLIP-rev (vLIP-rev), pRB-1B vLIP-rev (vLIP-rev), pRB-1B vLIPMluI-SpeI (vLIP), and pRB-1B-vLIP S307A (vLIP S307A). Experiments 2 and 3 included vLIP-rev, vLIP, and vLIP S307A. In experiment 2, 8 birds per group were infected, and in experiment 3, 10 birds per group were used. Experiments were allowed to proceed for 8 weeks, after which all surviving birds were sacrificed and necropsies were performed in a randomized, blind manner. Birds that died within the first week of the experiments were excluded from the studies. During the course of the experiments, birds that presented with Marek's disease symptoms were sacrificed, and necropsies were performed. Tumor incidence, morbidity, and mortality were recorded. In the third experiment, blood samples were obtained from all birds at days 4, 12, 15, and 19 postinfection (p.i.) by wing vein puncture. Data were analyzed using the SAS package for Windows, v8.2 (SAS Institute, Cary, NC), using the chi-square and Fisher's exact tests.

    Real-time qPCR. Real-time quantitative PCR (qPCR) was performed on an ABI PRISM 7700 Sequence Detector (Applied Biosystems), and data were analyzed using ABI PRISM SDS 1.9.1 software on a Macintosh computer. For all reactions, TaqMan Universal PCR Master Mix, No AmpErase UNG (Applied Biosystems), was used. Reactions were run in MicroAmp 96-well plates (Applied Biosystems) in a final volume of 25 μl per reaction. Reaction conditions were 50°C for 2 min, 95°C for 10 min, followed by 40 cycles of 15 s at 95°C and 1 min at 60°C. Samples were measured in duplicate on DNA samples obtained from 10 μl of whole chicken blood. DNA was prepared using the QIAGEN DNeasy tissue kit using a final elution volume of 400 μl, and 5 μl was used in each qPCR. The qPCR probes and primer sets used in this study were configured to detect the MDV-1 gD gene or the chicken genomic inducible nitric oxide synthase (iNOS) gene exactly as described previously (Table 1) (36, 43). Probes for detection of gD and iNOS were labeled with 6-carboxyfluorescein (FAM) as the reporter and with Black Hole Quencher-1 (BHQ1) as the quencher dye (Eurogentec N.A., San Diego, CA). The primer pairs used were ck-iNOS F-ck-iNOS R and gD F-gD R (Table 1).

    Standards of known copy numbers were run in triplicate to generate a standard curve for each qPCR and consisted of a log10 serial dilution ranging from 5 x 106 to 5 x 101 copies in the case of the ck-iNOS target plasmid and from 5 x 106 to 5 x 100 copies in the case of MDV gD. The plasmid target used for generation of a standard curve for quantitation of chicken DNA copies in the unknown samples was plasmid pBS-iNOS, which harbors cDNA of the chicken iNOS gene (obtained from Keith Jarosinski) (50). The target DNA used for the generation of the gD standard curve was the MDV BAC20 clone (72). Probes were used at a final concentration of 250 nM, and primers were used at a final concentration of 900 nM. Threshold cycle number values (CT) were calculated to correspond to the upper 10-fold standard deviation of the background fluorescence signal.

    RESULTS

    Characterization of the vLIP transcript. The vLIP ORF had been identified in close proximity to the terminal repeat long (TRL) region in the UL region of the MDV genome (Fig. 1a). The lymphoblastoid MSB-1 cell line was used as a source of RNA samples for cDNA cloning and Northern blot analysis. A cDNA of the vLIP transcript spanning the first coding sequences to the poly(A) site was obtained by RT-PCR on RNA from sodium butyrate-induced MSB-1 cells. In contrast, vLIP could not be amplified from RNA samples derived from noninduced cells, indicating that vLIP is not transcribed during latency, and Northern blot analysis confirmed such an assertion (see below). Comparison of the vLIP cDNA to the corresponding genomic DNA sequence, shown in Fig. 1a, revealed that a 70-bp intron was removed from the vLIP transcript by splicing, which brings sequences encoding the signal peptide in frame with the rest of the vLIP ORF.

    In order to characterize the size and kinetic class of the vLIP transcript, RNAs from MSB-1 cells induced with butyrate were collected after treatment for 30 h with PFA, an inhibitor of viral DNA polymerase. A 32P-labeled riboprobe antisense to a 376-bp region of the vLIP reading frame, corresponding to vLIP amino acids 476 to 601, resulted in the detection of a single sodium butyrate-inducible, foscarnet-sensitive transcript of 2.4 kb (Fig. 1b). RNA integrity was monitored by UV imaging of ethidium bromide-stained RNAs, both prior to and after capillary transfer; levels of intact rRNA were comparable across all samples (Fig. 1b). From the above experiments we concluded that the MDV vLIP ORF is 2,271 bp in length and is expressed with late kinetics from one major spliced mRNA transcript of 2.4 kb.

    Analysis of vLIP amino acid similarity to cellular lipases. As a first approach to elucidate the function of the protein encoded by vLIP, comprehensive structural predictions were performed. The homology between vLIP and cellular lipases is especially significant in that structural residues essential to lipase activity are—for the most part—well conserved in the viral protein. Figure 2a depicts an amino acid sequence alignment of vLIP against homologous residues from a wasp venom phospholipase and the human lipoprotein lipase. A secondary-structure prediction for vLIP, generated by GenTHREADER analysis (45), is also included in Fig. 2a to demonstrate the placement of catalytic-triad positions in relation to conserved secondary-structure elements found among pancreatic lipases. The oxyanion hole residues, Phe/Trp 77 and Leu153 (by HPL numbering), which serve to stabilize tetrahedral reaction intermediates during enzyme activity, are conserved in vLIP and its adenovirus homologues. The serine residue at position 307 of vLIP (HPL numbering 152), representing the triad nucleophile (15, 79, 81), is also well conserved and is found in a typical HPL-like GHSLG motif.

    In contrast, the aspartic acid position of the catalytic triad, located at position 335 on the vLIP chain (HPL numbering 176), bears an asparagine substitution in each of the MDV strains for which sequence data are available (Fig. 2a) (49, 78). The equivalent HPL mutation, D176N, has been shown to lead to a loss of enzyme activity in HPL (52). Furthermore, sequence homology of vLIP to cellular lipases is much lower in the region of the catalytic-triad histidine residue (Fig. 2b). In this region, pancreatic lipases conserve a "lid" or "flap" domain, which is delimited by a pair of cysteine residues that form the structure by participating in a disulfide bond (21, 81, 82). In some lipases, the lid regulates access to the enzyme active site and accounts for an observed preference for water-insoluble substrates known as "interfacial activation" (44). The triad histidine residue is normally located 2 amino acid positions downstream of the second cysteine residue of the pair that forms the lid. While a pair of cysteines and a histidine are present in the alignment of vLIP with relevant cellular lipases, the second cysteine involved in lid formation does not align favorably and the putative triad histidine residue would be positioned on the "lid," instead of just after it (Fig. 2b). Thus, based on the structural criteria, it is not clear a priori whether MDV vLIP would behave like a prototypic (or conventional) lipase.

    MDV vLIP lacks lipase activity. Lipoprotein lipases are a subset of the pancreatic lipase gene family; thus, the asparagine substitution observed in vLIP at the acid position (D335N [Fig. 2a]) would most likely render the protein incapable of lipase activity (52). To eliminate the possibility that vLIP is a functional lipase enzyme whose triad acid residue has been shifted from the preferred location in strand ?6 to a position following strand ?7 (71), functional assays were performed to demonstrate or exclude enzymatic activity.

    In order to obtain significant amounts of recombinant protein, wt vLIP and the vLIP serine nucleophile site mutant S307A, which would be unable to make nucleophilic attacks on lipid substrates, were purified from insect cell culture supernatants. A Coomassie-stained SDS-PAGE gel and Western blot analysis of the steps performed during the purification of the recombinant proteins are shown in Fig. 3a. Highly purified recombinant vLIP protein was tested on 14C-labeled phospholipid and 3H-labeled triacylglyceride substrates in vitro. Experiments were repeated several times, and representative results are shown as thin-layer chromatography demonstrating separation of substrate and reaction products for a phospholipid substrate, 1,2-dioleoylphosphatidylcholine ([1,2-14C]oleate), in which each fatty acid chain is radiolabeled (Fig. 3b) and as a phase separation-based assay for the triacylglyceride substrate glycerol tri[9,10(n)-3H]oleate (triolein) (Fig. 3c). In each reaction, 10 ng of insect cell-derived recombinant vLIP was used. Recombinant wt vLIP did not yield activity on phospholipid substrates above background levels observed in buffer alone, with a nonspecific protein control (K60), or with the S307A vLIP mutant (Fig. 3b). In contrast, phospholipid hydrolysis could readily be observed in samples to which 10 ng of bee venom phospholipase was added (Fig. 3b). As shown in Fig. 3c, vLIP did not exhibit hydrolytic activity on a triacylgyceride substrate above background levels, either. From the results of the (phospho)lipase assays, we concluded that vLIP, consistent with the unfavorable D335N substitution, lacked detectable hydrolytic activity on (phospho)lipid substrates.

    The vLIP gene product is secreted and undergoes N glycosylation. When vLIP cDNA was expressed in mammalian 293T cells (data not shown) or in chicken cells during virus replication (see below), the size of the mature, secreted protein was observed to be approximately 120 kDa, although the calculated molecular mass of vLIP (residues 31 to 756) is 81.8 kDa, taking into account signal peptide cleavage, which is predicted to remove the first 30 amino acids of the vLIP polypeptide (57, 58). To analyze putative vLIP glycosylation in more detail, vLIP was expressed in T. ni insect cells using a baculovirus expression system. SDS-PAGE, followed by Western blotting with a peptide-specific rabbit antibody, was used to compare the electrophoretic mobilities of purified vLIP protein before and after treatment with PNGase F. Untreated vLIP had an apparent molecular size of approximately 110 kDa, but treatment with PNGase F increased the electrophoretic mobility of vLIP, and the apparent molecular size was reduced to approximately 76 kDa (Fig. 4). These findings clearly indicated that the discrepancy between the 81.8-kDa predicted size of the vLIP protein chain and the size observed when vLIP was expressed in eukaryotic cells can be accounted for by N glycosylation.

    Construction and in vitro characterization of vLIP MDV mutants. The results obtained from the sequence analyses and the lipase assays suggested that vLIP may not serve as a conventional lipase enzyme. In addition, previous results had suggested that vLIP was not essential for viral replication in culture (39). In order to test whether the lipase homology of vLIP had any measurable role during in vivo replication of MDV, vLIP mutant MDVs were constructed from pathogenic strain RB-1B cloned as a BAC (pRB-1B) (64). As outlined in Fig. 5, by using shuttle mutagenesis techniques in E. coli, two vLIP mutants were prepared from pRB-1B: an S307A point mutant of the nucleophile site (vLIP S307A) and an in-frame deletion of 173 amino acids in the lipase homologous region, spanning positions 256 to 426 (vLIP). Native (vLIP-rev) and C-terminally FLAG epitope tagged (vLIP-rev) vLIP revertant clones were prepared from the vLIP mutant BAC, also by shuttle mutagenesis in E. coli (Fig. 5).

    The DNA of each pRB-1B mutant was analyzed by several restriction enzymes in order to confirm that the appropriate changes were present and that no spurious rearrangements took place during shuttle mutagenesis. The results of an EcoRI digest are shown in Fig. 6a, where the DNA fragment encoding the vLIP gene migrated at 3.743 kb in the digest of the vLIP mutant owing to the loss of 519 bp, instead of the usual 4.262 kb observed in the digests of parental pRB-1B or vLIP-rev. The vLIP-rev mutant exhibited a slight decrease in mobility in the fragment encoding vLIP, causing it to run as a doublet with the next largest band in the digest, which is 4.524 kb (Fig. 6a). This shift reflects the incorporation of an additional 24 bp of sequence encoding the 8-amino-acid FLAG epitope. DNA sequencing of a 4-kbp region of each mutant and revertant BAC was performed to further confirm the genetic makeup of the vLIP region of each mutant BAC; it demonstrated that only the intended modifications were introduced (data not shown).

    Virus was recovered from BAC DNA after transfection of CEC, and growth kinetics demonstrated that there were no significant defects in virus replication of either vLIP mutant in vitro (Fig. 6c). To confirm that a FLAG-tagged vLIP was indeed expressed in virus-infected cells, the FLAG epitope-tagged revertant, vLIP-rev, was grown on CEC, and supernatants and cell lysates were compared to those from parental pRB-1B virus. Anti-FLAG immunoprecipitation, as visualized by Western blotting using both anti-FLAG antibody and the vLIP-specific rabbit polyclonal antibody, clearly demonstrated a distinct band of approximately 120 kDa specific to vLIP-rev supernatants (Fig. 6b). However, vLIP was not detected in infected-cell lysates or in concentrated supernatants by either antibody. The latter result suggested that, consistent with the presence of a suboptimal cytosine at +4 in the Kozak sequence at the translation initiation site of vLIP (ACAATGC), vLIP is not highly expressed during MDV replication in cultured cells (47-49, 78). From these results we concluded that vLIP was secreted upon lytic infection of chicken cells and that deletion or mutation of vLIP sequences did not have a detrimental effect on MDV growth properties in vitro.

    vLIP is a determinant of virulence. In a preliminary experiment, each of the mutant and revertant viruses was used to infect a group of 15 chickens of the susceptible P2a line. Mock-infected birds and animals infected with parental virus reconstituted from pRB-1B were used as negative and positive controls, respectively. In this first experiment, increased survival and reduced incidence of Marek's disease, characterized by wasting, visceral tumors, and paralysis, were observed in the vLIP mutant virus groups relative to those groups infected with the parental or revertant virus (Tables 2 and 3). Because the observed differences in mortality in the first experiment amounted only to about a 30% reduction from the levels observed for parental and revertant viruses, two further animal studies were conducted to confirm the observed effects. In order to minimize the number of experimental animals, the pRB-1B and the untagged vLIP revertant (vLIP-rev) group were not included in any further experiments. The reduction in the number of experimental animals was possible because the first trial had demonstrated that the parental pRB-1B, the FLAG-tagged revertant vLIP-rev, and the native vLIP revertant vLIP-rev were equally virulent, inducing characteristic gross T-cell lymphoma at rates greater than 90% (Table 2). Therefore, the notion that the incorporation of a FLAG epitope-tag at the C terminus of the vLIP reading frame might abrogate pathogenesis in vivo was no longer a concern.

    The results of the second and third experiments suggested that the reduction in virulence observed in the first study was a repeatable phenotype for both the vLIP deletion and nucleophile mutant viruses. In Table 2, the observed defect of vLIP mutant viruses in causing Marek's disease is summarized, as measured by incidence of gross lymphoma in visceral organs or musculature at necropsy. In the course of the three experiments, the vLIP-rev revertant virus was observed to cause gross lymphoma at rates of 92 to 100%, but vLIP caused gross tumors in only 44 to 64% of birds, while the vLIP S307A point mutant virus caused tumors in 58 to 71% of birds. Statistical analyses using Fisher’s exact and 2 tests demonstrated that the reductions in tumor incidence in both groups were statistically significant (P < 0.004).

    In order to elucidate whether any measurable defect in early lytic replication of vLIP mutant viruses could be identified, samples of whole blood were taken at days 4, 7, 12, 15, and 19 p.i. from every bird in the third experiment, which included 9 to 10 birds per group. Blood samples (10 μl) were subjected to direct DNA purification, and aliquots of DNA, each equivalent to 0.125 μl of whole blood, were analyzed by real-time PCR. This small quantity of DNA sample, though it included erythrocyte DNA, routinely was measured to contain 7 x 105 to 1.6 x 106 chicken genome copies, and copies detected per sample were consistent within each time point.

    While virus DNA could not be detected in every sample collected on day 4 p.i., virus was reliably detected in all samples from day 7 and all later time points. The in vivo data on levels in blood shown in Fig. 7a indicate that, compared to chickens infected with revertant virus, vLIP mutant virus-infected chickens experienced reduced levels of viral DNA in peripheral blood at late, but not early, time points during lytic virus infection. Average levels of virus DNA in peripheral blood appeared nearly indistinguishable between all three groups at days 7 and 12, but vLIP-rev showed two- to threefold-higher average levels of viral DNA at days 15 and 19 than the mutants, which remained indistinguishable from each other (Fig. 7a).

    Notably, the trend seen in Fig. 7a is echoed in the survival curves in Fig. 7b, which shows the percent survival per week using the combined data from experiments 2 and 3. The revertant virus was observed to cause onset of Marek's disease more rapidly than the vLIP mutants, which each showed increased numbers of asymptomatic birds surviving the full 8 weeks to the end of the experiment (Fig. 7b). Although a subset of birds surviving to the end of the experiment invariably presented with characteristic Marek's disease tumors upon necropsy, over the three experiments the vLIP deletion and nucleophile substitution mutant groups each consistently contained a number of birds that survived to full term without any apparent Marek's disease pathogenesis at necropsy (Table 3). The percentages of birds surviving to full term without gross lymphoma or other MD symptoms at necropsy ranged from 20 to 29% for vLIP S307A and from 25 to 44% for vLIP. In contrast, in the vLIP-rev and vLIP-rev groups, completely asymptomatic birds were never observed among the few that survived to full term, and the scarce examples of birds free of gross lymphoma (n = 2) represented cases where birds were sacrificed prior to the end of the experiment because they had presented with severe paralytic symptoms. From the experiments summarized in Table 2, Table 3, and Fig. 7, we concluded that deletion of vLIP or mutation of the protein's serine nucleophile resulted in reduced lytic virus replication in vivo and significantly lower numbers of chickens that developed and/or succumbed to Marek's disease.

    DISCUSSION

    In this study, we have demonstrated that the MDV vLIP gene product is derived from a late transcript, is glycosylated and secreted, and is critical for maintaining the wild-type pathogenic potential of MDV.

    The data obtained from the vLIP cDNA sequence indicated a postsplicing coding sequence of 2,271 bp with a poly(A) signal and a polyadenylation site located 12 and 25 bp from the termination codon, respectively. Taken together with the size of 2.4 kb visualized for the vLIP transcript during Northern blot analysis, the data suggest that the TATA box present at –69 bp from the vLIP start codon is most likely used to initiate vLIP transcripts (Fig. 1a). The classification of vLIP as a foscarnet-sensitive, late () transcript suggests that vLIP is coexpressed with other secreted viral factors implicated in immune modulation, such as vIL-8 and glycoprotein C (31, 63). Therefore, vLIP is likely to be expressed as maturing virus particles are becoming ready for transfer to new host cells. Like vIL-8, another virally encoded soluble mediator which has been shown to be important for wild-type pathogenesis of MDV (26, 63), secreted vLIP may be involved either in the recruitment of new host cells or in the down-modulation of immune surveillance by cytotoxic T cells or NK cells.

    Structural and lipase assay data present the first clues that vLIP, and at least some of its homologues in other avian herpesviruses and adenoviruses, may not behave as a conventional lipase enzyme in vivo. Lipases of the pancreatic lipase gene family employ /? hydrolase folds where the catalytic-triad acid residues are positioned immediately after the ?6 strand and can be typically located in the context of a GLDP motif (Fig. 2a), but other /? hydrolase fold lipases position their acid residues following the ?7 strand (71). PSI-BLAST and manual sequence analysis each suggest that vLIP is derived from the pancreatic lipase gene family, specifically from a lipoprotein lipase. Therefore, the presence of an uncharged (asparagine) instead of an acidic (aspartic acid) residue at the catalytic-triad acid position in vLIP argues strongly against the possibility that vLIP is a conventional lipase (Fig. 2a) (52), and this contention is indeed supported by negative lipase assays with recombinant vLIP (Fig. 3b and c). In addition, an N335D mutant was prepared, restoring the acid residue of the catalytic triad; however, this mutant also was negative in lipase assays (J. P. Kamil, unpublished data). Moreover, other sequence features of vLIP further substantiate the notion that vLIP is not a lipase. Besides poor homology in the lid region containing the triad histidine, two amino acid substitutions downstream of the acid residue, A178Y and P180V in reference to HPL, are similar to those (A178V and P180A) that have been experimentally demonstrated to account for the absence of lipase activity in PLRP1s (Fig. 2a) (10, 25, 67).

    Based on negative lipase assays and an unfavorable substitution at the essential catalytic-triad acid position, we presently hypothesize that vLIP does not perform as a traditional lipase. Yet we have demonstrated that the well-conserved serine nucleophile is essential for the wild-type pathogenic potential of MDV, as evidenced by the fact that the vLIP deletion mutant vLIP and the serine nucleophile point mutant vLIP S307A showed statistically indistinguishable attenuated phenotypes in vivo (Tables 2 and 3). The data on in vivo replication, survival, and tumor incidence presented in this study all indicate that vLIP and vLIP S307A are equally attenuated relative to the vLIP-rev revertant virus (Tables 2 and 3; Fig. 7). If the lack of lipase activity observed in assays of vLIP protein meant that the lipase homology was irrelevant to vLIP function, then we would have expected the serine nucleophile mutant to behave like wild-type virus in vivo. Instead, the in vivo data indicate that the serine nucleophile mutant vLIP S307A and the vLIP deletion mutant are equally attenuated relative to revertant virus.

    The seemingly paradoxical conservation of significant amino acid homology to catalytic elements of known lipases in a viral protein that appears to be devoid of conventional lipase activity invites the notion that "tinkering" may be at play in this example of viral homology to a cellular gene. Tinkering explains molecular evolution as a process by which new functions are developed through the modification of preexisting genes rather than through de novo invention (42). It is possible that vLIP has evolved to use lipase homology in a manner related to the cellular lipase(s) from which it was derived, but no longer as a conventional lipase.

    Certain proteins, including a number of examples from herpesviruses, undergo fatty acylation, where myristoyl or palmitoyl fatty acids become covalently attached at specific amino acids through the activity of enzymes that act in trans (22, 34, 51, 54, 66). Such modifications usually serve to anchor proteins at intracellular membranes and are often essential to proper protein function (13, 66). The example of vLIP might point to an alternative method by which viruses may gain covalent attachment of fatty acids to secreted proteins: via piracy of the host cell lipase gene(s). The /? hydrolases are understood to employ catalytic strategies similar to those of the serine proteases, such as chymotrypsin (15, 38, 74). Therefore, as /? hydrolases, the lipases homologous to vLIP employ a catalytic mechanism which requires that a stable "acyl enzyme" intermediate be formed, where a fatty acid is left covalently bonded to the serine nucleophile (38). Only once a water molecule is activated to perform a nucleophilic substitution to release the bound fatty acid does the free serine hydroxyl group of the active site become regenerated. Therefore, it is conceivable that the lipase homology of vLIP functions in autoacylation rather than in bona fide lipid hydrolysis. In such a scenario, the lipase reaction would stop at the stable acyl enzyme intermediate of the lipase reaction mechanism rather than continuing to perform a complete hydrolysis.

    Autoacylation might allow for the targeting of a pathogenic effector function to the plasma membranes of cells, or perhaps to lipoprotein complexes circulating in the blood. The lipid bonding hypothesis is supported by the appearance of cysteines near the vLIP nucleophile position. Cysteine is a good nucleophile and is known to accept palmitoyl fatty acids in certain cellular and viral proteins, including the Marburg virus glycoprotein (32, 66). These cysteines are positioned at –4, +6, +10, and + 21 relative to the vLIP serine nucleophile position (Fig. 2a) and are absolutely conserved among vLIP homologues in the MD-like genus as well as in the avian adenoviruses (20, 23, 37, 59, 80). Cysteine residues are not found so close to the nucleophile position in any characterized cellular lipase. However, it is interesting that certain as yet uncharacterized lipase homologues found in the genomes of Drosophila melanogaster (NP_609442) and Anopheles gambiae (XP_319853) share cysteines with vLIP at the –4 and +6 positions.

    The hypothetical role of the vLIP nucleophile in fatty acid bonding will be addressed in future studies, and other explanations may resolve the question of why the vLIP S307A serine nucleophile mutant and lipase domain deletion mutants showed very similar, attenuated phenotypes in vivo. However, it is worth noting that there are other examples of enzymatically inactive lipase homologues in nature. For instance, PLRP1s conserve appropriate catalytic-triad residues at all three positions and are found alongside PLRP2s and classical PLs in the gastric juices of all mammals examined to date but do not show any lipase activity in vitro (24, 28). Furthermore, several insect yolk proteins are homologous to the pancreatic lipases (14, 69, 76). Although lacking the serine nucleophile and thus catalytically inactive, these yolk proteins are believed to have roles in binding lipid-conjugated steroid hormones to help regulate insect development (14, 69).

    While the exact role of vLIP remains unsolved, our data strongly indicate that vLIP provides an advantage for growth in vivo but is completely dispensable in cell culture, which may point to a role in immune modulation—especially since the protein is secreted. Furthermore, we have demonstrated that the lipase homologous residues are critical to the pathogenic effector function of vLIP in vivo. Since vLIP belongs to a group of lipase-like open reading frames found in several avian adenoviruses and throughout the Marek's disease-like genus of herpesviruses, we offer the first evidence for a direct role in virus replication for this novel family of virus genes.

    ACKNOWLEDGMENTS

    J.P.K. gratefully acknowledges the expert technical assistance of D. R. Robinson, A. Bora Inceoglu, and Arun Bruhn at the University of California, Davis. We thank Martin Messerle (University of Halle, Halle, Germany) for kindly providing recombinant plasmid pST76K-SR and Sanjay Reddy (Texas A&M University, College Station, TX) for cosmid sn5.

    The project was supported by the National Research Initiative of the USDA Cooperative State Research, Education, and Extension Service, grant 2004-35204-14653 (awarded to J.P.K.). Additional support was provided by NIH and USDA awards to H.-J.K.

    REFERENCES

    Adler, H., M. Messerle, M. Wagner, and U. H. Koszinowski. 2000. Cloning and mutagenesis of the murine gammaherpesvirus 68 genome as an infectious bacterial artificial chromosome. J. Virol. 74:6964-6974.

    Afonso, C. L., E. R. Tulman, Z. Lu, E. Oma, G. F. Kutish, and D. L. Rock. 1999. The genome of Melanoplus sanguinipes entomopoxvirus. J. Virol. 73:533-552.

    Afonso, C. L., E. R. Tulman, Z. Lu, L. Zsak, D. L. Rock, and G. F. Kutish. 2001. The genome of turkey herpesvirus. J. Virol. 75:971-978.

    Akiyama, Y., and S. Kato. 1974. Two cell lines from lymphomas of Marek's disease. Biken J. 17:105-116.

    Allal, C., C. Buisson-Brenac, V. Marion, C. Claudel-Renard, T. Faraut, P. Dal Monte, D. Streblow, M. Record, and J. L. Davignon. 2004. Human cytomegalovirus carries a cell-derived phospholipase A2 required for infectivity. J. Virol. 78:7717-7726.

    Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman. 1990. Basic local alignment search tool. J. Mol. Biol. 215:403-410.

    Altschul, S. F., T. L. Madden, A. A. Schaffer, J. Zhang, Z. Zhang, W. Miller, and D. J. Lipman. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25:3389-3402.

    Baek, S. H., J. Y. Kwak, S. H. Lee, T. Lee, S. H. Ryu, D. J. Uhlinger, and J. D. Lambeth. 1997. Lipase activities of p37, the major envelope protein of vaccinia virus. J. Biol. Chem. 272:32042-32049.

    Bawden, A. L., K. J. Glassberg, J. Diggans, R. Shaw, W. Farmerie, and R. W. Moyer. 2000. Complete genomic sequence of the Amsacta moorei entomopoxvirus: analysis and comparison with other poxviruses. Virology 274:120-139.

    Bezzine, S., A. Roussel, J. de Caro, L. Gastinel, A. de Caro, F. Carriere, S. Leydier, R. Verger, and C. Cambillau. 1998. An inactive pancreatic lipase-related protein is activated into a triglyceride-lipase by mutagenesis based on the 3-D structure. Chem. Phys. Lipids 93:103-114.

    Borst, E. M., G. Hahn, U. H. Koszinowski, and M. Messerle. 1999. Cloning of the human cytomegalovirus (HCMV) genome as an infectious bacterial artificial chromosome in Escherichia coli: a new approach for construction of HCMV mutants. J. Virol. 73:8320-8329.

    Bourne, Y., C. Martinez, B. Kerfelec, D. Lombardo, C. Chapus, and C. Cambillau. 1994. Horse pancreatic lipase. The crystal structure refined at 2.3 ? resolution. J. Mol. Biol. 238:709-732.

    Boutin, J. A. 1997. Myristoylation. Cell. Signal. 9:15-35.

    Bownes, M. 1992. Why is there sequence similarity between insect yolk proteins and vertebrate lipases? J. Lipid Res. 33:777-790.

    Brady, L., A. M. Brzozowski, Z. S. Derewenda, E. Dodson, G. Dodson, S. Tolley, J. P. Turkenburg, L. Christiansen, B. Huge-Jensen, L. Norskov, et al. 1990. A serine protease triad forms the catalytic centre of a triacylglycerol lipase. Nature 343:767-770.

    Briquet-Laugier, V., O. Ben-Zeev, and M. H. Doolittle. 1998. Determining lipoprotein lipase and hepatic lipase activity using radiolabeled substrates. Methods Mol. Biol. 109:81-95.

    Calnek, B. W. 2001. Pathogenesis of Marek's disease virus infection. Curr. Top. Microbiol. Immunol. 255:25-55.

    Calnek, B. W., H. K. Adldinger, and D. E. Kahn. 1970. Feather follicle epithelium: a source of enveloped and infectious cell-free herpesvirus from Marek's disease. Avian Dis. 14:219-233.

    Calnek, B. W., and R. L. Witter. 1997. Marek's disease, p. 369-413. In B. W. Calnek (ed.), Diseases of poultry. Iowa State University Press, Ames.

    Cao, J. X., P. J. Krell, and E. Nagy. 1998. Sequence and transcriptional analysis of terminal regions of the fowl adenovirus type 8 genome. J. Gen. Virol. 79:2507-2516.

    Carriere, F., C. Withers-Martinez, H. van Tilbeurgh, A. Roussel, C. Cambillau, and R. Verger. 1998. Structural basis for the substrate selectivity of pancreatic lipases and some related proteins. Biochim. Biophys. Acta 1376:417-432.

    Chung, T. D., J. P. Wymer, M. Kulka, C. C. Smith, and L. Aurelian. 1990. Myristylation and polylysine-mediated activation of the protein kinase domain of the large subunit of herpes simplex virus type 2 ribonucleotide reductase (ICP10). Virology 179:168-178.

    Clavijo, A., P. J. Krell, and E. Nagy. 1996. Molecular cloning and restriction enzyme mapping of avian adenovirus type 8 DNA. Virus Res. 45:93-99.

    Crenon, I., E. Foglizzo, B. Kerfelec, A. Verine, D. Pignol, J. Hermoso, J. Bonicel, and C. Chapus. 1998. Pancreatic lipase-related protein type I: a specialized lipase or an inactive enzyme. Protein Eng. 11:135-142.

    Crenon, I., S. Jayne, B. Kerfelec, J. Hermoso, D. Pignol, and C. Chapus. 1998. Pancreatic lipase-related protein type 1: a double mutation restores a significant lipase activity. Biochem. Biophys. Res. Commun. 246:513-517.

    Cui, X., L. F. Lee, W. M. Reed, H. J. Kung, and S. M. Reddy. 2004. Marek's disease virus-encoded vIL-8 gene is involved in early cytolytic infection but dispensable for establishment of latency. J. Virol. 78:4753-4760.

    Cygler, M., and J. D. Schrag. 1997. Structure as basis for understanding interfacial properties of lipases. Methods Enzymol. 284:3-27.

    De Caro, J., F. Carriere, P. Barboni, T. Giller, R. Verger, and A. De Caro. 1998. Pancreatic lipase-related protein 1 (PLRP1) is present in the pancreatic juice of several species. Biochim. Biophys. Acta 1387:331-341.

    Diaz, B. L., and J. P. Arm. 2003. Phospholipase A(2). Prostaglandins Leukot. Essent. Fatty Acids 69:87-97.

    Egmond, M. R., and C. J. van Bemmel. 1997. Impact of structural information on understanding of lipolytic function. Methods Enzymol. 284:119-129.

    Friedman, H. M., L. Wang, N. O. Fishman, J. D. Lambris, R. J. Eisenberg, G. H. Cohen, and J. Lubinski. 1996. Immune evasion properties of herpes simplex virus type 1 glycoprotein gC. J. Virol. 70:4253-4260.

    Funke, C., S. Becker, H. Dartsch, H. D. Klenk, and E. Muhlberger. 1995. Acylation of the Marburg virus glycoprotein. Virology 208:289-297.

    Girod, A., C. E. Wobus, Z. Zadori, M. Ried, K. Leike, P. Tijssen, J. A. Kleinschmidt, and M. Hallek. 2002. The VP1 capsid protein of adeno-associated virus type 2 is carrying a phospholipase A2 domain required for virus infectivity. J. Gen. Virol. 83:973-978.

    Harty, R. N., G. B. Caughman, V. R. Holden, and D. J. O'Callaghan. 1993. Characterization of the myristylated polypeptide encoded by the UL1 gene that is conserved in the genome of defective interfering particles of equine herpesvirus 1. J. Virol. 67:4122-4132.

    Heikinheimo, P., A. Goldman, C. Jeffries, and D. L. Ollis. 1999. Of barn owls and bankers: a lush variety of alpha/beta hydrolases. Structure Fold Des. 7:R141-R146.

    Hennig, H., N. Osterrieder, M. Muller-Steinhardt, H. M. Teichert, H. Kirchner, and K. P. Wandinger. 2003. Detection of Marek's disease virus DNA in chicken but not in human plasma. J. Clin. Microbiol. 41:2428-2432.

    Hess, M., H. Blocker, and P. Brandt. 1997. The complete nucleotide sequence of the egg drop syndrome virus: an intermediate between mastadenoviruses and aviadenoviruses. Virology 238:145-156.

    Holmquist, M. 2000. Alpha/beta-hydrolase fold enzymes: structures, functions and mechanisms. Curr. Protein Pept. Sci. 1:209-235.

    Isfort, R., D. Jones, R. Kost, R. Witter, and H. J. Kung. 1992. Retrovirus insertion into herpesvirus in vitro and in vivo. Proc. Natl. Acad. Sci. USA 89:991-995.

    Iyer, S. S., J. A. Barton, S. Bourgoin, and D. J. Kusner. 2004. Phospholipases D1 and D2 coordinately regulate macrophage phagocytosis. J. Immunol. 173:2615-2623.

    Izumiya, Y., H. K. Jang, M. Ono, and T. Mikami. 2001. A complete genomic DNA sequence of Marek's disease virus type 2, strain HPRS24. Curr. Top. Microbiol. Immunol. 255:191-221.

    Jacob, F. 1977. Evolution and tinkering. Science 196:1161-1166.

    Jarosinski, K. W., R. Yunis, P. H. O'Connell, C. J. Markowski-Grimsrud, and K. A. Schat. 2002. Influence of genetic resistance of the chicken and virulence of Marek's disease virus (MDV) on nitric oxide responses after MDV infection. Avian Dis. 46:636-649.

    Jennens, M. L., and M. E. Lowe. 1994. A surface loop covering the active site of human pancreatic lipase influences interfacial activation and lipid binding. J. Biol. Chem. 269:25470-25474.

    Jones, D. T. 1999. GenTHREADER: an efficient and reliable protein fold recognition method for genomic sequences. J. Mol. Biol. 287:797-815.

    Kingham, B. F., V. Zelnik, J. Kopacek, V. Majerciak, E. Ney, and C. J. Schmidt. 2001. The genome of herpesvirus of turkeys: comparative analysis with Marek's disease viruses. J. Gen. Virol. 82:1123-1135.

    Kozak, M. 1987. At least six nucleotides preceding the AUG initiator codon enhance translation in mammalian cells. J. Mol. Biol. 196:947-950.

    Kozak, M. 1986. Point mutations define a sequence flanking the AUG initiator codon that modulates translation by eukaryotic ribosomes. Cell 44:283-292.

    Lee, L. F., P. Wu, D. Sui, D. Ren, J. Kamil, H. J. Kung, and R. L. Witter. 2000. The complete unique long sequence and the overall genomic organization of the GA strain of Marek's disease virus. Proc. Natl. Acad. Sci. USA 97:6091-6096.

    Lin, A. W., C. C. Chang, and C. C. McCormick. 1996. Molecular cloning and expression of an avian macrophage nitric-oxide synthase cDNA and the analysis of the genomic 5'-flanking region. J. Biol. Chem. 271:11911-11919.

    Loomis, J. S., J. B. Bowzard, R. J. Courtney, and J. W. Wills. 2001. Intracellular trafficking of the UL11 tegument protein of herpes simplex virus type 1. J. Virol. 75:12209-12219.

    Lowe, M. E. 1992. The catalytic site residues and interfacial binding of human pancreatic lipase. J. Biol. Chem. 267:17069-17073.

    Lupiani, B., L. F. Lee, X. Cui, I. Gimeno, A. Anderson, R. W. Morgan, R. F. Silva, R. L. Witter, H. J. Kung, and S. M. Reddy. 2004. Marek's disease virus-encoded Meq gene is involved in transformation of lymphocytes but is dispensable for replication. Proc. Natl. Acad. Sci. USA 101:11815-11820.

    MacLean, C. A., B. Clark, and D. J. McGeoch. 1989. Gene UL11 of herpes simplex virus type 1 encodes a virion protein which is myristylated. J. Gen. Virol. 70:3147-3157.

    McDermott, M., M. J. Wakelam, and A. J. Morris. 2004. Phospholipase D. Biochem. Cell Biol. 82:225-253.

    Nardini, M., and B. W. Dijkstra. 1999. Alpha/beta hydrolase fold enzymes: the family keeps growing. Curr. Opin. Struct. Biol. 9:732-737.

    Nielsen, H., J. Engelbrecht, S. Brunak, and G. von Heijne. 1997. Identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites. Protein Eng. 10:1-6.

    Nielsen, H., J. Engelbrecht, S. Brunak, and G. von Heijne. 1997. A neural network method for identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites. Int. J. Neural Syst. 8:581-599.

    Ojkic, D., and E. Nagy. 2000. The complete nucleotide sequence of fowl adenovirus type 8. J. Gen. Virol. 81:1833-1837.

    Oliver, K. G., R. M. Buller, P. J. Hughes, J. W. Putney, Jr., and G. J. Palumbo. 1992. Inhibition of agonist-induced activation of phospholipase C following poxvirus infection. J. Biol. Chem. 267:25098-25103.

    Ollis, D. L., E. Cheah, M. Cygler, B. Dijkstra, F. Frolow, S. M. Franken, M. Harel, S. J. Remington, I. Silman, J. Schrag, et al. 1992. The alpha/beta hydrolase fold. Protein Eng. 5:197-211.

    Palumbo, G. J., W. C. Glasgow, and R. M. Buller. 1993. Poxvirus-induced alteration of arachidonate metabolism. Proc. Natl. Acad. Sci. USA 90:2020-2024.

    Parcells, M. S., S. F. Lin, R. L. Dienglewicz, V. Majerciak, D. R. Robinson, H. C. Chen, Z. Wu, G. R. Dubyak, P. Brunovskis, H. D. Hunt, L. F. Lee, and H. J. Kung. 2001. Marek's disease virus (MDV) encodes an interleukin-8 homolog (vIL-8): characterization of the vIL-8 protein and a vIL-8 deletion mutant MDV. J. Virol. 75:5159-5173.

    Petherbridge, L., A. C. Brown, S. J. Baigent, K. Howes, M. A. Sacco, N. Osterrieder, and V. K. Nair. 2004. Oncogenicity of virulent Marek's disease virus cloned as bacterial artificial chromosomes. J. Virol. 78:13376-13380.

    Reddy, S. M., B. Lupiani, I. M. Gimeno, R. F. Silva, L. F. Lee, and R. L. Witter. 2002. Rescue of a pathogenic Marek's disease virus with overlapping cosmid DNAs: use of a pp38 mutant to validate the technology for the study of gene function. Proc. Natl. Acad. Sci. USA 99:7054-7059.

    Resh, M. D. 1999. Fatty acylation of proteins: new insights into membrane targeting of myristoylated and palmitoylated proteins. Biochim. Biophys. Acta 1451:1-16.

    Roussel, A., J. de Caro, S. Bezzine, L. Gastinel, A. de Caro, F. Carriere, S. Leydier, R. Verger, and C. Cambillau. 1998. Reactivation of the totally inactive pancreatic lipase RP1 by structure-predicted point mutations. Proteins 32:523-531.

    Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.

    Sappington, T. W. 2002. The major yolk proteins of higher Diptera are homologs of a class of minor yolk proteins in lepidoptera. J. Mol. Evol. 55:470-475.

    Schrag, J. D., and M. Cygler. 1997. Lipases and alpha/beta hydrolase fold. Methods Enzymol. 284:85-107.

    Schrag, J. D., F. K. Winkler, and M. Cygler. 1992. Pancreatic lipases: evolutionary intermediates in a positional change of catalytic carboxylates? J. Biol. Chem. 267:4300-4303.

    Schumacher, D., B. K. Tischer, W. Fuchs, and N. Osterrieder. 2000. Reconstitution of Marek's disease virus serotype 1 (MDV-1) from DNA cloned as a bacterial artificial chromosome and characterization of a glycoprotein B-negative MDV-1 mutant. J. Virol. 74:11088-11098.

    Spiegel, S., D. Foster, and R. Kolesnick. 1996. Signal transduction through lipid second messengers. Curr. Opin. Cell Biol. 8:159-167.

    Stryer, L. 1995. Catalytic strategies, p. 207-237. In Biochemistry, 4th ed. W. H. Freeman, New York, N.Y.

    Svendsen, A. 2000. Lipase protein engineering. Biochim. Biophys. Acta 1543:223-238.

    Terpstra, P., and G. Ab. 1988. Homology of Drosophila yolk proteins and the triacylglycerol lipase family. J. Mol. Biol. 202:663-665.

    Tjoelker, L. W., C. Eberhardt, J. Unger, H. L. Trong, G. A. Zimmerman, T. M. McIntyre, D. M. Stafforini, S. M. Prescott, and P. W. Gray. 1995. Plasma platelet-activating factor acetylhydrolase is a secreted phospholipase A2 with a catalytic triad. J. Biol. Chem. 270:25481-25487.

    Tulman, E. R., C. L. Afonso, Z. Lu, L. Zsak, D. L. Rock, and G. F. Kutish. 2000. The genome of a very virulent Marek's disease virus. J. Virol. 74:7980-7988.

    van Tilbeurgh, H., M. P. Egloff, C. Martinez, N. Rugani, R. Verger, and C. Cambillau. 1993. Interfacial activation of the lipase-procolipase complex by mixed micelles revealed by X-ray crystallography. Nature 362:814-820.

    Washietl, S., and F. Eisenhaber. 2003. Reannotation of the CELO genome characterizes a set of previously unassigned open reading frames and points to novel modes of host interaction in avian adenoviruses. BMC Bioinformatics 4:55.

    Winkler, F. K., A. D'Arcy, and W. Hunziker. 1990. Structure of human pancreatic lipase. Nature 343:771-774.

    Withers-Martinez, C., F. Carriere, R. Verger, D. Bourgeois, and C. Cambillau. 1996. A pancreatic lipase with a phospholipase A1 activity: crystal structure of a chimeric pancreatic lipase-related protein 2 from guinea pig. Structure 4:1363-1374.

    Witter, R. L. 1983. Characteristics of Marek's disease viruses isolated from vaccinated commercial chicken flocks: association of viral pathotype with lymphoma frequency. Avian Dis. 27:113-132.

    Witter, R. L. 1997. Increased virulence of Marek's disease virus field isolates. Avian Dis. 41:149-163.(Jeremy P. Kamil, B. Karst)