Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2009 Nov 10.
Published in final edited form as: Virology. 2008 Sep 20;381(1):98–105. doi: 10.1016/j.virol.2008.08.024

Biological and genotypic properties of defective interfering particles of equine herpesvirus 1 that mediate persistent infection

Paul D Ebner 1,1, Seong K Kim 1, Dennis J O’Callaghan 1,*
PMCID: PMC2636567  NIHMSID: NIHMS77675  PMID: 18805562

Abstract

Infection with equine herpesvirus 1 (EHV-1) preparations enriched for defective interfering particles (DIP) leads to a state of persistent infection in which infected cells become lysis resistant and release both infectious (standard) virus and DIP. EHV-1 DIP are unique in that the recombination events that generate DIP genomes produce new open reading frames (ORFs; Hyb1.0 and Hyb2.0) consisting of 5’ sequences of varying lengths of the early regulatory gene IR4 fused to 3’ sequences of varying lengths of the UL5 regulatory gene. Only two additional ORFs (UL3 and UL4) are conserved. Because persistently infected cells release a heterogeneous mixture of DIP, characterization of the elements responsible for this altered state of infection has proved difficult. Here we describe a method for studying persistent infection using recombinant DIP (rDIP). Infection with rDIP resulted in the production of recombinant DIP that replicated faithfully to, at least, five passages and mediated a rapid progression to persistent infection as measured by: 1) production of cells resistant to lysis by the standard virus; and 2) infected cells that released both standard virus and DIP. High concentrations of rDIP also resulted in interference with the standard virus replication, another hallmark of persistent infection. rDIP deleted of UL3, UL4, and either Hyb gene, the only functional genes conserved in the DIP genome, replicated but exhibited markedly reduced ability to interfere with standard virus replication. Restoring only the Hyb genes (either Hyb1.0 or Hyb2.0), the IR4 gene, or specific portions of the IR4 gene restored interference. These data suggest that residues 144 to 196 of the IR4 protein within the HYB proteins are important for DIP interference and that persistent infection results from recombination events that produce DIP genomes.

Keywords: EHV-1, IR4, UL5, persistent infection, interference

Introduction

As with many viruses, high multiplicity infections with EHV-1 lead to the production of defective interfering particles (DIP; Huang and Baltimore, 1970; Campbell et al., 1976; Barrett and Dimmock, 1986; Deng et al., 2004; Ebner and O’Callaghan, 2006; O’Callaghan and Osterrieder, 2008). EHV-1 DIP result from recombination events that reduce the 155kbp EHV-1 genome to much smaller 6.0 to 7.5kbp DIP genomes that contain only three open reading frames (ORFs): UL3, UL4, and one of two unique hybrid ORFs (Hyb1.0 or Hyb2.0) which consist of portions of IR4, a homologue of herpes simplex virus-1 (HSV-1) ICP22, fused to portions of UL5, a homologue of HSV-1 ICP27. The origin of replication (ORIs) and cleavage/packaging sequences (CPS) are retained for replication and packaging (Baumann et al., 1984, 1986, 1987, 1989; Yalamanchili and O’Callaghan, 1990a, 1990b, 1991; Yalamanchili et al., 1990; Chen et al., 1996, 1999; Ebner and O’Callaghan, 2006). Comparisons between the standard virus genome and the DIP genomes are described schematically in Fig. 1.

Fig. 1. Organization of standard and defective EHV-1 genomes.

Fig. 1

Open in a new tab

(A) The standard EHV-1 genome is a two-isomer molecule of approximately 155kbp and is comprised of 78 known genes (Henry et al., 1981; Telford et al. 1991). (B) DIP genomes contain only UL3, UL4, a CPS, ORI, and one of two unique genes (Hyb1.0 and Hyb2.0) that form as a result of the recombination event that generates the defective genome. The Hyb genes contain sequences of IR4 fused to sequences of UL5. The dramatically truncated defective genomes (~6–7.5kbp) are repeated as a concatamer to produce packaged DNA approximately the size of the standard virus genome. Differences in DIP genome molecular weight are due to differences in the amounts of 3’ non-coding, repeat DNA sequences retained from the parent virus (Ebner and O’Callaghan, 2006). (C) HYB1.0 contains the amino 196aa of IR4P (from a total of 293aa) fused to the carboxy 68aa of UL5P (from a total of 470aa); HYB2.0 consists of the amino 228aa of IR4P fused to 15 aa* that are generated from a frameshifted reading of UL5 sequences. D) Organization of recombinant DIP plasmids. DIP genomes were inserted into pSV-β-galactosidase which contains a functional lacZ operon; DIP2 and DIP3 contain the Hyb1.0 gene while DIP1 contains the Hyb2.0 gene; ***DIP1, DIP2, and DIP3 each contain areas of non-coding sequences of different lengths (Ebner and O’Callaghan, 2006). See Table 1 for complete descriptions of each plasmid used in this study.

High concentrations of DIP lead to a state of persistent infection in which infected cells become lysis resistant but release both standard virus and DIP (Henry et al., 1979, 1980; Robinson et al., 1980; Dauenhauer et al., 1982). One focus of our laboratory has been the characterization of DIP elements responsible for this altered state of infection with an emphasis on the roles of the HYB proteins, the functions of which remain unclear. During lytic infection, both the IR4 protein (IR4P) and the UL5 protein (UL5P) are early auxiliary regulatory proteins that enhance immediate early protein (IEP)-mediated trans-activation of EHV-1 promoters (Holden et al., 1992, 1994, 1995; Zhao et al., 1995; Kim et al., 1997; Derbigny et al., 2000, 2002; Albrecht et al., 2004, 2005). In vitro reporter (chloramphenicol acetyl transferase) assays, however, showed that the HYB1.0 protein may have a negative effect on gene expression by independently down-regulating trans-activation of IE and IR4 promoters and reducing the ability of IR4P or UL5P to act synergistically with the IEP to up-regulate early EHV-1 promoters (Chen et al., 1996, 1999).

Persistently infected cells release a heterogenous mixture of DIP which has confounded efforts to identify DIP elements responsible for persistent infection (Henry et al., 1979, 1980; Robinson et al., 1980; Dauenhauer et al., 1982; Ebner and O’Callaghan, 2006). Here we describe a method of studying persistent infection that circumvents this and other obstacles to studying persistent infection by using recombinant DIP (rDIP). rDIP replicated faithfully and mimicked wild-type DIP in their ability to both produce lysis resistant cells and inhibit standard virus replication, which are two hallmarks of EHV-1 persistent infection. Mutant rDIP deleted of all functional genes (UL3, UL4, and either Hyb1.0 or Hyb2.0) replicated but did not interfere significantly with standard virus replication. Restoring sequences that encode either the IR4/UL5 hybrid protein or specific portions of the IR4 protein restored interference. Models describing persistent infection development and potential functions of HYB proteins during persistent infection are discussed.

Results

rDIP replicate in EHV-1- infected cells

Three distinct DIP genomes (DIP1, DIP2, and DIP3) were cloned into the expression plasmid pSV-β-galactosidase (Fig. 1). Mouse LM cells were transfected with the different rDIP constructs (rDIP1, rDIP2, rDIP3), subsequently infected with standard virus (MOI=0.1) to supply trans-acting factors and structural proteins, and passaged every 72 h (Fig. 2). The inclusion of a functional β-galactosidase gene in each rDIP construct allowed for rDIP quantitation. Supernatant from each passage was serially diluted and used to infect new monolayers. Helper (standard) virus was added to each sample at an MOI of 0.1. Infected monolayers were incubated for 5 days in medium containing a β-galactosidase substrate. Cells were fixed without staining and blue plaques were enumerated. rDIP replicated faithfully to high concentrations (>1.0 × 104 per ml) for the five passages tested (Table 2). In contrast, cells not transfected or transfected with only the empty parent vector (pSV-β-galactosidase) did not produce blue plaques at levels detectable by these methods (Table 2).

Fig. 2. Generation of recombinant defective interfering particles.

Fig. 2

Open in a new tab

LM cells were transfected with plasmids containing DIP genomes. Cells were subsequently infected with the standard virus to supply trans-acting factors and structural proteins, resulting in the production of recombinant DIP and the rapid progression to persistent infection. Details are given in the text and in Materials and Methods.

Table 2.

rDIP Replication in EHV-1 Infected Cells

Blue Plaque Forming Units Per mL
Pass no. Empty Vector DIP1 DIP2 DIP3 Control
1 None 3.4 × 106 3.2 × 106 1.2×104 None
2 None 1.2 × 107 3.2 × 107 2.0 × 106 None
3 None 1.3 × 107 8.9 × 106 1.7 × 106 None
4 None 2.6 × 106 7.7 × 106 5.7 × 106 None
5 None 6.0 × 104 6.0 × 104 2.4 × 104 None
Open in a new tab

rDIP were prepared, passaged, and quantitated as described in Materials and Methods. Empty vector = cells were transfected with only the parent vector pSV-β-galactosidase; DIP1, DIP2, and DIP3 = cells were transfected with DIP genomes cloned into pSV-β-galactosidase; Control = cells were not transfected prior to infection.

rDIP mediate a progression to persistent infection

Two hallmarks of EHV-1 persistently infected cells are: 1) resistance to lysis by the standard virus; and 2) the ability to release both standard virus and DIP (Henry et al., 1979, 1980; Robinson et al., 1980; Dauenhauer et al., 1982). To determine whether rDIP mediated a progression to persistent infection, supernatant from cells transfected with rDIP constructs was collected at each passage (72 h) and used to infect fresh monolayers. Monolayers were subsequently infected with standard virus and monitored for the development of lysis resistant colonies. Cells treated with supernatant from cells transfected with plasmids containing whole DIP genomes produced lysis resistant colonies at levels greater than cells treated with empty vector preparations or cells infected with only wild-type (wt) virus (Fig. 3).

Fig. 3. Persistent infection mediated by wild-type and recombinant EHV-1 DIP.

Fig. 3

Open in a new tab

Monolayers were infected with the standard virus, supernatant from cells transfected with the empty parent vector (negative control), supernatant of wild-type DIP infected cells, or supernatant of recombinant DIP (rDIP3 shown here) infected cells. Lysis resistant colonies were stained for visualization. Cells infected with standard virus or supernatant from cells transfected with only the empty parent vector and infected with the standard virus were completely lysed. In contrast, cells infected with supernatant from rDIP or wt DIP infected cells formed lysis resistant colonies.

Supernatant from each preparation was used to inoculate fresh monolayers to confirm the presence of standard virus (Table 3). In each case, standard virus was isolated from both rDIP and wt DIP preparations. Thus, rDIP mimic wt DIP in their ability to mediate a progression to persistent infection as measured by production of cells that are lysis resistant yet also release both infectious virus and DIP.

Table 3.

Release of Standard Virus by wtDIP- and rDIP-Infected Cells

Virus/DIP Preparation Virus in Supernatant log pfu/mL±SD
Standard EHV-1 2.9±0.05
Wt DIP 6.1±0.10
Empty vector 2.2±0.04
rDIP1 3.7±0.06
rDIP2 2.7±0.09
rDIP3 5.2±0.02
Open in a new tab

Persistent infection mediated by rDIP set up as described in the Materials and Methods. Monolayers of RK-13 cells were infected with the supernatant from cells infected with the standard virus, wild-type DIP generated from repeated high multiplicity infection with the standard virus (DI-172; 172nd high multiplicity passage) or recombinant DIP preparations (rDIP1, rDIP2, rDIP3; see Table 1 for genotypes). Plaques were quantitated after 24 h. Empty vector = cells transfected with pSV-β-galactosidase containing no DIP gene sequences (parent vector).

rDIP interfere with standard virus replication

A third hallmark of persistent infection is DIP-mediated reduction in standard virus replication. The ability of rDIP to interfere with virus replication was measured by quantifying the concentration of standard virus in rDIP infected cells at each passage. Supernatant was collected, serially diluted, and used to infect fresh RK-13 monolayers, and plaques were counted after 24 h. Standard virus replication was inhibited as early as the third passage in cells transfected with rDIP constructs. In cells transfected with rDIP1 or rDIP2, standard virus replication was reduced 1000-fold by the fourth passage (Fig. 4). In contrast, standard virus replication remained high (>108 pfu/mL) in infected cells not transfected with rDIP constructs and in infected cells transfected with only the parent vector (negative controls).

Fig. 4. EHV rDIP interfere with the replication of standard EHV-1.

Fig. 4

Open in a new tab

LM cells were transfected with plasmids containing different EHV-1 rDIP genomes and subsequently infected with the standard virus (MOI=1.0). Standard virus was quantitated at each passage (72h) by plaque assay. Empty vector = cells transfected with pSV-β-galactosidase; DIP1, DIP2, DIP3 = cells transfected with EHV-1 rDIP genomes cloned into pSV-β-galactosidase; Control = cells not transfected prior to infection.

In the case of rDIP1 and rDIP2 infected cells, standard virus replication increased to initial levels by the fifth passage. This same phenomenon (Von Magnus effect; Pauker et al., 1959) is seen in wt EHV-1 persistent infections where concentrations of standard virus increase and decrease as the concentrations of DIP reach and fall below inhibitory levels.

Interference during persistent infection is mediated by HYB proteins

To determine which DIP elements are involved in interference during persistent infection, rDIP containing deletions in UL3, UL4 and either Hyb1.0 (DIP2KO) or Hyb2.0 (DIP1KO) were generated (see Table 1 and Fig. 1 for full genotypic descriptions). The Hyb1.0 or 2.0 gene was restored to the DIP2KO and DIP1KO constructs to generate DIP2HybR and DIP1HybR, respectively. The absence of Hyb protein expression by DIP1KO and DIP2KO was confirmed by western blot analyses (Fig 5C and 5D, respectively). Likewise the restoration of Hyb protein expression by DIP1HybR and DIP2HybR was also confirmed by western blot analysis (Fig. 5C and 5D, respectively).

Table 1.

Plasmid nomenclature and characteristics

Plasmids Characteristics
pDIP1 contains genome of EHV-1 defective interfering particle #1 (see Fig. 1)
pDIP1KO contains genome of EHV-1 defective interfering particle #1 deleted of UL3, UL4, and Hyb2.0
pDIP1HybR pDIP1KO with Hyb2.0 restored
pDIP2 contains genome of EHV-1 defective interfering particle #2 (see Fig. 1)
pDIP2KO contains genome of EHV-1 defective interfering particle #2 deleted of UL3, UL4, Hyb1.0
pDIP2HybR pDIP2KO with Hyb1.0 restored
pDIP3 contains genome of EHV-1 defective interfering particle #3 (see Fig. 1)
pGFP vector pIRES-hrGFP-2a (Stratagene, CA)
pGFP-DIP1 pGFP vector with genome of DIP1
pGFP-DIP1KO pGFP vector with genome of DIP1KO
pGFP-DIP1KO-IR4(1–271) pGFP-DIP1KO with IR4(1–271)
pGFP-DIP1KO-IR4(1–196) pGFP-DIP1KO with IR4(1–196)
pGFP-DIP1KO-IR4(1–143) pGFP-DIP1KO with IR4(1–143)
Open in a new tab

Fig. 5. Interference with standard virus replication during persistent infection can be mediated by HYB proteins.

Fig. 5

Open in a new tab

Cells were transfected with rDIP plasmids containing various deletions described previously. Deletion of UL3, UL4, and either Hyb2.0 (DIP1KO; panel A) or Hyb1.0 (DIP2KO; panel B) resulted in a loss of interference. Restoration of either Hyb gene alone restored interference (DIP1HybR and DIP2HybR). Panels C and D: Detection of the IR4 protein encoded by standard virus and detection of the Hyb proteins encoded by rDIP was carried out by western blot analysis using anti-IR4 protein antiserum.

When all rDIP were tested for interference activity, cells transfected with the rDIP1KO or rDIP2KO constructs that were deleted of all functional genes showed no interference through passage five (Fig. 5A and 5B). In contrast, cells transfected with Hyb1.0 or Hyb2.0 restoration constructs (i.e., constructs containing Hyb1.0 or Hyb2.0 but not UL3 or UL4) displayed interference levels similar to those observed for cells transfected with the standard rDIP, indicating that the HYB proteins are capable of mediating interference during persistent infection. rDIP were also quantified at each passage as previously described, and deletion of UL3, UL4, and either Hyb gene had no effect on rDIP replication as cells transfected with rDIP1KO or rDIP2KO constructs produced blue plaques at levels similar to those of cells transfected with other rDIP constructs (Table 4).

Table 4.

Deletion of UL3, UL4, and Either Hyb Gene Does Not Affect rDIP Replication

Blue Plaque Forming Units/mL
Passage Empty DIP1 DIP1KO DIP1Hyb DIP2 DIP2KO DIP2HYB
1 <100 1.2 × 106 1.0 × 106 1.7 × 106 8.4 × 105 5.8 × 105 3.8 × 106
2 <50 5.0 ×106 4.9 × 106 8.9 × 106 4.5 × 106 6.1 × 106 7.2 × 106
3 <50 2.6 × 106 6.0 × 106 1.6 × 105 5.9 × 106 6.8 × 106 4.0 × 106
4 <50 1.1 × 106 3.5 × 105 1.0 × 105 6.0 × 105 2.7 × 106 2.0 × 106
5 <50 6.0 × 105 1.8 × 105 7.7 × 104 8.8 × 104 3.0 × 105 2.1 × 106
Open in a new tab

Recombinant DIP were generated as described in the Materials and Methods. DIP quantitation was based on lacZ production and generation of blue plaques. DIP1KO = DIP1 deleted of UL3, UL4, and Hyb2.0; DIP1HybR = DIP1KO with Hyb2.0 restored; DIP2KO = DIP2 deleted of UL3, UL4, and Hyb1.0; DIP2HybR = DIP2KO with Hyb1.0 restored.

Since restoration of either Hyb1.0 or Hyb2.0 restored interference activity, the IR4 protein sequences (aa 1 to 196) conserved in the Hyb protein may be important for DIP-mediated interference. To address this possibility, DIP1 and DIP1KO were cloned into the bacterial plasmid pIRES-hrGFP-2a (Stratagene, CA), and portions of the IR4 gene were cloned into the multiple cloning site (Fig. 6; see Materials and Methods). The HA tagged IR4 proteins were expressed by the human cytomegalovirus (CMV) immediate-early (IE) promoter, and expression was confirmed by western blot analysis (Fig. 6B). The method for rDIP generation with pGFP-DIP1 plasmids was modified from that shown in Fig 2 in that the cells were washed at 4 h postinfection in the case of passages 2 to 5. In cells transfected with pGFP-DIP1 and infected with EHV-1, standard virus replication was reduced 320-fold by the third passage as compared to that of cells transfected with the GFP vector (Fig. 7A). Interestingly, interference with standard virus replication was achieved by some rDIP that expressed only a portion of the IR4 protein. Interference was achieved but was delayed to the fourth passage in the case of GFP-DIP1KO-IR4(1–271) and GFP-DIP1KO-IR4(1–196) (Fig. 7A and 7B). However, DIP1KO expressing IR4(1–143) did not interfere with standard virus replication (Fig. 7B). These results suggested that the IR4 protein residues 144 to 196 are important for DIP interference.

Fig. 6. Generation of DIP1KO plasmids expressing truncated IR4 proteins.

Fig. 6

Open in a new tab

(A) Structures of pGFP-DIP1 plasmids expressing the truncated forms of IR4P. Truncated IR4 genes were inserted into the multiple cloning site (MCS) of pIRES-hrGFP-2a (Stratagene, CA). hrGFP, humanized recombinant green fluorescent protein. (B) Western blot analysis of IR4P truncation mutants. NBL6 cells were nucleofected with 1 pmol of pGFP-DIP1KO-IR4(1–271), pGFP-DIP1KO-IR4(1–196), pGFP-DIP1KO-IR4(1–143), or pGFP-DIP1KO. Cytoplasmic (C) and nuclear (N) extracts were prepared and subjected to SDS-PAGE, and the proteins were blotted to nitrocellulose and stained with the monoclonal anti-HA antibody (12CA5; Santa Cruz Biotechnology, Inc., CA). The arrow indicates the IR4(1–143). The numbers on the left represent molecular mass standards (Bio-Rad, CA) in kilodaltons.

Fig. 7. Interference with standard virus replication during persistent infection can be mediated by IR4P aa 144 to 196.

Fig. 7

Open in a new tab

(A and B) DIP interference assays with pGFP-DIP1KO plasmids expressing IR4 mutants. LM cells were transfected with plasmids containing different EHV-1 DIP genomes and subsequently infected with the standard virus (MOI=1.0). Standard virus was quantitated at each passage (72h) by plaque assay. The data are representative of four independent experiments. (C) Schematic diagram of HYB proteins and the deletion mutants of the IR4P. The top diagram represents the 293-aa IR4P of EHV-1. The numbers refer to the number of amino acids from the N terminus of each protein. TBP, TATA box-binding protein; NLS, nuclear localization signal.

Discussion

The progression from lytic infection to persistent infection has been observed for a variety of different viruses (Huang and Baltimore, 1970; Campbell et al., 1979; Barrett and Dimmock 1984; Deng et al., 2004; O’Callaghan and Osterrieder 2006, Ebner and O’Callaghan, 2006). The biological properties of this altered state of infection, however, remain understudied and unclear. In the case of EHV-1, persistent infection arises simultaneously with the production of defective interfering particles. EHV-1 DIP are unique in that the recombination events that produce DIP genomes result in the generation of one of two unique hybrid genes, Hyb1.0 and Hyb2.0, both being fusions of portions of two well characterized regulatory genes, IR4 and UL5 (Yalamanchili et al., 1990; Chen et al., 1996, 1999; Fig. 1). To our knowledge, these are the only unique genes produced by defective particles of an animal virus.

Traditionally, the study of EHV-1 persistent infection in general and defective interfering particles in particular, has been hampered by the difficulty in separating infectious particles from defective particles. What is more, persistently infected cells release a heterogeneous mixture of DIP with, in many cases, large genotypic differences (Ebner and O’Callaghan, 2006). Here, these problems were circumvented by employing a system whereby persistent infection was mediated by recombinant DIP (Fig. 1 and Fig 2). Transfecting cells with plasmids containing DIP genomes and subsequently infecting with standard virus rapidly produced persistently infected cells that released recombinant DIP. Inclusion of a functional LacZ gene allowed the quantification of DIP based on the ability to produce blue plaques. In this system rDIP replicated to at least five passages at which time the experiments were terminated. Moreover, like wt DIP, rDIP mediated a progression to persistent infection as measured by: 1) lysis resistance; 2) release of both standard virus and DIP; and 3) inhibition of standard virus replication.

rDIP constructs deleted of UL3, UL4, and either Hyb gene exhibited a great reduction in the ability to interfere with standard virus replication. Restoring either Hyb1.0 or Hyb2.0 alone restored interference. Our data support a model whereby the standard virus genome undergoes recombination events that generate R4/UL5 hybrid genes that encode HYB proteins capable of reducing standard virus replication to sub-lethal levels, resulting in lysis resistant cells that release both infectious virus and a heterogeneous mixture of DIP. It follows that UL3 and UL4 do not appear necessary for interference, albeit their role in generating DIP genomes remains to be tested in future experiments.

While each Hyb gene contains 5` portions of IR4 and 3` portions of UL5, the majority of amino acid sequences in each of the HYB proteins comes from the IR4 protein; Hyb1.0 and Hyb2.0 contain 67% and 78% of the parent IR4 protein, respectively. What is more, the UL5 sequences retained in Hyb2.0 are not in-frame which results in an early termination and the addition of 15 amino acids not found in either parent protein. While each HYB protein retains several functional domains of the parent IR4 protein, namely domains that mediate self-dimerization and interactions with the IE protein (IEP) and cellular TATA-box binding protein, neither HYB protein retains an NLS (Holden et al., 1994, 1995; Derbigny et al., 2000, 2002; Ebner and O’Callaghan, 2006; Ebner and O’Callaghan unpublished results 2008). Indeed, previous studies showed that in contrast to both IR4P and UL5P, each HYB protein displays a diffuse cytoplasmic staining pattern. As such, it was originally hypothesized that the HYB proteins might mediate interference by acting as a sink for important regulatory proteins, preventing their nuclear entry. Our preliminary studies indicated, however, that the viral IE, UL5, and IR4 proteins as well as cellular TBP were able to localize to nuclei of cells also producing either the HYB1.0 or HYB2.0 protein (Ebner and O’Callaghan, 2006). Those studies, however, employed transiently transfected cells which may not be ideal for studying the properties of DIP-associated proteins as the amount of HYB protein produced during persistent infection may dwarf that produced in cells simply transfected with plasmids containing Hyb genes. As such, further experiments are planned to address this hypothesis in the hope that a better understanding of the unique anti-viral properties of these proteins is achieved.

Our findings (Fig. 7) suggested that the IR4P residues 144 to 196 that harbor domains for TBP-binding (aa 142–220) and IEP-binding (aa 142–239) are important for DIP interference (Derbigny, et al., 2000, 2002; Ebner and O’Callaghan, unpublished data). In related experiments, we have shown that the unique EHV-1 IR2 protein (IR2P) negatively regulates viral gene expression and production, and interacts with TFIIB and TBP (Kim et al., 2006). Our observations that both the IEP and IR2P interact with TFIIB and TBP led to the hypothesis that the IR2P down-regulates EHV-1 gene expression by binding to viral promoters to prevent IEP binding and/or by squelching cellular factors essential for viral gene expression. The HYB1.0 and 2.0 proteins contain IR4P aa 1–196 and aa 1–228, respectively. Our published data showed that the IR4P minimally trans-activates EHV-1 promoters and acts synergistically with IEP and UL5P to trans-activate early and late viral promoters (Holden et al., 1995; Derbigny et al., 2002; Kim et al., 1997). Two C-terminal mutants of the IR4P IR4(1–271) and IR4(1–196) still trans-activated the EHV-1 EICP0 promoter; however, IR4(1–143) failed to trans-activate the early EICP0 promoter (data not shown), indicating that the IR4P residues 144 to 196 harbor sequences that are essential to mediate EHV-1 gene activation. These results suggest that during persistent infection, the HYB proteins, which are produced in enormous quantities, could interfere with standard virus replication by squelching cellular and viral factors essential for viral gene expression. It follows that truncation of IR4P residues 144 to 196 may alter the conformation of the protein such that the active inhibition site is not functional due to gross changes in structure. It should be noted, however, that interference could also result from cis-acting elements or trans-acting RNAs associated with Hyb1.0 or Hyb2.0. While we do not have any evidence of this, future experiments will need to be conducted to completely rule out this scenario.

It is unclear as to the mechanism(s) behind lysis resistance. It is possible that the HYB proteins may be involved in this process as well but this question was outside the scope of the current manuscript. It follows that the UL3 and UL4 proteins, while appearing to be unnecessary for interference, could play roles in the establishment of lysis resistance as well. As such, our laboratory is currently characterizing the functions of UL3 and UL4 (both during lytic and persistent infection), and the method described here of using recombinant DIP to study persistent infection should be able to address these and other questions regarding this altered state of infection.

Materials and Methods

Virus and Cells

EHV-1 Kentucky A (KyA) strain was propagated in mouse LM fibroblasts (O’Callaghan et al., 1968). LM cells were used in transfection assays. Rabbit kidney (RK-13) cells were also used for both infectious virus and rDIP quantitation.

Plasmids and DNA Sequencing

Plasmid nomenclature and characteristics are described in Table 1. The insertion of three DIP genomes (DIP1, DIP2, and DIP3) into bacterial plasmids is described elsewhere (Baumann et al., 1986). XbaI fragments containing entire DIP genomes of each plasmid were inserted into the XbaI site of the plasmid vector pSV-β-galactosidase (Invitrogen, Carlsbad, CA), resulting in the rDIP plasmids pDIP1, pDIP2, and pDIP3. pDIP1KO and pDIP2KO were generated by digesting pDIP1 and pDIP2 with BamHI and XbaI which released fragments containing the DIP genomes deleted for UL3, UL4 and Hyb2.0. The truncated DIP genomes were then religated into the parent plasmid. pDIP1HybR and pDIP2HybR were generated by digesting pDIP1 and pDIP2 with SalI and BglII which removed UL3 and UL4. The ends were filled in with Klenow fragment, and the backbones were religated. Sequence arrangements were confirmed by DNA sequence and restriction enzyme analyses. Fig. 1 describes the organization of the rDIP plasmids. To generate pGFP-DIP1, the XbaI fragment (DIP1 portion) of pDIP1 was cloned into the XbaI site of pIRES-GFP-X which was generated by inserting an XbaI linker (5’-CTAGTCTAGACTAG-3’) into NsiI site of pIRES-hrGFP-2a (Stratagene, CA). The XbaI fragment of pDIP1KO was cloned into the XbaI site of pIRES-GFP-X to generate pGFP-DIP1KO. The EcoRI and NheI fragment of pcDR4 (Holden et al., 1994) was cloned into the EcoRI and NheI sites of pSVSPORT1 (GIBCO BRL) to generate pSPORT-IR4(1–271). Plasmids pSPORT-IR4(1–271) was digested with NheI, and filled with Klenow fragment, and a SalI linker (5’-CGGTCGACGG-3’) was inserted and the plasmid was designated pSPORT-IR4(1–271)S. The EcoRI and SalI fragment was cloned into the EcoRI and SalI sites of pGFP-DIP1KO to generate pGFP-DIPKO-IR4(1–271). To generate pGFP-DIP1KO-IR4(1–143), the EcoRI and XhoI fragment of pcDR4 was cloned into the EcoRI and SalI sites of pGFP-DIP1KO. The 0.6-kb N-terminal DNA fragment of the IR4 gene from pcDR4 was amplified by PCR with the primers IR4196-F (5’-AGAAGGTACGCCTGCAGGTACCGGTCCGGA-3’) and IR4196-R2 (5’-GGGGTCGACGGCCGCTCACGTCACACTCCTCCCCAAA-3’) and cloned into the EcoRI and SalI sites of pGFP-DIP1KO to generate pGFP-DIP1KO-IR4(1–196).

rDIP generation and quantitation (blue plaque assay)

LM cells were transfected with the different rDIP constructs using a previously described lipofectin technique (Chen et al., 1999). Briefly, cells (5 × 105) were seeded at 80% confluence, allowed to attach, washed three times, and then incubated for 5h with 1-10µg of DNA resuspended in 20% lipofectin (Invitrogen). The lipofectin/DNA solution was replaced with Eagle’s standard medium containing 5% fetal calf serum, and the cells were incubated for 24h. Transfected cells were subsequently infected with the standard virus (MOI=1.0) and incubated at 37°C. Infected cells were harvested at 72h and diluted 1:5 in medium containing 2 × 106 cells/mL. Cells were passaged similarly four more times every 72h. rDIP were quantitated by infecting RK-13 monolayers with serial dilutions of each passage sample. Each sample was then infected with helper virus (wild type EHV-1) at MOI = 1. Infected monolayers were incubated in medium containing Bluo-gal (Difco, Livonia, MI) for 5 days. Cells were fixed with formalin and blue plaques were visualized and quantitated over a light box. rDIP concentrations were expressed as blue plaque forming units/mL. The generation of recombinant DIP is described schematically in Fig. 2.

Virus quantitation

Serial dilutions of samples from each passage were used to inoculate fresh RK-13 monolayers. Infected monolayers were incubated in medium containing 1.5% methylcellulose. Plaques were quantitated after 3 days by fixing with 10% formalin and staining with 0.5% crystal violet.

rDIP mediated persistent infection

RK-13 cell monolayers were infected with the equivalent of 0.1 DI particle/cell for all preparations containing rDIP, and two ml of DI72 (wild-type DIP; 1.1×107 PFU/ml). Infectious EHV-1 was added if necessary so that each infection was standardized to MOI of 20 (infectious virus). At 6 h post infection, infected cells were trypsinized and reseeded at 1:4 dilution. Lysis resistant colonies were fixed at 2–3 weeks and stained with 0.5% crystal violet for visualization and quantitation. The number of cells in lysis resistant colonies was determined by treating colonies with trypsin, staining live cells with erythrocin, and counting cells using a microscope. The concentration of infectious virus in the supernatant of rDIP preparations was determined by standard plaque assay as previously described.

Interference and rDIP quantitation

Interference was measured by quantitating the concentration of standard virus at each passage by standard plaque assay as described previously.

Western blotting

Extracts from cells infected with rDIP and standard virus were collected at the third passage. Proteins were isolated in RIPA buffer, separated by SDS-PAGE, and electrophoretically transferred to a positively charged nylon membrane (Bio-Rad, Hercules, CA). Membranes were blocked in 1% skim milk, rinsed in TBST, and incubated with the primary antibody (α-IR4 polyclonal antiserum; 1:10,000; Holden et al., 1994). Membranes were then washed again and incubated with a 1:10,000 dilution of secondary antibody (alkaline phosphatase-conjugated goat α-rabbit antiserum; Sigma, St Louis, MO). Protein-antibody complexes were visualized by incubating the membranes in NBT/BCIP solution (Sigma) according to the instructions from the manufacturer.

Acknowledgement

We thank Mrs. Suzanne Zavecz for excellent technical assistance. We thank Drs. Nikolaus Osterrieder and John Sixbey for helpful discussion and suggestions. This investigation was supported by research grants AI-22001 and P20-RR018724 from the National Institutes of Health. Paul D. Ebner was supported by NIH National Research Service Award F32 AI060113.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Albrecht RA, Kim SK, O’Callaghan DJ. The EICP27 protein of equine herpesvirus 1 is recruited to viral promoters by its interaction with the immediate-early protein. Virology. 2005;333:74–87. doi: 10.1016/j.virol.2004.12.014. [DOI] [PubMed] [Google Scholar]
  2. Albrecht RA, Kim SK, Zhang Y, Zhao Y, O’Callaghan DJ. The equine herpesvirus 1 EICP27 protein enhances gene expression via an interaction with TATA box-binding protein. Virology. 2004;324:311–326. doi: 10.1016/j.virol.2004.03.040. [DOI] [PubMed] [Google Scholar]
  3. Barrett AD, Dimmock NJ. Defective interfering viruses and infections of animals. Curr. Top. Microbiol. Immunol. 1986;128:55–84. doi: 10.1007/978-3-642-71272-2_2. [DOI] [PubMed] [Google Scholar]
  4. Baumann RP, Dauenhauer SA, Caughman GB, Staczek J, O'Callaghan DJ. Structure and genetic complexity of the genome of herpesvirus defective-interfering particles associated with oncogenic transformation and persistent infection. J. Virol. 1984;50:13–21. doi: 10.1128/jvi.50.1.13-21.1984. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Baumann RP, Staczek J, O'Callaghan DJ. Cloning and fine mapping of the DNA of equine herpesvirus type one defective interfering particles. Virology. l986;153:l88–200. doi: 10.1016/0042-6822(86)90022-x. [DOI] [PubMed] [Google Scholar]
  6. Baumann RP, Staczek J, O'Callaghan DJ. Equine herpesvirus defective-interfering (DI) particle DNA structure: the central region of the inverted repeat is deleted from DI DNA. Virology. l987;159:137–146. doi: 10.1016/0042-6822(87)90356-4. [DOI] [PubMed] [Google Scholar]
  7. Baumann RP, Yalamanchili VR, O'Callaghan DJ. Functional mapping and DNA sequence of an equine herpesvirus 1 origin of replication. J. Virol. 1989;63:1275–1283. doi: 10.1128/jvi.63.3.1275-1283.1989. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Campbell DE, Kemp MC, Perdue ML, Randall CC, Gentry GA. Equine herpesvirus in vivo: cyclic production of a DNA density variant with repetitive sequences. Virology. 1976;69:737–750. doi: 10.1016/0042-6822(76)90502-x. [DOI] [PubMed] [Google Scholar]
  9. Caughman GB, Staczek J, O'Callaghan DJ. Equine herpesvirus type l infected cell polypeptides: evidence for immediate early/early/late regulation of viral gene expression. Virology. 1985;145:49–6l. doi: 10.1016/0042-6822(85)90200-4. [DOI] [PubMed] [Google Scholar]
  10. Chen M, Harty RN, Zhao Y, Holden VR, O'Callaghan DJ. Expression of an equine herpesvirus 1 ICP22/CP27 hybrid protein encoded by defective interfering particles associated with persistent infection. J. Virol. 1996;70:313–320. doi: 10.1128/jvi.70.1.313-320.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Chen M, Garko KA, Zhang Y, O’Callaghan DJ. The defective interfering particles of equine herpesvirus 1 encode an ICP22/ICP27 hybrid protein that alters viral gene regulation. Virus Res. 1999;59:149–164. doi: 10.1016/s0168-1702(98)00128-2. [DOI] [PubMed] [Google Scholar]
  12. Dauenhauer SA, Robinson RA, O'Callaghan DJ. Chronic production of defective-interfering particles by hamster embryo cultures of herpesvirus persistently infected and oncogenically transformed cells. J. Gen. Virol. l982;60:1–14. doi: 10.1099/0022-1317-60-1-1. [DOI] [PubMed] [Google Scholar]
  13. Deng JH, Zhang YJ, Wang XP, Gao SJ. Lytic replication-defective Kaposi’s sarcoma-associated herpesvirus: potential role in infection and transformation. J. Virol. 2004;78:11108–11120. doi: 10.1128/JVI.78.20.11108-11120.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Derbigny WA, Kim SK, Caughman GB, O’Callaghan DJ. The EICP22 protein of equine herpesvirus 1 physically interacts with the immediate-early protein and with itself to form dimers or higher-ordered complexes. J. Virol. 2000;74:1425–1435. doi: 10.1128/jvi.74.3.1425-1435.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Derbigny WA, Kim SK, Jang HK, O’Callaghan DJ. EHV-1 EICP22 protein sequences that mediate its physical interaction with the immediate-early protein are not sufficient to enhance the trans-activation activity of the IE protein. Virus Res. 2002;84:1–15. doi: 10.1016/s0168-1702(01)00377-x. [DOI] [PubMed] [Google Scholar]
  16. Ebner PD, O’Callaghan DJ. Genetic complexity of EHV-1 defective interfering particles and identification of novel IR4/UL5 hybrid proteins produced during persistent infection. Virus Genes. 2006;32:313–320. doi: 10.1007/s11262-005-6916-y. [DOI] [PubMed] [Google Scholar]
  17. Henry BE, Robinson RA, Dauenhauer SA, Atherton SS, Hayward GS, O'Callaghan DJ. Structure of the genome of equine herpesvirus type l. Virology. 1981;115:97–114. doi: 10.1016/0042-6822(81)90092-1. [DOI] [PubMed] [Google Scholar]
  18. Henry BE, Newcomb WW, O'Callaghan DJ. Biological and biochemical properties of defective interfering particles of equine herpesvirus type l. Virology. l979;92:495–506. doi: 10.1016/0042-6822(79)90152-1. [DOI] [PubMed] [Google Scholar]
  19. Henry BE, Newcomb WW, O'Callaghan DJ. Alterations in virus protein synthesis and capsid production in cells infected with DI particles of herpesvirus. J. Gen. Virol. 1980;47:343–353. doi: 10.1099/0022-1317-47-2-343. [DOI] [PubMed] [Google Scholar]
  20. Holden VR, Yalamanchili RR, Harty RN, O'Callaghan DJ. ICP22 homolog of equine herpesvirus 1: expression from early and late promoters. J. Virol. 1992;66:664–673. doi: 10.1128/jvi.66.2.664-673.1992. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Holden VR, Caughman GB, Zhao Y, Harty RN, O'Callaghan DJ. Identification and characterization of the ICP22 protein of equine herpesvirus 1. J. Virol. 1994;68:4329–4340. doi: 10.1128/jvi.68.7.4329-4340.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Holden VR, Zhao Y, Thompson Y, Caughman GB, Smith RH, O’Callaghan DJ. Characterization of the regulatory function of the ICP22 protein of equine herpesvirus type 1. Virology. 1995;210:273–282. doi: 10.1006/viro.1995.1344. [DOI] [PubMed] [Google Scholar]
  23. Huang AS, Baltimore D. Defective viral particles and viral disease processes. Nature. 1970;226:325–327. doi: 10.1038/226325a0. [DOI] [PubMed] [Google Scholar]
  24. Kim SK, Ahn BC, Albrecht RA, O'Callaghan DJ. The unique IR2 protein of equine herpesvirus 1 negatively regulates viral gene expression. J. Virol. 2006;80:5041–5049. doi: 10.1128/JVI.80.10.5041-5049.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kim SK, Holden VR, O’Callaghan DJ. The ICP22 protein of equine herpesvirus 1 cooperates with the IE protein to regulate viral gene expression. J. Virol. 1997;71:1004–1012. doi: 10.1128/jvi.71.2.1004-1012.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. O'Callaghan DJ, Hyde JM, Gentry GA, Randall CC. Kinetics of viral deoxyribonucleic acid, protein, and infectious particle production and alterations in host macromolecular synthesis in equine abortion (herpes) virus-infected cells. J. Virol. 1968;2:793–804. doi: 10.1128/jvi.2.8.793-804.1968. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. O'Callaghan DJ, Osterrieder N. Herpesviruses of horses. In: Mahy B, van Regenmortel M, editors. Encyclopedia of Virology. 3rd Edition. Vol. 2. Oxford, UK: Elsevier Publishing Co.; 2008. pp. 411–420. [Google Scholar]
  28. Pauker K, Birch-Anderson A, Von Magnus P. Studies on the structure of influenza virus. I. Components of infectious and incomplete particles. Virology. 1959;8:1–20. doi: 10.1016/0042-6822(59)90017-0. [DOI] [PubMed] [Google Scholar]
  29. Robinson RA, Vance RB, O'Callaghan DJ. Oncogenic transformation by equine herpesviruses. II. Coestablishment of persistent infection and oncogenic transformation of hamster embryo cells by equine herpesvirus type 1 preparations enriched for defective interfering particles. J. Virol. 1980;36:204–219. doi: 10.1128/jvi.36.1.204-219.1980. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Telford EA, Watson MS, Davison AJ. The DNA sequence of equine herpesvirus-1. Virology. 1992;189:304–316. doi: 10.1016/0042-6822(92)90706-u. [DOI] [PubMed] [Google Scholar]
  31. Yalamanchili RR, O'Callaghan DJ. Sequence and organization of the genomic termini of equine herpesvirus type 1. Virus Res. 1990a;15:149–161. doi: 10.1016/0168-1702(90)90005-v. [DOI] [PubMed] [Google Scholar]
  32. Yalamanchili RR, O'Callaghan DJ. Organization and function of the ORIs sequence in the genome of EHV-1 DI particles. Virology. 1990b;179:867–870. doi: 10.1016/0042-6822(90)90157-m. [DOI] [PubMed] [Google Scholar]
  33. Yalamanchili RR, O'Callaghan DJ. Equine herpesvirus 1 sequence near the left terminus codes for two open reading frames. Virus Res. 1991;18:109–116. doi: 10.1016/0168-1702(91)90012-k. [DOI] [PubMed] [Google Scholar]
  34. Yalamanchili R, Raengsakulrach B, Baumann RP, O'Callaghan DJ. Identification of the site of recombination in the generation of the genome of DI particles of equine herpesvirus type 1. Virology. 1990;175:448–455. doi: 10.1016/0042-6822(90)90429-u. [DOI] [PubMed] [Google Scholar]
  35. Zhao Y, Holden VR, Smith RH, O'Callaghan DJ. Regulatory function of the equine herpesvirus 1 ICP27 gene product. J. Virol. 1995;69:2786–2793. doi: 10.1128/jvi.69.5.2786-2793.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES