The novel HSV-1 US5-1 RNA is transcribed off a domain encoding US 5, US 4, US 3, US 2 and α22

Background The genome of herpes simplex virus 1 encodes at least 84 transcripts from which proteins are translated and several additional RNAs whose status as mRNAs is unknown. These RNAs include latency-associated transcript, OriS1 and OriS2 RNAs and in case of α4 null mutant additional transcript that spans the junction between L and S component of the HSV-1 genome. Current data do not suggest that a peptide is translated from these RNAs. Results We describe here a novel RNA designated US5-1 that spans 4.5 kb of the unique-short (US) region. The RNA initiates in US5 and terminates in the α22 open reading frame. It is expressed antisense to US5, US4, US3 and ICP22 mRNAs. This transcript is expressed with γ2 kinetics and has a half-life of 80 minutes. Conclusion These results identify a novel transcript encoded within HSV-1 genome. Since no major hypothetical open-reading frames are present in this transcript it is feasible that this RNA exerts its function as a non-coding RNA.


Background
The original report of the organization of the unique short (U S ) DNA sequences of the herpes simplex virus 1 (HSV-1) listed 12 open reading frames (ORFs) designated U S 1 (α22) through U S 12 (α47) [1]. Subsequent studies led to the discovery of three additional transcripts and at least two non coding transcripts. The coding transcripts were U S 8.5 mRNA co-terminal with U S 8 and U S 9 [2], U S 1.5 co-terminal with the α22 transcript [3], and U S 3.5 co-terminal with the U S 3 mRNA [4]. The non coding transcript Ori S 1 initiated in α22 or α47 ORFs, run across the origins of DNA synthesis and terminated at the transcription termination site of the α4 gene. Ori S 2 terminated at or near the transcription initiation site of the α22 or α47 gene [5]. Latency-associated transcript (LAT) is the only viral transcript detected in latently infected neurons [6]. Although several potential ORFs can be found within LAT sequence, none of them are transcribed in the context of viral infection [7,8]. In addition, Schaffer and colleagues [9] reported transcripts in cells infected with Δα4 mutant that spanned the junction between the L and S components of HSV-1 DNA. While it is suspected that LAT plays a role in the establishment and maintenance of latent infection, its function is not well established. Also, the roles of Ori S transcripts or Δα4-specific transcript are unknown. In this report we describe an additional long transcript designated U S 5-1 RNA. As the designation implies, this transcript originates in the U S 5 ORF and terminates in the α22 ORF.

RNA antisense to U S 3 is expressed in cells infected with wild-type virus
The experiments reported here resulted from the observation that a mutant derived from R7208 contained an unexplained 1.6 kb RNA antisense to the U S 3 ORF. Briefly, in this mutant the α22 gene was deleted and in the course of studies of a selected isolate of this mutant we found that it contained an insertion containing a stop codon upstream of the U S 3 ORF. To determine whether RNA antisense to U S 3 is expressed in cells infected with wild-type virus, rabbit skin cells (RSC) were exposed to 10 PFU of HSV-1(F) per cell, RNA was extracted 18 h later and subjected to RT-PCR or Northern blot analysis.
For RT-PCR analysis cDNA was generated from total RNA using primer 51(R), specific for RNAs antisense to U S 3. PCR reaction was performed using the 51(F) as forward and 51(R) as reverse primer. While reverse transcription with 51(R) primer would also generate cDNA from U S 2 transcript, this cDNA would not be amplified under PCR reaction used in this experiment since 51(F) primer lies upstream from U S 2 transcript (Fig. 1C). Results of RT-PCR analysis ( Fig. 2A) show that antisense RNA was also expressed in cells infected with wild-type virus. Northern blot analysis, performed using 1% agarose gel, (Fig. 2B) verified these results and additionally indicated that in cells infected with the wild type virus the RNA was approximately 4.5 kb long.

The U S 5-1 transcript spans genomic region between U S 5 and α22
Experiments were next carried out to determine the region of the S component spanned by the U S 5-1 mRNA.
To determine the location of the 5'-end we relied initially on observations carried out on the R7802 recombinant virus. In cells infected with this mutant, the truncated U S 5-1 RNA terminated at the polyadenylation signal inserted into the U S 2/U S 3 boundary (V. Jovasevic. and B. Roizman, manuscript in preparation). Knowing the size of this RNA (~1.6 kb), we estimated the location of its 5'end on the basis of the assumption that the sequences upstream of the insertion site would be identical in both R7802 and wild-type virus. In order to fine map the location of the 5'-end of the U S 5-1 transcript, we performed a series of northern blot analyses that helped us place the 5'-end within the sequences covered by the probe 3 (data not shown). To determine more precisely the location of the 5'-end of the U S 5-1 transcript we performed an RNase-protection assay using probe 3. The results of this experiment reveal the presence of a protected fragment with size of about 270 bases (Fig. 3A, arrow). In figure 1A the sequence of the probe 3 is shown in red and underlined are the first 270 bases of the probe 3 protected in the RNase protection assay. A potential TATA box sequence can be observed upstream from the last underlined base (boxed sequence), making it more likely that the underlined sequence of the probe 3 is the 5' terminus of the U S 5-1 transcript.
On the basis of the size of the U S 5-1 transcript and the location of its 5'-end we estimated the location of the 3'end to be within the ICP22 ORF. To determine more precisely the location of the 3'-end, RSC were infected with 10 PFU of HSV-1(F) per cell, RNA extracted 18 h later and subjected to RNase protection assay using a probe that spanned the predicted termination site identified within ICP22 ORF (probe 1). The probe 1 protected frag-ment of approximately 180 bases long (Fig. 3B, arrow). In figure 1B the sequence of the probe 1 is shown in red and underlined are the last 180 bases of the probe 1 protected in the RNase protection assay. In the vicinity of the identified 3'-end we observed a sequence that could potentially serve as a polyadenylation sequence for the U S 5-1 transcript (boxed sequence). In addition to the 180 bases long fragment we also observed another protected fragment that was equal in length to the full-length probe. This fragment does not correspond to the U S 5-1 transcript, considering that the U S 5-1 transcript is ~4.5 kb long. It is possible that a different RNA, expressed antisense to ICP22, is present in infected cells. Figure 1C illustrates schematically the domains of transcripts mapped in the S component to date. Singlestranded RNA probes used for Northern blot and RNase protection assay are represented by green arrows. Primers used for RT-PCR analysis are represented by blue arrowheads.

The U S 5-1 transcript is expressed with γ 2 kinetics
In the experiments described next, we evaluated the timing and requirements for the expression of the U S 5-1 RNA. In the first series of experiments, RSC were exposed to 10 PFU of HSV-1(F) per cell, RNA extracted 1, 3, 6, or 9 h later and subjected to northern blot analysis using probe 2 for detection of the U S 5-1 transcript. The results (figure 4A) show that the U S 5-1 could not be detected at 1 or 3 h after infection. It was first detected at 6 h after infection.
In the next series of experiments we determined whether the onset of synthesis of U S 5-1 requires de novo viral protein synthesis. RSC were infected with 10 PFU of HSV-1(F) per cell in the presence or absence of 100 μg/ml cycloheximide. Total RNA was extracted 9 h later and subjected to northern blot analysis. Addition of cycloheximide completely abrogated the expression of the U S 5-1 transcript (Fig. 4B), indicating that de novo protein synthesis is essential for its expression.
The purpose of the next series of experiments was to determine whether viral DNA synthesis was required for the synthesis of U S 5-1 RNA. RSC were exposed to 10 PFU of HSV-1(F) per cell in the presence or absence of 300 μg of PAA per ml. Total RNA was extracted 9 h later and subjected to northern blot analysis. PAA completely abrogated the expression of the U S 5-1 transcript (Fig.  4C), indicating that the replication of viral DNA is essential for its expression, and therefore that the U S 5-1 transcript is expressed with γ kinetics. Since the expression of the RNA was completely abrogated in the presence of PAA and not just diminished we conclude that the U S 5-1 transcript is expressed with γ 2 kinetics.

The U S 5-1 transcript has a half-life of about 80 minutes
In the final set of experiments we proceeded to evaluate the stability of the U S 5-1 transcript. RSC were infected with 10 PFU of HSV-1(F) per cell. At 6 h after infection the cells were exposed to 100 μg of actinomycin D per ml of medium. The RNA was extracted at 0, 1, 2, 3 or 4 h after the addition of the actinomycin D and subjected to northern blot analysis (Fig. 4D). The intensity of individual bands was measured by a densitometer and plotted as a function of time (Fig. 4E). The results shown in figure  4E indicate that the half-life of the U S 5-1 transcript was approximately 80 min.

Discussion
Relevant to this report are the following: (i) One possible explanation for the failure to detect the 4.5 kb RNA in earlier studies is the failure of transfer of high molecular weight RNA from gels containing high concentrations of agarose. In the studies reported here, we detected the 4.5 kb RNA on transfer from gels containing 1%, or less, of agarose (Fig. 2C), but not following transfer from gel containing 1.2% agarose (data not shown).
(ii). The U S 5-1 RNA is dependent on viral DNA synthesis for its accumulation. In contrast, the Ori S 1 RNA does not require de novo viral protein or DNA synthesis for its accumulation [10]. The measured half-life (80 min.) is in the line with that of other HSV-1 RNAs reported to be between 60 and 150 minutes regardless of the kinetic group to which a gene belongs [11].
(iii). In principle, viruses evolve continuously and do not retain DNA sequences that have no functions related to their survival in nature. The function of U S 5-1 RNA is not known. The U S 5-1 RNA contains only two large ORFs of 300+ codons. An ORF of approximately the same size is present in the corresponding sequences of HSV-2 DNA. However, In HSV-1 DNA the first large ORF initiates with the fifth methionine codon and its predicted amino acid sequences is not conserved in the corresponding HSV-2 ORF. We cannot exclude the possibility that the very small ORFs upstream of the large one are expressed. It seems it would be very unlikely that an RNA 4.5 kb long would encode a peptide with less than 100 amino acids, as it would be highly energy inefficient. However, it would not be unprecedented, since Drosophila's tal gene expresses 1.5 kb long transcript that codes for several 11 amino acid long peptides [12]. Overall, the data do not currently support the hypothesis that the U S 5-1 directs protein synthesis. (iv) As noted earlier, the accumulation of HSV-1 non coding RNAs is not unprecedented. There is considerable current interest in non coding RNAs inasmuch as they frequently serve as precursors of microRNAs or directly regulate the expression of genes located antisense to the RNA. We should note however that none of the studies published to date reported microRNA derived from the domain of the U S 5-1 RNA.

Conclusion
In this report we identified a novel HSV-1 transcript, expressed from the unique-short region of the viral genome. It spans the region from the U S 5 to the α22 genes and is transcribed antisense to U S 5, U S 4, U S 3 and α22 mRNAs, with γ2 kinetics. Our data suggest that U S 5-1 is a long non-coding RNA. The role of this transcript is currently unknown, but it is plausible that, similarly to other long non-coding RNAs, it is involved in the regulation of expression of viral genes.

Cells and viruses
Rabbit skin cells (RSC) were originally obtained from J. McClaren. The cells were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 5% fetal bovine serum. HSV-1(F) is the prototype HSV-1 strain used in the laboratory [13]. Mutant virus R7802, which has the deletion of the entire ICP22 ORF has been described elsewhere [14].

Northern blot
Cultures of RSC in 25-cm 2 flasks were either mock infected or infected with 10 PFU of virus per cell and maintained at 37°C in medium 199 V consisting of a mixture of 199 supplemented with 1% calf serum. The cells