Figure 1.
Efficient translation of ORF36, but not ORF37, occurs from the full-length ORF35–37 tricistronic transcript.
(A) A schematic presentation of the ORF34–37 genetic locus showing the previously identified ORF34–37 and ORF35–37 polycistronic mRNAs with thin and thick lines respectively. Coding potentials are indicated on the right. The ORF37-specific transcript is denoted as a dotted line. Start sites (SS) are indicated for each transcript according to the nucleotide position described by Russo et al. [18]. The single poly(A) signal used by all four ORFs for transcription termination is shown. (B–C) TREx BCBL1-RTA cells were mock treated (latent) or lytically reactivated for the indicated times. RNA was then isolated and Northern blotted with a 32P-labeled ORF36 (B) or ORF37 (C) strand-specific riboprobe. An additional higher molecular weight 293T-specific cross-reacting band was also detected in the ORF36 control lane, denoted by *. (D–E) 293T cells were transfected with the indicated plasmid, and total RNA and protein were isolated 24 h later. Protein lysates were resolved by SDS-PAGE and detected by Western blot with antibodies against ORF36 (D) or ORF37 (E). Actin served as a loading control. To verify transcript integrity, RNA was Northern blotted with 32P-labeled ORF36 (D) or ORF37 (E) DNA probes or with a probe against the GFP co-transfection control.
Figure 2.
The ORF35–37 mRNA is functionally bicistronic.
(A) Western blot analysis of 293T cells transfected with either N-terminally HA-tagged ORF35 with the native 5′ UTR (ORF35), C-terminally HA tagged ORF36 (ORF36) or the full length 5′ UTR HA-ORF35-ORF36-HA (ORF35/ORF36) DNA constructs. Equivalent amounts of protein lysates were resolved by SDS-PAGE and detected with anti-HA antibodies. (B) 293T cells were transfected with the indicated in vitro transcribed capped and polyadenylated RNA. Protein lysates were harvested 4 h post-transfection, resolved by SDS-PAGE and detected with anti-HA antibodies. The ribosomal protein S6RP served as a loading control for both experiments.
Figure 3.
Translation of ORF36 is independent of IRES activity and dependent on the 5′ mRNA cap.
(A) Diagram of dual luciferase transcripts. (B) The indicated in vitro transcribed, polyadenylated transcripts were electroporated into lytically reactivated TREx BCBL1-RTA cells. A dual luciferase assay was performed 4 h post-electroporation to determine the relative levels of firefly and Renilla luciferase activity. Experiment was performed in triplicate, error bars represent the standard deviation between replicates. (C) Schematic of 5′ UTR HA-ORF35-ORF36-HA containing a ΔG = −61 kcal/mol hairpin (Hp7) inserted after nucleotide position 32 in the native 5′ UTR. (D) 293T cells were transfected with the indicated WT or Hp7 plasmid shown in (C), and equivalent amounts of protein lysates were resolved by SDS-PAGE and Western blotted with anti-HA antibodies. S6RP served as a loading control. RNA samples were examined by Northern blot analysis with a 32P-labeled ORF36 DNA probe. GFP served as a co-transfection control. 18S rRNA was used as a loading control. (E) Schematic of 5′ UTR HA-ORF35-ORF36-HA indicating the nucleotide mutated to weaken the Kozak context flanking the ORF35 AUG (35 KCS wkn). (F) 293T cells were transfected with the indicated WT or 35 KCS wkn plasmid shown in (E), and equivalent amounts of protein lysates were resolved by SDS-PAGE and Western blotted with anti-HA antibodies. S6RP served as a loading control. RNA samples were examined by Northern blot analysis with a 32P-labeled ORF36 DNA probe. GFP served as a co-transfection control. 18S rRNA was used as a loading control.
Figure 4.
Two uORFs mediate translational control of ORF35 and ORF36.
(A) Schematic representation of the uORF organization indicating the nucleotides mutated to disrupt the uORF1 AUG (Δ1). (B, D, F, H) 293T cells were transfected with the indicated wild-type or mutant plasmids, and 24 h post-transfection protein lysates were resolved by SDS-PAGE and Western blotted with anti-HA antibodies. S6RP or actin served as a loading control. RNA samples were examined by Northern blot analysis with a 32P-labeled ORF36 or GFP DNA probe. GFP served as a co-transfection control in B, D and F. 18S rRNA was used as a loading control. (C) Diagram indicating the nucleotide mutations used to disrupt (Δ2) or weaken (KCS2 wkn) the context of the uORF2 start codon. (E) The ORF35 start codon mutant (AUG→AGA; Δ35) and uORF fusion reporter RNAs are depicted schematically. uORF1-Δ35 has the uORF1 stop codon disrupted (UGA→UGG) while uORF2-Δ35 has one nucleotide deleted from uORF2 to shift the reading frame +1 (A→Δ). (G) Schematic of the bicistronic plasmid in which the ORF36 coding region was replaced with GFP. Because ORF36 partially overlaps with ORF35, this required truncating the C-terminus of ORF35. The uORF2 AUG mutation to AGA is also shown.
Figure 5.
Ribosomal access to the ORF36 start codon occurs via linear scanning after termination of uORF translation.
(A) Schematic of the elongation of uORF2. The uORF2 stop codon and the four subsequent in-frame stop codons were mutated, artificially lengthening uORF2 from 11 to 64 amino acids. (B, D, F, H) 293T cells were transfected with the indicated wild-type or mutant plasmids, and 24 h post-transfection protein lysates were resolved by SDS-PAGE and Western blotted with anti-HA antibodies. S6RP served as a loading control. RNA samples were examined by Northern blot analysis with a 32P-labeled ORF36 DNA probe. GFP served as a co-transfection control. 18S rRNA was used as a loading control. (C) Schematic of AUG insertions at two locations in the ORF35 coding region, placed out of frame with ORF36. All AUGs were designed to have the two dominant Kozak consensus sequence nucleotides (A at −3 and G at +4). (E) Schematic of the wild-type 5′ UTR-HA-ORF35-ORF36-HA construct showing the location of the Hp7 insertion into the 5′ or 3′-proximal region of the ORF35 coding region. (G) Schematic of the wild-type 5′ UTR-HA-ORF35-ORF36-HA construct showing the location of the native AUG within the ORF35 codon region which has been mutated to AGA to generate the MidMut construct.
Figure 6.
Disruption of uORF2 alters ORF36 expression during lytic infection.
iSLK-PURO cells stably harboring the WT KSHV BAC16, a uORF2 mutant BAC16 (BAC16-Δ2), or a mutant rescue BAC16 (BAC16-Δ2-MR) were either untreated or lytically reactivated for 48 h. Protein lysates were Western blotted with antibodies against ORF36, the viral latent protein LANA and a viral lytic protein ORF57. S6RP served as a loading control.
Figure 7.
uORF1 and uORF2 are conserved among select γ-herpesviruses.
Alignment using ClustalW2 of (A) uORF1 or (B) uORF2 from KSHV, EBV, SaHV-2, AtHV-3 and RRV. Consensus nucleotides are indicated (three: asterisk; two: dot). uORF length is indicated on the right, and the uORF start codons are boxed.
Figure 8.
Model of the mechanisms of translation initiation used to translate ORF35, ORF36 and ORF37.
ORF35 translation is repressed by uORF1 and uORF2. ORF36 is translated from a termination-reinitiation event after translation of uORF2. ORF37 is translated from an ORF37-specific transcript generated from a promoter located within the coding region of ORF36.