1 Kaposi’s sarcoma-associated herpesvirus ORF66 is essential for late gene 1 expression and virus production via interaction with ORF34 2 3 4 Tadashi Watanabe 1 , Mayu Nishimura 1 , Taisuke Izumi 2+ , Kazushi Kuriyama 1 , 5 Yuki Iwaisako 1 , Kouhei Hosokawa 1 , Akifumi Takaori-Kondo 2 and Masahiro 6 Fujimuro 1 * 7 8 1: Department of Cell Biology, Kyoto Pharmaceutical University, Misasagi- 9 Shichono-cho 1, Yamashina-ku, Kyoto, Japan. 10 2: Department of Hematology and Oncology, Graduate School of Medicine, 11 Kyoto University, 54 Shogoin Kawasaki-cho, Kyoto, Japan 12 +: Present Address 13 Henry M. Jackson Foundation for the Advancement of Military Medicine Inc., in 14 Support of Military HIV Research Program, Walter Reed Army Institute of 15 Research, Silver Spring, MD, United States 16 17 Running title: KSHV ORF66 is essential for late gene expression 18 19 *: Address correspondence to Masahiro Fujimuro, Ph.D. 20 Department of Cell Biology, Kyoto Pharmaceutical University, Misasagi- 21 Shichono-cho 1, Yamashina-ku, Kyoto 607-8412, Japan; Tel: +81-75-595-4717 22 E-mail: [email protected]-phu.ac.jp 23 24 not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which was this version posted August 7, 2019. . https://doi.org/10.1101/728147 doi: bioRxiv preprint
44
Embed
associated herpesvirus ORF66 is essential for late gene ... · 8/7/2019 · 4 66. Introduction. 67 KSHV, (also known as human herpesvirus 8 or HHV-8), causes 68 Kaposi’s sarcoma,
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
Kaposi’s sarcoma-associated herpesvirus ORF66 is essential for late gene 1
expression and virus production via interaction with ORF34 2
3
4
Tadashi Watanabe1, Mayu Nishimura1, Taisuke Izumi2+, Kazushi Kuriyama1, 5
Yuki Iwaisako1, Kouhei Hosokawa1, Akifumi Takaori-Kondo2 and Masahiro 6
Fujimuro1* 7
8
1: Department of Cell Biology, Kyoto Pharmaceutical University, Misasagi-9
Shichono-cho 1, Yamashina-ku, Kyoto, Japan. 10
2: Department of Hematology and Oncology, Graduate School of Medicine, 11
Kyoto University, 54 Shogoin Kawasaki-cho, Kyoto, Japan 12
+: Present Address 13
Henry M. Jackson Foundation for the Advancement of Military Medicine Inc., in 14
Support of Military HIV Research Program, Walter Reed Army Institute of 15
Research, Silver Spring, MD, United States 16
17
Running title: KSHV ORF66 is essential for late gene expression 18
19
*: Address correspondence to Masahiro Fujimuro, Ph.D. 20
Department of Cell Biology, Kyoto Pharmaceutical University, Misasagi-21
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
Kaposi’s sarcoma-associated herpesvirus (KSHV) is closely 26
associated with B-cell and endothelial cell malignancies. After the initial 27
infection, KSHV retains its viral genome in the nucleus of the host cell and 28
establishes a lifelong latency. During lytic infection, KSHV encoded lytic-related 29
proteins are expressed in a sequential manner and are classified as immediate 30
early, early, and late gene transcripts. The transcriptional initiation of KSHV late 31
genes is thought to require the complex formation of the virus specific pre-32
initiation complex (vPIC), which may consist of at least 6 transcription factors 33
(ORF18, 24, 30, 31, 34, and 66). However, the functional role of ORF66 in vPIC 34
during KSHV replication remains largely unclear. Here, we generated ORF66-35
deficient KSHV using a BAC system to evaluate its role during viral replication. 36
While ORF66-deficient KSHV demonstrated mainly attenuated late gene 37
expression and decreased viral production, viral DNA replication was 38
unaffected. CHIP analysis showed that ORF66 bound to the promoters of late 39
gene (K8.1), but did not to those of latent gene (ORF72), immediate early gene 40
(ORF16) and early gene (ORF46/47). Furthermore, we found that three highly 41
conserved C-X-X-C sequences and a conserved leucine-repeat in the C-42
terminal region of ORF66 were essential for interaction with ORF34 and viral 43
production. The interaction between ORF66 and ORF34 occurred in a zinc-44
dependent manner. Our data support a model, in which ORF66 serves as a 45
critical vPIC component to promote late viral gene expression and viral 46
production. 47
48
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
KSHV ORF66, a late gene product, and vPIC are thought to contribute 50
significantly to late gene expression during the lytic replication. However, the 51
physiological importance of ORF66 in terms of viral replication and vPIC 52
formation remains poorly understood. Therfore, we generated a ORF66-53
deficient BAC clone and evaluated its viral replication. Results showed that 54
ORF66 played a critical role in virus production and the transcription of L genes. 55
To our knowledge, this is the first report showing ORF66 function in virus 56
replication using ORF66-deficient KSHV. We also clarified that ORF66 57
interacted with the transcription start site of K8.1 gene, a late gene. 58
Furthermore, we identified the ORF34-binding motifs in the ORF66 C- terminus: 59
three C-X-X-C sequences and a leucine-repeat sequence, which are highly 60
conserved among - and -herpesviruses. Our study provides insights into the 61
regulatory mechanisms of not only the late gene expression of KSHV but also 62
those of other herpesviruses. 63
64
65
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
KSHV, (also known as human herpesvirus 8 or HHV-8), causes 67
Kaposi’s sarcoma, primary effusion lymphoma (PEL), and multicentric 68
Castleman’s disease (1-3). Since the first discovery of KSHV DNA fragments in 69
a Kaposi's sarcoma lesion of an AIDS patient in 1994 (4), over a quarter of a 70
century has passed, and several aspects of KSHV pathogenesis, life-cycle and 71
viral protein function have been elucidated. Compared with other viruses, 72
herpesvirus has a large number of viral genes in its genome. KSHV encodes 73
not only viral proteins but also miRNAs and LncRNAs. These viral molecules 74
are thought to be essential for KSHV replication and its pathogenesis. One 75
characteristic of the KSHV life-cycle is the establishment of a lifelong latency in 76
the infected individual leading to KSHV-associated malignancy in patients with 77
severe immunosuppression by drugs after organ transplantation or AIDS (1-5). 78
Development of KSHV-associated neoplasm occurs due to infected cells which 79
express few latent KSHV genes (including LANA, v-FLIP, Kaposin, and 80
miRNAs) (6). These genes modulate cell-proliferation and apoptosis pathways 81
(7-11). Another characteristic of the KSHV life-cycle is active virus production, 82
known as lytic infection. The genes related to lytic infection have been 83
categorized into into three groups, immediate early (IE), early (E), and late (L) 84
(12, 13). Sequential and temporal expression of KSHV genes are key to induce 85
efficient viral production during lytic infection. The IE-gene product RTA/ORF50 86
is a transcription factor for triggering transcriptional activation of another IE 87
genes and E genes, and RTA/ORF50 initiates the shift from latency to lytic 88
infection (13, 14). The transcribed products of E genes start viral genome DNA 89
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
replication. Finally, the transcriptional products of L genes, contribute to viral 90
particle formation by its encoding viral structure proteins (15). 91
The viral pre-initiation complex (vPIC) has recently been proposed to 92
regulate L gene expression(16). vPIC is composed of several viral proteins 93
conserved among - and -herpesviruses (17), and has functional homology 94
with the host pre-initiation complex, which consists of TATA-binding protein 95
(TBP) and general transcriptional factors (GTFs). Whereas the host pre-96
initiation complex accumulates on the TATA-box of transcriptional start site 97
(TSS) and initiates cellular RNA polymerase II (RNA pol II)-mediated 98
transcription, vPIC accumulates on the “TATT”-box of viral gene TSS and 99
initiates RNA pol II-mediated transcription (17). 100
The function of vPIC machinery and its components has been 101
extensively studied in the context of EBV, MHV-68, CMV and KSHV(16). In 102
KSHV, at least 6 viral proteins contribute to vPIC formation. Viral TBP homolog 103
KSHV ORF24 directly binds to the “TATT”-box on the promoter sequences of 104
the KSHV genome and is essential for the recruitment of host RNA pol II (18, 105
19). We and other groups revealed that KSHV ORF34 acts as a hub for 106
interaction between ORF24 and other vPIC components such as ORF18, 107
ORF30, ORF31, and ORF66 (18, 20, 21). Split luciferase assays and co-108
immunoprecipitation experiments revealed that ORF34 directly/or indirectly 109
interacts with ORF 18, 30, 31 and 66 (18, 20-22). ORF24 binds to the promoter 110
of the L genes with RNA pol II, and ORF34 serves as a bridge between ORF24 111
and a complex of ORF18, 30, 31 and 66 (18, 20, 21). Furthermore, ORF18 (23), 112
ORF30 (21), ORF31 (23) and ORF34 (20) are essential for virus replication and 113
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
gene expression. Although ORF66 appears to be a vPIC component (18, 20, 114
22), the importance and its function during KSHV replication remains unknown. 115
Therefore, we established ORF66-deficient KSHV, and evaluated its 116
physiological role during viral replication. Here, we show that ORF66 is 117
essential for virus production and L gene expression via interaction with ORF34. 118
119
120
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
BAC16 or ORF66-BAC16, was transfected into iVero or iSLK cells, and then 137
selected with hygromycin to generate recombinant KSHV-inducible stable cell 138
lines, iVero-WT, iVero-ORF66, iSLK-WT, and iSLK-ORF66. To evaluate 139
whether ORF66 is critical for KSHV replication, virus production and virus 140
genome replication in iSLK-ORF66 and iVero-ORF66 cells were analyzed. 141
iSLK-WT, iSLK-ORF66, iVero-WT, and iVero-ORF66 cells were treated with 142
Dox and NaB, and culture supernatants were harvested. The amount of WT-143
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
KSHV and ORF66-KSHV in iSLK or iVero culture supernatants were 144
measured by real-time PCR (Fig. 2a and 2e). As a result, the production of 145
ORF66-KSHV in iSLK and iVero cells were about 1000-fold and 100-fold lower 146
than WT-KSHV. In contrast, there was no significant difference in cell-147
associated KSHV DNA levels between WT-KSHV and ORF66-KSHV 148
producing cells (Fig. 2b and 2f). Next, to clarify that the reduction of virus 149
production in iSLK-ORF66 and iVero-ORF66 cells is caused by ORF66-150
deficiency, we tested whether exogenous ORF66 expression could rescue virus 151
production of iSLK- and iVero-ORF66 cells. The iSLK-ORF66 and iVero-152
ORF66 cells stably transfected with empty plasmid or ORF66 expression 153
plasmid were cloned by neomycin (G418) or blastcidin S. The protein 154
expression levels of exogenous 3xFLAG-tagged ORF66 were confirmed by 155
Western blotting (Fig.2c and 2g). Virus production from iSLK-ORF66 and 156
iVero-ORF66 cell lines was partially but significantly recovered when 3xFLAG-157
ORF66 was exogenously expressed (Fig.2d and 2h). These results indicate that 158
ORF66 is crucial for KSHV replication and may function in steps following viral 159
DNA replication. 160
To evaluate the contribution of ORF66 on KSHV gene expression, we 161
performed an RT-qPCR array on viral gene. Total RNA was extracted from 162
iSLK-WT and iSLK-ORF66, stimulated with NaB and Dox for 72 hours. RNA 163
was subjected to RT-qPCR array as previously reported (26). Our data showed 164
a broad reduction of KSHV mRNA in iSLK-ORF66 compared to iSLK-WT 165
(Fig.3), where 7 out of the top 10 down-regulated genes were late genes (K8.1, 166
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
ORF27, and ORF25 have been previously reported by Nandakumar et al (27) 168
as direct vPIC targets. In contrast to L genes, Latent and IE genes were mildly 169
down-regulated. 170
If ORF66 functions as a vPIC component, we hypothesized that it 171
would be recruited to the “TATT”-box, which is known to be located 172
approximately 30 bp upstream of the transcription start site (TSS) of L genes 173
(27). To demonstrate this, we evaluated whether ORF66 interacts with L gene 174
promoters by ChIP assay. iSLK-ORF66 and iVero-ORF66 cells stably 175
expressing 3xFLAG-ORF66 were treated with or without NaB and Dox to induce 176
lytic infection, and then cells were subjected to ChIP assay using anti-FLAG 177
antibody (Fig. 4). As a result, immunoprecipitated ORF66 was found to be 178
bound to the promoters of K8.1 (L gene) but not to the promoters of latent gene 179
(ORF72), IE gene (ORF16) and E gene (ORF46/47). These results indicated 180
that a protein complex bearing ORF66 may interact with L gene promoters via 181
an interaction between ORF24 and ORF66. 182
183 The interaction of ORF66-ORF34 and its function in viral production.184
ORF66 was reported to interact with other vPIC components such as 185
ORF31, ORF18 and ORF34 by us and others (18, 20, 22). We also found that 186
ORF34 operated as the vPIC hub, by bridging ORF24 and vPIC components 187
including ORF66. Furthermore, these interactions are thought to be important 188
for functions of vPIC in gene expression. To gain further insight into the 189
interaction between ORF66 and ORF34, we identified the responsible region 190
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
within ORF66 for binding with ORF34. We made ORF66 truncated mutants that 191
each had approximately 60 amino acid deletions spanning from the N-terminus 192
to C-terminus of ORF66 (Fig. 5a). The interaction between ORF66 mutants and 193
ORF34 was assayed by pull-down experiments (Fig. 5b). 293T cells were co-194
transfected with 2xS-ORF66 mutants and 6xMyc-ORF34 plasmids, and 2xS-195
ORF66 were precipitated from cell extracts by S-protein-immobilized beads. As 196
a result, ORF66 1-4 truncated mutants interacted with ORF34. However, no 197
interaction with ORF66 5-7 truncated mutants was observed, meaning that 198
the C-terminal region (241a.a.- 429a.a.) of ORF66 is critical for ORF34 binding. 199
Furthermore, we performed a trans-complementation assay using these 200
truncation ORF66 mutants to assess how the loss of ORF34-ORF66 interaction 201
affects virus production. ORF66 mutant plasmids were transfected into iSLK-202
ORF66 and iVero-ORF66, and recovery of virus production was measured. 203
Wild type ORF66 expression significantly increased viral production in iSLK-204
ORF66 and iVero-ORF66 cells, while recovery of virus production was not 205
detected by expressing any of the ORF66 truncated mutants (Fig. 5c and 5d). 206
These results indicate that not only the ORF66 C-terminal domain (i.e., ORF34-207
binding region) but also the entire structure of ORF66 are indispensable for 208
virus production. 209
An amino-acid sequence alignment of the ORF66 C-terminal domain 210
of KSHV and other herpes virus homologs are depicted in Fig. 6a. Conserved 211
amino acids are indicated with a gray background. To identify the key residues 212
in the ORF66 C-termial region for interaction with ORF34, we constructed block 213
alanine-scanning mutants (CR1mut to CR9mut), where several neighboring 214
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
conserved amino-acids were substituted to alanine (Fig. 6a), except for proline 215
to avoid disruption the whole protein structure. These mutants were subjected 216
to pull-down assays to investigate the association of ORF34. Results showed 217
that ORF66 CR2, 3, 4 and 6 mutants bind to ORF34, whereas ORF66 CR1, 5, 218
7, 8 and 9mut did not bind (Fig. 6b). Therefore, we focused on specific amino 219
acid sequences of mutants losing the ability to bind to ORF34. In particular, 220
CR1, 5, and 9mut have mutations in the C-X-X-C consensus sequence. On the 221
other hand, CR7 contains a single cysteine residue, and CR8 contains three 222
leucine residues. 223
To obtain farther insight into ORF34 binding via CR1 to CR9 of ORF66 224
in viral production, ORF66 alanine substitution mutants (CR1 to CR9) were 225
subjected to trans-complementation assay using iSLK-ORF66 and iVero-226
ORF66 cells (Fig. 7a and 7b). Compared with ORF66 wild type expression, 227
virus production in both iSLK-ORF66 and iVero-ORF66 cells could not be 228
recovered by the expression of CR1, 5, 7, 8 and 9mut, which failed to interact 229
with ORF34 (Fig. 6b). On the other hand, expression of CR2, 3, 4 and 6mut that 230
maintained interaction with ORF34 showed varying levels of recovery, indicated 231
by red columns (Fig. 7a-b). CR3 and 6mut showed almost the identical levels of 232
recovery compared with ORF66 wild-type in iVero-ORF66 cells and partial 233
recovery in iSLK-ORF66 cells. Recovery by the CR4mut was significant, 234
however, lower than those of CR3 and 6mut. CR2mut had no effect. These data 235
revealed that the binding between ORF66 and ORF34 via the conserved 236
amino-acids of CR1, 5, 7, 8 and 9 regions in ORF66 is necessary, but not 237
sufficient for virus replication. 238
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
The C-X-X-C sequences and Zinc binding of ORF66 are important for 240
association with ORF34. 241
Based on our results of ORF66 alanine scanning mutants, we found 242
that CR1, CR5 and CR9 regions of ORF66 contain a C-X-X-C sequence which 243
is critical for ORF34 binding (Fig. 6a). Therefore, we generated expression 244
plasmids with single amino acid mutants ORF66 C295A, C298A, C341A, 245
C344A, C424A and C427A in which a cysteine in C-X-X-C was substituted with 246
an alanine. Plasmids of ORF66 G299A, S342A, G345A and G345A mutants in 247
which the conserved glycine or serine around the C-X-X-C sequence was 248
substituted with alanine were also generated. The plasmids of 6xMyc-ORF34 249
and ORF66 alanine mutants were co-transfected into cells, and cell extracts 250
were subjected to affinity purification using S-protein immobilized beads, 251
followed by Western blotting. As a result, C295A, C298A, C341A, C344A, 252
C424A and C427A ORF66 mutants failed to interact with ORF34 (Fig 8a, b and 253
d). Furthermore, two leucine residues at 412 and 413 in CR8 of ORF66 were 254
also important for the association with ORF34 (Fig 8c). These results suggest 255
that a conserved leucine-repeat (412L and 413L) in CR8 and three conserved 256
C-X-X-C sequences in CR1, 5 and 9 are needed for binding to ORF34, leading 257
to the appropriate formation of vPIC. Because the C-X-X-C motif in proteins is 258
often related to the binding to bivalent-cation, we speculate that C-X-X-C 259
sequence contributes to form a higher order structure such as a zinc-finger 260
domain. Next, to elucidate whether capturing zinc ion within ORF66 is 261
necessary for binding to ORF34, we performed pull-down assays using zinc 262
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
chelator TPEN and 2xS-ORF66-conjugated S-protein beads. The ORF66-263
conjugated beads were mixed with the cell extracts including overexpressed 264
Myc tagged ORF34 in the presence of TPEN and were pull-down. The 265
precipitates were subjected to western blotting with anti-Myc. As a result, TPEN 266
decreased the interaction of ORF66 with ORF34, indicating that ORF66 may be 267
a zinc-binding protein (Fig. 8e). 268
269
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
The C-terminus of ORF66 is highly conserved among herpesvirus 271
homologs compared with the N-terminus (Fig.6a). We prepared alanine-272
scanning mutants of ORF66 for amino acid residues that are conserved among 273
MHV, BHV, EBV, HCMV, and HHV-6. The conserved amino acid regions of 274
CR1, 5, 7, 8 and 9 in the ORF66 C-terminal domain is necessary for the binding 275
between ORF66 and ORF34, and virus production. In addition, three conserved 276
C-X-X-C sequences in ORF66 CR1, 5 and 9 were needed for binding to ORF34 277
(Fig. 8). Our results indicate that ORF66 leads to appropriate vPIC formation by 278
the interaction of ORF34 via C-X-X-C sequences in the C-terminal domain of 279
ORF66. We speculate that the conserved C-X-X-C sequences could form a 280
higher order structure such as a zinc-finger domain. Therefore, we approached 281
the protein structure and functional prediction of ORF66 by a server-based 282
helical protein structure simulation, I-TASSER (Iterative Threading ASSEmbly 283
Refinement) (28-30) and generated a full-length homology model of ORF66 284
(Fig. 9). According to a meta-server approach to protein-ligand binding site 285
prediction (COACH) (31, 32), four cysteine residues, C295, C298, C341, and 286
C344 are predicted to associate with a zinc ion. Based on the location of each 287
cysteine residue in the homology model, a pair of cysteines, C295/C298 in 288
CR1/CR2 or C341/C344 in CR5, might chelate a single molecule of zinc, which 289
correlates with our experimental observation using a zinc chelator, TPEN (Fig. 290
8). The zinc chelator TPEN inhibited association between ORF66 and ORF34. 291
This result indicates that ORF66 binds zinc, which is important for its interaction 292
with ORF34. We also performed homology searching of ORF66 by SWISS-293
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
have low homology with the TFIIB zinc-ribbon domain of hyperthermophilic 295
archaea Pyrococcus furiosus. As these zinc-associated cysteine residues were 296
mostly located inside the protein structure, except for C341 (Fig. 9; middle 297
panel), it is implies that CR1, CR2, and CR5 residues are involved in 298
maintaining the functional structure formation of ORF66 via zinc interaction. 299
Two leucine residues (L412 and L413) in LLQL of CR8 are essential 300
for binding between ORF66 and ORF34 (Figure 6b and 8c), suggesting that the 301
hydrophobicity of a leucine-rich sequence is also indispensable for ORF66 to be 302
functional. In fact, L412 and L413 residues in CR8 are predicted to be fully 303
exposed on the protein surface, which represented a high degree of 304
hydrophobicity by our structural model (Fig. 9). Two cytosine residues (C424 305
and C427) in CR9 were not predicted to be a zinc-binding domain by our 306
protein-ligand binding prediction, however these residues in ORF66 were also 307
responsible for ORF34 interaction (Figure 6). These residues are exposed on 308
the protein surface (Figure 9; middle panel), and located at the hydrophobic 309
surface including LLQL motif (Figure 9; right panel). Thus, the hydrophobic 310
region surrounding two leucine residues in CR8 and cysteine residues in CR9 is 311
expected to make up the binding surface and be directly involved in the 312
hydrophobic interaction with ORF34. 313
We evaluated the viral replication in KSHV ORF66-producing iVero 314
and iSLK cells, which were stably integrated with an ORF66-dificient KSHV 315
BAC clone. ORF66 plays a critical role in virus production and the transcription 316
of L genes. KSHV ORF24 binds to ORF34, RNA pol II and the TATT-box of the 317
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
transcriptional start site (TSS) of the L gene (18, 20, 21). Because direct or 318
indirect interaction of ORF34 with ORF 18, 30, 31 and 66 were indicated by co-319
immunoprecipitation and split luciferase experiments (18, 20, 21), ORF34 has 320
been thought to function as a hub for interactions between ORF24 and other 321
vPIC components. Therefore, these reports, in addition to our viral replication 322
kinetics and viral gene expression data imply that ORF66 engages in L gene 323
expression as a vPIC component. These results are in line with other 324
herpesvirus homologs of KSHV ORF66. For instance, EBV BFRF2 is essential 325
for virus production and contributes to vPIC formation (33). HCMV UL49 is also 326
essential for replication in human foreskin fibroblasts (34, 35). 327
To gain a better understanding of ORF66 within the vPIC complex, we 328
attempted to search for ORF34-binding regions within ORF66. Pull-down 329
assays using truncated ORF66 mutants showed that the C-terminal region 330
(from 241 a.a. to 429 a.a/the C-terminus end) of ORF66 was responsible for 331
binding to ORF34 (Fig. 5b). And the truncated ORF66 mutants were subjected 332
to a trans-complementation assay. However, all truncated ORF66 mutants did 333
not rescue virus productions in both KSHV-ORF66 iVero and iSLK cell lines 334
(Fig. 5c and 5d). Therefore, we speculate that the entire structure of ORF66 is 335
necessary for virus production through vPIC formation. Considering the 336
complex structure of human PIC components (TBP and GTFs) and RNAPII 337
(36), vPIC is might be a crowded complex that consists of not only vPIC factors 338
but also host proteins such as RNAPII, RNAPII binding proteins, and other 339
unknown host. Lacking large regions of ORF66 may also influence the overal 340
structure of the protein and affect interaction with its binding partners. Another 341
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
possibility is that the N-terminal and center regions of ORF66 are related to 342
interaction with other vPIC components or unknown host factors. 343
To evaluate the physiological function of conserved amino acids, block 344
alanine-scanning mutants of ORF66 were subjected to a trans-complementation 345
assay. The ORF66 mutants (CR1, 5, 7, 8 and 9mut) failed to associate with 346
ORF34 (Fig. 6) and could not rescue virus production in KSHV-ORF66 cell 347
lines (Fig. 7). In contrast, some ORF66 mutants (CR3 and 6mut), which 348
associate with ORF34, showed full or partial rescue activities, while CR2mut did 349
not. Presumably, the conserved sequence (CLNxG) in the CR2 region is related 350
to binding to other vPIC associated factors. ORF66 rescued activity in iSLK/ 351
66 and iVero/ 66 cell lines, however, recovery rates were lower in iSLK/ 66. 352
The differences in the recovery rates of both cell lines may be due to KSHV 353
production potential and/ or characteristics of each cell line. Efficiency of KSHV 354
production in iSLK is 100-fold higher than iVero. Furthermore, iVero is derived 355
from non-human primates, African green monkey. Slight differences of mutated 356
sites structure and/or species differences of host factors may influence the vPIC 357
formation and accumulation of other host factors, resulting in differences in in 358
the recovery rates. Altogether, association between ORF66 and ORF34 is 359
necessary but not sufficient for virus production. 360
Our results show that the importance and molecular machinery of 361
ORF66 in viral replication and L gene expression, as a vPIC component. 362
Herpesvirus vPICs consist of viral factors as well as RNAPII and several host 363
factors that are engaged in a complex. In MCMV, RNA helicase and cellular 364
factors relating to splicing and translation interact with vPIC (37). Regulation of 365
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
vPIC occurs through post-transcriptional modification of vPIC components, such 366
as phosphorylation, which is known to contribute to physiological functions of 367
vPIC in EBV (38). As the dynamics of vPIC in viral replication, vPIC target 368
promoters on KSHV genome has more complexity than estimated. It 369
inextricably linked to genome DNA replication (27). Our efforts to unveil vPIC 370
machinery help to shed light on why - and -herpesviruses have incorporated 371
vPIC machinery into its genome for survival. 372
373
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
fragments were obtained by PCR from pLKO.1-blast (Addgene plasmid # 382
26655, a kind gift from Dr. Keith Mostov. (39)) using primer sets as noted in 383
Table 1, and were replaced NeoR in pCI-neo mammalian expression vector 384
(Promega, WI, USA). pCI-neo-3xFLAG and pCI-neo-3xFLAG-ORF66 were 385
digested with NheI and NotI sites, and the protein coding fragments were 386
inserted into MCS of pCI-blast to construct of pCI-blast-3xFLAG and pCI-blast-387
3xFLAG-ORF66. 388
389
Mutagenesis of KSHV BAC16. 390
KSHV BAC16 was a kind gift from Jae U. Jung and mutagenesis of KSHV 391
BAC16 was performed according to previous publications (24, 25). The primers 392
for mutagenesis sequences are noted in Table 1. Insertion and deletion of 393
kanamycin resistance cassettes (KanR) in each mutant were analyzed by 394
digestion of EcoRV and agarose-gel electrophoresis. Mutated sites of each 395
BAC clone were confirmed by Sanger sequencing. 396
397
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
3xFLAG-ORF66 stable cell lines, iSLK harboring each KSHV BAC clone was 417
selected and maintained in 1.5 mg/mL of G418. 418
419
420
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
iVero; NaB 1.5 mM/ Dox 8 μg/mL) for 72 hours to induce to lytic replication and 426
production of recombinant KSHV, and culture supernatants were harvested. 427
Culture supernatants (220 μL) were treated with DNase I (NEB, MA, USA) to 428
obtain only enveloped and encapsidated viral genomes. Viral DNA was purified 429
and extracted from 200 μL of DNase I-treated culture supernatant using the 430
QIAamp DNA blood mini kit (QIAGEN, CA, USA). To quantify viral DNA copies, 431
SYBR green real-time PCR was performed using KSHV-encoded ORF11 432
specific primers. 433
For measurement of KSHV genome replication, each KSHV producing 434
cell line was treated with doxycycline and NaB for 48 hours to induce lytic 435
replication, and harvested. Total cellular DNA containing the KSHV genome 436
were purified and extracted from washed cells using the QIAamp DNA blood 437
mini kit (QIAGEN). Cellular KSHV genome copies were determined by SYBR 438
green real-time PCR and normalized to total DNA. 439
440
Recovery of exogenous gene in BAC harboring cells. 441
The iSLK cells were transfected with pCI-neo-3xFLAG as a control plasmid, and 442
3xFLAG-tagged ORF66 full length, truncated mutant, alanine mutant plasmids 443
using Screenfect A plus (Wako, Tokyo, JAPAN) according to the manufacturer’s 444
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
cells were stimulated with NaB 0.5 mM / Dox 8 μg/mL containing medium. After 448
three days of stimulation, viral supernatant was harvested and KSHV genome 449
was evaluated by real-time PCR. 450
451
RT real-time PCR (RT-qPCR) array. 452
mRNA was extracted from iSLK cells treated with Dox and NaB using FastGene 453
RNA Premium Kit (Nippon Genetics Co. Ltd., Tokyo, Japan). cDNA was 454
synthesized by ReverTra Ace RT-qPCR kit (TOYOBO, Osaka, Japan) and 455
subjected to SYBR green real-time PCR. Gene expression was analyzed by 456
qPCR using specific primers designed by Fakhari and Dittmer (26). Relative 457
KSHV mRNA expression levels were determined by GAPDH expression and 458
Ct methods. 459
460
Chromatin Immuno-precipitation (ChIP) assay. 461
ChIP assay was performed as described previously (43) with slight 462
modifications. Briefly, iSLK-WT/ Control and iSLK-ORF66/ 3xFLAG-ORF66 463
cells were treated with or without 4 μg/mL of Dox and 0.75 mM NaB for 72 464
hours. Formaldehyde-fixed cells were lysed by farnham lysis buffer (5 mM 465
PIPES pH 8.0 / 85 mM KCl / 0.5% NP-40) and the nuclear pellet was collected. 466
The pellet was lysed in SDS lysis buffer and sonicated. The supernatant 467
containing DNA was diluted by CHIP dilution buffer and then subjected to 468
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
immunoprecipitation with anti-FLAG (DDDDK-tag) monoclonal antibody (FLA-1; 469
MBL, Nagoya, Japan) or mouse control IgG (Santa-Cruz). Immunoprecipitates 470
containing chromatin and viral DNA were subjected to SYBR green real-time 471
PCR for measuring the amount of promoter DNA of each gene. The amount of 472
immunoprecipitated viral DNA was normalized to 1% of input DNA. The 473
sequences of qPCR primer sets for each transcription start site of ORFs are 474
noted in Table 1. 475
476
Pull-down assay, Western blot, and antibodies. 477
Western blots were performed as described previously (20). For pull-down 478
assays, transfected 293T cells (RCB2202; RIKEN Bio Resource Center, 479
Tsukuba, Japan) were lysed by HNTG buffer (20 mM HEPES (pH 7.9), 0.18 M 480
NaCl, 0.1% NP-40, 0.1 mM EDTA, 10% Glycerol) with protease inhibitors and 481
sonicated. The cell extracts were subjected to affinity purification using S-482
protein immobilized beads (Novagen, MA, USA), and purified proteins 483
(containing 2xS-tagged ORF66 or mutants) were subjected to western blotting. 484
For Zinc chelator TPEN (N,N,N',N'-Tetrakis (2-pyridylmethyl) 485
ethylenediamine; TCI, Tokyo, Japan) treatment, 3xFLAG-taggged ORF66 486
overexpressed in 293T cells, was purified with S-protein immobilized beads in 487
the presence of each dose of TPEN or vehicle (ethanol) for 2 hours. The beads 488
were mixed with cell lysate of 6xMyc-ORF66 overexpressed in 293T in the 489
presence of each dose of TPEN or vehicle (ethanol) for 2 hours. The beads 490
were washed 4 times and subjected to western blotting. 491
Anti-Myc (9E10; Santa-Cruz, CA, USA), Anti-S-tag pAb (MBL, Nagoya, 492
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
were used as the primary antibodies. HRP linked anti-mouse IgG antibody (GE 494
Healthcare UK Ltd., Buckinghamshire, UK) or HRP linked anti-rabbit IgG 495
antibody (GE healthcare UK Ltd.) was used as the secondary antibody. Anti-496
FLAG-HRP (M2; Sigma-aldrich, MO, USA) was also used. 497
498
Homology modeling 499
The template structure for ORF66 was initially identified by collecting high-500
scoring structural templates from Local Meta-Threading Sever to generate a 3D 501
structural model. The protein-ligand predictions were then derived by threading 502
the 3D models through a protein function database, BioLiP. All procedures were 503
automatically processed on I-TASSER server program provided by the Zhang 504
lab (URL: https://zhanglab.ccmb.med.umich.edu/I-TASSER/) (28-32). The 505
visualization of the homology model was performed by molecular visualization 506
open-source software, PyMOL. The hydrophobicity in each amino acid was 507
determined by the Eisenberg hydrophobicity scale 508
(https://web.expasy.org/protscale/pscale/Hphob.Eisenberg.html) (44) and 509
visualized by running a color_h.py script on PyMOL. 510
511
Statistics. 512
The two-tailed student’s t-test was used to indicate the differences between the 513
groups. P values are shown in each figure. 514
515
Acknowledgements 516
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
The BAC16, KSHV BAC clone, was a kind gift from Dr. Kevin Brulois and Dr. 517
Jae U Jung (U.S.C., US). We thank Dr. Gregory A. Smith (Northwestern Univ., 518
US) for the E. coli strain GS1783, and Dr. Nikolaus Osterrieder (Cornell Univ., 519
US) for the plasmid pEP-KanS. We thank Dr. Peter Gee for scientific advice and 520
critical proofreading of the manuscript. This work was supported in-part by a 521
Grant-in-Aid for Scientific Research (C) (18K06642), Young Scientists (B) 522
(16K18925) and Young Scientists (18K14910) from the Ministry of Education, 523
Culture, Sports, Science and Technology of Japan. 524
525
526
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
Fig.1. Construction of recombinantORF66 KSHV BAC 528
(a) Schematic illustration of the KSHV genome including the ORF66 coding 529
region. Using a two-step Red recombination system, three stop codons were 530
inserted into the ORF66 coding region of KSHV BAC16 (nt113417 - nt113416; 531
Accession number: GQ994935) to construct ORF66-deficient BAC clone 532
(ORF66-BAC16). (b) Agarose gel electrophoresis of the recombinant KSHV 533
BACmids, digested with EcoRV. The asterisks (*) indicate insertion and deletion 534
of a kanamycin-resistance cassette in each BAC clone. (c) DNA sequencing 535
results of ORF66 mutagenesis sites in ORF66-BAC16. 536
537
Fig.2. ORF66 is essential for virus production but not DNA replication of 538
KSHV. 539
Virus production in (a) iSLK-WT, iSLK-ORF66 and (e) iVero-WT, iVero-540
ORF66. Each cell line was cultured for 72 h with medium containing NaB and 541
Dox. KSHV DNA was purified from capsidated KSHV virions in culture 542
supernatants, and KSHV genome copies were determined by real-time PCR. 543
Virus production in (b) iSLK-WT, iSLK-ORF66 and (f) iVero-WT, iVero-544
ORF66. Each cell line was cultured for 48 h with medium containing of NaB 545
and Dox. Cellular DNA containing KSHV genomic DNA was purified from each 546
cell line. KSHV genome copies were determined by real-time PCR and 547
normalized by the total DNA amount. Establishment of exogenous ORF66 548
expressing (c) iSLK-ORF66 and (h) iVero-ORF66 stable cell lines. Western 549
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
blot shows exogenous FLAG-tagged ORF66. Rescue of virus production in (d) 550
iSLK-ORF66 and (h) iVero-ORF66 cells by exogenous ORF66 expression. 551
Each stable cell line was cultured with NaB and Dox-containing medium for 3 552
days, and culture supernatant containing virus was harvested and quantified. 553
(a-b, d-f, h) Three or four independent samples were evaluated by real-time 554
PCR. The error bars indicate standard deviations. 555
556
Fig.3. ORF66 is required for the late gene expression 557
iSLK-WT and iSLK-ORF66 cells were cultured for 72h in media with NaB and 558
Dox to induce a lytic state. Total RNA was extracted from cells and subjected to 559
RT-qPCR. The mRNA expression level of each viral gene was normalized by 560
GAPDH expression, and columns indicated fold changes of iSLK-ORF66 561
transcripts compared with iSLK-WT transcripts. Classification of KSHV genes 562
was performed according to Arias C. et al. (45). White, light gray, dark gray, 563
black, and light blue columns indicate Latent, Immeidiate early, Early, Late, and 564
Not classified genes, respectively. Expression levels were assessed using three 565
independent samples, and error bars indicate ± standard deviations. 566
567
Fig.4. ORF66 associates with the transcriptional start site of Late genes 568
iSLK-WT/ Control and iSLK-ORF66/ 3xFLAG-ORF66 cells were treated with 569
(or without) Dox and NaB for 72 hours and subjected to ChIP-qPCR. 3xFLAG-570
ORF66 protein was immunoprecipitated by anti-FLAG or control IgG antibody, 571
and precipitates including chromatin and viral DNA were subjected to SYBR 572
green real-time PCR for measuring the amount of promoter DNA of ORF72 573
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
(Latent gene), ORF16 (IE gene), ORF46/47 (E gene) or K8.1 (L gene). The 574
levels of immunoprecipitated viral promoter were normalized to total input DNA. 575
576
Fig.5. ORF66 physically interacts with ORF34 via its C-terminus regions 577
and the entire structure of ORF66 is indispensable for virus production. 578
(a) Schematic representation of tagged ORF34 deletion mutants used in the 579
mapping experiments. The truncated amino acids are numbered next to its 580
corresponding mutant. (b) 293T cells were co-transfected with expression 581
plasmids of 2xS-ORF66 truncated mutant and 6xMyc-ORF34. Transfected cells 582
were lysed, and cell lysates were subjected to pull-down assays using S-583
protein-immobilized beads that capture the 2xS-ORF66. Obtained precipitates 584
including 2xS-ORF66 truncated mutants were probed with indicated antibodies 585
to detect interactions. 586
The (c) iSLK-ORF66 or (d) iVero-ORF66 cells were transfected with control, 587
ORF66 or ORF66-truncated mutant plasmid. Transfected cells were stimulated 588
for 3 days. Progeny KSHV was purified from harvested culture supernatant, and 589
the KSHV genome was quantified by real-time PCR. Viral productivity was 590
assessed using four independent samples, and error bars indicate standard 591
deviations. Transfected virus producing cells were lysed and subjected to 592
Western blotting to confirm ORF66 mutant expression. 593
594
Fig.6. Several conserved residues of ORF66 are essential for physical 595
association with ORF34. 596
(a) Amino acid sequence alignment of C-terminus of ORF66 (241 a.a.- 429 597
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
(b) Block alanine scanning mutants were co-transfected with expression 613
plasmids of 6xMyc-ORF34. Cell lysates were subjected to pull-down assays 614
using S-protein immobilized beads. 615
616
Fig.7. Association between ORF66 and ORF34 is necessary but not 617
sufficient for virus production 618
The (a) iSLK-ORF66 or (b) iVero-ORF66 cells were transfected with control, 619
ORF66 or ORF66 block alanine scanning mutant plasmid. Progeny KSHV was 620
purified and the KSHV genome was quantified. Viral productivities were 621
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
influences ORF66 binding abilities to ORF34. Pull-down assay using S-tagged 641
ORF66 binding beads in the presence of Zinc chelator TPEN. 642
643
Fig.9. The protein structure and functional prediction of ORF66. 644
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
The structural models of ORF66 simulated by I-TASSER, the server completed 645
protein structural and functional prediction, are shown. The cartoon model 646
indicates the position of responsible residues for interaction with ORF34 (left 647
panel). The surface model represents the exposed residues on the protein 648
surface (middle panel). The hydrophobicity of each residue is shown by a red 649
color gradient followed by the Eisenberg hydrophobicity scale (right panel). The 650
predicted Zinc ion-binding cysteins (C295 and C298, C341 and C344) are 651
highlighted with brown. C344 in CR5 is highlighted in magenta. The LLQL in 652
CR8 and C341 in CR5 are shown in cyan and green, respectively. 653
654
References 655
1. Nador RG, Cesarman E, Chadburn A, Dawson DB, Ansari MQ, Sald J, 656 Knowles DM. 1996. Primary effusion lymphoma: a distinct 657 clinicopathologic entity associated with the Kaposi's sarcoma-associated 658 herpes virus. Blood 88:645-56. 659
2. Soulier J, Grollet L, Oksenhendler E, Cacoub P, Cazals-Hatem D, 660 Babinet P, d'Agay MF, Clauvel JP, Raphael M, Degos L, et al. 1995. 661 Kaposi's sarcoma-associated herpesvirus-like DNA sequences in 662 multicentric Castleman's disease. Blood 86:1276-80. 663
3. Russo JJ, Bohenzky RA, Chien MC, Chen J, Yan M, Maddalena D, Parry 664 JP, Peruzzi D, Edelman IS, Chang Y, Moore PS. 1996. Nucleotide 665 sequence of the Kaposi sarcoma-associated herpesvirus (HHV8). Proc 666 Natl Acad Sci U S A 93:14862-7. 667
4. Chang Y, Cesarman E, Pessin MS, Lee F, Culpepper J, Knowles DM, 668 Moore PS. 1994. Identification of herpesvirus-like DNA sequences in 669 AIDS-associated Kaposi's sarcoma. Science 266:1865-9. 670
5. Cesarman E. 2014. Gammaherpesviruses and lymphoproliferative 671 disorders. Annu Rev Pathol 9:349-72. 672
6. Toth Z, Brulois K, Jung JU. 2013. The chromatin landscape of Kaposi's 673 sarcoma-associated herpesvirus. Viruses 5:1346-73. 674
7. Uppal T, Jha HC, Verma SC, Robertson ES. 2015. Chromatinization of 675 the KSHV Genome During the KSHV Life Cycle. Cancers (Basel) 7:112-676 42. 677
8. Ballestas ME, Chatis PA, Kaye KM. 1999. Efficient persistence of 678 extrachromosomal KSHV DNA mediated by latency-associated nuclear 679 antigen. Science 284:641-4. 680
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
9. Fujimuro M, Wu FY, ApRhys C, Kajumbula H, Young DB, Hayward GS, 681 Hayward SD. 2003. A novel viral mechanism for dysregulation of beta-682 catenin in Kaposi's sarcoma-associated herpesvirus latency. Nat Med 683 9:300-6. 684
10. Ashizawa A, Higashi C, Masuda K, Ohga R, Taira T, Fujimuro M. 2012. 685 The Ubiquitin System and Kaposi's Sarcoma-Associated Herpesvirus. 686 Front Microbiol 3:66. 687
11. Mesri EA, Cesarman E, Boshoff C. 2010. Kaposi's sarcoma and its 688 associated herpesvirus. Nat Rev Cancer 10:707-19. 689
12. Jenner RG, Alba MM, Boshoff C, Kellam P. 2001. Kaposi's sarcoma-690 associated herpesvirus latent and lytic gene expression as revealed by 691 DNA arrays. J Virol 75:891-902. 692
13. Guito J, Lukac DM. 2012. KSHV Rta Promoter Specification and Viral 693 Reactivation. Front Microbiol 3:30. 694
14. Sun R, Lin SF, Gradoville L, Yuan Y, Zhu F, Miller G. 1998. A viral gene 695 that activates lytic cycle expression of Kaposi's sarcoma-associated 696 herpesvirus. Proc Natl Acad Sci U S A 95:10866-71. 697
15. Damania B, Cesarman E. 2013. Kaposi’s Sarcoma–Associated 698 Herpesvirus, p 2080-2128. In Knipe DM, Howley PM (ed), Fields 699 Virology, 6th Edition, vol 2. Lippincott Williams & Wilkins. 700
16. Gruffat H, Marchione R, Manet E. 2016. Herpesvirus Late Gene 701 Expression: A Viral-Specific Pre-initiation Complex Is Key. Front Microbiol 702 7:869-869. 703
17. Gruffat H, Marchione R, Manet E. 2016. Herpesvirus Late Gene 704 Expression: A Viral-Specific Pre-initiation Complex Is Key. Front Microbiol 705 7. 706
18. Davis ZH, Hesser C, Park J, Glaunsinger BA. 2015. Interaction Between 707 ORF24 and ORF34 in the Kaposi's Sarcoma-Associated Herpesvirus 708 Late Gene Transcription Factor Complex is Essential For Viral Late Gene 709 Expression. J Virol doi:10.1128/jvi.02157-15. 710
19. Davis Zoe H, Verschueren E, Jang Gwendolyn M, Kleffman K, Johnson 711 Jeffrey R, Park J, Von Dollen J, Maher MC, Johnson T, Newton W, Jäger 712 S, Shales M, Horner J, Hernandez Ryan D, Krogan Nevan J, 713 Glaunsinger Britt A. 2015. Global Mapping of Herpesvirus-Host Protein 714 Complexes Reveals a Transcription Strategy for Late Genes. Mol Cell 715 57:349-360. 716
20. Nishimura M, Watanabe T, Yagi S, Yamanaka T, Fujimuro M. 2017. 717 Kaposi's sarcoma-associated herpesvirus ORF34 is essential for late 718 gene expression and virus production. Sci Rep 7:329. 719
21. Brulois K, Wong LY, Lee HR, Sivadas P, Ensser A, Feng P, Gao SJ, Toth 720 Z, Jung JU. 2015. Association of Kaposi's Sarcoma-Associated 721 Herpesvirus ORF31 with ORF34 and ORF24 Is Critical for Late Gene 722 Expression. J Virol 89:6148-54. 723
22. Castañeda AF, Glaunsinger BA. 2019. The Interaction between ORF18 724 and ORF30 Is Required for Late Gene Expression in Kaposi's Sarcoma-725 Associated Herpesvirus. J Virol 93:e01488-18. 726
23. Gong D, Wu NC, Xie Y, Feng J, Tong L, Brulois KF, Luan H, Du Y, Jung 727
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
JU, Wang CY, Kang MK, Park NH, Sun R, Wu TT. 2014. Kaposi's 728 sarcoma-associated herpesvirus ORF18 and ORF30 are essential for 729 late gene expression during lytic replication. J Virol 88:11369-82. 730
24. Tischer BK, Smith GA, Osterrieder N. 2010. En passant mutagenesis: a 731 two step markerless red recombination system. Methods Mol Biol 732 634:421-30. 733
25. Brulois KF, Chang H, Lee AS, Ensser A, Wong LY, Toth Z, Lee SH, Lee 734 HR, Myoung J, Ganem D, Oh TK, Kim JF, Gao SJ, Jung JU. 2012. 735 Construction and manipulation of a new Kaposi's sarcoma-associated 736 herpesvirus bacterial artificial chromosome clone. J Virol 86:9708-20. 737
27. Nandakumar D, Glaunsinger B. 2019. An integrative approach identifies 741 direct targets of the late viral transcription complex and an expanded 742 promoter recognition motif in Kaposi's sarcoma-associated herpesvirus. 743 PLoS Pathog 15:e1007774. 744
28. Zhang Y. 2008. I-TASSER server for protein 3D structure prediction. 745 BMC Bioinformatics 9:40. 746
29. Roy A, Kucukural A, Zhang Y. 2010. I-TASSER: a unified platform for 747 automated protein structure and function prediction. Nat Protoc 5:725-38. 748
30. Yang J, Yan R, Roy A, Xu D, Poisson J, Zhang Y. 2015. The I-TASSER 749 Suite: protein structure and function prediction. Nat Methods 12:7-8. 750
31. Yang J, Roy A, Zhang Y. 2013. Protein-ligand binding site recognition 751 using complementary binding-specific substructure comparison and 752 sequence profile alignment. Bioinformatics 29:2588-95. 753
32. Yang J, Roy A, Zhang Y. 2013. BioLiP: a semi-manually curated 754 database for biologically relevant ligand-protein interactions. Nucleic 755 Acids Res 41:D1096-103. 756
33. Aubry V, Mure F, Mariame B, Deschamps T, Wyrwicz LS, Manet E, 757 Gruffat H. 2014. Epstein-Barr virus late gene transcription depends on 758 the assembly of a virus-specific preinitiation complex. J Virol 88:12825-759 38. 760
34. Dunn W, Chou C, Li H, Hai R, Patterson D, Stolc V, Zhu H, Liu F. 2003. 761 Functional profiling of a human cytomegalovirus genome. Proc Natl Acad 762 Sci U S A 100:14223-14228. 763
35. Zhang W, Li H, Li Y, Zeng Z, Li S, Zhang X, Zou Y, Zhou T. 2010. 764 Effective inhibition of HCMV UL49 gene expression and viral replication 765 by oligonucleotide external guide sequences and RNase P. Virol J 7:100. 766
36. Murakami K, Tsai K-L, Kalisman N, Bushnell DA, Asturias FJ, Kornberg 767 RD. 2015. Structure of an RNA polymerase II preinitiation complex. 768 Proceedings of the National Academy of Sciences 112:13543-13548. 769
37. Chapa TJ, Du Y, Sun R, Yu D, French AR. 2017. Proteomic and 770 phylogenetic coevolution analyses of pM79 and pM92 identify 771 interactions with RNA polymerase II and delineate the murine 772 cytomegalovirus late transcription complex. The Journal of general 773 virology 98:242-250. 774
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
38. Sato Y, Watanabe T, Suzuki C, Abe Y, Masud H, Inagaki T, Yoshida M, 775 Suzuki T, Goshima F, Adachi J, Tomonaga T, Murata T, Kimura H. 2019. 776 S-Like-Phase Cyclin-Dependent Kinases Stabilize the Epstein-Barr Virus 777 BDLF4 Protein To Temporally Control Late Gene Transcription. J Virol 93. 778
39. Bryant DM, Datta A, Rodriguez-Fraticelli AE, Peranen J, Martin-Belmonte 779 F, Mostov KE. 2010. A molecular network for de novo generation of the 780 apical surface and lumen. Nat Cell Biol 12:1035-45. 781
41. Watanabe T, Nakamura S, Ono T, Ui S, Yagi S, Kagawa H, Watanabe H, 786 Ohe T, Mashino T, Fujimuro M. 2014. Pyrrolidinium fullerene induces 787 apoptosis by activation of procaspase-9 via suppression of Akt in primary 788 effusion lymphoma. Biochem Biophys Res Commun 451:93-100. 789
42. Katoh Y, Nozaki S, Hartanto D, Miyano R, Nakayama K. 2015. 790 Architectures of multisubunit complexes revealed by a visible 791 immunoprecipitation assay using fluorescent fusion proteins. J Cell Sci 792 128:2351-62. 793
43. Agata Y, Katakai T, Ye SK, Sugai M, Gonda H, Honjo T, Ikuta K, Shimizu 794 A. 2001. Histone acetylation determines the developmentally regulated 795 accessibility for T cell receptor gamma gene recombination. J Exp Med 796 193:873-80. 797
44. Eisenberg D, Schwarz E, Komaromy M, Wall R. 1984. Analysis of 798 membrane and surface protein sequences with the hydrophobic moment 799 plot. J Mol Biol 179:125-42. 800
45. Arias C, Weisburd B, Stern-Ginossar N, Mercier A, Madrid AS, Bellare P, 801 Holdorf M, Weissman JS, Ganem D. 2014. KSHV 2.0: a comprehensive 802 annotation of the Kaposi's sarcoma-associated herpesvirus genome 803 using next-generation sequencing reveals novel genomic and functional 804 features. PLoS Pathog 10:e1003847. 805
806
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 7, 2019. . https://doi.org/10.1101/728147doi: bioRxiv preprint