Characterization of a live-attenuated HCMV-based vaccine platform

Caposio, Patrizia; van den Worm, Sjoerd; Crawford, Lindsey; Perez, Wilma; Kreklywich, Craig; Gilbride, Roxanne M.; Hughes, Colette M.; Ventura, Abigail B.; Ratts, Robert; Marshall, Emily E.; Malouli, Daniel; Axthelm, Michael K.; Streblow, Daniel; Nelson, Jay A.; Picker, Louis J.; Hansen, Scott G.; Früh, Klaus

doi:10.1038/s41598-019-55508-w

Download PDF

Article
Open access
Published: 17 December 2019

Characterization of a live-attenuated HCMV-based vaccine platform

Patrizia Caposio¹^na1,
Sjoerd van den Worm¹^na1^nAff2,
Lindsey Crawford ORCID: orcid.org/0000-0003-1248-253X¹,
Wilma Perez¹,
Craig Kreklywich¹,
Roxanne M. Gilbride¹,
Colette M. Hughes¹,
Abigail B. Ventura¹,
Robert Ratts¹^nAff3,
Emily E. Marshall¹^nAff3,
Daniel Malouli ORCID: orcid.org/0000-0003-1287-7283¹,
Michael K. Axthelm¹,
Daniel Streblow¹,
Jay A. Nelson¹,
Louis J. Picker¹,
Scott G. Hansen¹ &
…
Klaus Früh ORCID: orcid.org/0000-0001-8014-3877¹

Scientific Reports volume 9, Article number: 19236 (2019) Cite this article

3603 Accesses
17 Citations
24 Altmetric
Metrics details

Subjects

Abstract

Vaccines based on cytomegalovirus (CMV) demonstrate protection in animal models of infectious disease and cancer. Vaccine efficacy is associated with the ability of CMV to elicit and indefinitely maintain high frequencies of circulating effector memory T cells (T_EM) providing continuous, life-long anti-pathogen immune activity. To allow for the clinical testing of human CMV (HCMV)-based vaccines we constructed and characterized as a vector backbone the recombinant molecular clone TR3 representing a wildtype genome. We demonstrate that TR3 can be stably propagated in vitro and that, despite species incompatibility, recombinant TR3 vectors elicit high frequencies of T_EM to inserted antigens in rhesus macaques (RM). Live-attenuated versions of TR3 were generated by deleting viral genes required to counteract intrinsic and innate immune responses. In addition, we eliminated subunits of a viral pentameric glycoprotein complex thus limiting cell tropism. We show in a humanized mouse model that such modified vectors were able to establish persistent infection but lost their ability to reactivate from latency. Nevertheless, attenuated TR3 vectors preserved the ability to elicit and maintain T_EM to inserted antigens in RM. We further demonstrate that attenuated TR3 can be grown in approved cell lines upon elimination of an anti-viral host factor using small interfering RNA, thus obviating the need for a complementing cell line. In sum, we have established a versatile platform for the clinical development of live attenuated HCMV-vectored vaccines and immunotherapies.

Highly stable and immunogenic CMV T cell vaccine candidate developed using a synthetic MVA platform

Article Open access 30 March 2024

MCMV-based vaccine vectors expressing full-length viral proteins provide long-term humoral immune protection upon a single-shot vaccination

Article Open access 07 January 2022

Optimising T cell (re)boosting strategies for adenoviral and modified vaccinia Ankara vaccine regimens in humans

Article Open access 12 October 2020

Introduction

Vaccines using cytomegalovirus (CMV) as antigen-carrying vectors have shown great promise in animal models of infectious diseases and cancer^1,2,3. Using rhesus CMV (RhCMV) as our model system we demonstrated unprecedented control of infection by highly pathogenic SIV upon repeated low dose challenge^4,5. Moreover, we recently also reported that RhCMV-vectors expressing antigens derived from Mycobacterium tuberculosis (TB) protected against intrabronchial challenge with TB to which RM are exquisitely susceptible⁶. Finally, we demonstrated that RhCMV-based vaccines eliciting T cells against antigens of the malaria parasite Plasmodium knowlesi strongly reduced the release of liver stage parasites into the blood⁷. Taken together these studies demonstrate that CMV-vectors represent a novel vaccine platform for many applications.

Since RhCMV-based vectors elicit little to no antibody responses to the inserted antigens, the protection elicited by these vectors is almost certainly attributable to cellular immunity^4,6,7. Indeed, one of the most unique aspects of RhCMV-based vectors is their ability to elicit and indefinitely maintain high frequencies of circulating and tissue-resident effector memory CD4+ and CD8+ T cells (T_EM) to the inserted antigens^4,5. The likely mechanism of T cell mediated protection was illustrated in the SIV model where 50% of RhCMV/SIV vaccinated animals were initially infected with SIV, as documented by cell-associated, replication-competent SIV and/or by the development of T cell responses to SIV antigens not included in the vaccine. However, animals remained aviremic and went on to eventually clear the SIV infection to below detection limits of all available virological measurements⁵. A very similar result was obtained when anti-retroviral treatment was started within 4–5 days of SIV challenge strongly suggesting that RhCMV/SIV elicited T cell immunity provided an early intercept of SIV infection that prevents the seeding of a long-lived latent SIV reservoir⁸. Thus, CMV-elicited T_EM provide a rapid interception and control of pathogens at the portal of pathogen entry and maintain control over time.

Since T cell effector differentiation is antigen-driven, it is likely that CMV-induced T_EM are maintained by continuous or recurring antigen exposure due to viral persistence and reactivation in antigen presenting cells (APC)⁹. Surprisingly however, this immune stimulation does not seem to require viral dissemination within the host as long as latency is established. In murine models it was shown previously that MCMV deleted for essential viral genes was still able to elicit and maintain T_EM despite being spread-deficient^10,11. More recently, we demonstrated that RhCMV lacking the tegument protein pp71 is highly debilitated in its ability to spread and was no longer transmitted either through secretions or by blood transfusions¹². Nevertheless, above a given dose threshold, pp71-deleted RhCMV elicited immune responses that retained all features described above¹². Moreover, pp71-deleted RhCMV/SIV vaccines protected against homologous and heterologous challenge with SIV and most of the protected animals were able to control SIV infection again when re-challenged years later¹³.

CMV species co-evolved with their individual host species and no naturally occurring instances of cross species infections have been observed¹⁴. Thus, CMV vectors have to be based on a HCMV vector backbone to maintain the desired immunological features of CMV-based vectors for human vaccines and immunotherapies. Since disseminating HCMV can cause serious disease in individuals with an immature or compromised immune system¹⁵, HCMV-based vaccine vectors intended for general prophylactic use in human need to be attenuated. Spread-deficient animal CMV species that maintain all unique T cell immunity features thus provide a blueprint for the design of highly attenuated HCMV-vectors for human use. To permit the genetic modifications required to insert heterologous antigens as well as safety features the selected HCMV strain needs to be amenable for genetic manipulation while maintaining genetic stability and manufacturability. Here we describe the novel HCMV-based vaccine platform TR3 that, starting from a complete viral genome representative of the low passage isolate HCMV TR¹⁶, can be genetically modified to introduce heterologous antigens as well as specific deletions that impact vector safety and immunogenicity. Using a humanized mouse model system we demonstrate that deletion or inactivation of the pp71-encoding gene UL82 renders TR3 reactivation-deficient while maintaining the ability to establish latency. We further demonstrate that TR3 carrying heterologous antigens can elicit and maintain T_EM to these antigens in RM even when deleted for pp71. We further show that pentameric complex (PC)-deleted TR3 maintains vector function in vivo either alone or in combination with pp71 deletion. Thus, we developed a robust and versatile live attenuated vaccine vector platform that is suitable for clinical testing.

Results

Construction and characterization of the HCMV-TR3 vector backbone

To translate the RhCMV-vector results into clinical testing we selected an HCMV strain that was likely to establish persistent infection since we posit that viral persistence is required to elicit and maintain potent T_EM-responses. At the same time the vector needed to be amenable to the genetic manipulations essential for the insertion of heterologous antigens as well as the introduction of safety and immunogenicity features. Low passage HCMV strains have the highest likelihood of maintaining the ability of the parental isolate to persist upon infection. However, many clinical isolates are genetically unstable upon prolonged passage in tissue culture rendering them unsuitable as a vaccine platform¹⁷. We previously described the clinical isolate HCMV TR as being representative for low passage isolates while lacking typical genetic adaptations to tissue culture¹⁶. To evaluate the suitability of HCMV TR as a vaccine vector backbone we modified a previously described recombinant that was cloned as a bacterial artificial chromosome (BAC) which facilitates genetic manipulation¹⁶. In this original clone (TR-BAC) a non-excisable BAC-cassette containing the bacterial origin of replication and a bacterial resistance marker was inserted into the US2-6 genomic region thus deleting parts of US2 and US6, and all of US3, US4 and US5¹⁶ (Fig. 1A). To restore expression of the US2-6 genes, Lauron et al. inserted the US2-US7 region from HCMV strain AD169 as well as loxP sites to allow the excision of the BAC cassette upon Cre-expression in mammalian cells. In addition, a GFP-expression cassette was inserted downstream of US8, thereby deleting the TR US7 gene¹⁸ (Fig. 1A). Since the US2-11 homologous genomic region in RhCMV was shown to be important for superinfection¹⁹ we used this US2-7-repaired version (TR-GFP) for further modifications. Using the galK BAC recombineering system²⁰ we deleted the GFP marker and inserted the Cre recombinase gene into the BAC cassette which renders the cassette self-excising upon viral reconstitution²¹. Since HCMV TR was resistant to Ganciclovir (GCV) due to mutations in the viral kinase UL97 that is required to phosphorylate the pro-drug we replaced the gene with UL97 from AD169 which is GCV sensitive (Fig. 1A). The final BAC, termed TR3, was sequenced by next generation sequencing (NGS) (Fig. S1). Upon reconstitution in human fibroblasts, viral genomes were also sequenced by NGS which confirmed the excision of the BAC cassette as well as the presence of US2-7 region of AD169, a loxP site between US7 and US8, and the UL97 gene from AD169 (Fig. S2). To demonstrate that insertion of AD169 UL97 conferred GCV sensitivity to TR3, we performed a plaque reduction assay in fibroblasts in the presence of increasing GCV concentrations (Fig. 1B). We next verified expression of US6, an inhibitor of the peptide transporter associated with antigen presentation²², that is encoded in the AD169-derived US2-7 region (Fig. 1C). Expression levels were comparable with the US6 expression observed in the original clinical isolate TR whereas US6 is absent from the original BAC-cloned TR-BAC. Taken together these data demonstrate that we successfully reconstructed a vector backbone with a genome configuration that is representative of low passage HCMV clinical isolates. A single loxP site is the only remaining heterologous sequence upon viral reconstitution in vitro (Figs. S1, S2).

Generation of a live-attenuated HCMV vector platform by deletion or inactivation of UL82 (pp71)

Given the possibility of HCMV to cause disease in individuals with an immature or compromised immune system, HCMV-based vectors need to be attenuated in order to be safe for use as a vaccine platform. We recently reported that RhCMV lacking the viral tegument protein pp71 was highly attenuated in vivo, as evident from reduced dissemination, lack of shedding as well as lack of natural transmission or transmission upon transfusion¹². Nevertheless, above the effective dose threshold, pp71-deleted RhCMV elicited and maintained T_EM responses that were similar to that of the parental strain and these immune responses were able to confer long-term protection against homologous and heterologous challenge with highly pathogenic SIV_mac239¹³. Based on these results, we selected deletion of pp71 as our primary attenuation strategy. While pp71-deleted RhCMV can be recovered upon transfection of BAC into fibroblasts without the need of complementation¹², it was previously reported that pp71-deleted HCMV could only be recovered in fibroblasts expressing exogenous pp71²³. A major function of pp71 is to counteract the host repressor DAXX (Death-domain associated protein 6) which otherwise limits viral immediate early (IE) gene expression^24,25,26. Since IE expression can be restored in cells infected with pp71-deleted HCMV treated with small interfering RNA (siRNA) targeting DAXX²⁵ we hypothesized that it is possible to recover and grow UL82-deleted TR3 in the presence of siRNA targeting DAXX. This alternative strategy would obviate the need for a complementing cell line. To generate a UL82-deleted HCMV we replaced the pp71-encoding ORF with that of HIVgag (Fig. 1D) a strategy we have used in the past to express SIV antigens in RhCMV^12,13. In the pp71-intact control vector we replaced UL78, a viral G protein coupled receptor not required for growth in fibroblasts²⁷, with HIVgag thus using the UL78 promoter to drive HIVgag expression (Fig. 1D). We previously reported the induction of SIV-antigen specific T cell responses by RhCMV expressing an SIV antigen in the UL78-homologous gene of RhCMV¹².

The sequences of the BACs for TR3ΔUL78gag and TR3ΔUL82gag were confirmed by NGS (Fig. S1). We were able to recover virus when siRNA targeting DAXX was transfected into fibroblasts together with the BAC for TR3ΔUL82gag (Fig. 1E). To monitor the impact of pp71 deletion on viral protein expression we monitored viral protein expression as well as the expression of HIVgag by immunoblotting of cell lysates obtained 6 days post infection (dpi) with each individual recombinant virus. In cells infected with TR3ΔUL82gag in the absence of DAXX siRNA we did not detect pp71 expression or the late protein pp28 whereas low amounts of IE72 and the E protein UL44 were observed (Fig. 1E). In contrast, treatment with DAXX siRNA resulted in protein expression levels comparable to pp71-intact TR3. Importantly, HIVgag expression was observed in the absence of pp71 and even when expression was driven by the UL82 promoter although levels were lower compared to pp71-expressing controls. These data indicate that HIVgag might continue to be expressed at a low level in vivo even when viral replication is severely attenuated in the absence of DAXX siRNA.

In vitro growth of live-attenuated TR3

To further characterize the growth defect of pp71-deficient TR3 in different cell lines we needed a cell line that allows for the accurate titration of UL82-deleted HCMV. To this end, we introduced a UL82 expression vector into human foreskin BJ fibroblasts that are life-extended by telomerase expression²⁸ (see Methods). To accurately determine the titers of pp71-deficient virus stocks or culture supernatants we monitored the expression of viral IE proteins in BJ-pp71 cells by immunocytochemistry. Using this assay we thus compared in vitro growth of pp71-intact and pp71-deficient TR3 at different MOIs upon infection of MRC-5 cells. As shown in Fig. 2A transfection of the MRC-5 cells with DAXX siRNA resulted in final titers, determined as focus forming units (FFU), of TR3ΔUL82gag that were comparable to pp71-intact TR3ΔUL78gag even at a very low MOI = 0.001. As reported previously for AD169, deletion of pp71 from TR3 did not impact viral growth at high MOI²³. The impact of pp71 deletion on TR3 growth was further investigated in endothelial, epithelial, and astrocytic cells. HUVEC, ARPE-19, and CCF-STTG1 cells were infected at high and low MOI with TR3 or TR3ΔUL82gag in the absence of DAXX siRNA. In all three cell types we observed a growth defect even at high MOI = 1 and no viral growth at MOI = 0.1 (Fig. 2B). Taken together these data demonstrate that the absence of pp71 profoundly affects growth of TR3 in vitro consistent with previously reported growth defects of AD169.

Stability of live-attenuated HCMV TR3 upon repeated passaging in vitro

Most clones of low passage HCMV isolates harbouring complete viral genomes rapidly mutate in vitro, including point mutations, gene deletions and rearrangements of sub-genomic regions²⁹. Such tissue culture adaptations have rendered it difficult if not impossible to use complete viral genomes as backbones for vectors or vaccines. This is particularly true when growing clinical isolates in human fibroblasts where it was shown that upon transfection of a full-length BAC clone of the clinical isolate Merlin the viral progeny was rapidly selected for mutations in the gene RL13 followed by loss of the UL128, UL130 or UL131 subunits of the pentameric complex (PC)¹⁷. Similarly, we observed that the PC was rapidly lost upon reconstitution of a BAC-cloned version of the clinical isolate Toledo in fibroblasts³⁰. The PC, consisting of the viral proteins gH/gL/UL128/UL130/UL131, facilitates entry into non-fibroblast cells such as endothelial cells, epithelial cells and myeloid cells by endocytosis^31,32. Host cell proteins that are specifically expressed in these cells, including neuropilin-2, ORF14I1 and CD147, have been identified in playing a crucial role in PC-mediated entry^33,34,35.

To evaluate the genetic integrity of our live attenuated HCMV platform we monitored the viral genomic sequences of TR3ΔUL82gag over a series of 20 passages in MRC-5 fibroblasts. MRC-5 cells were infected at each passage with TR3ΔUL82gag in the presence of DAXX siRNA at an MOI = 0.01. After every 5 passages cells and supernatants were harvested and analyzed by immunoblot whereas passages 6, 9 and 20 were analyzed by NGS. Immunoblot analysis revealed stable expression of HIVgag over all passages (Fig. 3A). Genomic sequence analysis indicated that each passage increased the number of single nucleotide polymorphisms (SNPs) which is expected from a replicating viral genome resulting in random, non-debilitating mutations. However, most of these SNPs remained at low frequency with the exception of two SNPs in glycoprotein B (gB, UL55) and in nucleotide reductase (UL45) which increased in frequency in later passages (Fig. 3B). Similarly, we observed an enrichment of a different SNP in gB when unmodified TR3 was passaged 20 times (Fig. 3C) The impact of these SNPs on gB function has not been evaluated, but since gB is an essential gene these mutations are unlikely to be detrimental for viral growth. Interestingly, it was recently reported that the gB of the highly passaged AD169 strain contains a SNP (D275Y) that seems to accelerate viral entry into fibroblasts by membrane fusion³⁶. Conceivably, the SNPs observed in TR3 gB might thus reflect a similar adaptation to fibroblast growth. However, TR3 clearly did not seem to undergo the same selection pattern described for other clinical isolates that result in the rapid selection of RL13 and PC mutations upon passaging in fibroblasts. Moreover, we did not observe any deletions or any major genomic rearrangements despite extensive passaging. Based on these results we conclude that TR3 is a genetically stable vector platform that retains the genomic configuration of a low passage clinical isolate.

Pp71-deficient HCMV-TR3 establishes a persistent infection but does not reactivate in humanized mice

We previously demonstrated that HCMV TR and other clinical isolates can establish persistent infection in a humanized mouse model in which human CD34+ hematopoietic precursor cells (HPC) are engrafted into NOD-scidIL2Rγc null (NSG) mice^37,38,39,40. Moreover, mobilization of human monocyte/macrophages by granulocyte-colony stimulating factor (G-CSF) reactivates virus in peripheral organs³⁹. This model thus represented an opportunity to evaluate the attenuation of pp71-deficient TR3 in vivo. HuNSG mice (n = 10/group) were inoculated intraperitoneally with human fibroblasts infected with TR3, or TR3ΔUL82gag and, after 8 weeks, 5 animals in each group were treated with G-CSF. One week later, animals were sacrificed and viral genome copy numbers were determined in spleen and liver by quantitative PCR. As reported previously for the parental TR strain³⁹, latent TR3 was detected in spleen and liver and viral genome copy numbers increased upon G-CSF treatment consistent with reactivation from latency (Fig. 4A). In contrast, while latent genomes for pp71-deficient TR3 vector were measured in the same range as pp71-intact TR3, G-CSF treatment did not increase genome copy numbers suggesting that virus was unable to reactivate in the absence of pp71. Different from pp71-deficient TR3 we observed that the highly passaged laboratory strain AD169 was not able to establish latent infection even when containing an intact PC (Fig. 4B) suggesting that other genes deleted or mutated in AD169 are required for the establishment and/or maintenance of latent infection. We therefore interpret our results as evidence that pp71-deleted TR3 maintains the ability to establish latency but is limited in its ability to reactivate. This interpretation is consistent with observations reported for pp71-deficient HCMV in in vitro latency models^41,42.

Live-attenuated HCMV TR3 elicits effector memory T cell responses in non-human primates

We used humanized mice to monitor viral latency and reactivation in vivo since HCMV replication is highly species-specific. Therefore, animal models generally rely on studying the CMV of the respective animal species, e.g. RhCMV in rhesus macaques (RM) or MCMV in mice. However, it has been reported that a conditionally replicating, pentamer-intact AD169 (termed V160), elicited HCMV–specific T cell and antibody responses in RM^43,44. Conditional replication was achieved by fusion of a degradation domain derived from the FK506-binding protein (FKBP) to the essential viral gene products IE1/2 and UL51 so that viral replication occurs only in the presence of a stabilizing ligand for FKBP. We previously reported that replication competent (pentamer-deficient) AD169 can infect RM fibroblasts, but does not produce viral progeny⁴⁵. Similarly, we observed that HCMV TR3ΔUL78gag was able to infect RM fibroblasts resulting in low level viral and HIVgag protein expression, but did not generate infectious progeny (Fig. S3). These results suggest that even replication competent HCMV will be unable to disseminate in RM suggesting that RM cannot be used to monitor pathogenesis and attenuation of HCMV.

However, despite this species-specificity, V160 was reported to elicit T cell responses to IE1/2, but not to the FKBP moiety⁴⁴. Therefore, we sought to determine whether TR3-based vectors would be able to elicit T cell responses to inserted heterologous antigens in RM. Two RM were subcutaneously inoculated with 5 × 10⁶ FFU of TR3ΔUL78gag and monitored the CD4+ and CD8+ memory T cell response in peripheral blood mononuclear cells (PBMC) to HCMV lysate or to pooled overlapping 15mer peptides covering the entire HIVgag sequence by intracellular cytokine staining for TNFα and IFNγ. As shown in Fig. 5A, both RM elicited high frequencies of both CD4+ and CD8+ T cell responses to HCMV and HIVgag. Of note, there was no detectable HCMV-specific T cell response prior to inoculation despite the fact that all animals are naturally infected with RhCMV suggesting little or no shared T cell epitope sequences. Next, we determined whether pp71-deficient vectors would elicit HCMV and HIVgag-specific T cell responses and whether this ability would be vector dose dependent by inoculating 8 RM with 10² FFU (n = 2), 10⁴ FFU (n = 2) or 10⁶ FFU (n = 4) of TR3ΔUL82gag. All RM developed robust HCMV-lysate and HIVgag-specific responses upon inoculation with TR3ΔUL82gag at 10⁴ and 10⁶ FFU, but not 10² FFU (Fig. 5B). Phenotyping of the HIVgag responsive CD8+ T cells confirmed that the majority of T cells, particularly CD8+ T cells, elicited by the pp71-deficient vectors displayed a T_EM phenotype (Fig. 5C). Taken together these data demonstrate that despite an inability to replicate in RM cells, HCMV TR3 vectors were capable of eliciting robust effector memory T cell responses in RM to a heterologous antigen inserted into the viral genome at a subcutaneous dose as low at 10⁴ FFU.

T cell immunogenicity of pentamer-deficient TR3

While the PC subunits UL128, UL130 and UL131 are dispensable for viral growth in fibroblasts, the glycoproteins gH and gL are essential since they form a trimeric complex with gO that binds to the cellular platelet-derived growth factor receptor α and this interaction is required for entry into fibroblasts^33,46. Given the important role of the PC in viral tropism, HCMV strains that lack the non-essential subunits are likely attenuated in vivo. Indeed, strain 68-1 RhCMV that lacks PC subunits homologous to UL128 and UL130 shows decreased viremia and shedding upon primary infection of RM⁴⁷. Interestingly however, despite lacking a functional PC, 68-1 RhCMV-based vaccines protect against HIV, TB and malaria^5,6,7.

To examine the role of the non-essential subunits UL128, UL130 or UL131 for TR3-based vector immunogenicity we initially deleted the entire coding regions for UL128, UL130, and UL131 by replacing the region with HIVgag (Fig. S3). As expected, TR3ΔUL128-131gag grew similar to parental TR3 in fibroblasts (Fig. 6A). However, when TR3ΔUL128-131gag was inoculated into RM, we did not observe HCMV-specific or HIVgag-specific T cell responses suggesting that this recombinant had lost the ability to infect RM (Fig. 6B). Since the homologue of UL131 is intact in 68-1 RhCMV whereas UL130 is partially and UL128 fully deleted⁴⁸ we next evaluated whether deletion of UL128 and/or UL130 would similarly affect vector function. As expected, TR3ΔUL78gag additionally deleted for UL128, UL130 or UL128+ UL130 grew similar to TR3ΔUL78gag in fibroblasts, but not in epithelial cells (Fig. 6C). We next inoculated two RM with each PC-deficient vector and monitored the HIVgag and HCMV-specific CD4+ and CD8+ T cell response in PBMC by intracellular cytokine staining over time. As shown in Fig. 6D, all RM developed robust HIVgag- and HCMV-specific T cell responses despite PC deficiency. Moreover, the majority of the HIVgag-specific CD8+ T cells displayed a T_EM phenotype (Fig. 6E). Thus, PC-deficient TR3 maintained the ability to elicit T cell responses in RM. However, the UL131 gene product seems to act not only as a PC subunit but seems to provide an additional, as yet unknown, function that is essential for virus infectivity in RM and possibly humans.

Attenuation of TR3 by the combined deletion of pentamer subunits and pp71

Since the limited tissue tropism of a PC-deficient vector would likely further increase the safety profile of vectors that are live-attenuated by deletion or inactivation of pp71 we wanted to determine the impact of a combined PC and pp71-deletion on cell-tropism, viral latency and cellular immune responses. Therefore, we deleted UL128 and UL130 from TR3ΔUL82gag using BAC mutagenesis. The deletion was confirmed by PCR analysis and NGS (Fig. S4). Upon reconstitution of virus in the presence of DAXX siRNA we confirmed loss of UL130 and pp71 by immunoblot of infected cell lysates (Fig. 7A) and demonstrated that virus growth in MRC-5 fibroblasts was impaired at low MOI unless cells were transfected with DAXX siRNA (Fig. 7B). As expected, the PC and pp71-deficient TR3 vector did not grow in non-fibroblast cells even at high MOI (Fig. S5). To determine whether the combined pp71 and PC deficiency affected the ability of TR3 to establish latency we inoculated humanized mice either with unmodified TR3 or with TR3ΔUL82gagΔUL128-130 and measured genome copy numbers in spleen and liver in the presence or absence of G-CSF. In the absence of G-CSF, similar amounts of latent viral genomes were detected for both TR3 and the PC-deficient recombinant (Fig. 7C). However, upon G-CSF treatment only TR3 displayed elevated viral genome copies consistent with viral reactivation, whereas TR3ΔUL82gagΔUL128-130 remained latent similar to the pp71-deleted parental recombinant. Taken together these data suggest that, while the PC is required for infection of non-fibroblast cells with cell-free virus in vitro, this complex does not seem to be essential for the transfer of virus from infected fibroblasts into human myeloid-lineage cells in vivo. Furthermore, once these cells are infected, the PC does not impact the establishment of latency. The lack of reactivation in response to G-CSF treatment is likely primarily due to lack of pp71.

To determine whether PC and pp71-deleted TR3 vectors retained the ability to elicit T cell responses to HIVgag we inoculated RM with different doses of TR3ΔUL82gagΔUL128-130 and monitored both HIVgag-specific and HCMV-specific T cell responses. As shown in Fig. 7D, we did not detect HIVgag-specific or HCMV-specific T cell responses in RM inoculated with PC and pp71-deficient TR3 at 10⁴ and 5 × 10⁴ FFU. However, all RM developed both HCMV and HIV-gag-specific T cell responses above a dose of 10⁵ FFU, although T cell responses were detectable about one week earlier in animals that received higher doses (Fig. 7D). Compared to pp71-intact TR3 it seems that deletion of the PC-subunits decreased the ability of pp71-deficient TR3 to elicit T cell responses by about 1 log. However, above a threshold of approximately 10⁵ FFU the resulting live attenuated vector was able to elicit and maintain effector memory T cell responses to the inserted heterologous antigen (Fig. 7E).

To determine whether the TR3 vectors elicited antibodies we performed enzyme linked immunosorbent assays (ELISAs) using lysates of HCMV-infected cells as antigens. We examined RM inoculated with TR3ΔUL78gag and increasing doses of TR3ΔUL82gag or TR3ΔUL82gagΔUL1280-130 as shown above. All animals that developed T cell responses to HIVgag and HCMV also showed above-background HCMV-reactive antibodies that peaked around day 42 post-inoculation (Fig. S6). However, unlike the T cell responses, which remained stable, the antibody response waned over time consistent with a lack of viral replication in the animals. Together these data suggest that similar to our observations in RhCMV, TR3 represents a robust T cell-inducing viral vector platform, even when highly attenuated, whereas the ability to elicit antibodies is limited, at least at the relatively low doses used here.

Discussion

As a first step towards achieving the goal of clinical testing of HCMV vectors we designed, constructed and characterized a novel live-attenuated HCMV-based vaccine platform. The attenuation strategy was derived from extensive studies in the RhCMV model demonstrating that deletion of the gene encoding the tegument protein pp71 rendered vectors deficient for inter- and intra-host spreading while maintaining all of the desired immunological characteristics as well as protection against SIV challenge^12,13. Since pp71 counteracts host transcriptional co-repressors of the promyelocytic leukemia protein (PML)-nuclear body complex⁴⁹ (sometimes referred to as “intrinsic” immunity²⁴), tumor suppressors⁵⁰ as well as innate immune signalling cascades⁵¹, this strategy thus engages several conserved antiviral host mechanisms to limit viral lytic replication. Deletion of pp71 as our primary attenuation strategy thus limits, if not completely eliminates, the possibility of HCMV pathogenesis in immunocompromised individuals which requires lytic replication for viral dissemination.

Since pp71 facilitates the expression of IE proteins which in turn activate viral gene expression and genome replication, it is considered a key protein in deciding establishment of latency versus activation of the lytic transcriptional cascade⁴². Similar to other tegument proteins, pp71 is delivered into the cytosol during viral entry. However, while pp71 travels to the nucleus in cells that are permissive for lytic replication, it remains cytoplasmic in incompletely differentiated myeloid cells as well as CD34+ hematopoietic stem cells resulting in lack of IE expression and viral latency in vitro⁵². Thus, by silencing IE expression cellular intrinsic defence proteins likely play a key role in the establishment and maintenance of HCMV latency. Consistent with this hypothesis we observed that pp71-deleted TR3 was able to establish latency in human CD34+ cells but could not be reactivated upon G-CSF treatment. In addition, it is possible that viral reactivation was further inhibited because deletion of the UL82 gene also eliminates the latent undefined nuclear antigen (LUNA) encoded on the opposite strand of UL82, a feature that is not conserved in the RhCMV homologue Rh110⁵³. LUNA was recently shown to de-sumoylate PML thus dispersing PML-bodies, a process that seems to be important for viral reactivation in response to external stimuli⁵⁴. Thus, UL82/LUNA-deleted TR3 preferentially establishes latency over lytic replication and is deficient in its ability to reactivate from latency.

The ability of pp71-deleted TR3 to persist in vivo is one of the desired features for a HCMV-based vector platform since viral persistence enables the maintenance of T_EM that require repeated antigen exposure. In our prototype vectors, the HIVgag antigen was expressed by replacing the coding regions of UL78 and UL82 which are expressed with early to late kinetics in vitro⁵⁵. Since early and late genes require transcriptional activation by viral IE protein, HIVgag would not be expected to be expressed during viral latency. Nevertheless, we observed robust and lasting HIVgag and HCMV-specific T_EM responses in RM inoculated with as little as 10⁴ FFU suggesting that some viral and heterologous insert gene expression occurs despite limited IE expression, perhaps as a consequence of partial reactivation. Thus, similar to our observation in RhCMV, we conclude that replacement of UL82 with a given transgene provides sufficient amount of antigen to elicit and maintain T_EM. The UL82-replacement design has the additional advantage to eliminate the possibility of reversal of UL82-deletion by homologous recombination with endogenous HCMV in chronically infected individuals. Although recombination with endogenous virus is likely a very rare event in immunocompetent individuals, as we have documented for pp71-deleted in RhCMV¹², there is clear evidence for recombination between different HCMV strains in highly viremic immunosuppressed individuals under the selective pressure of antiviral drugs⁵⁶. Since repair of UL82 by homologous recombination would result in an antigen-less TR3, the net result of homologous recombination would be infection by an additional HCMV variant in individuals that already carry HCMV, similar to the frequently observed superinfection with HCMV⁵⁷.

In addition to pp71-deletion we explored attenuation by deletion of the UL128, UL130 and UL131 subunits of PC. Unlike gH/gL, these subunits are non-essential for growth in vitro and, in fact, are often spontaneously deleted upon propagation of clinical isolates in fibroblasts⁵⁸. However, different from other clinical isolates, the TR3 PC seems to be stable upon repeated passages as demonstrated by NGS of passage 20 of TR3ΔUL82gag and TR3. This inherent stability is possibly due to the ability of HCMV TR to generate relatively high amounts of cell free virus which reduces the selection pressure for PC-deficient HCMV⁵⁹. Nevertheless, the PC is clearly functional in vitro, as demonstrated by increased infection of non-fibroblast cells, as well as in vivo, since PC-intact TR3 elicited T cell responses in RM at a lower threshold compared to PC-deleted TR3. Surprisingly, we observed that deletion of all three subunits resulted in vectors that had lost the ability to elicit T cell responses in RM even at the highest dose tested whereas deletion of UL128 and/or UL130 only increased the dose required to elicit T cell responses. Since all of these mutants are PC-deficient, this result suggests a PC-independent role for UL131 in promoting infection. At present this function is unknown. However, this observation is reminiscent of our previous finding that the NK cell evasion protein Rh159 of RhCMV is non-essential for growth in vitro, but required for infection in vivo⁶⁰ suggesting that UL131 might contribute to viral evasion of innate immune control. Further studies will be required to elucidate this unexpected, PC-independent function of UL131.

A second important rationale for pentamer deletion is our observation that RhCMV vectors derived from strain 68-1 elicit CD8+ T cells recognizing peptides in the context of major histocompatibility complex (MHC) class II or the non-polymorphic, highly conserved MHC class Ib molecule MHC-E instead of classical, polymorphic MHC-Ia molecules^61,62. In contrast, wildtype RhCMV or vectors based on strain 68-1.2 in which the RhCMV homologs of UL128 and UL130 were reinserted into 68-1, elicit CD8+ T cells restricted by MHC-Ia. Thus, it will be important to evaluate both PC-intact and PC-deleted HCMV-vectors in clinical trials to determine whether this “immune programming” can be translated from RM to human. The data presented here suggest that it will be possible to evaluate PC-deficient HCMV that is additionally attenuated by pp71-deletion although it might require a higher dose.

Most previously reported clinical studies with attenuated HCMV involved PC-deficient HCMV strains^16,63,64. In all of these cases only very modest T cell responses that did not show the CMV-typical T_EM bias were observed^63,65,66. The lack of T_EM bias indicates an inability for persistent antigen presentation due to a general inability to establish a persistent infection and to repeatedly provide viral antigens to professional antigen presenting cells. It is possible that this lack of T_EM bias was a consequence of PC-deficiency. However, we observed robust T_EM responses elicited and maintained by PC-deficient TR3 in RM, even upon additional attenuation by pp71-deletion. Similarly, we reported extensively on the robust T cell responses elicited by PC-deficient RhCMV to heterologous antigens inserted into strain 68-1^6,7,67. However, in addition to PC-deficiency, all previous clinical studies relied on HCMV strains and clones that display multiple genetic alterations as a results of passaging in tissue culture, including the low passage isolate Toledo³⁰. Our data thus suggest that the lack of PC was likely not the main reason for the lack of robust and durable T_EM responses (as well as unconventionally restricted CD8+ T cells⁶⁶) in previous clinical trials. This conclusion is supported by the finding that even a PC-intact version of the laboratory strain AD169 was unable to establish latency in humanized mice. Our strategy of using a wildtype-like vector backbone that is attenuated by a very specific molecular mechanism and contains a heterologous antigen will thus allow us to specifically address the role of the PC in the elicitation and maintenance of T cell responses, including unconventionally restricted CD8+ T cells.

One of the challenges of manufacturing viral vaccines that are attenuated by deletion of essential genes is the need for a production cell line that is compatible with vaccine manufacturing procedures. Generally, this is done by providing the deleted viral gene product in trans either by transient transfection or by generating a stable complementing cell line⁶⁸. While these approaches are well-established for transformed cell lines, efficient growth of HCMV only occurs in primary human cells which are difficult to transfect transiently with high efficiency and it is even more difficult to generate cloned, immortalized cell lines that maintain high levels of viral transgene expression. One possible strategy to overcome this limitation is to generate conditionally replicating HCMV by fusion of a degradation domain to essential viral proteins which allows the production of virus in primary cells upon addition of a stabilizing agent⁴⁴. However, since the long-term maintenance of T_EM responses likely requires the replication of viral genomes, at least at a low level, this approach runs the risk of eventually selecting for revertants in vivo upon viral genome replication. We demonstrate here an alternative production method that relies on siRNA-mediated suppression of anti-viral host cell proteins thus eliminating the need for the viral protein to support viral growth in vitro. Specifically, we show that by eliminating DAXX expression we can achieve viral titers for UL82-deleted TR3 that approach those of UL82-intact TR3. This approach is generally applicable to viral gene products whose main function is to counteract an anti-viral host cell protein. Importantly siRNA mediated DAXX knockdown was highly efficient in primary human cell lines, including MRC-5 cells which are approved for the manufacturing of viral vaccines for human use. Recently, this manufacturing strategy was used in the development of improved formulations for TR3ΔUL82gagΔUL1280-130 with respect to freeze-thaw characteristics and short-term liquid stability⁶⁹.

In sum, the TR3-based live attenuated HCMV vaccine vector platform not only maintains key immunological characteristics such as the ability to elicit lasting T_EM to an inserted heterologous antigen, but is also stable over many passages and can be manufactured in approved cell lines. As such, this vector platform is ideally suited to evaluate the unique aspects of CMV-based immunity in humans.

Material and Methods

Cell lines

Fetal human lung fibroblasts (MRC-5) were a gift from the international AIDS vaccine initiative (IAVI), human umbilical vein endothelial cells (HUVEC) were purchased from Lonza (cc-2519), human retinal pigmented epithelial cells (ARPE-19) and human Astrocytoma cell line (CCF-STTG1) were obtained from ATCC (CRL-2302 and CRL-1718). Primary rhesus fibroblasts (RF) were generated as described⁷⁰. MRC-5, ARPE-19 and RF were grown as monolayers in Dulbecco’s modified Eagle’s medium (DMEM) (Mediatech) supplemented with 10% fetal bovine serum (FBS) (HyClone), 4.5 g/l glucose, L-glutamine and sodium pyruvate, and antibiotics [penicillin (10 units/ml) and streptomycin (10 μg/ml)]. HUVEC were cultured in endothelial growth medium EGM-2 (Clonetics) containing 2% FBS, human recombinant vascular endothelial growth factor, basic fibroblast growth factor, human epidermal growth factor, insulin-like growth factor-1, hydrocortisone, ascorbic acid, gentamicin and amphotericin B (1 μg/ml each) (high-serum medium). CCF-STTG1 were grown in RPMI-1640 (HyClone) medium supplemented with 10% FBS, 4.5 g/l glucose, L-glutamine and sodium pyruvate, and antibiotics [penicillin (10 units/ml) and streptomycin (10 μg/ml)].

Primary human foreskin fibroblasts immortalized with telomerase (BJ-5ta) were described previously²⁸ and were purchased from ATCC (CRL-4001). BJ-5ta cells were grown in four parts DMEM, containing 4 mM L-glutamine, 4.5 g/l glucose and 1.5 g/l sodium bicarbonate, and one part of Medium 199 (Corning), supplemented with 10 µg/ml hygromycin B to maintain the telomerase expression, and 10% FBS.

To generate BJ-pp71 cells, BJ-5ta cells were first transduced with pLVX-blast (Clontech) into which the pp71-encoding UL82 gene (obtained by PCR from HCMV AD169) was inserted under control of a tetracycline-regulated-transactivator (tTa)-dependent promoter. Transductants were selected with 5 µg/ml blasticidin followed by a second transduction of pLVX-Tet-On-Advanced (Clontech), harboring the tTa and the neomycin gene which allows for selection by 100 µg/ml G418. Pp71 expression is inducible by the addition of 1 µg/ml doxycycline (DOX). All cells were tested for mycoplasma contamination using Universal Mycoplasma Detection Kit (ATCC 30-1012K^TM).

HCMV

HCMV AD169 and AD169rUL131, both of which express GFP in the UL21.5 locus, were described previously⁷¹ and were kindly provided by Dr. Tom Shenk, Princeton University. The HCMV isolate TR was isolated from an AIDS patient as previously reported reported⁷². TR-BAC (GenBank Accession Number: AC146906.1) was cloned by substituting non-excisable BAC DNA into the US2-6 region of the HCMV TR isolate as described¹⁶. TR-GFP was generated by inserting a GFP expression cassette into US7 together with a loxP site downstream of the BAC cassette and the US2-US7 genes from AD169 together with a second loxP site upstream of the BAC-cassette as described previously previously¹⁸. The BAC of TR-GFP was kindly provided by Dong Yu, Washington University.

TR3 was derived from TR-GFP by introducing UL97 from HCMV strain AD169 to restore ganciclovir sensitivity and by making the BAC cassette self-excising in human fibroblasts by inserting Cre via Lambda (λ) Red-mediated homologous recombination²⁰. In the first step we replaced the UL97 ORF with a galactokinase (galK) and kanamycin (Kan) selection cassette generated by PCR using primers containing 50 bp homology to regions flanking the UL97 ORF. E. coli strain SW105 containing the BAC of TR-GFP was electroporated with the PCR product for recombination and selected for the presence of galK on M63 minimal plates containing galactose. Next, we generated a recombination fragment that contained AD169 UL97 flanked by a DNA sequence homologous to flanking regions of TR UL97. This DNA fragment was then used to replace the GalK/Kan selection cassette with the AD169 UL97 by homologous recombination through selecting for loss of galK on 2-deoxy-galactose (DOG) plates. The resulting BAC construct was analyzed by restriction digest and the UL97 sequence was verified by sequencing of the corresponding PCR product. To generate a self-excising BAC cassette we inserted the Cre recombinase gene under control of the SV40 promoter downstream of the chloramphenicol resistance gene into the BAC cassette. Using a series of molecular cloning steps we generated a recombination fragment that contained the SV40/Cre expression and the galK/Kan resistance cassettes flanked upstream by a DNA sequence homologous to the BAC cassette and downstream by a loxP site and a DNA sequence homologous to a gene region downstream of the GFP gene. This DNA fragment was inserted into UL97-repaired TR-GFP using homologous recombination and selection for Kan resistance. In the next step, the galK/Kan selection cassette was removed by homologous recombination with a 100 bp double stranded DNA fragment with 50 bp homology to flanking regions. The resulting recombinant HCMV TR3 was then counter-selected on DOG plates for loss of galK. The final TR3 construct was analyzed by restriction digest, PCR and next generation sequencing (NGS) (GenBank Accession number: MN075802).

Recombinant HIVgag expressing vectors were constructed as follows: The HIVgag insert (GenBank Accession number: MN193852) was sub-cloned from a multigene construct containing the gag, reverse transcriptase, integrase and nef (GRIN) sequences of an HIV-1 subtype A isolate⁷³, (kindly provided by IAVI). TR3ΔUL78gag was generated by homologous recombination in E. coli strain EL250 containing heat-inducible λ red genes and an arabinose-inducible FLP recombinase⁷⁴. Using primers containing 50 bp homology to regions flanking UL78, HIVgag was amplified by PCR together with a Kan resistance cassette flanked by FRT sites from plasmid pCP015. EL250 bacteria containing TR3 BAC were electroporated with the PCR product for recombination. Recombinants were selected for Kan resistance, followed by selection of FRT-mediated loss of Kan upon arabinose induction of FLP recombinase.

TR3ΔUL82gag was generated using the galK/Kan two-step recombination and selection system. In step 1 UL82 was replaced by homologous recombination in TR3-BAC-containing SW105 bacteria with a PCR product containing a galK/Kan expression cassette with 50 bp flanking homology to UL82. Upon positive selection for Kan and chloramphenicol, UL82 replacement was confirmed by restriction digest and diagnostic PCR. In step 2 the galK/Kan cassette was replaced with a PCR product containing the HIVgag insert and the same flanking homology arms. Recombinants were selected on DOG and chloramphenicol minimal media with glycerol as the carbon source. Using the galK/Kan selection method we introduced deletions of UL128 and/or UL130 into TR3ΔUL78gag or TR3ΔUL82gag to generate TR3ΔUL78gagΔUL128, TR3ΔUL78gagΔUL130, and TR3ΔUL78gagΔUL128-130 or TR3ΔUL82gagΔUL128-130. To generate TR3ΔUL128-131gag we used galK/Kan to replace the entire gene region spanning UL131, UL130 and UL128 with the HIVgag ORF. All BAC constructs were verified by NGS.

Viruses were generated by electroporation of the BAC DNA into MRC-5 cells. TR3ΔUL82 was recovered in the presence of 5 nM DAXX siRNA (duplex of 5′-GCUACAAGCUGGAGAAUGAUU-3′ and 3′-UUCGAUGUUCGACCUCUUACU-5′) (Dharmacon). Viral stocks were prepared in MRC-5 cells cultured in DMEM with 10% FBS upon transfection of 5 nM DAXX siRNA with Lipofectamine 2000 (Invitrogen) 24 h prior to infection with an MOI = 0.01. DAXX siRNA was re-transfected on day 10 and, at full cytopathic effect (CPE), the supernatant from infected cells was harvested and purified on a 20% sorbitol cushion. Purified virus was resuspended in DMEM with 2% FBS and stored at −80 °C.

Next generation sequencing

BAC DNA was prepared using the NucleoBond PC 100 kit (Macherey-Nagel) following the manufacturer’s instructions. To generate purified viral DNA for NGS, we used a modified extraction protocol originally designed for the extraction of Polyoma virus DNA from mouse cells⁷⁵. Viral supernatants were harvested at full CPE, cellular contaminants and residual cells were removed by centrifugation initially at 2,000 × g for 10 min at 4 °C and subsequently at 7,500 × g for 15 min. The virus was pelleted from the clarified medium by overlaying a sorbitol cushion (20% D-sorbitol, 50 mM Tris [pH 7.4], 1 mM MgCl₂) and centrifuging at 82,000 × g for 1 h at 4 °C in a Beckman SW28 rotor. The generated pellet was resuspended in 500 μl 10.1 TE Buffer (10 mM Tris, pH 8.0; 0.1 mM EDTA, pH 8.0) and 500 μl 2x lysis buffer (20 mM Tris-Cl, pH 8.0; 50 mM EDTA, pH 8.0; 200 mM NaCl; 1.2% w/v SDS) was added. Finally, to digest the purified virions, 250 μg Proteinase K was added and the solution was incubated for 2 h at 37 °C. The viral DNA was phenol/chloroform extracted twice and precipitated with absolute ethanol at −80 °C overnight. The DNA was pelleted for 20 min at 13,200 × g at 4 °C, washed once with 70% ethanol, and subsequently resuspended in DNase-free water. DNA concentrations were determined using a ND-1000 Spectrophotometer (NanoDrop Technologies).

Illumina sequencing libraries were generated as previously described⁷⁶. Briefly, DNA was fragmented using an S2 Sonicator and was then converted to libraries using the standard TruSeq protocol. Libraries were examined on a Bioanalyzer (Agilent) and the concentration was determined using real time PCR and SYBR Green fluorescence. NGS was performed using a MiSeq NGS System (Illumina). Libraries were loaded into a MiSeq reagent cartridge at a concentration of 9 pM and single read sequencing was performed for 300 cycles with 6 additional cycles of index reads. The resulting data was imported into Geneious and the sequencing reads were trimmed of all regions exceeding the error probability limit of 0.1% to minimize sequencing errors. All reads with a total length of fewer than 50 bp after quality control were eliminated from further analysis to increase the likelihood of specific alignments during de novo and reference guided assemblies. Viral genomes were first de novo assembled using the processed sequencing data, and subsequently all reads were aligned to the generated consensus sequence in a reference guided assembly to examine potential SNPs.

Titration of pp71-deleted TR3

BJ-pp71 cells were seeded at approximately 12,000 cells/well into 96-well plates. The next day, viral supernatants were diluted in warm media (DMEM supplemented with 10% heat-inactivated tetracycline free FBS (Atlanta Biologicals), 1% 100X Pen-Strep-Glutamate (Invitrogen) and 1 µg/ml doxycycline). The cells were infected with serial virus dilutions and incubated for ~18 hours. Methanol (at −20 °C) was gently added to the side of each well. Plates were sealed and placed at −20 °C for 10–20 min followed by removal of the methanol. Using 200 to 300 μl of wash buffer (Dulbecco’s phosphate buffered saline (DPBS) with 0.05% Tween), plates were washed 3 times for 5 min. 100 μl of blocking solution (DPBS with 10% FBS) was added per well and incubated at 37 °C for 60 min. The blocking solution was removed and 50 μl of primary antibody solution (1:1000 mouse anti-HCMV (0898) antibody (Santa Cruz, Catalog No. sc-58118) in blocking solution) was added followed by incubation at 37 °C for 60 min. Using 200 to 300 μl of wash buffer per well the plates were washed 3 × 5 min. and 50 µl of the secondary antibody solution (1:1000 goat anti-mouse horseradish peroxidase antibody (Santa Cruz, Catalog No. sc-2005) in blocking buffer) was added to each well followed by incubation at 37 °C for 60 min. Plates were washed 3x for 5 min using 200 to 300 μl of wash buffer per well. Wash buffer was carefully removed after the final wash. 50 µl of TrueBlue Stain (KPL) was added to each well. Staining was monitored by microscopy and, once developed, the staining solution was removed and discarded. Each well was washed twice with 200 μl of deionized water to stop the reaction. Images were captured for each well and number of distinct blue foci were counted. All dilutions in the quantifiable range (1 to 200 foci/well) were counted by fluorescent microscopy, and for analysis of linearity, each dilution’s titers were independently calculated. The concentration of focus-forming units (FFU/ml) were calculated by the following formula: FFU/ml = foci count × dilution factor × 10.

Growth curves

For each time point, cells (MRC-5, HUVEC, ARPE-19, or CCF-STTG1) were plated separately in triplicate into 24-well plates. For each cell type, the media used at this step was the same as the cell media used for that cell type’s growth and propagation. After 24 h, each cell type was infected with recombinant HCMV at the indicated MOI. After 2 h, the inoculum was removed, cells were washed 3 times with phosphate buffered saline (PBS), and fresh media was added. Supernatants were collected at the indicated time after the relevant HCMV infection and either frozen at approximately −80 °C or titrated on BJ-pp71 cells.

Immunoblotting

Total proteins were extracted from HCMV infected fibroblasts using Cell Lysis Buffer (Cell Signaling Technologies). Extracted proteins were quantified by the BCA method (Pierce BCA Protein Assay kit, Thermo Scientific) and diluted to the same protein concentrations. Equivalent amounts of each group were separated on 8–12% SDS polyacrylamide gels. Proteins were transferred to Immobilon-P Transfer Membranes and probed with the following specific antibodies: rabbit polyclonal immediate-early (IE)-specific antibody⁷⁷; mouse monoclonal anti HCMV UL44 (CH16; Virusys Corporation), mouse monoclonal anti HCMV UL99 (pp28)⁷⁸, mouse monoclonal anti HCMV UL82 (pp71))⁷⁹ (kindly provided by Dr. Tom Shenk), rabbit polyclonal HIV-1 SF2 HIVgag p24 (#4250, NIH AIDS Reagent Program), mouse monoclonal anti-actin (MA5-11869; Thermo Scientific), rabbit polyclonal anti HCMV US6²², rabbit polyclonal anti HCMV UL130³² (kindly provided by David Johnson), rabbit anti DAXX (D7810, Sigma). Secondary antibodies: goat anti-mouse IgG-HRP (sc-2005; Santa Cruz) and goat anti-rabbit IgG-HRP (111-035-003; Jackson Immunoresearch). Membranes were stained multiple times after been stripped with Restore Western Blot Stripping buffer (Thermo Scientific).

Engraftment and infection of humanized mice

All animal experiments were conducted under the approved Oregon Health and Science University (OHSU) Institutional Animal Care and Use Committee (IACUC) protocol (3498). All mice in this study were managed in accordance with the NIH Office of Laboratory Animal Welfare: “PHS Policy on the Humane Care and Use of Research Animals” and the recommendations of the American Association for Accreditation of Laboratory Animal Care (AAALAC): “The Guide for the Care and Use of Laboratory Animals, 8^th edition”. NOD-scidIL2Rγc null (NSG) mice were purchased from Jackson Laboratories and bred in-house. All mice were housed in micro-isolator cages in a designated specific pathogen-free facility at OHSU and fed sterile food and water. Mice were euthanized via CO₂ administration according to AAALAC euthanasia guidelines. Engraftment of NSG mice with human CD34+ hematopoietic stem cells was performed as described previously³⁹. These huNSG mice were infected with HCMV after pre-treatment with 1 mL of 4% Thioglycollate (Brewer’s Media; BD) by intraperitoneal (IP) injection to recruit monocyte/macrophages. At 24 h post-treatment, humanized mice were infected with HCMV-infected fibroblasts from two T150 flasks (approximately 10⁵ FFU of cell-associated virus per mouse) via IP injection. To promote HCMV reactivation, mice were treated with G-CSF and AMD3100 as previously described³⁹. Briefly, G-CSF (100 µL at 300 µg/mL; Amgen) was delivered via an osmotic pump (1007D; Azlet) surgically implanted in the subcutaneous space beneath the dorsal skin on the mouse for 7 days. Additionally, on the same day as pump implantation, mice were also given a single IP injection of 125 µg of AMD3100 (1,1′-[1,4-Phenylenebis(methylene)]bis-1,4,8,11-tetraazacyclotetradecane octahydrochloride (Sigma).

Quantitative PCR to determine HCMV genome copy numbers

Total DNA was extracted from approximately 1 mm² sections of mouse spleen or liver using the DNAzol reagent (Life Technologies). Quantitative PCR (TaqMan) was performed on 1 µg of total DNA using TaqMan FastAdvance PCR Master Mix (Applied Biosystems), according to the manufacturer’s instructions. Primers and a probe recognizing HCMV UL141 were used to quantify HCMV genomes (probe: 5′CGAGGGAGAGCAAGTT; forward primer: 5′GATGTGGGCCGAGAATTATGA and reverse primer: 5′ATGGGCCAGGAGTGTGTCA). The probe contains a 5′ FAM reporter molecule and a 3′ quencher molecule (Applied Biosystems). The reaction was activated at 95 °C for 10 minutes followed by 40 cycles (15 s at 95 °C and 1 minute at 60 °C) using a StepOnePlus TaqMan PCR machine. Results were analyzed using ABI StepOne software (Applied Biosystems).

T cell responses to TR3 in rhesus macaques

A total of 30 purpose-bred male or female rhesus macaques (RM) (Macaca mulatta) of Indian genetic background were used in this study. All RM were classified as specific pathogen free (SPF) as defined by being free of cercopithicine herpesvirus 1, D-type simian retrovirus, simian T-lymphotrophic virus type 1, SIV, and Mycobacterium tuberculosis, but were naturally infected with RhCMV. All RM used in this study were housed at the Oregon National Primate Research Center (ONPRC) in Animal Biosafety level (ABSL)-2 rooms. RM care and all experimental protocols and procedures were approved by the ONPRC IACUC. The ONPRC is a Category I facility. The Laboratory Animal Care and Use Program at the ONPRC is fully accredited by AAALAC and has an approved assurance number (#A3304-01) for the care and use of animals on file with the NIH Office for Protection from Research Risks. The IACUC adheres to national guidelines established in the Animal Welfare Act (7 U.S.C. Sections 2131–2159) and the Guide for the Care and Use of Laboratory Animals (8th Edition) as mandated by the U.S. Public Health Service Policy.

RM were inoculated sub-cutaneously with 10³-5 × 10⁶ FFU of TR3-derived recombinant HCMV expressing HIVgag. To monitor T cell and antibody responses, animals were subjected to bi-weekly blood draws and serum as well as PBMC were recovered. CD4+ and CD8+ T cell responses were measured in PBMC by intracellular cytokine staining^4,19,67. For HIVgag-specific responses, PBMC were incubated with consecutive 15mer peptide mixes (11 amino acid overlap) comprising the inserted HIVgag sequence. For HCMV lysates, human fibroblasts infected with HCMV TR were scraped and pelleted to collect the cells at the time of maximum cytopathic effect. The cell pellet was resuspended and lysed by three cycles of freeze thaw and clarified by centrifugation at 3,000 × g for 10 min. Clarified supernatants were pelleted by ultracentrifugation through a 10% sorbitol cushion at 22,000 rpm in a Beckman SW32 rotor to concentrate the virus. The virus pellet was resuspended in the clarified cell lysate and mixed by vortexing. The mixture was aliquoted and frozen at −80 °C prior to use in the T cell assay.

For intracellular cytokine staining, the co-stimulatory molecules CD28 and CD49d (BD Biosciences) were added for 1 h, followed by addition of Brefeldin A (Sigma-Aldrich) for an additional 8 h. Co-stimulation without peptides served as background control. Stimulated cells were fixed, permeabilized and stained^4,19,67 using combinations of the following fluorochrome-conjugated mAbs: SP34-2 (CD3; Pacific Blue, Alexa700), L200 (CD4; AmCyan, BV510), SK-1 (CD8α; PerCP-Cy5.5), MAB11 (TNFα; FITC, PE), B27 (IFNγ; APC), FN50 (CD69; PE-TexasRed), B56 (Ki-67; FITC), and in polycytokine analyses, JES6-5H4 (IL2; PE Cy-7). Data was collected on an LSR-II (BD Biosciences). Analysis was performed using FlowJo software (Tree Star). Lymphocytes were gated for CD3+ and progressive gating on CD4+ and CD8+ T cell subsets. Antigen-responding cells in both CD4+ and CD8+ T cell populations were determined by their intracellular expression of CD69 and one or more cytokines. After subtracting background, the raw response frequencies were memory corrected^4,19,67 using combinations of the following mAbs to define the memory vs. naïve subsets: SP34-2 (CD3; Alexa700, PerCP-Cy5.5), L200 (CD4; AmCyan), SK-1 (CD8α; APC, PerCP-Cy-5.5), MAB11 (TNFα; FITC), B27 (IFNγ; APC), FN50 (CD69; PE), CD28.2 (CD28; PE-TexasRed), DX2 (CD95; PE), 15053 (CCR7; Pacific Blue), and B56 (Ki-67; FITC). For memory phenotype and polycytokine analysis of antigen-specific T cells, all cells expressing CD69 plus one or more cytokines were first Boolean gated, and then this overall Ag-responding population was subdivided into the subsets of interest on the basis of surface phenotype or cytokine production pattern^4,19,67.

Antibody responses to TR3 in rhesus macaques

End-point titer Enzyme-Linked Immunosorbent Assay (ELISA) was used to quantify anti-HCMV antibodies in plasma from RM. High binding ELISA 96-well plates (Costar) were coated overnight with 100 µl of lysates generated from HCMV TR-infected fibroblasts at 10 µg/ml diluted in PBS. Plates were incubated for 1 hour with 200 µl of blocking buffer (2% milk powder resuspended in PBST (PBS plus 0.1% Tween-20) and then washed three times with PBST. Serial two-fold dilutions of primary plasma samples, starting at 1:100, were generated in blocking buffer and then incubated for 2 h at room temperature. Plates were washed three times with PBST and then incubated with secondary anti-human IgGAM (Rockland) conjugated with horseradish peroxidase. Plates were washed three times with PBST and bound secondary antibody was detected using the OPD substrate (Life Technologies) followed by addition of HCl to stop the reaction. The plates were read at 490 nm using a Synergy HTX Microplate Reader (BioTek). End-point antibody binding titers were calculated by Log/Log transformation and analyzed using GraphPad Prism v6 software.

Data availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

References

Mendez, A. C., Rodriguez-Rojas, C. & Del Val, M. Vaccine vectors: the bright side of cytomegalovirus. Medical microbiology and immunology (2019).
Früh, K. & Picker, L. CD8+ T cell programming by cytomegalovirus vectors: applications in prophylactic and therapeutic vaccination. Curr Opin Immunol 47, 52–56 (2017).
Article PubMed PubMed Central CAS Google Scholar
Jarvis, M. A., Hansen, S. G., Nelson, J. A., Picker, L. J. & Früh, K. In Cytomegaloviruses: From Molecular Pathogenesis to Intervention Vol. 2 CYTOMEGALOVIRUSES: Molecular Biology and Immunology (ed Reddehase, M. J.) Ch. 21, (Caister Academic Press, 2013).
Hansen, S. G. et al. Profound early control of highly pathogenic SIV by an effector memory T-cell vaccine. Nature 473, 523–527 (2011).
Article ADS CAS PubMed PubMed Central Google Scholar
Hansen, S. G. et al. Immune clearance of highly pathogenic SIV infection. Nature 502, 100–104 (2013).
Article ADS CAS PubMed PubMed Central Google Scholar
Hansen, S. G. et al. Prevention of tuberculosis in rhesus macaques by a cytomegalovirus-based vaccine. Nature medicine 24, 130–143 (2018).
Article CAS PubMed PubMed Central Google Scholar
Hansen, S. G. et al. Cytomegalovirus vectors expressing Plasmodium knowlesi antigens induce immune responses that delay parasitemia upon sporozoite challenge. PLoS One 14, e0210252 (2019).
Article CAS PubMed PubMed Central Google Scholar
Okoye, A. A. et al. Early antiretroviral therapy limits SIV reservoir establishment to delay or prevent post-treatment viral rebound. Nature medicine 24, 1430–1440 (2018).
Article CAS PubMed PubMed Central Google Scholar
Klenerman, P. The (gradual) rise of memory inflation. Immunological reviews 283, 99–112 (2018).
Article CAS PubMed PubMed Central Google Scholar
Mohr, C. A. et al. A spread-deficient cytomegalovirus for assessment of first-target cells in vaccination. J Virol 84, 7730–7742 (2010).
Article CAS PubMed PubMed Central Google Scholar
Snyder, C. M., Cho, K. S., Bonnett, E. L., Allan, J. E. & Hill, A. B. Sustained CD8+ T cell memory inflation after infection with a single-cycle cytomegalovirus. PLoS Pathog 7, e1002295 (2011).
Article CAS PubMed PubMed Central Google Scholar
Marshall, E. E. et al. Enhancing safety of cytomegalovirus-based vaccine vectors by engaging host intrinsic immunity. Sci Transl Med 11 (2019).
Article PubMed CAS PubMed Central Google Scholar
Hansen, S. G. et al. A live-attenuated RhCMV/SIV vaccine shows long-term efficacy against heterologous SIV challenge. Sci Transl Med 11 (2019).
Article PubMed CAS PubMed Central Google Scholar
Leendertz, F. H. et al. Novel cytomegaloviruses in free-ranging and captive great apes: phylogenetic evidence for bidirectional horizontal transmission. J Gen Virol 90, 2386–2394 (2009).
Article CAS PubMed Google Scholar
Pass, R. F. Cytomegalovirus infection. Pediatr Rev 23, 163–170 (2002).
Article PubMed Google Scholar
Murphy, E. et al. Coding potential of laboratory and clinical strains of human cytomegalovirus. Proc Natl Acad Sci USA 100, 14976–14981 (2003).
Article ADS CAS PubMed PubMed Central Google Scholar
Stanton, R. J. et al. Reconstruction of the complete human cytomegalovirus genome in a BAC reveals RL13 to be a potent inhibitor of replication. The Journal of clinical investigation 120, 3191–3208 (2010).
Article CAS PubMed PubMed Central Google Scholar
Lauron, E. J., Yu, D., Fehr, A. R. & Hertel, L. Human cytomegalovirus infection of langerhans-type dendritic cells does not require the presence of the gH/gL/UL128-131A complex and is blocked after nuclear deposition of viral genomes in immature cells. J Virol 88, 403–416 (2014).
Article PubMed PubMed Central CAS Google Scholar
Hansen, S. G. et al. Evasion of CD8+ T cells is critical for superinfection by cytomegalovirus. Science 328, 102–106 (2010).
Article ADS CAS PubMed PubMed Central Google Scholar
Warming, S., Costantino, N., Court, D. L., Jenkins, N. A. & Copeland, N. G. Simple and highly efficient BAC recombineering using galK selection. Nucleic acids research 33, e36 (2005).
Article PubMed PubMed Central CAS Google Scholar
Yu, D., Smith, G. A., Enquist, L. W. & Shenk, T. Construction of a self-excisable bacterial artificial chromosome containing the human cytomegalovirus genome and mutagenesis of the diploid TRL/IRL13 gene. J Virol 76, 2316–2328 (2002).
Article CAS PubMed PubMed Central Google Scholar
Ahn, K. et al. The ER-luminal domain of the HCMV glycoprotein US6 inhibits peptide translocation by TAP. Immunity 6, 613–621 (1997).
Article CAS PubMed Google Scholar
Bresnahan, W. A. & Shenk, T. E. UL82 virion protein activates expression of immediate early viral genes in human cytomegalovirus-infected cells. Proc Natl Acad Sci USA 97, 14506–14511 (2000).
Article ADS CAS PubMed PubMed Central Google Scholar
Saffert, R. T. & Kalejta, R. F. Inactivating a cellular intrinsic immune defense mediated by Daxx is the mechanism through which the human cytomegalovirus pp71 protein stimulates viral immediate-early gene expression. J Virol 80, 3863–3871 (2006).
Article CAS PubMed PubMed Central Google Scholar
Preston, C. M. & Nicholl, M. J. Role of the cellular protein hDaxx in human cytomegalovirus immediate-early gene expression. J Gen Virol 87, 1113–1121 (2006).
Article CAS PubMed Google Scholar
Cantrell, S. R. & Bresnahan, W. A. Human cytomegalovirus (HCMV) UL82 gene product (pp71) relieves hDaxx-mediated repression of HCMV replication. J Virol 80, 6188–6191 (2006).
Article CAS PubMed PubMed Central Google Scholar
O’Connor, C. M. & Shenk, T. Human cytomegalovirus pUL78 G protein-coupled receptor homologue is required for timely cell entry in epithelial cells but not fibroblasts. J Virol 86, 11425–11433 (2012).
Article PubMed PubMed Central CAS Google Scholar
Bodnar, A. G. et al. Extension of life-span by introduction of telomerase into normal human cells. Science 279, 349–352 (1998).
Article ADS CAS PubMed Google Scholar
Wilkinson, G. W. et al. Human cytomegalovirus: taking the strain. Medical microbiology and immunology 204, 273–284 (2015).
Article CAS PubMed PubMed Central Google Scholar
Vashee, S. et al. Cloning, Assembly, and Modification of the Primary Human Cytomegalovirus Isolate Toledo by Yeast-Based Transformation-Associated Recombination. mSphere 2 (2017).
Hahn, G. et al. Human cytomegalovirus UL131-128 genes are indispensable for virus growth in endothelial cells and virus transfer to leukocytes. J Virol 78, 10023–10033 (2004).
Article CAS PubMed PubMed Central Google Scholar
Ryckman, B. J. et al. Characterization of the human cytomegalovirus gH/gL/UL128-131 complex that mediates entry into epithelial and endothelial cells. J Virol 82, 60–70 (2008).
Article CAS PubMed Google Scholar
Martinez-Martin, N. et al. An Unbiased Screen for Human Cytomegalovirus Identifies Neuropilin-2 as a Central Viral Receptor. Cell 174, 1158–1171 e1119 (2018).
Article CAS PubMed Google Scholar
Vanarsdall, A. L. et al. CD147 Promotes Entry of Pentamer-Expressing Human Cytomegalovirus into Epithelial and Endothelial Cells. MBio 9 (2018).
E, X. et al. OR14I1 is a receptor for the human cytomegalovirus pentameric complex and defines viral epithelial cell tropism. Proc Natl Acad Sci USA (2019).
Tang, J., Frascaroli, G., Lebbink, R. J., Ostermann, E. & Brune, W. Human cytomegalovirus glycoprotein B variants affect viral entry, cell fusion, and genome stability. Proc Natl Acad Sci USA (2019).
Caviness, K. et al. Complex Interplay of the UL136 Isoforms Balances Cytomegalovirus Replication and Latency. MBio 7, e01986 (2016).
Article CAS PubMed PubMed Central Google Scholar
Crawford, L. B., Streblow, D. N., Hakki, M., Nelson, J. A. & Caposio, P. Humanized mouse models of human cytomegalovirus infection. Curr Opin Virol 13, 86–92 (2015).
Article CAS PubMed PubMed Central Google Scholar
Smith, M. S. et al. Granulocyte-colony stimulating factor reactivates human cytomegalovirus in a latently infected humanized mouse model. Cell Host Microbe 8, 284–291 (2010).
Article CAS PubMed PubMed Central Google Scholar
Umashankar, M. et al. A novel human cytomegalovirus locus modulates cell type-specific outcomes of infection. PLoS Pathog 7, e1002444 (2011).
Article CAS PubMed PubMed Central Google Scholar
Penkert, R. R. & Kalejta, R. F. Nuclear localization of tegument-delivered pp71 in human cytomegalovirus-infected cells is facilitated by one or more factors present in terminally differentiated fibroblasts. J Virol 84, 9853–9863 (2010).
Article CAS PubMed PubMed Central Google Scholar
Penkert, R. R. & Kalejta, R. F. Tale of a tegument transactivator: the past, present and future of human CMV pp71. Future Virol 7, 855–869 (2012).
Article CAS PubMed PubMed Central Google Scholar
Fu, T. M. et al. Restoration of viral epithelial tropism improves immunogenicity in rabbits and rhesus macaques for a whole virion vaccine of human cytomegalovirus. Vaccine 30, 7469–7474 (2012).
Article CAS PubMed Google Scholar
Wang, D. et al. A replication-defective human cytomegalovirus vaccine for prevention of congenital infection. Sci Transl Med 8, 362ra145 (2016).
Article PubMed CAS Google Scholar
DeFilippis, V. & Früh, K. Rhesus cytomegalovirus particles prevent activation of interferon regulatory factor 3. J Virol 79, 6419–6431 (2005).
Article CAS PubMed PubMed Central Google Scholar
Wu, K., Oberstein, A., Wang, W. & Shenk, T. Role of PDGF receptor-alpha during human cytomegalovirus entry into fibroblasts. Proc Natl Acad Sci USA 115, E9889–E9898 (2018).
Article CAS PubMed PubMed Central Google Scholar
Oxford, K. L. et al. Open reading frames carried on UL/b’ are implicated in shedding and horizontal transmission of rhesus cytomegalovirus in rhesus monkeys. J Virol 85, 5105–5114 (2011).
Article CAS PubMed PubMed Central Google Scholar
Oxford, K. L. et al. Protein coding content of the UL)b’ region of wild-type rhesus cytomegalovirus. Virology 373, 181–188 (2008).
Article CAS PubMed Google Scholar
Cantrell, S. R. & Bresnahan, W. A. Interaction between the human cytomegalovirus UL82 gene product (pp71) and hDaxx regulates immediate-early gene expression and viral replication. J Virol 79, 7792–7802 (2005).
Article CAS PubMed PubMed Central Google Scholar
Kalejta, R. F. & Shenk, T. Proteasome-dependent, ubiquitin-independent degradation of the Rb family of tumor suppressors by the human cytomegalovirus pp71 protein. Proc Natl Acad Sci USA 100, 3263–3268 (2003).
Article ADS CAS PubMed PubMed Central Google Scholar
Fu, Y. Z. et al. Human Cytomegalovirus Tegument Protein UL82 Inhibits STING-Mediated Signaling to Evade Antiviral Immunity. Cell Host Microbe 21, 231–243 (2017).
Article CAS PubMed Google Scholar
Saffert, R. T., Penkert, R. R. & Kalejta, R. F. Cellular and viral control over the initial events of human cytomegalovirus experimental latency in CD34+ cells. J Virol 84, 5594–5604 (2010).
Article CAS PubMed PubMed Central Google Scholar
Reeves, M. B. & Sinclair, J. H. Analysis of latent viral gene expression in natural and experimental latency models of human cytomegalovirus and its correlation with histone modifications at a latent promoter. J Gen Virol 91, 599–604 (2010).
Article CAS PubMed Google Scholar
Poole, E. L. et al. A Virally Encoded DeSUMOylase Activity Is Required for Cytomegalovirus Reactivation from Latency. Cell Rep 24, 594–606 (2018).
Article CAS PubMed PubMed Central Google Scholar
Chambers, J. et al. DNA microarrays of the complex human cytomegalovirus genome: profiling kinetic class with drug sensitivity of viral gene expression. J Virol 73, 5757–5766 (1999).
CAS PubMed PubMed Central Google Scholar
Cudini, J. et al. Human cytomegalovirus haplotype reconstruction reveals high diversity due to superinfection and evidence of within-host recombination. Proc Natl Acad Sci USA 116, 5693–5698 (2019).
Article CAS PubMed PubMed Central Google Scholar
Pokalyuk, C. et al. Characterizing human cytomegalovirus reinfection in congenitally infected infants: an evolutionary perspective. Mol Ecol 26, 1980–1990 (2017).
Article PubMed Google Scholar
Dargan, D. J. et al. Sequential mutations associated with adaptation of human cytomegalovirus to growth in cell culture. J Gen Virol 91, 1535–1546 (2010).
Article CAS PubMed PubMed Central Google Scholar
Murrell, I. et al. Impact of sequence variation in the UL128 locus on production of human cytomegalovirus in fibroblast and epithelial cells. J Virol 87, 10489–10500 (2013).
Article CAS PubMed PubMed Central Google Scholar
Sturgill, E. R. et al. Natural Killer Cell Evasion Is Essential for Infection by Rhesus Cytomegalovirus. PLoS Pathog 12, e1005868 (2016).
Article PubMed PubMed Central CAS Google Scholar
Hansen, S. G. et al. Cytomegalovirus vectors violate CD8+ T cell epitope recognition paradigms. Science 340, 1237874 (2013).
Article PubMed CAS Google Scholar
Hansen, S. G. et al. Broadly targeted CD8(+) T cell responses restricted by major histocompatibility complex E. Science 351, 714–720 (2016).
Article ADS CAS PubMed PubMed Central Google Scholar
Adler, S. P. et al. A Phase 1 Study of Four Live, Recombinant Human Cytomegalovirus Towne/Toledo Chimera Vaccines in CMV Seronegative Men. J Infect Dis (2016).
Schleiss, M. R. Cytomegalovirus vaccines under clinical development. J Virus Erad 2, 198–207 (2016).
PubMed PubMed Central Google Scholar
Jacobson, M. A. et al. Antigen-specific T cell responses induced by Towne cytomegalovirus (CMV) vaccine in CMV-seronegative vaccine recipients. J Clin Virol 35, 332–337 (2006).
Article CAS PubMed Google Scholar
Murray, S. E. et al. Fibroblast-adapted human CMV vaccines elicit predominantly conventional CD8 T cell responses in humans. J Exp Med (2017).
Hansen, S. G. et al. Effector memory T cell responses are associated with protection of rhesus monkeys from mucosal simian immunodeficiency virus challenge. Nature medicine 15, 293–299 (2009).
Article CAS PubMed PubMed Central Google Scholar
van der Loo, J. C. & Wright, J. F. Progress and challenges in viral vector manufacturing. Hum Mol Genet 25, R42–52 (2016).
Article PubMed CAS Google Scholar
Kumru, O. S. et al. Stabilization and formulation of a recombinant Human Cytomegalovirus vector for use as a candidate HIV-1 vaccine. Vaccine (2019).
Chang, W. L., Kirchoff, V., Pari, G. S. & Barry, P. A. Replication of rhesus cytomegalovirus in life-expanded rhesus fibroblasts expressing human telomerase. J Virol Methods 104, 135–146 (2002).
Article CAS PubMed Google Scholar
Wang, D. & Shenk, T. Human cytomegalovirus UL131 open reading frame is required for epithelial cell tropism. J Virol 79, 10330–10338 (2005).
Article CAS PubMed PubMed Central Google Scholar
Smith, I. L. et al. Clinical failure of CMV retinitis with intravitreal cidofovir is associated with antiviral resistance. Arch Ophthalmol 116, 178–185 (1998).
CAS PubMed Google Scholar
Keefer, M. C. et al. A phase I double blind, placebo-controlled, randomized study of a multigenic HIV-1 adenovirus subtype 35 vector vaccine in healthy uninfected adults. PLoS One 7, e41936 (2012).
Article ADS CAS PubMed PubMed Central Google Scholar
Lee, E. C. et al. A highly efficient Escherichia coli-based chromosome engineering system adapted for recombinogenic targeting and subcloning of BAC DNA. Genomics 73, 56–65 (2001).
Article CAS PubMed Google Scholar
Hirt, B. Selective extraction of polyoma DNA from infected mouse cell cultures. J Mol Biol 26, 365–369 (1967).
Article CAS PubMed Google Scholar
Burwitz, B. J. et al. Cross-Species Rhesus Cytomegalovirus Infection of Cynomolgus Macaques. PLoS Pathog 12, e1006014 (2016).
Article PubMed PubMed Central CAS Google Scholar
Jupp, R. et al. Direct interaction of the human cytomegalovirus IE86 protein with the cis repression signal does not preclude TBP from binding to the TATA box. J Virol 67, 5595–5604 (1993).
CAS PubMed PubMed Central Google Scholar
MacManiman, J. D. et al. Human cytomegalovirus-encoded pUL7 is a novel CEACAM1-like molecule responsible for promotion of angiogenesis. MBio 5, e02035 (2014).
Article PubMed PubMed Central CAS Google Scholar
Kalejta, R. F., Bechtel, J. T. & Shenk, T. Human cytomegalovirus pp71 stimulates cell cycle progression by inducing the proteasome-dependent degradation of the retinoblastoma family of tumor suppressors. Mol Cell Biol 23, 1885–1895 (2003).
Article CAS PubMed PubMed Central Google Scholar

Download references

Acknowledgements

We thank the International AIDS Vaccine Initiative (IAVI) for providing low passage MRC-5 cells and the GRIN plasmid. We thank Dr. Thomas Shenk, Princeton University, for recombinant AD169 and for anti-pp71 antibody. The following reagent was obtained through the NIH AIDS Reagent Program, Division of AIDS, NIAID, NIH: Anti-HIV-1 SF2 p24 Polyclonal. We also thank Dr. David Johnson, OHSU for anti-UL130 and Dr. Dong Yu, Washington University for TR-GFP. We are grateful to Jennie Womack for generating figures and graphics and to Dave Stein for help with the siRNA design. We acknowledge the generation of libraries for NextGen Illumina sequencing by the OHSU Massively Parallel Sequencing Shared Resource and NextGen sequencing on the Illumina MiSeq platform by the ONPRC Molecular Technologies Core. This work was supported by the Bill and Melinda Gates Foundation grants OPP1107409 and OPP1033121 to L.J.P. and K.F.; by National Institute of Allergy and Infectious Diseases grants R01 AI095113, P01 AI094417 and R37 AI054292 to L.J.P.; and by the National Institutes of Health Office of the director grant P51OD011092.

Author information

Sjoerd van den Worm
Present address: Batavia Biosciences B.V., Zernikedreef 16, 2333 CL, Leiden, Netherlands
Robert Ratts & Emily E. Marshall
Present address: Vir Biotechnology, 4640, SW Macadam Avenue, Portland, OR, 97239, USA
These authors contributed equally: Patrizia Caposio and Sjoerd van den Worm.

Authors and Affiliations

Vaccine and Gene Therapy Institute, Oregon Health & Science University, Beaverton, Oregon, 97006, USA
Patrizia Caposio, Sjoerd van den Worm, Lindsey Crawford, Wilma Perez, Craig Kreklywich, Roxanne M. Gilbride, Colette M. Hughes, Abigail B. Ventura, Robert Ratts, Emily E. Marshall, Daniel Malouli, Michael K. Axthelm, Daniel Streblow, Jay A. Nelson, Louis J. Picker, Scott G. Hansen & Klaus Früh

Authors

Patrizia Caposio
View author publications
You can also search for this author in PubMed Google Scholar
Sjoerd van den Worm
View author publications
You can also search for this author in PubMed Google Scholar
Lindsey Crawford
View author publications
You can also search for this author in PubMed Google Scholar
Wilma Perez
View author publications
You can also search for this author in PubMed Google Scholar
Craig Kreklywich
View author publications
You can also search for this author in PubMed Google Scholar
Roxanne M. Gilbride
View author publications
You can also search for this author in PubMed Google Scholar
Colette M. Hughes
View author publications
You can also search for this author in PubMed Google Scholar
Abigail B. Ventura
View author publications
You can also search for this author in PubMed Google Scholar
Robert Ratts
View author publications
You can also search for this author in PubMed Google Scholar
Emily E. Marshall
View author publications
You can also search for this author in PubMed Google Scholar
Daniel Malouli
View author publications
You can also search for this author in PubMed Google Scholar
Michael K. Axthelm
View author publications
You can also search for this author in PubMed Google Scholar
Daniel Streblow
View author publications
You can also search for this author in PubMed Google Scholar
Jay A. Nelson
View author publications
You can also search for this author in PubMed Google Scholar
Louis J. Picker
View author publications
You can also search for this author in PubMed Google Scholar
Scott G. Hansen
View author publications
You can also search for this author in PubMed Google Scholar
Klaus Früh
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

K.F., L.J.P., S.G.H., J.A.N. and P.C. designed and supervised various parts of this project. R.R., S.v.d.W., E.E.M., D.M. and W.P. were involved in vector construction and in vitro characterization. P.C. and L.C. performed the humanized mouse experiments. S.G.H., R.M.G., C.M.H., A.B.V. and M.K.A. performed the non-human primate experiments including monitoring of the T cell response. C.K. and D.S. measured antibody responses.

Corresponding authors

Correspondence to Scott G. Hansen or Klaus Früh.

Ethics declarations

Competing interests

O.H.S.U. and P.C., S.G.H., W.P., R.R., E.E.M., D.M., J.A.N., L.J.P. and K.F. have financial interests in Vir Biotechnology, Inc., a company that may have a commercial interest in the results of this research and technology. This potential conflict of interest has been reviewed and managed by O.H.S.U. All other authors declare no conflict of interest.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplemental information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Caposio, P., van den Worm, S., Crawford, L. et al. Characterization of a live-attenuated HCMV-based vaccine platform. Sci Rep 9, 19236 (2019). https://doi.org/10.1038/s41598-019-55508-w

Download citation

Received: 22 August 2019
Accepted: 29 November 2019
Published: 17 December 2019
DOI: https://doi.org/10.1038/s41598-019-55508-w

This article is cited by

Alternative splicing and genetic variation of mhc-e: implications for rhesus cytomegalovirus-based vaccines
- Hayden Brochu
- Ruihan Wang
- Xinxia Peng
Communications Biology (2022)

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.