Editorial: Quality versus quantity in myeloid infection by a herpesvirus: more than one way to skin the CCAAT?

Viruses are obligate intracellular pathogens that subvert their host cells to support persistence and replication. In this issue of Journal of Leukocyte Biology, Qin and colleagues [1] demonstrate that KSHV expresses miRNAs, which induce IL-6 and IL-10 secretion in infected monocytes and macrophages. The miRNAs specifically inhibit expression of the LIP isoform of C/EBP-β, a transcriptional repressor protein. This paper thus reports a novel mechanism for dysregulating cellular cytokine expression with important implications for pathogenesis of cancers associated with KSHV infection.

The seminal example for viral subversion of host growth control is inactivation of the host protein pRb by proteins expressed from three small DNA tumor viruses: SV40, HPV, and adenovirus [2] (Table 1). Inactivation of pRb relieves its inhibition of the host transcription factor E2F, leading to S-phase entry of the host cell and activation of E2F-dependent viral promoters. The consequence of inhibiting E2F is to optimize the cellular environment for viral replication. In uninfected cells, pRb is physiologically inactivated by cyclin-dependent phosphorylation, and inactivating mutations of pRb contribute to human cancers. Thus, the fundamental discovery that small DNA tumor viruses inactivate pRb has permitted virology to proceed hand-in-hand with cell biology.

TABLE 1.

Cellular Targets of Viral Subversion Molecules

Cellular target	Viral dysregulator	Type of dysregulation	Effect on virus	Effect on host	References
LIP	KSHV miR-K12-3	Antagonists	?	Increased secretion of IL-6 and IL-10	[1]
	KSHV miR-K12-7
pRb	SV40 Large T Ag	Antagonists	Promotes transcription	Stimulates G1-to-S cell-cycle transition	[2]
	HPV E7
	Adenovirus E1A
MICB	KSHV miR-K12-7	Antagonists	Escapes NK detection	Reduces NK cell killing	[4]
	KSHV K5 protein
	HCMV miR-UL112
	HCMV UL16 protein
	EBV miR-BART2-5p
BCLAF	KSHV miR-K5	Antagonist	Potentiates reactivation	Stimulates apoptosis	[5]
BACH1	KSHV miR-K12-11	Antagonist	?	Transcriptional derepression?	[6, 7]
THBS1	KSHV miR-K12-1	Antagonists	?	Reduced TGF-β activity stimulates cell growth?	[8]
	KSHV miR-K12-3-3p
	KSHV miR-K12-6-3p
	KSHV miR-K12-11
miR155	KSHV miR-K12-11	Mimic	?	Stimulates cell growth	[6, 7, 9]
	MDV-1 mir-M4
IL-6	KSHV vIL-6	Mimic	?	Inhibits apoptosis stimulates cell growth, VEGF secretion, hematopoiesis, angiogenesis	[10]

Ag, antigen; BACH1, BRCA1-associated C-terminal helicase; VEGF, vascular endothelial growth factor. 

The recent explosion of biologic sequence information has facilitated bioinformatic approaches to investigating other host–virus interactions, particularly of KSHV (also known as human herpesvirus-8) [10]. KSHV is the etiologic agent of three hyperproliferative disorders: Kaposi’s sarcoma, an endothelial neoplasm; primary effusion lymphoma, a non-Hodgkin’s B cell lymphoma; and some forms of the lymphoproliferative disorder multicentric Castleman’s disease. Critical components of the pathogenesis of all three diseases are dysregulation of autocrine and paracrine cytokines and chemokines. The prevalence and aggressiveness of all three disorders are increased in immunocompromised hosts, similar to other herpesviruses. KSHV encodes at least 13 homologs of cellular genes that appear to have been captured or “pirated” by the virus from the host [3]. The human counterparts of these genes function as growth-regulating transcription factors, antiapoptotic proteins, chemokines, cytokines, signal transduction receptors, or immune-evasion factors [10]. Further studies showed that these cell-homologous proteins have evolved “resistance” to cellular, physiologic controls of their activities. Thus, one mechanism by which KSHV subverts normal host physiology is to express constitutively active viral protein mimics of host-cell proteins.

Qin and colleagues’ paper [1] highlights an emerging mechanism of host-cell subversion by viruses: expression of viral miRNAs, which are short (∼22 nt), single-stranded RNAs that bind sequence specifically to mRNAs to inhibit their translation or reduce their abundance. KSHV expresses 17 miRNAs from 12 pre-miRNAs, and the authors demonstrate that the effect is mediated primarily by two of them: miR-K12-3 and miR-K12-7. Furthermore, miRs-K1, -9, and -11 cooperate with -3 and -7 to induce IL-6 and -10 but have no effect individually. Importantly, Qin et al. [1] demonstrate these effects by ectopically expressing the viral miRNAs and by inhibiting specific miRNAs with antagomirs in infected cells. This publication describes the first example of dysregulating cytokine expression with a viral miRNA.

The authors also demonstrate the mechanism by which miR-K12-3 and miR-K12-7 subvert macrophage physiology [1]. Using bioinformatics, Qin and colleagues [1] identify potential target sites for both miRNAs in the 3′-untranslated region of the cellular gene C/EBP-β (Fig. 1A). The C/EBP-β mRNA expresses three protein isoforms by alternative translational start sites; remarkably, accumulation of only one of the isoforms is reduced by expression of KSHV miR-K12-3 and miR-K12-7 (Fig. 1A). The reduced isoform is called LIP and has been characterized previously as a dominant-negative transcriptional repressor of other C/EBP isoforms (Fig. 1B). As C/EBP-β has been implicated previously in regulating transcription of IL-6 and -10, the paper suggests that under physiologic conditions in macrophages, LIP represses IL-6 and -10 secretion by inhibiting C/EBP-β transactivation (Fig. 1B). Qin and colleagues [1] show that infection by KSHV induces IL-6 and -10, in part, by inactivating LIP through translational inhibition mediated by viral miRNAs.

Figure 1.

KSHV miRNAs stimulate secretion of IL-6 and IL-10. (A) KSHV miR-K12-3 and -12-7 inhibit expression of the protein LIP. C/EBP-β forms homodimers that transactivate the IL-6 and IL-10 promoters. (B) In uninfected myeloid cells, LIP forms inactive homodimers or heterodimers with C/EBP-β, which are unable to transactivate the IL-6 and IL-10 promoters. AD, Transcriptional activation domain.

Why might KSHV have evolved to derepress C/EBP transactivation? It is known that efficient KSHV gene expression requires C/EBP [10], so it is likely that KSHV inhibits LIP to support efficient viral replication. However, inhibiting LIP can be considered in the broader context of consequences for the infected host: IL-6 and IL-10 contribute to the pathogenesis of KSHV cancers. Both cytokines have roles in promoting growth of tumor cells, enhancing angiogenesis, suppressing activation of T cells, and inhibiting maturation of dendritic cells. The paper thus implicates KSHV-infected macrophages as a source of these cytokines [1]. Interestingly, the authors show that the miRNAs also induce receptors for both cytokines (IL-6Rα and IL-10R1), suggesting that the virus sets up an autocrine loop in infected cells. Furthermore, the literature shows that KSHV has conserved many additional genes that contribute to IL-6 expression and secretion, including expressing a promiscuous homologue of IL-6, named “vIL-6” [10]. Therefore, a likely reason for inhibiting LIP is that LIP is a node that controls expression of IL-6, IL-10, and possibly other cytokines. It follows that expression of these cytokines establishes a permissive cellular microenvironment for KSHV persistence and replication. The more insidious consequence is that this microenvironment can promote cancer, especially in the context of immunosuppression.

The picture that has emerged during the last few years is that herpesviruses have evolved to express self-encoded miRNAs as previously unrecognized mechanisms to subvert the host [9]. More than 50 viral mRNAs have been identified in five human herpesviruses, which establish latent (nonproductive) infections for the lifetime of their host, a strategy for persistence particularly dependent on promoting survival of the infected cell and avoiding immune elimination [10].

Identifying the cellular targets of viral miRNAs has not only revealed target redundancy within the miRNAs but also between miRNAs and proteins. Individual viruses use more than one gene product to subvert the same cellular proteins, and different viruses target identical proteins. For example, as discussed above, IL-6 signaling is activated by multiple mechanisms in KSHV infection. In another example, NK cell surveillance is inhibited by not only KSHV miRNA-mediated repression of MICB but also by a KSHV protein that inhibits cell-surface expression of MICB [4, 10]. MICB is also targeted by miRNAs expressed by two other herpesviruses—HCMV and EBV—resulting in inhibiting NK surveillance [4] (Table 1). Thus, both viral proteins and miRNAs can function as antagonists of cellular physiologic pathways.

Viruses also subvert the host by expressing miRNAs that mimic cellular molecules. For example, the viral target MICB is regulated by a cellular miRNA called miR93 [9]. The oncogenic cellular miRNA, miR-155, is mimicked by miRNAs expressed by KSHV and the oncogenic chicken herpesvirus, MDV-1 [6, 7, 9] (Table 1). Mimicry of miR-155 by two different herpesviruses suggests that miR-155 regulates a cellular pathway essential for viral replication. However, it is currently unclear why subversion of host growth control is targeted by viral miRNAs in some cases but viral proteins in others.

Additional examples show that herpesviruses subvert different targets with opposing effects on the host. For example, KSHV miRNAs induce apoptosis by inhibiting BCLAF but inhibit the tumor suppressor THBS1 [5, 8] (Table 1). KSHV also expresses eight proteins that inhibit apoptosis, apparently to ensure survival of the infected cell [10].

The recognition of miRNA-based strategies for viral subversion of the host has vastly increased the complexity of predicting consequences for viral replication and pathogenesis. For example, each KSHV miRNA has additional targets, demonstrated with less detail than the examples above or predicted bioinformatically [5,6,7,8]. Current bioinformatic approaches to predict mRNA targets of miRNAs are driven mostly by identifying potential binding sites for the miRNA “seed” sequence (nt 2–8). However, the current set of targets for each viral miRNA is probably underestimated: Target recognition that depends on nonseed sequences is not well-understood, and full complementarity of the seed and target is not essential for function of the miRNA. This is perhaps demonstrated most clearly by the observation that multiple viral miRNAs with nonidentical seed sequences cooperate to subvert common mRNA targets [5,6,7,8] (Table 1).

It is clear that cellular targets or pathways with multiple viral antagonists are key regulators of cellular physiology and critical targets for subversion by viruses. It is currently less obvious whether the cellular targets with fewer viral modulators are subtle effectors required for specific viruses or are downstream of the widely targeted nodes but are subverted in a virus-specific manner. These observations raise a number of questions. Do herpesviruses use miRNAs in subversion mechanisms only because they are not immunogenic, or do miRNAs provide a qualitative or quantitative mechanism for regulating the cell that differs from protein-based subversion? As additional targets of the viral miRNAs are discovered, will we be able to predict consequences for viral replication and pathogenesis based on how common the targets are? Are the less-common viral targets better candidates for directed antiviral therapies than the more common viral targets? Are targets identified by studying viral infections suitable for therapies directed against diseases with nonviral etiologies?

These questions might be answered by further study of C/EBP transactivation and IL-6 regulation. Qin and colleagues’ work [1] is the first description of LIP as a herpesviral target for subversion. By inhibiting the C/EBP repressor LIP, this mechanism for host-cell subversion seems to fit the paradigm established for small DNA tumor viruses: inhibiting a repressor protein whose cellular targets are required for efficient viral replication. Will other viruses also target LIP, suggesting that it’s a critical node, or does targeting LIP simply represent another mechanism by which to induce or maintain elevated IL-6? Is redundancy required so that different subversion mechanisms for IL-6 regulation can be used in latency versus reactivation? A better understanding of the basic biochemistry of miRNA target selection may facilitate identification of additional targets of miR-K12-3 and miR-K12-7 and additional subverters of LIP.