Date Published: February 6, 2014
Publisher: Public Library of Science
Author(s): Kai A. Kropp, Ana Angulo, Peter Ghazal, Glenn F. Rall.
The inflammatory milieu is the natural habitat for a pathogenic infection, characterised by activity of pro-inflammatory signalling pathways and inflammatory cytokines. Viral entry rapidly activates a range of innate-immune signalling events such as the activation of Pattern Recognition Receptors (PRRs) –. A virus must therefore counteract intrinsic cellular and innate-immune responses to successfully complete the replication cycle. Frequently this is accomplished by encoding viral effector molecules that block these cellular responses by working as either structural or functional mimics of host target proteins –. Nuclear DNA viruses are dependent on the host transcriptional machinery to express the first viral genes; for example the immediate-early (IE) control elements of DNA viruses are by definition absolutely dependent on host transcription factors (TF) . Therefore, these viruses are particularly hostage to their host transcriptional environment , . Here we propose that mimicry of regulatory DNA sequences by viral regulatory regions may also provide an additional strategy to counteract at IE times of infection the innate-immune response. In this context, viral IE control elements might functionally mimic innate-immune enhancers, taking advantage of the activated immune signalling TFs for promoting viral IE gene expression.
To investigate if there is any similarity of primary sequences and therefore structural mimicry between the selected viral and cellular enhancers, we used the BLAST tool to compare the sequences against each other (Table 1) and applied an exhaustive pairwise multi-way alignment (CloneManager suite 7.0) to search for similarities in this group of sequences (Figure 1A). While multi-way alignment of the various selected viral and cellular promoter-regulatory regions (Figure 1A, top panel) reveals a lack of extended primary sequence homology, the pairwise BLAST comparison showed that small islands of sequence identity or high similarity are present (Table 1). We randomly compared some of these short sequence motifs with the JASPAR CORE (Vertebrae) database  and found that all checked motifs have similarities with consensus binding motifs for TFs (e.g., AP1, SP1, YY1, or RelA with relative scores of >0.8). This finding raises the question of whether there might be functional similarity. We therefore consider in the next section how convergent evolution of viral enhancers may have resulted in functional mimicry of the transcription control elements of innate-immune genes, providing a co-opting strategy for immune evasion.
There are two principal genetic mechanisms that could lead to viral mimicry of host enhancers, horizontal transfer of cellular sequences to viral genomes or genetic drift of viral sequences. The first possibility, acquisition of cellular sequences through horizontal sequence transfer, could arise through illegitimate recombination with host DNA, for example by retro-transposition of non-coding RNA transcripts, resulting in the virus hijacking host transcription control sequences. If this were the general case, we would, however, expect significant structural similarity, which we did not find in our analysis. Alternatively, but not mutually exclusive from horizontal transfer, viral enhancer mimics could arise through neutral evolution and genetic drift by sequence duplication or accumulation of point mutations. Duplicated sequence features are hallmarks for many viral and cellular enhancers –. For instance, deletion or loss of enhancer sequences in SV40 and JC polyomavirus promotes restoration of enhancer function through duplication of flanking sequences –. A third possibility is the accumulation of point mutations in enhancer sequences and subsequent fixation . It has recently been described for a wide range of species that evolution of host-cell transcriptional control can occur in relatively short time spans and is mainly driven by the rapid and flexible emergence or loss of binding motifs rather than by evolution of the TF proteins themselves –. The described mechanisms of rapid enhancer evolution argue that viral enhancers could acquire functionality that mimics innate-immune enhancers without any extensive sequence homology, and this is consistent with the comparison of cellular and viral enhancers shown in Figure 1A. This possibility is underscored by the fact that promoter sequences seem to be poorly conserved even among members within a virus-family yet share many of the same regulatory elements . For example the MIE enhancers of cytomegaloviruses show low levels of primary sequence similarity between the different species strains (Figure 1A, lower panel). Despite these differences, functionality of the enhancers is conserved between hosts for different CMV species strains, e.g., the human CMV enhancer can functionally complement deletion of the murine CMV enhancer  and human CMV enhancer sequences recapitulate in vivo biological sites of infection in species from mice to zebra fish –.
Since our work and that of others discussed so far indicates that viral enhancers are functional rather than structural mimics of host innate-immune enhancers, we suggest four principal hallmarks of functional enhancer mimicry. These are: 1) shared TF interactions independent of sequence structure, 2) similar kinetics of gene induction between cellular innate-immune and viral IE genes, 3) positive responsiveness to immune-stimulatory ligands, and 4) susceptibility to inhibition of inflammatory signalling. In the following section we briefly discuss these hallmarks.
TFs activating innate-immune genes are regulated by PRR signalling that cannot be efficiently inhibited by viruses as their activation occurs during the viral entry process. Mimicking an innate-immune enhancer therefore has the advantage that TFs, already activated by the viral entry process, can be directly utilised in a time restricted manner to ensure viral gene expression at IE times. We hope this opinion opens debate and provides new insights for either reexamination or future-based investigations toward understanding viral gene activation and latency. Indeed we believe that the principle of viruses co-opting host-innate regulatory signals has broad implications toward understanding the biological role of viral enhancers, in acute and latent viral infections, and prospective host-directed antiviral therapeutic and vaccine strategies.