What does a given cell compartment actually contribute to infection and immunity?
Asking what one cell type contributes to an infection is normally done by removing the cell type or by deleting a gene it needs, and both perturbations remove far more than the infection. A depleted dendritic cell compartment is not an animal in which dendritic cells are present but uninfected. Langlois 2012 in PNAS states the limitation in those terms, noting that the available tools delete host genes or host cells and so cannot ask what a single infected compartment contributes while leaving the rest of the animal intact.
Detection has the mirror problem. Pham 2012 sets out the difficulty for dengue virus, where viral antigen or RNA in a tissue can reflect replication, or entry without replication, or uptake of virus or debris by a bystander, and where reports of endothelial infection had been questioned on those grounds. A third version appears in herpesvirus genetics. Møller 2018 notes that cytomegalovirus recombinants must be rescued in fibroblasts, so a gene essential for acute fibroblast infection cannot be deleted even when its function in the myeloid compartment is the question.
All three are one problem stated from different sides. The variable of interest is viral replication in a defined cell type, and no conventional genetic tool isolates it.
The contribution is to make viral tropism an experimental variable set independently of host genotype, so that one animal carries both a permissive and a nonpermissive compartment and the contribution of one can be read against an otherwise identical infection.
Langlois 2012 in PNAS is the clearest execution. Four perfect miR-142 sites were placed in a duplicated packaging region of the influenza nucleoprotein segment, giving a virus silenced in hematopoietic cells and unaffected in epithelium. Silencing was verified at the level of small viral RNA, cRNA, mRNA, nucleoprotein protein, multicycle growth, surface hemagglutinin on lung CD45-positive CD11c-positive cells, and infectious virus in the lung draining lymph node, and escape was excluded by sequencing. That validation chain is itself a contribution, because it sets what such a virus must demonstrate before any phenotype is interpretable.
The result splits an expectation. Mice infected with the targeted virus lost weight, cleared virus, and generated nucleoprotein- and polymerase acidic-specific CD8 T cells indistinguishably from controls, even though direct presentation was lost in a miR-142-expressing antigen presenting cell line, which the authors read as sufficiency of cross-presentation. What changed instead was innate signalling, with interferon beta and IRF-7 induction reduced in macrophages and in whole lung. The authors treat the whole lung deficit as the central result and state plainly that they cannot explain, from these data, why closing a numerically minor compartment produces so large a loss of total interferon. The connection between the lost interferon and RIG-I in hematopoietic cells is inferred, since the sensor requirement was established in cultured primary cells rather than in a hematopoietic-restricted knockout animal.
Pham 2012 runs the same subtraction against dissemination. In interferon receptor deficient mice the miR-142-targeted dengue virus lost replication in sorted CD11b-positive, CD11c-positive and CD45-positive splenocytes and lost roughly three logs of titre in spleen and liver by three routes. The strongest evidence is the escape sequencing, since a population that has entirely excised its targeting cassette is hard to reconcile with a productive alternative reservoir in which the intact virus could have grown. The authors phrase the conclusion conditionally, stating that either non-hematopoietic cells are not productively infected or that macrophages and dendritic cells are required for spread, and the data constrain dissemination rather than intrinsic permissiveness of every other cell type.
Møller 2018 turns the principle into a conditional genetics tool and extends it to a large DNA virus. Four miR-142 sites in the noncoding 3-prime untranslated region of cytomegalovirus IE2 were installed in the TB40/E bacterial artificial chromosome by a single galK and loxP recombination step, leaving the open reading frame untouched. The virus grew normally in fibroblasts, which lack miR-142, and IE2 was silenced in THP-1-derived macrophages. Loss of IE2 did not abolish replication there but sustained it two to three logs above a control that declined over ten days, with IE1 strongly overproduced through relief of cis-repression, and with more than 750 host genes changed in macrophages against roughly fifty in miR-142-expressing fibroblasts. Essentiality determined in fibroblasts does not transfer to the myeloid lineage.
The method moves from attenuation to interrogation to conditional genetics. Pham 2012 and Langlois 2012 in PNAS appeared in the same year and use the same microRNA, and neither is derivative of the other so much as two applications of one construct logic to two questions, dissemination and immune induction. Møller 2018 changes what is being asked. In 2012 the targeted gene is essential and the question is what the compartment does. In 2018 the targeted gene is a viral regulator and the question is what that gene does in a compartment where it could not previously be studied.
A limitation runs through all three and none of them overcomes it. MicroRNA silencing is potent but incomplete, and it falls off as target abundance falls. Møller 2018 says explicitly that residual near-undetectable IE2 cannot be excluded as sufficient for an early essential step, so the macrophage phenotype is a strong lineage-restricted knockdown and not a null. Because miR-142 is expressed across hematopoietic lineages, none of the three papers can assign a phenotype to a single subset.
Langlois 2012 in PNAS, Pham 2012 and Møller 2018 are all lab-led. Møller 2018 is co-senior with Domenico Tortorella, whose group supplied the cytomegalovirus rescue procedure.
The construct logic comes from Perez 2009 and is treated in the MicroRNA-Mediated Viral Attenuation theme. The interferon result in Langlois 2012 in PNAS belongs jointly to the Calibration of the Interferon Response In Vivo theme in the Innate Immune Signaling and the Interferon Response area. Heaton 2014, in the Lineage Tracing of Infected Cells theme, addresses a related question with the opposite design, marking infected cells permanently rather than excluding infection from a compartment.
A one-step recombineering strategy that inserts hematopoietic-specific miR-142 target sites into the untranslated region of the human cytomegalovirus IE2 transcript permits virus rescue in fibroblasts while silencing IE2 selectively in myeloid cells, revealing that IE2 loss raises rather than abolishes replication in macrophages.
Influenza A virus engineered to be silenced by the hematopoietic microRNA miR-142 shows that replication inside antigen presenting cells is dispensable for CD8 T cell priming and clearance but required for much of the RIG-I-dependent type I interferon response in vivo.
Inserting hematopoietic-specific miR-142 target sites into the dengue virus 3-prime untranslated region excludes replication from macrophages and dendritic cells in vivo and abolishes spread to other tissues, identifying hematopoietic cells as the dominant amplification compartment.