STAT3 regulates antiviral immunity by suppressing excessive interferon signaling.

This study identifies interleukin-6 (IL-6)-independent phosphorylation of STAT3 Y705 at the early stage of infection with several viruses, including influenza A virus (IAV). Such activation of STAT3 is dependent on the retinoic acid-induced gene I/mitochondrial antiviral-signaling protein/spleen tyrosine kinase (RIG-I/MAVS/Syk) axis and critical for antiviral immunity. We generate STAT3Y705F/+ knockin mice that display a remarkably suppressed antiviral response to IAV infection, as evidenced by impaired expression of several antiviral genes, severe lung tissue injury, and poor survival compared with wild-type animals. Mechanistically, STAT3 Y705 phosphorylation restrains IAV pathogenesis by repressing excessive production of interferons (IFNs). Blocking phosphorylation significantly augments the expression of type I and III IFNs, potentiating the virulence of IAV in mice. Importantly, knockout of IFNAR1 or IFNLR1 in STAT3Y705F/+ mice protects the animals from lung injury and reduces viral load. The results indicate that activation of STAT3 by Y705 phosphorylation is vital for establishment of effective antiviral immunity by suppressing excessive IFN signaling induced by viral infection.

Initial activation of STAT2 induced by IAV infection is critical for innate antiviral immunity.

STAT2 is an important transcription factor activated by interferons (IFNs) upon viral infection and plays a key role in antiviral responses. Interestingly, here we found that phosphorylation of STAT2 could be induced by several viruses at early infection stage, including influenza A virus (IAV), and such initial activation of STAT2 was independent of type I IFNs and JAK kinases. Furthermore, it was observed that the early activation of STAT2 during viral infection was mainly regulated by the RIG-I/MAVS-dependent pathway. Disruption of STAT2 phosphorylation at Tyr690 restrained antiviral response, as silencing STAT2 or blocking STAT2 Y690 phosphorylation suppressed the expression of several interferon-stimulated genes (ISGs), thereby facilitating viral replication. In vitro experiments using overexpression system or kinase inhibitors showed that several kinases including MAPK12 and Syk were involved in regulation of the early phosphorylation of STAT2 triggered by IAV infection. Moreover, when MAPK12 kinase was inhibited, expression of several ISGs was clearly decreased in cells infected with IAV at the early infection stage. Accordingly, inhibition of MAPK12 accelerated the replication of influenza virus in host. These results provide a better understanding of how initial activation of STAT2 and the early antiviral responses are induced by the viral infection.

A parsimonious personalized dose-finding model via dimension reduction.

Learning an individualized dose rule in personalized medicine is a challenging statistical problem. Existing methods often suffer from the curse of dimensionality, especially when the decision function is estimated nonparametrically. To tackle this problem, we propose a dimension reduction framework that effectively reduces the estimation to a lower-dimensional subspace of the covariates. We exploit that the individualized dose rule can be defined in a subspace spanned by a few linear combinations of the covariates, leading to a more parsimonious model. Also, our framework does not require the inverse probability of the propensity score under observational studies due to a direct maximization of the value function. This distinguishes us from the outcome weighted learning framework, which also solves decision rules directly. Under the same framework, we further propose a pseudo-direct learning approach focuses more on estimating the dimensionality-reduced subspace of the treatment outcome. Parameters in both approaches can be estimated efficiently using an orthogonality constrained optimization algorithm on the Stiefel manifold. Under mild regularity assumptions, the asymptotic normality results of the proposed estimators can are established, respectively. We also derive the consistency and convergence rate for the value function under the estimated optimal dose rule. We evaluate the performance of the proposed approaches through extensive simulation studies and a warfarin pharmacogenetic dataset.