Current address: Department of Microbiology and Immunobiology, Harvard Medical School, Boston, Massachusetts, United States of America
Conceived and designed the experiments: GK TPR MR PJMR CAMH LM. Performed the experiments: GK TPR MR GCMG FEC. Analyzed the data: GK TPR MR PJMR CAMH LM. Contributed reagents/materials/analysis tools: MER FEC. Wrote the paper: GK TPR MR CAMH LM.
The authors have declared that no competing interests exist.
Immunological checkpoints, such as the inhibitory CD200 receptor (CD200R), play a dual role in balancing the immune system during microbial infection. On the one hand these inhibitory signals prevent excessive immune mediated pathology but on the other hand they may impair clearance of the pathogen. We studied the influence of the inhibitory CD200-CD200R axis on clearance and pathology in two different virus infection models. We find that lack of CD200R signaling strongly enhances type I interferon (IFN) production and viral clearance and improves the outcome of mouse hepatitis corona virus (MHV) infection, particularly in female mice. MHV clearance is known to be dependent on Toll like receptor 7 (TLR7)-mediated type I IFN production and sex differences in TLR7 responses previously have been reported for humans. We therefore hypothesize that CD200R ligation suppresses TLR7 responses and that release of this inhibition enlarges sex differences in TLR7 signaling. This hypothesis is supported by our findings that
Immune responses need to be carefully orchestrated to prevent disease due to an overactive immune system. Immunological checkpoints are provided by immune inhibitory receptors, which set a threshold for activation and dampen the immune system. In the case of a viral infection, this prevents pathology induced by the immune system, but on the other hand may prevent adequate removal of the virus. In this paper, we show that removal of such an immunological checkpoint in mice leads to rapid removal of corona virus, but also to more immune-induced disease symptoms in case of influenza virus infection. We observe this predominantly in female mice. We demonstrate that this particular checkpoint inhibits anti-viral responses that are naturally stronger in females. Release of this checkpoint enlarges these sex differences. Our findings have major implications for therapeutic use of blockers of this pathway, which are currently in clinical trials for the treatment of cancer, as we predict that female patients will have a stronger response to such therapeutics.
To generate an appropriately controlled response during infections, the immune system is balanced by the action of activating and inhibitory receptors. Lack of inhibition leads to excessive inflammation and autoimmunity and other severe disease symptoms. One of the receptors regulating this balance is CD200 Receptor (CD200R)
The signal that triggers CD200R and results in delivery of an inhibitory intracellular signal to the cell is given by its ligand CD200, which has a short intracellular tail devoid of any known signaling motifs. CD200 is expressed on thymocytes, activated T cells, B cells, dendritic cells (DCs), vascular endothelial cells, hair follicular cells, in the central nervous system and in the retina (reviewed in
Mouse hepatitis coronavirus (MHV) is an accepted model for the most illustrious coronavirus (CoV): severe acute respiratory syndrome (SARS)-CoV. Host control of MHV infection is completely dependent on an immediate type I IFN response, initiated upon TLR7 triggering by viral RNA. Mice lacking this pathway show massive MHV replication and fatal infection within a few days
We here report that lack of CD200R signaling has a more profound effect on the beneficial but also on the pathological immune responses to viruses in female mice as compared to male mice, which can be attributed to the capacity of CD200R to inhibit TLR7 responses.
To determine the role of CD200-CD200R signaling in CoV infection, we intraperitoneally inoculated male and female wild type (WT) and
Clearance of MHV critically depends on TLR7-mediated type I IFN production by hematopoietic cells
Sex differences in TLR7 responses have previously been reported for humans
We next tested whether CD200R-mediated signaling directly inhibits signals transduced via TLR7. We generated a chimeric construct containing a LAIR-1 receptor in which the intracellular tail was replaced by that of CD200R. This allows for efficient cross-linking using anti-LAIR-1 antibodies to induce signaling via the CD200R cytoplasmic tail. We transfected HEK 293 cells with plasmids encoding a LAIR-1-CD200R chimeric receptor, human TLR7 and a luciferase reporter under control of a NF-κB driven promoter. Cross-linking of the chimeric receptor by anti-LAIR-1 antibody, but not by isotype-matched control antibody, resulted in robust inhibition of imiquimod-induced NF-κB activity (
A strong anti-viral response can also cause immune mediated pathology that can be detrimental to the host. We therefore moved to a virus infection model in which immune pathology is known to be important for clinical outcome. Upon intranasal infection with influenza A virus we again observed a sex bias in the viral load, measured in the lungs at day 8 post infection (
Naïve or influenza A virus-infected mice were sampled at indicated time points.
Confirming previous reports, we found a significantly enhanced body weight loss in female
In agreement with the increased neutrophil counts we measured elevated levels of KC (IL-8) in the BAL fluid (
Thus, from two different viral infection models we can conclude that sex has a profound effect on type I IFN production and viral clearance. This study is the first to report a significantly enhanced viral clearance in female mice due to a sex bias in TLR7 responses. Sex differences in TLR7 induced type I IFN production have previously been reported for humans
We demonstrate direct inhibition of TLR7 signaling through CD200R. Previously, CD200R-mediated inhibition of LPS-induced cytokine production was reported
Upon influenza A virus infection, CD200-deficiency strongly enhances neutrophil influx into the lungs of female mice possibly leading to pathology, but it does not affect viral clearance and type I IFN production. This implies that, for influenza virus, the sex-biased type I IFN production and viral clearance are not regulated by CD200R, while the events leading to increased neutrophil recruitment and lung pathology are. Neutrophil responses to influenza virus infection were shown to be dependent on TLR7
There is emerging evidence that tumor cells employ immunological checkpoints for their benefit. As a result of this, inhibitory immune pathways have become therapeutic targets to strengthen anti-tumor responses and develop (adjuvant) therapeutic strategies in cancer treatment. The successful application of anti-CTLA4 (Cytotoxic T-Lymphocyte Antigen 4) in melanoma is followed up with blocking agents for other checkpoints, among which the CD200-CD200R immune inhibitory pathway. Strong evidence for a role for CD200 in tumor progression comes from studies in patients. Expression of CD200 is an independent prognostic factor for multiple myeloma and acute myeloid leukemia predicting worse overall survival of these patients
Our finding that the combination of lack of CD200R signaling and female sex has such a profound impact on the control of virus infection as well as on immune pathology raises some important issues. We are the first to demonstrate a strong sex bias in type I IFN production and viral clearance in mice utilizing two different models of virus infection. This is of importance for scientists studying these widely used models and may result in a completely different interpretation of data obtained, depending on the sex of the mice used. Moreover, sex biased clinical responses to virus infections have been reported in humans
Wild-type C57BL/6J mice and
In additional experiments we injected the mice intraperitoneally with the TLR7 agonist imiquimod (Invivogen; 50 µg in 200 µl PBS). The Utrecht University Ethical Committee for Animal Experimentation approved the animal study protocols, in accordance with the advice of the Central Committee on Animal Experimentation (20 januari 1997) and the Dutch Law on Animal Experimentation (art. 18a).
After MHV-EFLM injection (day 0), mice were imaged at day 2 and day 4 as described previously
Whole livers or left lungs were dissected from the mice. The tissues were processed in Lysing Matrix D tubes (MP Biomedical), containing 1 ml of PBS, using a FastPrep instrument (MP Biomedical). The tissues were homogenized at 3300× g for 40 sec and immediately placed on ice. Subsequently, the homogenates were centrifuged at 18600× g for 10 minutes at 4°C and supernatants were harvested and stored at −80°C. Total RNA was isolated from the homogenates using the TRIzol reagent (Invitrogen) according to manufacturer's instructions.
Gene expression levels of IFN-α, IFN-β1, and TLR7 were measured by quantitative PCR using LightCycler 480 RNA Master Hydrolysis Probes in combination with a LightCycler 480 system (both from Roche Applied Science), according to the manufacturer's instructions. The housekeeping gene GAPDH was used as a reference in all experiments, and expression of this gene was found relatively constant among samples. The amounts of MHV RNA were determined by quantitative RT-PCR using primers and probe directed against the N gene of MHV-A59
Bronchoalveolar lavage (BAL) fluid was obtained by flushing the lungs two times with 1 ml PBS using a canula inserted into the trachea, yielding around 1.7 ml BAL fluid. Pelleted cells from BAL fluid were counted and cytospins were prepared and stained with May-Grunwald/Giemsa and neutrophils were scored on the basis of morphology (Dade Behring, Switzerland). BAL fluids were kept on ice or stored at −80°C until further processing. BAL fluid was centrifuged, and 20 µl of aliquot was used to determine the protein concentration with a BCA kit (Pierce) according to the manufacturer's instructions.
To measure the interferon concentration in the sera, blood was sampled from naïve mice or four days after MHV infection or one hour after imiquimod treatment. Sera were separated by spinning the blood at 2300× g for 15 minutes at 4°C. For measurement of cytokines in BAL, samples were prepared by spinning 5 minutes at 530× g. IFN-α was measured with a mouse interferon alpha ELISA kit (PBL Interferon Source). For the IL-6, IL-8 and TNF-alpha measured by Mouse IL-6 Mni ELISA Development Kit, Murine KC (IL-8) ELISA Development Kit and Murine TNF-alpha Mini ELISA Development Kit (PeproTech) respectively. Experiments were done according to manufacturer's instructions.
Livers of MHV-infected mice were sampled, fixed in 4% neutral buffered formalin, and embedded in paraffin. Seven µm liver sections were stained with hematoxylin and eosin. Total liver sections were examined by light microscopy and foci of hepatocellular necrosis and inflammation were scored in a semi-quantitative manner.
HEK 293 T cells were transiently co-transfected with: human TLR7 (kindly provided by Rogier Sanders, AMC, Amsterdam, the Netherlands), and NF-κB-reporter or IFNα-reporter constructs, kindly provided by Dr Paul Moynagh (National University of Ireland). A chimeric construct containing the extra cellular region of human LAIR-1 (amino acids 1–160) fused with the transmembrane and intracellular rat CD200R (rCD200R) (amino acids 236–327) was cloned into pcDNA3.1/zeo (Invitrogen, Breda, the Netherlands). A tyrosine (Y) to phenylalanine (F) mutant of tyrosines 287, 290, and 298 in the intracellular rCD200R tail were generated with PCR-based mutagenesis. The mutant was cloned into the same vector and all sequences were confirmed by automated DNA sequencing. The LAIR-CD200R plasmid was co-transfected with the TLR7 and reporter constructs. Twenty-four hours after transfection cells were trypsinized and seeded in 48-well plates coated with 3 ug/ml anti-LAIR-1 monoclonal antibody (clone 8A8). Forty-eight hrs after transfection cells were stimulated with imiquimod 3 µg/ml (Invivogen) in PBS. On the next day, cells were lysed with Passive Lysis Buffer (Promega), luciferase activity was measured on a luminometer (Berthold technologies Centro LB 960), and data were analyzed with Microwin software. Total protein content was determined with a Pierce BCA Protein Assay (Thermo Scientific). All luciferase values were normalized to protein concentration. Alternatively, we used HEK 293 cells stably expressing human TLR7 (Invivogen).
Significance was calculated with Mann-Whitney test using GraphPad Prism software.
(DOC)
(DOC)
(DOC)
We thank Hans Clevers, Lewis Lanier, Louis Bont, José Borghans and Erik Hack for critical discussions and review of versions of the manuscript. We also thank Eva Rijkers, Rogier Sanders and Paul Moynagh for providing plasmids, Guus Rimmelzwaan for influenza A virus, Julia Drylewicz for advice on statistics and Alan Rigter, Robert de Vries and Marco Viveen for their technical assistance.