香京julia种子在线播放

Mice were infected with escalating doses of viruses reflective of recent circulating influenza viral strains and or widely used laboratory strains. For H1N1, mice were infected with influenza A/England/195/2009 (12) (clinical isolate: Figure 1A) or PR8 (Lab strain: Figure 1B). Animals infected with both the seasonal and laboratory strains of H1N1 lost weight proportionally to the infectious dose of virus. Comparing the response by dose of plaque forming units would suggest PR8 causes more disease per PFU used, but there may be limitations in using PFU for comparisons. For H3N2, mice were infected with A/England/691/2010 (Clinical isolate: Figure 1C) or A/X31 (Lab strain), which consists of HA and NA molecules from A/Hong Kong/1/68 (H3N2) on a PR8 background (Figure 1D). While animals infected with the laboratory H3N2 strain, X31, lost weight after infection, mice infected with the current seasonal H3N2 virus (A/England/691/2010) did not lose weight at the doses used. To test responses to influenza B, mice were infected with virus isolates that are close to current circulating strains, B/Florida/04/06 representing the Yamagata lineage (Figure 1E) and B/Brisbane/60/2008 (Figure 1F) representing the Victoria lineage. Infection with the Yamagata but not the Victoria lineage influenza B led to weight loss, but a larger dose of virus may be required for the Victoria lineage virus.

Having observed that infection with some strains of influenza virus caused signs of disease, we wished to confirm that these viruses were able to replicate in mouse lungs and wanted to investigate the histological and immunological correlations of disease. Mice were challenged intranasally with representative H1N1 (Eng/195), Flu B (Flo/04), and H3N2 (A/X-31) strains and monitored over 7 days. A control group of mice were given sterile PBS intranasally. All influenza challenged mice lost significant amounts of weight compared to the control group (Figure 2A). Temperature was also measured, but no significant differences were observed (Figure 2B). Lung viral load was assessed via plaque assay (all groups) or influenza A M gene RNA qPCR (H1N1, X31, and control) (Figures 2C,D). Virus was detected in the lungs via plaque assay on day 4 for all infected mice (Figure 2C). Viral RNA was quantified for the influenza A infected groups and was detected on day 4 for both H1N1 and A/X-31 (Figure 2D). At day 7, virus and viral RNA was only detected in the H1N1 infected mice (data not shown).

Lung inflammation was investigated as a measure of disease pathology. There were no significant differences in lung inflammation at day 4 after challenge in either the upper and lower airways (data not shown). However, 7 days after challenge there was significantly more inflammation in both the upper (Figure 2E) and lower (Figure 2F) airways of infected animals. This lower airway inflammation was reflected by an increased cell recovery on day 7 after infection (Figure 2G). Airway inflammation correlated with lung cell counts (Figure 2H). The composition of the lung cellular infiltrate was assessed by flow cytometry. Four days after influenza challenge, there was a significant increase in the percentage of NK cells in infected mice compared to PBS controls (Figure 2I). The infiltrate at day 7 was predominantly made up of CD8 T cells, with no differences in the number of CD4 T cells (Figure 2J) but significantly higher levels of CD8 T cells in the lungs of infected mice than controls (Figure 2K).

Having developed infectious challenge models, we wished to determine the efficacy of seasonal influenza vaccines in mice. Mice were intramuscularly immunized with purified surface antigens from A/California/7/2009, which was the (H1N1)pdm09 strain used in the trivalent vaccine from 2010 to 2016. The aim of the study was to find the lowest protective dose of vaccine, mice were given increasing doses from 0.02 to 1.5 µg A/California/7/2009 influenza hemagglutinin (as part of a mixture of viral surface antigens), for reference the human vaccine dose is 15 µg. The antibody response was proportional to the immunization dose, with most in the 1.5 µg immunized group (Figure 3A). Mice received a single dose of vaccine and were challenged with 2.5 × 10⁵ PFU A/California/7/2009 H1N1 4 weeks later. Mice immunized with 1.5 or 0.5 µg lost up to 15% body weight peaking day 5 after infection. These mice were partially protected compared to the naïve animals, losing significantly less weight than naïve animals on day 6 after infection (Figure 3B). Mice immunized once with 0.02 µg did not produce antibodies and were not protected against challenge. To test whether repeat immunization affected the dose required, mice were immunized with 0.02, 0.01, or 0.005 µg (20, 10, or 5 ng) of A/California/7/2009 H1N1 hemagglutinin on days 0 and 21. Antibody responses were significantly greater in mice immunized with 0.02 µg than 0.005 µg or the naïve animals (Figure 3C). Mice were protected against infection when mice were vaccinated twice in a prime boost regime with a dose of 0.02 or 0.01 µg and partial protection was seen after immunization with 0.005 µg (Figure 3D). From these studies, we observe that immunization with a very low dose of protein can protect mice against homologous influenza challenge.

The biggest recent change in vaccine strains occurred with the emergence of the H1N1 pandemic strain in 2009. Because the strain recommendation preceded the emergence of the virus in 2009, the 2009–2010 vaccine did not contain the (H1N1)pdm09 like strain. However, from 2010 to 2016, the H1N1 subtype strain included in the virus was A/California/7/2009 (H1N1pdm09). By comparison, in the seven seasons since the emergence of the strain of H1N1 influenza (2009 pandemic strain) to the winter of 2017, the H3N2 component was changed four times (Table 1). In the same time period, the B component has changed between representative Yamagata and Victoria lineage reference strains in trivalent vaccines (13); after 2012 quadrivalent vaccines with two B strains were recommended. We wished to determine the genetic and antigenic drift of the (H1N1)pdm09 viral strains since its emergence in 2009.

We performed an integrated phylogenetic and antigenic cartography analysis (14) using hemagglutinin sequence data and HI titers for 61 (H1N1)pdm09 viruses collected between 2009 and 2016, comprising 53 viruses collected from England, 2 vaccine strains, and 6 other WHO reference viruses (Table S1 in Supplementary Material). Analysis of these genetic and antigenic data showed gradual genetic drift (Figure 4A) as well as gradual antigenic change (Figure 4B) of (H1N1)pdm09 viruses since 2009. The minimum antigenic distinction for when an influenza vaccine update is recommended is generally a difference between a vaccine strain and circulating strains of 2 antigenic map units, representing a fourfold drop in heterologous HI titer (15). Viruses circulating in England with at least 3 antigenic units (>8-fold drop in HI titer) difference from A/California/07/2009 only emerged from 2015 onward. A similar pattern was seen using multidimensional scaling (MDS) (Figure S1 in Supplementary Material). These recent viruses are antigenically similar to the updated H1N1 subtype component vaccine strain, A/Michigan/45/2015. Additionally, the recently circulating viruses include the emergence of one genetically distinct group of viruses, which are also genetically similar to the A/Michigan/45/2015 vaccine strain.

Since the H1 component of vaccine in use was unchanged from the initial wave of the pandemic, we wished to see whether the protection efficacy changed as the virus changed. Mice were immunized with 0.5 µg A/California/7/2009 antigens and then challenged either with a matched isolate from the initial wave of the pandemic in 2009 (A/England/195/2009), or drift isolates from 2010 (A/England/672/2010) or 2015 (A/England/336/2015). Immunized mice lost significantly less weight than control mice when infected with the 2009 (Figure 5A) or 2010 (Figure 5B) isolates. However mice infected with a 2015 isolate were not initially protected compared to the control animals, but they recovered slightly more rapidly than the unimmunized mice (Figure 5C). There was no significant difference in the antibody response to the immunizing antigen in the mice, suggesting that viral escape from this antigen drives the reduction in protection (Figure 5D).

Seasonal influenza viruses (Table 2) were isolated by Public Health England (UK). The England strains of H1N1, A/England/195/2009, A/England/672/2010, and A/England/336/2015 were isolated in SIAT-MDCK cells (12). B viruses were expanded in eggs prior to being grown in Madin-Darby Canine Kidney (MDCK) cells. Prior to use in mice, viruses were propagated in MDCK cells, in serum-free DMEM supplemented with 1 µg/ml trypsin. The virus was harvested 3 days after inoculation and stored at -80°C. Viral titer was determined by plaque assay as described below.

6–10-week-old female CB6F1 mice were obtained from Harlan UK Ltd. (Horsham, UK) or from an internal breeding colony and kept in specific-pathogen-free conditions in accordance with the United Kingdom’s Home Office guidelines and all work was approved by the Animal Welfare and Ethical Review Board (AWERB) at Imperial College London. Studies followed the ARRIVE guidelines. Mice were immunized intramuscularly (i.m.) with purified surface antigens from influenza strain H1N1 A/California/7/2009 (GSK Vaccines, Siena, Italy) in 50 µl, either once (prime only) or twice (prime boost). For infections, mice were anesthetized using isoflurane and infected intranasally (i.n.) with 100 µl influenza virus or sterile PBS. Body surface temperature was taken from the xiphoid process using a hand-held infrared thermometer.

Mice were culled using 100 µl intraperitoneal pentobarbitone (20 mg dose, Pentoject, Animalcare Ltd., UK) and tissues collected as previously described (29). Blood was collected from femoral veins and sera isolated after clotting by centrifugation. Lungs were removed and homogenized by passage through 100-µm cell strainers, then centrifuged at 200 × g for 5 min. Supernatants were removed and the cell pellet treated with red blood cell lysis buffer (ACK; 0.15 M ammonium chloride, 1 M potassium hydrogen carbonate, and 0.01 mM EDTA, pH 7.2) before centrifugation at 200 × g for 5 min. The remaining cells were resuspended in RPMI 1640 medium with 10% fetal calf serum, and viable cell numbers determined by trypan blue exclusion.

Upper and lower regions of paraformaldehyde-fixed left lung lobes were processed and embedded in paraffin. Sections of 3 µm were stained with hematoxylin and eosin and the entire section was scanned at ×20 magnification so that the area with the greatest inflammation could be assigned the inflammation score. The degree of airway inflammation was assessed in a blinded manner using a modified system described previously (30). Briefly, the degree of inflammation in the peribronchiolar, perivascular, and interstitial regions of both the upper and lower airways was assessed. A value of 0 (none), 1 (minimal), 2 (mild), 3 (moderate), or 4 (severe) was given to each histological site and the sum of these scores was used as the total upper/lower respiratory inflammation score.

Antibodies specific to influenza H1N1 were measured using a standardized ELISA (32). IgG responses were measured in sera. MaxiSorp 96-well plates (Nunc) were coated with 1 µg/ml surface proteins or a combination of anti-murine lambda and kappa light chain specific antibodies (AbDSerotec, Oxford, UK) and incubated overnight at 4°C. Plates were blocked with 1% BSA in PBS. Bound IgG was detected using HRP-conjugated goat anti-mouse IgG (AbD Serotec). A dilution series of recombinant murine IgG was used as a standard to quantify specific antibodies. TMB with H₂SO₄ as stop solution was used to detect the response and optical densities read at 450 nm.

A Bayesian multidimensional scaling (BMDS) model (14) of antigenic cartography (15) was implemented in BEAST v1.8.4 (33) to jointly infer antigenic and phylogenetic relationships, as previously described (14, 34). Briefly, an antigenic dataset composed of available HA gene sequences and corresponding HI measurements was assembled for 61 H1N1pdm09 viruses collected between 2009 and 2016: 53 viruses circulating in England, 2 vaccine strains, and 6 WHO reference viruses. Viruses were originally isolated from clinical specimens either by WHO NICs or by the WHO Collaborating Center and the corresponding HI measurements were obtained either from the Francis Crick Institute Influenza Interim Reports or were kindly provided by John McCauley. HA sequences were downloaded from either the Influenza Research Database (35) or the EpiFlu database (36). A phylogenetic tree of the HA sequences was estimated using BEAST (33) which incorporated the HKY substitution model, a coalescent model with constant effective population size and a strict molecular clock. Markov chain Monte Carlo (MCMC) was run for 15 million steps and trees were logged every 1,500 steps, with a burn-in of 5 million steps, resulting in 10,000 trees. This posterior set of 10,000 trees was used with the HI data to implement the full BMDS model infer virus and serum locations in two antigenic dimensions, as well as virus avidities, serum potencies, MDS precision, and virus and serum location precisions in BEAST. MCMC chains were run for 500 million states with sampling every 200,000 states with 10% burn-in, and run convergence was checked in Tracer v1.6 (http://tree.bio.ed.ac.uk/software/tracer/). A maximum clade credibility tree was summarized in TreeAnnotator v1.8.4 (33) and visualized using FigTree v1.4.3 (http://tree.bio.ed.ac.uk/software/figtree/). Antigenic map plots were generated using custom Python scripts with the matplotlib library (37).

Calculations as described in figure legends were performed using Prism 6 (GraphPad Software Inc., La Jolla, CA, USA).