Research ArticleInfluenza

Antibodies to influenza nucleoprotein cross-react with human hypocretin receptor 2

See allHide authors and affiliations

Science Translational Medicine  01 Jul 2015:
Vol. 7, Issue 294, pp. 294ra105
DOI: 10.1126/scitranslmed.aab2354
  • Fig. 1. Alignment studies with influenza sequences and components of the HCRT neurotransmission system demonstrated a shared motif between influenza NP and HCRT receptors.

    The influenza NP peptide contained isoleucine (in natural virus and X-179A vaccine reassortant) or methionine (in X-181 vaccine reassortant). Isoleucine in the figure is underlined because of its structural similarity to leucine. Adding in additional flanking residues to the N or C terminus (up to 11 residues total) of the fragments did not extend the alignment beyond the motif indicated. Subscript numbers indicate amino acid positions.

  • Fig. 2. Reactivity of clinical sera with cell lines expressing human HCRT receptor 2.

    (A) Narcoleptic patients with a history of Pandemrix vaccination in 2009 (red color), subjects at day 22 after vaccination (light blue color) or day 202 after vaccination (dark blue color) with Focetria, individuals with A(H1N1)pdm09 infection (pink color), and Finnish children not known to have narcolepsy in 2004/2005 (yellow color) demonstrated IgG binding to HCRT receptor 2 (data are means ± SD of triplicate experiments; inter-assay variability was zero). BK indicates non–HCRT-specific in-cell ELISA background verified by microscopic studies (fig. S4). The proportion of sera from *0602-positive Pandemrix-vaccinated narcoleptic patients, 1:800 dilution, with reactivity to HCRT receptor 2 (17 of 20 sera) was found to be significantly larger than that calculated for Focetria-vaccinated subjects (0 of 6 sera, P < 0.001, χ2 test for proportions) and for individuals with A(H1N1)pdm09 infection (5 of 20, P < 0.001, χ2 test for proportions). (B) Differential interference contrast microscopy of cell line engineered to express human HCRT receptor 2. (C) Nuclear staining with DAPI (4′,6-diamidino-2-phenylindole) (blue color). (D) Punctate membrane staining with serum, 1:400 dilution, from a narcoleptic patient with a history of Pandemrix vaccination (green color). (E) Punctate membrane staining with commercial antibody to HCRT receptor 2, 1:400 dilution, consistent with recognition of the extracellular domain of membrane-bound HCRT receptor 2 (red color). (F) Merged image (yellow) from double-labeling experiment with narcoleptic patient serum (green) and commercial antibody to HCRT receptor 2 (red). ***P < 0.001.

  • Fig. 3. Inhibition of narcoleptic patient serum IgG binding to HCRT receptor 2.

    (A) Sera from narcoleptic patients with a history of Pandemrix vaccination and HCRT receptor 2 antibodies (red color) or without detectable HCRT receptor 2 antibodies (brown color) were preincubated with recombinant peptide comprising the extracellular domain of HCRT receptor 2, peptides for influenza NP (amino acids 106 to 126) containing isoleucine (LILY I) or methionine (LILY M), or a scrambled peptide of the NP peptide containing isoleucine (final serum dilution 1:800). HCRT receptor 2–specific ELISA signal detected previously with patient 120, 136, and 137 sera was subsequently inhibited 62 to 74% (P < 0.001, multiple regression analysis) by either HCRT receptor 2 or NP peptide preincubation confirming the existence of cross-reactive antibodies (data are means ± SD of triplicate experiments, inter-assay variability less than 20%). No significant inhibition was detected using patient 119, 140, and 154 sera. No significant inhibition was detected when using the scrambled peptide. (B to E) HCRT receptor 2–specific punctate staining by serum, 1:400 dilution, from a narcoleptic patient with a history of Pandemrix vaccination (B) was substantially reduced when preincubated with (C) recombinant HCRT receptor 2 peptide, (D) NP peptide containing isoleucine, or (E) NP peptide containing methionine. (F and G) HCRT receptor 2–specific punctate staining by commercial antibody to HCRT receptor 2, 1:400 dilution (F), was inhibited by recombinant HCRT receptor 2 peptide (G). Source data are provided in fig. S7. **P < 0.01; ***P < 0.001.

  • Fig. 4. Differential NP content in inactivated seasonal influenza vaccines.

    (A) SDS-PAGE lanes loaded per micrograms of HA: (1 and 2) Fluzone trivalent, 1.8 and 0.9 μg; (3 and 4) Fluzone quadrivalent, 2.4 and 1.2 μg; (5 and 6) Fluarix trivalent, 1.8 and 0.9 μg; (7 and 8) Fluarix quadrivalent, 2.4 and 1.2 μg; (9 and 10) Afluria trivalent, 0.75 and 0.375 μg—lower amount of HA was loaded for Afluria to enable resolution of clear bands; (11 to 15) NP standard, 4, 2, 1, 0.5, and 0.25; (16) Agrippal trivalent, 1.8 μg; (17) Chiromas trivalent, 1.8 μg; and (18) Fluvirin trivalent, 1.8 μg. (B) Western blotting with NP-specific monoclonal antibody, lnA245. Same vaccines and loading as in A except (9) 0.9 μg, (10) 0.45 μg, and (11 to 15) NP standard, 5, 2.5, 1.25, 0.625, and 0.3125 μg. (C) Western blotting with another NP-specific monoclonal antibody, InA108. Loading reduced for lanes 1 to 10 due to more sensitive antibody and intentionally biased against Agrippal, Chiromas, and Fluvirin as follows: (1 to 10) 0.3 and 0.15 μg, respectively; (11 to 15) NP standard, 1, 0.5, 0.25, 0.125, and 0.0625 μg; (16) Agrippal, 1.8 μg; (17) Chiromas, 1.8 μg; and (18) Fluvirin, 1.2 μg. NP content in the subunit vaccines Agrippal and Chiromas (Chiromas is similar to the MF59-adjuvanted Fluad) was below the level of detection of SDS-PAGE and Western blotting. Complete image of blots for (B) and (C) is shown in fig. S9.

  • Fig. 5. Influenza A virus NP antibody inhibition test.

    (A to E) Undiluted, 1:160 diluted, and 1:320 diluted serum samples demonstrate presence of NP antibodies of different titers in serum from (A) narcoleptic patients with a history of Pandemrix vaccination in 2009 (about day 500 after vaccination), (B and C) subjects vaccinated with Focetria in 2009 (day 202 after vaccination or day 22 after vaccination, respectively), (D) individuals with A(H1N1)pdm09 infection (18 days after infection), and (E) Finnish children not known to have narcolepsy in 2004/2005. NPRI indicated in the figure reflects the presence of NP-specific antibody in sera that results in inhibition of NP influenza antigen in capture ELISA kit (Virusys Corporation). Only in individuals with A(H1N1)pdm09 infection (D) the proportions of positive results using undiluted sera and sera diluted by the factor 1:320 were not statistically different (P = 0.1), indicating that infection elicited very high antibody titers. At serum dilution of 1:320, antibody concentrations in (D) individuals with A(H1N1)pdm09 infection were significantly greater (16 of 20, 80% with positive NPRI) than those from (C) time-matched Focetria vaccinated subjects (1 of 6 sera, 17% with positive NPRI, P < 0.05, χ2 squared test). *P < 0.05.

  • Fig. 6. Influenza NP-associated immune response in subjects vaccinated with Focetria in 2009.

    Sera (1:80 dilution) at baseline, day 8, day 22, and day 202 after Focetria vaccination were tested for antibodies to NP using a commercial ELISA kit (quadruplicate experiments, inter-assay variability less than 30%). Antibody responses (y axis) represent NPRI described in Fig. 5. Day 202 values (July 2010) served to establish background antibodies in the absence of infection, suggesting that in December 2009, subjects 2, 4, and 5 had previously been exposed to influenza A virus (subject labels capitalized to facilitate quick identification). The significant response to NP in subject 1 at day 8 (P < 0.01, unpaired Student’s t test) was transient with a decrease in NP antibodies already at day 22 (P < 0.01, unpaired Student’s t test), suggesting suboptimal amounts of NP to elicit a durable immune response to NP. Subjects 3 and 6 demonstrated no statistically significant NP antibody differences over time. Asterisks directly over time point indicate significance relative to day 1 time point. Asterisks centered over horizontal bar indicate significance between indicated time points. Subject numbers were also indicated in parentheses near asterisks to indicate significance of those time points for each indicated subject. Logarithms of the data were represented for each subject and time point along with their arithmetic means and 95% confidence intervals. Source data are provided in table S7. *P < 0.05; **P < 0.01.

  • Table 1. Influenza virus and vaccine reassortant strain differences in protein sequence.

    Influenza sequences were retrieved from the National Center for Biotechnology Information (NCBI) Influenza Virus Resource (75) querying by the strains for A(H1N1)pdm09 virus (A/California/07/2009) and for vaccine reassortants (X-179A and X-181) with duplicate sequences removed. Each influenza protein expected to be present in the vaccine preparations (M1, NP, HA, and NA) was compared across strains, and identified three sequences where X-181 differed from X-179A by at least one residue. Sequences identified for X-179A were similar to those of A(H1N1)pdm09 virus. Residues that vary are underlined.

    ProteinGenbank accessionSourceSequence
    HAACR47014X-179A136 KTSSWPNHDSNKGVTAACPHA
    AFM72842X-181136 KTSSWPNHDSDKGVTAACPHA
    NPADE29096X-179A106 RELILYDKEEIRRIWRQANNG
    AFM72846X-181106 RELILYDKEEMRRIWRQANNG
    ADE29096X-179A130 WRQANNGDDAAAGLTHMMIWH
    AFM72846X-181130 WRQANNGDDATAGLTHMMIWH
  • Table 2. Quantification of NP in influenza vaccines by mass spectrometry.

    For each vaccine, the proteins were quantified by spectral count of related strains indicated in the package insert. Because commercial vaccines indicate primarily micrograms of HA content, mass spectrometry–quantitated NP spectral counts were expressed in relation to HA spectral count. NP content in seasonal split-vaccines was about 3 to 12 times greater than that detected in seasonal subunit vaccines, with the exception of Fluvirin. Testing of Focetria by mass spectrometry demonstrated NP-to-HA ratios similar to those seen with Agrippal and Chiromas. NP content of the A(H1N1)pdm09 split-virion vaccines Arepanrix and Pandemrix was about 2.9 and 3.7 times greater, respectively, than that detected in the A(H1N1)pdm09 subunit vaccine Focetria.

    VaccineNP (spectra
    count)
    HA (spectra
    count)
    NP per HA (ratio)
    Fluzone
    trivalent
    3282581.27
    Fluzone
    quadrivalent
    3203350.96
    Fluarix
    trivalent
    1462820.52
    Fluarix
    quadrivalent
    1421870.76
    Afluria
    trivalent
    2126040.35
    Agrippal
    trivalent
    433930.11
    Chiromas
    trivalent
    464520.10
    Fluvirin
    trivalent
    3263410.96
    Focetria
    lot 2009
    634670.13
    Focetria
    lot 2010
    604980.12
    Arepanrix
    2010
    4887030.69
    Pandemrix
    2009
    4464650.96

Supplementary Materials

  • www.sciencetranslationalmedicine.org/cgi/content/full/7/294/294ra105/DC1

    Materials and Methods

    Fig. S1. Visualization of three-dimensional proteins and surface-exposed domains from published crystal structures of human HCRT receptor 2 and trimer NP structure from the 1/Wilson-Smith/1933 influenza strain.

    Fig. S2. Trimer NP structure from the 1/Wilson-Smith/1933 influenza strain interrogated for structural chains, known T cell epitopes, sequence conservation, and functional determinants.

    Fig. S3. The conservation of influenza NP epitopes identified in this study with other influenza strains from 1902 to 2013.

    Fig. S4. Clinical serum samples with lower signal on ELISA confirmed to be background staining (1:400 dilution) based on microscopic patterns of reactivity.

    Fig. S5. Microscopy of cell lines engineered to express HCRT receptor 1.

    Fig. S6. Inhibition of IgG binding to HCRT receptor 2 by isoleucine variant of influenza NP peptide.

    Fig. S7. Source data for blocking experiments done in triplicate on six subjects with narcolepsy and history of Pandemrix vaccination.

    Fig. S8. Modeling of influenza NP peptide (amino acids 111 to 122) fit within the HLA-DQB1*0602 (allele strongly associated with narcolepsy).

    Fig. S9. Western blots for influenza NP using monoclonal antibody lnA245 or monoclonal antibody lnA108.

    Fig. S10. Extracted ion chromatogram (EIC) generated for the influenza NP peptides ELILYDKEEIR (isoleucine variant), ELILYDKEEMR (methionine variant), and IVVDYMMQKPGK (control sequence from influenza HA contained in all vaccines).

    Fig. S11. Lineage of influenza vaccine reassortants and laboratory strains generated through crosses with high-yielding donor strains.

    Fig. S12. Crosses of X-157 high-yielding donor strain with other historical influenza strains.

    Fig. S13. The identity of the “X” contained in the influenza NP YDKEEXR sequence from NYMC X-157 (CY095712).

    Fig. S14. Influenza HA–associated immune response in subjects vaccinated with Focetria in 2009.

    Table S1. Percent of individuals with HLA-DQB1*0602 allele associated with narcolepsy.

    Table S2. In-cell ELISA IgG binding value (absorbance 450 nm/615 nm) for sera with non–HCRT receptor 2–specific background staining.

    Table S3. HLA-DQA1*0102:DQB1*0602 (allele tightly associated with narcolepsy-cataplexy) binding to 21- and 15-mer peptides.

    Table S4. T cell studies published in the literature on the influenza NP motif “YDKEEIRRIWRQ” providing functional evidence for HLA class II processing of this epitope.

    Table S5. Influenza A virus NP antibody inhibition test with sera from individuals with A(H1N1)pdm09 infection.

    Table S6. Influenza A virus NP antibody inhibition test with sera from Finnish children not known to have narcolepsy in 2004/2005.

    Table S7. Source data for influenza NP-associated immune response in subjects vaccinated with Focetria in 2009.

    Movie S1. Influenza A virus NP indicating exposed surface location of NP mimic of HCRT receptors 2 and 1.

    References (7679)

  • Supplementary Material for:

    Antibodies to influenza nucleoprotein cross-react with human hypocretin receptor 2

    Syed Sohail Ahmed,* Wayne Volkmuth, José Duca, Lorenzo Corti, Michele Pallaoro, Alfredo Pezzicoli, Anette Karle, Fabio Rigat, Rino Rappuoli, Vas Narasimhan, Ilkka Julkunen, Arja Vuorela, Outi Vaarala, Hanna Nohynek, Franco Laghi Pasini, Emanuele Montomoli, Claudia Trombetta, Christopher M. Adams, Jonathan Rothbard, Lawrence Steinman*

    *Corresponding author. E-mail: sohail.q.ahmed{at}gsk.com (S.S.A.); steinman{at}stanford.edu (L.S.)

    Published 1 July 2015, Sci. Transl. Med. 7, 294ra105 (2015)
    DOI: 10.1126/scitranslmed.aab2354

    This PDF file includes:

    • Materials and Methods
    • Fig. S1. Visualization of three-dimensional proteins and surface-exposed domains from published crystal structures of human HCRT receptor 2 and trimer NP structure from the 1/Wilson-Smith/1933 influenza strain.
    • Fig. S2. Trimer NP structure from the 1/Wilson-Smith/1933 influenza strain interrogated for structural chains, known T cell epitopes, sequence conservation, and functional determinants.
    • Fig. S3. The conservation of influenza NP epitopes identified in this study with other influenza strains from 1902 to 2013.
    • Fig. S4. Clinical serum samples with lower signal on ELISA confirmed to be background staining (1:400 dilution) based on microscopic patterns of reactivity.
    • Fig. S5. Microscopy of cell lines engineered to express HCRT receptor 1.
    • Fig. S6. Inhibition of IgG binding to HCRT receptor 2 by isoleucine variant of influenza NP peptide.
    • Fig. S7. Source data for blocking experiments done in triplicate on six subjects with narcolepsy and history of Pandemrix vaccination.
    • Fig. S8. Modeling of influenza NP peptide (amino acids 111 to 122) fit within the HLA-DQB1*0602 (allele strongly associated with narcolepsy).
    • Fig. S9. Western blots for influenza NP using monoclonal antibody lnA245 or monoclonal antibody lnA108.
    • Fig. S10. Extracted ion chromatogram (EIC) generated for the influenza NP peptides ELILYDKEEIR (isoleucine variant), ELILYDKEEMR (methionine variant), and IVVDYMMQKPGK (control sequence from influenza HA contained in all vaccines).
    • Fig. S11. Lineage of influenza vaccine reassortants and laboratory strains generated through crosses with high-yielding donor strains.
    • Fig. S12. Crosses of X-157 high-yielding donor strain with other historical influenza strains.
    • Fig. S13. The identity of the “X” contained in the influenza NP YDKEEXR sequence from NYMC X-157 (CY095712).
    • Fig. S14. Influenza HA–associated immune response in subjects vaccinated with Focetria in 2009.
    • Table S1. Percent of individuals with HLA-DQB1*0602 allele associated with narcolepsy.
    • Table S2. In-cell ELISA IgG binding value (absorbance 450 nm/615 nm) for sera with non–HCRT receptor 2–specific background staining.
    • Table S3. HLA-DQA1*0102:DQB1*0602 (allele tightly associated with narcolepsy-cataplexy) binding to 21- and 15-mer peptides.
    • Table S4. T cell studies published in the literature on the influenza NP motif “YDKEEIRRIWRQ” providing functional evidence for HLA class II processing of this epitope.
    • Table S5. Influenza A virus NP antibody inhibition test with sera from individuals with A(H1N1) pdm09 infection.
    • Table S6. Influenza A virus NP antibody inhibition test with sera from Finnish children not known to have narcolepsy in 2004/2005.
    • Table S7. Source data for influenza NP-associated immune response in subjects vaccinated with Focetria in 2009.
    • References (7679)

    [Download PDF]

    Other Supplementary Material for this manuscript includes the following:

    • Movie S1 (Microsoft Word format). Influenza A virus NP indicating exposed surface location of NP mimic of HCRT receptors 2 and 1.

    [Download Movie S1]

Navigate This Article