Bispecific antibody generated with and click chemistry has broad antiinfluenza virus activity

Koen Wagnera,1, Mark J. Kwakkenbosa,1, Yvonne B. Claassena, Kelly Maijoora, Martino Böhnea, Koenraad F. van der Sluijsb, Martin D. Wittec,2, Diana J. van Zoelend, Lisette A. Cornelissend, Tim Beaumonta, Arjen Q. Bakkera, Hidde L. Ploeghc, and Hergen Spitsa,3

aAIMM Therapeutics, 1105 BA Amsterdam, The Netherlands; bLaboratory of Experimental Intensive Care and Anesthesiology, Academic Medical Center, University of Amsterdam, 1105 AZ Amsterdam, The Netherlands; cWhitehead Institute for Biomedical Research, Cambridge, MA 02142; and dDepartment of Virology, Central Veterinary Institute, Wageningen University and Research Centre, 8200 AB Lelystad, The Netherlands

Edited by K. Christopher Garcia, Stanford University, Stanford, CA, and approved October 21, 2014 (received for review May 9, 2014) Bispecific antibodies have therapeutic potential by expanding the Here, we present a bispecific antibody format, in which two functions of conventional antibodies. Many different formats of antibodies are fused at their C termini, using a combination of bispecific antibodies have meanwhile been developed. Most are sortase transpeptidation and click chemistry (20), to create an IgG genetic modifications of the antibody backbone to facilitate heterodimer. This C-C fusion does not require mutations within the incorporation of two different variable domains into a single antibody constant domains that might interfere with Fc-receptor molecule. Here, we present a bispecific format where we have binding or that would compromise antibody stability. Thus, the fused two full-sized IgG antibodies via their C termini using sor- native antibody structure is fully retained in our format. tase transpeptidation and click chemistry to create a covalently C-to-C fusion is a two-step process (Fig. 1), using a combina- linked IgG antibody heterodimer. By linking two potent anti- tion of sortase transpeptidation and click chemistry (20). Sortase influenza A antibodies together, we have generated a full anti- is a bacterial that functions to attach cell surface body dimer with bispecific activity that retains the activity and bearing an “LPXTG” motif to the of Gram-positive bac- stability of the two fusion partners. teria via transacylation (21, 22). Sortase-catalyzed transpeptidation allows for efficient site-specific modifications under physiological

antibody engineering | immunotherapy | influenza conditions, with excellent specificity and near-quantitative yields INFLAMMATION IMMUNOLOGY AND (23–25). To facilitate site-specific linking of the C termini of two ith a steady increase of antibodies and antibody derivatives antibodies, the fusion partners are labeled with either an azide or Wsuch as antibody drug conjugates and bispecific antibodies a cyclooctyne (DIBAC) functional group. The modified proteins entering the clinic, monoclonal human antibodies are now an are then conjugated via a strain-promoted cycloaddition between established source of new therapeutic agents (1, 2). The development the azide and the cyclooctyne. This reaction is highly specific and of bispecific antibodies has generated particular interest, because readily proceeds at room temperature in aqueous environments at it allows expansion of basic antibody functions (3, 4). Through neutral pH (26), allowing for efficient fusion under mild conditions. binding two (or more) different targets, a bispecific antibody can To test the robustness of this process and determine the features simultaneously engage two epitopes of a disease agent, block/acti- of this bispecific antibody format, we fused two potent anti- influenza antibodies, each active against a different subgroup of vate multiple ligands/receptors at once, or recruit immune effector the influenza A virus. Based on the hemagglutinin (HA) cells (i.e., T cells or B cells) to a specific (tumor) site (5). There is a growing interest in bispecific antibodies with anticancer proper- ties, which has led to an increase in bispecifics that have entered Significance preclinical testing (5, 6). Bispecific antibodies with defined functions are generated by Bispecific antibodies expand the function of conventional anti- means of genetic or biochemical engineering. Many different bodies. However, therapeutic application of bispecifics is ham- methods exist to engineer immunoglobulins, with more than 45 pered by the reduced physiochemical stability of such molecules. bispecific antibody formats at last count (reviewed in ref. 5). We present a format for bispecific antibodies, fusing two full- These bispecific antibody formats fall into three broad subclasses sized antibodies via their C termini. This format does not require mutations in the antibody constant domains beyond installation (5): (i) single-chain double variable domain formats (50–100 of a five-residue tag, ensuring that the native antibody structure kDa) (7–9): Generally these bispecifics consist of multiple vari- ii is fully retained in the bispecific product. We have validated the able domains that are connected via peptide linkers. ( ) IgG with approach by linking two anti-influenza A antibodies, each active multiple variable domains: In this type of bispecific antibody, against a different subgroup of the virus. The bispecific antibody a second variable domain is genetically linked to any desirable dimer retains the activity and the stability of the two original position in the IgG molecule (i.e., the C or N terminus of either antibodies. the IgG heavy or light chain) (10–12). (iii) Asymmetric IgG molecules: In an asymmetric IgG antibody, two different vari- Author contributions: K.W., M.J.K., T.B., H.L.P., and H.S. designed research; K.W., M.J.K., able domains are incorporated into a single, asymmetric, anti- Y.B.C., K.M., M.B., and A.Q.B. performed research; K.F.v.d.S., M.D.W., D.J.v.Z., and L.A.C. contributed new reagents/analytic tools; K.W., M.J.K., and H.S. analyzed data; and K.W., body molecule via heterodimerization of the constant domains. M.J.K., H.L.P., and H.S. wrote the paper. Heterodimerization may be achieved through engineering the – Conflict of interest statement: K.W., M.J.K., Y.B.C., K.M., M.B., T.B., A.Q.B., and H.S. are CH3domain(13 16) or the hinge region of the antibody (17, employees of AIMM Therapeutics. 18). Depending on the engineering method, asymmetric IgGs This article is a PNAS Direct Submission. can be made with a common light chain or with two different 1K.W. and M.J.K. contributed equally to this work. light chains (19). 2Present address: Bio-Organic Chemistry, Stratingh Institute for Chemistry, University of Each of these formats has its specific advantages and draw- Groningen, 9747 AG Groningen, The Netherlands. backs. Most of the limitations arise from the fact that their for- 3To whom correspondence should be addressed. Email: [email protected]. mats deviate significantly from the natural, highly stable, IgG This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. structure, which compromises stability and ease of manufacture. 1073/pnas.1408605111/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1408605111 PNAS Early Edition | 1of6 Downloaded by guest on September 23, 2021 this antibody binds full-length H5 HA protein and the HA2 subunit. The subunits of HA are described in SI Appendix, Table S2. AT10-005 contains the IGHV1-69 gene segment and harbors the hydrophobic signature commonly found in group 1-specific antibodies (SI Appendix,TableS3) (38, 39). Antibody competition using AT10-005 and the stem-binding antibody CR6261 (40) (SI Appendix,Fig.S1) for H1 binding on H1N1 (A/Hawaii/31/2007)- infected cells confirms that both antibodies bind similar regions. AT10-002 is specific for HA proteins of group 2 viruses (SI Appendix, Table S1) and shows neutralizing activity against four group 2 viruses (two H3N2, one H7N1, and one H7N7 virus) (Table 1). The antibody binds full-length H3 but not the separate HA1 portion (SI Appendix, Table S4). In addition, AT10-002 competes with the group 2 HA stem-specific antibody CR8020 (31) for binding to H3N2-infected cells (SI Appendix, Fig. S2). To further analyze the of AT10-002, we have isolated a third HA-specific antibody: AT10-003. AT10-003 was found to – Fig. 1. Approach for synthesis of C-to-C fused antibodies. (A) Antibodies bind to three H3 viruses (SI Appendix, Table S5), and reacted are labeled at the C terminus either with an azide (N ) or DIBAC with a click 3 with both the full-length H3 and the HA1 portion of the mole- peptide by using sortase. (B) Click-labeled antibodies are fused via the click SI reaction. cule indicating that the HA head is sufficient for binding ( Appendix, Table S4). Notably, AT10-003 was unable to block binding of AT10-002 to H3N2-infected cells (SI Appendix, Fig. sequence, there are 18 different subtypes of influenza A, divided S2). Therefore, the finding that AT10-002 binding is blocked by into two subgroups (27, 28). The HA protein is the common target CR8020 and not by AT10-003 suggests that the stem region of of almost all neutralizing antibodies, and several antibodies with group 2 HA influenza has the largest contribution to the AT10- broadly neutralizing activity between influenza A subtypes in the 002 epitope. Linking the broadly reacting antibodies AT10-005 same group exist (29–33). Combining two such potent subgroup- and AT10-002 would potentially result in a molecule active specific antibodies may result in an IgG heterodimer with even against a broad spectrum of group 1 and group 2 influenza A broader anti-influenza A activity. This type of molecule would viruses. have therapeutic relevance in a passive immunization setting, Synthesis of the C-to-C Fused Bispecific Antibody Dimer. To enable because influenza viruses continue to cause significant morbidity the C-to-C protein fusion, we modified the C termini of the and mortality, despite efforts to contain them with seasonal heavy chains of both antiinfluenza antibodies with a small tag, vaccines (34). Because these vaccines are typically only effective consisting of a GGGGS (G4S) linker sequence, followed by the against the specific seasonal viral strain, there is an urgent unmet sortase recognition site LPETGG and a His6 tag. The His tag is medical need for new treatments active against multiple subtypes used to monitor the sortase labeling reaction, because it will be of the influenza virus (35). removed upon sortase-catalyzed transpeptidation. Triglycine Results peptides containing either an azide or a DIBAC moiety were synthesized as described (20). We chose to label AT10-002 with Isolation and Characterization of Potent Antiinfluenza Antibodies. To DIBAC and AT10-005 with azide. The extent of sortase labeling obtain broadly neutralizing influenza A antibodies, we isolated was monitored by using a fluorescent azide-containing nucleo- memory B cells from influenza-vaccinated individuals. These B phile (GGG-TAMRA-azide): We observed excellent labeling cells were transduced with human Bcl6 and Bcl-xL as described with the azide-containing probe after 4 h at 41 °C in 25 mM Tris in Kwakkenbos et al. (36, 37) and screened for HA binding. B buffer, pH 8.0 and 150 mM NaCl (Fig. 2A). For the TAMRA- cells that recognized HA molecules from multiple influenza DIBAC-modified peptide, sortase labeling is somewhat less ef- subtypes were cloned and the antibody derived from them pro- ficient (Fig. 2B). We attribute this difference to the nonspecific duced in 293T HEK cells. This approach resulted in the iden- thiol-yne coupling reaction between DIBAC and free thiol tification of two broadly neutralizing antibodies: AT10-002 groups (41), such as the unpaired cysteine residues in the anti- and AT10-005. body or in sortase (Fig. 2A). Because the of sortase AT10-005 neutralizes two H1N1 and one H5N1 virus (Table contains a free thiol (42), sortase itself would be particularly 1) and binds four additional group 1 viruses (two H1N1, one vulnerable to the thiol-yne reaction (43). Nonetheless, we H5N1, and one H9N2) (SI Appendix, Table S1). In an ELISA, observed efficient installation of DIBAC after 4-h incubation at 33 °C, in 25 mM Tris buffer, pH 6.8 and 150 mM NaCl, demonstrating that a lower incubation temperature and pH Table 1. Neutralizing activity of AT10-002 and AT10-005 reduces thiol-yne coupling. Group Virus AT10-002 AT10-005 Following sortase labeling, the antibodies were separated from sortase, excess triglycine peptide, and other reaction products 1 H1N1 (A/Hawaii/31/2007) — 1.29 (± 0.53) by size exclusion chromatography. AT10-002–DIBAC was readily 1 H1N1 (A/Neth./602/2009) — 16.8 (± 1.8) coupled to AT10-005–azide at 20 °C. Antibody dimers were re- 1 H5N1 (A/Turkey/Turkey/04) — 8.2 (± 6.1) solved from any remaining antibody monomers by size exclu- 2 H3N2 (A/Neth./177/2008) 0.76 (± 0.27) — sion chromatography (Fig. 2C). After prolonged incubation of 2 H3N2 (A/Swine/St. Oedenrode/1996) 1.79 (± 0.35) — click-labeled antibodies, a second peak was visible in the elution 2 H7N1 (A/Chicken/Italy/1067/99) 22.3 (± 8.7) — profile (Fig. 2C). This second peak consists of higher order an- 2 H7N7 (A/Chicken/Neth./621557/03) 0.78 (± 0.36) — tibody oligomers that may form when the two click-linked C — termini of one antibody connect with two different antibodies (as Shown are IC50 values in nanomolars. , no neutralizing activity D detected. SI Appendix, Table S1 shows a complete overview of the reactivity shown in Fig. 2 ). Formation of higher-order complexes also of AT10-002 and AT10-005. Data represent mean ± SD of at least two in- occurred when the click reactions were performed at higher dependent experiments. temperatures or at a higher antibody concentration, suggesting

2of6 | www.pnas.org/cgi/doi/10.1073/pnas.1408605111 Wagner et al. Downloaded by guest on September 23, 2021 INFLAMMATION IMMUNOLOGY AND

Fig. 2. Preparation of BiFlu. (A) Determining optimal reaction conditions for sortase labeling. ST-tagged antibody AT10-005-ST (1.0 μM) is mixed with sortase

(2.0 μM) and GGG-TAMRA-azide or GGG-TAMRA-DIBAC (125 μM), in labeling buffer (25 mM Tris·HCl, 150 mM NaCl, and 10 mM CaCl2) and incubated for 6 h at the indicated pH and temperature. C, control reaction (incubation with GGG-TAMRA-DIBAC at 37 °C, pH 7.5). After incubation, the reaction mixture is

analyzed with reducing fluorescent SDS/PAGE (λex, 532 nm; λem, 580 nm). (B) Quantification of sortase-labeling results. Fluorescence is measured as relative to labeling with GGG-TAMRA-azide at 37 °C, pH 7.5 (set at 1.00 RU). (C) Gel filtration chromatogram of click-reaction products. AT10-002-DIBAC (5.0 μM) is mixed with AT10-005-aizde (5.0 μM). After the indicated time, a sample is analyzed with gel filtration chromatography (column volume: 120 mL). (D) Coomassie-stained SDS/PAGE gel of purified BiFlu (1.5 μg). (E) Anti-His Western blot of purified BiFlu (0.5 μg). HC, antibody heavy chain; LC, antibody light chain; M, Dual Color Protein Standard (Bio-Rad); S, sortase. Products of thiol-yne coupling are indicated with asterisks.

that this reaction requires careful control by choice of temper- the stability of the resulting structure, we assessed the stability ature and antibody concentration, to obtain the optimal yield of our antibody dimer (BiFlu) by means of dynamic scanning of the click-linked antibody dimer. fluorescence (DSF). We obtained pure and fully intact click-linked antibody dimers The DSF curves (Fig. 3A) showed that the melting properties as judged from analysis by SDS/PAGE (Fig. 2D). The presence of BiFlu are adequately described by summation of the curves of of a ∼100-kDa polypeptide (100 kDa equals twice the size of the the two individual antibodies. Thermal stability of the two anti- antibody heavy chain) in the reduced dimer sample (Fig. 2D) bodies is therefore unchanged when they are linked via their C confirms covalent fusion at the C termini of the heavy chains. termini. In view of its small size (24 aa + the click product) The reduced dimer samples also showed some monomeric heavy relative to the mass of BiFlu (±320 kDa), the linker sequence chains (Fig. 2D). The presence of these monomeric heavy chains that connects the two antibodies contributes little if at all to the upon SDS/PAGE of the dimer sample demonstrates that not all DSF curve of BiFlu. antibody heavy chains have fused. This finding is not unexpected, We further assessed the stability of the antibody dimer over because a single covalent HC-HC link suffices to generate an a more extended observation window. After 3 wk of incubation at antibody dimer. This single HC-HC linkage is not due to in- 37 °C in PBS (Fig. 3B) or in PBS plus (antibody-depleted) human complete sortase labeling; anti-His Western blotting demon- serum (Fig. 3C), the majority of BiFlu remains intact, suggesting strates a near-complete loss of the His tag in the purified dimer that the antibody dimer would also be stable at physiological con- (Fig. 2E). Apparently, after the first HC-HC linkage, a second ditions. Only a small amount of BiFlu (<< 10%) has decomposed click coupling between the two remaining modified HC C ter- into the separate IgG monomers. The covalent link between the two mini is disfavored, presumably due to steric hindrance. C termini in the antibody dimer is stable under these conditions. In this bispecific full antibody dimer format, the two anti- BiFlu Retains Structural Stability and Fc-Receptor Binding. To ensure bodies are linked via the C termini of the heavy chain. Because that the C-to-C fusion of two antibodies does not compromise the Fc portion contains the binding sites for the IgG Fc receptors

Wagner et al. PNAS Early Edition | 3of6 Downloaded by guest on September 23, 2021 its functional activity. The binding of BiFlu to HA proteins was tested with capture surface plasmon resonance (SPR), using ei- ther heavy or light chain-specific anti-IgG antibodies, followed by subsequent injections with H1 or H3 HA protein. H1 HA represents group 1 Influenza A viruses and should be detected by BiFlu component AT10-005; H3 is a representative of group 2 Influenza A and should be recognized by BiFlu component AT10-002 (Fig. 4A). When immobilized via a heavy chain-specific antibody, BiFlu bound both H1 and H3 HA with similar, low picomolar (KD =± 15 pM for both antigens) affinity, as did the parental antibodies Fig. 3. Stability of the BiFlu antibody dimer. (A) DSF curves of BiFlu and A μ (Fig. 4 and Table 2). In the same setup, we determined binding monomeric IgG (25 g/mL) in PBS buffer. (B) Long-term stability test of to H7 and H9 HA protein, finding that BiFlu binds these anti- BiFlu and monomeric IgG in PBS buffer. BiFlu or monomeric IgG (AT10-002 or AT10-005) is diluted into PBS buffer (final concentration: 250 μg/mL) and gens with an affinity similar to that of the parental antibodies incubated at 37 °C. After the indicated number of days, 2.5 μgofantibody (Table 2). is heated for 10 min at 55 °C (the lower temperature is used to minimize When using anti-light chain antibodies for immobilization, antibody breakdown) and analyzed with SDS/PAGE. (C) Long-term stability BiFlu could be captured on both anti-kappa and anti-lambda, test of BiFlu and monomeric IgG in IgG-depleted human serum. BiFlu or because AT10-002 and AT10-005 have different light chains monomeric IgG (mixture of AT10-002 and AT10-005) is diluted into IgG- (lambda and kappa, respectively). BiFlu captured via its light μ depleted human serum (final concentration: 250 g/mL) and incubated at chains binds both H1 and H3 HA (Fig. 4B), demonstrating that 37 °C. After the indicated time, 0.25 μgofantibody(+ serum) is heated for 10minat55°Candanalyzedwithanti-IgGHCWesternblotting.InB and C, BiFlu is a bivalent heterodimer. The fact that light chain-cap- BiFlu and monomeric IgG are not fully denatured; therefore, they run at tured BiFlu binds both antigens with similar kinetics as the single a lower molecular mass than expected (IgG molecular mass: 160 kDa). antibodies (here: AT10-002 + H3) indicates that, in the BiFlu sample, the two antibodies are linked in a 1:1 ratio and that homodimers must be absent. We estimate that BiFlu contains > (FcγRs) and the neonatal Fc receptor (FcRn) (44), we verified 95% heterodimers. that such a fusion did not impair interaction of the antibody We then tested the in vitro neutralization activity of BiFlu dimer with Fc receptors. We measured binding of BiFlu to sol- against H3N2 and H1N1. BiFlu neutralized both strains effi- uble Fc receptors by ELISA and found that BiFlu binds FcRn ciently (Fig. 5) with an IC50 value of ∼1.0 nM, similar to the two and all three Fcγ receptors with similar affinity as the parental single antibodies tested separately (Table 3). We then tested the IgGs (SI Appendix, Fig. S3). We also tested binding of BiFlu to in vivo activity of our antibodies in a murine H1N1 (A/PR/8/ THP-1 cells, which express Fc receptors on their cell surface 1934) challenge model. Prophylactic administration of 1 mg/kg (45), and observed equivalent binding of BiFlu and the parental AT10-005 protected the mice against lethal infection, all mice antibodies (SI Appendix, Fig. S3). pretreated with AT10-002 or rituximab lost more then 25% of BiFlu retains FcR-binding activity, implying that it should be their body weight and were killed by day 8 (SI Appendix, Fig. S7). capable of engaging FcR effector functions. If in vivo neutralization We examined the in vivo activity of BiFlu in the same challenge of virus by the single antibodies were to somehow involve engage- model, injecting mice with either BiFlu (2 mg/kg) or a mixture ment with Fc receptors, then the bispecific antibody retains func- of AT10-005 and AT10-002 (1 mg/kg each). All BiFlu-treated mice + tionality for this parameter as well. were protected against H1N1 challenge, similar to the AT10-002 AT10-005–treated mice, whereas 50% of the mice treated with BiFlu Retains Functional Activity and Neutralizes Influenza A. Having rituximab failed to recover from the infection (Fig. 6A) [the determined stability and receptor binding of BiFlu, we measured control group survival is different from the first experiment (SI Appendix, Fig. S7); we attribute this result to variation in the model]. Mice injected with BiFlu or AT10-002 + AT10-005 also Table 2. Kinetic constants for HA binding showed a significantly lower body weight loss compared with the control group (Fig. 6B). Antibody ka kd KD

H1 (A/New Caledonia/20/1999) AT10-002 —— — AT10-005 19.5 (± 1.8) 0.29 (± 0.10) 15.0 (± 5.8) BiFlu 21.4 (± 2.8) 0.34 (± 0.17) 16.2 (± 8.4) H3 (A/Wyoming/03/2003) AT10-002 19.2 (± 1.5) 0.23 (± 0.08) 11.8 (± 5.9) AT10-005 —— — BiFlu 22.2 (± 2.5) 0.33 (± 0.16) 14.7 (± 9.2) H7 (A/Netherlands/219/03) AT10-002 0.59 (± 0.15) 0.25 (± 0.05) 441 (± 109) AT10-005 —— — BiFlu 0.57 (± 0.12) 0.21 (± 0.01) 375 (± 88) H9 (A/Hong Kong/1073/1999) AT10-002 —— —Fig. 4. Capture SPR analysis of HA binding. (A) HA binding of BiFlu, AT10- AT10-005 0.81 (± 0.06) 21.6 (± 0.1) 27,000 (± 2,000) 002, and AT10-005, immobilized on an anti-HC–coated spot. (B) HA binding BiFlu 0.94 (± 0.02) 27.1 (± 0.2) 29,000 (± 1,000) of BiFlu, AT10-002, and AT10-005, immobilized on an anti-lambda light- chain coated spot. First, antibody is injected, followed by HA antigens (H3, 4 −1 −1 −5 −1 ka in 10 sec ·M , kd in 10 sec , KD in picomolars; —, no binding then H1), and then anti-light chain antibody (either anti-kappa or anti- detected. Data represent mean ± SD of at least two independent experi- lambda). Because BiFlu is twice as large as IgG, BiFlu gives a greater capture ments. SPR curves and fits are shown in SI Appendix, Figs. S4 (H1 + H3 response when binding to the capture antibody (anti-HC or anti-lambda). binding), S5 (H7), and S6 (H9). Kinetic constants are shown in Table 2.

4of6 | www.pnas.org/cgi/doi/10.1073/pnas.1408605111 Wagner et al. Downloaded by guest on September 23, 2021 Table 3. IC50 values for influenza neutralization Virus AT10-002 AT10-005 BiFlu

H1N1 (A/Hawaii/31/2007) — 1.29 (± 0.53) 0.36 (± 0.25) H3N2 (A/Neth./177/2008) 0.76 (± 0.27) — 1.37 (± 0.41)

IC50 values in nanomolars. —, no neutralizing activity detected. Data rep- resent mean ± SD of at least two independent experiments.

exploration of many different combinations of C-C–linked bis- Fig. 5. Virus neutralization assays. (A) Neutralization of influenza A H1N1 pecific antibodies than genetic fusions would allow. virus (strain: A/Hawaii/31/2007). (B) Neutralization of influenza A H3N2 virus Antibodies against the stem region of Influenza virus HA (strain: A/Netherlands/177/2008). Data represent mean ± SD of at least two independent experiments. antigens provide protection against virus in a prophylactic setting in animal models (30, 51) and based on this property, they are being tested in clinical trials. BiFlu combines the activities of two To test the integrity of the BiFlu molecule at the time of viral broadly neutralizing antibodies into a single unit. From a de- infection, we performed an anti-human IgG HC Western blot velopmental and regulatory perspective, combining two anti- (Fig. 6C), demonstrating that BiFlu remained intact as a dimer. bodies into a single drug makes development less complex and The human IgG concentration in the mice, 24 h after injection, more cost-effective, because preclinical and clinical testing will was determined with ELISA (Fig. 6D). For both the BiFlu and be reduced to a single molecule. the antibody mixture group, we found approximately 9.5 μgof Materials and Methods human IgG/mL, indicating that BiFlu remained in the circulation at similar levels as the single antibodies. Isolation and Selection of Antiinfluenza Antibodies from Human B Cells. Im- mortalization of human memory B cells was performed as described (36, 37). BiFlu has the ability to bind H1, H3, H7, and H9 HAs, and Human memory B cells were isolated using FACS, out of peripheral blood exhibits neutralization potency against both H1 and H3. The activity mononucleated cells (PBMCs) from an influenza vaccinated donor, and im- of BiFlu is consistent with the combined activities of the individual mortalized through retroviral transduction with a bicistronic construct cod- parental antibodies, which neutralize a wide range of influenza ing for Bcl6 and Bcl-xL. The use of human PBMCs was approved by the INFLAMMATION

subtypes. Thus, we have created a bispecific antibody dimer capable Medical Ethical Committee of the Academic Medical Center and was IMMUNOLOGY AND of broad HA binding and potentially broad neutralization potency. Discussion We have presented a bispecific antibody format, in which two full-length IgG antibodies are joined at their C termini. One advantage of this format is the stability of the C-C–linked IgG heterodimer, produced with minimal modification of the native IgG structure. The chemical structure of the C-C linkage includes the sortase recognition site, plus a triazole moiety resulting from the click reaction and eventual linker peptides. This product, like any other nonnative protein modification, could be immunogenic, a notion that would require testing in a human host. The C-C–linked IgG heterodimer stands out from other IgG- scFv formats; the latter bispecifics, in which the extra domains are genetically fused to the antibody backbone, are often un- stable and aggregation-prone (46, 47). Several formats for asym- metric bispecific IgG antibody formats now exist. Antibody asymmetry is facilitated through engineering of the CH3 domain (13–16) or the hinge region of the antibody (17, 18), promoting heterodimerization of the constant domains. Some of the constant domain mutations required to enable IgG heterodimerization compromise stability and may affect binding to Fc receptors as well (48). Solving these issues requires extensive antibody engi- neering (48). Also, an asymmetric IgG binds monovalently to its target because it contains only one copy of each variable domain, which may affect its activity. Production of these asymmetric IgG Fig. 6. In vivo protective activity of BiFlu. Kaplan–Meier survival curves (A) antibodies requires either a mild-reduction step to convert and body weight loss (mean ± SD) of C57BL/6J mice (B) that were i.v. injected + homodimers into heterodimers (17, 18) or coexpression of two with either BiFlu (2 mg/kg), AT10-002 AT10-005 (1 mg/kg each), or rituximab (1 mg/kg). Twenty-four hours later, the mice were challenged different antibodies (13–16), adding further complication. 4.5 intranasally with 50 μLofa10 TCID50 H1N1 (A/PR/8/1934) preparation. In contrast, the preparation of the C-C–fused IgG hetero- Compared with mice treated with control antibody (rituximab), the survival dimers lends itself to large-scale manufacturing without loss of and body weight loss of mice treated with BiFlu or the AT10-002 + AT10-005 product quality or the need for elaborate optimization. The antibody mixture was significantly improved (survival: P < 0.05; body weight antibodies to be joined can be expressed separately as full-length BiFlu P < 0.01 from day 6, AT10-002 + AT10-005 P < 0.01 from day 3). No + – antibodies, and the coupling reactions occur under physiological significant difference was found between BiFlu- and AT10-002 AT10-005 treated mice. (C) Integrity of BiFlu in mouse plasma, 24 h after injection. conditions. Likewise, sortase-catalyzed reactions enable large- Samples containing 0.25 μg of antibody are heated for 10 min at 55 °C and scale preparation of modified biomolecules (49) and immuno- analyzed with anti-IgG HC Western blotting. (D) Antibody concentration in toxins (50). A panel of sortase-modified antibodies equipped mouse plasma, 24 h after injection. IgG concentration of all mice in each with click handles far more readily facilitates combinatorial group (n = 6) is determined by ELISA.

Wagner et al. PNAS Early Edition | 5of6 Downloaded by guest on September 23, 2021 contingent on informed consent. B cells with reactivity to more than one HA length HA-antigens are injected over the chip surface in cycles of concate- type were characterized for HA recognition by ELISA and binding to HA- nated injections. Data is processed with SprintX software (IBIS Technologies). expressing cells. Virus Neutralization Assay. MDCK-SIAT cells are incubated with virus and Preparation of SDS-PAGE and Western Blot Samples. Unless indicated other- antibody. Cells are fixed 24 h after infection. The amount of infected cells is wise, samples are prepared in 1x XT-sample buffer (BIORAD) and heated for detected with FITC-labeled antiinfluenza nuclear protein (NP) antibody; total 10 min at 95 °C. If indicated, samples are reduced by adding 10 mM DTT. cell count is measured with DAPI staining. Mini-Protean TGX precast 4-20% gradient gels (BIORAD) were used for Additional detailed information is described in SI Appendix, SI Materials electrophoresis. and Methods.

SPR. SPR is performed on an IBIS Mx96 instrument (IBIS Technologies). Anti- ACKNOWLEDGMENTS. We thank Carla Guimaraes and Juan-Jose Cragnolini for their assistance with the purification of sortase and setting up the sortase human IgG heavy and anti-human light chain antibodies are immobilized on labeling assays. Chris Theile is acknowledged for synthesis of GGG-DIBAC. an amine-specific E2S gold-film SPR chip (Ssens Technologies) using a CFM This work is supported by the FLUNIVAC programme, European Commission microfluidics spotter device (Wasatch Microfluidics). Antibodies and full- FP7, Project no. 602604.

1. Nelson AL, Dhimolea E, Reichert JM (2010) Development trends for human mono- 27. Nobusawa E, et al. (1991) Comparison of complete sequences and clonal antibody therapeutics. Nat Rev Drug Discov 9(10):767–774. receptor-binding properties among 13 serotypes of hemagglutinins of influenza 2. Reichert JM (2013) Which are the antibodies to watch in 2013? MAbs 5(1):1–4. Aviruses.Virology 182(2):475–485. 3. Reichert JM (2011) Bispecific antibodies and ADCs: Once and future kings? MAbs 3(4): 28. Webster RG, Bean WJ, Gorman OT, Chambers TM, Kawaoka Y (1992) Evolution and 329–330. ecology of influenza A viruses. Microbiol Rev 56(1):152–179. 4. Vincent KJ, Zurini M (2012) Current strategies in antibody engineering: Fc engineer- 29. Sui J, et al. (2009) Structural and functional bases for broad-spectrum neutralization ing and pH-dependent antigen binding, bispecific antibodies and antibody drug of avian and human influenza A viruses. Nat Struct Mol Biol 16(3):265–273. – conjugates. Biotechnol J 7(12):1444 1450. 30. Ekiert DC, et al. (2009) Antibody recognition of a highly conserved influenza virus 5. Kontermann RE (2012) Dual targeting strategies with bispecific antibodies. MAbs 4(2): epitope. Science 324(5924):246–251. – 182 197. 31. Ekiert DC, et al. (2011) A highly conserved neutralizing epitope on group 2 influenza 6. May C, Sapra P, Gerber HP (2012) Advances in bispecific biotherapeutics for the A viruses. Science 333(6044):843–850. – treatment of cancer. Biochem Pharmacol 84(9):1105 1112. 32. De Marco D, et al. (2012) A non-VH1-69 heterosubtypic neutralizing human mono- 7. Holliger P, Prospero T, Winter G (1993) “Diabodies”: Small bivalent and bispecific clonal antibody protects mice against H1N1 and H5N1 viruses. PLoS ONE 7(4):e34415. antibody fragments. Proc Natl Acad Sci USA 90(14):6444–6448. 33. Friesen RHE, et al. (2014) A common solution to group 2 influenza virus neutraliza- 8. Schoonjans R, et al. (2001) A new model for intermediate molecular weight re- tion. Proc Natl Acad Sci USA 111(1):445–450. combinant bispecific and trispecific antibodies by efficient heterodimerization of 34. Lambert LC, Fauci AS (2010) Influenza vaccines for the future. N Engl J Med 363(21): single chain variable domains through fusion to a Fab-chain. Biomol Eng 17(6): 2036–2044. 193–202. 35. Wang TT, Palese P (2009) Universal epitopes of influenza virus hemagglutinins? Nat 9. Kellner C, et al. (2008) A novel CD19-directed recombinant bispecific antibody de- Struct Mol Biol 16(3):233–234. rivative with enhanced immune effector functions for human leukemic cells. 36. Kwakkenbos MJ, et al. (2010) Generation of stable monoclonal antibody-producing B J Immunother 31(9):871–884. cell receptor-positive human memory B cells by genetic programming. Nat Med 16(1): 10. Coloma MJ, Morrison SL (1997) Design and production of novel tetravalent bispecific – antibodies. Nat Biotechnol 15(2):159–163. 123 128. 11. Wu C, et al. (2007) Simultaneous targeting of multiple disease mediators by a dual- 37. Kwakkenbos MJ, et al. (2014) Genetic manipulation of B cells for the isolation of rare – variable-domain immunoglobulin. Nat Biotechnol 25(11):1290–1297. therapeutic antibodies from the human repertoire. Methods 65(1):38 43. 12. Dong J, et al. (2011) Stable IgG-like bispecific antibodies directed toward the type I 38. Lingwood D, et al. (2012) Structural and genetic basis for development of broadly – insulin-like growth factor receptor demonstrate enhanced ligand blockade and anti- neutralizing influenza antibodies. Nature 489(7417):566 570. tumor activity. J Biol Chem 286(6):4703–4717. 39. Avnir Y, et al. (2014) Molecular signatures of hemagglutinin stem-directed hetero- 13. Ridgway JBB, Presta LG, Carter P (1996) ’Knobs-into-holes’ engineering of antibody subtypic human neutralizing antibodies against influenza A viruses. PLoS Pathog CH3 domains for heavy chain heterodimerization. Prot. Eng 9(7):617–621. 10(5):e1004103. 14. Davis JH, et al. (2010) SEEDbodies: Fusion proteins based on strand-exchange en- 40. Throsby M, et al. (2008) Heterosubtypic neutralizing monoclonal antibodies cross- + gineered domain (SEED) CH3 heterodimers in an Fc analogue platform for asymmetric protective against H5N1 and H1N1 recovered from human IgM memory B cells. PLoS binders or immunofusions and bispecific antibodies. Protein Eng Des Sel 23(4): ONE 3(12):e3942. 195–202. 41. van Geel R, Pruijn GJM, van Delft FL, Boelens WC (2012) Preventing thiol-yne addition 15. Gunasekaran K, et al. (2010) Enhancing antibody Fc heterodimer formation through improves the specificity of strain-promoted azide-alkyne cycloaddition. Bioconjug electrostatic steering effects: Applications to bispecific molecules and monovalent Chem 23(3):392–398. IgG. J Biol Chem 285(25):19637–19646. 42. Ilangovan U, Ton-That H, Iwahara J, Schneewind O, Clubb RT (2001) Structure of 16. Schaefer W, et al. (2011) Immunoglobulin domain crossover as a generic approach for sortase, the transpeptidase that anchors proteins to the cell wall of Staphylococcus the production of bispecific IgG antibodies. Proc Natl Acad Sci USA 108(27): aureus. Proc Natl Acad Sci USA 98(11):6056–6061. – 11187 11192. 43. Ekkebus R, et al. (2013) On terminal alkynes that can react with active-site cysteine 17. Strop P, et al. (2012) Generating bispecific human IgG1 and IgG2 antibodies from any nucleophiles in . J Am Chem Soc 135(8):2867–2870. – antibody pair. J Mol Biol 420(3):204 219. 44. Shields RL, et al. (2001) High resolution mapping of the binding site on human IgG1 18. Labrijn AF, et al. (2013) Efficient generation of stable bispecific IgG1 by controlled for Fc γ RI, Fc γ RII, Fc γ RIII, and FcRn and design of IgG1 variants with improved Fab-arm exchange. Proc Natl Acad Sci USA 110(13):5145–5150. binding to the Fc γ R. J Biol Chem 276(9):6591–6604. 19. Spiess C, et al. (2013) Bispecific antibodies with natural architecture produced by co- 45. Tsuchiya S, et al. (1980) Establishment and characterization of a human acute culture of expressing two distinct half-antibodies. Nat Biotechnol 31(8): monocytic leukemia cell line (THP-1). Int J Cancer 26(2):171–176. 753–758. 46. Demarest SJ, Glaser SM (2008) Antibody therapeutics, antibody engineering, and the 20. Witte MD, et al. (2012) Preparation of unnatural N-to-N and C-to-C protein fusions. merits of protein stability. Curr Opin Drug Discov Devel 11(5):675–687. Proc Natl Acad Sci USA 109(30):11993–11998. 47. Miller BR, et al. (2010) Stability engineering of scFvs for the development of bispecific 21. Navarre WW, Schneewind O (1999) Surface proteins of gram-positive bacteria and and multivalent antibodies. Protein Eng Des Sel 23(7):549–557. mechanisms of their targeting to the cell wall envelope. Microbiol Mol Biol Rev 63(1): 48. Von Kreudenstein TS, et al. (2013) Improving biophysical properties of a bispecific 174–229. antibody scaffold to aid developability: quality by molecular design. MAbs 5(5): 22. Popp MW, Ploegh HL (2011) Making and breaking peptide bonds: Protein engi- – neering using sortase. Angew Chem Int Ed Engl 50(22):5024–5032. 646 654. 23. Antos JM, et al. (2009) Site-specific N- and C-terminal labeling of a single polypeptide 49. Steinhagen M, Zunker K, Nordsieck K, Beck-Sickinger AG (2013) Large scale modifi- using of different specificity. J Am Chem Soc 131(31):10800–10801. cation of biomolecules using immobilized sortase A from . Bio- – 24. Möhlmann S, Mahlert C, Greven S, Scholz P, Harrenga A (2011) In vitro sortagging of org Med Chem 21(12):3504 3510. an antibody fab fragment: Overcoming unproductive reactions of sortase with water 50. Kornberger P, Skerra A (2014) Sortase-catalyzed in vitro functionalization of a HER2- and lysine side chains. ChemBioChem 12(11):1774–1780. specific recombinant Fab for tumor targeting of the plant cytotoxin gelonin. MAbs 25. Guimaraes CP, et al. (2013) Site-specific C-terminal and internal loop labeling of 6(2):354–366. proteins using sortase-mediated reactions. Nat Protoc 8(9):1787–1799. 51. Koudstaal W, et al. (2009) Pre- and postexposure use of human monoclonal antibody 26. Becer CR, Hoogenboom R, Schubert US (2009) Click chemistry beyond metal-catalyzed against H5N1 and H1N1 influenza virus in mice: Viable alternative to oseltamivir. cycloaddition. Angew Chem Int Ed Engl 48(27):4900–4908. J Infect Dis 200(12):1870–1873.

6of6 | www.pnas.org/cgi/doi/10.1073/pnas.1408605111 Wagner et al. Downloaded by guest on September 23, 2021