- Email: [email protected]
Designer Proteins to Trigger Cell Death
Mycobacterium (M. leprae), involves loss of intraepidermal nerve fibers and pain receptors in clinical skin biopsies (Facer et al., 2000). The absence of inflammatory pain may thus result from effects of mycolactone on both inflammatory cells and cutaneous nociceptors. The mycolactone concentrations required to stimulate AT2 receptors are in the 3 mg/ml range (4 mmol/l), whereas the level required to cause hyperpolarization in vitro and that in the tissue after infection are one to two orders of magnitude lower. Clearly, more evidence is needed to confirm that mycolactone truly is a bona fide AT2 receptor agonist at relevant in vivo levels. It is also important to know the exact location of the AT2 receptors, and why mycolactone, despite binding to AT1 receptors, does not stimulate these receptors. Studies with well-established AT2 receptor agonists like C21
and clinicopathological correlations may help to answer these questions.
rion, D., Clere, N., Paille, V., et al. (2014). Cell 157, this issue, 1565–1576.
ACKNOWLEDGMENTS
Noe¨l, J., Sandoz, G., and Lesage, F. (2011). Channels (Austin) 5, 402–409.
P.A. is a member of Spinifex Pharmaceuticals Scientific Advisory Board and has participated in preclinical and clinical studies with EMA401.
Rice, A.S., Dworkin, R.H., McCarthy, T.D., Anand, P., Bountra, C., McCloud, P.I., Hill, J., Cutter, G., Kitson, G., Desem, N., and Raff, M.; EMA401-003 study group (2014). Lancet 383, 1637–1647.
REFERENCES Anand, U., Facer, P., Yiangou, Y., Sinisi, M., Fox, M., McCarthy, T., Bountra, C., Korchev, Y.E., and Anand, P. (2013). Eur. J. Pain 17, 1012–1026.
Rompe, F., Artuc, M., Hallberg, A., Alterman, M., Stro¨der, K., Tho¨ne-Reineke, C., Reichenbach, A., Schacherl, J., Dahlo¨f, B., Bader, M., et al. (2010). Hypertension 55, 924–931. Verdonk, K., Danser, A.H.J., and van Esch, J.H.M. (2012a). Expert Opin. Investig. Drugs 21, 501–513.
Facer, P., Mann, D., Mathur, R., Pandya, S., Ladiwala, U., Singhal, B., Hongo, J., Sinicropi, D.V., Terenghi, G., and Anand, P. (2000). Pain 85, 231–238.
Verdonk, K., Durik, M., Abd-Alla, N., Batenburg, W.W., van den Bogaerdt, A.J., van Veghel, R., Roks, A.J., Danser, A.H.J., and van Esch, J.H.M. (2012b). Hypertension 60, 722–729.
Hall, B.S., Hill, K., McKenna, M., Ogbechi, J., High, S., Willis, A.E., and Simmonds, R.E. (2014). PLoS Pathog. 10, e1004061.
Zavattaro, E., Boccafoschi, F., Borgogna, C., Conca, A., Johnson, R.C., Sopoh, G.E., Dossou, A.D., Colombo, E., Clemente, C., Leigheb, G., and Valente, G. (2012). Histopathology 61, 224–236.
Marion, E., Song, O.-K., Christophe, T., Babonneau, J., Fenistein, D., Eyer, J., Letournel, F., Hen-
Designer Proteins to Trigger Cell Death Wayne J. Fairbrother1,* and Avi Ashkenazi2,* 1Department
of Early Discovery Biochemistry Immunology Genentech, Inc., 1 DNA Way, South San Francisco, CA 94080, USA *Correspondence: [email protected] (W.J.F.), [email protected] (A.A.) http://dx.doi.org/10.1016/j.cell.2014.06.010 2Cancer
Efforts to generate biologically active proteins by de novo computational design have been limited to creating functional sites within pre-existing scaffolds. Procko et al. use an innovative computational design approach coupled with in-vitro-targeted evolution to produce a potent polypeptide inhibitor of a viral Bcl-2-like protein. This novel inhibitor triggers apoptosis of virus-infected cells. Metazoan organisms employ a distinct type of programmed cell death, called apoptosis, to eliminate severely damaged or infected cells. The Bcl-2 gene family encodes proteins that control apoptotic signaling via the cell-intrinsic, mitochondrial pathway (Cory and Adams, 2002). Two structural subclasses of this family share Bcl-2 homology (BH) motifs: (1) BH3 proteins (e.g., Bim, Bid, Bad, Bmf,
Puma, and Noxa), which harbor a single BH motif and promote apoptosis; and (2) multi-BH proteins, which possess three or four BH regions and act either as apoptosis activators (e.g., Bax and Bak) or inhibitors (e.g., Bcl-2, Bcl-xL, Bcl-w, and Mcl-1). Some BH3 factors promote apoptosis by sequestering prosurvival Bcl-2 proteins from Bax and Bak, whereas others bind directly to Bax or
1506 Cell 157, June 19, 2014 ª2014 Elsevier Inc.
Bak to drive activation. The BH3 motif— an 26 amino acid a-helical peptide— interacts with a hydrophobic groove on the cognate binding partner. Viruses often encode orthologs of cellular antiapoptotic proteins to prevent host cell death and extend viral replication. Epstein-Barr virus (EBV) latently infects human B cells, contributing to cancers such as Burkitt’s lymphoma.
EBV encodes a Bcl-2-like (Azzarito et al., 2013). The de decoy, called BHRF1, which novo design approach preblocks host-cell apoptosis sented by Procko et al. (Henderson et al., 1993). (2014) has one important relaAlthough small-molecule intive advantage: the stabilizing hibitors of Bcl-2 proteins are protein scaffold provides being developed in the clinic favorable interactions with (Billard, 2013), polypeptide the target in addition to inhibitors might have differenthose provided by the functiating qualities; however, tional peptide motif itself. computational efforts to Indeed, the crystal structure design biologically active of the BHRF1:BINDI complex proteins, including Bcl-2 inshows that the incorporated hibitors, have been limited Bim-BH3 motif contributes (Fleishman et al., 2011; only 40% of the total Procko et al., 2013). In this BHRF1-interaction surface. issue of Cell, Procko et al. Importantly, directed evolu(2014) employ a compution of the designed scaffold tational design method (Corimproves stability, further reia et al., 2014), combined enhancing affinity. with in-vitro-directed evoluThe de novo computational tion, to generate a potent design of a protein that and selective polypeptide inbinds tightly and selectively hibitor of BHRF1 (Procko to the surface of a target and et al., 2014). exerts a specific biologic The computational design activity is an important and procedure developed by exciting advance—it imProcko et al. (2014) (Figproves substantially over preure 1)—dubbed Fold From vious approaches of transLoops (FFL)—requires as planting functional binding input the structure of a funcepitopes into an existing protional motif and a template tein scaffold. Upon modificaFigure 1. Schematic Overview of the Fold From Loops Computational Design Procedure as Applied to Generation of the BHRF1 scaffold topology that will tion to facilitate intracellular Inhibitor ‘‘fold’’ around this motif and entry, Procko et al. (2014) stabilize it. Procko et al. find that their inhibitor induces (2014) select the Bim-BH3 apoptosis of EBV-infected helix from the BHRF1:Bim-BH3 cocrystal (Leaver-Fay et al., 2011), and the designs cells. Moreover, on further modification structure (Kvansakul et al., 2010) as the are filtered for interface quality and to enable systemic delivery, they observe functional motif and folding nucleus. monomer stability. Ultimately, the au- that the inhibitor attenuates tumor growth Structural homologs of the three-helix thors test the 74 designs with the lowest and extends host survival in a mouse bundle target scaffold—a domain number of buried, unsatisfied hydrogen- xenograft model of EBV-positive human derived from a bacterial ribosome-recy- bonding atoms in the unbound monomer lymphoma. This represents the successcling factor—are then assembled around in a yeast display system. Of these, two ful design of an entirely novel protein the functional motif using a fragment- variants bind BHRF1 with apparent Kd inhibitor with a meaningful, on-target based approach. Subsequently, they values of 60 nM. Notably, the best biologic activity. perform multiple rounds of sequence computational predictor of binding activIn future studies, it will be interesting minimization and design. The template ity is the degree of similarity between the to explore whether this strategy can be scaffold guides the assembly of the conformations predicted ab initio and the applied to diverse types of protein unique sequences through both atom- designed structure. BHRF1 binders are interactions with different template scafpair distance constraints and topology. further optimized for affinity, specificity, folds. A remaining key challenge is the They then select 1,000 structures pos- and stability via directed evolution by development of clinically viable methods sessing the lowest energy from an yeast display to yield a stable, high-affinity for intracellular or systemic (preferably ensemble of 5,000 FFL-assembled (Kd = 220 pM) variant named BINDI. nonimmunogenic) delivery of designer Other methods seeking to constrain proteins. Assessing whether the new homologs and align them to the BimBH3 peptide in the BHRF1:Bim-BH3 conformations of functional protein motifs approach can be employed not only complex. The interface residues outside include stabilizing the helical arrangement for functional modulation but also the Bim-BH3 motif in the docked struc- of BH3 peptides by covalently linking for wider tasks, such as payload tures are optimized using ROSETTA pairs of residues on one face of the a helix delivery of chemotherapeutics or other Cell 157, June 19, 2014 ª2014 Elsevier Inc. 1507
biologically active compounds, will also be important.
Cory, S., and Adams, J.M. (2002). Nat. Rev. Cancer 2, 647–656.
REFERENCES
Fleishman, S.J., Corn, J.E., Strauch, E.-M., Whitehead, T.A., Karanicolas, J., and Baker, D. (2011). J. Mol. Biol. 413, 1047–1062.
Azzarito, V., Long, K., Murphy, N.S., and Wilson, A.J. (2013). Nat. Chem. 5, 161–173. Billard, C. (2013). Mol. Cancer Ther. 12, 1691– 1700. Correia, B.E., Bates, J.T., Loomis, R.J., Baneyx, G., Carrico, C., Jardine, J.G., Rupert, P., Correnti, C., Kalyuzhniy, O., Vittal, V., et al. (2014). Nature 507, 201–206.
Henderson, S., Huen, D., Rowe, M., Dawson, C., Johnson, G., and Rickinson, A. (1993). Proc. Natl. Acad. Sci. USA 90, 8479–8483. Kvansakul, M., Wei, A.H., Fletcher, J.I., Willis, S.N., Chen, L., Roberts, A.W., Huang, D.C.S., and Colman, P.M. (2010). PLoS Pathog. 6, e1001236.
1508 Cell 157, June 19, 2014 ª2014 Elsevier Inc.
Leaver-Fay, A., Tyka, M., Lewis, S.M., Lange, O.F., Thompson, J., Jacak, R., Kaufman, K., Renfrew, P.D., Smith, C.A., Sheffler, W., et al. (2011). Methods Enzymol. 487, 545–574. Procko, E., Hedman, R., Hamilton, K., Seetharaman, J., Fleishman, S.J., Su, M., Aramini, J., Kornhaber, G., Hunt, J.F., Tong, L., et al. (2013). J. Mol. Biol. 425, 3563–3575. Procko, E., Berguig, G.Y., Shen, B.W., Song, Y., Frayo, S., Convertine, A.J., Margineantu, D., Booth, G., Correia, B.E., Cheng, Y., et al. (2014). Cell 157, this issue, 1644–1656.
and clinicopathological correlations may help to answer these questions.
rion, D., Clere, N., Paille, V., et al. (2014). Cell 157, this issue, 1565–1576.
ACKNOWLEDGMENTS
Noe¨l, J., Sandoz, G., and Lesage, F. (2011). Channels (Austin) 5, 402–409.
P.A. is a member of Spinifex Pharmaceuticals Scientific Advisory Board and has participated in preclinical and clinical studies with EMA401.
Rice, A.S., Dworkin, R.H., McCarthy, T.D., Anand, P., Bountra, C., McCloud, P.I., Hill, J., Cutter, G., Kitson, G., Desem, N., and Raff, M.; EMA401-003 study group (2014). Lancet 383, 1637–1647.
REFERENCES Anand, U., Facer, P., Yiangou, Y., Sinisi, M., Fox, M., McCarthy, T., Bountra, C., Korchev, Y.E., and Anand, P. (2013). Eur. J. Pain 17, 1012–1026.
Rompe, F., Artuc, M., Hallberg, A., Alterman, M., Stro¨der, K., Tho¨ne-Reineke, C., Reichenbach, A., Schacherl, J., Dahlo¨f, B., Bader, M., et al. (2010). Hypertension 55, 924–931. Verdonk, K., Danser, A.H.J., and van Esch, J.H.M. (2012a). Expert Opin. Investig. Drugs 21, 501–513.
Facer, P., Mann, D., Mathur, R., Pandya, S., Ladiwala, U., Singhal, B., Hongo, J., Sinicropi, D.V., Terenghi, G., and Anand, P. (2000). Pain 85, 231–238.
Verdonk, K., Durik, M., Abd-Alla, N., Batenburg, W.W., van den Bogaerdt, A.J., van Veghel, R., Roks, A.J., Danser, A.H.J., and van Esch, J.H.M. (2012b). Hypertension 60, 722–729.
Hall, B.S., Hill, K., McKenna, M., Ogbechi, J., High, S., Willis, A.E., and Simmonds, R.E. (2014). PLoS Pathog. 10, e1004061.
Zavattaro, E., Boccafoschi, F., Borgogna, C., Conca, A., Johnson, R.C., Sopoh, G.E., Dossou, A.D., Colombo, E., Clemente, C., Leigheb, G., and Valente, G. (2012). Histopathology 61, 224–236.
Marion, E., Song, O.-K., Christophe, T., Babonneau, J., Fenistein, D., Eyer, J., Letournel, F., Hen-
Designer Proteins to Trigger Cell Death Wayne J. Fairbrother1,* and Avi Ashkenazi2,* 1Department
of Early Discovery Biochemistry Immunology Genentech, Inc., 1 DNA Way, South San Francisco, CA 94080, USA *Correspondence: [email protected] (W.J.F.), [email protected] (A.A.) http://dx.doi.org/10.1016/j.cell.2014.06.010 2Cancer
Efforts to generate biologically active proteins by de novo computational design have been limited to creating functional sites within pre-existing scaffolds. Procko et al. use an innovative computational design approach coupled with in-vitro-targeted evolution to produce a potent polypeptide inhibitor of a viral Bcl-2-like protein. This novel inhibitor triggers apoptosis of virus-infected cells. Metazoan organisms employ a distinct type of programmed cell death, called apoptosis, to eliminate severely damaged or infected cells. The Bcl-2 gene family encodes proteins that control apoptotic signaling via the cell-intrinsic, mitochondrial pathway (Cory and Adams, 2002). Two structural subclasses of this family share Bcl-2 homology (BH) motifs: (1) BH3 proteins (e.g., Bim, Bid, Bad, Bmf,
Puma, and Noxa), which harbor a single BH motif and promote apoptosis; and (2) multi-BH proteins, which possess three or four BH regions and act either as apoptosis activators (e.g., Bax and Bak) or inhibitors (e.g., Bcl-2, Bcl-xL, Bcl-w, and Mcl-1). Some BH3 factors promote apoptosis by sequestering prosurvival Bcl-2 proteins from Bax and Bak, whereas others bind directly to Bax or
1506 Cell 157, June 19, 2014 ª2014 Elsevier Inc.
Bak to drive activation. The BH3 motif— an 26 amino acid a-helical peptide— interacts with a hydrophobic groove on the cognate binding partner. Viruses often encode orthologs of cellular antiapoptotic proteins to prevent host cell death and extend viral replication. Epstein-Barr virus (EBV) latently infects human B cells, contributing to cancers such as Burkitt’s lymphoma.
EBV encodes a Bcl-2-like (Azzarito et al., 2013). The de decoy, called BHRF1, which novo design approach preblocks host-cell apoptosis sented by Procko et al. (Henderson et al., 1993). (2014) has one important relaAlthough small-molecule intive advantage: the stabilizing hibitors of Bcl-2 proteins are protein scaffold provides being developed in the clinic favorable interactions with (Billard, 2013), polypeptide the target in addition to inhibitors might have differenthose provided by the functiating qualities; however, tional peptide motif itself. computational efforts to Indeed, the crystal structure design biologically active of the BHRF1:BINDI complex proteins, including Bcl-2 inshows that the incorporated hibitors, have been limited Bim-BH3 motif contributes (Fleishman et al., 2011; only 40% of the total Procko et al., 2013). In this BHRF1-interaction surface. issue of Cell, Procko et al. Importantly, directed evolu(2014) employ a compution of the designed scaffold tational design method (Corimproves stability, further reia et al., 2014), combined enhancing affinity. with in-vitro-directed evoluThe de novo computational tion, to generate a potent design of a protein that and selective polypeptide inbinds tightly and selectively hibitor of BHRF1 (Procko to the surface of a target and et al., 2014). exerts a specific biologic The computational design activity is an important and procedure developed by exciting advance—it imProcko et al. (2014) (Figproves substantially over preure 1)—dubbed Fold From vious approaches of transLoops (FFL)—requires as planting functional binding input the structure of a funcepitopes into an existing protional motif and a template tein scaffold. Upon modificaFigure 1. Schematic Overview of the Fold From Loops Computational Design Procedure as Applied to Generation of the BHRF1 scaffold topology that will tion to facilitate intracellular Inhibitor ‘‘fold’’ around this motif and entry, Procko et al. (2014) stabilize it. Procko et al. find that their inhibitor induces (2014) select the Bim-BH3 apoptosis of EBV-infected helix from the BHRF1:Bim-BH3 cocrystal (Leaver-Fay et al., 2011), and the designs cells. Moreover, on further modification structure (Kvansakul et al., 2010) as the are filtered for interface quality and to enable systemic delivery, they observe functional motif and folding nucleus. monomer stability. Ultimately, the au- that the inhibitor attenuates tumor growth Structural homologs of the three-helix thors test the 74 designs with the lowest and extends host survival in a mouse bundle target scaffold—a domain number of buried, unsatisfied hydrogen- xenograft model of EBV-positive human derived from a bacterial ribosome-recy- bonding atoms in the unbound monomer lymphoma. This represents the successcling factor—are then assembled around in a yeast display system. Of these, two ful design of an entirely novel protein the functional motif using a fragment- variants bind BHRF1 with apparent Kd inhibitor with a meaningful, on-target based approach. Subsequently, they values of 60 nM. Notably, the best biologic activity. perform multiple rounds of sequence computational predictor of binding activIn future studies, it will be interesting minimization and design. The template ity is the degree of similarity between the to explore whether this strategy can be scaffold guides the assembly of the conformations predicted ab initio and the applied to diverse types of protein unique sequences through both atom- designed structure. BHRF1 binders are interactions with different template scafpair distance constraints and topology. further optimized for affinity, specificity, folds. A remaining key challenge is the They then select 1,000 structures pos- and stability via directed evolution by development of clinically viable methods sessing the lowest energy from an yeast display to yield a stable, high-affinity for intracellular or systemic (preferably ensemble of 5,000 FFL-assembled (Kd = 220 pM) variant named BINDI. nonimmunogenic) delivery of designer Other methods seeking to constrain proteins. Assessing whether the new homologs and align them to the BimBH3 peptide in the BHRF1:Bim-BH3 conformations of functional protein motifs approach can be employed not only complex. The interface residues outside include stabilizing the helical arrangement for functional modulation but also the Bim-BH3 motif in the docked struc- of BH3 peptides by covalently linking for wider tasks, such as payload tures are optimized using ROSETTA pairs of residues on one face of the a helix delivery of chemotherapeutics or other Cell 157, June 19, 2014 ª2014 Elsevier Inc. 1507
biologically active compounds, will also be important.
Cory, S., and Adams, J.M. (2002). Nat. Rev. Cancer 2, 647–656.
REFERENCES
Fleishman, S.J., Corn, J.E., Strauch, E.-M., Whitehead, T.A., Karanicolas, J., and Baker, D. (2011). J. Mol. Biol. 413, 1047–1062.
Azzarito, V., Long, K., Murphy, N.S., and Wilson, A.J. (2013). Nat. Chem. 5, 161–173. Billard, C. (2013). Mol. Cancer Ther. 12, 1691– 1700. Correia, B.E., Bates, J.T., Loomis, R.J., Baneyx, G., Carrico, C., Jardine, J.G., Rupert, P., Correnti, C., Kalyuzhniy, O., Vittal, V., et al. (2014). Nature 507, 201–206.
Henderson, S., Huen, D., Rowe, M., Dawson, C., Johnson, G., and Rickinson, A. (1993). Proc. Natl. Acad. Sci. USA 90, 8479–8483. Kvansakul, M., Wei, A.H., Fletcher, J.I., Willis, S.N., Chen, L., Roberts, A.W., Huang, D.C.S., and Colman, P.M. (2010). PLoS Pathog. 6, e1001236.
1508 Cell 157, June 19, 2014 ª2014 Elsevier Inc.
Leaver-Fay, A., Tyka, M., Lewis, S.M., Lange, O.F., Thompson, J., Jacak, R., Kaufman, K., Renfrew, P.D., Smith, C.A., Sheffler, W., et al. (2011). Methods Enzymol. 487, 545–574. Procko, E., Hedman, R., Hamilton, K., Seetharaman, J., Fleishman, S.J., Su, M., Aramini, J., Kornhaber, G., Hunt, J.F., Tong, L., et al. (2013). J. Mol. Biol. 425, 3563–3575. Procko, E., Berguig, G.Y., Shen, B.W., Song, Y., Frayo, S., Convertine, A.J., Margineantu, D., Booth, G., Correia, B.E., Cheng, Y., et al. (2014). Cell 157, this issue, 1644–1656.