USO08067339B2

(12) United States Patent (10) Patent No.: US 8,067,339 B2 Prinz et al. (45) Date of Patent: Nov. 29, 2011

(54) SURFACE DISPLAY OF WHOLE 2002.0068325 A1 6/2002 Ng et al. ANTIBODIES IN EUKARYOTES 2003. O1863.74 A1 10, 2003 Hufton et al. 2004, OO18590 A1 1/2004 Gerngross et al. 2004f0074458 A1 4/2004 Nakamura et al. (75) Inventors: Bianka Prinz, Lebanon, NH (US); 2004/02196.11 A1 11/2004 Racher Natarajan Sethuraman, Hanover, NH 2004/0229.306 A1 11, 2004 Nett 2004/0230.042 A1 11/2004 Hamilton (US); Dongxing Zha, Etna, NH (US); 2005/0170452 A1 8, 2005 Wildt et al. Stefan Wildt, Lebanon, NH (US); Piotr 2005/0216958 A1 9, 2005 Yamane et al. Bobrowicz, Hanover, NH (US) 2005/0260729 A1 11/2005 Hamilton 2006,0040353 A1 2/2006 Davidson et al. Assignee: Merck Sharp & Dohme Corp., 2006, O141540 A1 6/2006 Miltenyi et al. (73) 2006/0211085 A1 9, 2006 Bobrowicz Rahway, NJ (US) 2006/0286637 A1 12/2006 Hamilton 2007/0020260 A1 1, 2007 Presta (*) Notice: Subject to any disclaimer, the term of this 2007/0O37248 A1 2/2007 Bobrowicz et al. patent is extended or adjusted under 35 2007/OO72262 A1 3/2007 Nett et al. 2009,0005264 A1 1/2009 Rakestraw et al. U.S.C. 154(b) by 0 days. 2009.0137416 A1 5/2009 Fandlet al. (21) Appl. No.: 12/489,900 2010/0331192 A1* 12/2010 Zha et al...... 506.1 FOREIGN PATENT DOCUMENTS (22) Filed: Jun. 23, 2009 EP 1324040 T 2003 EP 1415 158 6, 2005 (65) Prior Publication Data WO OOf 42176 T 2000 WO O2/O57423 T 2002 US 201O/OOO9866A1 Jan. 14, 2010 WO 2004/057 002 T 2004 WO 2007/061631 5/2007 Related U.S. Application Data WO 2007 136865 11, 2007 WO 2008/OO6554 1, 2008 (60) Provisional application No. 61/208,583, filed on Feb. WO 2009/036379 3, 2009 25, 2009, provisional application No. 61/134.331, WO 2009/085135 T 2009 filed on Jul. 9, 2008. WO 2009/105.357 8, 2009 (51) Int. C. OTHER PUBLICATIONS CI2N 15/87 (2006.01) Weaver-Feldhaus, “Yeast mating for combinatorial Fab library gen C4OB 30/04 (2006.01) eration ... ', FEBS Letters (2004), vol. 564, pp. 24-34. GOIN33/53 (2006.01) De Wildt, “Heavy chain CDR3 optimization of a germline (52) U.S. Cl...... 506/9:435/7.2:435/455 encoded. . .”. Protein Engineering (1997), vol. 10, pp. 835-841. (58) Field of Classification Search ...... 506/9; 435/72, Berens, “Gene regulation by tetracyclines”, Eur, J. Biochem. (2003), 435/455 vol. 270, pp. 3109-3121. See application file for complete search history. Bessette, “Rapid isolation of high-affinity protein binding . . . . Protein Engineering. Design & Selection (2004), vol. 17, pp. 731 References Cited 739. (56) Bobrowicz, "Engineering of an artificial glycosylation pathway . . . . U.S. PATENT DOCUMENTS Glycobiology (2004), vol. 14, pp. 757-766. Bobrowicz, “Isolation of three contiguous genes...”. Yeast (1997), 4,816,397 A 3, 1989 BOSS et al. 4,816,567 A 3/1989 Cabilly et al. vol. 13, pp. 819-828. 4,818,700 A 4, 1989 Cregg et al. Boder, “Yeast surface display for screening...”. Nature Biotechnol 5,693,762 A 12/1997 Queen et al. ogy (1997), vol. 15, pp. 553-557. 5,700,678 A 12/1997 Toyoshima et al. Boder, “Directed evolution of antibody fragments . . . . PNAS 5,733,757 A 3/1998 Barbas, III et al. (2000), vol. 97, pp. 10701-10705. 5,772,245 A 6, 1998 Muhlhausen 5,843,708 A 12/1998 Hardman et al. (Continued) 5,874,247 A 2/1999 Toyoshima et al. 5,985,626 A 1 1/1999 Barbas, III et al. 6,114,147 A 9, 2000 Frenken et al. Primary Examiner — Lynn Bristol 6,180,370 B1 1/2001 Queen et al. (74) Attorney, Agent, or Firm — John David Reilly 6,300,065 B1 10/2001 Kieke et al. 6,368,839 B1 4/2002 Barbas, III et al. 6,632,927 B2 10/2003 Adair et al. 6,699,658 B1 3/2004 Wittrup et al. (57) ABSTRACT 6,872,392 B2 3/2005 Nakamura et al. 6,919, 183 B2 7/2005 Fandlet al. Methods for display of recombinant whole immunoglobulins 6,949,372 B2 9/2005 Betenbaugh et al. or immunoglobulin libraries on the surface of eukaryote host 7,029,872 B2 4/2006 Gerngross cells, including yeast and filamentous fungi, are described. 7,105,554 B2 9, 2006 Orchard et al. The methods are useful for screening libraries of recombinant 7,132,273 B1 1 1/2006 Choi et al. 7,166,423 B1 1/2007 Miltenyi et al. immunoglobulins in eukaryote host cells to identify immu 7,198.921 B2 4/2007 Miura et al. noglobulins that are specific for an antigen of interest. 7,259,007 B2 8, 2007 Bobrowicz et al. 7,435,553 B2 10/2008 Fandlet al. 7,479,389 B2 1/2009 Nett et al. 11 Claims, 22 Drawing Sheets US 8,067,339 B2 Page 2

OTHER PUBLICATIONS Markland, “Selection for Protease Inhibitors...'. Methods in Enzy mology (1996), vol. 267, pp. 28-51. Caldas, “Design and synthesis of germline-based hemi-humanized Marks, “By-passing immunization: . . . . J. Mol. Biol. (1991), vol. single chain...”. Protein Engineering (2000), vol. 13, pp. 353-360. 222, pp. 581-597. Chiba, “Production of human compatible high mannose-type...”. J. Matthews, "Substrate phage: selection of protease substrates . . .”. Biol. Chem. (1998), vol. 273. pp. 26298-26304. Choi. “Use of combinatorial genetic libraries...”. PNAS (2003), vol. Science (1993), vol. 260, pp. 1113-1117. 100, pp. 5022-5027. Mazor, “Isolation of engineered, full-length antibodies...”. Nature Choo, “Designing DNA-binding proteins on the surface...”. Curr. Biotech. (2007), vol. 25, pp. 563-565. Opin. in Biotech. (1995), vol. 6, pp. 431-436. Mergler, “Development of a bisphenol A-adsorbing yeast. . . . Appl. Coloma, “Position effects of variable region carbohydrate”. J. of Microbiol. Biotech. (2004), vol. 63, pp. 418-421. Immunol. (1999), vol. 162, pp. 2162-2170. Mille, “Identification of a new family of genes ...”. J. Biol. Chem. Cosano, "Cloning and sequence analysis of the Pichia pastoris . . . . (2008), vol. 283, pp. 9724-9736. Yeast (1998), vol. 14, pp. 861-867. Mottershead, “Baculoviral display of functional sclfv...”. Biochem. Cox, 'Phagocytic signaling strategies: . ..”. Immunology (2001), and Biophys. Res. Comm. (2000), vol. 275, pp. 84-90. vol. 13, pp. 339-345. Nett, “Cloning and disruption of the PpURA5 gene . . . . Yeast Daeron, “Fc Receptor biology”. Ann. Rev. Immunol. (1997), vol. 15. (2003), vol. 20, pp. 1279-1290. p. 203-234. Nett, “Cloning and disruption of the Pichia pastoris . . . . Yeast Damasceno, "Cooverexpression of chaperones for enhanced secre (2005), vol. 22, pp. 295-304. tion...”. Appl Microbiol. Biotechnol. (2007), vol. 74, pp. 381-389. Nizard, "Prolonged display or rapid internalization...”. Protein Eng. Daugherty, “Quantitative analysis of the effect of the mutation...”. (2001), vol. 14, pp. 439-446. PNAS (2000), vol.97, pp. 2029-2034. Phizicky, "Protein-protein interactions:...'. Microbiol. Rev. (1995), De Groot, Genome-wide identification of fungal . . . . Yeast (2003), vol. 59, pp. 94-123. vol. 20, pp. 781-796. DiRienzo, “The outer membrane proteins of gram-negative . . . . Rakestraw, "A flow cytometric assay for Screening . . . . Biotech. Ann. Rev. Biochem. (1978), vol. 47, pp. 481-532. Prog. (2006), vol. 22, pp. 1200-1208. Ellman, “Combinatorial thinking in chemistry . . .”. PNAS (1997), Ravetch, “Fc receptors'. Curr. Opin. in Immunol. (1997), vol. 9, pp. vol. 94, pp. 2779-2782. 121-125. Francisco, “Production and fluorescence-activated cell sorting . . . . Rehberg, “Specific molecular activities of recombinant...”. J. Biol. PNAS (1993), vol. 90, pp. 10444-10448. Chem. (1982), vol. 257, pp. 11497-11502. Yamane-Ohnuki, "Establishment of FUT8 knockout Chinese ham Ren, “Displayofadenoregulin with a novel Pichiapastoris...'. Mol. ster...”. Biotech. and Bioeng., vol. 87 (2004), pp. 614-622. Biotech. (2007), vol. 35, pp. 103-108. Geoffroy, “A new phage display system...”. Gene, vol. 151 (1994), Riechmann, 'Phage display and selection of a site-directed random pp. 109-113. ized ... ', Biochemistry (1993), vol. 32, pp. 8848-8855. Georgiou, "Display of heterologous proteins . . .”. Nature Biotech. Ryckaert, “Fishing for lectins from diverse sequence (1997), vol. 15, pp. 29-34. libraries: . . . .” 191st meeting of the Belgian Society of Biochemistry Hamilton, “Production of complex human...”. Science (2003), vol. and Molecular Biology. . . . . Vrije Universiteit Brussel, Belgium 301, pp. 1244-1246. (Dec. 2, 2005), Abstract. Hamilton, “Humanization of yeast to produce ...”. Science (2006), Sblattero, “Exploiting recombination in single bacteria...”. Nature vol. 313, pp. 1441-1443. Biotech. (2000), vol. 18, pp. 75-80. Heyman, "Feedback regulation by IgG . . . . Immunology Letters Stemmer, “Rapid evolution of a protein...”. Nature (1994), vol.370, (2003), vol. 88, pp. 157-161. Holler, “In vitro evolution of a T cell...”. PNAS (2000), vol.97, pp. pp. 389-391. 5387-5392. Stemmer, “DNA shuffling by random fragmentation . . , PNAS Hoogenboom, “Designing and optimizing library . . . . Trends in (1994), vol. 91, pp. 10747-10751. Biotech. (1997), vol. 15, pp. 62-70. Streuli. “Target cell specificity of two species of human interferon Huo, “Co-expression of human protein...”. Protein Exp. and Purif. alpha ..., PNAS (1981), vol. 78, pp. 2848-2852. (2007), vol. 54, pp. 234-239. Swers, "Shuffled antibody libraries created by in vivo...”. Nucleic Inan, “Enhancement of protein secretion . . . . Biotech. and Bioeng. Acids Res. (2004), vol. 32, pp. 1-8. (2006), vol. 93, p. 771-778. Tanino, "Construction of a Pichia pastoris cell-surface display sys Jacobs, “Pichia surface display. . . .”. Biotech. Letters (2008), vol. tem...”. Biotech. Prog. (2006), vol. 22, pp. 989-993. 30, pp. 2173–218.1. Toman, “Production of recombinant human type 1 . . . . J. Biol. Jacobs, “Pichia surface display. . . . . Pichia Protein Expression Chem. (2000), vol. 275, pp. 23303-23309. Conference, San Diego, CA (Oct. 8-11, 2006), Abstract T23. Ulrich, “Expression studies of catalytic antibodies'. PNAS (1995), Jeffreris, "Glycosylation of recombinant...”. Biotech. Prog. (2005), vol.92, pp. 11907-1 1911. vol. 21, pp. 11-16. Vad, "Engineering of a Pichia pastoris expression system. . . . J. of Kanda, "Comparison of cell lines...”. Biotech. and Bioeng. (2006), Biotech. (2005), vol. 116, pp. 251-260. vol. 94, pp. 680-688. Walker, “Effect of redox environment on the in vitro . . . . J. Biol. Kanda, “Comparison of biological activity . . . . Glycobiology Chem. (1994), vol. 269, pp. 28487-28493. (2006), vol. 17, pp. 104-118. Wang, “Phage display of proteases . . . . Methods in Enzymology Keizer-Gunnink, ''Accumulation of properly folded human type (1996), vol. 267, pp. 52-68. III ... ', Matrix Biology (2000), vol. 19, pp. 29-36. Wang, "A new yeast display vector . . . . Protein Eng., Design & Knappik, "Engineered turns of a recombinant antibody. . . . Protein Selection (2005), vol. 18, pp. 337-343. Eng. (1995), vol. 8, pp. 81-89. Waterhouse, "Combinatorial infection and in vivo Kohler, “Continuous cultures of fused cells...”. Nature (1975), vol. recombination: . . . . Nucleic Acids Res. (1993), vol. 21, pp. 2265 256, pp. 495-497. 2266. Ladner, "Constrained peptides as binding entities'. Trends in Wysocki. “The Saccharomyces cerevisiae ACR3 gene...”. J. Biol. Biotech. (1995), vol. 13, pp. 426-430. Chem. (1997), vol. 272, pp. 30061-30066. Li, "Optimization of humanized IgGs . . .”. Nature Biotech. (2006), Wildt, “The humanization of N-glycosylation . . ... ', Nature Rev. vol. 24, p. 210-215. (2005), vol. 3, pp. 119-128. Lin Cereghino, “New selectable marker/auxotrophic hoststrain...”. Yoshino, “Efficient and stable display of functional proteins . . .”. Gene (2001), vol. 263, pp. 159-169. Appl. and Environ. Microbiol. (2006), vol. 72, pp. 465-471. Lowman, “Selecting high-affinity binding proteins . . . . Biochem Zhang, “Enhanced secretion of heterologous proteins. . . . Biotech. istry (1991), vol. 30, pp. 10832-10838. Prog. (2006), vol. 22, pp. 1090-1095. Maras, "Filamentous fungi as production organisms . . . . Kennard, “GPI-Anchored fusion proteins”. Methods in Biotech. Glycoconjugate Journa (1999), vol. 16, pp. 99-107. (1999), vol. 8, pp. 187-200. US 8,067,339 B2 Page 3

Ward, “The effector functions of immunoglobulins: ...”. Therapeu- Accession No. X56180, (Apr. 18, 2005), “P pastoris HIS4 gene for tic Immunology, (1995), vol. 2, pp. 77-94. trifunctional enzyme . . . . pp. 1-3. Chao, “Isolating and engineering human antibodies . . . . Nature Protocols (2006), vol. 1, pp. 755-768. * cited by examiner U.S. Patent US 8,067,339 B2

U.S. Patent Nov. 29, 2011 Sheet 2 of 22 US 8,067,339 B2

...... Si Construction of pGY642 PARG3-5 re P:GY24 bp) ScCYC FpGAPDH Prom Sii-A - Big NŠeys act repeat 27 pGLY620 pre?s Pme/6 (5100 bp) Pmel PppD1 (5)

x-EcoR -Sf

Xies pUC19 PpPDI; (5) -Phe

pGLY677 SecYC \\ Po: (7893 bp). N: g PpGAPDH Prom S. N tact repeot iac repeat Spe: S. RA PPD (5) pGLY617 5102 bp)

Continue on FG.2B FG.2A U.S. Patent Nov. 29, 2011 Sheet 3 of 22 US 8,067,339 B2

Fror F.G.2A

pGLY678 secyc II (8224 bp) Si- PCA}} p MYX YX YXY. Bg: PppD. (3) ce. ry Xho act repeat SPEE93

Afe/Paci Notif/ Polic SCo.MFpre hPD (Not/Blunt)

Sf

PppD. (5) pGLY642 SCYC \-Pac

St. (9759 bp) ScoMFpre+hPD ||

PpGAPDH Pro oci repeat

U.S. Patent Nov. 29, 2011 Sheet 4 of 22 US 8,067,339 B2

Construction of pGY2233 pGLY2231 (9039 bp) Sco8 pre-h: Rig

H Spe PpGAPDH Prof. s s oc fepeat

Not/fset ScotMF pre (Not/Blunt)

Sph: ScCYC TW-se pGLY2229 Si (1903.3 bp) FPEPs;pPEP4 (3)3 Scab Fpre-hCRP94. : Kpr: Spei Syg: s Spe: ot

pCLY2233 (982 bp) Scal pre-hGRP94ll

U.S. Patent Nov. 29, 2011 Sheet 6 of 22 US 8,067,339 B2

(OZ64)|008 U.S. Patent Nov. 29, 2011 Sheet 7 of 22 US 8,067,339 B2

inction meteor (29) 2-iss Bg (306) Pme (78)

Construction of p(LY 162 PpAOX

pGLY2028 -EcoRI (1246) (508 bp) NEcoR (258) Not (1285)

Soci (411) Si (420) Eg: (2240)

Fsei (1654) S&AA. Rsri (1761)(2044) EcoR blunt mei (2038 ABgill Afi (2050 Frag. Ngok V (2057) Sri (874, Al CF65 Pic 2. i (874) SpRO 10531 bp) i G Not blunt/Bgll

gArc: fkpni (3167) Xhoi (3507) & loci repeat PpGAPDH Prom Kpni (3547) R8 I repeat -2 Not (3905) $S- Bg|E (4398) Bornt; (4410) Xhoi (4404) Saci (41) SfS (420) EcoR (596) /fse (1654) Aoti (109) -Y \Kpni (1761) 269 5'-'80sriSN Rsri (1969)(2044) Hindi (8743)/ N Pme (2038) Sg8 (8720) AF: 2050) SF (3714) ANgok V (2057) EcoR (84.29) GY 62 Peci (2347) 3PRO p (Aria? Hind: (2854) Kpal (367)

icci pepeot Xho 3307) Spel (7276) PpIRA5 PpAOX Cé Kpni (3347) aca Fepeat Pmei (444) Kpnl (634) Sea q : (4849) FIG.5 Kont (6808), E" (486) U.S. Patent Nov. 29, 2011 Sheet 8 of 22 US 8,067,339 B2

yCLY24-1 (och1, pbs2, minn411, pno1, bmt2, suc2, ura5)

yGLY702 (PopditA:hPD) yGLY7045-FOA PopdiiA:thPD, uro5) pGLY1162 (PAOX1-TriMNS1) yGLY733 (PopditAihfD., PpPRO1:PAOX1-TriMNS1) pGF207t (PGAPDH-TriMNS1, PGAPDH-FB53 ) yGLY762 (Fppdf|A:thPD, PpPRO1:PAOX1-TriMNS1, PoPRO1: PGAPDH-TriMNS1: PGAPDH-FB53) 5-FOA yGLY2677 (Popdi1A:hPD., PpPRO1:PAOX1-TriMNS1, PppRO1:PGAPDH-TriMNS1: PGAPDH-FB53, ura5) pGLY2233 (Popep4A.h6RP94) yGLY2696 (PopditAihfpl. Pp.PRO1:PAOY1-TriMNS1, PpPRO1: PGAPDH-TriMNS1: PGAPDH-FB53, Popep44:hGRP94) Genealogy of Chaperone-humanized Strains F.G. 6

U.S. Patent Sheet 10 of 22

'//uolaIxo'da

!#auðu,fiuppauuado

*Es,*

(8/6/)!!!S U.S. Patent Nov. 29, 2011 Sheet 11 of 22 US 8,067,339 B2

U.S. Patent Nov. 29, 2011 Sheet 13 of 22 US 8,067,339 B2

U.S. Patent Nov. 29, 2011 Sheet 14 of 22 US 8,067,339 B2

yGLY2696/pGLY4136 10ng Her2 (prot A/SED1 fusion) Detection: goat anti-human IgG FIG. 12 U.S. Patent Nov. 29, 2011 Sheet 15 of 22 US 8 9 067,339 B2

U.S. Patent Nov. 29, 2011 Sheet 16 of 22 US 8,067,339 B2

yGLY2696/pGLY4116 50 ng Her2 (FoRIII/SED1 fusion protein) Detection: goat anti human IgG FIG. 14 U.S. Patent US 8,067,339 B2

(10%)íg

U.S. Patent Nov. 29, 2011 Sheet 18 of 22 US 8,067,339 B2

9L.''

U.S. Patent Nov. 29, 2011 Sheet 20 of 22 US 8,067,339 B2

8L U.S. Patent Nov. 29, 2011 Sheet 21 of 22 US 8,067,339 B2

0.02 WSCELLS 5 0.02 Y5757/pGLY4144-3 a 0.01 -Y5757/pGLY4144-2 -Y5757/pGLY4144 0.01 -Y2696/pGLY4144 s 0.00 to Too Too 53Onn yGLY5757/pGLY4144 anti-CD20 expressing strain F.G. 19A

0.02 WXXX XXXX XXXX XXX XXXW. re-r-s:CELS W. s s ...Y5434/p6LY4144-3 -Y5434/pcLY4144-2 -Y5434/pGLY4144-1 e EY2696/pcLY4144

"o "oo" too 530 in yGLY5434/pCLY4144 anti-Her2 expressing strain F.G. 19B U.S. Patent US 8,067,339 B2

(10%)|||ôg

US 8,067,339 B2 1. 2 SURFACE DISPLAY OF WHOLE land et al., Methods Enzymol. 267: 28-51 (1996): Matthews ANTIBODIES IN EUKARYOTES & Wells, Science 260: 1113-1117 (1993); Wang et al., Meth ods Enzymol. 267: 52-68 (1996)). CROSS REFERENCE TO RELATED Antibodies possessing desirable binding properties are APPLICATIONS selected by binding to immobilized antigen in a process called “panning. Phage bearing nonspecific antibodies are This patent application claims benefit of U.S. Provisional removed by washing, and then the bound phage are eluted and Application No. 61/208,583, filed 25 Feb. 2009 and U.S. amplified by infection of E. coli. This approach has been Provisional Application No. 61/134.331, filed 9 Jul. 2008. applied to generate antibodies against many antigens. 10 Nevertheless, phage display possesses several shortcom REFERENCE TO SEQUENCE LISTING ings. Although panning of antibody phage display libraries is SUBMITTED ELECTRONICALLY a powerful technology, it possesses several intrinsic difficul ties that limit its wide-spread successful application. For The sequence listing of the present application is Submitted example, some eukaryotic secreted proteins and cell Surface electronically via EFS-Web as an ASCII formatted sequence 15 proteins require post-translational modifications such as gly listing with a file name “GFIBIO0034USNP-SEQTXT cosylation or extensive disulfide isomerization, which are 09JUN2009.txt, creation date of Jun. 9, 2009, and a size of unavailable in bacterial cells. Furthermore, the nature of 61 KB. This sequence listing submitted via EFS-Web is part phage display precludes quantitative and direct discrimina of the specification and is herein incorporated by reference in tion of ligand binding parameters. For example, very high its entirety. affinity antibodies (Kds 1 nM) are difficult to isolate by pan ning, since the elution conditions required to break a very BACKGROUND OF THE INVENTION strong antibody-antigen interaction are generally harsh enough (e.g., low pH, high salt) to denature the phage particle (1) Field of the Invention sufficiently to render it non-infective. The present invention relates to methods for display of 25 Additionally, the requirement for physical immobilization whole immunoglobulins or libraries of immunoglobulins on of an antigen to a Solid Surface produces many artifactual the Surface of eukaryote host cells, including mammalian, difficulties. For example, high antigen Surface density intro plant, yeast, and filamentous fungal cells. The methods are duces avidity effects which mask true affinity. Also, physical useful for screening libraries of eukaryotic host cells that tethering reduces the translational and rotational entropy of produce recombinant immunoglobulins to identify particular 30 the antigen, resulting in a smaller DS upon antibody binding immunoglobulins with desired properties. The methods are and a resultant overestimate of binding affinity relative to that particularly useful for screening immunoglobulin libraries in for soluble antigen and large effects from variability in mix eukaryote host cells to identify host cells that express an ing and washing procedures lead to difficulties with repro immunoglobulin of interest at high levels, as well as host cells ducibility. Furthermore, the presence of only one to a few that express immunoglobulins that have high affinity for spe 35 antibodies per phage particle introduces substantial stochas cific antigens. tic variation, and discrimination between antibodies of simi (2) Description of Related Art lar affinity becomes impossible. For example, affinity differ The discovery of monoclonal antibodies has evolved from ences of six-fold or greater are often required for efficient hybridoma technology for producing the antibodies to direct discrimination (Riechmann & Weill, Biochem. 32: 8848-55 selection of antibodies from human cDNA or synthetic DNA 40 (1993)). Finally, populations can be overtaken by more rap libraries. This has been driven in part by the desire to engineer idly growing wild-type phage. In particular, since p is improvements in binding affinity and specificity of the anti involved directly in the phage life cycle, the presence of some bodies to improve efficacy of the antibodies. Thus, combina antibodies or bound antigens will prevent or retard amplifi torial library Screening and selection methods have become a cation of the associated phage. common tool for altering the recognition properties of pro 45 Additional bacterial cell surface display methods have teins (Ellman et al., Proc. Natl. Acad. Sci. USA 94: been developed (Francisco, et al., Proc. Natl. Acad. Sci. USA 2779-2782 (1997): Phizicky & Fields, Microbiol. Rev. 59: 90: 10444-10448 (1993); Georgiou et al., Nat. Biotechnol. 94-123 (1995)). The ability to construct and screen antibody 15:29-34 (1997)). However, use of a prokaryotic expression libraries in vitro promises improved control over the strength system occasionally introduces unpredictable expression and specificity of antibody-antigen interactions. 50 biases (Knappik & Pluckthun, Prot. Eng. 8: 81-89 (1995); The most widespread technique for constructing and Ulrich et al., Proc. Natl. Acad. Sci. USA 92: 11907-11911 screening antibody libraries is phage display, whereby the (1995); Walker & Gilbert, J. Biol. Chem 269: 28487-28493 protein of interest is expressed as a polypeptide fusion to a (1994)) and bacterial capsular polysaccharide layers presenta bacteriophage coat protein and Subsequently screened by diffusion barrier that restricts such systems to small molecule binding to immobilized or soluble biotinylated ligand. 55 ligands (Roberts, Annu. Rev. Microbiol. 50: 285-315 (1996)). Fusions are made most commonly to a minor coat protein, E. coli possesses a lipopolysaccharide layer or capsule that called the gene III protein (plII), which is present in three to may interfere sterically with macromolecular binding reac five copies at the tip of the phage. A phage constructed in this tions. In fact, a presumed physiological function of the bac way can be considered a compact genetic “unit', possessing terial capsule is restriction of macromolecular diffusion to the both the phenotype (binding activity of the displayed anti 60 cell membrane, in order to shield the cell from the immune body) and genotype (the gene coding for that antibody) in one system (DiRienzo et al., Ann. Rev. Biochem. 47: 481-532, package. Phage display has been Successfully applied to anti (1978)). Since the periplasm of E. coli has not evolved as a bodies, DNA binding proteins, protease inhibitors, short pep compartment for the folding and assembly of antibody frag tides, and enzymes (Choo & Klug, Curr. Opin. Biotechnol. 6: ments, expression of antibodies in E. coli has typically been 431-436 (1995); Hoogenboom, Trends Biotechnol. 15: 62-70 65 very clone dependent, with some clones expressing well and (1997); Ladner, Trends Biotechnol. 13: 426-430 (1995); others not at all. Such variability introduces concerns about Lowman et al., Biochemistry 30: 10832-10838 (1991); Mark equivalent representation of all possible sequences in an anti US 8,067,339 B2 3 4 body library expressed on the surface of E. coli. Moreover, closes a method for identifying cells that express a particular phage display does not allow some important posttransla protein by expressing in the cella Surface capture moiety and tional modifications such as glycosylation that can affect the protein wherein the capture moiety and the protein form a specificity or affinity of the antibody. About a third of circu complex which is displayed on the surface of the cell; U.S. lating monoclonal antibodies contain one or more N-linked 5 Pat. No. 6,114,147, which discloses a method for immobiliz glycans in the variable regions. In some cases it is believed ing proteins on the Surface of a yeast or fungal using a fusion that these N-glycans in the variable region may play a signifi protein consisting of a binding protein fused to a cell wall cant role in antibody function. protein which is expressed in the cell. The efficient production of monoclonal antibody therapeu The potential applications of engineering antibodies for the tics would be facilitated by the development of alternative test 10 diagnosis and treatment of human disease such as cancer systems that utilize lower eukaryotic cells, such as yeast cells. therapy, tumor imaging, sepsis are far-reaching. For these The structural similarities between B-cells displaying anti applications, antibodies with high affinity (i.e., Kds 10 nM) bodies and yeast cells displaying antibodies provide a closer and high specificity are highly desirable. Anecdotal evidence, analogy to in vivo affinity maturation than is available with as well as the a priori considerations discussed previously, filamentous phage. In particular, because lower eukaryotic 15 Suggests that phage display or bacterial display systems are cells are able to produce glycosylated proteins, whereas fila unlikely to consistently produce antibodies of sub-nanomolar mentous phage cannot, monoclonal antibodies produced in affinity. Also, antibodies identified using phage display or lower eukaryotic host cells are more likely to exhibit similar bacterial display systems may not be susceptible to commer activity in humans and other mammals as they do in test cial scale production in eukaryotic cells. To date, no system systems which utilize lower eukaryotic host cells. has been developed which can accomplish Such purpose, and Moreover, the ease of growth culture and facility of genetic be used. manipulation available with yeast will enable large popula Therefore, development of further protein expression sys tions to be mutagenized and screened rapidly. By contrast tems based on improved vectors and host cell lines in which with conditions in the mammalian body, the physicochemical effective protein display facilitates development of geneti conditions of binding and selection can be altered for a yeast 25 cally enhanced cells for recombinant production of immuno culture within a broad range of pH, temperature, and ionic globulins is a desirable objective. strength to provide additional degrees of freedominantibody engineering experiments. The development of yeast Surface BRIEF SUMMARY OF THE INVENTION display system for screening combinatorial protein libraries has been described. 30 One of the most powerful applications of the display sys U.S. Pat. Nos. 6,300,065 and 6,699,658 describe the devel temherein is its use in the arena of immunoglobulin engineer opment of a yeast surface display system for screening com ing. It has been shown that sclv antigen-binding units can be binatorial antibody libraries and a screen based on antibody expressed on the surface of lower eukaryote host cells with no antigen dissociation kinetics. The system relies on apparent loss of binding specificity and affinity (See for transfecting yeast with vectors that express an antibody or 35 example, U.S. Pat. No. 6,300,065). It has also been shown that antibody fragment fused to a yeast cell wall protein, using full-length antibodies can be captured and bound to the sur mutagenesis to produce a variegated population of mutants of face of hybridomas and CHO cells, for example (See U.S. Pat. the antibody or antibody fragment and then screening and Nos. 6,919,183 and 7,166,423). While antibodies and frag selecting those cells that produce the antibody or antibody ments thereof to many diverse antigens have been Success fragment with the desired enhanced phenotypic properties. 40 fully isolated using phage display technology, there is still a U.S. Pat. No. 7,132.273 discloses various yeast cell wall need for a robust display system for producing immunoglo anchor proteins and a Surface expression system that uses bulins in eukaryotic host cells and in particular, lower eukary them to immobilize foreign enzymes or polypeptides on the ote host cells. It is particularly desirable to have a robust cell wall. display system for producing immunoglobulins that have Of interest are Tanino et al. Biotechnol. Prog. 22:989-993 45 human-like glycosylation patterns. Genetically engineered (2006), which discloses construction of a Pichia pastoris cell eukaryote cells that produce glycoproteins that have various Surface display system using Flolpanchor System; Ren et al., human-like glycosylation patterns have been described in Molec. Biotechnol. 35:103-108 (2007), which discloses the U.S. Pat. No. 7,029,872 and for example in Choi et al., Hamil display of adenoregulin in a Pichia pastoris cell Surface dis ton, et al., Science 313; 1441 1443 (2006); Wildt and Gern play system using the Flo1p anchor system; Mergler et al., 50 gross, Nature Rev. 3: 119-128 (2005); Bobrowicz et al., Gly Appl. Microbiol. Biotechnol. 63:418-421 (2004), which dis coBiol. 757-766 (2004); Li et al., Nature Biotechnol. 24: closes display of K. lactis yellow enzyme fused to the C-ter 210-215 (2006); Chiba et al., J. Biol. Chem. 273: 26298 minus half of S. cerevisiae C.-agglutinin; Jacobs et al., 26304 (1998); and, Mara et al., Glycoconjugate J. 16:99-107 Abstract T23, Pichia Protein expression Conference, San (1999). Diego, Calif. (Oct. 8-11, 2006), which discloses display of 55 The methods disclosed herein are particularly suited for proteins on the Surface of Pichia pastoris using C-agglutinin; this application because it allows presentation of a vast Ryckaert et al., Abstracts BVBMB Meeting, Vrije Univer diverse repertoire of full-sized immunoglobulins having par siteit Brussel, Belgium (Dec. 2, 2005), which discloses using ticular glycosylation patterns on the Surface of the cell when a yeast display system to identify proteins that bind particular the host cells have been genetically engineered to have altered lectins; U.S. Pat. No. 7,166,423, which discloses a method for 60 or modified glycosylation pathways. In many respects the identifying cells based on the product secreted by the cells by Subject display System mimics the natural immune system. coupling to the cell Surface a capture moiety that binds the Antigen-driven stimulation can be achieved by selecting for secreted product, which can then be identified using a detec high-affinity binders from a display library of cloned anti tion means: U.S. Published Application No. 2004/021961 1, body Hand L chains. The large number of chain permutations which discloses a biotin-avidin system for attaching protein A 65 that occur during recombination of H and L chain genes in or G to the surface of a cell for identifying cells that express developing B cells can be mimicked by shuffling the cloned H particular antibodies; U.S. Pat. No. 6,919,183, which dis and L chains as DNA, and protein and through the use of US 8,067,339 B2 5 6 site-specific recombination (Geoffory et al. Gene 151: 109 expression of the variegated population of mutants of the 113 (1994)). The somatic mutation can also be matched by the immunoglobulin in the host cells; contacting the plurality of introduction of mutations in the CDR regions of the H and L host cells with a detection means that binds to the immuno chains. globulin of interest to identify host cells in the plurality of Immunoglobulins with desired binding specificity or affin 5 host cells that display the immunoglobulin of interest on the ity can be identified using a form of affinity selection known surface thereof. In further embodiments, the method further as “panning (Parmley & Smith, Gene 73:305-318 (1988)). includes isolating the host cells that display the immunoglo The library of immunoglobulins is first incubated with an bulin of interest on the surface of thereof to produce the host antigen of interest followed by the capture of the antigen with cells expressing the immunoglobulin of interest. the bound immunoglobulins. The immunoglobulins recov 10 In a further embodiment, provided is a method for produc ered in this manner can then be amplified and again gain ing eukaryotic host cells that express an immunoglobulin of selected for binding to the antigen, thus enriching for those interest, comprising: providing a host cell that includes a first immunoglobulins that bind the antigen of interest. One or nucleic acid molecule encoding a capture moiety comprising more rounds of selection will enable isolation of antibodies or a cell Surface anchoring protein fused to a binding moiety that fragments thereof with the desired specificity or avidity. 15 is capable of specifically binding an immunoglobulin oper Thus, rare host cells expressing a desired antibody or frag ably linked to a first regulatable promoter, transfecting the ment thereof can easily be selected from greater than 10 host cells with a one or more nucleic acid molecules encoding different individuals in one experiment. The primary struc the heavy and light chains of an immunoglobulin wherein at ture of the binding immunoglobulins is then deduced by least one of the heavy or light chain encoding nucleic acid nucleotide sequence of the individual host cell clone. When molecules is operably linked to a second regulatable promoter human VH and VL regions are employed in the displayed to generate a plurality of host cells encoding a variegated immunoglobulins, the Subject display systems allow selec population of mutants of the immunoglobulins; inducing tion of human immunoglobulins without further manipula expression of the capture moiety for a time Sufficient to pro tion of a non-human immunoglobulins. duce the capture moiety on the surface of the host cells; and Therefore, in one embodiment, provided is a method for 25 inhibiting expression of the capture moiety and inducing producing eukaryotic host cells that express an immunoglo expression of the variegated population of mutants of the bulin of interest, comprising providing host cells that include immunoglobulin in the host cells to produce the host cells. In a first nucleic acid molecule encoding a capture moiety com further embodiments, the method further includes contacting prising a cell Surface anchoring protein fused to a binding the host cells with a detection means that binds to the immu moiety that is capable of specifically binding an immunoglo 30 noglobulin of interest to identify host cells that display the bulin operably linked to a first regulatable promoter; trans immunoglobulin of interest on the Surface thereof, and iso fecting the host cells with a plurality of nucleic acid mol lating the host cells that display the immunoglobulin of inter ecules encoding a genetically diverse population of heavy and est on the surface of thereof to produce the host cells that light chains of an immunoglobulin wherein at least one of the express the immunoglobulin of interest. heavy or light chain encoding nucleic acid molecules is oper 35 In a further embodiment, provided is a method for produc ably linked to a second regulatable promoter to produce a ing eukaryotic host cells that express an immunoglobulin of plurality of genetically diverse host cells capable of display interest, comprising providing host cells that include a first ing an immunoglobulin on the Surface thereof; inducing nucleic acid molecule encoding a capture moiety comprising expression of the first nucleic acid molecule encoding the a cell Surface anchoring protein fused to a binding moiety that capture moiety for a time sufficient to produce the capture 40 is capable of specifically binding an immunoglobulin oper moiety on the Surface of the host cells; and inhibiting expres ably linked to a first regulatable promoter, transfecting the sion of the first nucleic acid molecule encoding the capture host cells with a plurality of nucleic acid molecules compris moiety and inducing expression of the nucleic acid molecules ing open reading frames (ORFs) encoding a genetically encoding the immunoglobulins in the host cells to produce the diverse population of heavy and light chains of an immuno host cells, which display the immunoglobulin of interest on 45 globulin wherein at least the ORFs encoding the heavy chain the surface of the cells. In further aspects, the method further are operably linked to a second regulatable promoter when includes contacting the host cells with a detection means that the capture moiety binds the heavy chain or at least the ORFs specifically binds to the immunoglobulin of interest dis encoding the light chain are operably linked to a second played on the Surface thereof, and isolating host cells in which regulatable promoter when the capture moiety binds the light the detection means is bound to select the host cells that 50 chain to produce a plurality of genetically diverse host cells express the immunoglobulin of interest. capable of displaying an immunoglobulin on the Surface In another embodiment, provided is a method for produc thereof inducing expression of the nucleic acid molecule ing eukaryotic host cells that express an immunoglobulin of encoding the capture moiety for a time sufficient to produce interest comprising providing a host cell that includes a first the capture moiety on the surface of the host cell; and inhib nucleic acid molecule encoding a capture moiety comprising 55 iting expression of the nucleic acid molecule encoding the a cell Surface anchoring protein fused to a binding moiety that capture moiety and inducing expression of the nucleic acid is capable of specifically binding an immunoglobulin oper molecules encoding the immunoglobulins in the host cells to ably linked to a first regulatable promoter, transfecting the produce the host cells. In further embodiments, the method host cell with one or more second nucleic acid molecules further includes contacting the host cells with a detection encoding an immunoglobulin wherein either the molecules 60 means that specifically binds to the immunoglobulin of inter encoding the light chain or the heavy chain are operably est displayed on the cell Surface of the host cells; and isolating linked to a second regulatable promoter, wherein mutagen host cells in which the detection means is bound to produce esis is used to generate a plurality of host cells encoding a the host cells that express the immunoglobulin of interest. variegated population of mutants of the immunoglobulin; In a further embodiment, provided is a method of produc inducing expression of the capture moiety for a time sufficient 65 ing eukaryote host cells that produce an immunoglobulin to produce the capture moiety on the surface of the host cells; having a VH domain and a VL domain and having an antigen inhibiting expression of the capture moiety and inducing binding site with binding specificity for an antigen of interest, US 8,067,339 B2 7 8 the method comprising (a) providing a library of eukaryote acid molecules encoding the light chains are operably linked host cells displaying on their surface an immunoglobulin to a second regulatable promoter and the nucleic acid mol comprising a VH domain and a VL domain, wherein the ecules encoding the heavy chain are operably linked to a third library is created by (i) providing eukaryote host cells that regulatable promoter or to a constitutive promoter. In particu express a capture moiety comprising a cell Surface anchoring lar aspects, the heavy and light chains are encoded by separate protein fused to a moiety capable of binding to an immuno open reading frames (ORFs) wherein each ORF is operably globulin wherein expression of the capture moiety is effected linked to a promoter. In other aspects, the heavy and light by a first regulatable promoter, and (ii) transfecting the host chains are encoded by a single ORF, which produces a single cells with a library of nucleic acid molecules encoding a fusion polypeptide comprising the heavy and light chains in a genetically diverse population of immunoglobulins, wherein 10 tandem orientation, and the ORF is operably linked to a the VH domains of the genetically diverse population of regulatable promoter. The single polypeptide is cleavable immunoglobulins are biased for one or more VH gene fami between the heavy and light chains to produce separate heavy lies and wherein expression of at least one of the heavy or and light chain proteins, which can then associate to form a light chains of the immunoglobulins is effected by a second functional antibody molecule. regulatable promoter to produce a plurality of host cells, each 15 In various aspects of any one of the above embodiments or expressing an immunoglobulin; (b) inducing expression of aspects, the binding moiety that binds the immunoglobulin the capture moiety in the host cells for a time sufficient to binds the Fc region of the immunoglobulin. Examples of such produce the capture moiety on the Surface of the host cells; (c) binding moieties include, but are not limited to those selected inhibiting expression of the capture moiety and inducing from the group consisting of protein A, protein AZZ domain, expression of the library of nucleic acid sequences in the host protein G, and protein Land fragments thereofthat retain the cells, whereby each host cell displays an immunoglobulin at ability to bind to the immunoglobulin. Examples of other the surface thereof to produce the host cells. In further binding moieties, include but are not limited to, Fc receptor embodiments, the method further includes (d) identifying (FcR) proteins and immunoglobulin-binding fragments host cells in the plurality of host cells that display immuno thereof. The FCR proteins include members of the Fc gamma globulins thereon that has a binding specificity for the antigen 25 receptor (FcyR) family, which bind gamma immunoglobulin of interest by contacting the plurality of host cells with the (IgG). Fc epsilon receptor (FceR) family, which bind epsilon antigen of interest and detecting the host cells that have the immunoglobulin (IgE), and Fc alpha receptor (FcCR) family, antigen of interest bound to the immunoglobulin displayed which bind alpha immunoglobulin (IgA). Particular FcR pro thereon to produce the host cells that produce the immuno teins that bind IgG that can comprise the binding moiety globulin having a VH domain and a VL domain and having 30 herein include at least the IgG binding region of FcyRI, the antigen binding site with binding specificity for the anti FcyRIIA, FcyRIIB1, FcyRIIB2, FcyRIIIA, FcyRTIIB, or gen of interest. FcyRn (neonatal). In a further aspect of the above embodiment, the immuno In further aspects of any one of the above embodiments or globulin comprises a synthetic human immunoglobulin VH aspects, detection means is an antigen that is capable of being domain and a synthetic human immunoglobulin VL domain 35 bound by the immunoglobulin of interest. In particular and wherein the synthetic human immunoglobulin VH aspects, the antigen is conjugated to or labeled with a fluo domain and the synthetic human immunoglobulin VL domain rescent moiety. In other aspects, the detection means further comprise framework regions and hyperVariable loops, includes a detection immunoglobulin that is specific for the wherein the framework regions and first two hypervariable immunoglobulin-antigen complex or is specific for another loops of both the VH domain and VL domain are essentially 40 epitope on the antigen and it is this detection immunoglobulin human germ line, and wherein the VH domain and VL that is conjugated to or labeled with a detection moiety Such domain have altered CDR3 loops. In a further aspect of the as a fluorescent moiety. above embodiment, in addition to having altered CDR3 loops In further aspects of any one of the above embodiments or the human synthetic immunoglobulin VH and VL domains aspects, the cell Surface anchoring protein is a Glycosylphos contain mutations in other CDR loops. In a further still aspect 45 phatidylinositol-anchored (GPI) protein. In particular of the above embodiment, each human synthetic immunoglo aspects, the cell Surface anchoring protein is selected from the bulin VH domain CDR loop is of random sequence, and in a group consisting of C-agglutinin, Cwplp, Cwp2p, Gaslip, further still aspect of the above embodiment, the human syn Yap3p, Flo1p, Crh2p, Pir1p, Pir4p, Sedlp, Tiplp, Wpip, thetic immunoglobulin VH domain CDR loops are of known Hpwplp. Als3p, and Rbt5p. In further aspects, the cell surface canonical structures and incorporate random sequence ele 50 anchoring protein is Sedlp. mentS. The host cell that can be used includes both lower and high In a further embodiment, provided is a eukaryote host cell eukaryote cells. Higher eukaryote cells include mammalian, comprising a nucleic acid molecule encoding a capture moi insect, and plant cells. In further aspects of any one of the ety comprising a cell Surface anchoring protein fused to a above embodiments or aspects, the eukaryote is a lower binding moiety capable of binding an immunoglobulin oper 55 eukaryote. In further aspects, the host cell is a yeast or fila ably linked to a regulatable promoter and one or more nucleic mentous fungi cell, which in particular aspects is selected acid molecules encoding the heavy and light chains of immu from the group consisting of Pichia pastoris, Pichia fin noglobulins, wherein at least one of the nucleic acid mol landica, Pichia trehalophila, Pichia koclamae, Pichia mem ecules encoding the heavy or light chains is operably linked to branaefaciens, Pichia minuta (Ogataea minuta, Pichia lind a second regulatable promoter. In particular embodiments, 60 neri), Pichia opuntiae, Pichia thermotolerans, Pichia the nucleic acid molecules encoding both the heavy and light salictaria, Pichia guercuum, Pichia piperi, Pichia stiptis, chains are operably linked to a second regulatable promoter. Pichia methanolica, Pichia sp., Saccharomyces cerevisiae, In other embodiments, the nucleic acid molecules encoding Saccharomyces sp., Hansenula polymorpha, Kluyveromyces the heavy chains are operably linked to a second regulatable sp., Kluyveromyces lactis, Candida albicans, Aspergillus promoter and the nucleic acid molecules encoding the light 65 nidulans, Aspergillus niger, Aspergillus Oryzae, Trichoderma chain are operably linked to a third regulatable promoter or to reesei, Chrysosporium lucknowense, Fusarium sp., a constitutive promoter. In other embodiments, the nucleic Fusarium gramineum, Fusarium venenatum and Neurospora US 8,067,339 B2 10 crassa. In particular aspects, the eukaryote is a yeast and in without inducing the expression of the first regulatable pro further aspects, the yeast is Pichia pastoris. While the meth moter. In further aspects, the inducer of the second regulat ods herein have been exemplified using Pichia pastoris as the able promoter inhibits transcription from the first regulatable host cell, the methods herein can be used in other lower promoter. In particular aspects in which the host cells are eukaryote or higher eukaryote cells for the same purposes 5 yeast, the first regulatable promoter is the GUT1 promoter and disclosed herein. the second regulatable promoter is the GADPH promoter. In In further aspects of any one of the aforementioned meth other aspects, the first regulatable promoter is the PCK1 pro ods, O-glycosylation of glycoproteins in the host cell is con moter and the second regulatable promoter is the GADPH trolled. That is, O-glycan occupancy and mannose chain promoter. length are reduced. In lower eukaryote host cells such as 10 In general, in the above embodiments or aspects, the immu yeast, O-glycosylation can be controlled by deleting the noglobulin will be an IgG molecule and can include IgG1. genes encoding one or more protein O-mannosyltransferases IgG2, IgG3, and IgG4 immunoglobulins and subspecies (Dol-P-Man:Protein (Ser/Thr) Mannosyl Transferase genes) thereof. However, in particular aspects of the above, the (PMTs) or by growing the host in a medium containing one or immunoglobulin is selected from the group consisting of IgA, more Pmtp inhibitors. In further aspects, the host cell includes 15 IgM, IgE, camel heavy chain, and llama heavy chain. a deletion of one or more of the genes encoding PMTs and the The information derived from the host cells and methods host cell is cultivated in a medium that includes one or more herein can be used to produce affinity matured immunoglo Pmtp inhibitors. Pmtp inhibitors include but are not limited to bulins, derivatives of the antibodies, and modified immuno a benzylidene thiazolidinedione. Examples of benzylidene globulins or the nucleic acid encoding the desired immuno thiazolidinediones that can be used are 5-3,4-bis(phenyl globulin can be subcloned into another host cell for methoxy) phenylmethylene-4-oxo-2-thioxo-3-thiazo production or affinity maturation of the immunoglobulin. lidineacetic Acid; 5-3-(1-Phenylethoxy)-4-(2-phe Therefore, further provided is a host cell that expresses an nylethoxy)phenyl)methylene-4-oxo-2-thioxo-3- immunoglobulin that had been identified using any one of the thiazolidineacetic Acid; and 5-3-(1-Phenyl-2-hydroxy) aforementioned methods but does not necessarily have to be ethoxy)-4-(2-phenylethoxy)phenylmethylene-4-oxo-2- 25 the host cell that was used to identify the immunoglobulin. thioxo-3-thiazolidineacetic Acid. In further still aspects, the The host cell can be a prokaryote or eukaryote host cell. host cell further includes a nucleic acid that encodes an alpha Further provided is an immunoglobulin produced by any 1.2-mannosidase that has a signal peptide that directs it for one of the above embodiments or aspects. secretion. The following terms, unless otherwise indicated, shall be In further aspects of any one of the aforementioned meth 30 understood to have the following meanings: ods, host cells further include lower eukaryote cells (e.g., As used herein, the terms “N-glycan and “glycoform” are yeast such as Pichia pastoris) that are genetically engineered used interchangeably and refer to an N-linked oligosaccha to eliminate glycoproteins having C-mannosidase-resistant ride, e.g., one that is attached by an asparagine-N-acetylglu N-glycans by deleting or disrupting one or more of the cosamine linkage to an asparagine residue of a polypeptide. B-mannosyltransferase genes (e.g., BMT1, BMT2, BMT3, 35 N-linked glycoproteins contain an N-acetylglucosamine resi and BMT4)(See, U.S. Published Patent Application No. due linked to the amide nitrogen of an asparagine residue in 2006/0211085) or abrogating translation of RNAs encoding the protein. The predominant Sugars found on glycoproteins one or more of the B-mannosyltransferasesusing interfering are glucose, galactose, mannose, fucose, N-acetylgalac RNA, antisense RNA, or the like. tosamine (GalNAc), N-acetylglucosamine (GlcNAc) and In further aspects of any one of the methods herein, the host 40 sialic acid (e.g., N-acetyl-neuraminic acid (NANA)). The cells can further include lower eukaryote cells (e.g., yeast processing of the Sugar groups occurs co-translationally in Such as Pichia pastoris) that are genetically engineered to the lumen of the ER and continues in the Golgi apparatus for eliminate glycoproteins having phosphomannose residues by N-linked glycoproteins. deleting or disrupting one or both of the phosphomannosyl N-glycans have a common pentasaccharide core of transferase genes PNO1 and MNN4B (See for example, U.S. 45 Man3GlcNAc2 (“Man’ refers to mannose; “Glc” refers to Pat. Nos. 7,198.921 and 7.259,007), which in further aspects glucose; and “NAc refers to N-acetyl; GlcNAc refers to can also include deleting or disrupting the MNN4A gene or N-acetylglucosamine). N-glycans differ with respect to the abrogating translation of RNAS encoding one or more of the number of branches (antennae) comprising peripheral Sugars phosphomannosyltransferases using interfering RNA, anti (e.g., GlcNAc, galactose, fucose and sialic acid) that are sense RNA, or the like. 50 added to the ManaGlcNAc2(“Mang') core structure which is In further still aspects, the host cell has been genetically also referred to as the “trimannose core', the “pentasaccha modified to produce glycoproteins that have predominantly ride core' or the “paucimannose core. N-glycans are classi an N-glycan selected from the group consisting of complex fied according to their branched constituents (e.g., high man N-glycans, hybrid N-glycans, and high mannose N-glycans nose, complex or hybrid). A "high mannose' type N-glycan wherein complex N-glycans are selected from the group con 55 has five or more mannose residues. A "complex” type N-gly sisting of Man-GlcNAc, GlcNAC-Man, GlcNAc, can typically has at least one GlcNAc attached to the 1.3 Galla GlcNAc-Man, GlcNAc2, and NANA Galla mannose arm and at least one GlcNAc attached to the 1,6 Man-GlcNAc, hybrid N-glycans are selected from the group mannose arm of a “trimannose core. Complex N-glycans consisting of Mans GlcNAc, GlcNAcMans GlcNAc, may also have galactose (“Gal') or N-acetylgalactosamine GalGlcNAcMans GlcNAc, and 60 (“GalNAc) residues that are optionally modified with sialic NANAGalGlcNAcMan GlcNAc, and high Mannose N-gly acid orderivatives (e.g., “NANA’ or “NeuAc', where “Neu” cans are selected from the group consisting of MancGlcNAc, refers to neuraminic acid and “Ac' refers to acetyl). Complex Man-GlcNAc, Mans.GlcNAc, and Man-GlcNAc N-glycans may also have intrachain Substitutions comprising In any one of the above embodiments or aspects, the first “bisecting GlcNAc and core fucose (“Fuc'). Complex regulatable promoter is a promoter that is inducible without 65 N-glycans may also have multiple antennae on the “triman inducing expression of the second regulatable promoter. The nose core, often referred to as “multiple antennary glycans.” second regulatable promoter is a promoter that is inducible A “hybrid N-glycan has at least one GlcNAc on the terminal US 8,067,339 B2 11 12 of the 1.3 mannose arm of the trimannose core and Zero or Pichia sp., Saccharomyces cerevisiae, Saccharomyces sp., more mannoses on the 1.6 mannose arm of the trimannose Hansenula polymorpha, Kluyveromyces sp., Kluyveromyces core. The various N-glycans are also referred to as 'glyco lactis, Candida albicans, Aspergillus nidulans, Aspergillus forms.' niger; Aspergillus Oryzae, Trichoderma reesei, Chrysospo Abbreviations used herein are of common usage in the art, rium lucknowense, Fusarium sp., Fusarium gramineum, see, e.g., abbreviations of Sugars, above. Other common Fusarium venematum, Physcomitrella patens and Neuro abbreviations include “PNGase', or “glycanase' or “glucosi spora crassa. Pichia sp., any Saccharomyces sp., Hansenula dase' which all refer to peptide N-glycosidase F (EC polymorpha, any Kluyveromyces sp., Candida albicans, any 3.2.2.18). Aspergillus sp., Trichoderma reesei, Chrysosporium luckno The term “operably linked expression control sequences 10 wense, any Fusarium sp. and Neurospora crassa. refers to a linkage in which the expression control sequence is As used herein, the terms “antibody.” “immunoglobulin.” contiguous with the gene of interest to control the gene of “immunoglobulins' and “immunoglobulin molecule' are interest, as well as expression control sequences that act in used interchangeably. Each immunoglobulin molecule has a trans or at a distance to control the gene of interest. unique structure that allows it to bind its specific antigen, but The term “expression control sequence' or “regulatory 15 all immunoglobulins have the same overall structure as sequences' are used interchangeably and as used herein refer described herein. The basic immunoglobulin structural unit is to polynucleotide sequences which are necessary to affect the known to comprise a tetramer of subunits. Each tetramer has expression of coding sequences to which they are operably two identical pairs of polypeptide chains, each pair having linked. Expression control sequences are sequences which one “light' chain (about 25 kDa) and one “heavy chain control the transcription, post-transcriptional events and (about 50-70kDa). The amino-terminal portion of each chain translation of nucleic acid sequences. Expression control includes a variable region of about 100 to 110 or more amino sequences include appropriate transcription initiation, termi acids primarily responsible for antigen recognition. The car nation, promoter and enhancer sequences; efficient RNA pro boxy-terminal portion of each chain defines a constant region cessing signals such as splicing and polyadenylation signals; primarily responsible for effector function. Light chains are sequences that stabilize cytoplasmic mRNA: sequences that 25 classified as either kappa or lambda. Heavy chains are clas enhance translation efficiency (e.g., ribosome binding sites); sified as gamma, mu, alpha, delta, or epsilon, and define the sequences that enhance protein stability; and when desired, antibody's isotype as IgG, IgM, IgA, Ig), and IgE, respec sequences that enhance protein secretion. The nature of Such tively. control sequences differs depending upon the host organism; The light and heavy chains are subdivided into variable in prokaryotes, such control sequences generally include pro 30 regions and constant regions (See generally, Fundamental moter, ribosomal binding site, and transcription termination Immunology (Paul, W., ed., 2nded. Raven Press, N.Y., 1989), sequence. The term "control sequences” is intended to Ch. 7. The variable regions of each light/heavy chain pair include, at a minimum, all components whose presence is form the antibody binding site. Thus, an intact antibody has essential for expression, and can also include additional com two binding sites. Except in bifunctional or bispecific anti ponents whose presence is advantageous, for example, leader 35 bodies, the two binding sites are the same. The chains all sequences and fusion partner sequences. exhibit the same general structure of relatively conserved The term “recombinant host cell (“expression host cell'. framework regions (FR) joined by three hypervariable “expression host system', 'expression system” or simply regions, also called complementarity determining regions or “host cell), as used herein, is intended to refer to a cell into CDRs. The CDRs from the two chains of each pair are aligned which a recombinant vector has been introduced. It should be 40 by the framework regions, enabling binding to a specific understood that such terms are intended to refer not only to epitope. The terms include naturally occurring forms, as well the particular subject cell but to the progeny of Such a cell. as fragments and derivatives. Included within the scope of the Because certain modifications may occur in Succeeding gen term are classes of immunoglobulins (Igs), namely, IgG, IgA, erations due to either mutation or environmental influences, IgE. IgM, and Ig|D. Also included within the scope of the Such progeny may not, in fact, be identical to the parent cell, 45 terms are the Subtypes of IgGs, namely, IgG1, IgG2, IgG3. but are still included within the scope of the term “host cell and IgG4. The term is used in the broadest sense and includes as used herein. A recombinant host cell may be an isolated cell single monoclonal antibodies (including agonist and antago or cell line grown in culture or may be a cell which resides in nist antibodies) as well as antibody compositions which will a living tissue or organism. bind to multiple epitopes or antigens. The terms specifically The term “transfect”, “transfection”, “transfecting and the 50 cover monoclonal antibodies (including full length mono like refer to the introduction of a heterologous nucleic acid clonal antibodies), polyclonal antibodies, multispecific anti into eukaryote cells, both higher and lower eukaryote cells. bodies (for example, bispecific antibodies), and antibody Historically, the term “transformation” has been used to fragments so long as they contain or are modified to contain at describe the introduction of a nucleic acid into a yeast or least the portion of the CH2 domain of the heavy chain immu fungal cell; however, herein the term “transfection' is used to 55 noglobulin constant region which comprises an N-linked gly refer to the introduction of a nucleic acid into any eukaryote cosylation site of the CH2 domain, or a variant thereof. cell, including yeast and fungal cells. Included within the terms are molecules comprising only the The term “eukaryotic” refers to a nucleated cell or organ Fc region, such as immunoadhesins (U.S. Published Patent ism, and includes insect cells, plant cells, mammalian cells, Application No. 20040136986), Fc fusions, and antibody animal cells and lower eukaryotic cells. 60 like molecules. The term “lower eukaryotic cells' includes yeast and fila The term “Fc fragment refers to the fragment crystal mentous fungi. Yeast and filamentous fungi include, but are lized C-terminal region of the antibody containing the CH2 not limited to Pichia pastoris, Pichia finlandica, Pichia tre and CH3 domains. The term “Fab' fragment refers to the halophila, Pichia koclamae, Pichia membranaefaciens, fragment antigen binding region of the antibody containing Pichia minuta (Ogataea minuta, Pichia lindneri), Pichia 65 the VH, CH1, VL and CL domains. Opuntiae, Pichia thermotolerans, Pichia salictaria, Pichia The term “monoclonal antibody” (mAb) as used herein guercuum, Pichia piperi, Pichia stiptis, Pichia methanolica, refers to an antibody obtained from a population of Substan US 8,067,339 B2 13 14 tially homogeneous antibodies, i.e., the individual antibodies nantly species A, and species B would be the next most comprising the population are identical except for possible predominant species. Some host cells may produce compo naturally occurring mutations that may be present in minor sitions comprising neutral N-glycans and charged N-glycans amounts. Monoclonal antibodies are highly specific, being Such as mannosylphosphate. Therefore, a composition of gly directed against a single antigenic site. Furthermore, in con coproteins can include a plurality of charged and uncharged trast to conventional (polyclonal) antibody preparations or neutral N-glycans. In the present invention, it is within the which typically include different antibodies directed against context of the total plurality of neutral N-glycans in the com different determinants (epitopes), each mAb is directed position in which the predominant N-glycan determined. against a single determinant on the antigen. In addition to Thus, as used herein, "predominant N-glycan’ means that of their specificity, monoclonal antibodies are advantageous in 10 the total plurality of neutral N-glycans in the composition, the that they can be synthesized by hybridoma culture, uncon predominant N-glycan is of a particular structure. taminated by other immunoglobulins. The term “mono As used herein, the term “essentially free of a particular clonal indicates the character of the antibody as being Sugar residue, such as fucose, orgalactose and the like, is used obtained from a Substantially homogeneous population of to indicate that the glycoprotein composition is substantially antibodies, and is not to be construed as requiring production 15 devoid of N-glycans which contain such residues. Expressed of the antibody by any particular method. For example, the in terms of purity, essentially free means that the amount of monoclonal antibodies to be used in accordance with the N-glycan structures containing such Sugar residues does not present invention may be made by the hybridoma method first exceed 10%, and preferably is below 5%, more preferably described by Kohleret al., (1975) Nature, 256:495, or may be below 1%, most preferably below 0.5%, wherein the percent made by recombinant DNA methods (See, for example, U.S. ages are by weight or by mole percent. Thus, Substantially all Pat. No. 4,816,567 to Cabilly et al.). of the N-glycan structures in a glycoprotein composition The term “fragments' within the scope of the terms “anti according to the present invention are free of fucose, or galac body' or “immunoglobulin' include those produced by tose, or both. digestion with various proteases, those produced by chemical As used herein, a glycoprotein composition "lacks' or "is cleavage and/or chemical dissociation and those produced 25 lacking a particular Sugar residue. Such as fucose or galac recombinantly, so long as the fragment remains capable of tose, when no detectable amount of Such Sugar residue is specific binding to a target molecule. Among such fragments present on the N-glycan structures at any time. For example, are Fic, Fab, Fab'. Fv, F(ab')2, and single chain Fv (sclv) in preferred embodiments of the present invention, the gly fragments. Hereinafter, the term “immunoglobulin' also coprotein compositions are produced by lower eukaryotic includes the term “fragments' as well. 30 organisms, as defined above, including yeast (for example, Immunoglobulins further include immunoglobulins or Pichia sp., Saccharomyces sp.: Kluyveromyces sp.: Aspergil fragments that have been modified in sequence but remain lus sp.), and will “lack fucose.” because the cells of these capable of specific binding to a target molecule, including: organisms do not have the enzymes needed to produce fuco interspecies chimeric and humanized antibodies; antibody sylated N-glycan structures. Thus, the term "essentially free fusions; heteromeric antibody complexes and antibody 35 of fucose' encompasses the term “lacking fucose.” However, fusions, such as diabodies (bispecific antibodies), single a composition may be “essentially free of fucose' even if the chain diabodies, and intrabodies (See, for example, Intracel composition at one time contained fucosylated N-glycan lular Antibodies: Research and Disease Applications, structures or contains limited, but detectable amounts of fuco (Marasco, ed., Springer-Verlag New York, Inc., 1998). Sylated N-glycan structures as described above. The term “catalytic antibody' refers to immunoglobulin 40 The interaction of antibodies and antibody-antigen com molecules that are capable of catalyzing a biochemical reac plexes with cells of the immune system and the variety of tion. Catalytic antibodies are well known in the art and have responses, including antibody-dependent cell-mediated cyto been described in U.S. patent application Ser. Nos. 7.205, toxicity (ADCC) and complement-dependent cytotoxicity 136; 4,888,281; 5,037,750 to Schochetman et al., U.S. patent (CDC), clearance of immunocomplexes (phagocytosis), anti application Ser. Nos. 5,733,757; 5,985,626; and 6,368,839 to 45 body production by B cells and IgG serum half-life are Barbas, III et al. defined respectively in the following: Daeron et al., 1997, As used herein, the term “consisting essentially of will be Annu. Rev. Immunol. 15:203-234; Ward and Ghetie, 1995, understood to imply the inclusion of a stated integer or group Therapeutic Immunol. 2:77-94: Cox and Greenberg, 2001, of integers; while excluding modifications or other integers Semin. Immunol. 13: 339-345; Heyman, 2003, Immunol. which would materially affect or alter the stated integer. With 50 Lett. 88:157-161; and Ravetch, 1997, Curr. Opin. Immunol. respect to species of N-glycans, the term "consisting essen 9: 121-125. tially of a stated N-glycan will be understood to include the N-glycan whether or not that N-glycan is fucosylated at the BRIEF DESCRIPTION OF THE DRAWINGS N-acetylglucosamine (GlcNAc) which is directly linked to the asparagine residue of the glycoprotein. 55 FIG. 1 illustrates the general operation of the method using As used herein, the term “predominantly' or variations an embodiment wherein the immunoglobulin (Ig) light and such as “the predominant’ or “which is predominant will be heavy chains are separately expressed and detection of cells understood to mean the glycan species that has the highest that express the immunoglobulin of interest is via a labeled mole percent (%) of total neutral N-glycans after the glyco antigen. protein has been treated with PNGase and released glycans 60 FIG. 2 illustrates the construction of plasmid vector analyzed by mass spectroscopy, for example, MALDI-TOF pGLY642. MS or HPLC. In other words, the phrase “predominantly' is FIG. 3 illustrates the construction of plasmid vector defined as an individual entity, Such as a specific glycoform, pGLY2233 is present in greater mole percent than any other individual FIG. 4 illustrates the construction of plasmid vector entity. For example, if a composition consists of species A in 65 pGFI207t. 40 mole percent, species B in 35 mole percent and species C FIG. 5 illustrates the construction of plasmid vector pGLY in 25 mole percent, the composition comprises predomi 162. US 8,067,339 B2 15 16 FIG. 6 illustrates the genealogy of some of the yeast strains FIG. 19 shows the results of FACS sorting of the cells used to demonstrate operation of the present invention. shown in FIG. 18. The redline represents the negative control FIG. 7 shows a map of plasmid vector pGLY2988. without co-expression of antibody. The blue line represents FIG. 8 shows a map of plasmid vector pGLY3200. colonies of anti-Her2 or anti-CD20 expressing strains. FIG. 9 shows maps of plasmid vectors pGLY4136 and FIG. 20 shows a map of plasmid vector pGLY3033. pGLY4124. FIG. 10 shows maps of plasmid vectors pGLY4116 and DETAILED DESCRIPTION OF THE INVENTION pGLY4137. FIG. 11 shows fluorescence microscopy results of strain The present invention provides a protein display system yGLY4134 (expresses anti-Her2 antibody), strainyGLY2696 10 that is capable of displaying diverse libraries of immunoglo (empty strain) transfected with pGLY4136 encoding Protein bulins on the surface of a eukaryote host cell. The composi A/SED1 fusion protein, and strain yCLY4134 (expresses tions and methods are particularly useful for the display of anti-Her2 antibody) transfected with pGLY4136 encoding collections of immunoglobulins in the context of discovery Protein A/SED1 fusion protein incubated with goat anti-hu (that is, Screening) or molecular evolution protocols. A salient man IgG (H+L)-Alexa 488. 15 feature of the method is that it provides a display system in FIG. 12 shows fluorescence microscopy results of strain which a whole, intact immunoglobulin molecule of interest yGLY2696 (empty strain) transfected with pGLY4136 can be displayed on the surface of a host cell without having encoding the Protein A/SED1 fusion protein incubated with to express the immunoglobulin molecule of interest either as anti-Her2 antibody. Goat anti-human IgG (H+L)-Alexa 488 fusion protein in which it is fused to a surface anchor protein was used for detection of anti-antibody bound to the Protein or other moiety that enables capture of the immunoglobulin A/SED1 fusion protein anchored to the cell surface. by a capture moiety bound to the cell surface. Another feature FIG. 13 shows fluorescence microscopy results of strain of the method is that it enables screening diverse libraries of yGLY2696 (empty strain) transfected with pGLY4136 immunoglobulins in host cells for a host cell in the library that encoding Protein A/SED 1 fusion protein, strain yOLY3920 produces an immunoglobulin of interest and then enables the (expresses anti-CD20 antibody) transfected with pGLY4136 25 host cell to be separated from the other host cells in the library encoding Protein A/SED1 fusion protein, and strain that do not express the immunoglobulin of interest. Impor yGLY4134 (expresses anti-Her2 antibody) transfected with tantly, the isolated host cell can then be used for production of pGLY4136 encoding Protein A/SED1 fusion protein incu the immunoglobulin of interest for use in therapeutic or diag bated with anti-Her2 antibody. Goat anti-human IgG (H+L)- nostic applications. This is an improvement over phage and Alexa 488 was used for detection of anti-antibody bound to 30 yeast display methods wherein a diverse library of scFV or the Protein A/SED1 fusion protein anchored to the cell Sur Fab fragments are screened for a host cell that expresses an face. scFV or Fab of interest, which is then used in a series of steps FIG. 14 shows fluorescence microscopy results of strain to construct a mammalian host cell that expresses a whole yGLY2696 (empty strain) transfected with pGLY4116 immunoglobulin with the characteristics of the scFV or Fab encoding the FcRIII/SED1 fusion protein incubated with 35 of interest. These subsequent steps present the risk that the anti-Her2 antibody. Goat anti-human IgG (H+L)-Alexa 488 desired affinity or specificity of an scFV or Fab that has been was used for detection of anti-antibody bound to the Protein identified during the maturation process of converting the A/SED1 fusion protein anchored to the cell surface. sch V or Fab into a whole immunoglobulin could be abrogated FIG. 15 shows maps of plasmid vectors pGLY439 and or diminished. pGLY4144. 40 While current phage-based methods provide substantial FIG. 16 shows fluorescence microscopy results of strain library diversity and have greatly improved the processes for yGLY4134 (AOX promoter-anti-Her2 antibody) transfected developing immunoglobulins, a disadvantage is that the with pGLY4136 (AOX promoter-Protein A/SED1 fusion pro prokaryotic host cells used to construct the libraries do not tein), strain yGLY4134 (AOX promoter-anti-Her2 antibody) produce N-linked glycosylated glycoproteins. Posttransla transfected with pGLY4139 (GAPDH promoter-Protein 45 tional modifications such as glycosylation can affect speci A/SED1 fusion protein), and strainyGLY5434(GAPDH pro ficity or affinity of the immunoglobulin. It is estimated that moter-anti-Her2 antibody) transfected with pGLY4139 about 15-20% of circulating monoclonal antibodies derived (GUT1 promoter-Protein A/SED1 fusion protein). Goat anti entirely in mammalian cells contain one or more N-linked human IgG (H+L)-Alexa 488 was used for detection of anti glycans in the variable regions. (Jefferis, Biotechnol Progress antibody bound to the Protein A/SED1 fusion protein 50 21: 11-16 (2005)) In some cases it is believed that these anchored to the cell surface. N-glycans in the variable region may play a significant role in FIG. 17 illustrates the hypothetical expression of Protein immunoglobulin function. For example, both positive and A/SED1 fusion protein and antibody under the control of negative influences on antigen binding have been seen in different combinations of promoters. antibody molecules with variable region N-glycosylation. FIG. 18 shows fluorescence microscopy results of strains 55 N-glycosylation consensus sites added within the CDR2 yGLY5757 (expresses anti-CD20 antibody under control of region of an anti-dextran antibody were filled with carbohy the GAPDH promoter) and yCLY5434 (expresses anti-Her2 drates of varying structure and showed changes in affinity, antibody under control of the GAPDH promoter), each trans half-life and tissue targeting in a site dependent manner (Co fected with pGLY4144 encoding Protein A/SED1 fusion pro loma et al., The Journal of Immunology 162: 2162-2170 teinunder the control of the GUT1 promoter. Protein A/SED1 60 (1999)). Therefore, libraries produced and screened in fusion protein expression (GUT1 promoter) was induced first prokaryotic host cells will tend to be biased against immuno under glycerol conditions; then antibody expression from the globulin species that might have glycosylation in the variable GAPDH promoter was induced under dextrose conditions, region. Thus, immunoglobulins that might have particularly which also inhibits expression of the Protein A/SED1 fusion desirable specificity or affinity due in whole or in part to protein. Goat anti-human IgG (H+L)-Alexa 488 was used for 65 glycosylation of one or more sites in the variable regions will detection of anti-antibody bound to the Protein A/SED1 not be identified. Conversely, antibodies identified through fusion protein anchored to the cell surface. prokaryotic screening methods may, when expressed in a US 8,067,339 B2 17 18 eukaryotic host, have glycosylation structures that unfavor encoding a heavy chain is operably linked to a second regu ably impact folding or affinity. The methods and systems latable promoter when the capture moiety binds the heavy herein for the first time enable libraries of immunoglobulins chain or at least the nucleic acid encoding a light chain is to be screened wherein the libraries include populations of operably linked to a second regulatable promoter when the immunoglobulins that are glycosylated in the variable region. 5 capture moiety binds the light chain. This produces a plurality This has the potential effect of increasing the diversity of the of host cells wherein each host cell in the plurality of host library over what would be expected if the diversity of the cells capable of displaying an immunoglobulin on the Surface library was based solely on sequence. This improvement is thereof and each host cell in the plurality of host cells is expected to increase the ability to develop immunoglobulins capable of displaying a particular distinct immunoglobulin that have greater specificity or affinity than current methods 10 species. In general, the diversity of the host cell population in permit. the plurality of host cells will depend on the diversity of the The methods and systems herein also provide another library of nucleic acid molecules that was transfected into the advantage over current methods in that eukaryote host cells host cells. that have been genetically engineered to produce glycopro In another aspect, the host cell is propagated in a culture to teins that have predominantly particular N-glycan structures 15 provide a multiplicity of host cells, which are then transfected can be used. The N-glycan structures include any of the with one or more nucleic acid second molecules encoding the N-glycan structures currently found on human immunoglo heavy and/or light chains of an immunoglobulin wherein at bulins or N-glycan structures that lack features not found in least the nucleic acid encoding a heavy chain is operably glycoproteins from higher eukaryotes. For example, in the linked to a second regulatable promoter when the capture case of yeast, the host cells can be genetically engineered to moiety binds the heavy chain or at least the nucleic acid produce immunoglobulins wherein the N-glycans are not encoding a light chain is operably linked to a second regulat hypermannosylated. The host cells can be genetically engi able promoter when the capture moiety binds the light chain neered to limit the amount of O-glycosylation or to modify to provide a multiplicity of host cells that are capable of O-glycosylation to resemble O-glycosylation in mammalian displaying the encoded immunoglobulin on the Surface cells. 25 thereof. Mutagenesis of the multiplicity of host cells is used to A significant advantage of the methods and systems is that generate a plurality of host cells that encode a variegated the host cell identified in the library to produce a desired population of mutants of the immunoglobulin. The diversity immunoglobulin can be used without further development or is dependent on the mutagenesis method used. Suitable meth manipulation of the host cell or the nucleic acid molecule ods for mutagenesis include but are not limited to cassette encoding the immunoglobulin for production of the immu 30 mutagenesis, error-prone PCR, chemical mutagenesis, or noglobulin. That is, cultivating the host cells identified herein shuffling to generate a refined repertoire of altered sequences as expressing the desired immunoglobulin under conditions that resemble the parent nucleic acid molecule. that induce expression of the desired immunoglobulin with In further aspects, the host cell is propagated in a culture to out inducing expression of the capture moiety either before, provide a multiplicity of host cells, which are then transfected after, or at the same time: the cells secrete the desired immu 35 with a plurality of second nucleic acid molecules, each noglobulin, which can then be recovered from the culture nucleic acid molecule encoding the heavy and/or light chains medium using methods well known in the art. An important of an immunoglobulin wherein at least the nucleic acid element is that the immunoglobulin that is produced is a encoding a heavy chain is operably linked to a second regu whole, intact immunoglobulin molecule. This ability to use latable promoter when the capture moiety binds the heavy library cells to produce whole, intact immunoglobulins is not 40 chain or at least the nucleic acid encoding a light chain is possible with the current phage-based or yeast-based sys operably linked to a second regulatable promoter when the tems. In those systems, the nucleic acid molecules encoding capture moiety binds the light chain to produce a plurality of the desired Fab or sch V has to be further manipulated to host cells that are capable of displaying an immunoglobulin construct a nucleic acid molecule that encodes a whole, intact on the Surface thereof. Mutagenesis is then used to generate immunoglobulin, which is then transfected into a mammalian 45 further increase the diversity of the plurality of host cells that cell for production of the whole, intact immunoglobulin. are capable of displaying an immunoglobulin on the Surface Thus, the methods and systems herein provide significant thereof. improvements to the development and production of immu In particular embodiments, the nucleic acid molecules noglobulins for therapeutic or diagnostic purposes. encoding both the heavy and light chains are operably linked What is provided then is a method for constructing and 50 to a second regulatable promoter. In other embodiments, the isolating a eukaryotic host cell expressing an immunoglobu nucleic acid molecules encoding at least one of the heavy lin of interest from a library of host cells expressing a plurality chains are operably linked to a second regulatable promoter of immunoglobulins. The method enables the construction and the nucleic acid molecules encoding the light chain are and selection of immunoglobulins with desirable specificity operably linked to a third regulatable promoter or to a con and/or affinity properties. In general, the method comprises 55 stitutive promoter. In particular aspects, a plurality of nucleic providing a host cell that comprises a first nucleic acid mol acids encoding Sub-populations of heavy chains are provided ecule encoding a capture moiety comprising a cell Surface wherein expression of each sub-population is effected by a anchoring protein fused to a binding moiety that is capable of second, third, or more regulatable promoter such that differ specifically binding an immunoglobulin operably linked to a ent Sub-populations can be expressed at a particular time first regulatable promoter. The host cell can be further geneti 60 while other Sub-populations are not expressed at that time. cally engineered to produce immunoglobulins having par In general, the heavy and light chains are encoded by ticular predominant N-glycan structures. separate open reading frames (ORFs) wherein each ORF is In one aspect, the host cell is propagated in a culture to operably linked to a promoter. However, in other aspects, the provide a multiplicity of host cells, which are then transfected heavy and light chains are encoded by a single ORF, which with a plurality of second nucleic acid molecules, each 65 produces a single fusion polypeptide comprising the heavy nucleic acid molecule encoding the heavy and/or light chains and light chains in a tandem orientation, and the ORF is of an immunoglobulin wherein at least the nucleic acid operably linked to a regulatable promoter. The single US 8,067,339 B2 19 20 polypeptide is cleavable between the heavy and light chains to an epitope other than the epitope bound by the detection produce separate heavy and light chain proteins, which can immunoglobulin. In another aspect, the detection immuno then associate to form a functional antibody molecule. (See globulin is specific for the immunoglobulin-antigen complex. for example, U.S. Published Application No. 2006/0252096). Regardless of the detection means, a high occurrence of the In any one of the above aspects, the expression of the first 5 label indicates the immunoglobulin of interest has desirable nucleic acid molecule encoding the capture moiety is induced binding properties and a low occurrence of the label indicates for a time sufficient to produce the capture moiety and allow the immunoglobulin of interest does not have desirable bind it to be transported to and thenbound to the surface of the host ing properties. cell Such that the capture moiety is capable of binding immu Detection moieties that are suitable for labeling are well noglobulin molecules as they are secreted from the host cell. 10 known in the art. Examples of detection moieties, include but Expression of the capture moiety is then reduced or inhibited are not limited to, fluorescein (FITC), Alexa Fluors such as and expression of the nucleic acid molecules encoding the Alexa Fuor 488 (Invitrogen), green fluorescence protein heavy and/or light chains of the immunoglobulins operably (GFP), Carboxyfluorescein succinimidyl ester (CFSE), linked to the second regulatable promoter is induced. While DyLight Fluors (Thermo Fisher Scientific), HyLite Fluors expression of both the heavy and light chains can be induced, 15 (AnaSpec), and phycoerythrin. Other detection moieties in particular aspects, the expression of the heavy chain is include but are not limited to, magnetic beads which are induced and expression of the light chain is constitutive. In coated with the antigen of interest or immunoglobulins that other aspects, when the capture moiety binds the light chain, are specific for the immunoglobulin of interest or immuno expression of the light chain can be regulated and expression globulin-antigen complex. In particular aspects, the magnetic of the heavy chain can be constitutive. Thus, whether it is the beads are coated with anti-fluorochrome immunoglobulins heavy chain or the light chain that is captured determines specific for the fluorescent label on the labeled antigen or whether it is the light chain or the heavy chain whose expres immunoglobulin. Thus, the host cells are incubated with the sion is regulated. labeled-antigen or immunoglobulin and then incubated with Inhibition of expression of the capture moiety can be the magnetic beads specific for the fluorescent label. effected by no longer providing the inducer than induces 25 Analysis of the cell population and cell sorting of those expression of the capture moiety, or by providing an inhibitor host cells that display the immunoglobulin of interest based of the first regulatable promoter that inhibits expression of the upon the presence of the detection moiety can be accom capture moiety, or by using an inducer of expression of the plished by a number of techniques known in the art. Cells that immunoglobulins heavy and/or light chains operably linked display the immunoglobulin of interest can be analyzed or to a second or more inducible promoter that also inhibits 30 Sorted by, for example, flow cytometry, magnetic beads, or expression of the capture moiety. Inhibition can be complete fluorescence-activated cell sorting (FACS). These techniques repression of expression or a reduction in expression to an allow the analysis and sorting according to one or more amount wherein expression of the capture moiety is such that parameters of the cells. Usually one or multiple secretion it does not interfere with the processing and transport of the parameters can be analyzed simultaneously in combination heavy and light chains through the secretory pathway. The 35 with other measurable parameters of the cell, including, but expressed immunoglobulin heavy and/or light chains are pro not limited to, cell type, cell Surface antigens, DNA content, cessed and transported to the cell surface via the host cell etc. The data can be analyzed and cells that display the immu secretory pathway where they are captured by the capture noglobulin of interest can be sorted using any formula or moiety bound to the host cell surface for display. The plurality combination of the measured parameters. Cell sorting and of host cells with the expressed immunoglobulins displayed 40 cell analysis methods are known in the art and are described thereon are then screened using a detection means that will in, for example, The Handbook of Experimental Immunol bind to the immunoglobulin of interest but not to other immu ogy, Volumes 1 to 4. (D. N. Weir, editor) and Flow Cytometry noglobulins to identify the host cells that display the immu and Cell Sorting (A. Radbruch, editor, Springer Verlag, noglobulin of interest on the surface thereof from those host 1992). Cells can also be analyzed using microscopy tech cells that do not display the immunoglobulin of interest. Host 45 niques including, for example, laser Scanning microscopy, cells that express and display the immunoglobulin of interest fluorescence microscopy; techniques such as these may also are separated from the host cells that do not express and be used in combination with image analysis systems. Other display the immunoglobulin of interest to produce a popula methods for cell sorting include, for example, panning and tion of host cells comprising exclusively or enriched for the separation using affinity techniques, including those tech host cells displaying the immunoglobulin of interest. These 50 niques using Solid Supports such as plates, beads, and col separated host cells can be propagated and used to produce S. the immunoglobulin of interest in the quantities needed for In further aspects, provided is a library method for identi the use intended. The nucleic acid encoding the immunoglo fying and selecting cells that produce an immunoglobulin bulin can be determined and an expression vector encoding having a desired specificity and/or affinity for a particular the heavy and light chains of the immunoglobulin can be 55 antigen. The method comprises providing a library of eukary constructed and used to transfect another host cell, which can ote host cells displaying on their surface an immunoglobulin be a prokaryotic or eukaryotic host cell. comprising a VH domain and a VL domain, wherein the Detection and analysis of host cells that express the immu library is created by (i) providing eukaryote host cells that noglobulin of interest can be achieved by labeling the host express a capture moiety comprising a cell Surface anchoring cells with an antigen that is specifically recognized by the 60 protein fused to a moiety capable of binding an immunoglo immunoglobulin of interest. In particular aspects, the antigen bulin wherein expression of the capture moiety is effected by is labeled with a detection moiety. In other aspects the antigen a first regulatable promoter; and (ii) transfecting the host cells is unlabeled and detection is achieved by using a detection with a library of nucleic acid sequences encoding a geneti immunoglobulin that is labeled with a detection moiety and cally diverse population of immunoglobulins, wherein the binds an epitope of the antigen that is not bound by the 65 VH domains of the genetically diverse population of immu immunoglobulin of interest. This enables selection of host noglobulins are biased for one or more VH gene families and cells that produce immunoglobulins that bind the antigen at wherein expression of at least one or more heavy or light US 8,067,339 B2 21 22 chains is effected by a second regulatable promoter to pro molecules. In either case, these nucleic acid molecules may duce a plurality of host cells, each host cell in the plurality of further include when desirable codon optimizations to host cells expresses an immunoglobulin species. Expression enhance translation of the mRNA encoding the immunoglo of the capture moiety is induced in the plurality of host cells bulins in the host cell chosen. The nucleic acid molecule may for a time sufficient to produce the capture moiety on the further include when desirable replacement of endogenous surface of the host cells. Then expression of the of the capture signal peptides with signal peptides that are appropriate for moiety while expression of the library of nucleic acid the host cell chosen. sequences is induced in the plurality of host cells to produce In one aspect, the above nucleic acid molecule can com a plurality of host cells wherein each host cell displays an prise a single expression cassette operably linked to a second immunoglobulin species at the Surface thereof. Host cells in 10 regulatable promoter wherein the open reading frames the plurality of host cells that display immunoglobulins (ORFs) for the light and heavy chains are in frame and sepa thereon that has a binding specificity for the antigen of inter rated by a nucleic acid molecule encoding inframe a protease est are identified by contacting the plurality of host cells with cleavage site that upon expression produces a fusion protein the antigen of interestand detecting the host cells that have the that is processed post-translationally with a protease specific antigen of interest bound to the immunoglobulin displayed 15 for the protease cleavage site to produce the light and heavy thereon to produce the host cells that produce the immuno chains of the immunoglobulin. Examples of these expression globulin having a VH domain and a VL domain and having cassettes can be found in for example, U.S. Publication No. the antigen binding site with binding specificity for the anti 20060252096. In another aspect, the heavy and light immu gen of interest. In particular aspects, the immunoglobulin noglobulin chains are expressed from separate expression comprises a synthetic human immunoglobulin VH domain cassettes wherein the ORF encoding each of the light and and a synthetic human immunoglobulin VL domain and fur heavy chains is operably linked to a second regulatable pro ther, the synthetic human immunoglobulin VH domain and moter. Examples of these expression cassettes can be found in the synthetic human immunoglobulin VL domain comprise for example, U.S. Pat. Nos. 4,816,567 and 4,816,397. In a framework regions and hyperVariable loops, wherein the further aspect, the heavy and light immunoglobulin chains are framework regions and first two hypervariable loops of both 25 expressed from separate expression cassettes wherein the the VH domain and VL domain are essentially human germ ORF encoding the heavy chain is operably linked to a second line, and wherein the VH domain and VL domain have altered regulatable promoter and the ORF encoding the light chain is CDR3 loops. operably linked to a constitutive promoter. This provides a library of host cells that are capable of In particular aspects, the encoded immunoglobulin com expressing a plurality of immunoglobulin molecules, which 30 prises a synthetic human immunoglobulin VH domain and a can be captured and displayed on the cell Surface for detection synthetic human immunoglobulin VL domain and wherein by a detection means that can bind an immunoglobulin spe the synthetic human immunoglobulin VH domain and the cific for a particular antigen and thereby enable the host cell synthetic human immunoglobulin VL domain comprise expressing the immunoglobulin to be identified from the plu framework regions and hyperVariable loops, wherein the rality of host cells in the library. In general, the detection 35 framework regions and first two hypervariable loops of both means will usually use the antigen that has been labeled with the VH domain and VL domain are essentially human germ a detection moiety. These host cells can be isolated from the line, and wherein the VH domain and VL domain have altered plurality of host cells by any means currently used for selec CDR3 loops. In further still aspects, in addition to having tion of particular cells in a population of cells, e.g., FACS altered CDR3 loops, the human synthetic immunoglobulin Sorting. 40 VH and VL domains contain mutations in other CDR loops. Thus, the method comprises at least two components. The In further aspects, each human synthetic immunoglobulin VH first component is a helper vector that contains an expression domain CDR loop is of random sequence. In further still cassette comprising the first nucleic acid molecule that aspects, the human synthetic immunoglobulin VH domain encodes and expresses a capture moiety that in particular CDR loops are of known canonical structures and incorporate embodiments comprises a cell Surface anchoring protein or 45 random sequence elements. cell wall binding protein that is capable of binding or inte Both of the components can be provided in vectors which grating to the surface of the host cell fused at its N- or C-ter integrate the nucleic acid molecules into the genome of the minus to a binding moiety capable of binding an immunoglo host cell by homologous recombination. Homologous recom bulin. The binding moiety is located at the end of the cell bination can be double crossover or single crossover homolo Surface anchoring protein that is exposed to the extracellular 50 gous recombination. Roll-in single crossover homologous environment Such that the binding moiety is capable of inter recombination has been described in Nett et al., Yeast 22: acting with an immunoglobulin. The immunoglobulin bind 295-304 (2005). Each component can be integrated in the ing moiety includes the immunoglobulin binding domains same locus in the genome or in separate loci in the genome. from Such molecules as protein A, protein G, protein L, or the Alternatively, one or both components can be transiently like or an Fc receptor. 55 expressed in the host cell. The second component is one or more vectors that contain FIG. 1 illustrates the general operation of the method using expression cassettes that encode and express the heavy and an embodiment wherein the immunoglobulin light and heavy light chains of an immunoglobulin of interest or libraries of chains are separately expressed and detection is via a labeled which the immunoglobulin of interest is to be selected (for antigen. FIG. 1 shows an expression cassette encoding the example, a library of vectors expressing immunoglobulins). 60 capture moiety fusion protein operably linked to promoter A In particular aspects, the nucleic acid molecule encoding the and expression cassettes encoding the immunoglobulin (Ig) immunoglobulin may include the nucleotide sequences light and heavy chains, each operably linked to promoter B. encoding both the heavy and the light chains of the immuno As shown, the host cell is transfected with the expression globulins, e.g., an immunoglobulin having a VH domain and cassettes and the transformed cells grown under conditions aVL domain and having an antigenbinding site with binding 65 that induce expression of the capture moiety fusion protein specificity for an antigen of interest. In other aspects, the via promoter A. The capture moiety fusion protein is heavy and light chains are encoded on separate nucleic acid anchored to the cell surface. Then the cells are grown under US 8,067,339 B2 23 24 conditions that inhibit or reduce expression of the capture dominantly Mans GlcNAc or Gal-GlcNAc-Man-GlcNAc moiety fusion protein but induce expression of the immuno N-glycosylation is selected and a library of vectors encoding globulin light and heavy chains via promoter B. The immu GPI proteins fused to one or more immunoglobulin capture noglobulins are secreted from the cells and captured by the moieties is then provided (GPI-IgG capture moiety). A library capture moiety fusion protein anchored to the cell Surface. of host cells is then constructed wherein each host cell to The cells with the captured immunoglobulins are then make up the library is transfected with one of the vectors in screened for the Ig of interest using a antigen labeled with a the library of vectors encoding GPI-IgG capture moiety detection moiety. As shown, not all cells will produce the fusion proteins such that each host cell species in the library immunoglobulin of interest. Cells that bind the labeled anti will express one particular GPI-IgG capture moiety fusion gen are selected and separated from cells that do not produce 10 protein. Each host cell species of the library is then trans the immunoglobulin of interest. This produces cells that fected with a vector encoding the desired particular immuno express the immunoglobulin of interest. These cells can be globulin. The host cell that results in the best presentation of used for producing the immunoglobulin for use in therapeutic the particular immunoglobulin on the surface of the host cell or diagnostic applications. Alternatively, the cells can is selected as the host cell for producing the particular immu undergo mutagenesis that introduces mutations into the 15 noglobulin. expression cassettes encoding the immunoglobulins and the In general, the GPI protein used in the methods disclosed cells screened for cells that produce immunoglobulins with herein is a chimeric protein or fusion protein comprising the properties that have been modified or altered from those prop GPI protein fused at its N-terminus to the C-terminus of an erties in the immunoglobulin prior to mutagenesis and which immunoglobulin capture moiety. The N-terminus of the cap are desired. Cells that express immunoglobulins having ture moiety is fused to the C-terminus of a signal sequence or modified or altered but desired properties can be separated peptide that enables the GPI-IgG capture moiety fusion pro from the other cells and used for producing the immunoglo tein to be transported through the secretory pathway to the bulin for therapeutic or diagnostic applications. cell surface where the GPI-IgG capture moiety fusion protein Glycosylphosphatidylinositol-anchored (GPI) proteins is secreted and then bound to the cell Surface. In some aspects, provide a Suitable means for tethering the capture moiety to 25 the GPI-IgG capture moiety fusion protein comprises the the surface of the host cell. GPI proteins have been identified entire GPI protein and in other aspects, the GPI-IgG capture and characterized in a wide range of species from humans to moiety fusion protein comprises the portion of the GPI pro yeast and fungi. Thus, in particular aspects of the methods tein that is capable of binding to the cell surface. disclosed herein, the cell surface anchoring protein is a GPI The immunoglobulin capture moiety can comprise any protein or fragment thereofthat can anchor to the cell Surface. 30 molecule that can bind to an immunoglobulins. A multitude Lower eukaryotic cells have systems of GPI proteins that are of Gram-positive bacteria species have been isolated that involved in anchoring or tethering expressed proteins to the express surface proteins with affinities for mammalian immu cell wall so that they are effectively displayed on the cell wall noglobulins through interaction with their heavy chains. The of the cell from which they were expressed. For example, 66 best known of these immunoglobulin binding proteins are putative GPI proteins have been identified in Saccharomyces 35 type 1 Staphylococcus Protein A and type 2 Streptococcus cerevisiae (See, de Groot et al., Yeast 20:781-796 (2003)). Protein G which have been shown to interact principally GPI proteins which may be used in the methods herein through the C2-C3 interface on the Fc region of human immu include, but are not limited to, Saccharomyces cerevisiae noglobulins. In addition, both have also been shown to inter CWP1, CWP2, SED1, and GAS1; Pichia pastoris SP1 and act weakly to the Fab region, but again through the immuno GAS1; and H. polymorpha TIP1. Additional GPI proteins 40 globulin heavy chain. may also be useful. Additional suitable GPI proteins can be Recently, a novel protein from Peptococcusmagnums, Pro identified using the methods and materials of the invention tein L, has been reported that was found to bind to human, described and exemplified herein. rabbit, porcine, mouse, and rat immunoglobulins uniquely The selection of the appropriate GPI protein will depend on through interaction with their light chains. In humans this the particular recombinant protein to be produced in the host 45 interaction has been shown to occur exclusively to the kappa cell and the particular post-translation modifications to be chains. Since both kappa and lambda light chains are shared performed on the recombinant protein. For example, produc between different classes, Protein L binds strongly to all tion of immunoglobulins with particular glycosylation pat human classes, in particular to the multi-subunit IgM, and terns will entail the use of recombinant host cells that produce similarly is expected to bind to all classes in species that show glycoproteins having particular glycosylation patterns. The 50 Protein L light chain binding. GPI protein most suitable in a system for producing antibod Examples of other binding moieties, include but are not ies or fragments thereof that have predominantly limited to, Fc receptor (FcR) proteins and immunoglobulin Mans.GlcNAc N-glycosylation many not necessarily be the binding fragments thereof. The FCR proteins include mem GPI protein most suitable in a system for producing antibod bers of the Fc gamma receptor (FcyR) family, which bind ies O thereof having predominantly 55 gamma immunoglobulin (IgG). Fc epsilon receptor (FceR) GalGlcNAc-Man-GlcNAc,N-glycosylation. In addition, family, which bind epsilon immunoglobulin (IgE), and Fc the GPI most Suitable in a system for producing immunoglo alpha receptor (FcCR) family, which bind alpha immunoglo bulins specific for one epitope or antigen may not necessarily bulin (IgA). Particular FcR proteins that bind IgG and can be be the most suitable GPI protein in a system for producing used to comprise the capture moiety disclosed herein include immunoglobulins specific for another epitope or antigen. 60 at least the immunoglobulin binding portion of any one of Therefore, further provided is a library method for con FcyRI, FcyRIIA, FcyRIIB1, FcyRIIB2, FcyRIIIA, FcyRIIIB, structing the host cell that is to be used for producing a or FcyRn (neonatal). particular immunoglobulin. In general, the host cell that is Regulatory sequences which may be used in the practice of desired to produce the particular immunoglobulin is selected the methods disclosed herein include signal sequences, pro based on the desired characteristics that will be imparted to 65 moters, and transcription terminator sequences. It is generally the particular immunoglobulin produced by the host cell. For preferred that the regulatory sequences used be from a species example, a host cell that produces glycoproteins having pre orgenus that is the same as or closely related to that of the host US 8,067,339 B2 25 26 cellor is operational in the host cell type chosen. Examples of into nucleic acid molecules encoding a human immunoglo signal sequences include those of Saccharomyces cerevisiae bulin framework to produce a cDNA library encoding a plu invertase; the Aspergillus niger amylase and glucoamylase; rality of humanized immunoglobulins (See, for example, human serum albumin; Kluyveromyces maxianus inulinase; U.S. Pat. Nos. 6, 180,370; 6,632,927; and, 6,872,392). Chi and Pichia pastoris mating factor and Kar2. Signal sequences meric immunoglobulin-encoding cDNA libraries can be con shown herein to be useful in yeast and filamentous fungi structed by PCR amplifying the variable regions from the include, but are not limited to, the alpha mating factor prese cDNAs in the cDNA library from one species and integrating quence and preprosequence from Saccharomyces cerevisiae: the nucleic acid molecules encoding the PCR-amplified vari and signal sequences from numerous other species. able regions onto nucleic acid molecules encoding immuno Examples of promoters include promoters from numerous 10 globulin constant regions from another species to produce a species, including but not limited to alcohol-regulated pro cDNA library encoding a plurality of chimeric immunoglo moter, tetracycline-regulated promoters, steroid-regulated bulins (See, for example, U.S. Pat. No. 5,843.708). Various promoters (e.g., glucocorticoid, estrogen, ecdysone, retinoid, methods that have been developed for the creation of diversity thyroid), metal-regulated promoters, pathogen-regulated pro within protein libraries, including random mutagenesis moters, temperature-regulated promoters, and light-regu 15 (Daugherty et al., Proc. Natl Acad. Sci. USA,97:, 2029-2034 lated promoters. Specific examples of regulatable promoter (2000); Boder et al., Proc. Natl Acad. Sci. USA,97:, 10701 systems well known in the art include but are not limited to 10705 (2000); Holler et al., Proc. Natl Acad. Sci. USA, 97:, metal-inducible promoter systems (e.g., the yeast copper 5387-5392 (2000)), in vitro DNA shuffling (Stemmer, metallothionein promoter), plant herbicide safner-activated Nature, 370:,389-391 (1994); Stemmer, Proc. Natl Acad. Sci. promoter systems, plant heat-inducible promoter systems, USA, 91:, 10747-10751 (1994)), in vivo DNA shuffling plant and mammalian steroid-inducible promoter systems, (Swers et al., Nucl. Acid Res. 32: e36 (2004)), and site Cym repressor-promoter system (Krackeler Scientific, Inc. specific recombination (Rehberg et al., J. Biol. Chem. 257. Albany, N.Y.), RheoSwitch System (New England Biolabs, 11497-11502 (1982); Streuli et al., Proc. Natl Acad. Sci. Beverly Mass.), benzoate-inducible promoter systems (See USA, 78:, 2848-2852 (1981); Waterhouse et al., (1993) Nucl. WO2004/043885), and retroviral-inducible promoter sys 25 Acids Res., 21:, 2265-2266 (1993); Sblattero & Bradbury, tems. Other specific regulatable promoter systems well Nat. Biotechnol., 18:, 75-80 (2000)) can be used or adapted to known in the art include the tetracycline-regulatable systems produce the plurality of host cells disclosed herein that (See for example, Berens & Hillen, Eur J Biochem 270: express immunoglobulins and the capture moiety comprising 3109-3121 (2003)), RU486-inducible systems, ecdysone a cell Surface anchoring protein fused to a binding moiety that inducible systems, and kanamycin-regulatable system. 30 is capable of specifically binding an immunoglobulin. Lower eukaryote-specific promoters include but are not lim Production of active immunoglobulins requires proper ited to the Saccharomyces cerevisiae TEF-1 promoter, Pichia folding of the protein when it is produced and secreted by the pastoris GAPDH promoter, Pichia pastoris GUT1 promoter, cells. In E.coli, the complexity and large size of an antibody PMA-1 promoter, Pichia pastoris PCK-1 promoter, and presents an obstacle to proper folding and assembly of the Pichia pastoris AOX-1 and AOX-2 promoters. For temporal 35 expressed light and heavy chain polypeptides, resulting in expression of the GPI-IgG capture moiety and the immuno poor yield of intact antibody. The presence of effective globulins, the Pichia pastoris GUT1 promoter operably molecular chaperone proteins may be required, or may linked to the nucleic acid molecule encoding the GPI-IgG enhance the ability of the cell to produce and secrete properly capture moiety and the Pichia pastoris GAPDH promoter folded proteins. The use of molecular chaperone proteins to operably linked to the nucleic acid molecule encoding the 40 improve production of immunoglobulins in yeast has been immunoglobulin are shown in the examples herein to be disclosed in U.S. Pat. Nos. 5,772,245; 5,700,678 and 5,874, useful. 247; U.S. Application Publication No. 2002/0068325; Toman Examples of transcription terminator sequences include et al., J. Biol. Chem. 275: 23303-23309 (2000); Keizer-Gun transcription terminators from numerous species and pro nink et al., Martix Biol. 19: 29-36 (2000); Vad et al., J. teins, including but not limited to the Saccharomyces cerevi 45 Biotechnol. 116: 251-260 (2005); Inana et al., Biotechnol. siae cytochrome C terminator; and Pichia pastoris ALG3 and Bioengineer. 93: 771-778 (2005); Zhang et al., Biotechnol. PMA1 terminators. Prog. 22: 1090-1095 (2006); Damasceno et al., Appl. Micro Nucleic acid molecules encoding immunoglobulins can be biol. Biotechnol. 74: 381-389 (2006); Huo et al., Protein obtained from any Suitable source including spleen and liver Express. Purif. 54: 234-239 (2007); and copending applica cells and antigen-stimulated antibody producing cells, 50 tion Ser. No. 61/066,409, filed 20 Feb. 2008. obtained from either in vivo or in vitro sources. Regardless of As used herein, the methods can use host cells from any source, the cellular VH and VL mRNAs are reverse tran kind of cellular system which can be modified to express a scribed into VHandVL cDNA sequences. Reverse transcrip capture moiety comprising a cell Surface anchoring protein tion may be performed in a single step or in an optional fused to a binding moiety capable of binding an immunoglo combined reverse transcription/PCR procedure to produce 55 bulin and whole, intact immunoglobulins. Within the scope of cDNA libraries containing a plurality of immunoglobulin the invention, the term “cells' means the cultivation of indi encoding DNA molecules. (See, for example, Marks et al., J. vidual cells, tissues, organs, insect cells, avian cells, reptilian Mol. Biol. 222:581-596 (1991)). Nucleic acid molecules can cells, mammalian cells, hybridoma cells, primary cells, con also be synthesized de novo based on sequences in the scien tinuous cell lines, stem cells, plant cells, yeast cells, filamen tific literature. Nucleic acid molecules can also be synthe 60 tous fungal cells, and/or genetically engineered cells, such as sized by extension of overlapping oligonucleotides spanning recombinant cells expressing and displaying a glycosylated a desired sequence (See, e.g., Caldas et al., Protein Engineer immunoglobulin. ing, 13: 353–360 (2000)). Humanized immunoglobulin-en In a further embodiment, lower eukaryotes such as yeast or coding cDNA libraries can be constructed by PCR amplifying filamentous fungi are used for expression and display of the the complementary-determining regions (CDR) from the 65 immunoglobulins because they can be economically cul cDNAS in one or more libraries from any source and integrat tured, give high yields, and when appropriately modified are ing the PCR amplified CDR-encoding nucleic acid molecules capable of suitable glycosylation. Yeast particularly offers US 8,067,339 B2 27 28 established genetics allowing for rapid transfections, tested to the ER or Golgi apparatus of the host cell. Passage of a protein localization strategies and facile gene knock-out tech recombinant glycoprotein through the ER or Golgi apparatus niques. Suitable vectors have expression control sequences, of the host cell produces a recombinant glycoprotein com Such as promoters, including 3-phosphoglycerate kinase or prising a Mans GlcNAc glycoform, for example, a recombi other glycolytic enzymes, and an origin of replication, termi 5 nant glycoprotein composition comprising predominantly a nation sequences and the like as desired. Mans.GlcNAc glycoform. For example, U.S. Pat. No. 7,029. Host cells useful in the present invention include Pichia 872 and U.S. Published Patent Application Nos. 2004/ pastoris, Pichia finlandica, Pichia trehalophila, Pichia koc 0018590 and 2005/0170452 disclose lower eukaryote host lamae, Pichia membranaefaciens, Pichia minuta (Ogataea cells capable of producing a glycoprotein comprising a minuta, Pichia lindneri), Pichia opuntiae, Pichia thermotol 10 Mans.GlcNAc glycoform. erans, Pichia salictaria, Pichia guercuum, Pichia piperi, In a further embodiment, the immediately preceding host Pichia stiptis, Pichia methanolica, Pichia sp., Saccharomy cell further includes a GlcNAc transferase I (GnTI) catalytic ces cerevisiae, Saccharomyces sp., Hansenula polymorpha, domain fused to a cellular targeting signal peptide not nor Kluyveromyces sp., Kluyveromyces lactis, Candida albicans, mally associated with the catalytic domain and selected to Aspergillus nidulans, Aspergillus niger; Aspergillus Oryzae, 15 target GlcNAc transferase I activity to the ER or Golgi appa Trichoderma reesei, Chrysosporium lucknowense, Fusarium ratus of the host cell. Passage of the recombinant glycoprotein sp., Fusarium gramineum, Fusarium venematum and Neuro through the ER or Golgi apparatus of the host cell produces a spora Crassa. Various yeasts, such as K. lactis, Pichia pas recombinant glycoprotein comprising a toris, Pichia methanolica, and Hansenula polymorpha are GlcNAcMans.GlcNAc glycoform, for example a recombi particularly suitable for cell culture because they are able to nant glycoprotein composition comprising predominantly a grow to high cell densities and secrete large quantities of GlcNAcMan-GlcNAc, glycoform. U.S. Pat. No. 7,029,872 recombinant protein. Likewise, filamentous fungi. Such as and U.S. Published Patent Application Nos. 2004/0018590 Aspergillus niger, Fusarium sp., Neurospora crassa and oth and 2005/0170452 disclose lower eukaryote host cells ers can be used to produce glycoproteins of the inventionatan capable of producing a glycoprotein comprising a industrial scale. In the case of lower eukaryotes, cells are 25 GlcNAcMans.GlcNAc glycoform. The glycoprotein pro routinely grown from between about 1.5 to 3 days under duced in the above cells can be treated in vitro with a hexam conditions that induce expression of the capture moiety. The inidase to produce a recombinant glycoprotein comprising a induction of immunoglobulin expression while inhibiting Mans.GlcNAc glycoform. expression of the capture moiety is for about 1 to 2 days. In a further embodiment, the immediately preceding host Afterwards, the cells are analyzed for those cells that display 30 cell further includes a mannosidase II catalytic domain fused the immunoglobulin of interest. to a cellular targeting signal peptide not normally associated Lower eukaryotes, particularly yeast and filamentous with the catalytic domain and selected to target mannosidase fungi, can be genetically modified so that they express gly II activity to the ER or Golgi apparatus of the host cell. coproteins in which the glycosylation pattern is human-like Passage of the recombinant glycoprotein through the ER or or humanized. In this manner, glycoprotein compositions can 35 Golgi apparatus of the host cell produces a recombinant gly be produced in which a specific desired glycoform is pre coprotein comprising a GlcNAcMan-GlcNAc glycoform, dominant in the composition. Such can be achieved by elimi for example a recombinant glycoprotein composition com nating selected endogenous glycosylation enzymes and/or prising predominantly a GlcNAcMan-GlcNAc glycoform. genetically engineering the host cells and/or supplying exog U.S. Pat. No. 7,029,872 and U.S. Published Patent Applica enous enzymes to mimic all or part of the mammalian glyco 40 tion No. 2004/0230042 discloses lower eukaryote host cells sylation pathway as described in US 2004/0018590. If that express mannosidase II enzymes and are capable of pro desired, additional genetic engineering of the glycosylation ducing glycoproteins having predominantly a can be performed, such that the glycoprotein can be produced GlcNAc-Man-GlcNAc glycoform. The glycoprotein pro with or without core fucosylation. Use of lower eukaryotic duced in the above cells can be treated in vitro with a hexam host cells is further advantageous in that these cells are able to 45 inidase to produce a recombinant glycoprotein comprising a produce highly homogenous compositions of glycoprotein, Man-GlcNAc glycoform. Such that the predominant glycoform of the glycoprotein may In a further embodiment, the immediately preceding host be present as greater than thirty mole percent of the glyco cell further includes GlcNAc transferase II (GnT II) catalytic protein in the composition. In particular aspects, the predomi domain fused to a cellular targeting signal peptide not nor nant glycoform may be present in greater than forty mole 50 mally associated with the catalytic domain and selected to percent, fifty mole percent, sixty mole percent, seventy mole target GlcNAc transferase II activity to the ER or Golgi appa percent and, most preferably, greater than eighty mole per ratus of the host cell. Passage of the recombinant glycoprotein cent of the glycoprotein present in the composition. through the ER or Golgi apparatus of the host cell produces a Lower eukaryotes, particularly yeast, can be genetically recombinant glycoprotein comprising a modified so that they express glycoproteins in which the 55 GlcNAc-Man-GlcNAc glycoform, for example a recombi glycosylation pattern is human-like or humanized. Such can nant glycoprotein composition comprising predominantly a be achieved by eliminating selected endogenous glycosyla GlcNAc-Man-GlcNAc glycoform. U.S. Pat. No. 7,029,872 tion enzymes and/or Supplying exogenous enzymes as and U.S. Published Patent Application Nos. 2004/0018590 described by Gerngross et al., US 20040018590. For and 2005/0170452 disclose lower eukaryote host cells example, a host cell can be selected or engineered to be 60 capable of producing a glycoprotein comprising a depleted in 1.6-mannosyltransferase activities, which would GlcNAc-Man-GlcNAc glycoform. The glycoprotein pro otherwise add mannose residues onto the N-glycan on a gly duced in the above cells can be treated in vitro with a hexam coprotein. inidase to produce a recombinant glycoprotein comprising a In one embodiment, the host cell further includes an O. 12 Man-GlcNAc glycoform. mannosidase catalytic domain fused to a cellular targeting 65 In a further embodiment, the immediately preceding host signal peptide not normally associated with the catalytic cell further includes a galactosyltransferase catalytic domain domain and selected to target the C.1.2-mannosidase activity fused to a cellular targeting signal peptide not normally asso US 8,067,339 B2 29 30 ciated with the catalytic domain and selected to target galac binant glycoprotein through the ER or Golgi apparatus of the tosyltransferase activity to the ER or Golgi apparatus of the host cell produces a recombinant glycoprotein comprising a host cell. Passage of the recombinant glycoprotein through NANAGalGlcNAcMans.GlcNAc glycoform. the ER or Golgi apparatus of the host cell produces a recom Various of the preceding host cells further include one or binant glycoprotein comprising a more sugar transporters such as UDP-GlcNAc transporters GalGlcNAc-Man-GlcNAc or Gal-GlcNAc-Man-GlcNAc (for example, Kluyveromyces lactis and Mus musculus UDP glycoform, or mixture thereof for example a recombinant GlcNAc transporters), UDP-galactose transporters (for glycoprotein composition comprising predominantly a example, Drosophila melanogaster UDP-galactose trans GalGlcNAc-Man-GlcNAc glycoform O porter), and CMP-sialic acid transporter (for example, human GalGlcNAc-Man-GlcNAc glycoform or mixture thereof. 10 U.S. Pat. No. 7,029,872 and U.S. Published Patent Applica sialic acid transporter). Because lower eukaryote host cells tion No. 2006/0040353 discloses lower eukaryote host cells Such as yeast and filamentous fungi lack the above transport capable of producing a glycoprotein comprising a ers, it is preferable that lower eukaryote host cells such as GalGlcNAc-Man-GlcNAc glycoform. The glycoprotein yeast and filamentous fungi be genetically engineered to produced in the above cells can be treated in vitro with a 15 include the above transporters. galactosidase to produce a recombinant glycoprotein com Host cells further include lower eukaryote cells (e.g., yeast prising a GlcNAc-Man-GlcNAc glycoform, for example a Such as Pichia pastoris) that are genetically engineered to recombinant glycoprotein composition comprising predomi eliminate glycoproteins having C-mannosidase-resistant nantly a GlcNAc-Man-GlcNAc glycoform. N-glycans by deleting or disrupting one or more of the In a further embodiment, the immediately preceding host B-mannosyltransferase genes (e.g., BMT1, BMT2, BMT3, cell further includes a sialyltransferase catalytic domain and BMT4)(See, U.S. Published Patent Application No. fused to a cellular targeting signal peptide not normally asso 2006/0211085) and glycoproteins having phosphomannose ciated with the catalytic domain and selected to target Sia residues by deleting or disrupting one or both of the phospho lytransferase activity to the ER or Golgi apparatus of the host mannosyl transferase genes PNO1 and MNN4B (See for cell. Passage of the recombinant glycoprotein through the ER 25 example, U.S. Pat. Nos. 7,198,921 and 7.259,007), which in or Golgi apparatus of the host cell produces a recombinant further aspects can also include deleting or disrupting the glycoprotein comprising predominantly a MNN4A gene. Disruption includes disrupting the open read NANAGalGlcNAc-Man-GlcNAc glycoform O ing frame encoding the particular enzymes or disrupting NANAGal-GlcNAc-Man-GlcNAc glycoform or mixture expression of the open reading frame or abrogating transla thereof. For lower eukaryote host cells such as yeast and 30 tion of RNAS encoding one or more of the B-mannosyltrans filamentous fungi, it is useful that the host cell further include ferases and/or phosphomannosyltransferases using interfer a means for providing CMP-sialic acid for transfer to the ing RNA, antisense RNA, or the like. The host cells can N-glycan. U.S. Published Patent Application No. 2005/ further include any one of the aforementioned host cells 0260729 discloses a method for genetically engineering modified to produce particular N-glycan structures. lower eukaryotes to have a CMP-sialic acid synthesis path 35 Host cells further include lower eukaryote cells (e.g., yeast way and U.S. Published Patent Application No. 2006/ Such as Pichia pastoris) that are genetically modified to con 0286637 discloses a method for genetically engineering trol O-glycosylation of the glycoprotein by deleting or dis lower eukaryotes to produce sialylated glycoproteins. The rupting one or more of the protein O-mannosyltransferase glycoprotein produced in the above cells can be treated in (Dol-P-Man:Protein (Ser/Thr) Mannosyl Transferase genes) vitro with a neuraminidase to produce a recombinant glyco 40 (PMTs) (See U.S. Pat. No. 5,714,377) or grown in the pres protein comprising predominantly a ence of Pmtp inhibitors and/or an alpha-mannosidase as dis GalGlcNAc-Man-GlcNAc glycoform O closed in Published International Application No. WO GalGlcNAc-Man-GlcNAc glycoform or mixture thereof. 2007061631, or both. Disruption includes disrupting the open Any one of the preceding host cells can further include one reading frame encoding the Pmtp or disrupting expression of or more GlcNAc transferase selected from the group consist 45 the open reading frame or abrogating translation of RNAS ing of GnT III, GnT IV. GnTV, GnT VI, and GnT IX to encoding one or more of the PmtpS using interfering RNA, produce glycoproteins having bisected (GnT III) and/or mul antisense RNA, or the like. The host cells can further include tiantennary (GnT IV, V, VI, and IX) N-glycan structures such any one of the aforementioned host cells modified to produce as disclosed in U.S. Published Patent Application Nos. 2004/ particular N-glycan structures. O74458 and 2007/OO37248. 50 Pmtp inhibitors includebut are not limited to a benzylidene In further embodiments, the host cell that produces glyco thiazolidinediones. Examples of benzylidene thiazolidinedi proteins that have predominantly GlcNAcMans.GlcNAc ones that can be used are 5-3,4-bis(phenylmethoxy) phenyl N-glycans further includes a galactosyltransferase catalytic methylene-4-oxo-2-thioxo-3-thiazolidineacetic Acid; 5-3- domain fused to a cellular targeting signal peptide not nor (1-Phenylethoxy)-4-(2-phenylethoxy)phenylmethylene mally associated with the catalytic domain and selected to 55 4-oxo-2-thioxo-3-thiazolidineacetic Acid; and 5-3-(1- target Galactosyltransferase activity to the ER or Golgi appa Phenyl-2-hydroxy)ethoxy)-4-(2-phenylethoxy)phenyl ratus of the host cell. Passage of the recombinant glycoprotein methylene-4-oxo-2-thioxo-3-thiazolidineacetic Acid. through the ER or Golgi apparatus of the host cell produces a In particular embodiments, the function or expression of at recombinant glycoprotein comprising predominantly the least one endogenous PMT gene is reduced, disrupted, or GalGlcNAcMans GlcNAc glycoform. 60 deleted. For example, in particular embodiments the function In a further embodiment, the immediately preceding host or expression of at least one endogenous PMT gene selected cell that produced glycoproteins that have predominantly the from the group consisting of the PMT1, PMT2, PMT3, and GalGlcNAcMans.GlcNAc, N-glycans further includes a sia PMT4 genes is reduced, disrupted, or deleted; or the host cells lyltransferase catalytic domain fused to a cellular targeting are cultivated in the presence of one or more PMT inhibitors. signal peptide not normally associated with the catalytic 65 In further embodiments, the host cells include one or more domain and selected to target sialytransferase activity to the PMT gene deletions or disruptions and the host cells are ER or Golgi apparatus of the host cell. Passage of the recom cultivated in the presence of one or more Pmtp inhibitors. In US 8,067,339 B2 31 32 particular aspects of these embodiments, the host cells also methods herein to identify and select antibodies with speci express a secreted alpha-1,2-mannosidase. ficity for a particular antigen based upon affinity or avidity of PMT deletions or disruptions and/or Pmtp inhibitors con the antibody for the antigen without identification and selec trol O-glycosylation by reducing O-glycosylation occupancy, tion of the antibody being influenced by the O-glycosylation that is by reducing the total number of O-glycosylation sites system of the host cell. Thus, controlling O-glycosylation on the glycoprotein that are glycosylated. The further addi further enhances the usefulness of yeast or fungal host cells to tion of an alpha-1,2-mannsodase that is secreted by the cell identify and selectantibodies that will ultimately be produced controls O-glycosylation by reducing the mannose chain in a mammalian cell line. length of the O-glycans that are on the glycoprotein. Thus, Yield of antibodies can in some situations be improved by combining PMT deletions or disruptions and/or Pmtp inhibi 10 overexpressing nucleic acid molecules encoding mammalian tors with expression of a secreted alpha-1,2-mannosidase or human chaperone proteins or replacing the genes encoding controls O-glycosylation by reducing occupancy and chain one or more endogenous chaperone proteins with nucleic acid length. In particular circumstances, the particular combina molecules encoding one or more mammalian or human chap tion of PMT deletions or disruptions, Pmtp inhibitors, and erone proteins. In addition, the expression of mammalian or alpha-1,2-mannosidase is determined empirically as particu 15 human chaperone proteins in the host cell also appears to lar heterologous glycoproteins (antibodies, for example) may control O-glycosylation in the cell. Thus, further included are be expressed and transported through the Golgi apparatus the host cells herein wherein the function of at least one with different degrees of efficiency and thus may require a endogenous gene encoding a chaperone protein has been particular combination of PMT deletions or disruptions, reduced or eliminated, and a vector encoding at least one Pmtp inhibitors, and alpha-1,2-mannosidase. In another mammalian or human homolog of the chaperone protein is aspect, genes encoding one or emore endogenous mannosyl expressed in the host cell. Also included are host cells in transferase enzymes are deleted. This deletion(s) can be in which the endogenous host cell chaperones and the mamma combination with providing the secreted alpha-1,2-mannosi lian or human chaperone proteins are expressed. In further dase and/or PMT inhibitors or can be in lieu of providing the aspects, the lower eukaryotic host cell is a yeast or filamen secreted alpha-1,2-mannosidase and/or PMT inhibitors. 25 tous fungi host cell. Examples of the use of chaperones of host Thus, the control of O-glycosylation can be useful for cells in which human chaperone proteins are introduced to producing particular glycoproteins in the host cells disclosed improve the yield and reduce or control O-glycosylation of herein in better total yield or in yield of properly assembled recombinant proteins has been disclosed in U.S. Provisional glycoprotein. The reduction or elimination of O-glycosyla Application Nos. 61/066409 filed Feb. 20, 2008 and 61/188, tion appears to have a beneficial effect on the assembly and 30 723 filed Aug. 12, 2008. Like above, further included are transport of whole antibodies as they traverse the secretory lower eukaryotic host cells wherein, in addition to replacing pathway and are transported to the cell surface. Thus, in cells the genes encoding one or more of the endogenous chaperone in which O-glycosylation is controlled, the yield of properly proteins with nucleic acid molecules encoding one or more assembled antibodies fragments is increased over the yield mammalian or human chaperone proteins or overexpressing obtained in host cells in which O-glycosylation is not con 35 one or more mammalian or human chaperone proteins as trolled. described above, the function or expression of at least one In addition, O-glycosylation may have an effect on an endogenous gene encoding a protein O-mannosyltransferase antibody's affinity and/or avidity for an antigen. This can be (PMT) protein is reduced, disrupted, or deleted. In particular particularly significant when the ultimate host cell for pro embodiments, the function of at least one endogenous PMT duction of the antibody is not the same as the host cell that was 40 gene selected from the group consisting of the PMT1, PMT2, used for selecting the antibody. For example, O-glycosylation PMT3, and PMT4 genes is reduced, disrupted, or deleted. might interfere with an antibody's affinity for an antigen, thus Therefore, the methods disclose herein can use any host an antibody that might otherwise have high affinity for an cell that has been genetically modified to produce glycopro antigen might not be identified because O-glycosylation may teins that have no N-glycans compositions wherein the pre interfere with the ability of the antibody to bind the antigen. In 45 dominant N-glycan is selected from the group consisting of other cases, an antibody that has high avidity for an antigen complex N-glycans, hybrid N-glycans, and high mannose might not be identified because O-glycosylation interferes N-glycans wherein complex N-glycans are selected from the with the antibody's avidity for the antigen. In the preceding group consisting of Man-GlcNAc, GlcNAC. two cases, an antibody that might be particularly effective Man-GlcNAc, Gala GlcNAc-Man, GlcNAc, and when produced in a mammalian cell line might not be iden 50 NANA Gala Man-GlcNAca; hybrid N-glycans are tified because the host cells for identifying and selecting the selected from the group consisting of Mans GlcNAc, antibody was of another cell type, for example, a yeast or GlcNAcMans.GlcNAc, GalGlcNAcMans GlcNAc, and fungal cell (e.g., a Pichia pastoris host cell). It is well known NANAGalGlcNAcMans.GlcNAc, and high Mannose N-gly that O-glycosylation in yeast can be significantly different cans are selected from the group consisting of Man GlcNAc from O-glycosylation in mammalian cells. This is particu 55 Man,GlcNAc, MansGlcNAc, and Man-GlcNAca. In par larly relevant when comparing wild type yeast o-glycosyla ticular aspects, the composition of N-glycans comprises tion with mucin-type or dystroglycan type O-glycosylation in about 39% GlcNAC-Man-GlcNAc 40% mammals. In particular cases, O-glycosylation might Gal GlcNAC-Man-GlcNAc and 6% enhance the antibody's affinity or avidity for an antigen GalGlcNAC-Man-GlcNAc O about 60% instead of interfere. This effect is undesirable when the pro 60 GlcNAC. Man-GlcNAc; 17% GalGlcNAC. Man-GlcNAc: duction host cell is to be different from the host cell used to and 5% Gal-GlcNAC, Man-GlcNAc, or mixtures in identify and select the antibody (for example, identification between. and selection is done in yeast and the production host is a In the above embodiments in which the yeast cell does not mammalian cell) because in the production host the O-gly display 1.6-mannosyltransferase activity (that is, the OCHI cosylation will no longer be of the type that caused the 65 gene encoding ochlp has been disrupted or deleted), the host enhanced affinity or avidity for the antigen. Therefore, con cell is not capable of mating. Thus, depending on the effi trolling O-glycosylation can enable use of the materials and ciency of transformation, the potential library diversity of US 8,067,339 B2 33 34 light chains and heavy chains appears to be limited to a heavy sylation at position 297 of the antibody or fragment thereof chain library of between about 10 to 10° diversity and a light when the nucleic acid molecule encoding the antibody chain library of about 10 to 10° diversity. However, in a yeast thereof is introduced into a host cell that has been engineered host cell that is capable of mating, the diversity can be to make glycoproteins that have hybrid or complex N-glycans increased to about 10 to 10' because the host cells express as discussed previously. ing the heavy chain library can be mated to host cells express The cell systems used for recombinant expression and ing the light chain library to produce host cells that express display of the immunoglobulin can also be any higher eukary heavy chain/light chain library. Therefore, in particular ote cell, tissue, organism from the animal kingdom, for embodiments, the host cell is a yeast cell such as Pichia example transgenic goats, transgenic rabbits, CHO cells, pastoris that displays 1.6-mannosyl transferase activities 10 insect cells, and human cell lines. Examples of animal cells (that is, has an OCH1 gene encoding a function ochlp) but include, but are not limited to, SC-I cells, LLC-MK cells, which is modified as described herein to display antibodies or CV-I cells, CHO cells, COS cells, murine cells, human cells, fragments thereof on the cell surface. In these embodiments, HeLa cells, 293 cells, VERO cells, MDBK cells, MDCK the host cell can be a host cell with its native glycosylation cells, MDOK cells, CRFK cells, RAF cells, TCMK cells, pathway. 15 LLC-PK cells, PK15 cells, WI-38 cells, MRC-5 cells, T-FLY Yeast selectable markers that can be used in the present cells, BHK cells, SP2/0, NSO cells, and derivatives thereof. invention include drug resistance markers and genetic func Insect cells include cells of Drosophila melanogaster origin. tions which allow the yeast host cell to synthesize essential These cells can be genetically engineered to render the cells cellular nutrients, e.g. amino acids. Drug resistance markers capable of making immunoglobulins that have particular or which are commonly used in yeast include chloramphenicol, predominantly particular N-glycans. For example, U.S. Pat. kanamycin, methotrexate, G418 (geneticin), Zeocin, and the No. 6,949,372 discloses methods for making glycoproteins in like. Genetic functions which allow the yeast host cell to insect cells that are sialylated. Yamane-Ohnuki et al. Biotech synthesize essential cellular nutrients are used with available nol. Bioeng. 87: 614-622 (2004), Kanda et al., Biotechnol. yeast strains having auxotrophic mutations in the correspond Bioeng. 94: 680-688 (2006), Kanda et al., Glycobiol. 17: ing genomic function. Common yeast selectable markers pro 25 104-118 (2006), and U.S. Pub. Application Nos. 2005/ vide genetic functions for synthesizing leucine (LEU2), tryp 0216958 and 2007/0020260 disclose mammalian cells that tophan (TRP1 and TRP2), proline (PRO1), uracil (URA3, are capable of producing immunoglobulins in which the URA5, URA6), histidine (HIS3), lysine (LYS2), adenine N-glycans thereon lack fucose or have reduced fucose. (ADE1 or ADE2), and the like. Other yeast selectable mark In particular embodiments, the higher eukaryote cell, tis ers include the ARR3 gene from S. cerevisiae, which confers 30 Sue, organism can also be from the plant kingdom, for arsenite resistance to yeast cells that are grown in the presence example, wheat, rice, corn, tobacco, and the like. Alterna of arsenite (Bobrowicz et al., Yeast, 13:819-828 (1997); tively, bryophyte cells can be selected, for example from Wysocki et al., J. Biol. Chem. 272:30061-30066 (1997)). A species of the genera Physcomitrella, Funaria, Sphagnum, number of suitable integration sites include those enumerated Ceratodon, Marchantia, and Sphaerocarpos. Exemplary of in U.S. Published application No. 2007/0072262 and include 35 plant cells is the bryophyte cell of Physcomitrella patens, homologs to loci known for Saccharomyces cerevisiae and which has been disclosed in WO 2004/057002 and WO2008/ other yeast or fungi. Methods for integrating vectors into 006554. Expression systems using plant cells can further yeast are well known, for example, see U.S. Pat. No. 7,479, manipulated to have altered glycosylation pathways to enable 389, WO2007136865, and PCT/US2008/13719. Examples the cells to produce immunoglobulins that have predomi of insertion sites include, but are not limited to, Pichia ADE 40 nantly particular N-glycans. For example, the cells can be genes; Pichia TRP (including TRP1 through TRP2) genes: genetically engineered to have a dysfunctional or no core Pichia MCA genes; Pichia CYM genes: Pichia PEP genes: fucosyltransferase and/or a dysfunctional or no Xylosyltrans Pichia PRB genes; and Pichia LEU genes. The Pichia ADE1 ferase, and/or a dysfunctional or no B1,4-galactosyltrans and ARG4 genes have been described in Lin Cereghino et al., ferase. Alternatively, the galactose, fucose and/or Xylose can Gene 263:159-169 (2001) and U.S. Pat. No. 4,818,700, the 45 be removed from the immunoglobulin by treatment with HIS3 and TRP1 genes have been described in Cosano et al., enzymes removing the residues. Any enzyme resulting in the Yeast 14:861-867 (1998), HIS4 has been described in Gen release of galactose, fucose and/or Xylose residues from Bank Accession No. X56180. N-glycans which are known in the art can be used, for In embodiments that express whole antibodies, the nucleic example O-galactosidase, B-Xylosidase, and O-fucosidase. acid molecule encoding the antibody or heavy chain fragment 50 Alternatively an expression system can be used which syn thereof is modified to replace the codon encoding an aspar thesizes modified N-glycans which can not be used as Sub agine residue at position 297 of the molecule (the glycosyla strates by 1.3-fucosyltransferase and/or 1.2-Xylosyltrans tion site) with a codon encoding any otheramino acid residue. ferase, and/or 1,4-galactosyltransferase. Methods for Thus, the antibody that is produced in the host cell is not modifying glycosylation pathways in plant cells has been glycosylated. In this embodiment, the host cell displaying the 55 disclosed in U.S. Published Application No. 2004/0018590. heavy chain library is mated to the host cell displaying the The methods disclosed herein can be adapted for use in light chain library and the resulting combinatorial library is mammalian, insect, and plant cells. The regulatable promot screened as taught herein. Because the antibodies lack N-gly ers selected for regulating expression of the expression cas cosylation, the non-human yeast N-glycans of the host cell settes in mammalian, insect, or plant cells should be selected which might interfere with antibody affinity for a desired 60 for functionality in the cell-type chosen. Examples of suitable antigen are not present on the recombinant antibodies. Cells regulatable promoters include but are not limited to the tetra producing antibodies that have desired affinity for an antigen cycline-regulatable promoters (See for example, Berens & of interest are selected. The nucleic acid molecules encoding Hillen, Eur. J. Biochem. 270: 3109-3121 (2003)), RU486 the heavy and light chains of the antibody thereofare removed inducible promoters, ecdysone-inducible promoters, and from the cells and the nucleic acid molecule encoding the 65 kanamycin-regulatable systems. These promoters can heavy chain is modified to reintroduce an asparagine residue replace the promoters exemplified in the expression cassettes at position 297. This enables appropriate human-like glyco described in the examples. The capture moiety can be fused to US 8,067,339 B2 35 36 a cell Surface anchoring protein Suitable for use in the cell TTACA GTTCA TCATG CACAG CTTTC TGATCAT 3 type chosen. Cell Surface anchoring proteins including GPI (SEQID NO: 2), Pfuturbo DNA polymerase (Stratagene, La proteins are well known for mammalian, insect, and plant Jolla, Calif.), and a human liver cDNA (BD Bioscience, San cells. GPI-anchored fusion proteins has been described by Jose, Calif.). The PCR conditions were 1 cycle of 95°C. for Kennard et al., Methods Biotechnol. Vo. 8: Animal Cell Bio two minutes, 25 cycles of 95°C. for 20 seconds, 58° C. for 30 technology (Ed. Jenkins. Human Press, Inc., Totowa, N.J.) seconds, and 72° C. for 1.5 followed by one cycle of 72°C. for pp. 187-200 (1999). The genome targeting sequences for 10 minutes. The resulting PCR product was cloned into plas integrating the expression cassettes into the host cell genome mid vector pCR2.1 to make plasmid vector pGLY618. The for making stable recombinants can replace the genome tar nucleotide and amino acid sequences of the human PDI1 geting and integration sequences exemplified in the 10 (SEQ ID NOS:39 and 40, respectively) are shown in Table 1. examples. Transfection methods for making stable and tran The nucleotide and amino acid sequences of the Pichia siently transfected mammalian, insect, plant host cells are pastoris PDI1 (SEQ ID NOs:41 and 42, respectively) are well known in the art. Once the transfected host cells have shown in Table 1. Isolation of nucleic acid molecules com been constructed as disclosed herein, the cells can be prising the Pichia pastoris PDI5' and 3' regions was per screened for expression of the immunoglobulin of interest 15 formed by PCR amplification of the regions from Pichia and selected as disclosed herein. pastoris genomic DNA. The 5' region was amplified using The present invention also encompasses kits containing the primers PB248: 5 ATGAA TTCAG GCCAT ATCGG expression and helper vectors of this invention in suitable CCATT GTTTACTGTG CGCCCACAGT AG 3' (SEQ ID packaging. Each kit necessarily comprises the reagents which NO: 3); PB249: 5' ATGTT TAAAC GTGAG GATTA render the delivery of vectors into a host cell possible. The CTGGT GATGA AAGAC 3' (SEQID NO: 4). The 3' region selection of reagents that facilitate delivery of the vectors may was amplified using primers PB250: 5' AGACT AGTCT vary depending on the particular transfection or infection ATTTG GAGAC ATTGACGGAT CCAC 3' (SEQ ID NO: method used. The kits may also contain reagents useful for 5); PB251: 5' ATCTC GAGAG GCCAT GCAGG CCAAC generating labeled polynucleotide probes or proteinaceous CACAA GATGA ATCAA ATTTT G-3' (SEQ ID NO: 6). probes for detection of exogenous sequences and the protein 25 Pichia pastoris strain NRRL-11430 genomic DNA was used product. Each reagent can be supplied in a Solid form or for PCR amplification. The PCR conditions were one cycle of dissolved/suspended in a liquid buffer suitable for inventory 95°C. for two minutes, 25 cycles of 95°C. for 30 seconds, 55° storage, and later for exchange or addition into the reaction C. for 30 seconds, and 72°C. for 2.5 minutes, and followed by medium when the experiment is performed. Suitable packag one cycle of 72° C. for 10 minutes. The resulting PCR frag ing is provided. The kit can optionally provide additional 30 ments, PpPDI1 (5') and PpPDI1 (3'), were separately cloned components that are useful in the procedure. These optional into plasmid vector pCR2.1 to make plasmid vectors components include, but are not limited to, buffers, capture pGLY620 and pGLY617, respectively. To construct reagents, developing reagents, labels, reacting Surfaces, pGLY678, DNA fragments Pp.ARG3-5 and PpARG-3' of means for detection, control samples, instructions, and inter integration plasmid vector pGLY24, which targets the plas pretive information. 35 mid vector to Pichia pastoris ARG3 locus, were replaced All publications, patents, and other references mentioned with DNA fragments PpPDI (5') and PpPDI (3'), respectively, herein are hereby incorporated by reference in their entireties. which targets the plasmid vector pGLY678 to the PDI1 locus The following examples are intended to promote a further and disrupts expression of the PDI1 locus. understanding of the present invention. The nucleic acid molecule encoding the human PDI was 40 then cloned into plasmid vector pGLY678 to produce plasmid EXAMPLE1 vector pGLY642 in which the nucleic acid molecule encoding the human PDI was placed under the control of the Pichia Utility of the invention was demonstrated using Pichia pastoris GAPDH promoter (PpGAPDH). Expression/inte pastoris as a model. The glycoengineered Pichia pastoris gration plasmid vector pGLY642 was constructed by ligating strain yCLY2696 was the background strain used. In strain 45 a nucleic acid molecule encoding the Saccharomyces cerevi yGLY2696, the gene encoding the endogenous PDI replaced siae alpha mating factor (MF) presequence signal peptide with a nucleic acid molecule encoding the human PDI and a (ScoMFpre-signal peptide) having a Not restriction enzyme nucleic acid molecule encoding the human GRP94 protein site at the 5' end and a blunt 3' end and the expression cassette inserted into the PEP4 locus. The strain was further engi comprising the nucleic acid molecule encoding the human neered to alter the endogenous glycosylation pathway to pro 50 PDI released from plasmid vector pGLY618 with Afel and duce glycoproteins that have predominantly Mans.GlcNAc PacI to produce a nucleic acid molecule having a blunt 5' end N-glycans. Strain YGLY2696 has been disclosed in co-pend and a PacI site at the 3' end into plasmid vector pGLY678 ing Application Ser. No. 61/066,409, filed 20 Feb. 2008. This digested with Not and PacI. The resulting integration/ex strain was shown to be useful for producing immunoglobulins pression plasmid vector pGLY642 comprises an expression and for producing immunoglobulins that have reduced O-gly 55 cassette encoding a human PDI1/ScC.MFpre-signal peptide cosylation. Construction of strain yOLY2696 involved the fusion protein operably linked to the Pichia pastoris pro following steps. moter and nucleic acid molecule sequences to target the plas Construction of expression/integration plasmid vector mid vector to the Pichia pastoris PDI1 locus for disruption of pGLY642 comprising an expression cassette encoding the the PDI1 locus and integration of the expression cassette into human PDI protein and nucleic acid molecules to target the 60 the PDI1 locus. FIG. 2 illustrates the construction of plasmid plasmid vector to the Pichia pastoris PDI1 locus for replace vector pGLY642. The nucleotide and amino acid sequences ment of the gene encoding the Pichia pastoris PDI1 with a of the ScC.MFpre-signal peptide are shown in SEQID NOs: nucleic acid molecule encoding the human PDI was as fol 27 and 28, respectively. lows and is shown in FIG.8. cDNA encoding the human PDI1 Construction of expression/integration vector pGLY2233 was amplified by PCR using the primers hPDI/UP1: 5' 65 encoding the human GRP94 protein was as follows and is AGCGC TGACG CCCCC GAGGA GGAGG ACCAC 3' shown in FIG.3. The human GRP94 was PCR amplified from (SEQ ID NO: 1) and hPDI/LP-PacI: 5' CCTTA ATTAA human liver cDNA (BD Bioscience) with the primers US 8,067,339 B2 37 38 hGRP94/UP 1 : 5'-AGCGC TGACG ATGAA GTTGA to arsenite. This was accomplished by isolating the ScARR3 TGTGG ATGGT ACAGT AG-3"; (SEQ ID NO: 15); and gene from pGFI 166 digested with AscI and the AscI ends hGRP94/LP1: 5'-GGCCG GCCTTACAATTCATCATGTT made blunt) and BglII, and inserting the fragment into CAGCT GTAGA TTC 3"; (SEQ ID NO: 16). The PCR con pGLY1896 that digested with Spel and the Spel ends made ditions were one cycle of 95°C. for two minutes, 25 cycles of 5 blunt and BglII. Integration of the plasmid vector is to the 95° C. for 20 seconds, 55° C. for 20 seconds, and 72° C. for Pichia pastoris PRO1 locus and selection is using the Sac 2.5 minutes, and followed by one cycle of 72° C. for 10 charomyces cerevisiae ARR3 gene. A map of plasmid vector minutes. The PCR product was cloned into plasmid vector pGFI2007 t is shown in FIG. 4. The ARR3 gene from S. pCR2.1 to make plasmid vector pGLY2216. The nucleotide cerevisiae confers arsenite resistance to cells that are grown in and amino acid sequences of the human GRP94 (SEQ ID 10 the presence of arsenite (Bobrowicz et al., Yeast, 13:819-828 NOs:43 and 44, respectively) are shown in Table 1. (1997); Wysocki et al., J. Biol. Chem. 272:30061-066 The nucleic acid molecule encoding the human GRP94 (1997)). was released from plasmid vector pGLY2216 with Afel and Yeast transfections with the above expression/integration FseI. The nucleic acid molecule was then ligated to a nucleic vectors were as follows. Pichiapastoris Strains were grown in acid molecule encoding the ScCMPpre-signal peptide having 15 50 mL YPD media (yeast extract (1%), peptone (2%), dex NotI and blunt ends as above and plasmid vector pGLY223 1 trose (2%)) overnight to an OD of between about 0.2 to 6. digested with NotI and Fsel carrying nucleic acid molecules After incubation on ice for 30 minutes, cells were pelleted by comprising the Pichia pastoris PEP4 5' and 3' regions centrifugation at 2500-3000 rpm for 5 minutes. Media was (PpPEP4-5' and PpPEP4-3' regions, respectively) to make removed and the cells washed three times with ice cold sterile plasmid vector pGLY2229. Plasmid vector pGLY2229 was 1 Msorbitol before resuspending in 0.5 ml ice cold sterile 1 M digested with BglII and Notland a DNA fragment containing sorbitol. Ten uL linearized DNA (5-20 lug) and 100 uL cell the PpPDI1 promoter was removed from plasmid vector Suspension was combined in an electroporation cuvette and pGLY2187 with BglII and NotI and the DNA fragment incubated for 5 minutes on ice. Electroporation was in a ligated into pGLY2229 to make plasmid vector pGLY2233. Bio-Rad GenePulser Xcell following the preset Pichia pas Plasmid vector pGLY2233 encodes the human GRP94 fusion 25 toris protocol (2 kV. 25uF, 200 S2), immediately followed by protein under control of the Pichia pastoris PDI promoter and the addition of 1 mLYPDS recovery media (YPD media plus includes the 5' and 3' regions of the Pichia pastoris PEP4 gene 1 M sorbitol). The transfected cells were allowed to recover to target the plasmid vector to the PEP4 locus of genome for for four hours to overnight at room temperature (26° C.) disruption of the PEP4 locus and integration of the expression before plating the cells on selective media. cassette into the PEP4 locus. FIG. 3 illustrates the construc 30 Generation of Cell Lines was as follows and is shown in tion of plasmid vector pGLY2233. FIG. 6. The strain yGLY24-1 (ura5A::MET1 och1A:lacZ Construction of plasmid vectors pGLY 162, p.GLY1896, bmt2A:lacZ/KIMNN2-2/mnn4L1A::lacZ/MmSLC35A3 and pGF1207t was as follows. All Trichoderma reesei C-1,2- pnol Amnn4A:lacZ met16A:lacZ), was constructed using mannosidase expression plasmid vectors were derived from methods described earlier (See for example, Nett and Gern pGFI 165, which encodes the T. reesei C-1,2-mannosidase 35 gross, Yeast 20:1279 (2003); Choi et al., Proc. Natl. Acad. Sci. catalytic domain (See published International Application USA 100:5022 (2003); Hamilton et al., Science 301:1244 No. WO2007061631) fused to S. cerevisiae C.MATpresignal (2003)). The BMT2 gene has been disclosed in Mille et al., J. peptide (ScCMPpre-signal peptide) herein expression is Biol. Chem. 283: 9724-9736 (2008) and U.S. Published under the control of the Pichia pastoris GAP promoter and Application No. 20060211085. The PNO1 gene has been wherein integration of the plasmid vectors is targeted to the 40 disclosed in U.S. Pat. No. 7,198.921 and the minn4L1 gene Pichia pastoris PRO1 locus and selection is using the Pichia (also referred to as mnn4b) has been disclosed in U.S. Pat. No. pastoris URA5 gene. A map of plasmid vector pGFI165 is 7.259,007. The minn4 refers to minn4L2 or minn4a. In the shown in FIG. 4. genotype, KIMNN2-2 is the Kluveromyces lactis GlcNAc Plasmid vector pGLY 1 162 was made by replacing the transporter and MmSLC35A3 is the Mus musculus GlcNAc GAP promoter in pGFI165 with the Pichia pastoris AOX1 45 transporter. The URA5 deletion renders the yGLY24-1 strain (Pp.AOX1) promoter. This was accomplished by isolating the auxotrophic for uracil (See U.S. Published application No. Pp.AOX1 promoter as an EcoRI (made blunt)-BglII fragment 2004/0229306) and was used to construct the humanized from pCLY2028, and inserting into pGFI 165 that was chaperone strains that follow. While the various expression digested with NotI (made blunt) and BglII. Integration of the cassettes were integrated into particular loci of the Pichia plasmid vector is to the Pichia pastoris PRO1 locus and 50 pastoris genome in the examples herein, it is understood that selection is using the Pichia pastoris URA5 gene. A map of the operation of the invention is independent of the loci used plasmid vector pGLY 1 162 is shown in FIG. 5. for integration. Loci other than those disclosed herein can be Plasmid vector pGLY1896 contains an expression cassette used for integration of the expression cassettes. Suitable inte encoding the mouse C-1.2-mannosidase catalytic domain gration sites include those enumerated in U.S. Published fused to the S. cerevisiae MNN2 membrane insertion leader 55 application No. 20070072262 and include homologs to loci peptide fusion protein (See Choi et al., Proc. Natl. Acad. Sci. known for Saccharomyces cerevisiae and other yeast or fungi. USA 100: 5022 (2003)) inserted into plasmid vectorpGFI165 StrainsyGLY702andyGLY704 were generated in order to (FIG. 5). This was accomplished by isolating the GAPp test the effectiveness of the human PDI1 expressed in Pichia ScMNN2-mouse MNSI expression cassette from pGLY1433 pastoris cells in the absence of the endogenous Pichia pas digested with XhoI (and the ends made blunt) and PmeI, and 60 toris PDI gene. Strains yCLY702 and yCLY704 (huPDI) inserting the fragment into pGFI 165 that digested with PmeI. were constructed as follows. Strain yGLY702 was generated Integration of the plasmid vector is to the Pichia pastoris by transfecting yGLY24-1 with plasmid vector pGLY642 PRO1 locus and selection is using the Pichia pastoris URA5 containing the expression cassette encoding the human PDI gene. A map of plasmid vector pGLY 1896 is shown in FIG. 4. under control of the constitutive PpGAPDH promoter. Plas Plasmid vector pGFI207t is similar to pGLY1896 except 65 mid vector pGLY642 also contained an expression cassette that the URA5 selection marker was replaced with the S. encoding the Pichia pastoris URA5, which rendered strain cerevisiae ARR3 (ScARR3) gene, which confers resistance yGLY702 prototrophic for uracil. The URA5 expression cas US 8,067,339 B2 39 40 sette was removed by counterselecting yGLY702 on 5-FOA ing the LC fusion protein under the control of the Pichia plates to produce strainyGLY704 in which, so that the Pichia pastoris AOX1 promoter and Saccharomyces cerevisiae pastoris PDI1 gene has been stably replaced by the human CYC terminator was removed from plasmid vector PDI gene and the strain is auxotrophic for uracil. pGLY2338 by digesting with BamHI and NotI and then clon StrainyCLY733 was generated by transfecting with plas ing the DNA fragment into plasmid vector pGLY2987 mid vector pGLY1162, which comprises an expression cas digested with BamH1 and Not1, thus generating the final sette that encodes the Trichoderma Reesei mannosidase expression plasmid vector pGLY2988 (FIG. 7). (TrMNS1) operably linked to the Pichia pastoris AOX1 pro Expression/integration plasmid vector pGLY3200 (map is moter (Pp.AOX1-TrMNS1), into the PRO1 locus of identical to pGLY2988 except LC and HCare anti-CD20 with yGLY704. This strain has the gene encoding the Pichiapas 10 C.-amylase signal sequences). Anti-CD20 sequences were toris PD1 replaced with the expression cassette encoding the from GenMab sequence 2C6 except Light chain (LC) frame human PDI1, has the Pp.AOX1-TrMNS1 expression cassette work sequences matched those from VKappa 3 germline. integrated into the PRO1 locus, and is a URA5 auxotroph. Heavy (HC) and LC variable sequences fused at the N-termi The Pp.AOX1 promoter allows overexpression when the cells nus to the C-amylase (from Aspergillus niger) signal peptide are grown in the presence of methanol. 15 were synthesized by GeneArtAG. The nucleotide and amino StrainyGLY762 was constructed by integrating expression acid sequences for the C-amylase signal peptide are shown in cassettes encoding TrMNS1 and mouse mannosidase IA SEQID NOS:33 and 34. Each was synthesized with unique 5' (MuMNS1A), each operably linked to the Pichia pastoris EcoR1 and 3' KpnI sites which allowed for the direct cloning GAPDH promoter in plasmid vector pGFI207t into control of variable regions into expression vectors containing the strainyGLY733 at the 5' PRO1 locus UTR in Pichia pastoris IgG1 and V. kappa constant regions. The nucleotide and genome. This strain has the gene encoding the Pichia pastoris amino acid sequences of the anti-CD20 HC are shown in SEQ PD 1 replaced with the expression cassette encoding the ID Nos:37 and 38, respectively. The nucleotide and amino human PDI1, has the PpGAPDH-TrMNS1 and PpGAPDH acid sequences of the anti-CD20 LC are shown in SEQ ID MuMNS1A expression cassettes integrated into the PRO1 Nos:35 and 36, respectively. Both HC and LC fusion proteins locus, and is a URA5 auxotroph. 25 were subcloned into IgG1 plasmid vector pGLY3184 and Strain yOLY2677 was generated by counterselecting VKappa plasmid vector pGLY2600, respectively, (each plas yGLY762 on 5-FOA plates. This strain has the gene encoding mid vector contains the Pichia pastoris TRP2 targeting the Pichia pastoris PD1 replaced with the expression cassette nucleic acid molecule and Zeocin-resistance marker) to form encoding the human PDI1, has the Pp.AOX1-TrMNS1 plasmid vectors pGLY3192 and pGLY3196, respectively. expression cassette integrated into the PRO1 locus, has the 30 The LC expression cassette encoding the LC fusion protein PpGAPH-TrMNS1 and PpGAPDH-MuMNS1 A expression under the control of the Pichia pastoris AOX1 promoter and cassettes integrated into the PRO1 locus, and is a URA5 Saccharomyces cerevisiae CYC terminator was removed prototroph. from plasmid vector pGLY3196 by digesting with BamHI Strains yCLY2696 was generated by integrating plasmid and NotI and then cloning the DNA fragment into plasmid vector pGLY2233, which encodes the human GRP94 protein, 35 vector pGLY3192 digested with BamH1 and Not1, thus gen into the PEP4 locus. This strain has the gene encoding the erating the final expression plasmid vector pGLY3200 (FIG. Pichia pastoris PD1 replaced with the expression cassette 8). encoding the human PDI1, has the Pp.AOX1-TrMNS1 Transfection of strainyGLY2696 with the above anti-Her2 expression cassette integrated into the PRO1 locus, has the or anti-CD20 antibody expression/integration plasmid vec PpGAPDH-TrMNS1 and PpGAPDH-MuMNS1A expres 40 tors was performed essentially as follows. Appropriate Pichia sion cassettes integrated into the PRO1 locus, has the human pastoris strains were grown in 50 mL YPD media (yeast GRP64 integrated into the PEP4 locus, and is a URA5 pro extract (1%), peptone (2%), dextrose (2%)) overnight to an totroph. The genealogy of this chaperone-humanized strain is OD of between about 0.2 to 6. After incubation on ice for 30 shown in FIG. 6. minutes, cells were pelleted by centrifugation at 2500-3000 45 rpm for 5 minutes. Media were removed and the cells washed EXAMPLE 2 three times with ice cold sterile 1M sorbitol before resuspend ing in 0.5 mL ice cold sterile 1 M sorbitol. Ten uL linearized Expression vectors encoding an anti-Her2 antibody and an DNA (5-20 ug) and 100 uL cell suspension was combined in anti-CD20 antibody were constructed as follows. an electroporation cuvette and incubated for 5 minutes on ice. Expression/integration plasmid vector pGLY2988 con 50 Electroporation was in a Bio-Rad GenePulser Xcell follow tains expression cassettes encoding the heavy and light chains ing the preset Pichia pastoris protocol (2 kV. 25 LF, 200 S2), of an anti-Her2 antibody. Anti-Her2 heavy (HC) and light immediately followed by the addition of 1 mL YPDS recov (LC) chains fused at the N-terminus to C-MAT presignal ery media (YPD media plus 1 M sorbitol). The transfected peptide were synthesized by GeneArtAG. The nucleotide and cells were allowed to recover for four hours to overnight at amino acid sequences for the C-amylase signal peptide are 55 room temperature (26°C.) before plating the cells on selec shown in SEQID NOS: 27 and 28. Each was synthesized with tive media. Strain yGLY2696 transfected with pGLY2988 unique 5' EcoR1 and 3' Fsel sites. The nucleotide and amino encoding the anti-HER2 antibody was designated acid sequences of the anti-Her2 HC are shown in SEQ ID yGLY4134. Strain yOLY2696 transfected with pGLY3200 Nos:29 and 30, respectively. The nucleotide and amino acid encoding the anti-CD20 antibody was designated sequences of the anti-Her2 LC are shown in SEQID Nos:31 60 yGLY3920. and 32, respectively. Both nucleic acid molecule fragments encoding the HC and LC fusion proteins were separately EXAMPLE 3 subcloned using 5' EcoR1 and 3' Fsel unique sites into an expression plasmid vector pGLY2198 (contains the Pichia This example describes the construction of plasmids com pastoris TRP2 targeting nucleic acid molecule and the Zeo 65 prising expression cassettes encoding cell Surface anchoring cin-resistance marker) to form plasmid vector pGLY2987 and proteins fused to binding moieties capable of binding an pGLY2338, respectively. The LC expression cassette encod immunoglobulin, which are Suitable for use in Pichia pas US 8,067,339 B2 41 42 toris. The plasmids comprise a nucleic acid molecule encod shown in SEQID NO:46. The nucleic acid molecule encod ing Sedlp, a cell Surface anchoring protein that inherently ing the apre-5xBD-Htag fusion protein was digested with contains an attached glycophosphotidylinositol (GPI) post EcoRI and SalI and the fragment cloned into pGLY3033, translational modification that anchors the protein in the cell which had been digested with EcoRI and SalI to remove the wall. The nucleic acid molecule encoding the Sedlp was GR2 coiled coil encoding fragment. This produced plasmid linked in frame to a nucleic acid molecule encoding an anti pGLY4136, which contains operably linked to the Pp.AOX1 body-binding moiety that is capable of binding whole, intact promoter, the nucleic acid molecule encoding the apre-5x antibodies. BD-Htag fusion protein linked in-frame to the nucleic acid Four plasmids were constructed containing antibody bind molecule encoding the SED1 fragment. The plasmid is an ing moiety/cell Surface anchor fusion protein expression cas 10 integration/expression vector that targets the plasmid to the settes. Plasmid pGLY4136 encodes the five Fc binding URA6 locus. The fusion protein expressed by this integra domains of Protein A fused to the Saccharomyces cerevisiae tion/insertion plasmid is referred to herein as the Protein SED1 (ScSED1) gene followed by the CYC terminator, all A/SED1 fusion protein. under the control of the AOX promoter (FIG. 9). Plasmid To put the Protein A/SED1 fusion protein under the control pGLY4116 encodes the Fc receptor III (FcRIII (LF)) fused to 15 of the GAPDH promoter, plasmid pGLY4136 was digested the ScSED1 gene (FIG. 10). Plasmid pGLY4137 encodes Fc with BglII and EcoRI to release the AOX1 promoter and to receptor I (FcRI) fused to the ScSED1 gene (FIG. 10) and insert the Pichia pastoris GAPDH promoter from pGLY880. plasmid pGLY4124 (FIG. 9) encodes the ZZ-domain from This produced plasmid pGLY4139. Protein A fused to the ScSED1 gene. The ZZ-domain consists Plasmid pGLY4124 comprising an expression cassette of two of the five Fc binding domains. All four plasmids encoding the Protein A ZZ domain fused to the SED1 frag contain a pUC19 E. coli origin and an arsenite resistance ment under the control of the AOX1 promoter was con marker and are integrated into the Pichia pastoris genome at structed as follows. The ZZ-domain from GeneArt plasmid the URA6 locus. 0706208 ZZHAtag was PCR amplified using the following Plasmid pGLY3033 comprising an expression cassette primers: primer alpha-amy-ProtAZZ/up: CGGAATTCac encoding a fusion protein comprising the Saccharomyces 25 gATGGTCGCTTGGTGGTCTTTGTTTCTG cerevisiae SED1 GPI anchoring protein without its endog TACGGTCTTCAGGTCGCTGCA CCTGCTTTGGCT enous signal peptide (SED1 fragment) has been described in TCTGGTGGTGTTACTCCAGCTGCTAACGCTGCTCA copending Application Ser. No. 61/067.965 filed Mar. 3, ACACG (SEQ ID NO:47) and HA-ProtAZZ-Xho1ZZ/lp: 2008. The SED1 amino acid sequence without its endogenous GCCTCGAGAGCGTAGTCTGGAACATCG signal peptide is shown in SEQ ID NO:60. A nucleic acid 30 TATGGGTAACCACCACCAGCATC (SEQID NO:48). The molecule encoding the SED1 fragment was synthesized by alpha-amy-ProtAZZ/up primer includes in-frame the coding GeneArt AG. The codons encoding the fragment had been sequence for the first 20 amino acids of the Aspergillus niger optimized for expression in Pichia pastoris. The nucleotide C.-amylase signal peptide (underlined). The primers intro sequence encoding the SED1 fragment is shown in SEQ ID duce an EcoRI site at the 5' end of the coding region and a NO:61). The Pichia pastoris URA6 locus was chosen as an 35 XhoI site at the 3' end. The nucleic acid sequence of the integrating site for the GPI anchoring protein expression cas ZZ-domain as an EcoRI/XhoI fragment is shown in SEQID sette. The URA6 gene was PCR amplified from Pichia pas NO:49. The amino acid sequence of the ZZ-domain is shown toris genomic DNA and cloned into pCR2.1 TOPO (Invitro in SEQID NO:50. The PCR conditions were one cycle of 95° gen, La Jolla, Calif.) to produce plasmid pGLY1849. The C. for 2 minutes, 20 cycles of 98° C. for 10 seconds, 65° C. for BglII and EcoR1 sites within the gene were mutated by silent 40 10 seconds, and 72° C. for 1 minute, and followed by one mutation for cloning purposes. The TRP2 targeting nucleic cycle of 72° C. for 10 minutes. acid molecule of plasmid pGLY2184 was replaced with the The PCR fragment was cloned into plasmid pCR2.1 TOPO Pichia pastoris URA6 gene from pCLY1849. In addition, the and the cloned fragment sequenced to confirm the sequence Pichia pastoris ARG1 selection marker was replaced with the encoded the Protein AZZ domain. The ZZ-domain fragment Arsenite marker cassette from plasmid pCFI8. The final plas 45 was extracted from the pCR2.1 TOPO vector by EcoRI and mid was named pCFI30t and was used to make plasmid XhoI digest and the EcoRI/XhoI fragment was cloned into pGLY3033 (FIG. 20), containing an expression cassette com plasmidpCLY3033, which had been digested with EcoRI and prising a nucleic acid molecule encoding the SED1 fragment SalI to remove the GR2 coiled coil encoding fragment. This protein fused at its amino terminus to a GR2 coiled-coil produced plasmid pGLY4124, which contains operably peptide and Aspergillus niger alpha-amylase signal peptide 50 linked to the Pp.AOX1 promoter, the nucleic acid molecule operably linked to the Pp.AOX1 promoter. The GR2 coiled encoding the Protein AZZ domain-alpha amylase signal pep coil and signal peptide encoding fragment can be removed by tide fusion protein linked in-frame to the nucleic acid mol EcoRI and SalI digestion and replaced with an antibody cap ecule encoding the SED1 fragment. The plasmid is an inte ture moiety to make a fusion protein in which the capture gration/expression vector that targets the plasmid to the moiety is fused to a cell Surface anchoring protein. 55 URA6 locus. The fusion protein expressed by this integra Plasmid pGLY4136 comprising an expression cassette tion/insertion plasmid is referred to herein as the ZZ/SED1 encoding the five Fc binding domains of protein Afused to the fusion protein. SED1 fragment under the control of the AOX1 promoter was Plasmid pGLY4116 comprising an expression cassette constructed as follows. A nucleic acid molecule fragment encoding the FcRIIIa LF receptor fused to the SED1 fragment encoding the five Fc binding domains from protein A was 60 under the control of the AOX1 promoter was constructed as synthesized by GeneArt to encode the five Fc binding follows. A nucleic acid molecule encoding the FcRIIIa LF domains fused to the Saccharomyces cerevisiae C-Mating receptor was PCR amplified from plasmid pGLY3247 Factor presignal sequence at the N-terminus and an HA and (FcRIIIa LF) as an EcoRI/SalI fragment. In plasmid 9xHIS Tag sequence at the C-terminus and to have an EcoRI pGLY3247, the FcRIIIa LF receptor is a fusion protein in 5' end and a SalI 3' end. The fragment apre-5xBD-Htag has 65 which the endogenous signal peptide had been replaced with the nucleotide sequence shown in SEQID NO:45. The apre the C-MFpre-pro. The 5' primer anneals to the sequence 5xBD-Htag fusion protein has the amino acid sequence encoding the signal peptide and the 3' primer anneals to the US 8,067,339 B2 43 44 His-tag at the end of the receptor and omits the stop codon for were each transfected with pGLY4116 (expresses FcRIII the receptor. The 5' primer was 5Ecoapp: AACGGAAT receptor/SED fusion protein), pGLY4136 (expresses Protein TCATGAGATTTCCTTCAATTTTTAC (SEQ ID NO:51) A/SED fusion protein), pGLY4124 (expresses Protein A ZZ and the 3' primer was 3HtagSal CGATGTCGACGTGATG domain/SED fusion protein), or pGLY4137 (expresses FcRI GTGATGGTGGTGATGATGATGACCACC (SEQ ID receptor/SED fusion protein). YGLY2696 was also trans NO:52). The PCR conditions were one cycle of 95° C. for 2 fected with each of the above four expression/integration minutes, 25 cycles of 95° C. for 30 seconds, 58° C. for 30 vectors. For transfection, the strains are grown in 50 mL seconds, and 72°C. for 70 seconds, and followed by one cycle of 72° C. for 10 minutes. BMGY media until the culture reached a density of about The PCR fragment encoding the receptor fusion protein OD600=2.0. The cells are washed three times with 1 M Sor was cloned into plasmid pCR2.1 TOPO and the cloned frag 10 bitol and resuspended in 1 mL 1 M sorbitol. About 1 to 2 ug ment sequenced to confirm the sequence encoded the recep oflinearized plasmid are mixed with the cells. Transfection is tor. The nucleotide sequence of the FcRIII(LF) as an EcoRI/ performed with a BioRad electroporation apparatus using the SalI fragment is shown in SEQID NO:53. The amino acid manufacturer's program specific for electroporation of sequence of the FcRIII(LF) with C. MF pre-signal sequence is nucleic acid molecules into Pichia pastoris. One mL of shown in SEQID NO:54. 15 recovery media is added to the cells, which are then plated out Plasmid pCR2.1 TOPO was digested with EcoRI and Sall on YPG (yeast extract: peptone:glycerol medium) with 50 and the EcoRI/SalI fragment encoding the receptor was g/mL arsenite. cloned into pGLY3033, which had been digested with EcoRI and SalI to remove the GR21 coiled coil encoding fragment. Cell surface labeling was as follows. Strain yCLY4134 This produced plasmid pGLY4116, which contains operably (expresses anti-Her2 antibody), strainyGLY4134 transfected linked to the Pp.AOX1 promoter, the nucleic acid molecule with pGLY4136 (expresses anti-Her2 antibody and Protein encoding the FcRIIIa LF/O-MF pre-pro signal peptide fusion A/SED1 fusion protein, and strain YGLY2696 transfected protein linked in-frame to the nucleic acid molecule encoding with pGLY4136 (expresses Protein A/SED1 fusion protein) the SED1 fragment. The plasmid is an integration/expression were grown in 600 uL BMGY (buffered minimal glycerol vector that targets the plasmid to the URA6 locus. The fusion medium-yeast extract, Invitrogen) in a 96 deep well plate or protein expressed by this integration/insertion plasmid is 25 referred to herein as the FcRIIIa fusion protein. 50 mL BMGY in a 250 mL shake flask for two days. The cells Plasmid pGLY4137 encoding the FcRI receptor fused to were collected by centrifugation and the Supernatant was the SED1 fragment was constructed as follows. A nucleic acid discarded. The cells were induced by incubation in 300 uL or molecule encoding the FcRI receptor was PCR amplified 25 mL BMMY with Pmti-3 inhibitor overnight following the from plasmid pGLY3248 as an EcoRI/SalI fragment. In plas methods taught in WO2007/061631. Pmti-3 is 3-hydroxy-4- midpCLY3248, the FcRI receptor is a fusion protein in which 30 (2-phenylethoxy)benzaldehyde: 3-(1-phenylethoxy)-4-(2- the endogenous signal peptide had been replaced with the C.-MFpre-pro. The 5' primer anneals to the sequence encod phenylethoxy)-benzaldehyde, which as been described in ing the signal peptide and the 3' primeranneals to the His-tag U.S. Pat. No. 7,105,554 and Published International Appli at the end of the receptor and omits the stop codon for the cation No. WO 2007061631. The Pmti-3 inhibitor reduces the receptor. The 5' primer was 5Ecoapp: AACGGAATTCAT 35 O-glycosylation occupancy, that is the number of total O-gly GAGATTTCCTTCAATTTTTAC (SEQ ID NO:51) and the cans on the antibody molecule. The cell further express a T. 3' primer was 3HtagSal CGATGTCGACGTGATGGT reesei alpha-1,2-mannsodase catalytic domain linked to the GATGGTGGTGATGATGATGACCACC (SEQID NO:52). The PCR conditions were one cycle of 95°C. for 2 minutes, Saccharomyces cerevisiea CMAT presignal peptide to con 25 cycles of 95°C. for 30 seconds, 58° C. for 30 seconds, and trol the chain length of those O-glycans that are on the anti 72°C. for 70 seconds, and followed by one cycle of 72°C. for 40 body molecule. 10 minutes. Induced cells were labeled with goatanti-human heavy and The PCR fragment encoding the receptor fusion protein light chain (H+L) Alexa 488 (Invitrogen, Carlsbad, Calif.) was cloned into plasmid pCR2.1 TOPO and the cloned frag conjugated antibody and viewed using fluorescence micros ment sequenced to confirm the sequence encoded the recep copy as follows. After induction, cells at density of about tor. The nucleic acid sequence of the FcRI as an EcoRI/SalI 45 0.5-1.0 OD600 were collected by centrifugation in a 1.5-mL fragment is shown in SEQ ID NO:55. The amino acid tube. The cells were rinsed twice with 1 mL PBS and 0.5 mL sequence of the FcRI with C.MF pre-signal sequence is shown in SEQID NO:56. goat anti-human IgG (H+L)-Alexa 488 (1:500 in 1% BSA in Plasmid pCR2.1 TOPO was digested with EcoRI and Sall PBS) was added. The tubes were rotated for one hour at 37° and the EcoRI/SalI fragment encoding the receptor was 50 C., centrifuged, and rinsed 3x with 1 mL PBS to remove the cloned into pGLY3033, which had been digested with EcoRI detection antibody. The cells were resuspended in about and SalI to remove the GR21 coiled coil encoding fragment. 50-100 uL of PBS and a 10 uL aliquot viewed with a fluores This produced plasmid pGLY4116, which contains operably cence microscope and photographed (FIG. 2). As expected, linked to the Pp.AOX1 promoter, the nucleic acid molecule both the anti-Her2 antibody expressing strain yCLY4134 encoding the FcRI/O.-MF pre-pro signal peptide fusion pro 55 without pGLY4136 encoding the protein A/SED1 fusion pro tein linked in-frame to the nucleic acid molecule encoding the tein and yCLY2696 with pGLY4136 encoding the Protein SED1 fragment. The plasmid is an integration/expression vector that targets the plasmid to the URA6 locus. The fusion A/SED1 fusion protein but no anti-Her2 antibody showed no protein expressed by this integration/insertion plasmid is surface labeling. The weak labeling that was visible on the referred to herein as the FcRI fusion protein. cells of yOLY2696 transfected with pGLY4136 might be due 60 to cross reaction of the goat antihuman heavy and light chain EXAMPLE 4 (H+L) Alexa 488 conjugated antibody to the expressed Pro tein A. However, as can also be seen in FIG. 11, co-expression Co-Expression of antibody and antibody binding moiety/ of the Protein A/SED1 fusion protein and the anti-Her2 anti cell Surface anchor fusion protein in Pichia pastoris was as body (strainyGLY4134 transfected with pGLY4136) did not follows. 65 result in displayed antibody on the cell surface and showed Pichia pastoris strains yCLY4134 (expresses anti-HER2 only background labeling. This result Suggested that simul antibody) and yCLY3920 (expresses anti-CD20 antibody) taneously expressing the antibody and Protein A/SED1 pro US 8,067,339 B2 45 46 tein interfered with display of the antibody on the cell surface yGLY4134 (expresses anti-Her2 antibody) and yOLY3920 or the Protein A/SED1 protein was not properly anchored to (expresses anti-CD20 antibody) were separately transfected the cell surface. with each of plasmids pGLY4116 (encodes FcRIII/SED1 fusion protein), pGLY4124 (encodes Protein A ZZ domain/ EXAMPLE 5 SED1 fusion protein), and pGLY4136 (encodes Protein A/SED1 fusion protein), were grown, induced and labeled as This example demonstrates that the Protein A/SED1 fusion in Example 4. protein is properly anchored to the cell Surface and that co The results for the ZZ-domain were similar to those for expressing the anti-Her2 antibody and Protein A/SED1 Protein Aalbeit the staining was somewhat weaker. This fusion protein at the same time interfere with capture and 10 Suggests that two Fc binding domains have a lower affinity for display of the antibody on the cell surface. the antibody compared to the intact Protein A, which has five To test whether the Protein A/SED1 fusion protein itself is displayed on the cell surface, strain yGLY2696 transfected Fc binding domains. pGLY4136 encoding the Protein A/SED1 fusion protein was Co-expression of the FcRIII/SED1 fusion protein and anti grown and induced as described in the previous example. At 15 body resulted in a lack of cell Surface staining. Strain a cell density of about 0.5-1.0 OD600, cells were collected by yGLY2696 transfected with pGLY4116 (encodes FcRIII/ centrifugation in a 1.5-mL tube and rinsed twice with 1 mL SED1 fusion protein) was grown and induced as described in PBS. Either 10 or 50 ng of anti-Her2 antibody was added Example 4 and the cells were incubated with 10 or 50 ng externally to the cells and the cells incubated for one hour. externally added anti-Her2 antibody. Contrary to the results Afterwards, the cells were washed 3x in 1 ml PBS and labeled from strains that expressed the Protein A/SED1 fusion pro with goat anti human H--L as described in the previous tein, cell Surface staining was absent while some intracellular example. The results showed that the anti-Her2 antibody was staining is observed (FIG. 14). The results suggest that while captured and displayed on the surface of the cells. This can be the FcRIII/SED1 fusion protein may be expressed in the cell, seen in FIG. 12, which shows strong cell surface staining. The it did not appear to be secreted. results confirm that the Protein A/SED1 fusion protein is 25 expressed, the expressed fusion protein is properly inserted EXAMPLE 7 into the cell Surface, and the fusion protein is able to capture and display antibodies on the cell Surface. This example demonstrates that temporal expression of the To determine whether co-expression interfered with dis Protein A/SED1 fusion protein and the antibody enables play of the antibody on the cell surface, strain yCLY2696 30 proper expression and capture of the secreted antibody on the transfected with pGLY4136 (empty strain that expresses Pro cell surface. tein A/SED1 fusion protein), strain yGLY4134 transfected The above experiments suggested that co-expression of the with pGLY4136 (strain expresses anti-Her2 antibody and antibody binding moiety/cell Surface anchor fusion protein Protein A/SED1 fusion protein), and strainyGLY3920 trans and the antibody together does not allow the anchor to be fected with pGLY4136 (strain expresses anti-CD20 antibody 35 displayed at the cell surface. In the above experiments, both and Protein A/SED1 fusion protein) were grown and induced the antibody binding moiety/cell Surface anchor fusion pro as in the previous example. Cells were incubated with 10 ng tein and antibody were expressed from nucleic acid mol externally added anti-Her2 antibody, labeled, and detected as ecules operably linked to the strong AOX inducible promoter. in the previous example. FIG. 13 illustrates strong cell surface It was hypothesized that inducing expression of the antibody labeling of the empty strain expressing only the Protein 40 binding moiety/cell Surface anchor fusion protein first, then A/SED1 fusion protein (yGLY2696 transfected with after sufficient antibody binding moiety/cell surface anchor pGLY4136), but only weak staining in the strains when the fusion protein had been made and anchored to the cell Sur Protein A/SED1 fusion protein and the antibody were co face, inhibiting expression of the antibody binding moiety/ expressed (yGLY4134 transfected with pGLY4136 and cell Surface anchor fusion protein and inducing expression of yGLY3920 transfected with pGLY4136). Cells expressing 45 the antibody, would enable the antibody that is made to be the Protein A/SED1 fusion protein were able to capture exter captured at the cell surface by the antibody binding moiety/ nally added antibody and display it while cells co-expressing cell surface anchor fusion protein. Therefore, different pro antibody and Protein A/SED1 fusion protein were unable to moters that would allow temporal expression of the nucleic capture externally added antibody nor display their own acid molecules encoding the antibody binding moiety/cell secreted antibody. 50 Surface anchor fusion protein and antibody were tested. These results suggested that the Protein A/SED1 fusion The GUT1 promoter is a promoter that is induced in cells protein is not displayed well on the cell surface in an antibody grown in the presence of glycerol and repressed when the co-expressing strain. This may be because co-expression of cells are Switched to a medium that lacks glycerol but con the Protein A/SED1 fusion protein and the antibody from the tains dextrose. PCR was used to amplify the GUT1 promoter strong AOX promoter under methanol induction may lead to 55 from genomic DNA of Pichia pastoris as BglII/EcoRI frag aggregation of the antibody—Protein A/SED1 fusion protein ment using primer 5gutBglII ATTGAGATCT complex in the ER and degradation. Alternatively, the anti ACCCAATTTAGCAGCCTGCATTCTC (SEQID NO:57) body—Protein A/SED1 fusion protein complex produced in and primer 3gutEcoRI GTCAGAATTC ATCTGTGGTA the ER may not secrete well because of its molecular weight TAGTGTGAAAAAGTAG (SEQID NO:58). The PCR frag or steric hindrance. 60 ment was then cloned into the pCR2.1 TOPO vector, and then sequenced to confirm the sequence. The GUT1 promoter EXAMPLE 6 fragment was extracted from the pCR2.1 TOPO vector by BglII/EcoRI digest and cloned into pGLY4136 digested with Other antibody binding moieties were tested for their abil BglII/EcoRI to exchange the AOX1 promoter by the GUT1 ity to display antibody on the cell surface of P. pastoris. These 65 promoter. The nucleotide sequence of the GUT1 promoter include the Fc receptor I (FcRI), the Fc receptor III (FcRIII) including the BglI and EcoRI ends is shown in SEQ ID and the Protein A ZZ-domain. Strains yCLY2696 (empty), NO:59. US 8,067,339 B2 47 48 The AOX promoter from the Protein A/SED1 fusion pro glycerol growth phase and induced in the methanol induction tein plasmid pGLY4136 was replaced either by the phase, led to no detectable cell Surface display. Likely, co PpGAPDH promoter resulting in plasmid pGLY4139 or the expression leads to a Protein A/SED1 fusion protein—anti GUT1 promoterproducing the plasmidpCLY4144 (FIG. 15). body complex in the ER, which does not secrete to the cell The PpGAPDH promoter is induced in dextrose and at about 5 Surface or is degraded. 80% of that level in glycerol, while the GUT1 promoter is Expression of the Protein A/SED1 fusion protein under the induced in glycerol and repressed in dextrose. pGLY4139 GAPDH promoter during growth in glycerol and expression was transfected into yCLY4134, expressing anti-Her2 anti of the antibody under the AOX promoter during induction in body under control of the AOX promoter. Additionally, pGLY4144 has been transfected into strain yCLY5434 10 methanol resulted in weak cell Surface display. In this case, (yGLY2696 transfected with pGLY4142), in which anti-Her2 the Protein A/SED1 fusion protein is still expressed at some expression is regulated by the GAPDH promoter. level during induction of the antibody because the GAPDH Strain yCLY4134 transfected with pGLY4136, in which promoter is not repressed completely under methanol induc expression of the Protein A/SED1 fusion protein and the tion conditions. This means that under induction conditions, anti-Her2 antibody are both regulated by the AOX promoter, 15 there might be complex formation between the Protein was grown in 600 uL BMGY (glycerol as carbon source) in a A/SED1 fusion protein and the antibody in the ER, which 96 deep well plate or 50 mL BMGY in a 250 mL shake flask then clogs the Secretory pathway leading to only a small for two days. The cells were collected by centrifugation and amount of Protein A/SED1 fusion protein at the cell surface. the supernatant was discarded. The cells were induced by Expression of the Protein A/SED1 fusion protein under the incubation overnight in 300 uL or 25 mL BMMY (methanol 20 GUT1 promoter during growth in glycerol followed by as carbon source) with PMTi inhibitor. expression of the antibody under the GAPDH promoter while Strain yCLY4134 transfected with pGLY4139, in which simultaneously repressing expression of the Protein A/SED1 expression of the Protein A/SED1 fusion protein is regulated fusion protein during induction of antibody expression with by the PpGAPDH promoter and expression of the anti-Her2 dextrose led to strong cell surface display. Thus, when the antibody regulated by the AOX promoter, was grown in 25 Protein A/SED1 fusion protein is expressed first and then BMGY (glycerol as carbon source) and induced in BMMY completely repressed during antibody induction, the Protein with PMTi inhibitor (methanol as carbon source). A/SED1 fusion protein is secreted to the cell wall where it can Strain yCLY5434 transfected with pGLY4144, in which capture the antibody when it is secreted. Although the anti expression of the Protein A/SED1 fusion protein is regulated body is expressed at some level during Protein A/SED1 by the GUT1 promoter and expression of the anti-Her2 anti- 30 body is regulated by the GAPDH promoter, was grown in fusion protein growth because the GAPDH promoter is not BMGY (glycerol as carbon source) and induced in BMDY repressed under glycerol, the level of expression of the anti with PMTi inhibitor (dextrose as carbon source). Dextrose body appears to be low enough to not interfere with the inhibits transcription from the GUT1 promoter. After induc Protein A/SED1 fusion protein secretion. tion, all three strains were labeled with goat anti human IgG 35 To demonstrate that the cell surface display of whole anti (H+L)-Alexa 488 as described in Example 1. In general, body by Protein A/SED1 fusion protein regulated under the growth can be between 1.5 days to 3 days and induction GUT1 promoter is functional for different antibodies, the between 1 to 2 days. Strains are usually grown for 2 days and anti-CD20 antibody expressing strain yCLY5757 was also then induced for another 2 days: afterwards the analysis is transfected with plasmid pGLY4144, which encodes Protein done. 40 A/SED1 fusion protein whose expression is regulated the FIG. 16 illustrates the results of cell surface staining of the GUT1 promoter. StrainyGLY5757 is strainyGLY2696 trans above strains. As was shown in Example 5, co-expression of fected with the plasmid pGLY4078. Plasmid pGLY4078 the Protein A/SED1 fusion protein and anti-Her2 antibody, encodes the heavy and light chain of the anti-CD20 antibody both under the strong AOX promoter (yGLY4134 transfected under the regulation of the GAPDH promoter. with pGLY4136) does not show any cell surface labeling. 45 Strain yOLY5757 expressing the anti-CD20 antibody Expression of the Protein A/SED1 fusion protein under the operably linked to the GAPDH promoter and transfected with GAPDH promoter during growth in glycerol and the expres pGLY4144 (encodes Protein A/SED1 fusion protein under sion of anti-Her2 antibody regulated by the AOX promoter control of the GUT1 promoter) and strain yCLY5434 during induction with methanol (yGLY4134 transfected with expressing the anti-Her2 antibody operably linked to the pGLY4139) shows some weak but visible cell surface label- 50 GAPDH promoter transfected with pGLY4144 were grown ing. In this case the Protein A/SED1 fusion protein is still expressed at Some level during induction of the antibody in glycerol for Protein A/SED1 fusion protein expression because the GAPDH promoter is not completely repressed followed by induction in dextrose for antibody expression under methanol induction conditions. However, expression of and secretion as described for FIG. 6. Strong cell surface the Protein A/SED1 fusion protein under the GUT1 promoter 55 staining was observed for both antibodies (FIG. 18). This during growth in glycerol followed by induction of the anti demonstrates that temporal regulation enables different anti Her2 antibody regulated by the GAPDH promoter during bodies and not just the anti-Her2 antibodies to be displayed on induction in dextrose (YGLY5434 transfected with the yeast Surface by an anchored antibody binding moiety. pGLY4144) showed strong cell surface labeling. In this case, FIG. 19 shows the results of FACS sorting of the samples the Protein A/SED1 fusion protein was not expressed under 60 from FIG. 8. The anti-Her2 expressing strain yCLY5757 antibody induction conditions because the GUT1 promoter is transfected with pGLY4144, the anti-CD20 expressing strain completely repressed in dextrose. yGLY5434 transfected with pGLY4144 and the empty strain FIG. 17 is a chart that illustrates the expected expression yGLY2696 transfected with pGLY4144 were grown in glyc patterns of Protein A/SED1 fusion protein and antibody under erol and then induced in dextrose. Cells were labeled with the control of different combinations of promoters. Expres- 65 goatantihuman IgG (H+L)-Alexa 488 and analyzed by FACS sion of the Protein A/SED1 fusion protein and the antibody sorting. As shown in FIG. 9, the empty strain without anti under the strong AOX promoter, which is repressed in the body expression displayed background fluorescent staining US 8,067,339 B2 49 50 while for three clones of the anti-CD20 expressing strain, the ing, which could be a false positive from a transfection that fluorescence was shifted to the right showing cell surface does not express the antibody or the anchor. These results labeling. The same was also seen for the anti-Her2 expressing demonstrate that the cells displaying whole antibodies can be strain. One clone of this strain showed no cell surface label Sorted using FACS Sorting. TABLE 1.

BRIEF DESCRIPTION OF THE SEQUENCES SEQ ID NO : De scription Sequence

1. PC R primer AGCGCTGACGCCCCCGAGGAGGAGGACCAC hP DI/UP1

PC R primer CCTTAATTAATTACAGTTCATCATGCACAGCTTTCTGATCAT hP DIALP-PacI

PC R primer ATGAATTCAGGCCATATCGGCCATTGTTTACTGTGCGCCCACAG PB248 TAG

PC R primer ATGTTTAAACGTGAGGATTACTGGTGATGAAAGAC PB249

PC R primer AGACTAGTCTATTTGGAGACATTGACGGATCCAC PB250

PC R primer ATCTCGAGAGGCCATGCAGGCCAACCACAAGATGAATCAAATTT PB25

PC R primer GGTGAGGTTGAGGTCCCAAGTGACTATCAAGGTC Pp PDI/UPi-1

PC R primer GACCTTGATAGTCACTTGGGACCTCAACCTCACC Pp PDIALPi-1

PC R primer CGCCAATGATGAGGATGCCTCTTCAAAGGTTGTG Pp PDI/UPi-2

PC R primer CACAACCTTTGAAGAGGCATCCTCATCATTGGCG Pp PDIALPi-2

PC R primer GGCGATTGCATTCGCGACTGTATC Pp PDI-5"/UP

PC R primer CCTAGAGAGCGGTGGCCAAGATG hP DI-3'ALP

PC R primer GTGGCCACACCAGGGGGCATGGAAC hP DI/UP

PC R primer CCTAGAGAGCGGTGGCCAAGATG hP DI-3'ALP

PC R primer AGCGCTGACGATGAAGTTGATGTGGATGGTACAGTAG G RP94/UP1

PC R primer GGCCGGCCTTACAATTCATCATGTTCAGCTGTAGATTC RP94/LP1

PC R primer TGAACCCATCTGTAAATAGAATGC

PC R primer GTGTCACCTAAATCGTATGTGCCCATTTACTGGAAGCTGCTAAC C

PC R primer CTCCCTATAGTGAGTCGTATTCATCATTGTACTTTGGTATATTG G

PC R primer TATTTGTACC TOCGTCCTGTTTGC

21 PC R primer CACATACGATTTAGGTGACAC PR29

22 PC R primer AATACGACTCACTATAGGGAG PR32

23 PC R primer TGCTCTCCGCGTGCAATAGAAACT PMT4 - KO1

US 8,067,339 B2 63 TABLE 1 - continued

BRIEF DESCRIPTION OF THE SEQUENCES SEQ ID NO: Description Sequence

GTTGCGGAGCGTCAAAAGTTTGCCCGATCTGATAGCTTGCAAGA TGCCACCGCTTATCCAACGCACTTCAGAGAGCTTGCCGTAGAAA GAACGTTTTCCTCGTAGTATTCCAGCACTTCATGGTGAAGTCGC TATTTCACCGAAGGGGGGGTATTAAGGTTGCGCACCCCCTCCCC ACACCCCAGAATCGTTTATTGGCTGGGTTCAATGGCGTTTGAGT TAGCACATTTTTTCCTTAAACACCCTCCAAACACGGATAAAAAT GCATGTGCATCCTGAAACTGGTAGAGATGCGTACTCCGTGCTCC GATAATAACAGTGGTGTTGGGGTTGCTGTTAGCTCACGCACTCC GTTTTTTTTTCAACCAGCAAAATTCGATGGGGAGAAACTTGGGG TACTTTGCCGACTCCTCCACCATACTGGTATATAAATAATACTC GCCCACTTTTCGTTTGCTGCTTTTATATTTCAAGGACTGAAAAA AGACTCTTCTTCTACTTTTTCACACTATACCACAGATGAATTC

60 S. cerevisiae VDOFSNSTSASSTDWTSSSSISTSSGSVTITSSEAPESDNGTST SED1 (without AAPTETSTEAPTTAIPTNGTSTEAPTTAIPTNGTSTEAPTDTTT endogenous EAPTTALPTNGTSTEAPTDTTTEAPTTGLPTNGTTSAFPPTTSL leader sequence PPSNTTTTPPYNPSTDYTTDYTCCTEYTTYCPEPTTFTTNGKTY TVTEPTTLTITDCPCTIEKPTTTSTTEYTV VTEYTTYCPEPTTF TTNGKTYTVTEPTTLTITDCPCTEKSEAPESSWPWTESKGTTT KETGVTTKOTTANPSLTWSTVWPVSSSASSHSVWINSNGANVVV PGALGLAGWAMIFL

61 S. cerevisiae GTCGACCAATTCTCTAACTCTACTTCCGCTTCCTCTACTGACGT SED1, DNA TACTTCCTCCTCC TO TATTTCTACTTCCTCCGGTTCCGT TACTA sequence TTACTTCCTCTGAGGCTCCAGAATCTGACAACGGTACTTCTACT GCTGCTCCAACTGAAACTTCTACTGAGGCTCCTACTACTGCTAT TCCAACTAACGGAACTTCCACAGAGGCTCCAACAACAGCTATCC CTACAAACGGTACATCCACTGAAGCTCCTACTGACACTACTACA GAAGCTCCAACTACTGCTTTGCCTACTAATGGTACATCAACAGA GGCTCCTACAGATACAACAACTGAAGCTCCAACAACTGGATTGC CAACAAACGGTACTACTTCTGCTTTCCCACCAACTACTTCCTTG CCACCATCCAACACTACTACTACTCCACCATACAACCCATCCAC TGACTACACTACTGACTACACAGTTGTTACTGAGTACACTACTT ACTGTCCAGAGCCAACTACTTTCACAACAAACGGAAAGACTTAC ACTGTTACTGAGCCTACTACTTTGACTATCACTGACTGTCCATG TACTATCGAGAAGCCAACTACTACTTCCACTACAGAGTATACTG TTGTTACAGAATACACAACATATTGTCCTGAGCCAACAACATTC ACTACTAATGGAAAAACATACACAGTTACAGAACCAACTACATT GACAATTACAGATTGTCCTTGTACAATTGAGAAGTCCGAGGCTC CTGAATCTTCTGTTCCAGTTACTGAATCCAAGGGTACTACTACT AAAGAAACTGGTGTTACTACTAAGCAGACTACTGCTAACCCATC CTTGACTGTTTCCACTGTTGTTCCAGTTTCTTCCTCTGCTTCTT CCCACTCCGTTGTTATCAACTCCAACGGTGCTAACGTTGTTGTT CCTGGTGCTTTGGGATTGGCTGGTGTTGCTATGTTGTTCTTGTA A.

While the present invention is described herein with refer- 45 ence to illustrated embodiments, it should be understood that the invention is not limited hereto. Those having ordinary skill in the art and access to the teachings herein will recognize additional modifications and embodiments within the scope thereof. Therefore, the present invention is limited only by the claims attached herein.

SEQUENCE LISTING

<16O is NUMBER OF SEO ID NOS: 61

<210s, SEQ ID NO 1 &211s LENGTH: 30 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer hPDI/UP1

<4 OOs, SEQUENCE: 1 agcgctgacg CCC ccgagga ggaggaccac 3 O US 8,067,339 B2 65 66 - Continued

<210s, SEQ ID NO 2 &211s LENGTH: 42 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer hPDI/LP-PacI <4 OOs, SEQUENCE: 2 ccittaattaa ttacagttca totatgcacag ctittctgatc at 42

<210s, SEQ ID NO 3 &211s LENGTH: 47 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer PB248 <4 OOs, SEQUENCE: 3 atgaatticag gccatat cqg C cattgttta Ctgtgcgc.cc acagtag 47

<210s, SEQ ID NO 4 &211s LENGTH: 35 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer PB249 <4 OOs, SEQUENCE: 4 atgtttaaac gtgaggatta Ctggtgatga aagac 35

<210s, SEQ ID NO 5 &211s LENGTH: 34 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer PB250 <4 OOs, SEQUENCE: 5 agacitagt ct atttggagac attgacggat C cac 34

<210s, SEQ ID NO 6 &211s LENGTH: 46 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer PB251 <4 OOs, SEQUENCE: 6 atct cagag gcc atgcagg C calaccacaa gatgaat caa attittg 46

<210s, SEQ ID NO 7 &211s LENGTH: 34 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer PpPDI/UPi-1 <4 OO > SEQUENCE: 7 ggtgaggttg aggtoccaag tact at Caa ggt C 34

<210s, SEQ ID NO 8 &211s LENGTH: 34 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer PpPDI/LPi-1 US 8,067,339 B2 67 68 - Continued

<4 OOs, SEQUENCE: 8 gaccttgata gtcacttggg acct caac ct cacc 34

<210s, SEQ ID NO 9 &211s LENGTH: 34 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer PpPDI/UPi-2

<4 OOs, SEQUENCE: 9 cgc.caatgat gaggatgcct Cttcaaaggt ttg 34

<210s, SEQ ID NO 10 &211s LENGTH: 34 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer PpPDI/LPi-2

<4 OOs, SEQUENCE: 10 caca acctitt gaagaggcat cot catcatt gg.cg 34

<210s, SEQ ID NO 11 &211s LENGTH: 24 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer PpPDI-5"/UP

<4 OOs, SEQUENCE: 11 ggcgattgca ttcgcgactg. t at C 24

<210s, SEQ ID NO 12 &211s LENGTH: 23 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer hPDI-3"/LP <4 OOs, SEQUENCE: 12

Cctagaga.gc ggtggccaag atg 23

<210s, SEQ ID NO 13 &211s LENGTH: 25 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer hPDI/UP <4 OOs, SEQUENCE: 13 gtggcc acac Caggggggat ggaac 25

<210s, SEQ ID NO 14 &211s LENGTH: 23 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer hPDI-3"/LP <4 OOs, SEQUENCE: 14

Cctagaga.gc ggtggccaag atg 23

<210s, SEQ ID NO 15 &211s LENGTH: 37 US 8,067,339 B2 69 70 - Continued

&212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer hoRP94/UP1 <4 OOs, SEQUENCE: 15 agcgctgacg atgaagttga tigtggatggit acagtag 37

<210s, SEQ ID NO 16 &211s LENGTH: 38 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer hoRP94/LP1 <4 OOs, SEQUENCE: 16 gg.ccggcctt acaatt catc atgttcagct gtagattic 38

<210s, SEQ ID NO 17 &211s LENGTH: 24 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer PMT1-KO1 <4 OOs, SEQUENCE: 17 tgaacc catc totaaataga atgc 24

<210s, SEQ ID NO 18 &211s LENGTH: 45 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer PMT1-KO2 <4 OOs, SEQUENCE: 18 gtgtcaccita aatcg tatgt gcc catttac toggaagctgc taacc 45

<210s, SEQ ID NO 19 &211s LENGTH: 45 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer PMT1-KO3 <4 OOs, SEQUENCE: 19 citcc ctatag tdagt.cgitat t catcattgt actittggitat attgg 45

<210s, SEQ ID NO 2 O &211s LENGTH: 24 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer PMT1-KO4 <4 OOs, SEQUENCE: 2O tatttgtacc togcgt.cctgt ttgc 24

<210s, SEQ ID NO 21 &211s LENGTH: 21 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer PR29 <4 OOs, SEQUENCE: 21 cacatacgat ttaggtgaca c 21 US 8,067,339 B2 71 72 - Continued

<210s, SEQ ID NO 22 &211s LENGTH: 21 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer PR32 <4 OOs, SEQUENCE: 22 aatacgactic actataggga g 21

<210s, SEQ ID NO 23 &211s LENGTH: 24 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer PMT4 - KO1 <4 OOs, SEQUENCE: 23 tgct ct cogc gtgcaataga aact 24

<210s, SEQ ID NO 24 &211s LENGTH: 45 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer PMT4 - KO2 <4 OOs, SEQUENCE: 24 citcc ctatag tdagt.cgitat t cacagtgta ccatctitt catct co 45

<210s, SEQ ID NO 25 &211s LENGTH: 45 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer PMT4 - KO3 <4 OOs, SEQUENCE: 25 gtgtcaccita aatcg tatgt gaacctaact ctaattic titc aaag.c 45

<210s, SEQ ID NO 26 &211s LENGTH: 24 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: PCR primer PMT4 - KO4 <4 OOs, SEQUENCE: 26 actagggitat ataatticcca aggt 24

<210s, SEQ ID NO 27 &211s LENGTH: 57 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: Pre-pro alpha-mating factor signal peptide (ScaMTprepro) (DNA)

<4 OOs, SEQUENCE: 27 atgagattico catccatctt cactgctgtt ttgttcgctg. citt cittctgc tittggct

<210s, SEQ ID NO 28 &211s LENGTH: 19 212. TYPE: PRT <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: US 8,067,339 B2 73 - Continued <223> OTHER INFORMATION: Pre-pro alpha-mating factor signal peptide (protein)

<4 OOs, SEQUENCE: 28 Met Arg Phe Pro Ser Ile Phe Thr Ala Val Lieu Phe Ala Ala Ser Ser 1. 5 1O 15

Ala Lieu Ala

<210s, SEQ ID NO 29 &211s LENGTH: 1353 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: Anti-Her2 Heavy chain (VH + IgG1 constant region)

<4 OOs, SEQUENCE: 29 gaggttcagt tttgaatc tigaggagga ttggttcaac Ctggtggitt C tittgagattg 6 O t cct gtgctg citt coggittt caa.catcaag gacacttaca tocactgggit tag acaa.gct 12 O cCaggaaagg gattggagtg ggttgctaga atctacccala Ctaacggitta Cacaagatac 18O gctgactic cq tta agggaag attcactato tctgctgaca citt coaagaa cactgcttac 24 O ttgcagatga act cottgag agctgaggat actgctgttt act actgtt C Cagatggggt 3OO ggtgatggitt t ctacgct at ggact actgg ggt Caaggaa ctittggittac ttitt cotcC 360 gcttctacta agggaccatc tdttitt coca ttggctic cat cittctaagtic tact tcc.ggit 42O gg tactgctg ctittgggatgtttggittaaa gac tact tcc cagagc.cagt tactgtttct 48O tggaactic cq gtgctittgac ttctggtgtt cacactitt.cc cagctgttitt gcaatct tcc 54 O ggtttgtact citttgtcc to cqttgttact gttccatcct citt cottggg tact cagact 6OO tacatctgta acgittalacca caa.gc.catcc aacactaagg ttgacaagaa gottgagc.ca 660 aagt cctgtg acaagactica tacttgtc.ca C catgtc.cag Ctccagaatt gttgggtggit 72 O cct tcc.gttt ttttgttc.cc accaaag.cca aagga cactt tdatgat ct c cagaact coa 78O gaggittacat gtgttgttgt tacgtttct cacgaggacc cagaggittaa gttcaactgg 84 O tacgttgacg gtgttgaagt t cacaacgct aagactalagc caa.gagagga gcagtacaac 9 OO tccact taca gagttgtttc. c9ttittgact gttittgcacc aggattggitt gaacggaaag 96.O gagtacaagt gtaaggtttc caacaaggct ttgc.cagctic caatcgaaaa gacitat citcc 1 O2O aaggctaagg gtcaac Caag agagccacag gtttacactt to Caccat C cagagatgag 108 O ttgactaaga accaggtttc cittgacttgt ttggittaagg gattic taccc atc.cgacatt 114 O gctgttgaat gggagt ctaa C9gtcaiacca gagaacaact acaagactac toccacctgtt 12 OO ttgg actctg acggttcc tt tttcttgtact coaagttga citgttgacaa gtccagatgg 126 O caac agggta acgttittct c ctdtt cogitt atgcatgagg ctittgcacaa ccact acact 132O caaaagt cct tdt citttgtc. ccctggtaag taa 1353

<210s, SEQ ID NO 3 O &211s LENGTH: 450 212. TYPE: PRT <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: Anti-Her2 Heavy chain (VH + IgG1 constant region)

<4 OOs, SEQUENCE: 30 Glu Val Glin Lieu Val Glu Ser Gly Gly Gly Lieu Val Glin Pro Gly Gly 1. 5 1O 15 US 8,067,339 B2 75 76 - Continued

Ser Luell Arg Luell Ser Ala Ala Ser Gly Phe Asn Ile Lys Asp Thir 2O 25 3O

Ile His Trp Wall Arg Glin Ala Pro Gly Lys Gly Lell Glu Trp Wall 35 4 O 45

Ala Arg Ile Pro Thir Asn Gly Tyr Thir Arg Tyr Ala Asp Ser Wall SO 55 6 O

Lys Gly Arg Phe Thir Ile Ser Ala Asp Thir Ser Asn Thir Ala Tyr 65 70

Lell Glin Met Asn Ser Lell Arg Ala Glu Asp Thir Ala Wall Tyr Cys 85 90 95

Ser Arg Trp Gly Gly Asp Gly Phe Tyr Ala Met Asp Trp Gly Glin 105 11 O

Gly Thir Luell Wall Thir Wall Ser Ser Ala Ser Thir Gly Pro Ser Wall 115 12 O 125

Phe Pro Luell Ala Pro Ser Ser Ser Thir Ser Gly Gly Thir Ala Ala 13 O 135 14 O

Lell Gly Luell Wall Lys Asp Phe Pro Glu Pro Wall Thir Wall Ser 145 150 155 160

Trp Asn Ser Gly Ala Lell Thir Ser Gly Wall His Thir Phe Pro Ala Wall 1.65 17O

Lell Glin Ser Ser Gly Lell Tyr Ser Luell Ser Ser Wall Wall Thir Wall Pro 18O 185 19 O

Ser Ser Ser Luell Gly Thir Glin Thir Ile Asn Wall Asn His 195 2OO

Pro Ser Asn Thir Wall Asp Wall Glu Pro Ser Asp 21 O 215 22O

Lys Thir His Thir Pro Pro Pro Ala Pro Glu Lell Luell Gly Gly 225 23 O 235 24 O

Pro Ser Wall Phe Lell Phe Pro Pro Pro Asp Thir Luell Met Ile 245 250 255

Ser Arg Thir Pro Glu Wall Thir Wall Wall Wall Asp Wall Ser His Glu 26 O 265 27 O

Asp Pro Glu Wall Lys Phe Asn Trp Wall Asp Gly Wall Glu Wall His 285

Asn Ala Thir Pro Arg Glu Glu Glin Tyr Asn Ser Thir 29 O 295 3 OO

Wall Wall Ser Wall Lell Thir Wall Luell His Glin Asp Trp Lell Asn Gly Lys 3. OS 310 315

Glu Lys Wall Ser Asn Ala Luell Pro Ala Pro Ile Glu 3.25 330 335

Thir Ile Ser Ala Gly Glin Pro Arg Glu Pro Glin Wall Tyr 34 O 345 35. O

Thir Luell Pro Pro Ser Arg Asp Glu Luell Thir Asn Glin Wall Ser Luell 355 360 365

Thir Cys Luell Wall Lys Gly Phe Pro Ser Asp Ile Ala Wall Glu Trp 37 O 375

Glu Ser Asn Gly Glin Pro Glu Asn Asn Tyr Lys Thir Thir Pro Pro Wall 385 390 395 4 OO

Lell Asp Ser Asp Gly Ser Phe Phe Luell Tyr Ser Lell Thir Wall Asp 4 OS 415

Ser Arg Trp Glin Glin Gly Asn Wall Phe Ser Ser Wall Met His 42O 425 43 O

Glu Ala Luell His Asn His Thir Glin Ser Lell Ser Luell Ser Pro US 8,067,339 B2 77 - Continued

435 44 O 445 Gly Lys 450

<210s, SEQ ID NO 31 &211s LENGTH: 645 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: Anti-Her2 light chain (VL + Kappa constant region)

<4 OOs, SEQUENCE: 31 gacatccaaa tact caatc cc catcttct ttgttctgctt cogttggtga cagagttact 6 O at cacttgta gagct tcc.ca ggacgittaat actgctgttg cittggitatica acagaa.gc.ca 12 O ggaaaggctic caaagttgtt gat ct actico gottcct tct td tactctgg togttc catcc 18O agattct citg gttccagatc cqg tactgac titcactittga citat citcctic cittgcaacca 24 O gaagattt cq c tact tacta citgtcagoag cactacacta citccaccaac titt cqgacag 3OO gg tactaagg ttgagat caa gagaactgtt gctgctic cat cogttitt cat titt cocacca 360 tcc.gacgaac agttgaagtic togg tacagct tcc.gttgttt gtttgttgaa caactitctac 42O cCaagagagg Ctalaggttca gtggalaggtt gacaacgctt to aatc.cgg taact cocaia 48O gaatcc.gtta citgagcaaga citctaaggac tocactitact cottgtc.ctic cactittgact 54 O ttgtcCaagg Ctgattacga galagcacaag gtttacgctt gtgaggittac acatcagggit 6OO ttgtcc tocc cagttact aa gtCct tcaac agaggagagt gttaa 645

<210s, SEQ ID NO 32 &211s LENGTH: 213 212. TYPE: PRT <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: Anti-Her2 light chain (VL + Kappa constant region)

<4 OOs, SEQUENCE: 32 Asp Ile Gln Met Thr Glin Ser Pro Ser Ser Leu Ser Ala Ser Val Gly 1. 5 1O 15 Asp Arg Val Thir Ile Thr Cys Arg Ala Ser Glin Asp Wall Asn. Thir Ala 2O 25 3O Val Ala Trp Tyr Glin Gln Llys Pro Gly Lys Ala Pro Llys Lieu. Lieu. Ile 35 4 O 45 Tyr Ser Ala Ser Phe Lieu. Tyr Ser Gly Val Pro Ser Arg Phe Ser Gly SO 55 6 O Ser Arg Ser Gly Thr Asp Phe Thr Lieu. Thir Ile Ser Ser Leu Gln Pro 65 70 7s 8O Glu Asp Phe Ala Thr Tyr Tyr Cys Glin Glin His Tyr Thr Thr Pro Pro 85 90 95 Thr Phe Gly Glin Gly Thr Lys Val Glu Ile Lys Arg Thr Val Ala Ala 1OO 105 11 O Pro Ser Val Phe Ile Phe Pro Pro Ser Asp Glu Gln Leu Lys Ser Gly 115 12 O 125 Thir Ala Ser Val Val Cys Lieu. Asn. Asn. Phe Tyr Pro Arg Glu Ala Lys 13 O 135 14 O Val Glin Trp Llys Val Asp Asn Ala Lieu. Glin Ser Gly Asn. Ser Glin Glu 145 150 155 160

Ser Val Thr Glu Glin Asp Ser Lys Asp Ser Thr Tyr Ser Leu Ser Ser US 8,067,339 B2 79 80 - Continued

1.65 17O 17s Thir Lieu. Thir Lieu. Ser Lys Ala Asp Tyr Glu Lys His Llys Val Tyr Ala 18O 185 19 O Cys Glu Val Thr His Glin Gly Lieu Ser Ser Pro Val Thr Lys Ser Phe 195 2OO 2O5 Asn Arg Gly Glu. Cys 21 O

<210s, SEQ ID NO 33 &211s LENGTH: 60 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: Alpha amylase signal peptide (from Aspergillus niger -amylase)

<4 OOs, SEQUENCE: 33 atggttgctt ggtggit cott gttcttgtac ggattgcaag ttgctgct CC agctttggct 6 O

<210s, SEQ ID NO 34 &211s LENGTH: 2O 212. TYPE: PRT <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: Alpha amylase signal peptide (from Aspergillus niger -amylase)

<4 OOs, SEQUENCE: 34 Met Val Ala Trp Trp Ser Lieu. Phe Lieu. Tyr Gly Lieu. Glin Val Ala Ala 1. 5 1O 15

Pro Ala Lieu Ala 2O

<210s, SEQ ID NO 35 &211s LENGTH: 397 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: Anti-CD2O Light chain Variable Region <4 OOs, SEQUENCE: 35 gagat.cgttt tdacacagtic cccagctact ttgtc.tttgt ccc.caggtga aagagctaca 6 O ttgtcc tigta gagct tcc.ca atctgtttico toc tacttgg cittggitatica acaaaagcca 12 O ggacaggctic Caagattgtt gatctacgac gct tccalata gagct actgg tat cocagot 18O agattct citg gttctggttc cqg tactgac titcactittga citat citcttic cittggaacca 24 O gaggactt.cg Ctgttt acta Ctgtcagoag agat.ccalatt ggc cattgac titt.cggtggit 3OO gg tactaagg ttgagat caa gcgtacggitt gctgctic citt cogttitt cat titt cocacca 360 tcc.gacgaac aattgaagtic togg tacccaa titcgc cc 397

<210s, SEQ ID NO 36 &211s LENGTH: 132 212. TYPE: PRT <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: Anti-CD2O Light chain Variable Region <4 OOs, SEQUENCE: 36 Glu Ile Val Lieu. Thr Glin Ser Pro Ala Thr Lieu Ser Leu Ser Pro Gly 1. 5 1O 15

Glu Arg Ala Thr Lieu. Ser Cys Arg Ala Ser Glin Ser Val Ser Ser Tyr 2O 25 3O US 8,067,339 B2 81 82 - Continued

Lieu Ala Trp Tyr Glin Glin Llys Pro Gly Glin Ala Pro Arg Lieu. Lieu. Ile 35 4 O 45

Tyr Asp Ala Ser Asn Arg Ala Thr Gly Ile Pro Ala Arg Phe Ser Gly SO 55 6 O

Ser Gly Ser Gly Thr Asp Phe Thr Lieu Thir Ile Ser Ser Luell Glu Pro 65 70 7s 8O

Glu Asp Phe Ala Val Tyr Tyr Cys Glin Glin Arg Ser Asn Trp Pro Leu 85 90 95

Thr Phe Gly Gly Gly Thr Lys Val Glu Ile Llys Arg Thir Wall Ala Ala 1OO 105 11 O

Pro Ser Wall Phe Ile Phe Pro Pro Ser Asp Glu Glin Lell Lys Ser Gly 115 12 O 125

Thir Glin Phe Ala 13 O

<210s, SEQ ID NO 37 &211s LENGTH: 445 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: OTHER INFORMATION: Anti-CD2O Heavy chain Variable Region <4 OO > SEQUENCE: 37 gctgttcagc tiggttgaatc tigtggtgga ttggttcaac Ctggtagat C Cttgagattg 6 O

citt coggittt tactitt.cggit gactacacta tgcactgggt tagacaa.gct 12 O cCaggaaagg gattggaatg ggttt CC99t atttcttgga acticcggttc cattggttac 18O gctgatt.ccg tta agggaag attcactato tccagaga.ca acgctaagaa ctic cttgtac 24 O ttgcagatga act cottgag agctgaggat actgctttgt act actgtac taaggacaac 3OO caatacggitt ctggttccac ttacggattg ggagtttggg gacagggaac tittggittact 360 gtct cagtg cittctact aa goggaccatcc gtttitt.ccat tggct coat c ctictaagtict actitccggtg gtacccaatt cqcco 445

SEQ ID NO 38 LENGTH: 148 TYPE : PRT ORGANISM: Artificial Sequence FEATURE: OTHER INFORMATION: Anti-CD2O Heavy chain Variable Region SEQUENCE: 38

Ala Wall Glin Lieu Wall Glu Ser Gly Gly Gly Lieu. Wall Glin Pro Gly Arg 1. 5 1O 15

Ser Luell Arg Luell Ser Cys Ala Ala Ser Gly Phe Thir Phe Gly Asp Tyr 2O 25

Thir Met His Trp Val Arg Glin Ala Pro Gly Lys Gly Lell Glu Trp Val 35 4 O 45

Ser Gly Ile Ser Trp Asn Ser Gly Ser Ile Gly Tyr Ala Asp Ser Wall SO 55 6 O

Lys Gly Arg Phe Thir Ile Ser Arg Asp Asn Ala Asn Ser Leu Tyr 65 70 8O

Lell Glin Met Asn. Ser Lieu. Arg Ala Glu Asp Thr Ala Lell Tyr Tyr Cys 85 90 95

Thir Asp ASn Glin Tyr Gly Ser Gly Ser Thir Gly Luell Gly Val 1OO 105 11 O

Trp Gly Glin Gly Thr Leul Wall. Thir Wall Ser Ser Ala Ser Thir

US 8,067,339 B2 85 86 - Continued

Phe Ala Glu Ala Lell Ala Ala His Lys Tyr Pro Pro Wall Phe His 2O 25

Ala Pro Trp Gly His Lys Ala Luell Ala Pro Glu Tyr Ala 35 4 O 45

Ala Ala Gly Lell Ala Glu Gly Ser Glu Ile Arg Luell Ala SO 55 6 O

Wall Asp Ala Thir Glu Ser Asp Luell Ala Glin Glin Tyr Wall Arg 65

Gly Tyr Pro Thir Ile Phe Phe Arg Asn Gly Asp Thir Ser Pro 85 90 95

Glu Tyr Thir Ala Gly Arg Glu Ala Asp Asp Ile Wall Trp Luell 105

Arg Thir Gly Pro Ala Ala Thir Thir Luell Pro Asp Ala Ala 115 12 O 125

Ala Glu Ser Luell Wall Glu Ser Ser Glu Wall Ala Wall Ile Phe Phe 13 O 135 14 O

Lys Asp Wall Glu Ser Asp Ser Ala Glin Phe Lell Glin Ala Glu 145 150 155 160

Ala Ile Asp Asp Ile Pro Phe Gly Ile Thir Ser Asn Ser Asp Wall Phe 1.65 17s

Ser Glin Lell Asp Asp Gly Wall Wall Lell Phe Lys Phe 18O 185 19 O

Asp Glu Gly Arg Asn Asn Phe Glu Gly Glu Wall Thir Lys Glu Asn Luell 195

Lell Asp Phe Ile His Asn Glin Luell Pro Luell Wall Ile Glu Phe Thir 21 O 215 22O

Glu Glin Thir Ala Pro Lys Ile Phe Gly Gly Glu Ile Thir His Ile 225 23 O 235 24 O

Lell Luell Phe Luell Pro Ser Wall Ser Asp Asp Gly Lel Ser 245 250 255

Asn Phe Thir Ala Ala Glu Ser Phe Gly Ile Luell Phe Ile 26 O 265 27 O

Phe Ile Asp Ser Asp His Thir Asp Asn Glin Arg Ile Lell Glu Phe Phe 27s 285

Gly Luell Glu Glu Cys Pro Ala Wall Arg Lell Ile Thir Lel Glu 29 O 295 3 OO

Glu Glu Met Thir Tyr Pro Glu Ser Glu Glu Lell Thir Ala Glu 3. OS 310 315

Arg Ile Thir Glu Phe His Arg Phe Luell Glu Gly Ile Lys Pro 3.25 330 335

His Luell Met Ser Glin Glu Lell Pro Glu Asp Trp Asp Glin Pro Wall 34 O 345 35. O

Wall Luell Wall Gly Asn Phe Glu Asp Wall Ala Phe Asp Glu 355 360 365

Asn Wall Phe Wall Glu Phe Ala Pro Trp Cys Gly His 37 O 375

Glin Luell Ala Pro Ile Trp Asp Luell Gly Glu Thir Asp His 385 390 395 4 OO

Glu Asn Ile Wall Ile Ala Met Asp Ser Thir Ala Asn Glu Wall Glu 4 OS 415

Ala Wall Wall His Gly Phe Pro Thir Luell Gly Phe Phe Pro Ala Ser 42O 425 43 O

Ala Asp Arg Thir Wall Ile Asp Asn Gly Glu Arg Thir Luell Asp Gly

US 8,067,339 B2 89 90 - Continued

1O 15

Lell Thir Luell Ala Glin Ala Ser Asp Glin Glu Ala Ile Ala Pro Glu Asp 25 3O

Ser His Wall Wall Lys Lell Thir Glu Ala Thir Phe Glu Ser Phe Ile Thir 35 4 O 45

Ser Asn Pro His Wall Lell Ala Glu Phe Phe Ala Pro Trp Gly His SO 55 6 O

Cys Luell Gly Pro Glu Luell Wall Ser Ala Ala Glu Ile Luell Lys 65 70

Asp Asn Glu Glin Wall Ile Ala Glin Ile Asp Thir Glu Glu 85 90 95

Glu Luell Glin Gly Tyr Glu Ile Lys Gly Tyr Pro Thir Luell Wall 105 11 O

Phe His Gly Glu Wall Glu Wall Pro Ser Asp Glin Gly Glin Arg Glin 115 12 O 125

Ser Glin Ser Ile Wall Ser Tyr Met Luell Glin Ser Lell Pro Pro Wall 13 O 135 14 O

Ser Glu Ile Asn Ala Thir Asp Luell Asp Asp Thir Ile Ala Glu Ala 145 150 155 160

Glu Pro Wall Ile Wall Glin Wall Luell Pro Glu Asp Ala Ser Asn Luell 1.65 17O 17s

Glu Ser Asn Thir Thir Phe Gly Wall Ala Gly Thir Lell Arg Glu 18O 185 19 O

Phe Thir Phe Wall Ser Thir Ser Thir Asp Ala Lys Thir 195

Ser Asp Ser Thir Pro Ala Tyr Luell Luell Wall Arg Pro Gly Glu Glu Pro 21 O 215 22O

Ser Wall Ser Gly Glu Glu Luell Asp Glu Thir His Lell Wall His Trp 225 23 O 235 24 O

Ile Asp Ile Glu Ser Pro Luell Phe Gly Asp Ile Asp Gly Ser Thir 245 250 255

Phe Ser Tyr Ala Glu Ala Asn Ile Pro Luell Ala Tyr Phe Tyr 26 O 265 27 O

Glu Asn Glu Glu Glin Arg Ala Ala Ala Ala Asp Ile Ile Pro Phe 285

Ala Lys Glu Glin Arg Gly Lys Ile Asn Phe Wall Gly Lell Asp Ala Wall 29 O 295 3 OO

Lys Phe Gly His Ala Asn Luell Asn Met Asp Glu Glu Luell 3. OS 310 315

Pro Luell Phe Wall Ile His Asp Luell Wall Ser ASn Phe Gly Wall 3.25 330 335

Pro Glin Asp Glin Glu Lell Thir Asn Lys Asp Wall Thir Glu Luell Ile Glu 34 O 345 35. O

Phe Ile Ala Gly Glu Ala Glu Pro Ile Wall Ser Glu Pro Ile 355 360 365

Pro Glu Ile Glin Glu Glu Lys Wall Phe Luell Wall Gly Ala His 37 O 375 38O

Asp Glu Wall Wall Phe Asp Glu Ser Asp Wall Lell Wall Tyr 385 390 395 4 OO

Ala Pro Trp Gly His Arg Met Ala Pro Ala Glu Glu 4 OS 41O 415

Lell Ala Thir Luell Tyr Ala Asn Asp Glu Asp Ala Ser Ser Lys Wall Wall 42O 425 43 O US 8,067, 339 B2 91 92 - Contin lued Ile Ala Lys Lieu. Asp His Thr Lieu. Asn Asp Val Asp Asn. Wall Asp Ile 435 44 O 445

Gln Gly Tyr Pro Thr Lieu. Ile Leu Tyr Pro Ala Gly Asp Llys Ser Asn 450 45.5 460

Pro Glin Lieu. Tyr Asp Gly Ser Arg Asp Lieu. Glu Ser Lieu Ala Glu Phe 465 470 47s

Val Lys Glu Arg Gly. Thir His Lys Val Asp Ala Lieu Ala Lieu. Arg Pro 485 490 495

Val Glu Glu Glu Lys Glu Ala Glu Glu Glu Ala Glu Ser Glu Ala Asp SOO 505 51O Ala His Asp Glu Lieu. 515

<210s, SEQ ID NO 43 &211s LENGTH: 2349 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223 is OTHER INFORMATION: Part of human GRP94 Gene

<4 OOs, SEQUENCE: 43 gatgatgaag ttgacgttga cgg tactgtt gaagaggact tgggaaagtic tagagagggit 6 O tccagaactg acgacgaagt tgttcagaga gaggaagagg ctatt cagtt ggacggattg 12 O aacgct tccc aaatcagaga gttgagaga.g aagtc.cgaga agttcgctitt cCaagctgag 18O gtta acagaa tdatgaaatt gattatcaac tcc ttgtaca agaacaaaga gattittcttg 24 O agagagttga t ct ctaacgc ttctgacgct ttggaca aga t cagattgat Ctccttgact 3OO gacgaaaacg Ctttgtc.cgg taacgaagag ttgactgtta agatcaagtg tgacaaagag 360 aagaacttgttgcacgttac tgacactggit gttggaatga Ctagaga aga gttggittaag aacttgggta citat cqctaa gtctgg tact tcc.gagttct tgaacaagat gactgaggct

Caagaagatg gtcaatccac titc.cgagttg attggtcagt tcggtgttgg titt cit actic c 54 O gctttcttgg ttgctgacaa ggittatcgtt act tccaagc acaacaacga Cactcaa.cac atttgggaat cogatt coaa cgagttct co gtt attgctg acccalagagg taacactittg 660 ggtagaggta Ctact atcac tittggittittg aaagaagagg citt cogacta Cttggagttg 72 O gacactatica agaacttggit taagaagtac tcc cagttca toalactt CCC aatctatgtt tggit cotcca agactgagac tgttgaggaa ccalatggaag aagaagaggc tgctaaagaa 84 O gagaaagagg aatctgacga cgaggctgct gttgaagaag aggaagaaga aaagaa.gc.ca 9 OO alagact aaga aggttgaaaa gactgtttgg gactgggagc titatgaacga catcaa.gc.ca 96.O atttggcaga gaccatccaa agaggttgag gaggacgagt acaaggctitt ctacaagt cc ttct coaaag aatccgatga cc caatggct taCat CCaCt t cactgctga gggtgaagtt actittcaagt ccatcttgtt cgttccaact tctgctic caa gaggattgtt cgacgagtac 14 O ggttctaaga agt ccgacta Cat Caaactt tatgttagaa gagttitt cat Cactgacgac 2OO titccacgata tatgccaaa gtacttgaac titcgitta agg gtgttgttga titc.cgatgac 26 O ttgc cattga acgttt coag agagactittg Cagcagcaca agttgttgaa ggittatcaga 32O aagaaacttg ttagaaagac tittggacatg at Caagaaga tcgctgacga caagtacaac gacactittct ggaaagagtt cggalactaac at Caagttgg gtgtt attga ggaccacticc 44 O alacagaacta gattggctaa gttgttgaga titccagt cct Ctcat CaccC aactgacatc SOO actitcc ttgg accagtacgt. tgagagaatg aaagaga agc aggacaaaat c tact tcatg 560 US 8,067,339 B2 93 - Continued gctggttcct Ctagaaaaga ggctgaatcc tcc cc attcg ttgagagatt gttgaagaag 162O ggittacgagg titatic tact t gactgagc.ca gttgacgagt actgtat coa ggctittgc.ca 168O gagtttgacg gaaagagatt ccagaacgtt gctaaagagg gtgttaagtt cgacgaatcC 1740 gaaaagacta aagaatccag agaggctgtt gagaaagagt tcq agcc att gttgaactgg 18OO atgaaggaca aggctttgaa gga caagatc gagaaggctg ttgtttic cca gagattgact 1860 gaatcc cc at gtgctttggit tgct tcc.caa tacggatgga gtggtaiacat ggaaagaatc 1920 atgaaggctic aggcttacca aactggaaag gacat ct coa Ctalactact a cgct tcc cag 198O aagaaaactt tcgagat Caa cc caagacac c cattgatca gagacatgtt gagaagaatc aaagaggacg aggacgacaa gactgttittg gatttggctg ttgttttgtt cgagactgct 21OO actittgagat ccggitt actt gttgc.ca.gac actaaggctt acggtgacag aatcgagaga 216 O atgttgagat tgtc.cttgaa cattgaccca gacgctalagg ttgaagaaga accagaagaa 222 O gagc.ca.gagg aaactgctga agatact act gagga cactg aacaagacga ggacgalaga.g 228O atggatgttg gtactgacga agaggalagag acagcaaagg aatccactgc tgaac acgac 234 O gagttgtaa 2349

SEQ ID NO 44 LENGTH: 782 TYPE : PRT ORGANISM: Artificial Sequence FEATURE: OTHER INFORMATION: Part of human GRP94 Gene

<4 OOs, SEQUENCE: 44

Asp Asp Glu Val Asp Val Asp Gly Thir Wall Glu Glu Asp Luell Gly Lys 1. 5 1O 15

Ser Arg Glu Gly Ser Arg Thr Asp Asp Glu Val Wall Glin Arg Glu Glu 25

Glu Ala Ile Glin Lieu. Asp Gly Lieu. Asn Ala Ser Glin Ile Arg Glu Lieu. 35 4 O 45

Arg Glu Ser Glu Lys Phe Ala Phe Glin Ala Glu Wall Asn Arg Met SO 55 6 O

Met Luell Ile Ile Asn. Ser Lieu Tyr Lys Asn Glu Ile Phe Lieu. 65 70 8O

Arg Glu Luell Ile Ser Asn Ala Ser Asp Ala Lieu. Asp Ile Arg Lieu. 85 90 95

Ile Ser Luell Thir Asp Glu Asn Ala Lieu. Ser Gly Asn Glu Glu Lieu. Thir 105 11 O

Wall Ile Lys Cys Asp Llys Glu Lys Asn Lieu. Lell His Wall Thir Asp 115 12 O 125

Thir Gly Wall Gly Met Thr Arg Glu Glu Lieu. Wall Lys Asn Luell Gly Thr 13 O 135 14 O

Ile Ala Ser Gly Thr Ser Glu Phe Lieu. Asn Met Thir Glu Ala 145 150 155 160

Glin Glu Gly Glin Ser Thr Ser Glu Lieu. Ile Gly Glin Phe Gly Val 1.65 17O 17s

Gly Phe Ser Ala Phe Lieu Wall Ala Asp Llys Wall Ile Wall Thir Ser 18O 185 19 O

His Asn Asn Asp Thr Glin His Ile Trp Glu Ser Asp Ser Asn. Glu 195

Phe Ser Wall Ile Ala Asp Pro Arg Gly Asn Thr Lell Gly Arg Gly Thr 21 O 215 22O US 8,067,339 B2 95 96 - Continued

Thir Ile Thir Luell Wall Lell Glu Glu Ala Ser Asp Tyr Luell Glu Luell 225 23 O 235 24 O

Asp Thir Ile Asn Lell Wall Tyr Ser Glin Phe Ile Asn Phe 245 250 255

Pro Ile Wall Trp Ser Ser Thir Glu Thir Wall Glu Glu Pro Met 26 O 265 27 O

Glu Glu Glu Glu Ala Ala Glu Glu Glu Glu Ser Asp Asp Glu 27s 28O 285

Ala Ala Wall Glu Glu Glu Glu Glu Glu Pro Thir 29 O 295 3 OO

Wall Glu Thir Wall Trp Asp Trp Glu Luell Met Asn Asp Ile Pro 3. OS 310 315 32O

Ile Trp Glin Arg Pro Ser Glu Wall Glu Glu Asp Glu Tyr Lys Ala 3.25 330 335

Phe Ser Phe Ser Glu Ser Asp Asp Pro Met Ala Ile 34 O 345 35. O

His Phe Thir Ala Glu Gly Glu Wall Thir Phe Ser Ile Luell Phe Wall 355 360 365

Pro Thir Ser Ala Pro Arg Gly Luell Phe Asp Glu Tyr Gly Ser 37 O 375

Ser Asp Ile Lell Tyr Wall Arg Arg Wall Phe Ile Thir Asp Asp 385 390 395 4 OO

Phe His Asp Met Met Pro Luell Asn Phe Wall Gly Wall Wall 4 OS 415

Asp Ser Asp Asp Lell Pro Lell Asn Wall Ser Arg Glu Thir Luell Glin Glin 425 43 O

His Luell Luell Wall Ile Arg Luell Wall Arg Thir Luell 435 44 O 445

Asp Met Ile Lys Ile Ala Asp Asp Asn Asp Thir Phe Trp 450 45.5 460

Lys Glu Phe Gly Thir Asn Ile Luell Gly Wall Ile Glu Asp His Ser 465 470

Asn Arg Thir Arg Lell Ala Luell Luell Arg Phe Glin Ser Ser His His 485 490 495

Pro Thir Asp Ile Thir Ser Lell Asp Glin Tyr Wall Glu Arg Met Glu SOO 505 51O

Glin Asp Ile Phe Met Ala Gly Ser Ser Arg Glu Ala 515 525

Glu Ser Ser Pro Phe Wall Glu Arg Luell Luell Lys Lys Glu Wall 53 O 535 54 O

Ile Luell Thir Glu Pro Wall Asp Glu Tyr Cys Ile Ala Luell Pro 5.45 550 555 560

Glu Phe Asp Gly Lys Arg Phe Glin Asn Wall Ala Gly Wall 565 st O sts

Phe Asp Glu Ser Glu Thir Glu Ser Arg Glu Wall Glu 58O 585 59 O

Glu Phe Glu Pro Lell Lell Asn Trp Met Lys Asp Luell 595 6OO

Ile Glu Ala Wall Wall Ser Glin Arg Luell Thir Ser Pro 610 615 62O

Ala Luell Wall Ala Ser Glin Tyr Gly Trp Ser Gly Asn Met Glu Arg Ile 625 630 635 64 O

Met Ala Glin Ala Glin Thir Gly Lys Asp Ile Ser Thir Asn Tyr 645 650 655 US 8,067,339 B2 97 - Continued

Tyr Ala Ser Gln Lys Llys Thr Phe Glu Ile ASn Pro Arg His Pro Leu 660 665 67 O Ile Arg Asp Met Lieu. Arg Arg Ile Lys Glu Asp Glu Asp Asp Llys Thr 675 68O 685 Val Lieu. Asp Lieu Ala Val Val Lieu. Phe Glu Thir Ala Thr Lieu. Arg Ser 69 O. 695 7 OO Gly Tyr Lieu. Lieu Pro Asp Thir Lys Ala Tyr Gly Asp Arg Ile Glu Arg 7 Os 71O 71s 72O Met Lieu. Arg Lieu. Ser Lieu. Asn. Ile Asp Pro Asp Ala Lys Val Glu Glu 72 73 O 73 Glu Pro Glu Glu Glu Pro Glu Glu Thir Ala Glu Asp Thr Thr Glu Asp 740 74. 7 O Thr Glu Glin Asp Glu Asp Glu Glu Met Asp Val Gly Thr Asp Glu Glu 7ss 760 765 Glu Glu Thir Ala Lys Glu Ser Thr Ala Glu. His Asp Glu Lieu. 770 775 78O

<210s, SEQ ID NO 45 &211s LENGTH: 1035 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: ProteinA fusion protein (apre-5xBD-Htag) as EcoRI/Sall fragment, including alpha MF pre signal sequence, 5 Fc binding domains, and a HA and 9 x HIS tag at the C-terminus <4 OOs, SEQUENCE: 45 gaatticgaaa cqatgagatt cc catccatc titcactgctg ttttgttcgc tigcttcttct 6 O gctittggcgg cc.gctaatgc tigct Caacac gacgaagctic alacagaacgc tittctaccag 12 O gttittgaaca to caaactt gaacgctgac cagaggaatg gtttcatcca gtc.cttgaag 18O gatgacccat Ctcaatcc.gc taacgttittg ggtgaagctic agaagttgaa cacagt caa 24 O gctic ctaagg citgatgctica acaaaacaac ttcaacaagg accagcaatic togctttctac 3OO gaaatc.ttga atatgcctaa tittgaacgag got cagagaa atggatt cat t caat ctittg 360 aaagacgacc catcc.cagtic tactaatgtt ttgggagagg ctaagaaact taatgaaagt 42O caggct cota aagctgacaa caactittaac aaagagcago agaacgctitt titatgagatt 48O citta acatgc ctaacttgaa cqaagagcaa agaaacggitt ttatt caatc attgaaggac 54 O gatcct tcac agtctgctaa cittgttgtcc gaggctaaaa agttgaacga atcto aggct 6OO cctaaggctg ataataagtt caacaaagaa caacaaaatg ctittctacga gattittgcac 660 ttgccalaatt taatgagga acagagaaac ggttt tatto agt cattgaa ggatgaccct 72 O t cccaatctg ctaatttgtt ggctgaagct aagaaattga acgacgctica ggctic caaaa 78O gctgataa.ca aattcaacaa agagcaa.cag aacgctittct acgaaatctt gcatttgc.ca 84 O aacttgacag aagagcagag aaacggattic attcagt ctt taaggatga CCCtt CC9tt 9 OO tccaaagaga ttittggctgaggctaaaaag ttgaatgatg Ctcaa.gct cc aaaaggtggit 96.O ggittacccat acgatgttcc agattacgct ggagg to atc atcat cacca ccatcac cat 1 O2O Catggtggtg tcgac 1035

<210s, SEQ ID NO 46 &211s LENGTH: 341 212. TYPE: PRT <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: US 8,067,339 B2 99 100 - Continued <223> OTHER INFORMATION: ProteinA fusion protein (apre-5xBD-Htag) as EcoRI/Sall fragment, including alpha MF pre signal sequence, 5 Fc binding domains, and a HA and 9 x HIS tag at the C-terminus <4 OOs, SEQUENCE: 46 Met Arg Phe Pro Ser Ile Phe Thr Ala Val Lieu Phe Ala Ala Ser Ser 1. 5 1O 15 Ala Lieu Ala Ala Ala Asn Ala Ala Gln His Asp Glu Ala Glin Glin Asn 2O 25 3O Ala Phe Tyr Glin Val Lieu. Asn Met Pro Asn Lieu. Asn Ala Asp Glin Arg 35 4 O 45 Asn Gly Phe Ile Glin Ser Lieu Lys Asp Asp Pro Ser Glin Ser Ala Asn SO 55 6 O Val Lieu. Gly Glu Ala Glin Llys Lieu. Asn Asp Ser Glin Ala Pro Lys Ala 65 70 7s 8O Asp Ala Glin Glin Asn. Asn. Phe Asn Lys Asp Glin Glin Ser Ala Phe Tyr 85 90 95 Glu Ile Lieu. Asn Met Pro Asn Lieu. Asn. Glu Ala Glin Arg Asin Gly Phe 1OO 105 11 O Ile Glin Ser Lieu Lys Asp Asp Pro Ser Glin Ser Thr Asn Val Lieu. Gly 115 12 O 125 Glu Ala Lys Llys Lieu. Asn. Glu Ser Glin Ala Pro Lys Ala Asp Asn. Asn 13 O 135 14 O Phe Asn Lys Glu Gln Glin Asn Ala Phe Tyr Glu Ile Lieu. Asn Met Pro 145 150 155 160 Asn Lieu. Asn. Glu Glu Glin Arg Asn Gly Phe Ile Glin Ser Lieu Lys Asp 1.65 17O 17s Asp Pro Ser Glin Ser Ala Asn Lieu. Lieu. Ser Glu Ala Lys Llys Lieu. Asn 18O 185 19 O Glu Ser Glin Ala Pro Lys Ala Asp Asn Llys Phe Asn Lys Glu Glin Glin 195 2OO 2O5 Asn Ala Phe Tyr Glu Ile Lieu. His Lieu Pro Asn Lieu. Asn. Glu Glu Glin 21 O 215 22O Arg Asn Gly Phe Ile Glin Ser Lieu Lys Asp Asp Pro Ser Glin Ser Ala 225 23 O 235 24 O Asn Lieu. Lieu Ala Glu Ala Lys Llys Lieu. Asn Asp Ala Glin Ala Pro Llys 245 250 255 Ala Asp Asn Llys Phe Asn Lys Glu Glin Glin Asn Ala Phe Tyr Glu Ile 26 O 265 27 O Lieu. His Lieu Pro Asn Lieu. Thr Glu Glu Glin Arg Asn Gly Phe Ile Glin 27s 28O 285 Ser Lieu Lys Asp Asp Pro Ser Val Ser Lys Glu Ile Lieu Ala Glu Ala 29 O 295 3 OO Llys Llys Lieu. Asn Asp Ala Glin Ala Pro Lys Gly Gly Gly Tyr Pro Tyr 3. OS 310 315 32O Asp Val Pro Asp Tyr Ala Gly Gly His His His His His His His His 3.25 330 335 His Gly Gly Val Asp 34 O

<210s, SEQ ID NO 47 &211s LENGTH: 111 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: alpha-amylase-ProtAZZ/up primer US 8,067,339 B2 101 102 - Continued

<4 OOs, SEQUENCE: 47 cgga attcac gatggit cqct ttggtctt tdtttctgta C9gtc.tt cag gtc.gctgcac 6 O

Ctgctttggc titctggtggt gtt acticcag Ctgctaacgc tigct Caacac g 111

SEQ ID NO 48 LENGTH: SO TYPE: DNA ORGANISM: Artificial Sequence FEATURE: OTHER INFORMATION: HA-ProtAZZ-Xho12Z/lp primer

SEQUENCE: 48 gcct cagag cqtagt ctgg alacat cqtat ggg talaccac Caccagcatc SO

SEQ ID NO 49 LENGTH: 516 TYPE: DNA ORGANISM: Artificial Sequence FEATURE: OTHER INFORMATION: DNA sequence of the ZZ-domain as EcoRI/XhoI fragment with Alpha-amylase sequence

SEQUENCE: 49 gaattic acga tiggtc.gcttg gtggt Ctttg tittctgtacg gtc.tt caggit cqctgcacct 6 O gctittggctt Ctggtggtgt tactic cagct gctaacgctg citcaiacacga tigaagctgtt 12 O gacaacaagt t caacaaaga gcago agaac gctttctacg agat cittgca cittgccaaac 18O ttgaacgaag agcaaagaaa cgctitt catc cagtic cttga aggatgaccc atctoaatcc 24 O gctaacttgt tggctgaggc taagaagttg aacgacgctic aagct coaaa ggtcgacaat 3OO aagtttaa.ca aagaacaa.ca aaatgcct tc tacgaaattic tgcatctgcc caaccittaac 360 gaggalacaga gaaacgc.citt cattcagagt ttgaaggacg atcct tcc.ca gtctgctaat

aagccaagaa attgaatgat gcc caggctic caaaagttga tigctggtggit ggittacccat acgatgttcc agacitacgct Ctcgag 516

SEO ID NO 5 O LENGTH: 169 TYPE : PRT ORGANISM: Artificial Sequence FEATURE: OTHER INFORMATION: Protein sequence of the ZZ-domain with Alpha-amylase leader

SEQUENCE: 5 O

Met Val Ala Trp Trp Ser Leu Phe Lieu. Tyr Gly Lell Glin Wall Ala Ala 1. 5 1O 15

Pro Ala Lieu Ala Ser Gly Gly Val Thr Pro Ala Ala Asn Ala Ala Glin 25 3O

His Asp Glu Ala Val Asp Asn Lys Phe Asn Lys Glu Glin Glin Asn Ala 35 4 O 45

Phe Tyr Glu Ile Lieu. His Lieu Pro Asn Luell Asn Glu Glu Glin Arg Asn SO 55 6 O

Ala Phe Ile Glin Ser Lieu Lys Asp Asp Pro Ser Glin Ser Ala Asn Lieu. 65 70 7s 8O

Lell Ala Glu Ala Lys Llys Lieu. Asn Asp Ala Glin Ala Pro Llys Val Asp 85 90 95

Asn Phe Asn Lys Glu Glin Glin Asn Ala Phe Glu Ile Lieu. His 105 11 O US 8,067,339 B2 103 104 - Continued

Lieu Pro Asn Lieu. Asn. Glu Glu Glin Arg Asn Ala Phe Ile Glin Ser Lieu. 115 12 O 125 Lys Asp Asp Pro Ser Glin Ser Ala Asn Lieu. Lieu Ala Glu Ala Lys Llys 13 O 135 14 O Lieu. Asn Asp Ala Glin Ala Pro Llys Val Asp Ala Gly Gly Gly Tyr Pro 145 150 155 160 Tyr Asp Val Pro Asp Tyr Ala Lieu. Glu 1.65

SEO ID NO 51 LENGTH: 33 TYPE: DNA ORGANISM: Artificial Sequence FEATURE: OTHER INFORMATION: 5Ecoapp primer

SEQUENCE: 51 aacggaattic atgagattitc cittcaattitt tac 33

SEO ID NO 52 LENGTH: 43 TYPE: DNA ORGANISM: Artificial Sequence FEATURE: OTHER INFORMATION: 3Htag Sal primer SEQUENCE: 52 cgatgtcgac gtgatggtga tiggtggtgat gatgatgacc acc 43

SEO ID NO 53 LENGTH: 882 TYPE: DNA ORGANISM: Artificial Sequence FEATURE: OTHER INFORMATION: FoRIII (LF) as EcoRIA SalI fragment

SEQUENCE: 53 gaatt catga gatttic ctitc aattitt tact gctgttt tat tcqcago atc citc.cgcatta 6 O gctgct coag tica acactac alacagaagat gaaacgg CaC aaatticcggc tgaagctgtc. 12 O atcggittact Cagatttaga aggggatttic gatgttg Ctg ttittgcc att titccaac agc 18O acaaataacg ggittattgtt tataaatact act attg CCa gcattgctgc taaagaagaa 24 O ggggitatic to tcgagaaaag agctggaatg agaactg agg acttgccaaa ggctgttgtt 3OO ttcttggagc cacagtggta cagagttittg gagaagg att cc.gttactitt gaagtgtcag 360 ggagct tact citccagaaga taact coact cagtggit to c acaacgaatc cittgatttct t ct caggctt cotcct actt cattgacgct gctactg ttg acgatt.ccgg tgagtacaga tgtcagacta acttgtccac tttgtc.cgac ccagttc aat tggaggttca catcggttgg 54 O ttgttgttgc aagctic caag atgggittitt C aaggagg agg accoa attca tittgagatgt cact cittgga agaacactgc tittgcacaaa gttactt act tgcagaacgg aaagggit aga 660 aagtatttico accacaactic cqactitctac atcc.caa agg c tactittgaa ggatt CC9gt 72 O t cct acttct gtagaggatt gttcggttcc aagaacg titt citt cogagac tgttaa.catc actato actic agggattggc tigttt coact atctott cott tott coccacci aggittat caa 84 O ggtggtggtc atcatcatca ccaccatcac cat cacg tog ac 882

SEO ID NO 54 LENGTH: 292 US 8,067,339 B2 105 106 - Continued

212. TYPE: PRT <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: FoRIII (LF) with alpha MF pre signal sequence and HIS Tag

<4 OOs, SEQUENCE: 54 Met Arg Phe Pro Ser Ile Phe Thr Ala Val Lieu Phe Ala Ala Ser Ser 1. 5 1O 15 Ala Lieu Ala Ala Pro Val Asn. Thir Thir Thr Glu Asp Glu Thir Ala Glin 2O 25 3O Ile Pro Ala Glu Ala Val Ile Gly Tyr Ser Asp Lieu. Glu Gly Asp Phe 35 4 O 45 Asp Wall Ala Val Lieu Pro Phe Ser Asn. Ser Thr Asn. Asn Gly Lieu. Lieu SO 55 6 O Phe Ile Asn. Thir Thir Ile Ala Ser Ile Ala Ala Lys Glu Glu Gly Val 65 70 7s 8O Ser Lieu. Glu Lys Arg Ala Gly Met Arg Thr Glu Asp Lieu Pro Lys Ala 85 90 95 Val Val Phe Lieu. Glu Pro Glin Trp Tyr Arg Val Lieu. Glu Lys Asp Ser 1OO 105 11 O Val Thr Lieu Lys Cys Glin Gly Ala Tyr Ser Pro Glu Asp Asn Ser Thr 115 12 O 125 Gln Trp Phe His Asn Glu Ser Lieu. Ile Ser Ser Glin Ala Ser Ser Tyr 13 O 135 14 O Phe Ile Asp Ala Ala Thr Val Asp Asp Ser Gly Glu Tyr Arg Cys Glin 145 150 155 160 Thr Asn Lieu Ser Thr Lieu Ser Asp Pro Val Glin Lieu. Glu Val His Ile 1.65 17O 17s Gly Trp Lieu. Lieu. Lieu. Glin Ala Pro Arg Trp Val Phe Lys Glu Glu Asp 18O 185 19 O Pro Ile His Lieu. Arg Cys His Ser Trp Lys Asn. Thir Ala Lieu. His Lys 195 2OO 2O5 Val Thr Tyr Lieu. Glin Asn Gly Lys Gly Arg Llys Tyr Phe His His Asn 21 O 215 22O Ser Asp Phe Tyr Ile Pro Lys Ala Thr Lieu Lys Asp Ser Gly Ser Tyr 225 23 O 235 24 O Phe Cys Arg Gly Lieu. Phe Gly Ser Lys Asn Val Ser Ser Glu. Thr Val 245 250 255 Asn Ile Thir Ile Thr Glin Gly Lieu Ala Val Ser Thr Ile Ser Ser Phe 26 O 265 27 O Phe Pro Pro Gly Tyr Glin Gly Gly Gly His His His His His His His 27s 28O 285 His His Val Asp 29 O

<210s, SEQ ID NO 55 &211s LENGTH: 1119 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: FoRIas EcoRI/Sall fragment: <4 OO > SEQUENCE: 55 gaatt catga gatttic ctitc aattitt tact gctgttittat t cqcagoat c ctic cqcatta 6 O gctgct coag tica acactac alacagaagat gaaacggcac alaatticcggc tigaagctgtc. 12 O atcggittact cagatttaga aggggatttic gatgttgctg ttittgccatt titcca acagc 18O US 8,067,339 B2 107 108 - Continued acaaataacg ggittattgtt tataaatact act attgc.ca gcattgctgc taaagaagaa 24 O ggggitatic to tcgagaaaag agctgatact actaaggctg titat cactitt gcaaccacca 3OO tgggtttc.cg ttitt coagga ggagactgtt actittgcact gtgaggittitt gcatttgcct 360 ggttcc tott c cacticagtg gttcttgaac ggtactgcta Ctcaaactt C CactC catcC tacagaatta citt cogctitc cgittaacgat tctggtgagt acagatgtca gagaggattg tctggtagat ccgacc caat t cagttggag att cacagag gatggttgtt gttgcaggitt 54 O t cct coagag ttitt cactga gggtgalacca ttggctttga gatgtcacgc ttggaaggac aagttggttt acaacgttitt gtact acaga aacggaaagg Ctttcaagtt citt coactgg 660 aact CCaact tgactatott gaaaactaac atctoccaca acgg tactta c cactgttct 72 O ggaatgggaa agcacagata cact tcc.gct ggt at ct cog ttactgttaa ggagttgttc c cagct coag ttittgaacgc titc.cgttact t ct coattgt tggagggaala cittggittact 84 O ttgtcc tigtg agactaaatt gttgttgcaa. agacCaggat tgcagttgta ottct cotto 9 OO tacatgggtt c caagactitt gagaggtaga aac actt CCt cc.gagtacca aatcttgact 96.O gctagaagag aggatt Cogg tttgtactgg tgttgaagctg Ctact gagga cggtaacgtt ttgaagagat c cc cagagtt ggagttgcaa. gttittgggat tgcaattgcc aactic caggit 108 O ggtggit catc atcatCacca CCatCaCCat cacgt.cgac 1119

<210s, SEQ I D NO 56 &211s LENGT H: 371 212. TYPE : PRT <213> ORGANISM: Artificial Sequence 22 Os. FEATU RE: <223> OTHER INFORMATION: FoRI with alpha MF pre signal sequence and HIS Tag

<4 OOs, SEQUENCE: 56

Met Arg Phe Pro Ser Ile Phe Thr Ala Wall Lieu. Phe Ala Ala Ser Ser 1. 1O 15

Ala Lieu Ala Ala Pro Wall Asn. Thir Thir Thr Gul Asp Glu Thir Ala Glin 25

Ile Pro Ala Glu Ala Val Ile Gly Tyr Ser Asp Lell Glu Gly Asp Phe 35 4 O 45

Asp Wall Ala Wall Leu Pro Phe Ser Asn. Ser Thr Asn Asn Gly Lieu. Luell SO 55 6 O

Phe Ile Asn Thir Thir Ile Ala Ser Ile Ala Ala Glu Glu Gly Val 65 70 7s 8O

Ser Lieu. Glu Lys Arg Ala Asp Thr Thir Lys Ala Wall Ile Thir Lieu. Glin 85 90 95

Pro Pro Trp Wall Ser Wall Phe Glin Glu Glu. Thir Wall Thir Luell His Cys 105 11 O

Glu Wall Lieu. His Leu Pro Gly Ser Ser Ser Thr Glin Trp Phe Luell Asn 115 12 O 125

Gly Thr Ala Thr Gin. Thir Ser Thr Pro Ser Tyr Arg Ile Thir Ser Ala 13 O 135 14 O

Ser Wall Asn Asp Ser Gly Glu Tyr Arg Cys Glin Arg Gly Luell Ser Gly 145 150 155 160

Arg Ser Asp Pro Ile Glin Lieu. Glu Ile His Arg Gly Trp Luell Lieu. Luell 1.65 17O 17s

Glin Wal Ser Ser Arg Val Phe Thr Glu Gly Glu Pro Lell Ala Lieu. Arg 18O 185 19 O US 8,067, 339 B2 109 110 - Contin lued His Ala Trp Lys Asp Llys Lieu. Val Tyr Asn Val Lieu. Tyr Tyr Arg 195 2O5

Asn Gly Lys Ala Phe Llys Phe Phe His Trp Asn Ser Asn. Luell Thir Ile 21 O 215 22O

Lell Llys Thr Asn. Ile Ser His Asn Gly Thr Tyr His Cys Ser Gly Met 225 23 O 235 24 O

Gly Llys His Arg Tyr Thr Ser Ala Gly Ile Ser Wall. Thir Wall Lys Glu 245 250 255

Lell Phe Pro Ala Pro Wall Lieu. Asn Ala Ser Wall Thir Ser Pro Lieu. Luell 26 O 265 27 O

Glu Gly Asn Lieu Val Thir Lieu. Ser Cys Glu Thr Llys Lieu. Lieu. Lieu. Glin 27s 285 Arg Pro Gly Lieu Gln Lieu. Tyr Phe Ser Phe Tyr Met Gly Ser Lys Thr 29 O 295 3 OO

Lell Arg Gly Arg Asn. Thir Ser Ser Glu Tyr Glin Ile Lieu. Thir Ala Arg 3. OS 310 315 32O

Arg Glu Asp Ser Gly Lieu. Tyr Trp Cys Glu Ala Ala Thr Glu Asp Gly 3.25 330 335

Asn Val Lieu Lys Arg Ser Pro Glu Lieu. Glu Lieu Glin Wall Lieu. Gly Lieu. 34 O 345 35. O

Glin Leu Pro Thr Pro Gly Gly Gly His His His His His His His His 355 360 365

His Val Asp 37 O

<210s, SEQ ID NO 57 &211s LENGTH: 35 &212s. TYPE: DNA ORGANISM: Artificial Sequence 22 Os. FEATURE: OTHER INFORMATION: 5gutBgl II primer

<4 OO > SEQUENCE: 57 attgagat ct acc caattta gcagoctdca ttct c 35

<210s, SEQ ID NO 58 &211s LENGTH: 36 &212s. TYPE: DNA ORGANISM: Artificial Sequence 22 Os. FEATURE: OTHER INFORMATION: 3gutEco RI primer

<4 OOs, SEQUENCE: 58 gtcagaattic atctgtggta tagtgtgaaa aagtag 36

<210s, SEQ ID NO 59 &211s LENGTH: 1010 &212s. TYPE: DNA ORGANISM: Artificial Sequence 22 Os. FEATURE: OTHER INFORMATION: Pichia pastoris GUT1 promoter <4 OO > SEQUENCE: 59 agat ct accc aatttagcag cctgcattct cittgattitta tgggggaaac taacaat agt 6 O gttgcCttga ttittaagtgg cattgttctt tgaaatcgaa attggggata acgt.cat acc 12 O gaaaggtaaa Caactt.cggg gaattgcc ct ggittaaacat ttattaag.cg agataaatag 18O gggatagcga gat agggggc ggagaagaag aagggtgtta aattgctgaa atctotcaat 24 O

Ctggaagaala C9gaataaat taact cotto ctgagataat aagat.ccgac tctgctatoga 3OO US 8,067,339 B2 111 112 - Continued CCCC acacgg tactgacctic ggcat acc cc attggatctg gtgcgaagca acaggtoctg 360 aaac ctittat cacgtgtagt agattgacct tccagoaaaa aaaggcatta tatattttgt 42O tgttgaaggg gtgaggggag gtgcaggtgg ttcttittatt citcttgtag ttaattitt CC 48O cggggttgcg gagct caaa agtttgc.ccg atctgatagc titgcaagatg ccaccgctta 54 O tccaacgcac titcagaga.gc titgcc.gtaga aagaacgttt to citcgtagt atticcagolac 6OO tt catggtga agt cqctatt t caccgaagg ggggg tatta aggttgcgca CCCCCtcCCC 660 acaccc.ca.ga atcgtttatt ggctgggttcaatggcgttt gagttagcac atttittt cot 72 O taalacaccct C caaac acgg ataaaaatgc atgtgcatcc taaactggit agagatgcgt. 78O actic.cgtgct C cataataa cagtggtgtt ggggttgctg. ittagct cacg cactic.cgttt 84 O tttitttcaac cagcaaaatt cqatiggggag aaacttgggg tactittgc.cg act cotccac 9 OO catactggta tataaataat act cqcccac ttitt cqtttg citgcttittat atttcaagga 96.O ctgaaaaaga citcttcttct actttitt cac actataccac agatgaattic 1010

<210s, SEQ ID NO 60 &211s LENGTH: 322 212. TYPE: PRT <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: S. cerevisiae SED1 (without endogenous leader sequence

<4 OOs, SEQUENCE: 60 Val Asp Glin Phe Ser Asn Ser Thr Ser Ala Ser Ser Thr Asp Val Thr 1. 5 1O 15 Ser Ser Ser Ser Ile Ser Thr Ser Ser Gly Ser Val Thr Ile Thir Ser 2O 25 3O Ser Glu Ala Pro Glu Ser Asp Asn Gly Thr Ser Thr Ala Ala Pro Thr 35 4 O 45 Glu Thir Ser Thr Glu Ala Pro Thr Thr Ala Ile Pro Thr Asn Gly Thr SO 55 6 O Ser Thr Glu Ala Pro Thir Thr Ala Ile Pro Thr Asn Gly Thr Ser Thr 65 70 7s 8O Glu Ala Pro Thr Asp Thr Thr Thr Glu Ala Pro Thr Thr Ala Leu Pro 85 90 95 Thr Asn Gly Thr Ser Thr Glu Ala Pro Thr Asp Thr Thr Thr Glu Ala 1OO 105 11 O Pro Thir Thr Gly Lieu Pro Thr Asn Gly Thr Thr Ser Ala Phe Pro Pro 115 12 O 125 Thir Thr Ser Leu Pro Pro Ser Asn. Thir Thr Thr Thr Pro Pro Tyr Asn 13 O 35 14 O Pro Ser Thr Asp Tyr Thr Thr Asp Tyr Thr Val Val Thr Glu Tyr Thr 145 150 155 160 Thr Tyr Cys Pro Glu Pro Thr Thr Phe Thr Thr Asn Gly Lys Thr Tyr 1.65 70 17s Thr Val Thr Glu Pro Thir Thr Lieu. Thir Ile Thr Asp Cys Pro Cys Thr 18O 185 19 O Ile Glu Lys Pro Thr Thr Thr Ser Thr Thr Glu Tyr Thr Val Val Thr 195 2OO 2O5 Glu Tyr Thr Thr Tyr Cys Pro Glu Pro Thir Thr Phe Thr Thr Asn Gly 21 O 215 22O Lys Thr Tyr Thr Val Thr Glu Pro Thr Thr Lieu. Thir Ile Thr Asp Cys 225 23 O 235 24 O US 8,067,339 B2 113 114 - Continued Pro Cys Thr Ile Glu Lys Ser Glu Ala Pro Glu Ser Ser Wall Pro Wall 245 250 255

Thr Glu Ser Lys Gly. Thir Thr Thr Lys Glu Thr Gly Wall Thir Thr Lys 26 O 265 27 O

Glin. Thir Thir Ala Asn. Pro Ser Lieu. Thir Wal Ser Thir Wall Wall Pro Wall 27s 285

Ser Ser Ser Ala Ser Ser His Ser Wall Wall Ile Asn Ser Asn Gly Ala 29 O 295 3 OO

Asn Val Val Val Pro Gly Ala Lieu. Gly Lieu Ala Gly Wall Ala Met Lieu. 3. OS 310 315 32O

Phe Lieu.

<210s, SEQ ID NO 61 &211s LENGTH: 969 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: 223 OTHER INFORMATION: S. cere visiae SED1 (without endogenous leader sequence

<4 OOs, SEQUENCE: 61 gtcgaccalat t ct ctaactic tact tcc.gct tcc to tactg acgttacttic ctic ct cotct 6 O atttct actt cotccggttc cgttactatt act tcct citg aggct Coaga atctgacaac 12 O gg tact tcta citgctgct co aactgaaact tct actgagg CtcCtactac tgctatt coa 18O actaacggaa citt coacaga ggctic caa.ca acagotaticc Ctacaaacgg taCat CCaCt 24 O gaagct cota Ctgacactac tacagaagct ccaactactg Ctttgcctac taatggtaca 3OO t caacagagg ct cctacaga tacaacaact gaagctic caa caactggatt gccaacaaac 360 gg tact actt ctdctitt.ccc accalactact tcc ttgccac CatCcaa.cac tact act act 42O

CCaC cataca accoat CCaC tgact acact actgactaca cagttgttac tgagtacact 48O actitactgtc. cagagccaac tactitt Caca acaaacggaa agacittacac tgttact gag 54 O cc tact actt tact atcac tgactgtc.ca tgtact atcg agaa.gc.caac tact act tcc. 6OO actacagagt at actgttgt tacagaatac acaa.catatt gtc.ctgagcc aacaa.catt C 660 actactaatg gaaaaacata cacagttaca gaaccaacta cattgacaat tacagattgt 72 O ccttgtacaa ttgagaagtic cgaggctic ct gaatc.ttctg titccagttac tgaatccaag 78O gg tact act a ctaaagaaac tggtgttact actalagcaga c tactgctaa cc catcc ttg 84 O actgtttcca citgttgttcc agtttct tcc totgcttctt cc cactic cqt tgttatcaac 9 OO tccaacggtg Ctalacgttgt tgttcctggit gctttgggat tgctatgttg 96.O ttcttgtaa 96.9

What is claimed: diverse yeast host cells capable of displaying immuno 1. A method for producing yeast host cells that express an 55 globulins on the surface thereof; immunoglobulin of interest, comprising: (c) growing the yeast host cells in the presence of glycerol (a) providing yeast host cells that include a first nucleic for a time sufficient to produce the capture moiety on the acid molecule operably linked to a GUT1 promoter and Surface of the yeast host cells and then growing the yeast which encodes a capture moiety comprising a cell Sur 60 cells in a medium that lacks glycerol but contains dex face anchoring protein fused to protein A; trose to produce yeast host cells in which the immuno (b) transfecting the yeast host cells with a plurality of globulins expressed in the yeast host cells are captured nucleic acid molecules encoding a genetically diverse by the capture moiety on the surface of the yeast host population of heavy and light chains of an immunoglo cells; bulin wherein at least one of the heavy or light chain 65 encoding nucleic acid molecules is operably linked to a (d) contacting the host cells with a detection means that GAPDH promoter to produce a plurality of genetically specifically binds to the immunoglobulin of interest; and US 8,067,339 B2 115 116 (e) isolating yeast host cells in which the detection means Pichia methanolica, Pichia sp., Saccharomyces cerevisiae, is bound to the immunoglobulin of interest to produce Saccharomyces sp., Hansenula polymorpha, Kluyveromyces the yeast host cells that express the immunoglobulin of sp., Kluyveromyces lactis, and Candida albicans. interest. 8. The method of claim 1, wherein, the yeast host cells are 2. The method of claim 1, wherein the cell surface anchor Pichia pastoris. ing protein is a GPI protein. 9. The method of claim 1, wherein the yeast host cells 3. The method of claim 1, wherein the detection means is include a deletion or disruption of one or more of the B-man an antigen that is capable of being bound by the immunoglo nosyltransferase genes. bulin of interest. 10. The method of claim 1, wherein the yeast host cells that 10 include the first nucleic acid molecule have been genetically 4. The method of claim 3, wherein the antigen is conju modified to produce glycoproteins that have predominantly gated to or labeled with a fluorescent moiety. an N-glycan selected from the group consisting of complex 5. The method of claim 1, wherein the cell surface anchor N-glycans, hybrid N-lycans, and high mannose N-glycans. ing protein is selected from the group consisting of C-agglu tinin, Cwplp, Cwp2p, Gas 1p, Yap3p, Flo1p, Crh2p, Pir1p, 11. The method of claim 1, wherein the yeast host cells that 15 include the first nucleic acid molecule have been genetically Pir4p, Sedlp, Tiplp, Hpwplp. Als3p, and Rbt5p. modified to produce glycoproteins that have predominantly 6. The method of claim 1, wherein the cell surface anchor an N-glycan selected from the group consisting of ing protein is GlcNAc, Man-GlcNAca, Gali-4ClcNAc-4) Sedlp. Man-GlcNAc, NANA Gala Man-GlcNAca. 7. The method of claim 1, wherein the yeast host cells are GlcNAcMans.GlcNAc GalGlcNAcMans GlcNAc, selected from the group consisting of Pichia pastoris, Pichia NANAGalGlcNAcMan-GlcNAc, ManGlcNAc, finlandica, Pichia trehalophila, Pichia koclamae, Pichia Mans.GlcNAc, Man GlcNAc, Man, GlcNAc, membranaefaciens, Pichia minuta (Ogataea minuta, Pichia Mans.GlcNAc, and Man-GlcNAc lindneri), Pichia opuntiae, Pichia thermotolerans, Pichia salictaria, Pichia guercuum, Pichia piperi, Pichia stiptis, k k k k k