<<

USO095931 60B2

(12) United States Patent (10) Patent No.: US 9,593,160 B2 Ingber et al. (45) Date of Patent: Mar. 14, 2017

(54) ENGINEERED MICROBE-TARGETING (56) References Cited MOLECULES AND USES THEREOF U.S. PATENT DOCUMENTS (75) Inventors: Donald E. Ingber, Boston, MA (US); 5,137,810 A 8, 1992 Sizemore et al. Michael Super, Lexington, MA (US); 5,270,199 A 12/1993 Ezekowitz Jeffrey Charles Way, Cambridge, MA 5,545,820 A 8, 1996 Gatehouse et al. 5,951,976 A 9/1999 Segal (US); Mark J. Cartwright, West 6,117,977 A 9/2000 Lasky et al. Newton, MA (US); Julia B. Berthet, 6,225,046 B1 5/2001 Vesey et al. Framingham, MA (US); Dinah R. 6,503,761 B1 1/2003 Koenig et al. Super, Lexington, MA (US); Martin 6,528,624 B1 3/2003 Idusogie et al. 6,562,784 B1 5/2003 Thiel et al. M. Rottman, Cambridge, MA (US); 6,733,753 B2 5, 2004 Boone et al. Alexander Watters, Melrose, MA (US) 6,846,649 B1 1/2005 Thiel et al. 6,900,292 B2 5/2005 Sun et al. (73) Assignee: PRESIDENT AND FELLOWS OF 7,202,207 B2 4/2007 Thiel et al. 7,211,396 B2 5/2007 Uttenthal HARVARD COLLEGE, Cambridge, 7,226,429 B2 6, 2007 Tullis MA (US) (Continued) (*) Notice: Subject to any disclaimer, the term of this patent is extended or adjusted under 35 FOREIGN PATENT DOCUMENTS U.S.C. 154(b) by 0 days. EP O375,736 B1 5, 1998 EP O861667 A2 9, 1998 (21) Appl. No.: 14/233,553 (Continued)

(22) PCT Filed: Jul. 18, 2012 OTHER PUBLICATIONS (86). PCT No.: Azevedo et al. 2003; Horseradish peroxidase: a valuable tool in biotechnology. Biotechnology Annual Review, 9: 199-247.* S 371 (c)(1), Functional MBL Elisa kit (ligand elisa); created Feb. 28, 2006 on (2), (4) Date: Apr. 18, 2014 the web at mastgrp.biz Sanquin Glossies/MBL%20flyerO6.pdf.* Brooks et al., “Expression and secretion of ficolin beta by porcine (87) PCT Pub. No.: WO2013/012924 neutrophils'. Biochim. Biophys. Acta 1624 (1-3):36-45 (2003). Fox et al., “Single amino acid substitutions on the Surface of PCT Pub. Date: Jan. 24, 2013 Escherichia coli maltose-binding protein can have a profound impact on the solubility of fusion proteins”. Protein Sci. 10(3):622 (65) Prior Publication Data 630 (2001). Grogl et al., “Leishmania braziliensis: protein, carbohydrate, and US 2014/0227723 A1 Aug. 14, 2014 antigen differences between log phase and stationary phase promastigotes in vitro”. Exp. Parasitol. 63(3):352-359 (1987). Related U.S. Application Data (Continued) (60) Provisional application No. 61/508,957, filed on Jul. Primary Examiner — Karen Cochrane Carlson 18, 2011, provisional application No. 61/605,052, (74) Attorney, Agent, or Firm — Nixon Peabody LLP filed on Feb. 29, 2012, provisional application No. 61/605,081, filed on Feb. 29, 2012. (57) ABSTRACT Described herein are engineered microbe-targeting or (51) Int. Cl. microbe-binding molecules, kits comprising the same and C07K I4/42 (2006.01) uses thereof. Some particular embodiments of the microbe C07K 6/12 (2006.01) targeting or microbe-binding molecules comprise a carbo A6 IK 47/48 (2006.01) hydrate recognition domain of mannose-binding lectin, or a GOIN 33/543 (2006.01) fragment thereof, linked to a portion of a Fc region. In some GOIN 33/569 (2006.01) embodiments, the microbe-targeting molecules or microbe (52) U.S. Cl. binding molecules can be conjugated to a Substrate, e.g., a CPC ...... C07K 16/12 (2013.01); A61K 47/4843 magnetic microbead, forming a microbe-targeting Substrate (2013.01); A61K 47/48369 (2013.01); C07K (e.g., a microbe-targeting magnetic microbead). Such 14/42 (2013.01); G0IN 33/54353 (2013.01); microbe-targeting molecules and/or Substrates and the kits G0IN 33/569 II (2013.01); G0IN 33/569 16 comprising the same can bind and/or capture of a microbe (2013.01); G0IN 33/56938 (2013.01); G0IN and/or microbial matter thereof, and can thus be used in 33/56961 (2013.01); C07K 2319/33 (2013.01); various applications, e.g., diagnosis and/or treatment of an G0IN 2333/42 (2013.01); G0IN 2333/4724 caused by microbes such as sepsis in a Subject or (2013.01) any environmental Surface. Microbe-targeting molecules (58) Field of Classification Search and/or Substrates can be regenerated after use by washing CPC C07K 2319/30; C07K 16/2866; C07K 14/42: with a low pH buffer or buffer in which calcium is insoluble. A61K 39/395 See application file for complete search history. 18 Claims, 40 Drawing Sheets US 9,593,160 B2 Page 2

(56) References Cited Jarva et al., “Streptococcus pneumoniae evades complement attack and opsonophagocytosis by expressing the pspC locus-encoded Hic U.S. PATENT DOCUMENTS protein that binds to short consensus repeats 8-11 of factor H'. J. Immunol. 168(4): 1886-1894 (2002). 7,439,224 B2 10/2008 Thiel et al. Kehres, “A kinetic model for binding protein-mediated arabinose 7.462,596 B2 12/2008 Larsen et al. transport', Protein Sci. 1 (12): 1661-1665 (1992). 7,695,937 B2 4/2010 Baum Ye et al., "Surface display of a glucose binding protein'. Journal of 7,763,436 B2 7, 2010 Das et al. Molecular Catalysis B: Enzymatic 28(4-6):201-206 (2004). 8,080.245 B2 12/2011 Visintin et al. Bangs Laboratories, Inc., “Protein Coated Microspheres'. Tech. 8,084.275 B2 12/2011 Hirai et al. Note #51 (1997). 8,088,596 B2 1/2012 Zeng et al. Brouwer et al., “Mannose-Binding Lectin (MBL) Facilitates 8,415, 118 B2 4/2013 Huang et al. Opsonophagocytosis of Yeasts but Not of Bacteria despite MBL 2003,0162248 A1 8/2003 Wakamiya Binding”. J. Immunol., 180(6):4124-32 (2008). 2003. O166878 A1 9/2003 Nishiya et al. Frakking et al., “Safety and pharmacokinetics of plasma-derived 2004O161861 A1* 8/2004 Levon ...... GO1N 33,569 11 mannose-binding lectin (MBL) substitution in children with che 436,518 motherapy-induced neutropaenia'. Eur, J. Cancer, 45(4):505-12 2004/02292.12 A1 11/2004 Thiel et al. (2009). 2005, OO14932 A1 1/2005 Imboden et al. Garred et al., “Mannose-binding lectin and its genetic variants'. 2005, 0037949 A1 2/2005 O'Brien et al. Genes and Immunity, 7(2):85-94 (2006). 2006,0040362 A1 2/2006 Wakamiya Jack et al., “Mannose-binding lectin: targeting the microbial world 2006/O104975 A1 5/2006 Geitenbeek et al. for complement attack and opsonophagocytosis'. Immunological 2006/0177879 A1 8/2006 Mayes et al. Reviews, 180:86-99 (2001). 2006, O188963 A1 8/2006 Kongerslev et al. Kang et al., “The human macrophage mannose receptor directs 2006,025 1580 A1 11/2006 Keppler et al. Mycobacterium tuberculosis lipoarabinomannan-mediated phago 2007/OO72247 A1 3/2007 Wong et al. some biogenesis”. J. Exp. Med., 202(7):987-99 (2005). 2007/0224640 A1 9/2007 Dahlgren Caldwell et al. Krarup et al., “Simultaneous Activation of Complement and Coagul 2007,02698.18 A1 11/2007 Savage 2007/0269878 A1 11/2007 Gerard et al. lation by MBL-Associated Serine Protease 2”. PLoS One, 2(7):e623 2008. O156736 A1 7/2008 Hirai et al. (2007). 2008/O182793 A1 7/2008 Baum Lo et al., “High level expression and secretion of Fc-X fusion 2008.0193965 A1 8/2008 Zeng et al. proteins in mammalian cells'. Protein Eng., 11 (6):495-500 (1998). 2009, O181041 A1 7/2009 Holgersson et al. Matsushita et al., “Activation of the Classical Complement Pathway 2009, 0220932 A1 9/2009 Ingber et al. by Mannose-binding Protein in Association with a Novel Cls-like 2009,0297516 A1 12/2009 Mayo et al. Serine Protease”. J. Exp. Med., 176(6): 1497-502 (1992). 2010/0266558 A1 10/2010 Zipori Nethet al., Mannose-Binding Lectin Binds to a Range of Clinically 2010.0323429 A1 12/2010 Hu et al. Relevant Microorganisms and Promotes Complement Deposition, 2010/0331240 A1 12/2010 Michelow et al. Infect. Immun., 68(2):688-93 (2000). 2011/0027267 A1 2/2011 Kyneb et al. Neth et al., “Enhancement of Complement Activation and 2011/0053145 A1 3/2011 Takakura et al. Opsonophagocytosis by Complexes of Mannose-Binding Lectin 2011/0053250 A1 3/2011 Takakura et al. with Mannose-Binding Lectin-Associated Serine Protease After 2011 O159000 A1 6, 2011 Silverman Binding to Staphylococcus aureus'. J. Immunol. 169:4430-4436 2011 0183398 A1 7/2011 Dasaratha et al. (2002). 2011 (0281792 A1 11/2011 Zion et al. Perham, "Domains, Motifs, and Linkers in 2-Oxo Acid 2012fO164628 A1 6, 2012 Duffin et al. Dehydrogenase Multienzyme Complexes: A Paradigm in the Design of a Multifunctional Protein'. Biochem. 30(35):8501-12 FOREIGN PATENT DOCUMENTS (1991). Rutishauser et al., “Amino Acid Sequence of the Fc Region of a EP O91597O B1 9, 2004 Human gamma(G Immunoglobulin'. Biochemistry, 61:1414-1421 EP 1812459 B1 3, 2011 (1968). WO 02A32292 A2 4/2002 Safarik et al., “The application of magnetic separations in applied WO O3,O14150 A2 2, 2003 microbiology”, Journal of Applied Bacteriology, 78:575-585 WO 03/054164 A2 T 2003 (1995). WO 2004/O18698 A2 3, 2004 Sprong et al., “Mannose-Binding Lectin is a Critical Factor in WO 2007 OO1332 A2 1, 2007 Systemic Complement Activation during Meningococcal Septic WO 2007/044642 A2 4/2007 WO 2007. 111496 A1 10/2007 Shock”. Clinical Infectious Diseases, 49(9): 1380-6 (2009). WO 2008, 130618 A1 10, 2008 Thiel et al., “A second serine protease associated with mannan WO 2009/062195 A2 5, 2009 binding lectin that activates complement”. Nature, 386:506-10 WO 2009/126346 A2 10/2009 (1997). WO 2011 (090954 A2 T/2011 Ward et al., “Characterization of Humanized Antibodies Secreted by WO 2011/09 1037 A2 T/2011 ApSergillus niger'. Applied and Environmental Microbiology, WO 201109.0954 A2 T/2011 70(5):2567-2576 (2004). WO 201210.0099 A2 T 2012 Wriggers et al., “Control of Protein Functional Dynamics by Peptide WO 2012/135834 A2 10/2012 Linkers”, Biopolymers, 80(6):736-46 (2005). Xia et al., "Combined microfluidic-micromagnetic separation of OTHER PUBLICATIONS living cells in continuous flow”. Biomed Microdevices, 8 (4):299 308 (2006). Holmskov et al., “Affinity and kinetic analysis of the bovine plasma Yung et al., “Micromagnetic-microfluidic blood cleansing device'. Lab Chip, 9:1171-7 (2009). C-type lectin collectin-43 (CL-43) interacting with mannan'. FEBS Chen et al., Angewandte Chemie Int. Ed., 47(45):8627-8630 (2008). Lett. 393(2-3):314-316 (1996). “Fabrication of an Oriented Fc-Fused Lectin Microarray through Huang et al., “Porcine DC-Sign: molecular cloning, gene structure, Boronate Formation.” tissue distribution and binding characteristics'. Dev. Comp. Linehan et al., Eur, J. Immunol. 31:1857-1866 (2001). “Endog Immunol. 33(4)464-480 (2009). enous Ligands of Carbohydrate Recognition Domains of the Man Hwang et al., “The pepper mannose-binding lectin gene CaMBL1 nose Receptor in Murine Macrophages, Endothelial Cells and is required to regulate cell death and defense responses to microbial Secretory Cells; Potential Relevance to Inflammation and Immu pathogens”. Plant Physiol. 155(1):447-463 (2011). nity.” US 9,593,160 B2 Page 3

(56) References Cited Arakawa et al., “Elution of antibodies from a protein-A column by aqueous arginine solutions.” Protein Expression and Purification OTHER PUBLICATIONS 36:244-248 (2004). Armour et al., “Recombinant human IgG molecules lacking Nisnevitch et al., J. Biochem. Biophys. Methods, 49:467-480 Ficgamma receptor I binding and monocyte triggering activities.” (2001). “The Solid Phase in Affinity Chromatography: Strategies for Antibody Attachment.” Eur, J. Immunol. 29:2613-2624 (1999). Terai et al., Eur, J. Immunol. 33:2755-2763 (2003). “Relationship Ashkenazi et al., “Immunoadhesins as research tools and therapeu between Gene Polymorphisms of Mannose-Binding Lectin (MBL) tic agents.” Curr Opin Immunol. 9(2):195-200 (1997). and Two Molecular Forms of MBL.” Chamow et al., “Immunoadhesins: principles and applications.” Witus et al., J. Am. Chem. Soc., 132: 16812-16817 (2010). “Iden Trends Biotechnol. 14(2):52-60 (1996). tification of Highly Reactive Sequences for PLP-Mediated Chang et al., “Crystallization and preliminary X-ray analysis of a Bioconjugation Using a Combinatorial Peptide Library.” trimeric form of human mannose binding protein.” J. Mol. Biol. Nadesalingam et al., “Mannose-binding lectin recognizes 241:125-127 (1994). peptidoglycan via the N-acetyl glucosamine moiety, and inhibits Foster “Immune evasion by Staphylococci.” Nature 3:948-958 ligand-induced proinflammatory effect and promotes chemokine (2005). production by macrophages.” Journal of Immunology 175: 1785 Hinton et al., “Engineered human IgG antibodies with longer serum 1794 (2005). half-lives in primates.” The Journal of Biological Chemistry Nakamura et al., “Characterization of the interaction between serum 279(8):6213-6216 (2004). mannan-binding protein and nucleic acid ligands.” Journal of Leu Idusogie et al., “Engineered antibodies with increased activity to kocyte Biology 86:737-748 (2009). recruit complement.” The Journal of Immunology 166 (4):2571-5 Ogden et al., “C1q and mannose binding lectin engagement of cell (2001). Surface calreticulin and CD91 initiates macropinocytosis and uptake Keen et al., “Interrelationship between pH and surface growth of of apoptotic cells.” J. Exp. Med. 194:781-795 (2001). nitrobacter.” Soil Biol. Biochem. 19(6):665-672 (1987). Stuart et al., “Mannose-binding lectin-deficient mice display defec Schmidt, "Fusion proteins as biopharmaceuticals—applications and tive apoptotic cell clearance but no autoimmune phenotype.” Jour challenges.” Current Opinion in Drug Discovery & Development nal of Immunology 174:3220-3226 (2005). Warwicket al., “Use of quantitative 16S ribosomal DNA detection 12(2):284-295 (2009). for diagnosis of central vascular catheter-associated bacterial infec Shields et al., “High resolution mapping of the binding site on tion.” Journal of Clinical Microbiology 42:1402-1408 (2004). human IgG1 for Fc gamma RI. Fc gamma RII, Fc gamma RIII, and Bossola et al., “Circulating bacterial-derived DNA fragments and FcRn and design of IgG1 variants with improved binding to the Fc markers of inflammation in chronic hemodialysis patients.” Clin J. gamma R.” The Journal of Biological Chemistry 276(9):6591-6604 Am. Soc. Nephrol. 4:379-385 (2009). (2001). Bayston et al., "Bacterial endotoxin and current concepts in the Steurer et al., “Ex vivo coating of islet cell allografts with murine diagnosis and treatment of endotoxaemia.” J. Med. Microbiol CTLA4/Fc promotes graft tolerance.” The Journal of Immunology 31:73-83 (1990). 155(3): 1165-1174 (1995). Michelow et al., “A novel L-ficolin?mannose-binding lectin chime Vaccaro et al., “Engineering the Fc region of immunoglobulin G to ric molecule with enhanced activity against Ebola virus.” Journal of modulate in vivo antibody levels.” Nature Biotechnology Biological Chemistry 285(32):24729-24739 (2010). 23(1): 1283-1288 (2005). Takahashi et al., “Mannose-binding lectin and its associated Wong et al., “Bioinspired self-repairing slippery Surfaces with proteases (MASPs) mediate coagulation and its deficiency is a risk pressure—stable omniphobicity.” Nature 477:443-447 (2011). factor in developing complications from infection, including dis Gouin et al., “Multimeric Lactoside “Click Clusters' as Tools to Seminated intravascular coagulation.” Immunobiology 216(1-2):96 Investigate the Effect of Linker Length in Specific Interactions with 102 (2011). Peanut Lectin, Galectin-1, and -3', ChemEBioChem 11(10): 1430 Invivo Gen Insight, “IgG-Fc Engineering for Therapeutic Use.” 1442 (2010). Apr./May 2006. Sibille et al., "Comparison of serological tests for the diagnosis of Product Data Sheet: Human Mannan Binding Lectin peptide (237 feline immunodeficiency virus infection of cats'. Veterinary Micro 248) (Carboxyterminal end) ab45655. biology 45(2-3): 259-267 1995). Loosdrecht et al., “Influence of interfaces on microbial activity.” Microbiological Reviews 54(1):75-87 (1990). * cited by examiner U.S. Patent Mar. 14, 2017 Sheet 1 of 40 US 9,593,160 B2

U.S. Patent Mar. 14, 2017. Sheet 2 of 40 US 9,593,160 B2

FIG 2 U.S. Patent Mar. 14, 2017. Sheet 3 of 40 US 9,593,160 B2

s

c: xes &S & ax. x w U.S. Patent Mar. 14, 2017. Sheet 4 of 40 US 9,593,160 B2

8

:

8. 88:

FIG. at U.S. Patent Mar. 14, 2017. Sheet 5 of 40 US 9,593,160 B2

KX 88 3 : 888 & 8, a & i88 8 88 { i ; iii. 888 8:8;

38

88 I.T.T. II 8. 8.2 . 8. 88: .8 8:E8888 gf8

F.G. 5 U.S. Patent Mar. 14, 2017. Sheet 6 of 40 US 9,593,160 B2

888: 888 . . .3388 i 8. 86. 88 888 58. 33.8 & 388 i.

i W. -- poro. 8 S8 g : 8 : 8 888 is . 8: 8:8; 8: 88: 8888

8; 8 888

FIG. A U.S. Patent Mar. 14, 2017. Sheet 7 of 40 US 9,593,160 B2

ii , $833 -Sisi. 8,

8,

8,

8. 8 8:8 8.88. i. 8. i SS S 3. * : * 3: FEERE80 SSSICY 38.88 six 8.38Y & E-REE, 38&E, i-EX.

FIG. E. U.S. Patent Mar. 14, 2017. Sheet 8 of 40 US 9,593,160 B2

its i is: : ; it i8 EFFES, BESE

8 38 - i. s: 6 xis 8.5

& 8.4 3.3 i. 8.

8.88 8S 8 : Yi 3ats, ai, is a EE SER; ; PiS.

FIG. 7 U.S. Patent Mar. 14, 2017. Sheet 9 of 40 US 9,593,160 B2

iS 8: 3 Siki is

8-3. sii is - - 8S

FIG. 8 U.S. Patent Mar. 14, 2017. Sheet 10 of 40 US 9,593,160 B2

S : i: 88

8 rew

8 s i --

. .

88: 888

FIG. 9 U.S. Patent Mar. 14, 2017. Sheet 11 of 40 US 9,593,160 B2

/

i88 $88.88 ((Y ) 3886:8:is Si:S 8888:8 &^

CAPTURE

DETECTION

888. 88: c8. i: 88 88: 8 88: 8. E88: 888. Sii 8:08.8883: 88: 88: 88:

y ~ F.G. 1 U.S. Patent Mar. 14, 2017. Sheet 12 of 40 US 9,593,160 B2

i88 t. iiii;S 83.

8 8.8 8,8. 8. 8.8 8.3 - 88 8. 83 8,8 8.8 U.S. Patent Mar. 14, 2017. Sheet 13 of 40 US 9,593,160 B2

is 60 ECOL xx. $3.88. “r-xx - - - - - , ...... ww.d.s. s. V. . . .& ------...... &

1,305.03 .00E04 1. GE-5 E-86 i. 8.

88: 8388

FIG, 12 U.S. Patent US 9,593,160 B2

U.S. Patent US 9,593,160 B2

38.833$$$3;&###

U.S. Patent Mar. 14, 2017. Sheet 16 of 40 US 9,593,160 B2

- Baa 8,888 3: 888 fitti

-- $83,88 - 8,888 88 383 8 888 Saifetis FG,

38, 8,888, 3.8gf8 .38g:

8,88g: 8.338gfa

FG. If U.S. Patent Mar. 14, 2017. Sheet 17 of 40 US 9,593,160 B2

3: SEREPRESS: 888 - 888

8 8 8 8 8 8

8 8

8 Sii. 8&S8 i:

8 8 8 8 xiii

i88 it 8

8 8

8 8 8

33i:

8 8 8 8 8

88: M 23

FIG. U.S. Patent Mar. 14, 2017. Sheet 18 of 40 US 9,593,160 B2

83-8 fe88 yesei 8EAS 8 is A. : :::: X 8, 8.3:

...s.-,-,-w

i a

ES * 8.8 8.8 8.

8, a

o : 888 i8 FIG. 8A

{X 88-3 88-88-fei.si. (BeS$ 8:3 EBA 8 gy. . ; : X 8.28 : . as- f s:

1:08 i.18:8 88 8 . . i: FIG. ISB U.S. Patent Mar. 14, 2017 Sheet 19 of 40 US 9,593,160 B2

U.S. Patent Mar. 14, 2017. Sheet 20 of 40 US 9,593,160 B2

kx fi: ;8:3, SSES g

it 8: 83 s 888 FIG. 2A

it - ; : iii. 3 is & S. asses 38 8.88;ingiitiis 3. -- KX & .388gis8S3 x 33 3 - 8:888 {} is:

*& '' '.v.r - s & ssix-g : s s W ...... :

e- 8-g 388-3 8- 88-3 8-8 SEE FIG. 20 U.S. Patent Mar. 14, 2017. Sheet 21 of 40 US 9,593,160 B2

{} R EPIRE 8 CS 88. i.S. atteis 888 x & 8

88.8 S.atists iii FIG. & A

iii. 3 & :: 8: 888

8 Sir & S is i. 3i88i8

ei FG, 2 B U.S. Patent Mar. 14, 2017. Sheet 22 of 40 US 9,593,160 B2

1 a capsults

E.8 s.

F. X. U.S. Patent Mar. 14, 2017 Sheet 23 of 40 US 9,593,160 B2

######## 83"?I),

3888S8 U.S. Patent Mar. 14, 2017 Sheet 24 of 40 US 9,593,160 B2

|34#623º

§§§3838?3334####8#######8########3

& ck

88: U.S. Patent Mar. 14, 2017. Sheet 25 of 40 US 9,593,160 B2

S. 8:888 :... ii. % 8&ERA 8: 3: EEE

Pas CRL i S8i 8 88.888;883 88: FIG. 25A

S. 88888

%. 888 ER 8:8; 9. E.I. is , i.

is 88. S88. 8 88.888 i: , 25 U.S. Patent Mar. 14, 2017 Sheet 26 of 40 US 9,593,160 B2

{{}{} 83% si??dneºs

#033 U.S. Patent Mar. 14, 2017 Sheet 27 of 40 US 9,593,160 B2

8193(91) ~~~~*~ U.S. Patent

$igi U.S. Patent Mar. 14, 2017 Sheet 29 of 40 US 9,593,160 B2

"?I),83 U.S. Patent Mar. 14, 2017 Sheet 30 of 40 US 9,593,160 B2

63™91),

§§§3 3#####3 $3&####3438}} umumama U.S. Patent Mar. 14, 2017. Sheet 31 of 40 US 9,593,160 B2

S88:

is si; ; ;

iii. 888 888

FIG. 3OA U.S. Patent Mar. 14, 2017 Sheet 32 of 40 US 9,593,160 B2

8108'91,1 U.S. Patent Mar. 14, 2017. Sheet 33 of 40 US 9,593,160 B2

U.S. Patent Mar. 14, 2017. Sheet 34 of 40 US 9,593,160 B2

As U.S. Patent Mar. 14, 2017 Sheet 35 of 40 US 9,593,160 B2

V38‘91). U.S. Patent Mar. 14, 2017 Sheet 36 of 40 US 9,593,160 B2

U.S. Patent Mar. 14, 2017. Sheet 37 of 40 US 9,593,160 B2

38: 88:

8 : Si 8-3, i-S8.

y3 is: 888

~ --~~ F.G. U.S. Patent Mar. 14, 2017. Sheet 38 of 40 US 9,593,160 B2

;: if

3888: 8: 8888: 888 : 8: -.

8 : 88: 8:3 88: 8 is 88:88-888 F.G. 4A :: 8888: 8888: ; : 388: 888: 8 {C} if:

8. a ar 3 :88 8 8

F.C. at U.S. Patent Mar. 14, 2017. Sheet 39 of 40 US 9,593,160 B2

s

8 :8: 8.88 s ki: FIG. 34Cs & iii. 8: 8 fit

* 8:8; 1888Sri

: 4- i-10 8. FIG. 34t U.S. Patent Mar. 14, 2017 Sheet 40 of 40 US 9,593,160 B2

t

s as s s a. & w w s s s: as: ex xces ex as se: s 3. & w sw: -w (.333i US 9,593,160 B2 1. 2 ENGINEERED MICROBE-TARGETING cal means (stain, culture, antigenemia or antigenuria, nucleic MOLECULES AND USES THEREOF acid detection) or pathognomonic signs of infection obtained by imaging or clinical examination. The infection can affect CROSS REFERENCE TO RELATED any organ system, but the more severe cases present as APPLICATIONS 5 septicemia (i.e., organisms, their metabolic end-products or toxins in the blood stream), bacteremia (i.e., bacteria in the This Application is a 35 U.S.C. S371 National Phase blood), toxemia (i.e., toxins in the blood), endotoxemia (i.e., Entry Application of International Application No. PCT/ endotoxin in the blood). Sepsis can also result from US2012/047201 filed Jul. 18, 2012, which designates the fungemia (i.e., fungi in the blood), Viremia (i.e., viruses or U.S., and which claims the benefit under 35 U.S.C S119(e) 10 virus particles in the blood), and parasitemia (i.e., helminthic of U.S. Provisional Application No. 61/508,957 filed Jul.18, or protozoan parasites in the blood). Thus, septicemia and 2011: 61/605,081 filed Feb. 29, 2012; and 61/605,052 filed septic shock (acute circulatory failure resulting from Septi Feb. 29, 2012, the contents of each of which are herein cemia often associated with multiple organ failure and a high incorporated by reference in their entirety. 15 mortality rate) may be caused by various microorganisms. There are three major types of sepsis characterized by the GOVERNMENT SUPPORT type of infecting organism. For example, gram-negative sepsis is the most frequently isolated (with a case fatality This invention was made with Government support under rate of about 35%). The majority of these are grant no. N66001-11-1-4180 awarded by DARPA. The caused by Escherichia coli, Klebsiella pneumoniae and Government has certain rights in the invention. 2O Pseudomonas aeruginosa. Gram-positive pathogens such as SEQUENCE LISTING the Staphylococci and Streptococci are the second major cause of sepsis. The third major group includes fungi, with The instant application contains a Sequence Listing which fungal infections causing a relatively small percentage of has been Submitted in ASCII format via EFS-Web and is 25 sepsis cases, but with a high mortality rate; these types of hereby incorporated by reference in its entirety. Said ASCII infections also have a higher incidence in immunocomprised copy, created on Jan. 15, 2014, is named 28671233.txt and patients. is 22,557 bytes in size. Some of these infections can be acquired in a hospital setting and can result from certain types of Surgery (e.g., TECHNICAL FIELD 30 abdominal procedures), immune Suppression due to cancer or transplantation therapy, immune deficiency diseases, and Described herein relates generally to molecules, products, exposure through intravenous catheters. Sepsis is also com kits and methods for detecting and/or removing microbes in monly caused by trauma, difficult newborn deliveries, and a sample or a target area, including bodily fluids such as intestinal torsion (especially in dogs and horses). Infections blood and tissues of a subject, food, water, and environmen- 35 in the lungs (pneumonia), bladder and kidneys (urinary tract tal Surfaces. infections), skin (cellulitis), abdomen (such as appendicitis), bone (osteomyelitis) and joints (arthritis) and other areas BACKGROUND (such as meningitis) can spread and also lead to sepsis. In Some circumstances, ingestion of microbe-contaminated Sepsis is a major cause of morbidity and mortality in 40 water, fluid or food, or contact with microbe-covered envi humans and other animals. In the United States, sepsis is the ronmental Surfaces can cause infections that lead to sepsis, second leading cause of death in intensive care units among and infection with food-borne and water-borne pathogens patients with non-traumatic illnesses. It is also the leading Such as Shigella spp., or certain serotypes of Escherichichia cause of death in young livestock, affecting 7.5-29% of coli (such as O157H7), Salmonella spp including Salmo neonatal calves, and is a common medical problem in 45 nella enterica serovar typhi or Listeria monocytogenes can neonatal foals. Despite the major advances of the past also lead to sepsis. several decades in the treatment of serious infections, the Many patients with septicemia or Suspected septicemia incidence and mortality due to sepsis continues to rise. exhibit a rapid decline over a 24-48 hour period. It has been Sepsis results from the systemic invasion of microorgan reported that patients with septic shock require adapted isms into blood and can present two distinct problems. First, 50 treatment in less than 6 hours in order to benefit from the growth of the microorganisms can directly damage antimicrobial therapy. Thus, rapid and reliable diagnostic tissues, organs, and vascular function. Second, toxic com and treatment methods are essential for effective patient ponents of the microorganisms can lead to rapid systemic care. Unfortunately, a confirmed diagnosis as to the type of inflammatory responses that can quickly damage vital infection, e.g., sepsis, traditionally requires microbiological organs and lead to circulatory collapse (i.e., septic shock) 55 analysis involving inoculation of blood cultures, incubation and, often times, death. for 18-24 hours, plating the causative microorganism on Sepsis is a systemic reaction defined by the American Solid media, another incubation period, and final identifica College of Chest Physicians and the Society of Critical Care tion 1-2 days later. Even with immediate and aggressive Medicine by a systemic inflammatory response (SIRS) in treatment, some patients can develop multiple organ dys response to a confirmed infectious process. SIRS is defined 60 function syndrome and eventually death. Hence, there by the presence of two or more of the following: altered remains a strong need for improved techniques for diagnosis body temperature (<36° C. or >38° C.), tachycardia (heart and treatment of patients with infectious diseases, blood rate-90/min), tachypnea (respiratory rate-20/min) or borne infections, sepsis, or systemic inflammatory response hypocapnia (PCO less than 4.3 kPa), leucopenia (white syndrome. The ability to rapidly detect infectious pathogens blood cells (WBCs)<4000 cells/mm or leucocytosis 65 in food, water, and/or environmental Surfaces would also (>12000 WBC/mm) or >10% band forms. The confirma have great value for preventing infections and sepsis in the tion of the infectious process is confirmed by microbiologi population. US 9,593,160 B2 3 4 SUMMARY for orienting the CRD away from the substrate; and (c) at least one linker between the CRD and the substrate-binding Embodiments described herein are based on, at least in domain. part, engineering a microbe-targeting molecule or a In some embodiments, the microbe-Surface binding microbe-binding molecule. For example, in one embodi domain comprises the full amino acid sequence of a carbo ment, a microbe-targeting molecule is engineered by fusing hydrate-binding protein. In some embodiments, the amino the carbohydrate recognition domain and neck region of a acid sequence of the carbohydrate-binding protein does not carbohydrate-binding protein (e.g., mannose-binding lectin) include a complement region. In some embodiments, the to the C-terminal of a Fc fragment of human IgG1. Such amino acid sequence of the carbohydrate-binding protein microbe-targeting molecules can be also modified to reduce 10 does not include a coagulation activation region. the complement activation and coagulation side effects In some embodiments of any aspects described herein, the which are present in the wild-type mannose-binding lectin, linker can comprise a portion of a Fc region of an immu and can complicate binding and detection. Further, the noglobulin, e.g., IgG1. In Such embodiments, the portion of microbe-targeting molecules described herein can be engi the Fc region can be linked, directly or indirectly, to N-ter neered, e.g., by inserting an AKT tripeptide to the N-termi 15 minal of the carbohydrate recognition domain. In some nal of the Fc fragment for site-specific biotinylation, such embodiments, the portion of the Fc region can be genetically that their carbohydrate recognition domains orient away modified, e.g., to increase half-life of the engineered mol from a substrate to which they attach, thus increasing the ecules, or modulate an immune response (e.g., antibody microbe-binding capacity. The microbe-targeting molecules dependent cell-mediated cytotoxicity and complement-de can be attached to various Substrates, e.g., a magnetic pendent cytotoxicity). microbead, in a multivalent oriented manner to form a In some embodiments of any aspects described herein, the microbe-targeting substrate. The term “microbead' as used Substrate-binding domain can comprise at least one oligo herein generally refers to a bead or a particle of any material peptide comprising an amino acid sequence of AKT. In other having a size of about 0.001 um to about 1000 um or about embodiments, the Substrate-binding domain can comprise a 0.001 um to about 100 um, or about 0.01 um to about 10 um. 25 biotin molecule. Depending on various applications, e.g., for In one embodiment, the microbead is a nanobead. The term use as a soluble protein in pharmaceutical compositions, the “nanobead” as used herein generally refers to a bead or Substrate-binding domain can become non-essential in some particle having a size ranging from about 1 nm to about 1000 embodiments of the engineered microbe-targeting mol nm, from about 10 nm to about 500 nm, from about 25 nm. ecules. Otherwise, the engineered microbe-targeting mol to about 300 nm, from about 40 nm to about 250 nm, or from 30 ecules can be used to coat various Substrates for a wide about 50 nm to about 200 nm. variety of applications. In some embodiments, the Substrate In some embodiments, the microbe-targeting molecules is a magnetic microbead, resulting in formation of a can be modified, e.g., to facilitate attachment of the microbe microbe-targeting magnetic microbead or opsonin. In some targeting molecules to a Substrate. For example, in one embodiments, the microbe-targeting magnetic microbead or embodiment, the microbe-targeting molecules can be bioti 35 opsonin can encompass a microbe-targeting nanobead. nylated, e.g., for attachment to an avidin- or avidin-like Not only can the microbe-targeting magnetic microbeads coated Substrate. Thus, the engineered microbe-targeting be used to remove microbes or pathogens in a sample, e.g., molecules described herein provide a valuable building blood and tissues, they can also be used to develop assays for block for various applications, e.g., diagnosis and/or treat detecting the presence or absence of, and/or differentiating ment of diseases caused by microbes or pathogens, removal 40 between, different microbes or pathogens. Accordingly, kits of microbes or pathogens from a sample, including bodily and assays for detecting the presence or absence of fluids and tissues of a Subject, foods, water, or an environ microbes, and/or differentiating between, different microbes mental Surface; and development of targeted drug delivery or pathogens in a test sample are also provided herein. In devices. Some embodiments, the kits comprise microbe-targeting Accordingly, provided herein is directed to an engineered 45 Substrates (e.g., but not limited to, one or more containers microbe-targeting molecule comprising: (a) at least one each containing a population of magnetic microbeads coated microbe Surface-binding domain; (b) a Substrate-binding with a plurality of the engineered microbe-targeting mol domain adapted for orienting the microbe Surface-binding ecules); and at least one reagent. In some embodiments, the domain away from the Substrate; and (c) at least one linker kits can further comprise one or more containers each between the microbe surface-binding domain and the sub 50 containing a population of detectable labels, wherein each of strate-binding domain. the detectable labels is conjugated to a molecule that binds In some embodiments, the microbe-Surface binding to the microbes or pathogens. Such kits can be used for domain can comprise a carbohydrate recognition domain analysis, e.g., by an enzyme-linked immunosorbent assay (CRD) or a fragment thereof. In some embodiments, the (ELISA), fluorescent linked immunosorbent assay (FLISA), microbe-Surface binding domain can further comprise a non 55 immunofluorescent microscopy, fluorescence in situ hybrid carbohydrate recognition domain or fragment thereof from ization (FISH), or any other radiological, chemical, enzy the carbohydrate-binding protein, e.g., a neck region of the matic or optical detection assays. In some embodiments, the carbohydrate-binding protein. As used herein, the term “non kits and assays described herein can be adapted for carbohydrate recognition domain refers to the portion or Susceptibility tests, e.g., to determine Susceptibility of a fragment of a carbohydrate-binding protein that does not 60 microbe in a test sample to one or more , regard directly bind with the microbe surface. less of whether the identity of the microbe is known or not. In some embodiments, the CRD or the carbohydrate Without limitations, in some embodiments, the engi binding protein can be derived from, e.g., mannose-binding neered microbe-targeting molecules can be formulated as an lectin. Hence, another aspect provided herein is directed to antibiotic or for use in various applications, e.g., an engineered mannose-binding lectin molecule comprising: 65 wound dressings, alone or in combination with other wound (a) at least one carbohydrate recognition domain (CRD) or dressing protocols, e.g., silver nanoparticles and other a fragment thereof; (b) a Substrate-binding domain adapted wound treatment. US 9,593,160 B2 5 6 BRIEF DESCRIPTION OF THE DRAWINGS 20), followed by FcMBL-based ELISA analysis. “2" run” corresponds to the “input' blood sample incubated with 20 FIGS. 1A-1C shows a general scheme of engineering one ul/ml MYONETM FCMBL microbeads for microbe capture or more embodiments of engineered microbe-targeting or (with mixing on a shaker 10'), followed by FcMBL-based microbe-binding molecules and microbe-targeting Sub 5 ELISA analysis. strates described herein. FIG. 1A is a diagrammatic view of FIG. 10 is a schematic diagram of an exemplary ELISA a native (wild-type) mannose-binding lectin (MBL). FIG. assay comprising engineered microbe-targeting magnetic 1B shows one or more embodiments of the engineered microbeads according to one or more embodiments. The microbe-targeting molecules or engineered-binding mol ELISA assay can be used for any diagnostic applications, ecules, e.g., engineered MBL molecules. FIG. 1C shows one 10 e.g., for sepsis tests. or more embodiments of the microbe-targeting or microbe FIG. 11 is a graph showing results of detecting C. albicans binding molecules conjugated to a Substrate, e.g., a magnetic in blood. Serial dilutions of C. albicans were spiked into microbead or nanobead, to form a microbe-targeting Sub blood, captured by AKT-FcMBL magnetic microbeads (1 Strate. um) and detected by an ELISA method using HRP-labeled FIG. 2 shows a crystal structure of a portion of a wild-type 15 FCMBL. MBL, which is the “neck and carbohydrate recognition FIG. 12 is a graph showing bacterial detection sensitivity domain (CRD) head.” The crystal structure depicts three of one or more embodiments of the FcMBL-based ELISA MBL heads, and calcium binding sites (Chang et al. (1994) assay. Serial dilutions of E. coli were spiked into a buffer, J Mol Biol. 241:125-7). captured by AKT-FcMBL magnetic microbeads (about 128 FIG. 3 is a schematic diagram showing an exemplary nm in size) and detected by an ELISA method using HRP Fc-X vector construct for one or more embodiments of the labeled FcMBL. In some embodiments, the limit of detec engineered microbe-targeting or microbe-binding molecules tion (LOD) of the FcMBL-based ELISA colorimetric assay described herein. is about or below 160 E. coli bacteria. FIG. 4 shows a Western blot image indicating expression FIG. 13 is a schematic diagram showing one or more of the purified wild-type MBL (MBLWT) proteins and one 25 embodiments of a dipstick assay for microbial detection. or more embodiments of the engineered microbe-targeting The FcMBL can be attached to a membrane (for example or microbe-binding molecules described herein (FcMBL.81: Biodyne membrane). The membrane can be mixed with a SEQ ID NO. 6). test sample (e.g., blood sample), washed, incubated with a FIG. 5 shows the mannan-binding results of various desired detecting protein (e.g., AP-labeled FcMBL or spe embodiments of the engineered microbe-targeting or 30 cific antibody for certain microbes, e.g., bacteria or fungus), microbe-binding molecules described herein in the presence washed and added with a readout reagent for colorimetric or absence of calcium ions. A chelating agent (e.g., EDTA) development. The dipstickassay can be performed manually can be added to the sample to remove calcium ions. or modified for automation. FIGS. 6A and 6B are bar graphs showing the results of FIG. 14 is a schematic diagram showing one or more capturing microbes, e.g., C. albicans, with one or more 35 embodiments of an ELISA-based test for microbial detec embodiments of the microbe-targeting Substrates (e.g., AKT tion. A test sample (e.g., blood sample) can be added into a FcMBL.81 conjugated to magnetic microbeads having a size single tube (e.g., a blood collection container Such as EDTA of about 1 um at various microbe densities. FIG. 6A shows VACUTAINER(R) containing lyophilized FcMBL magnetic the percentage of microbes bound to microbe-targeting microbeads or FcMBL-coated magnetic microbeads. An substrates and controls at a low microbe density (e.g., 1500 40 exemplary protocol for microbial capture and detection is C. albicans cells). FIG. 6B shows the amount of unbound described in Example 10. The ELISA-based test can be microbes remained in the microbe samples after treatment performed manually or modified for automation. In some with different magnetic microbeads (including the engi embodiments, the single-tube based ELISA assay can be neered microbe-targeting magnetic microbeads) when the used to detect microbes or pathogens such as S. aureus and microbe is present at a much higher microbe density (e.g., 45 E. coli. greater than 10 cells). FIG. 15 is an image showing direct detection of bacteria FIG. 7 shows the size effect of one or more embodiments on a membrane by AP-labeled FcMBL. Serial dilutions of E. of the microbe-targeting Substrates (e.g., microbe-targeting coli and S. aureus (10' to 10) were spotted directly onto magnetic microbeads such as AKT-FcMBL.81 magnetic a Biodyne membrane, blocked for about 30 mins in 1% microbeads, wherein the size of the microbeads were varied 50 casein, washed twice in TBST containing Ca" (5 mM), from about 100 nm to about 1000 nm diameter) on the incubated with AP-labeled FcMBL (1:10,000 dilution) in efficiency of capturing microbes or pathogens, e.g., Can 3% BSA 1XTBST containing Ca" (for about 20 min), dida. washed twice in TBST containing Ca" (5 mM) and once in FIG. 8 shows the amount of unbound microbes remained TBS containing Ca" (5 mM), and reacted with BCIP/NBT in the microbe samples after treatment with different mag 55 for about 20 mins to develop a colorimetric readout. In this netic microbeads (including the engineered microbe-target example, maximum dilution allowed for detection of both ing magnetic microbeads) when the microbes (e.g., Can species was 10' after 30 min development (corresponding dida) are growing in a log phase VS. in a saturated phase. to detection of 130 E. coli and 343 S. aureus cells). FIG. 9 shows the depletion of microbe/microbial matter FIG. 16 is an image showing capture and detection of S. from a blood sample from a human donor as measured using 60 aureus by dot blot using a membrane coupled with FcMBL. FcMBL, ELISA. In the figure, “input corresponds to undi Dilutions of S. aureus (10° and 10) were captured by luted EDTA donor blood spiked with S. aureus or E. coli, FcMBL immobilized on a Biodyne membrane. For example, and supplemented with Ca" (final Ca"=5 mM) and 5uL of two indicated concentrations of FcMBL were spotted heparin (4 mg/ml), before addition of any FcMBL micro onto a Biodyne membrane, allowed to dry, blocked in 1% beads. "1 run” corresponds to the "input' blood sample 65 casein, and washed twice in TBST containing Ca" (5 mM). incubated with 20 ul/mL MYONETM FCMBL microbeads Each FcMBL concentration was assessed for capture (~10 for microbe capture (with mixing on a HULAMIXERTM min) of serial dilutions of S. aureus, washed, and detected US 9,593,160 B2 7 8 with 1:10,000 dilution of AP-labeled FcMBL in 3% BSA FIG. 27 is a bar graph comparing different microbial or 1XTBST containing Ca" (~20 min). Excess AP-labeled pathogenic species captured on FcMBL-coated Substrates FcMBL was removed by washes (e.g., washing three times (e.g., magnetic microbeads) in the presence or absence of a with TBST containing Ca" (5 mM) and once with TBS chelating agent (e.g., EDTA) and various Ca" concentra containing Ca" (5 mM)). Colorimetric detection was devel tions. oped with BCIP/NBT for -20 min. FIG. 28 is an image showing colorimetric outcomes of the FIG. 17 is a schematic of an exemplary microbial detec tube-based ELISA assay for S. aureus and E. coli binding to tion process or diagnosis process. FcMBL-coated Substrates (e.g., magnetic microbeads) in the FIGS. 18A and 18B are line graphs showing ELISA of E. presence or absence of a chelating agent (e.g., EDTA) and/or 10 a low pH buffer. coli on two different FcMBL, microbead formats. FIG. 18A FIG. 29 is an image showing dot blot determination of E. corresponds to FcMBL directly coupled to MYONETM Tosyl coli and S. aureus with or without EDTA in the capture activated beads and FIG. 18B corresponds to biotinylated and/or wash buffer. AKT-FcMBL coupled to Streptavidin MYONETMT1 micro FIGS. 30A-30B are images showing binding of one or beads (~1000 nm diameter). Three different dilutions of an 15 more embodiments of microbe-targeting Substrates to E. coli overnight culture were captured on FcMBL micro microbial matter, including live microbes and/or fragments beads, washed with one of four elution buffers and then run or matter derived from microbes. FIG. 30A shows that through one or more embodiments of the ELISA protocol microbial outgrowth is observed when one or more microbe described herein. A decrease in signal corresponds to fewer targeting Substrates (e.g., FcMBL-coated fluorescent micro E. coli bound to the microbeads prior to the ELISA detec beads) bind(s) to at least one live microbe, e.g., E. coli. FIG. tion. 30B is a set of fluorescent images showing that FcMBL FIG. 19 is an image showing plating out of equal titers of coated fluorescent microbeads bind to microbial matter (left S. aureus either mixed with FcMBL microbeads or control panel) including live microbes (indicated by the middle without FcMBL microbeads. panel) and fragments or matter derived from microbes. The FIGS. 20A and 20B are line graphs showing capture 25 right panel is an overlay of the first two fluorescent images efficiency of engineered microbe-targeting or microbe-bind in addition to a bright-field image. ing molecules (e.g., FcMBL) in clinical isolates of different FIGS. 31A-31B are images showing capture of microbes microbial species. FIG. 20A shows data for capture effi or fragments thereof on one or more embodiments of ciency of FcMBL in the clinical isolates of S. aureus and microbe-targeting Substrates from fluid samples, followed methicillin-resistant S. aureus (MRSA). FIG. 20B shows 30 by antibody characterization. FIG. 31A shows capture of E. data for capture efficiency of FcMBL in the clinical isolates coli or fragments thereof on FcMBL-coated microbeads of S. aureus, MRSA, N. meningitidis, and P. aeroginosa. (e.g., magnetic or fluorescent microbeads) from heparinized FIGS. 21A-21B are line graphs showing capture effi blood, followed by incubation with an antibody against E. ciency of engineered microbe-targeting or microbe-binding coli lipopolysaccharide lipid A (anti-LPS lipid A antibody. molecules (e.g., FcMBL) in clinical isolates obtained from 35 FIG. 31B shows capture of E. coli or fragments thereof on different types of fluids. FIGS. 21A and 21B shows data for FcMBL-coated microbeads (e.g., magnetic or fluorescent capture efficiency of FcMBL in the clinical isolates of S. microbeads) from blood containing EDTA anticoagulation aureus and E. coli, respectively, obtained from other body agent, followed by incubation with an antibody against E. fluids, e.g., urine, cerebrospinal fluid (CSF), and sputum. coli lipopolysaccharide lipid A (anti-LPS lipid A antibody). FIG. 22 is a schematic diagram showing mechanism by 40 Both FIGS. 31A-31B show that the anti-LPS lipid A anti which S. aureus avoids opsonophagocytosis. See additional body does not bind to FcMBL-coated microbeads in the details in Fraser T., Nature Reviews Microbiology 2005: absence of E. coli or fragments thereof. 3(12):948-58. FIGS. 32A-32B are images showing capture of microbes FIG. 23 is a bar graph showing detection signals of on one or more embodiments of microbe-targeting Sub various concentrations of S. aureus captured by AKT-Fc 45 strates from samples of a rat sepsis model, followed by MBL 1 uM magnetic microbeads and detected by FcMBL antibody characterization. FIG. 32A shows capture of HRP ELISA. Sensitivity of this embodiment of the assay microbes or fragments thereof on FcMBL-coated micro was about 149 CFU/mL. beads (e.g., magnetic or fluorescent microbeads) from rat FIG. 24 is a bar graph showing elution of S. aureus and blood (upper panel) or pleural (lower panel) fluids after E. coli bacteria bound onto FcMBL-coated substrates (e.g., 50 24-hr infection, followed by incubation with an anti-LPS magnetic microbeads) with different treatments, including lipid A antibody. FIG. 32B shows capture of microbes or chelation, pH and salt washes. fragments thereof on FcMBL-coated microbeads (e.g., mag FIGS. 25A and 25B are bar graphs showing elution of E. netic or fluorescent microbeads) from rat blood (upper coli and S. aureus off FcMBL-coated Substrates (e.g., mag panel) or pleural (lower panel) fluids after 72-hr infection, netic microbeads) using chelators. FIG. 25A shows the 55 followed by incubation with an anti-LPS lipid A antibody. results in OD450 and FIG. 25B shows the results as a FIG. 33 is a set of images showing the use of specific percent of bound bacteria remained on the FcMBL-coated antibodies to microbes to allow further discrimination or substrates after treatment. identification of samples that indicate positive signals with FIGS. 26A and 26B show results of tube-based ELISA for one or more embodiments of microbe-targeting Substrates. S. aureus and E. coli binding to FcMBL-coated substrates 60 De-identified clinical blood samples were screened by (e.g., magnetic microbeads) in the presence of a chelating FcMBL, ELISA described herein and the captured microbial agent (e.g., EDTA). FIG. 26A is an image showing colori matters (including intact cells and fragments thereof) on the metric outcomes of the tube-based ELISA for S. aureus and FcMBL-coated microbeads were further screened by using E. coli binding to FcMBL-coated Substrates (e.g., magnetic an anti-LPS lipid A antibody. The top panel indicates that no microbeads) in the presence or absence of a chelating agent 65 detection of anti-LPS lipid A antibody signal was observed (e.g., EDTA). FIG. 26B is a bar graph showing quantitative in clinical samples with Substantially negative or negligible measurement of the color developed in FIG. 26A. signal from FcMBL, ELISA, indicative of no microbial US 9,593,160 B2 10 infection detected in the clinical samples. The middle panel tation as to the contents of these documents is based on the indicates that the microbial matter producing positive signal information available to the applicants and does not consti (OD=-1.69) in FcMBL ELISA bound to anti-LPS lipid A tute any admission as to the correctness of the dates or antibody, which indicates that the microbial matter could be contents of these documents. derived from E. coli, and that the corresponding clinical Unless defined otherwise, all technical and scientific samples had a gram-negative infection (e.g., E. coli infec terms used herein have the same meaning as those com tion). In contrast, the bottom panel indicates that the micro monly understood to one of ordinary skill in the art to which bial matter producing positive signal (OD->3.9) did not bind this invention pertains. Although any known methods, to anti-LPS lipid A antibody, which indicates that the micro devices, and materials may be used in the practice or testing bial matter could be derived from microbes other than E. 10 of the invention, the methods, devices, and materials in this coli, e.g., when the clinical samples were infected with a regard are described herein. gram-positive microbe. Described herein are engineered microbe-targeting or FIGS. 34A-34D are data graphs showing that use of microbe-binding molecules, compositions comprising the FcMBL magnetic microbeads is a more sensitive and reli same, processes or assays, and kits for separating microbes able measure of blood-borne pathogens (including live and 15 from a test sample in vivo, in situ or in vitro, and/or detecting non-viable pathogens Such as dead pathogens and endotox the presence or absence of the microbes in the test sample. ins) than conventional blood cultures. FIG. 34A is a bar The engineered microbe-targeting or microbe-binding mol graph showing results of anaerobe cultures at Day 4 of blood ecules can bind or capture at least one microbe, e.g., an collected from five rats developed with intra-abdominal intact microbe, and/or “microbial matter.” The term “micro abscesses. FIG. 34B is a plot comparing the microbe detec bial matter as used herein refers to any matter or component tion results based on colorimetric ELISA using FcMBL that is derived, originated or secreted from a microbe. For magnetic microbeads and conventional blood cultures and example, microbial matter or a component derived or their correlations with morbidity of the rats. FIG. 34C is a secreted from a microbe that can bind to an engineered line graph showing correlation of pathogen load determined microbe-targeting or microbe-binding molecule can include, by the ELISA using FcMBL magnetic microbeads with 25 but are not limited to, a cell wall component, an outer morbidity ranking. FIG. 34D is a bar graph comparing the membrane, a plasma membrane, a ribosome, a microbial microbe detection results based on colorimetric ELISA capsule, a pili or flagella, any fragments of the aforemen using FcMBL magnetic microbeads and conventional blood tioned microbial components, any nucleic acid (e.g., DNA, cultures in a separate experiment. including 16S ribosomal DNA, and RNA) derived from a FIG. 35 is a bar graph showing percentages of microbe 30 microbe, and microbial endotoxin (e.g., lipopolysaccharide). depletion by one or more embodiments of the microbe In addition, microbial matter can encompass non-viable targeting magnetic microbeads. FcMBL-coated magnetic microbial matter that can cause an adverse effect (e.g., microbeads of different sizes (~1 um, ~128 nm, and ~50 nm) toxicity) to a host or an environment. were used to capture E. coli and S. aureus that were initially In accordance with various embodiments described spiked into a buffered solution. The microbe-bound FcMBL 35 herein, the engineered microbe-targeting molecules or coated magnetic microbeads were then removed from the microbe-binding molecules comprise a microbe Surface buffered solution. After removal of the magnetic micro binding domain (e.g., a carbohydrate recognition domain), beads, the buffered solution was used for inoculation on LB directly or indirectly, conjugated to a linker (e.g., a Fc plates to determine the level of microbe depletion by fragment), which can further comprise a substrate-binding FcMBL-coated magnetic microbeads of different sizes. 40 domain for immobilization. Thus, the engineered microbe targeting molecules or microbe-binding molecules described DETAILED DESCRIPTION OF THE herein can be used as Soluble proteins, e.g., in therapeutic INVENTION compositions, or be immobilized to a Substrate for various applications ranging from diagnosis and/or treatment of a It should be understood that this invention is not limited 45 microbial infection or disease, to microbe-clearing compo to the particular methodology, protocols, and reagents, etc., sitions or devices, to drug delivery. described herein and as Such may vary. The terminology In one aspect, provided herein is an engineered microbe used herein is for the purpose of describing particular targeting molecule (or an engineered microbe-binding mol embodiments only, and is not intended to limit the scope of ecule) comprising at least one microbe Surface-binding the present invention, which is defined solely by the claims. 50 domain, a Substrate-binding domain adapted for orienting As used herein and in the claims, the singular forms the carbohydrate recognition domain away from the Sub include the plural reference and vice versa unless the context strate, and at least one linker between the microbe surface clearly indicates otherwise. Other than in the operating binding domain and the Substrate-binding domain. In some examples, or where otherwise indicated, all numbers embodiments, the microbe Surface-binding domain can expressing quantities of ingredients or reaction conditions 55 comprise a carbohydrate recognition domain or a fragment used herein should be understood as modified in all thereof. In some embodiments, the microbe Surface-binding instances by the term “about.” domain can further comprise at least a portion of mannose All patents and other publications identified are expressly binding lectin (MBL). Accordingly, another aspect provided incorporated herein by reference for the purpose of describ herein is an engineered MBL molecule comprising at least ing and disclosing, for example, the methodologies 60 a fragment of a carbohydrate recognition domain derived described in Such publications that might be used in con from MBL: a substrate-binding domain adapted for orient nection with the present invention. These publications are ing the carbohydrate domain away from the Substrate; and at provided solely for their disclosure prior to the filing date of least one linker between the fragment of the MBL carbo the present application. Nothing in this regard should be hydrate recognition domain and the Substrate-binding construed as an admission that the inventors are not entitled 65 domain. The terms “microbe-binding molecule(s) and to antedate such disclosure by virtue of prior invention or for “microbe-targeting molecule(s) are used interchangeably any other reason. All Statements as to the date or represen herein. US 9,593,160 B2 11 12 In some embodiments of any aspects described herein, the For example, an interspecies hybrid microbe-targeting mol Substrate-binding domain adapted for orienting the carbo ecule can contain a linker, e.g., a peptide linker, from a hydrate recognition domain away from the Substrate is not murine species, and a human sequence from a carbohydrate always necessary and thus can be excluded under certain recognition domain protein, provided that they do not pro circumstances, e.g., using the engineered microbe-targeting vide unacceptable levels of deleterious effects. The engi molecules in a soluble format, e.g., for therapeutic purposes. neered microbe-targeting molecules described herein can Further, it should be noted that the engineered microbe also include those that are made entirely from murine binding molecules excluding the Substrate-binding domain derived sequences or fully human. adapted for orienting the carbohydrate recognition domain Microbe Surface-Binding Domain and Carbohydrate Rec away from the Substrate does not necessarily mean that the 10 ognition Domain engineered microbe-binding molecules cannot bind to a As disclosed herein, an engineered microbe-targeting Substrate Surface. In some embodiments, the engineered molecule can comprise at least one microbe Surface-binding microbe-binding molecules excluding the Substrate-binding domain, including at least two, at least three, at least four, at domain adapted for orienting the carbohydrate recognition least five, at least six, at least seven, at least eight, at least domain away from the substrate can still bind to a substrate 15 nine, at least ten or more microbe Surface-binding domains. surface, but the orientation of the carbohydrate recognition The term “microbe surface-binding domain” as used herein domain relative to the Substrate Surface can be random. refers to any molecule or a fragment thereof that can In some embodiments of any aspects described herein, the specifically bind to the Surface of a microbe or pathogen, engineered microbe-targeting molecule can further comprise e.g., any component present on a surface of a microbe or a detectable label, e.g., to facilitate detection of the presence pathogen, and/or any microbial matter, e.g., any matter or or absence of a microbe and/or microbial matter. Detectable component/fragment that is derived, originated or secreted labels suitable for conjugation to some embodiments of the from a microbe. Molecules that can be used in the microbe engineered microbe-targeting molecule can include any Surface-binding domain can include, for example, but are composition detectable by spectroscopic, photochemical, not limited to, peptides, polypeptides, proteins, peptidomi biochemical, immunochemical, electrical, magnetic, optical 25 metics, antibodies, antibody fragments (e.g., antigen binding or chemical means, as well as any examples of detectable fragments of antibodies), carbohydrate-binding protein, e.g., labels described herein and any equivalent thereof. In some a lectin, glycoproteins, glycoprotein-binding molecules, embodiments, the detectable labels also encompass any amino acids, carbohydrates (including mono-, di-, tri- and imaging agent (e.g., but not limited to, a bubble, a liposome, poly-saccharides), lipids, steroids, hormones, lipid-binding a sphere, a contrast agent, or any detectable label described 30 molecules, cofactors, nucleosides, nucleotides, nucleic acids herein) that can facilitate imaging or visualization of a tissue (e.g., DNA or RNA, analogues and derivatives of nucleic or an organ in a subject, e.g., for diagnosis of an infection. acids, or aptamers), peptidoglycan, lipopolysaccharide, In some embodiments, the detectable label conjugated to Small molecules, and any combinations thereof. In some the engineered microbe-targeting molecule can include an embodiments, the microbe Surface-binding domain can enzyme of horseradish peroxidase (HRP), alkaline phos 35 comprise a carbohydrate recognition domain or a fragment phatase (AP), or any combinations thereof. Conjugation of thereof. In some embodiments, a microbe Surface-binding the detectable label (e.g., HRP or AP) to any proteins and domain can comprise a peptidomimetic that mimics any antibodies are known in the art. In one embodiment, molecule or a fragment thereof that can specifically bind to FcMBL-HRP or FcMBL-AP construct is generated using the Surface of a microbe or pathogen, and/or any microbial any art-recognized methods for direct coupling HRP or AP 40 matter. For example, a microbe Surface-binding domain can to FCMBL. comprise a peptidomimetic that mimics any carbohydrate In some embodiments, the detectable label conjugated to recognition domain or a fragment thereof, e.g., carbohydrate the engineered microbe-targeting molecule can include a recognition domain of MBL or a fragment thereof, or any microbial enzyme Substrate conjugated to a detectable agent. carbohydrate recognition domain that is known in the art or For example, the detectable agent can be any moiety that, 45 a fragment thereof. In some embodiments, the microbe when cleaved from a microbial enzyme substrate by the Surface binding domain comprises the full amino acid enzyme possessed or secreted by the microbe, forms a sequence of a carbohydrate-binding protein. detectable moiety (e.g., a light-emitting signal), but that is In some embodiments, the microbe Surface-binding not detectable in its conjugated State. The microbial enzyme domain can have an amino acid sequence of about 10 to Substrate is a Substrate specific for one or more types of 50 about 300 amino acid residues, or about 50 to about 150 microbes to be detected, and it can be selected depending amino acid residues. In some embodiments, the microbe upon what enzymes the microbe possesses or secretes. See, Surface-binding domain can have an amino acid sequence of e.g., International Patent Application: WO 2011/103144 for at least about 5, at least about 10, at least about 15, at least the use of such detectable label in detection of microbes, the about 20, at least about 30, at least about 40, at least about content of which is incorporated herein by reference. 55 50, at least about 60, at least about 70, at least about 80, at General methods of preparing any embodiments of the least about 90, at least about 100 amino acid residues or engineered microbe-targeting molecules are known in the art more. For any known sequences of microbe Surface-binding (Ashkenazi, A. and S. M. Chamow (1997), “Immunoad molecules, one of skill in the art can determine the optimum hesins as research tools and therapeutic agents. Curr. Opin. length of amino acid sequence for the microbe Surface Immunol. 9(2): 195-200, Chamow, S. M. and A. Ashkenazi 60 binding domain. (1996). “Immunoadhesins: principles and applications.” In some embodiments, the microbe Surface-binding Trends Biotechnol. 14(2):52-60). In one example, an engi domain can comprise an opsonin or a fragment thereof. The neered microbe-targeting molecule can be made by cloning term “opsonin’ as used herein refers to naturally-occurring into an expression vector Such as Fc-X vector as discussed and synthetic molecules which are capable of binding to or in Lo et al. (1998) 11:495 and Example 1. 65 attaching to the Surface of a microbe or a pathogen, of acting The engineered microbe-targeting molecules can contain as binding enhancers for a process of phagocytosis. sequences from the same species or from different species. Examples of opsonins which can be used in the engineered US 9,593,160 B2 13 14 molecules described herein include, but are not limited to, mannose-containing glycoproteins on their envelopes and/or vitronectin, fibronectin, complement components such as function as a receptor for several viruses such as HIV and Cld (including any of its component polypeptide chains A, Hepatitis C. B and C), complement fragments such as C3d, C3b and C4b, Collectins are soluble pattern recognition receptors mannose-binding protein, conglutinin, Surfactant proteins. A (PRRS) belonging to the Superfamily of collagen containing and D, C-reactive protein (CRP), alpha2-macroglobulin, and C-type lectins. Exemplary collectins include, without limi immunoglobulins, for example, the Fc portion of an immu tations, mannose-binding lectin (MBL) (also known as noglobulin. mannan-binding lectin, mannan-binding protein, or man In some embodiments, the microbe Surface-binding nose-binding protein), Surfactant protein A (SP-A), Surfac 10 tant protein D (SP-D), collectin liver 1 (CL-L1), collectin domain can comprise a carbohydrate recognition domain. In placenta 1 (CL-P1), conglutinin, collectin of 43 kDa (CL Some embodiments, the microbe Surface-binding domain 43), collectin of 46 kDa (CL-46), and a fragment thereof. can further comprise at least a portion of a carbohydrate Mannose-binding lectin (MBL), also known as mannose binding protein or a portion thereof. In some embodiments, binding protein (MBP), or mannan-binding lectin or man the portion of the carbohydrate-binding proteins can activate 15 nan-binding protein, is a calcium-dependent serum protein the complement system. In alternative embodiments, the that can play a role in the innate immune response by portion of the carbohydrate-binding protein cannot activate binding to carbohydrates on the Surface of a wide range of the complement system. In some embodiments, the portion microbes or pathogens (viruses, bacteria, fungi, protozoa) of the carbohydrate-binding protein can be selected or where it can activate the complement system. MBL can also configured such that it cannot activate the complement serve as a direct opsonin and mediate binding and uptake of system, e.g., via modification. Examples of carbohydrate pathogens by tagging the Surface of a pathogen to facilitate binding proteins include, but are not limited to, lectin, recognition and ingestion by phagocytes. collectin, ficolin, mannose-binding lectin (MBL), maltose MBL is a member of the collectin family of proteins. A binding protein, arabinose-binding protein, and glucose native MBL is a multimeric structure (e.g., about 650 kDa) binding protein. Additional carbohydrate-binding proteins 25 composed of subunits, each of which contains three identical that can be included in the microbe Surface-binding domain polypeptide chains (FIG. 1A). Each MBL polypeptide chain described herein can include, but is not limited to, lectins or (containing 248 amino acid residues in length with a signal agglutinins that are derived from a plant, e.g., Galanthus sequence: SEQ ID NO. 1) comprises a N-terminal cysteine inivalis agglutinin (GNA) from the Galanthus (Snowdrop) rich region, a collagen-like region, a neck region, and a plant, and peanut lectin. In some embodiments, pentraxin 30 carbohydrate recognition domain (CRD). The sequence of each region has been identified and is well known in the art. family members, e.g., C-reactive protein, can also be used as SEQID NO. 2 shows a full-length amino acid sequence of a carbohydrate-binding protein. Pentraxin family members MBL without a signal sequence. can generally bind capsulated microbes. The carbohydrate The Surface or carbohydrate recognition function of a binding proteins can be wild-type, recombinant or a fusion 35 native MBL is mediated by clusters of three C-type carbo protein. The respective carbohydrate recognition domains hydrate-recognition domains (CRDs) held together by for Such carbohydrate-binding proteins are known in the art, coiled-coils of a-helices. The N-terminal portion collagen and can be modified for various embodiments of the engi like domain is composed of Gly-X-Y triplets. The short neered microbe-targeting molecules described herein. In N-terminal domain contains several cysteine residues that Some embodiments, peptidomimetics or any structural mim 40 form interchain disulfide bonds. Serum MBLs assemble into ics mimicking a microbe Surface-binding domain (e.g., a larger forms containing 2-4 trimeric subunits in rodents and carbohydrate recognition domain or a fragment thereof) and as many as six subunits in humans. All three oligomeric capable of binding to a microbe Surface can also be used as forms of rat serum MBP, designated MBPA, can fix comple a microbe Surface-binding domain described herein. ment, although the larger oligomers have higher specific The term “lectin’ as used herein refers to any molecules 45 activity. Many species express a second form of MBP. In including proteins, natural or genetically modified (e.g., rats, the second form, MBP-C, is found in the liver. MBP-C recombinant), that interact specifically with Saccharides does not form higher oligomers beyond the simple Subunit (e.g., carbohydrates). The term “lectin’ as used herein can that contains three polypeptides. also refer to lectins derived from any species, including, but When a native MBL interacts with carbohydrates on the not limited to, plants, animals, insects and microorganisms, 50 Surface of microbes or pathogens, e.g., calcium-dependent having a desired carbohydrate binding specificity. Examples binding to the carbohydrates mannose, N-acetylglu of plant lectins include, but are not limited to, the Legumi cosamine, and/or fucose, it can form the pathogen recogni nosae lectin family, such as ConA, soybean agglutinin, tion component of the lectin pathway of complement acti peanut lectin, lentil lectin, and Galanthus nivalis agglutinin vation. The MBL binds to surface arrays containing repeated (GNA) from the Galanthus (snowdrop) plant. Other 55 mannose or N-acetylglucosamine residues. It circulates as a examples of plant lectins are the Gramineae and Solanaceae complex with one or more MBP-associated serine proteases families of lectins. Examples of animal lectins include, but (MASPs) that autoactivate when the complex binds to an are not limited to, any known lectin of the major groups appropriate surface. The MBL and associated MASP pro S-type lectins, C-type lectins, P-type lectins, and I-type teins can activate C2/C4 convertase leading to the deposition lectins, and galectins. In some embodiments, the carbohy 60 of C4 on the pathogen Surface and opSonization for phago drate recognition domain can be derived from a C-type cytosis. The native MBL can also activate coagulation lectin, or a fragment thereof. C-type lectin can include any function through MASP proteins. carbohydrate-binding protein that requires calcium for bind While native MBL can detect microbes or pathogens and ing. In some embodiments, the C-type lectin can include, but act as opsonins for tagging the microbes for phagocytosis, are not limited to, collectin, DC-SIGN, and fragments 65 native MBLs may not be desirable for use in treatment of thereof. Without wishing to be bound by theory, DC-SIGN microbe-induced inflammatory diseases or infections, e.g., can generally bind various microbes by recognizing high sepsis, because native MBLS can activate complement sys US 9,593,160 B2 15 16 tem and induce an inflammatory response. Provided herein In some embodiments, the nucleic acid encodes a carbo is an engineered MBL molecule that binds to microbes or hydrate recognition domain having greater than 50% homol pathogens, comprising at least one carbohydrate recognition ogy, including greater than 60%, greater than 70%, greater domain or a fragment thereof, e.g., derived from MBL. In than 80%, greater than 90% homology or higher, to a some embodiments, the engineered MBL molecule can fragment of at least 50, at least 60, at least 70, at least 80, at comprises at least two, at least three or at least four carbo least 90, at least 100, at least 150 contiguous amino acids or hydrate recognition domains or a fragment thereof. In some more, of any known carbohydrate-binding molecules (e.g., embodiments, the engineered MBL molecules do not acti mannose-binding lectins). vate complement system or coagulation side effects that are The term “carbohydrate recognition domain as used present in a native MBL. Such embodiments can be used as 10 herein refers to a region, at least a portion of which, can bind dominant-negative inhibitors of downstream responses in to carbohydrates on a Surface of microbes or pathogens. For Vivo or as microbe-binding proteins that do not induce example, as shown in FIG. 1B, the carbohydrate recognition coagulation or complement fixation in vitro. For example, domain, in Some embodiments, can encompass MBL CRD the engineered MBL molecules that do not have complement 102. However, in some embodiments, the carbohydrate fixation and/or coagulation domains can act as a dominant 15 recognition domain can be also construed to encompass a negative protein in terms of activating cytokine and/or neck region 104 in addition to MBL CRD 102. In some inflammatory cascades, and thus reduce system inflamma embodiments, the carbohydrate recognition domain can tory syndrome and/or sepsis symptoms. comprise at least about 50% of its domain, including at least FIG. 1B shows a diagrammatic view of a dimeric engi about 60%, at least about 70%, at least about 80%, at least neered MBL molecule 100 according to one or more about 90% or higher, capable of binding to carbohydrates on embodiments of the engineered MBL molecules described a microbe surface. In some embodiments, 100% of the herein. The dimeric molecule 100 comprises at least two carbohydrate recognition domain can be used to bind to carbohydrate recognition domains 102 (e.g., MBL CRD) microbes or pathogens. In other embodiments, the carbohy connected, directly or indirectly, to a linker, e.g., a Fc region drate recognition domain can comprise additional regions 106. The N-terminal of the Fc region 106 can further 25 that are not capable of carbohydrate binding, but can have comprise an oligopeptide 108, e.g., comprising an amino other characteristics or perform other functions, e.g., to acid sequence AKT. In some embodiments, the carbohydrate provide flexibility to the carbohydrate recognition domain recognition domains 102 can further comprise neck regions when interacting with microbes or pathogens. 104 such as MBL neck to provide flexibility of the CRD Accordingly, in some embodiments, the carbohydrate interacting with microbes. 30 recognition domain can further comprise a neck region of The full-length amino acid sequence of carbohydrate the MBL with an amino acid sequence pdgdsslaaserka recognition domain (CRD) of MBL is shown in SEQID NO. ldtema rikkwltfslgkg (SEQ ID NO. 15) or a fragment 4. The carbohydrate recognition domain of an engineered thereof. Without wishing to be bound by theory, the neck MBL described herein can have an amino acid sequence of region can provide flexibility and proper orientation of the about 10 to about 300 amino acid residues, or about 50 to 35 CRD to bind to a microbe surface. In some embodiments, about 160 amino acid residues. In some embodiments, the the carbohydrate recognition domain can comprises a full microbe Surface-binding domain can have an amino acid length CRD of MBL (SEQID NO. 4; termed as “CRD head” sequence of at least about 5, at least about 10, at least about 102) and the neck region thereof 104, as shown in FIG. 1B. 15, at least about 20, at least about 30, at least about 40, at The amino acid sequence encoding a full-length CRD of least about 50, at least about 60, at least about 70, at least 40 MBL and the neck region thereof is shown in SEQ ID NO. about 80, at least about 90, at least about 100, at least about 5. The crystal structure of a native MBL “neck and CRD 150 amino acid residues or more. Accordingly, in some head' has been previously shown in Chang et al. (1994) J embodiments, the carbohydrate recognition domain of the Mol Biol. 241:125-7 (FIG. 2). A skill artisan can readily engineered MBL molecule can comprise SEQID NO. 4. In modify the identified CRD and fragments thereof to modu Some embodiments, the carbohydrate recognition domain of 45 late its orientation and binding performance to carbohy the engineered MBL molecule can comprise a fragment of drates on a microbe Surface, e.g., by theoretical modeling SEQ ID NO. 4. Exemplary amino acid sequences of such and/or in vitro carbohydrate-binding experiments. In addi fragments include, but are not limited to, ND (SEQID NO. tion, based on the crystal structure of the native MBL “neck 10), EZN (SEQID NO. 11: where Z is any amino acid, e.g., and CRD head', peptidomimetics that can effectively mimic P), NEGEPNNAGS (SEQID NO. 12) or a fragment thereof 50 at least a fragment of the CRD head and optionally the neck comprising EPN, GSDEDCVLL (SEQ ID NO. 13) or a region can be also used as a carbohydrate recognition fragment thereof comprising E, and LLLKNGOWND domain of the engineered microbe-targeting molecule or VPCST (SEQID NO. 14) or a fragment thereof comprising MBL molecule described herein. One of skill in the art can ND. Modifications to such CRD fragments, e.g., by conser readily determine such peptidomimetic structure without vative substitution, are also within the scope described 55 undue experimentations, using any methods known in the art herein. In some embodiments, the MBL or a fragment and the known crystal structure. thereof used in the microbe surface-binding domain of the In some embodiments, the carbohydrate recognition engineered microbe-targeting molecules described herein domain of the microbe-targeting molecule can further com can be a wild-type molecule or a recombinant molecule. prise a portion of a carbohydrate-binding protein. However, The exemplary sequences provided herein for the carbo 60 in Some circumstances, complement or coagulation activa hydrate recognition domain of the engineered microbe tion induced by a carbohydrate-binding protein or a frag targeting molecules are not construed to be limiting. For ment thereof can be undesirable depending on various example, while the exemplary sequences provided herein applications, e.g., in vivo administration for treatment of are derived from a human species, amino acid sequences of sepsis. In such embodiments, the portion of the carbohy the same carbohydrate recognition domain in other species 65 drate-binding protein can exclude at least one of comple Such as mice, rats, porcine, bovine, feline, and canine are ment and coagulation activation regions. By way of known in the art and within the scope described herein. example, when the carbohydrate-binding protein is man US 9,593,160 B2 17 18 nose-binding lectin or a fragment thereof, the mannose Immun. 688 (2000). The capsular polysaccharide of Neis binding lectin or a fragment thereof can exclude at least one seria meningitides serogroup B. H. influenzae type b and of the complement and coagulation activation regions Cryptococcus neoformans are thought to decrease MBL located on the collagen-like region. In such embodiments, binding, as does bacterial endotoxin. Id.; Van Emmerik et the mannose-binding lectin or a fragment thereof can al., 97 Clin. Exp. Immunol. 411 (1994); Schelenz et al., 63 exclude at least about one amino acid residue, including at Infect. Immun. 3360 (1995). least about two amino acid residues, at least about three Others have reported that MBL facilitates opsonophago amino acid residues, at least about four amino acid residues, cytosis of yeasts but not of bacteria, despite MBL binding: at least about five amino acid residues, at least about six MBL (Lectin) pathway of complement was critical for the amino acid residues, at least about seven amino acid resi 10 opsonophagocytosis of yeast, but the classical complement dues, at least about eight amino acid residues, at least about pathway was critical for opsonophagocytosis of bacteria. nine amino acid residues, at least about ten amino acid Brouwer et al., 180 J. Immunol. 4124 (2008). It was not residues or more, around amino acid residue K55 or L56 of reported that MBL bound to the bacterial species tested, SEQ ID NO. 2. Exemplary amino sequences comprising however, only that MBL binding did not promote significant K55 or L56 of SEQID NO. 2 that can be excluded from the 15 complement activation and opsonophagocytosis. engineered MBL molecule include, but are not limited to, Derivatives of MBL with a particular specificity can be EPGQGLRGLQGPPGKLGPPGNPGPSGS (SEQ ID NO. isolated, e.g., by the following approach, which is a standard 16), GKLG (SEQ ID NO. 17), GPPGKLGPPGN (SEQ ID phage display strategy: First, express a set of MBL variants NO. 18), RGLQGPPGKL (SEQ ID NO. 19), GKLGPPGN from a phagemid vector; then bind this library to a target of PGPSGS (SEQ ID NO. 20), GLRGLQGPPGKLGPPGN interest (e.g., E. coli) and perform one or two rounds of PGP (SEQ ID NO. 21), or any fragments thereof. selection; and then perform a round of negative selection Further regarding the carbohydrate recognition domain against a related target (e.g., Candida), taking those (CRD) or a fragment thereof its binding characteristics can phagemids that fail to bind. These cycles of positive and be manipulated by directed evolution for altered binding negative selection are then repeated until a population of specificity. By way of example only, MBL can be modified 25 phages that generally bind to the target and do not bind to the so that it binds to a more limited set of sugars or other non-target is generated. This method can be applied to any molecular features, with the result that the modified MBL pair of microbial strains against which differential binding is will bind to a more limited set of microbes to provide a desired. Such as bacteria that are resistant and sensitive to a capability for pathogen class identification (e.g., one of given antibiotic. This positive/negative enrichment strategy virus, bacteria, fungi, or protozoan), Subclass typing (e.g., 30 can also be used with an antibody-phage display library, gram negative or gram positive bacteria) or specific species which is an even more standard way to isolate such specific determination. Numerous strategies of directed evolution are binders. available in the art. The directed evolution and selection approach described For example, a straightforward directed evolution strategy above also can potentially be used to generate human visually examines an atomic structure of MBL complexed 35 antibody fragments or peptides that provide the class, Sub with a Sugar, and then mutates appropriate amino acids that class and species specificity described above. make contact in a Sugar-specific manner, so that distinctive In some embodiments, at least two microbe Surface contacts are lost or particular types of steric hindrance are binding domains (e.g., carbohydrate recognition domains), created. The three dimensional structure of rat MBL has including at least three, at least four, at least five, at least six, been solved in a complex with a high-mannose oligosac 40 at least seven, at least eight, at least nine, at least ten or more charide and with N acetylglucosamine, a methylated fucose, microbe Surface-binding domains, can be linked together to and so on. His 189Val and Ile207Val are examples of sub form a multimeric microbe Surface-binding domain or car stitutions that modifications alter specificity. bohydrate recognition domain. In such embodiments, the In another strategy of directed evolution, the protein is distances between microbe Surface-binding domains (e.g., Subjected to random mutagenesis and the resulting proteins 45 carbohydrate recognition domains) can be engineered to are screened for desired qualities. This is a particularly match with the distance between the binding sites on the useful technology for affinity maturation of phage display target microbe Surface. antibodies, where the antibody complementary determining A multimeric microbe Surface-binding domain can have regions (CDRS) are mutated by Saturation mutagenesis and each of the individual microbe surface-binding domains the successful variants of the six CDRs are shuffled together to 50 same. Alternatively, a multimeric microbe Surface-binding form the highest affinity antibodies. domain can have at least one, at least two, or at least three The directed evolution paradigm can be applied to MBL microbe surface-binding domains different from the rest. In in order to select MBL variants with specific binding to, e.g., Such embodiments, microbe Surface-binding domains that but not limited to, yeast, gram-positive bacteria, gram share a common binding specificity for carbohydrates on a negative, coagulase negative, and aerobic bacteria. For this 55 microbe Surface can be used. By way of example only, the to work, however, the pattern and nature of the target Sugars fibrinogen-like domain of several lectins has a similar func or related Surface features on these target microorganisms tion to the CRD of C-type lectins including MBL, and can differ between the classes or species. function as pattern-recognition receptors to discriminate MBL is known to bind strongly to mannose and N-acetyl pathogens from self. One of Such lectins comprising the glucosamine Sugars on fungi, gram-positive, and gram 60 fibrinogen-like domain is serum ficolins. negative bacteria. For example, MBL binds strongly to Serum ficolins have a common binding specificity for Candida spp., Aspergillus filmigatus, Staphylococcus GlcNAc (N-acetylglucosamine), elastin or GalNAc aureus, and B hemolytic group A Streptococci. MBL has (N-acetyl-galactosamine). The fibrinogen-like domain is intermediate affinity to Escherichia coli, Klebsiella spp., and responsible for the carbohydrate binding. In human serum, Haemophilus influenzae type b. MBL binds weakly to B 65 two types of ficolin, known as L-ficolin (also called P35, hemolytic group B Streptococci, Streptococcus pneumoniae, ficolin L, ficolin 2 or hucolin) and H-ficolin (also called and Staphylococcus epidermidis. Neth et al., 68 Infect. & Hakata antigen, ficolin 3 or thermolabile b2-macroglyco US 9,593,160 B2 19 20 protein), have been identified, and both of them have lectin - Continued activity. L-ficolin recognises GlcNAc and H-ficolin recog nises GalNAc. Another ficolin known aS M-ficolin (also ciohydrate recognition domain (CRD) of MBL (SEQ called P3 5-related protein, ficolin 1 or ficolin A) is not ID NO. 4) : considered to be a serum protein and is found in leucocytes 5 VGNKFFLTNG EIMTFEKVKA LCVKFQASVA TPRNAAENGA and in the lungs. L-ficolin and H-ficolin activate the lectin complement pathway in association with MASPs. M-Fico IONLIKEEAF LGITDEKTEG OFVDLTGNRL TYTNWNEGEP lin, L-ficolin and H-ficolin has calcium-independent lectin NNAGSDEDCV LLLKNGOWND VPCSTSHLAV CEFPI activity. Accordingly, in Some embodiments, an engineered (v) 10 (v microbe-targeting, e.g., an engineered MBL molecule, Ca Neck + Carbohydrate recognition domain of MBL (SEQ comprise MBL and L-ficolin carbohydrate recognition ID NO. 5) : domains, MBL and H-ficolin carbohydrate recognition PDGDSSLAAS ERKALOTEMA RIKKWLTFSL GKOVGNKFFL domains, or a combination thereof. TNGEIMTFEK VKALCVKFOA SWATPRNAAE NGAIONLIKE Any art-recognized recombinant carbohydrate-binding 15 proteins or carbohydrate recognition domains can also be EAFLGITDEK TEGOFWDLTG NRLTYTNWNE GEPNNAGSDE used in the engineered microbe-targeting molecules. For DCVLLLKNGO WNDVPCSTSH LAVCEFPI example, recombinant mannose-binding lectins, e.g., but not limited to, the ones disclosed in the U.S. Pat. Nos. 5,270, FoMBL.(vi) 81 (SEQ ID NO. 6) : 199; 6,846,649; and U.S. Patent Application No. US 2004/ 20 EPKSSDKTHT CPPCPAPELL GGPsv FPP 0229212, the contents of which are incorporated herein by reference, can be used in constructing the engineered MBL KPKDTLMSR TPEWTCVWWD WSHEDPEWKFNWYWDGWEVH molecules described herein. NAKTKPREEO YNSTYRVVSV LTVLHODWLN GKEYKCKVSN In one embodiment, the microbe-binding molecule com- as prises an MBL, a carbohydrate recognition domain of an KALPAPIEKT ISKAKGOPRE POVYTLPPSR DELTKNOWSL MBL, or a genetically engineered version of MBL (FcMBL) TCLVKGFYPS DIAVEWESNG OPENNYKTTPPVLDSDGSFF as described in International Application No. WO 2011/ 090954, filed Jan. 19, 2011, the content of both of which is LYSKLTVDKS RWOOGNWFSC SVMHEALHNH YTOKSLSLSP incorporated herein by reference. Amino acid sequences for 30 GAPDGDSSLAASERKALQTE MARIKKWLTF SLGKQVGNKF MBL and engineered MBL include, but are not limited to: FLTNGEIMTF EKVKALCVKF OASWATPRNA. AENGAIONLI

KEEAFLGITD EKTEGOFWDL TGNRLTYTNW NEGEPNNAGS (i) MBL full length (SEQ ID NO. 1) : 35 DEDCVLLLKN GOWNDVPCST SHLAVCEFPI MSLFPSLPLL LISMWAASYS (vii) ETVTCEDAOK TCPAVIACSS PGINGFPGKD GRDGTKGEKG AKT-FoMBL (SEO ID NO. 7) : AKTEPKSSDKTHT CPPCPAPELL EPGQGLRGLQ GPPGKLGPPG NPGPSGSPGP KGOKGDPGKS GGPSWFLFPP KPKDTLMISR TPEWTCWWWD WSHEDPEWKF PDGDSSLAAS ERKALOTEMA RIKKWLTFSL GKOVGNKFFL 40 NWYWDGVEVH NAKTKPREEO YNSTYRVVSV LTVLHODWLN TNGEIMTFEK VKALCVKFOA SWATPRNAAE NGAIONLIKE GKEYKCKVSN KALPAPIEKT ISKAKGOPRE POVYTLPPSR EAFLGITDEK TEGOFWDLTG NRLTYTNWNE GEPNNAGSDE DELTKNOWSL TCLVKGFYPS DIAVEWESNG OPENNYKTTP DCVLLLKNGO WNDVPCSTSH LAVCEFPI 45 PVLDSDGSFF LYSKLTVDKS RWOOGNWFSC SVMHEALHNH (ii) MBL without the signal sequence (SEQ ID NO. 2) : YTOKSLSLSP GAPDGDSSLA ASERKALOTE MARIKKWLTF ETVTCEDAOK SLGKQVGNKF FLTNGEIMTF EKVKALCVKF OASWATPRNA TCPAVIACSS PGINGFPGKD GRDGTKGEKG EPGOGLRGLO 50 AENGAIONLI KEEAFLGITD EKTEGOFVDL TGNRLTYTNW GPPGKLGPPG NPGPSGSPGP KGOKGDPGKS PDGDSSLAAS NEGEPNNAGS DEDCVLLLKN GOWNDVPCST SHLAVCEFPI ERKALOTEMA RIKKWLTFSL GKOVGNKFFL TNGEIMTFEK (viii) WKALCVKFOA SWATPRNAAE NGAIONLIKE EAFLGITDEK 55 FoMBL. 111 (SEQ ID NO. 8) : EPKSSDKTHT CPPCPAPELL GGPSWFLFPP TEGOFVDLTG NRLTYTNWNE GEPNNAGSDE DCVLLLKNGO KPKDTLMSR TPEWTCVWWD WSHEDPEWKF NWYWDGWEVH WNDWPCSTSH LAWCEFPI NAKTKPREEO YNSTYRVVSV LTVLHODWLN GKEYKCKVSN (iii) Truncated MBL (SEQ ID NO. 3) : KALPAPIEKT ISKAKGQPRE POVYTLPPSR DELTKNOVSL AASERKALOT EMARIKKWLT TCLVKGFYPS DIAVEWESNG OPENNYKTTP PVLDSDGSFF FSLGKOVGNK FFLTNGEIMT FEKVKALCVK FOASWATPRN LYSKLTVDKS RWOOGNWFSC SVMHEALHNH YTOKSLSLSP AAENGAIONL IKEEAFLGIT DEKTEGOFWD LTGNRLTYTN 65 GATSKQVGNKF FLTNGEIMTF EKVKALCVKF QASVATPRNA WNEGEPNNAG SDEDCVLL.L.K NGOWNDVPCS TSHLAVCEFP I US 9,593,160 B2 21 22 - Continued (Thr), methionine (Met) or alanine (Ala), or at least one of AENGAIONLI KEEAFLGITD EKTEGOFVDL TGNRLTYTNW codon sequences encoding the aforementioned amino acids (i.e., Gly, Ser, Asn. Thr, Met or Ala). Such amino acids and NEGEPNNAGS DEDCVLLLKN GOWNDVPCST SHLAVCEFPI corresponding nucleic acid sequences are generally used to In some embodiments, a microbe-binding molecule com- 5 provide flexibility of a linker. However, in some embodi prises an amino acid sequence selected from SEQ ID NO. ments, other uncharged polar amino acids (e.g., Gln, CyS or 1-SEQ ID NO. 8. Tyr), nonpolar amino acids (e.g., Val, Leu, Ile, Pro, Phe, and Without wishing to be bound by a theory, microbe Trp), or nucleic acid sequences encoding the amino acids binding molecules comprising lectins or modified versions thereof can also be included in a linker sequence. In alter thereof can act as broad-spectrum pathogen binding mol 10 native embodiments, polar amino acids or nucleic acid ecules. Accordingly, microbes and/or microbial matter pres sequence thereof can be added to modulate the flexibility of ent in a test sample can be captured using lectin-based a linker. One of skill in the art can control flexibility of a microbe-binding molecules without identifying the microbe. linker by varying the types and numbers of residues in the Linkers linker. See, e.g., Perham, 30 Biochem. 8501 (1991); Wrig 15 gers et al., 80 Biopolymers 736 (2005). As used herein, the term “linker generally refers to a In alternative embodiments, a linker can be a chemical molecular entity that can directly or indirectly connect at linker of any length. In some embodiments, chemical linkers two parts of a composition, e.g., at least one microbe can comprise a direct bond or an atom such as oxygen or Surface-binding domain and at least one Substrate-binding sulfur, a unit such as NH, C(O), C(O)NH, SO, SO, SONH, domain. In some embodiments, the linker can directly or 20 or a chain of atoms, such as Substituted or unsubstituted indirectly connect to one or more microbe Surface-binding C-C alkyl, Substituted or unsubstituted C-C alkenyl, domains. Without limitations, in some embodiments, the substituted or unsubstituted C-C alkynyl, substituted or linker can also provide binding sites to one or more microbes unsubstituted C-C aryl, Substituted or unsubstituted and/or microbial matter. In such embodiments, the microbe Cs-C heteroaryl, Substituted or unsubstituted Cs-C het binding sites on the linker can bind to the same types and/or 25 erocyclyl. Substituted or unsubstituted C-C cycloalkyl, species of microbes as the microbes bind to a microbe where one or more methylenes can be interrupted or termi Surface-binding domain. Alternatively or additionally, the nated by O, S, S(O), SO, NH, or C(O). In some embodi microbe-binding sites on the linker can capture different ments, the chemical linker can be a polymer chain (branched types and/or species of microbes than the ones that bind to or linear). a microbe Surface-binding domain described herein. 30 In some embodiments where the linker is a peptide. Such Linkers can be configured according to a specific need, peptidyl linker can comprise at least a portion of an immu e.g., based on at least one of the following characteristics. noglobulin, e.g., IgA, Ig|D, IgE, IgG and IgM including their By way of example only, in some embodiments, linkers can Subclasses (e.g., IgG1), or a modified molecule or recom be configured to have a sufficient length and flexibility such binant thereof. In some embodiments, the peptide linker can that it can allow for a microbe Surface-binding domain to 35 comprise a portion of fragment crystallization (Fc) region of orient accordingly with respect to at least one carbohydrate an immunoglobulin or a modified thereof. In Such embodi on a microbe Surface. In some embodiments, linkers can be ments, the portion of the Fc region that can be used as a configured to allow multimerization of at least two engi linker can comprise at least one region selected from the neered microbe-targeting molecules (e.g., to from a di-, tri-, group consisting of a hinge region, a CH2 region, a CH3 tetra-, penta-, or higher multimeric complex) while retaining 40 region, and any combinations thereof. By way of example, biological activity (e.g., microbe-binding activity). In some in some embodiments, a CH2 region can be excluded from embodiments, linkers can be configured to facilitate expres the portion of the Fc region as a linker. In one embodiment, sion and purification of the engineered microbe-targeting Fc linker comprises a hinge region, a CH2 domain and a molecule described herein. In some embodiments, linkers CH3 domain, e.g., Fc IgG 106 as shown in FIG. 1B and FIG. can be configured to provide at least one recognition site for 45 3. Such Fc linker can be used to facilitate expression and proteases or nucleases. In addition, linkers are preferably purification of the engineered microbe-targeting molecules non-reactive with the functional components of the engi described herein. The N terminal Fc has been shown to neered molecule described herein (e.g., minimal hydropho improve expression levels, protein folding and secretion of bic or charged character to react with the functional protein the fusion partner. In addition, the Fc has a staphylococcal domains such as a microbe Surface-binding domain or a 50 protein A binding site, which can be used for one-step Substrate-binding domain). purification protein A affinity chromatography. See Lo KM In some embodiments, a linker can be configured to have et al. (1998) Protein Eng. 11:495-500. Further, the protein any length in a form of a peptide, peptidomimetic, an A binding site can be used to facilitate binding of protein aptamer, a protein, a nucleic acid (e.g., DNA or RNA), or A-expressing or protein G-expressing microbes in the any combinations thereof. In some embodiments, the pep- 55 absence of calcium ions. Such binding capability can be tidyl or nucleic acid linker can vary from about 1 to about used to develop methods for distinguishing protein A-ex 1000 amino acids long, from about 10 to about 500 amino pressing microbes (e.g., S. aureus) from non-protein A-ex acids long, from about 30 to about 300 amino acids long, or pressing or non-protein G-expressing microbes (e.g., E. coli) from about 50 to about 150 amino acids long. Longer or present in a test sample, and various embodiments of Such shorter linker sequences can be also used for the engineered 60 methods will be described in detail later. Further, such Fc microbe-targeting molecules described herein. In one linker have a molecule weight above a renal threshold of embodiment, the peptidyl linker has an amino acid sequence about 45 kDa, thus reducing the possibility of engineered of about 200 to 300 amino acids in length. microbe-targeting molecules being removed by glomerular In some embodiments, a peptide or nucleic acid linker can filtration. Additionally, the Fc linker can allow dimerization be configured to have a sequence comprising at least one of 65 of two engineered microbe-targeting molecules to form a the amino acids selected from the group consisting of dimer, e.g., the dimeric engineered MBL molecule 100 as glycine (Gly), serine (Ser), asparagine (ASn), threonine shown in FIG. 1B. US 9,593,160 B2 23 24 In some embodiments where the linker comprises a Fc effectively with microbes. In some embodiments, the dis region or a fragment thereof, the Fc region or a fragment tance between the microbe Surface-binding domain and the thereof can comprise at least one mutation, e.g., to modify substrate can range from about 50 angstroms to about 5000 the performance of the engineered microbe-targeting mol angstroms, from about 100 angstroms to about 2500 ang ecules. For example, in some embodiments, a half-life of the stroms, or from about 200 angstroms to about 1000 ang engineered microbe-targeting molecules described herein StrOmS. can be increased, e.g., by mutating an amino acid lysine (K) The linkers can be of any shape. In some embodiments, at the residue 232 of SEQ ID NO. 9 to alanine (A). Other the linkers can be linear. In some embodiments, the linkers mutations, e.g., located at the interface between the CH2 and can be folded. In some embodiments, the linkers can be CH3 domains shown in Hinton et al (2004) J Biol Chem. 10 branched. For branched linkers, each branch of a microbe 279:6213-6216 and Vaccaro C. et al. (2005) Nat Biotechnol. Surface-binding domain can comprise at least one microbe 23: 1283-1288, can be also used to increase the half-life of Surface-binding domain. In other embodiments, the linker the IgG1 and thus the engineered microbe-targeting mol adopts the shape of the physical Substrate. ecules. In some embodiments provided herein, the linker can In some embodiments, the linker can be albumin, trans 15 further comprise a detectable label. In some embodiments, ferrin or a fragment thereof. Such linkers can be used to the detectable label can be a chromogenic or fluorogenic extend the plasma half-life of the engineered microbe microbe enzyme substrate so that when a microbe binds to targeting molecules and thus are good for in vivo adminis the engineered microbe-targeting molecule, the enzyme that tration. See Schmidt S R (2009) Curr Opin Drug Discov the microbe releases can interact with the detectable label to Devel. 12: 284. induce a color change. Examples of Such microbe enzyme When the engineered microbe-targeting molecules are Substrate can include, but are not limited to, indoxyl used as therapeutics in vivo, the linker can be further butyrate, indoxyl glucoside, esculin, magneta glucoside, modified to modulate the effector function such as antibody red-f-glucuronide, 2-methoxy-4-(2-nitrovinyl) phenyl B-D- dependent cellular cytotoxicity (ADCC) and complement glu-copyranoside, 2-methoxy-4-(2-nitrovinyl) phenyl B-D- dependent cytotoxicity (CDC). By way of example only, the 25 cetamindo-2-deoxyglucopyranoside, and any other art-rec Fc region for use as a linker can mediate ADCC and CDC. ognized microbe enzyme Substrates. Such embodiments can In ADCC, the Fc region can generally bind to Fc receptors act as an indicator for the presence of a microbe or pathogen. on the Surface of immune effector cells such as natural Conjugation of Engineered Microbe-Targeting Molecules to killers and macrophages, leading to the phagocytosis or lysis a Substrate of a targeted cell. In CDC, the Fc region can generally 30 The engineered microbe-targeting molecules can be trigger the complement cascade at the cell Surface to kill the immobilized on any Substrate for various applications and/or targeted cell. Accordingly, modulating effector functions can purposes. For example, when the affinity of a single microbe be achieved by engineering the Fc region to either increase Surface-binding domain for a target molecule (e.g., a carbo or decrease their binding to the Fc receptors on the surface hydrate recognition domain for a Sugar/carbohydrate mol of the immune effector cells or the complement factors. For 35 ecule) is relatively low, and Such binding is generally driven example, numerous mutations within a Fc region for modu by avidity and multivalency, multivalency of Such engi lating ADCC and CDC are well known to a skilled artisan, neered microbe-targeting molecules can be effectively e.g., see Armour K L. et al. (1999) Eur J Immmunol 29: increased by attachment of a plurality of the engineered 2613-2624; Shields R L. et al. (2001) J Biol Chem. 276: microbe-targeting molecules (e.g., each with one or two or 6591-6604: Idusogie E. E. et al. (2001) J Immunol. 166: 40 more carbohydrate recognition domains) to a solid Substrate 2571-2575: Idusogie E. E. et al. (2000) J Immunol. 155: (e.g., a nanometer- or micrometer-sized bead) at a high 1165-1174; and Steurer W. et al. (1995) J Immunol. 155: density, which can be varied to provide optimal functional 1165-1674. In one embodiment, the amino acid asparagine ity. Alternatively, the engineered microbe-targeting mol (N) at the residue 82 of the SEQ ID NO. 6 can be mutated ecules can be immobilized on a solid substrate for easy to aspartic acid (D), e.g., to remove the glycosylation of Fc 45 handling during usage, e.g., for isolation, observation or and thus, in turn, reduce ADCC and CDC functions. microscopic imaging. In various embodiments, the N-terminus or the C-termi The attachment of the engineered microbe-binding mol nus of the linker, e.g., the portion of the Fc region, can be ecule (e.g., FcMBL) to a Substrate Surface (e.g., membrane modified. By way of example only, the N-terminus or the Surface, glass Surface, tubing Surface) can be performed with C-terminus of the linker can be extended by at least one 50 multiple approaches, for example, by direct cross-linking the additional linker described herein, e.g., to provide further engineered microbe-binding molecule (e.g., FcMBL) to the flexibility, or to attach additional molecules. In some Substrate Surface; cross-linking the engineered microbe embodiments, the N-terminus of the linker can be linked binding molecule (e.g., FcMBL) to the substrate surface via directly or indirectly (via an additional linker) with a sub a nucleic acid matrix (e.g., DNA matrix or DNA/oligonucle strate-binding domain adapted for orienting the carbohy 55 otide origami structures) for orientation and concentration to drate recognition domain away from the Substrate. increase detection sensitivity; cross-linking FcMBL to the In some embodiments, the linker can be embodied as part substrate surface via a dendrimer-like structure (e.g., PEG/ of the microbe Surface-binding domain, or part of the Chitin-structure) to increase detection sensitivity; attracting carbohydrate-binding protein (e.g., MBL and/or the neck FcMBL-coated magnetic microbeads to the substrate sur region thereof). 60 face with a focused magnetic field gradient applied to the In some embodiments, the linker can be a physical Substrate surface, attaching an engineered microbe-binding Substrate, e.g., microparticles or magnetic microbes, to molecule (e.g., FcMBL) to a substrate via biotin-avidin or which a plurality of microbe Surface-binding domains (in biotin-avidin-like interaction, or any other art-recognized cluding carbohydrate recognition domain) can bind, pro methods. vided that there is at least a certain distance between the 65 For engineered microbe-targeting molecules or mannose microbe Surface-binding domain and the Substrate surface binding lectin molecules to be immobilized on or conjugated sufficient for the microbe surface-binding domain to interact to a Substrate, the engineered molecules described herein can US 9,593,160 B2 25 26 further comprise at least one (e.g., one, two, three, four, five, ordinary skill in the art. The oligonucleotides including six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, aptamers can be of any length, e.g., from about 1 nucleotide fifteen, sixteen, seventeen, eighteen, nineteen, twenty or to about 100 nucleotides, from about 5 nucleotides to about more) Substrate-binding domain, e.g., adapted for orienting 50 nucleotides, or from about 10 nucleotides to about 25 the carbohydrate recognition domain away from the Sub nucleotides. Generally, a longer oligonucleotide for hybrid strate. Without limitations, exemplary types of substrates ization to a nucleic acid scaffold can generate a stronger can be a nucleic acid scaffold, a biological molecule (e.g., a binding strength between the engineered microbe Surface living cell), or a solid Surface. In some embodiments, the binding domain and Substrate. Solid Surface can be functionalized with a coupling mol Alternatively or additionally, the surface of a substrate can ecule, e.g., an amino group, to facilitate the conjugation of 10 be functionalized to include coupling molecules described engineered microbe Surface-binding domains to the Solid herein. As used herein, the term “coupling molecule” refers Surface. to any molecule or any functional group that is capable of As used herein, the term “substrate-binding domain selectively binding with an engineered microbe Surface refers to any molecule that facilitates the conjugation of the binding domain described herein. Representative examples engineered molecules described herein to a Substrate or a 15 of coupling molecules include, but are not limited to, functionalized Substrate. In some embodiments, the Sub antibodies, antigens, lectins, proteins, peptides, nucleic acids strate-binding domain can comprise at least one amino (DNA, RNA, PNA and nucleic acids that are mixtures group that can non-covalently or covalently coupled with thereof or that include nucleotide derivatives or analogs); functional groups on the surface of the substrate. For receptor molecules, such as the insulin receptor, ligands for example, the primary amines of the amino acid residues receptors (e.g., insulin for the insulin receptor); and biologi (e.g., lysine or cysteine residues) at the N-terminus or in cal, chemical or other molecules that have affinity for close proximity to the N-terminus of the engineered microbe another molecule, such as biotin and avidin. The coupling Surface-binding domains (e.g., engineered mannose-binding molecules need not comprise an entire naturally occurring lectins) can be used to couple with functional groups on the molecule but may consist of only a portion, fragment or Substrate Surface. 25 Subunit of a naturally or non-naturally occurring molecule, In some embodiments, the Substrate-binding domain can as for example the Fab fragment of an antibody. The comprise at least one, at least two, at least three or more coupling molecule can further comprise a detectable label. oligopeptides. The length of the oligonucleotide can vary The coupling molecule can also encompass various func from about 2 amino acid residues to about 10 amino acid tional groups that can couple the Substrate to the engineered residues, or about 2 amino acid residues to about 5 amino 30 microbe Surface-binding domains. Examples of Such func acid residues. Determination of an appropriate amino acid tional groups include, but are not limited to, an amino group, sequence of the oligonucleotide for binding with different a carboxylic acid group, an epoxy group, and a tosyl group. substrates is well within one of skill in the art. For example, In some embodiments, the engineered microbe-targeting as shown in FIG. 1B, according to one or more embodi molecule can be conjugated to a substrate surface through a ments, the Substrate-binding domain 108 can comprise an 35 covalent or non-covalent interaction. The engineered oligopeptide comprising an amino acid sequence of AKT, microbe-targeting molecule and/or coupling molecule can which provides a single biotinylation site for Subsequent be conjugated to the Surface of a solid Substrate covalently binding to streptavidin-coated Substrate, e.g., a magnetic or non-covalently using any of the methods known to those microbead 110. Such single biotinylation site can also of skill in the art. For example, covalent immobilization can enable the carbohydrate recognition domain of an engi 40 be accomplished through, for example, silane coupling. See, neered microbe Surface-binding domain to orient away from e.g., Weetall, 15 Adv. Mol. Cell Bio. 161 (2008); Weetall, 44 the Substrate, and thus become more accessible to microbes Meths. Enzymol. 134 (1976). The covalent interaction or pathogens. See, e.g., Witus et al. (2010).JACS 132: 16812. between the engineered microbe-targeting molecule and/or In some embodiments, the Substrate-binding domain can coupling molecule and the Surface can also be mediated by comprise at least one oligonucleotide. The sequence and 45 other art-recognized chemical reactions, such as NHS reac length of the oligonucleotides can be configured according tion or a conjugation agent. The non-covalent interaction to the types of the Substrate, binding density, and/or desired between the engineered microbe-targeting molecule and/or binding strength. For example, if the Substrate is a nucleic coupling molecule and the Surface can be formed based on acid scaffold, e.g., a DNA scaffold, the oligonucleotide ionic interactions, van der Waals interactions, dipole-dipole sequence of the Substrate-binding domain can be designed 50 interactions, hydrogen bonds, electrostatic interactions, and/ Such that it is complementary to a Sub-sequence of the or shape recognition interactions. nucleic acid scaffold to where the substrate-binding domain Without limitations, conjugation can include either a can hybridize. stable or a labile (e.g. cleavable) bond or conjugation agent. In some embodiments, the oligonucleotides can include Exemplary conjugations include, but are not limited to, aptamers. As used herein, the term "aptamer” means a 55 covalent bond, amide bond, additions to carbon-carbon single-stranded, partially single-stranded, partially double multiple bonds, azide alkyne Huisgen cycloaddition, Diels Stranded or double-stranded nucleotide sequence capable of Alder reaction, disulfide linkage, ester bond, Michael addi specifically recognizing a selected non-oligonucleotide mol tions, silane bond, urethane, nucleophilic ring opening reac ecule or group of molecules by a mechanism other than tions: epoxides, non-aldol carbonyl chemistry, cycloaddition Watson-Crick base pairing or triplex formation. Aptamers 60 reactions: 1,3-dipolar cycloaddition, temperature sensitive, can include, without limitation, defined sequence segments radiation (IR, near-IR, UV) sensitive bond or conjugation and sequences comprising nucleotides, ribonucleotides, agent, pH-sensitive bond or conjugation agent, non-covalent deoxyribonucleotides, nucleotide analogs, modified nucleo bonds (e.g., ionic charge complex formation, hydrogen tides and nucleotides comprising backbone modifications, bonding, pi-pi interactions, cyclodextrin/adamantly host branchpoints and nonnucleotide residues, groups or bridges. 65 guest interaction) and the like. Methods for selecting aptamers for binding to a molecule are As used herein, the term "conjugation agent’ means an widely known in the art and easily accessible to one of organic moiety that connects two parts of a compound. US 9,593,160 B2 27 28 Linkers typically comprise a direct bond or an atom Such as carbohydrate, enzyme-enzyme cofactor, enzyme-enzyme oxygen or sulfur, a unit such as NR", C(O), C(O)NH, SO, inhibitor, and complementary oligonucleotide pairs capable SO, SONH or a chain of atoms, such as substituted or of forming nucleic acid duplexes). The coupling molecule unsubstituted alkyl, substituted or unsubstituted alkenyl, pair can also include a first molecule that is negatively Substituted or unsubstituted alkynyl, arylalkyl, arylalkenyl, charged and a second molecule that is positively charged. arylalkynyl, heteroarylalkyl, heteroarylalkenyl, heteroaryl One example of using coupling pair conjugation is the alkynyl, heterocyclylalkyl, heterocyclylalkenyl, heterocy biotin-avidin or biotin-streptavidin conjugation. In this clylalkynyl, aryl, heteroaryl, heterocyclyl cycloalkyl, approach, one of the members of the coupling pair (e.g., a cycloalkenyl, alkylarylalkyl, alkylarylalkenyl, alkylarylal portion of the engineered microbe-targeting molecule Such kynyl, alkenylarylalkyl, alkenylarylalkenyl, alkenylarylal 10 as Substrate-binding domain, or a Substrate) is biotinylated kynyl, alkynylarylalkyl, alkynylarylalkenyl, alkynylarylal and the other (e.g., a Substrate or the engineered microbe kynyl, alkylheteroarylalkyl, alkylheteroarylalkenyl, targeting molecule) is conjugated with avidin or streptavi alkylheteroarylalkynyl, alkenylheteroarylalkyl, alkenylhet din. Many commercial kits are also available for biotinylat eroarylalkenyl, alkenylheteroarylalkynyl, alkynylheteroary ing molecules, such as proteins. For example, an aminooxy lalkyl, alkynylheteroarylalkenyl, alkynylheteroarylalkynyl, 15 biotin (AOB) can be used to covalently attach biotin to a alkylheterocyclylalkyl, alkylheterocyclylalkenyl, alkyl molecule with an aldehyde or ketone group. In one embodi hererocyclylalkynyl, alkenylheterocyclylalkyl, alkenylhet ment, AOB is attached to the Substrate-binding domain (e.g., erocyclylalkenyl, alkenylheterocyclylalkynyl, alkynylhet comprising AKT oligopeptide) of the engineered microbe erocyclylalkyl, alkynylheterocyclylalkenyl, targeting molecule. alkynylheterocyclylalkynyl, alkylaryl, alkenylaryl, alky One non-limiting example of using conjugation with a nylaryl, alkylheteroaryl, alkenylheteroaryl, alkynyl coupling molecule pair is the biotin-sandwich method. See, hereroaryl, where one or more methylenes can be interrupted e.g., Davis et al., 103 PNAS8155 (2006). The two molecules or terminated by O, S, S(O), SO, NH, C(O)N(R'), C(O), to be conjugated together are biotinylated and then conju cleavable linking group, Substituted or unsubstituted aryl, gated together using tetravalent Streptavidin. In addition, a substituted or unsubstituted heteroaryl, substituted or unsub 25 peptide can be coupled to the 15-amino acid sequence of an stituted heterocyclic; where R' is hydrogen, acyl, aliphatic acceptor peptide for biotinylation (referred to as AP; Chen et or substituted aliphatic. al., 2 Nat. Methods 99 (2005)). The acceptor peptide Without limitations, any conjugation chemistry known in sequence allows site-specific biotinylation by the E. coli the art for conjugating two molecules or different parts of a enzyme biotin ligase (BirA. Id.). An engineered microbe composition together can be used for linking at least one 30 Surface-binding domain can be similarly biotinylated for engineered microbe-targeting molecule to a substrate. conjugation with a solid Substrate. Many commercial kits Exemplary coupling molecules and/or functional groups for are also available for biotinylating proteins. Another conjugating at least one engineered microbe-targeting mol example for conjugation to a solid Surface would be to use ecule to a substrate include, but are not limited to, a PLP-mediated bioconjugation. See, e.g., Witus et al., 132 polyethylene glycol (PEG, NH-PEG-COOH which can 35 JACS 16812 (2010). As described earlier, an AKT sequence have a PEG spacer arm of various lengths X, where on the N terminal of the engineered microbe-targeting 1