US 201601 16468A1 (19) United States (12) Patent Application Publication (10) Pub. No.: US 2016/0116468 A1 RYU et al. (43) Pub. Date: Apr. 28, 2016

(54) METHOD OF ANALYZING BINDING ASPECT Publication Classification OF MEMBRANE IN A LIVING CELL (51) Int. Cl. GOIN 33/557 (2006.01) (71) Applicant: POSTECH ACADEMY-INDUSTRY FOUNDATION, Gyeongsangbuk-do (52) U.S. Cl. (KR) CPC ...... G0IN33/557 (2013.01) (72) Inventors: Sung Ho RYU, Gyeongsangbuk-do (KR); Dohyeon KIM, Seoul (KR): Nam (57) ABSTRACT Ki LEE, Gyeongsangbuk-do (KR); Dong Kyun KIM, Gyeongsangbuk-do (KR); Soyeon PARK, Seoul (KR): The present invention relates to a method for analyzing the Yonghoon KWON, Seoul (KR): Kai pattern of live intercellular membrane protein binding. The ZHOU, Gyeongsangbuk-do (KR) method for analyzing the pattern according to the present (21) Appl. No.: 14/891,555 invention can analyze accurately, sensitively, quickly, and readily the binding pattern of a target membrane protein and (22) PCT Fled: May 16, 2014 a candidate substance to be specifically bound therewith with (86) PCT NO.: PCTAKR2014/004.426 out tagging to a ligand, and thus measure directly and accu rately the position and quantitative information of the binding S371 (c)(1), (2) Date: Nov. 16, 2015 of the membrane protein and the target Substance. Such effects make it possible to apply the method for various uses (30) Foreign Application Priority Data Such as dissociation constant, mutant study, complex forma tion, and signal transduction. Moreover, it is expected to use May 16, 2013 (KR) ...... 10-2013-OO56008 Nov. 29, 2013 (KR) ...... PCTAKR2O13/O110O2 the method for searching out undiscovered membrane pro May 16, 2014 (KR) ...... 10-2014-OO59.027 teins and target substances. Patent Application Publication Apr. 28, 2016 Sheet 1 of 19 US 2016/0116468 A1

FIG. 1

extraceuiar intfacetia

rtation tita RYA

: refectic light intraceluiar

tipid bilayer igand tein) are 3rotei it joiescert proteir

FIG. 2

tle growth

is 888

&

..

dow condition stipe Safia ercia a tice &yeage aiyef age taking cS is a A. SS s refection microscope R & (ojective less {&rctic in 2 time (cycie Patent Application Publication Apr. 28, 2016 Sheet 2 of 19 US 2016/0116468 A1

FIG. 3 Cetuximab

xii)

Patent Application Publication Apr. 28, 2016 Sheet 3 of 19 US 2016/0116468 A1

FIG. 4 pMT NT PVT. Cetuximab 1.0 . . .

8.

s

( Cy

FIG.S. exist a

a. O

2 O

O PM EGFR Patent Application Publication Apr. 28, 2016 Sheet 4 of 19 US 2016/0116468 A1

FIG. 6)

heterotrimeric Gi potei AOP ribosylated Gigotei & Petassis oxii i-S

Fzb, Wy

FIG. 7

SS OO TX

SO

40

2

-2 O -- PMT FZD1 B2-AR Patent Application Publication Apr. 28, 2016 Sheet 5 of 19 US 2016/0116468 A1

FIG. 8

()meie, FIG.9)

ax batch at C2 at C3- at C3-2

different batches reasi?ed of the Sane day Patent Application Publication Apr. 28, 2016 Sheet 6 of 19 US 2016/0116468 A1

FIG. 10 a - -ex.283 (el b- irie 3 ce. & & ... , 58 SO

S. g 4{ 2 40 S. is 30 g 20a g 20. 10 10 S s s: 3 ice. o N. -- g ;: EGR g $ 83-8

C CO-K Ce: d as so S.se 50 s

s 40. 40

s 5 20 ; 20 s s . ... g oil , a lic YYYY YY YY s i -EKS3 i88 Chi-X s ce ce {&

FIG. 11

A R Patent Application Publication Apr. 28, 2016 Sheet 7 of 19 US 2016/0116468 A1

FIG. 12

pEGFR (Y1086 EGFR

FIG. 13

S O

4. O

2 O

is S T Patent Application Publication Apr. 28, 2016 Sheet 8 of 19 US 2016/0116468 A1

FIG. 14) - fluorescent protein

FIG. 15

. S

(arbitrary unit) Patent Application Publication Apr. 28, 2016 Sheet 9 of 19 US 2016/0116468 A1

FIG.16

50 PMT EGFR

2 5

O texifraix tasizia acti artiady

FIG. 17

Ceituxina Fata cetixifra Fab againe: Patent Application Publication Apr. 28, 2016 Sheet 10 of 19 US 2016/0116468 A1

FIG. 18)

EGFR 5 e

aaf{f C&tuxia (exifra Cei Xia Fa a2

O 40 8O 12O 16O molecular weight (kilodalton) Patent Application Publication Apr. 28, 2016 Sheet 11 of 19 US 2016/0116468 A1

FIG. 20

Patent Application Publication Apr. 28, 2016 Sheet 12 of 19 US 2016/0116468 A1

FIG. 21 O.02 (ugimitreated cetuximab concentration 1.0 - -2

O.8

is first measurement is second reasurement O6 treatment time point Patent Application Publication Apr. 28, 2016 Sheet 13 of 19 US 2016/0116468 A1

FIG.22) dissociatio constant Ks is 0.8 . , 8 y & s s^s / ^ 8s a S. c. 0.5 8 Sf :: s: .

werg 5 S S haa 3 . i. 8 12 ratic Ceix Sat-3rd GR rt also 2 & 8 O.8 8 Cettixinhab girl FIG. 23) stry frutant Str WT (wild type)

Patent Application Publication Apr. 28, 2016 Sheet 14 of 19 US 2016/0116468 A1

FIG. 24

S. Cettixia S A R

3. s

FIG. 25

s

s

O 5

o2 os os os to (, 3i Oi city&isian-Si 3FRy: olo oa o os os o 12 4 cetuximab (ign) Patent Application Publication Apr. 28, 2016 Sheet 15 of 19 US 2016/0116468 A1

FIG. 26)

cetaxia t secondary artibody t tertiary antibody

FIG. 27

, treatinent point

x: 8. te rt iary antibody

5 . ^ Ns

10 2 3. time (minute) Patent Application Publication Apr. 28, 2016 Sheet 16 of 19 US 2016/0116468 A1

FIG. 28 in treating point O O

& tertiary antibody W) 0.75 &

O.50

O O 2O 3O time (minute) FIG. 29) & fief:S SR

&ftistiy

Alexa of 64 Patent Application Publication Apr. 28, 2016 Sheet 17 of 19 US 2016/0116468 A1

FIG. 30

five-fescetice tail { ai: Cities: 8 is is: me a 38 fig to Céxi &i gigi) fii (fessi: (.. : . & A is is, 3 + £&iaxina (Exigiri) * ...etixis via {}{igin: f8:

i. R r

| 88

it (3 & $ 3 of: forescence ritersity ofescence intersity (arbitrary unit) (arbitsary unit

FIG. 31

5

...) 4. -

3.

2 Patent Application Publication Apr. 28, 2016 Sheet 18 of 19 US 2016/0116468 A1

FIG. 32 S is turboglobulin a tibody x CetuXiao

.

2

{. COS A43 MA-M8-23 BT20 CC8 Ce ce Ce Cei Ce

FIG.33)

Patent Application Publication Apr. 28, 2016 Sheet 19 of 19 US 2016/0116468 A1

FIG. 34 50 Cetuximab treatment

4. O

3 O

2 O

1 O

O BASC endothelial AT2 Clara Cei cel ce US 2016/0116468 A1 Apr. 28, 2016

METHOD OF ANALYZING BINDING ASPECT the candidate material in a living cell expressing the target OF MEMBRANE PROTEIN IN A LIVING membrane protein; and analyzing a change in diffusion coef CELL ficient of at least one target membrane protein obtained thereby. TECHNICAL FIELD 0008 Also, in another aspect, another exemplary embodi 0001. The present invention relates to a method of analyz ment of the present invention provides a method of screening ing a binding aspect between a membrane protein and a a drug for a target membrane protein, the method including: candidate material in a living cell. analyzing a binding aspect between a candidate material and a target membrane protein; and determining a drug specifi BACKGROUND ART cally acting on the target membrane protein among the can didate materials using the binding aspect. 0002. A cell membrane is an essential factor for maintain ing a cell, and membrane localized in the cell mem 0009. Hereinafter, the present invention will be described brane play an important role in intracellular and extracellular in detail. communication through dynamic interaction with ligands. 0010. The present invention provides a method of analyz Substantially, about a half of cellular proteins are known to be ing a binding aspect between a candidate material and a target able to interact with ligands in the cell membrane, but huge membrane protein, the method including: measuring a diffu parts still remain unknown. To figure out such interactions, a sion coefficient of the target membrane protein before and variety of techniques and attempts have been conducted over after the candidate material is treated in a living cell express the years, and researchers are still making efforts. ing the target membrane protein; and analyzing a change in 0003. Since an extracellular environment has a relatively diffusion coefficient of at least one target membrane protein very small viscosity, compared to an inside of the cell mem obtained thereby. brane, the influence of an extracellular domain of the mem 0011 A variety of dynamic interaction between mem brane protein which directly binds to ligands on the diffusiv brane proteins and ligands is one of the cellular responses to ity of target proteins has been ignored, and many researches various environmental changes. A complex of the membrane relating to the diffusion theory of membrane proteins, which protein bound with various ligands serves in various roles. is the Saffman-Delbruck model, focus on a transmembrane However, since the identification of such a membrane protein domain to determine such diffusivity of the membrane pro complex is technically limited, much of the complex is still teins and describe its importance on membrane protein dif unknown. To overcome Such a limitation, the inventors com fusion. Therefore, it has been considered that it is fundamen pleted a single molecule diffusional motion shift assay (sm tally difficult to measure the binding of ligands to the DIMSA) in which single particle tracking (SPT) combined extracellular domain of the target membrane proteinbased on with Super-resolution microscopy is utilized, and this method diffusion of membrane protein. However, since a system used can be applied as a method of analyzing a binding aspect in the conventional research could measure the interaction between a candidate material and a target membrane protein. between target membrane protein and ligand only in an arti Because smDIMSA sensitively detects a change in diffusion ficial and controlled environment using purified membrane coefficient in a living cell, it can be used to study a biological proteins in vitro or inserting the target membrane protein into reaction in an actual environment in which cells are alive and an artificially formed lipid membrane. Such previous properly functioning, not in an artificially-created environ researches cannot reflect the complicated structure of an ment. actual cell membrane, and thus the analysis of the interactions 0012. The analysis method of the present invention has the involving extracellular domain of membrane proteins with following characteristics (1) to (3) by detecting the fluores ligands has been limited. cence of a fluorescent protein directly conjugated to target membrane protein or fluorescent dyes linked to the probe DISCLOSURE which specifically binds to the target membrane protein: 0013 (1) Non-labeling of ligands and direct binding mea Technical Problem Surement, 0004. The present invention is directed to providing a 0014 (2) Ligand- specificity; and method of sensitively analyzing a binding aspect between a (0015 (3) Size sensitivity. target membrane protein and a candidate material that can 0016. With reference to (1), when conventional flow specifically bind thereto in a living single cell. cytometry, FRET or a different fluorescence-based assay is 0005. The present invention is also directed to providing a used, a candidate material, as well as a membrane protein, method of screening a drug for a target membrane protein. needs to be simultaneously labeled with a fluorescent mate 0006. However, the technical objectives to be accom rial. Therefore, when using various candidate materials, it plished by the present invention are not limited to the above requires separate labeling to each candidate materials. How described problems, and other objectives not described herein ever, since the candidate material does not need to be labeled will be fully understood by those of ordinary skill in the art in the method of the present invention, it has an advantage in from the following descriptions. that the binding can be determined only by the change in diffusion coefficient without additional fluorescence. Technical Solution (0017. With reference to (2), to use the method of the 0007. In one aspect, one exemplary embodiment of the present invention, a fluorescent protein linked directly to tar present invention provides a method of analyzing a binding get membrane protein or a fluorescent dye conjugated probe aspect between a candidate material and a target membrane which specifically binds to target membrane protein are uti protein, the method including: calculating a diffusion coeffi lized to detect target membrane protein, and therefore the cient of target membrane protein before and after treatment of method of the present invention has an advantage in that other US 2016/0116468 A1 Apr. 28, 2016 neighboring membrane proteins do not affect the measure cooperativity may also be obtained therefrom. Examples of ment results during observation. the process of forming a complex can be found in FIGS. 27 0018 With reference to (3), a change in diffusion coeffi and 28. It can be confirmed that the formation of an antibody cient which is related with the size is observed, and therefore complex is detected using a second antibody againt the pri the method of the present invention has an advantage in which mary antibody directly binding to a target membrane protein, the size of an unknown material can be estimated by the and a third antibody against the second antibody. change in diffusion coefficient. 0024. Meanwhile, in the present invention, the membrane 0019. Because of the characteristics (1) to (3), the method protein refers to a protein binding to a membrane of cells or of the present invention may be applied to a variety offields of cell organelles, and the target membrane protein refers to a membrane protein research including (a) a dissociation con desired membrane protein to be observed among numerous stant, (b) a mutant study, (c) complex formation, and (d) membrane proteins present in a membrane. The membrane signal transduction. protein includes an integral membrane protein, a peripheral 0020. With reference to (a), the dissociation constant is an membrane protein, a transmembrane protein, a membrane important parameter capable of estimating a binding strength glycoprotein, and a lipid anchored membrane protein. between a membrane protein and a ligand in the field of 0025. For example, the membrane protein may include a biochemistry or pharmacology. The actual dissociation con receptor tyrosine kinase (RTK), a coupled receptor stant is generally measured from a purified protein of interest (GPCR), an ion-channel, a pattern recognition receptor through surface plasmon resonance (SPR) or a filter binding (PRR), etc. as an integral membrane protein. assay. However, since the membrane protein is difficult to (0026. For example, the RTK includes RTK class I, RTK purify because of its characteristics, the measurement of dis class II, RTK class III, RTK class IV, RTK class V, RTK class Sociation constant of membrane protein has also been expe VI, RTK class VII, RTK class VIII, RTK class IX, RTK class riencing difficulties. If there are two different groups at the X, RTK class XI, RTK class XII, RTK class XIII, RTK class same time, a ratio between two groups may be estimated by XIV, RTK class XV, RTK class XVI, RTK class XVII, etc. the above-described method. Since, according to the method, The RTK class I is an EGF receptor family or ErbB family, a ligand-binding group and an unbound group have different and includes ErbB-1 (EGFR), ErbB-2, ErbB-3, and ErbB-4, sizes, the difference in diffusion coefficients between the etc., the RTK class II is an insulin receptor family and groups can be distinguished, and thus the dissociation con includes insulin receptor-A, insulin receptor-B, etc., the RTK stant can be quantitatively measured using the above ratio at class III is a PDGF receptor family and includes PDGF-A, a single cell level. PDGF-B, PDGF-C, and PDGF-D, etc., the RTK class IV is an 0021. With reference to (b), a mutant protein is used as an FGF receptor family and includes FGFR1, FGFR2, FGFR3, important marker for diagnosing a disease. If the binding FGFR4, FGFR6, etc., the RTK class V is a VEGF receptors between a mutant protein and a ligand in a cell is understood, family and includes VEGFR-1, VEGFR-2, VEGFR-3, etc., a mutant signal can be distinguished from signals of mixed the RTK class VI is a HGF receptor family and includes cells, and this can be of great help in disease research. Only HGFR, etc., the RTK class VII is a Trk receptor family and specific binding originating from mutant proteins with includes TrkA, TrkB, TrkC, etc., the RTK class VIII is an EPH respect to the mutant proteins that are mixed with normal receptor family and includes EPHA1, EPHA2, EPHA3, proteins may be observed by the method of the present inven EPHA4, EPHA5, EPHA6, EPHA7, EPHA8, EPHA9, tion, and therefrom mutant protein-specific characteristics EPHA10, EPHB1, EPHB2, EPHB3, EPHB4, EPHB5, may be analyzed. EPHB6, etc., the RTK class IX is an AXL receptor family and 0022. With reference to (c), proteins generally form an includes AXL, etc., the RTK class X is an LTK receptor enormous molecular complex and play specific roles. How family and includes LTK, etc., the RTK class XI is a TIE ever, since a membrane protein is difficult to purify and iso receptor family and includes TIE1, TIE2, etc., the RTK class late, it is difficult to carry out research on the formation of a XII is an ROR receptor family and includes ROR1, ROR2, molecular complex. However, according to the method of the etc., the RTK class XIII is a DDR receptor family and present invention, since the molecular size becomes larger includes DDR1, etc., the RTK class XIV is an RET receptor during the formation of such a complex and directly affects to family and includes RET, etc., the RTK class XV is a KLG a diffusion coefficient, a process of forming the complex can receptor family and includes PTK7, etc., the RTK class XVI be figured out. is an RYK receptor family and includes RYK, etc., and the 0023 The following examples show that binding between RTK class XVII is an MuSK receptor family and includes a target membrane protein and a candidate material can be MuSK, etc. measured by the analysis of a change in diffusion coefficient, (0027. The GPCR includes Class A (-like), for example, an EGFR antibody and its binding to EGFR is Class B ( family), Class C (Metabotropic identified by a change in diffusion coefficient (refer to FIGS. glutamate?), Class D (Fungal mating pheromone 5 and 13), and binding between GPCR and a G protein can be receptors), Class E (Cyclic AMP receptors), Class F detected by a chemical inhibitor treatment (refer to FIG. 7). (/), etc. Class A has Subfamilies A1 to Also, FIG. 19 shows that a molecular weight can be estimated A19, wherein Subfamily A1 includes chemokine (C-C motif) only by measuring a diffusion coefficient through a correla receptor 1, chemokine (C-C motif) receptor 2, chemokine tion between the change in diffusion coefficient and the (C-C motif) receptor 3, chemokine (C-C motif) receptor 4. molecular weight of the candidate material. Also, FIG. 21 chemokine (C-C motif) receptor 5, chemokine (C-C motif) shows that a ratio of cetuximab-bound EGFR can be analyzed receptor 8, chemokine (C-C motif) receptor-like 2, chemok by the result of measuring the change in diffusion coefficient ine (C motif) receptor 1, chemokine (C-X3-C motif) receptor by treating gradually increasing concentration of cetuximab 1, and GPR137B, etc.; Subfamily A2 includes chemokine (candidate material). Accordingly, the plot shown in FIG. 22 (C-C motif) receptor-like 1, chemokine (C-C motif) receptor may be obtained, and a dissociation constant Kd and binding 6, chemokine (C-C motif) receptor 7, chemokine (C-C motif) US 2016/0116468 A1 Apr. 28, 2016 receptor 9, chemokine (C-C motif) receptor 10, chemokine receptor 1, lysophosphatidic acid (C-X-C motif) receptor 3, chemokine (C-X-C motif) receptor receptor 2, lysophosphatidic acid receptor 3, Sphingosine 4, chemokine (C-X-C motif) receptor 5, chemokine (C-X-C 1-phosphate receptor 1, Sphingosine 1-phosphate receptor 2, motif) receptor 6, chemokine (C-X-C motif) receptor 7. sphingosine 1-phosphate receptor 3, Sphingosine 1-phos IL8R-O, IL8R-3, adrenomedulin receptor, Duffy blood phate receptor 4, sphingosine 1-phosphate receptor 5, mel group, , G protein-coupled receptor 30, anocortin1 receptor, 3 receptor, melanocortin 4 etc.; Subfamily A3 includes II receptor, , , ACTH receptor, GPR3, receptor, B1, . GPR6, GPR12, etc.; Subfamily A14 includes prostaglandin GPR15, GPR25, etc.; Subfamily A4 includes delta opioid D2 receptor, prostaglandin E1 receptor, prostaglandin E2 receptor, kappa , mu opioid receptor, nocice receptor, prostaglandin E3 receptor, prostaglandin E4 recep ptin receptor, 1. Somatostatin receptor tor, , prostaglandin 12 (prostacyclin) 2. . . Soma receptor, thromboxane A2 receptor, etc.; Subfamily A15 tostatin receptor 5, GPCR B/W receptor 1, includes lysophosphatidic acid receptor 4, lysophosphatidic neuropeptides B/W receptor 2, GPR1 , etc.; acid receptor 5, lysophosphatidic acid receptor 6, purinergic Subfamily A5 includes 1, , receptor P2Y 10, coagulation factor II (thrombin) receptor , cysteinyl 1, cysteinyl like 1, coagulation factor II (thrombin) receptor-like 2, coagu leukotriene receptor 2, , leukotriene lation factor II (thrombin) receptor-like 3, lymphocyte-spe B4 receptor 2, relaxin/insulin-like family receptor 1, cific G protein-coupled receptor, GPR4, GPR65, GPR68, relaxin/insulin-like family peptide receptor 2, relaxin/insu GPR132, GPR17, GPR18, GPR20, GPR35, GPR55, coagu lin-like family peptide receptor 3, relaxin/insulin-like family lation factor II receptor, etc.; Subfamily A16 includes rhodop peptide receptor 4, KiSS1-derived peptide receptor (GPR54), sin, 1 (cone pigments), opsin 3, panopsin, opsin 4. melanin-concentrating 1, urotensin-II receptor, etc.; Subfamily A6 includes cholecystokinin A , opsin 5, retinal G protein coupled receptor, and receptor, cholecystokinin B receptor, FF recep retinal pigment epithelium-derived rhodopsin homologs, tor 1, neuropeptide FF receptor 2, hypocretin (orexin) recep etc.; Subfamily A11 includes 5-HT2A, 5-HT2B, 5-HT2C, tor 1, hypocretin (orexin) receptor 2, arginine 5-HT6, alpha1A, alpha1B, alpha1D, alpha2A, alpha2B, receptor 1A, arginine 1B, arginine vaso alpha2C, beta1, beta2, beta3, D1, D2, D3, D4, D5, TAAR1, pressin receptor 2, gonadotrophin releasing hormone recep TAAR2, TAAR3, TAAR5, TAAR6, TAAR8, TAAR9, hista tor, pyroglutamylated RFamide peptide receptor, GPR22, mine H2 receptor, etc.; Subfamily A18 includes histamine H1 GPR176, etc.; Subfamily A7 includes bombesin-like receptor receptor, , , 3. , gastrin-releasing peptide receptor, (A1, A2a, A2b, A3), muscarinic acetyl type A, endothelin receptor type B, choline receptor (M1, M2, M3, M4, M5), GPR21, GPR27, GPR37, 1, , GPR45, GPR52, GPR61, GPR62, GPR63, GPR78, GPR84, 1, , thyrotropin GPR85, GPR88, GPR101, GPR161, GPR173, etc.; and Sub releasing hormone receptor, growth hormone secretagogue family A19 includes 5-Hydroxytryptamine (5-HT) receptors receptor, GPR39, , etc.; Subfamily A8 (5-HT1A, 5-HT1B, 5-HT1D, 5-HT1E, 5-HT1F, 5-HT4, includes , , chemokine-like receptor 5-HT5A, 5-HT7), and additionally, unclassified olfactory 1, 1, formyl peptide receptor-like 1, receptors vomeronasal receptors (VN1R1, VN1R2, VN1 R3, formyl peptide receptor-like 2, MAS1, MAS1L, GPR1, VN1R4, VN1R5), etc. Class B has Subfamilies B1 to B3, GPR32, GPR44, GPR77, etc.; Subfamily A9 includes mela wherein Subfamily B1 includes pituitary adenylate cyclase tonin receptor 1A, 1B, tachykinin recep activating polypeptide type 1 receptor (PACAPR), calcitonin tor 1, 2, , neuropep receptor (CALCR), corticotropin-releasing hormone recep tide Y receptor Y1, Y2, pancreatic tor (CRHR1, CRHR2), glucose-dependent insulinotropic polypeptide receptor 1, neuropeptide Y receptor Y5, prolac polypeptide receptor/gastric inhibitory polypeptide receptor tin-releasing peptide receptor, 1, proki (GIPR), -related (GLP1R, GLP2R), growth neticin receptor 2, GPR19, GPR50, GPR75, GPR83, etc.; hormone releasing hormone receptor (GHRHR), parathyroid Subfamily A10 includes FSH-receptor, luteinizing hormone/ hormone receptor (PTHR1, PTHR2), secretin receptor choriogonadotropin receptor, , leucine (SCTR), vasoactive intestinal peptide receptor (VIPR1, rich repeat-containing G protein-coupled receptor 4, leucine VIPR2), etc.; Subfamily B2 includes brain-specific angio rich repeat-containing G protein-coupled receptor 5, leucine genesis inhibitor (BAI1, BAI2, BAI3), CD97 antigen (CD7), rich repeat-containing G protein-coupled receptor 6, etc.; EMR hormone receptor (CELSR1, CELSR2, CELSR3, Subfamily A11 includes 1, free fatty EMR1, EMR2, EMR3, EMR4), GPR56 orphan receptor acid receptor 2, free fatty acid receptor 3, GPR42, purinergic IPRO03910 (GPR56, GPR64, GPR97, GPR110, GPR111, receptor P2Y1, P2Y2, purinergic receptor GPR112, GPR113, GPR114, GPR115, GPR123, GPR125, P2Y4, purinergic receptor P2Y6, purinergic receptor P2Y8. GPR126, GPR128, GPR133, GPR144, GPR157), Latrophi purinergic receptor P2Y 11, hydroxycarboxylic acid receptor lin receptor IPR003924 (ELTD1, LPHN1, LPHN2, LPHN3), 1, hydroxycarboxylic acid receptor 2, niacin receptor 1, etc.; subfamily B3 includes diuretic hormone receptor; and hydroxycarboxylic acid receptor 3, niacin receptor 2, GPR31, unclassified subfamilies includes Ig-hepta receptor GPR82, oxoglutarate (alpha-ketoglutarate) receptor 1. Succi (GPR116). Class C includes Group I (mGluR1, mGluR5), nate receptor 1, etc.; Subfamily A12 includes purinergic Group II (mGluR2, mGluR3), and Group III (mGluR4, receptor , purinergic receptor P2Y 13, purinergic mGluR6, mGluR7, mGluR8) as a metabotropic glutamate/ receptor P2Y14, GPR34, GPR87, GPR171, platelet-activat pheromone, etc., Class D includes STE2, STE3, etc. as a ing factor receptor, etc.; Subfamily A13 includes cannabinoid fungal mating pheromone receptor, Class E includes cyclic receptor 1 (brain), 2 (macrophage), AMP receptors, and Class F includes Frizzled-1, Frizzled-2, US 2016/0116468 A1 Apr. 28, 2016

Frizzled-3, Frizzled-4, Frizzled-5, Frizzled-6, Frizzled-7, the cell membrane. By using Such a TIRF microscope, images Frizzled-8, Frizzled-9, Frizzled-10, etc. as Frizzled/Smooth are obtained about the positions of the target membrane pro ened. tein over a series of time points, and the result is expressed as 0028. The ion-channels include voltage-gated ion chan coordinates, thereby obtaining a diffusion coefficient. nels (voltage-gated Sodium channel, Voltage-gated calcium 0033. The diffusion coefficient according to the present channel, Voltage-gated potassium channel, transient receptor invention may be obtained using the following Equations 1 potential channel, cyclic nucleotide-gated channel, and Volt and 2. For example, the fluorescence of a fluorescent protein age-gated proton channel), ligand-gated ion channels, verte linked to the target membrane protein is detected, thereby brate anionic cys-loop receptors (GABAA and GlyR), verte obtaining several sequential images of positions of the target brate cationic cys-loop receptors (serotonin, nicotinic membrane protein over time, that is, coordinates for a prede acetylcholine, and zinc-activated ion channel), ionotropic termined period of time, and the motion of Such coordinates glutamate receptors (AMPA, kainate, NMDA, and orphan), may be expressed as one trajectory of the target membrane and ATP-gated channels (P2X), etc. protein. Afterward, a diffusion coefficient with respect to one 0029. The PRRs include receptor kinases, toll-like recep trajectory may be obtained by Substituting the measured coor tors (TLR) (TLRs in drosophila immunity, TLR3, TLR 11), dinates of the trajectory over time to the following Equations and C-type lectin receptors (CLR) including Group I CLRs 1 and 2. To determine the position of one target membrane that is the mannose receptors, and Group II CLRs, which is protein particle by the above-described calculation, an expo the asialoglycoprotein receptor family that contains the clas Sure time for detecting fluorescence (a time required to image sic asialoglycoprotein receptor macrophage galactose-type the membrane-protein particle at one position in the coordi lectin (MGL), DC-SIGN (CLEC4L), Langerin (CLEC4K), nates) may be suitably changed depending on analysis con Myeloid DAP12-associating lectin (MDL)-1 (CLEC5A), ditions, and may be, for example, about 30 to 70 ms, prefer DC-associated C-type lectin 1 (Dectinl) subfamily (dectin ably, 50 ms. Also, the number of pairs of coordinates 1/CLEC7A, DNGR1/CLEC9A, Myeloid C-type lectin-like necessary to make one trajectory of one target membrane receptor (MICL) (CLEC12A), CLEC2, CLEC12B), and DC protein particle may be, but is not limited to. 5 or more. immunoreceptor (DCIR) subfamily (DCIR/CLEC4A, Dectin 2/CLEC6A, Blood DC antigen 2 (BDCA2) (CLEC4C), Mincle i.e. macrophage-inducible C-type lectin (CLEC4E)). 1 NA Equation 1 etc MSD(A) = X(x,-A-X, + y, A -y, 0030 Also, in the present invention, the candidate mate =l rial is a material binding to a target membrane protein to activate the target membrane protein, for example, a com 0034) A is the step size between coordinates of the target pound, a nucleic acid, a saccharide, a carbohydrate, a lipid, a membrane protein particle, wherein A is a natural number, peptide, or a protein. The protein may be, for example, an MSD(A) is the mean square displacement of a target mem antibody, and here, the antibody includes all or a part thereof, brane protein particle with respect to the step size between for example, a Fab domain or a domain from which only a coordinates of the target membrane protein particle, N is the hinge partis selectively removed (half-immunoglobulin frag total number of pairs of coordinates of the target membrane ment). In one exemplary embodiment, the candidate material protein particle to form one trajectory, (x, y) are the coor may be a nucleic acid, and here, the nucleic acid includes all dinates of the target membrane protein particle at an n-num of oligonucleotides including 2 to 200 bases or cDNAs bered position in one trajectory, (x,y) are the coordinates of thereof. Cells may be treated with the candidate material at an the target membrane protein particle at the start point in one amount Suitable for analyzing a binding aspect of a target trajectory, (x,y) are the coordinates of the target membrane membrane protein, and the treatment may be carried out by a protein particle at the end point in one trajectory, and (X, A, method known in the art. For example, 0.01 to 100 ug/ml of yA) are the coordinates of the target membrane protein the candidate material may be directly applied to a medium of particle at an n+A-numbered position in one trajectory, how the cells. ever, n+A is the same as or Smaller than N. 0031. In one exemplary embodiment, a diffusion coeffi Equation 2 cient of a target membrane protein may be obtained by detect ing a motion of the membrane protein in a cell membrane 0035. D is the diffusion coefficient, and A is the step size through single particle tracking (SPT). The SPT is technology between coordinates of the target membrane protein particle. of observing the motion of single particles in a certain 0036. According to such a calculation, MSD is referred to medium, presenting coordinates of the motion over time as a as a function of the step size between the coordinates of the trajectory, and therefrom analyzing a mode of the motion and target membrane protein particle. Assuming that the corre non-uniformity (Saxton, M. J., Jacobson, K. Single-particle sponding trajectory is subject to the Brownian motion, MSD tracking: Applications to membrane dynamics Annual is referred to as a linear function. Therefore, from the slope of Review of Biophysics and Biomolecular Structure 1997: 26 the linear function (f(X)=aX) when MSD of the trajectory of (373-399)). each membrane protein is fitted to the linear function by 0032. The motion of the membrane proteins in the cell Equation 2, a diffusion coefficient (a/4) of the trajectory may membrane may be detected by a method known in the art. For be obtained. Since MSD represents a mean value, more accu example, fluorescence may be detected by total internal rate diffusion coefficient may be obtained as there are more reflection fluorescence (TIRF). A TIRF microscope can coordinates of the target membrane protein particle forming observe only fluorescence in a specific range of 200 nm or less the corresponding trajectory (that is, as the tracked trajectory to the exclusion of interference of other light using total increases in length or a coordinate density is increased). internal reflection of light and therefore is suitable for Therefore, the calculation of the diffusion coefficient needs a research on a cell membrane and the events happening around trajectory with a predetermined length or more, and the fitting US 2016/0116468 A1 Apr. 28, 2016

of the linear function uses two or more frames of the step size 5% at a significance level, the candidate material may be between the coordinates of the target membrane protein par determined as the ligand binding to the target membrane ticle (because the fitting of the linear function is made pos protein. sible with two points), thereby increasing the accuracy in the 0041. The motion of the membrane protein in the cell measurement of the diffusion coefficient. Such diffusion membrane may be detected by labeling the membrane protein coefficients are averaged with respect to the total trajectory, a with an optical probe Such as a fluorescent probe, and as mean diffusion coefficient of the membrane protein in each described above, the detection does not require labeling the cell may be calculated. candidate material, or the binding between the membrane 0037 Also, a change in diffusion coefficient may be protein and the candidate material. obtained by the following Equation 3: 0042. In one exemplary embodiment of the present inven Change in diffusion coefficient (%)=100* |1-(D/ tion, the detection of the motion of the membrane protein in D1) Equation 3 the cell membrane may be carried out by detecting the fluo 0038. In this formula, D is the diffusion coefficient of the rescence of the fluorescent protein in cells expressing a fusion target membrane protein at a concentration c 1 of the candi protein of the target membrane protein and the fluorescent date material in a peripheral environment of the cell at a single protein. The fluorescent protein serves as a label to observe cell level, and D is the diffusion coefficient of the target the target membrane protein, and a type of the fluorescent membrane protein at a concentration c2 of the candidate protein is not particularly limited. For example, the fluores material in a peripheral environment of the cell at a single cell cent protein may include a green fluorescent protein (GFP) level. type, a blue fluorescent protein (BFP) type, a cyan type, a 0039. Since single membrane proteins on a living cell yellow fluorescent protein (YFP) type, a red fluorescent pro membrane of the present invention have a very high hetero tein (RFP) type, an orange type, a far-red type, a near-IR, a geneity even though they are the same type of membrane photoactivatable protein, a photoconvertible protein, a pho proteins, a single cell-level change in diffusion coefficient toswitchable protein, etc. Here, the photoactivatable protein may be analyzed by obtaining diffusion coefficients by track refers to a protein that does not fluoresce in normal circum ing more than one trajectory of target single membrane pro stances, but fluoresces in response to certain stimuli (e.g., tein. As the number of the detected target membrane proteins laser radiation). Also, the photoconvertible protein and the is increased, the analysis accuracy may increase. The number photoSwitchable protein are proteins that showed a specific of the detected target membrane proteins and time to obtain color of fluorescence in normal circumstances, and turned a the trajectory of the target membrane protein may vary different color of fluorescence because of certain stimuli like depending on a type and characteristics of the target mem the photoactivatable protein, but such conversion is irrevers brane protein and a type of cells, and may be Suitably adjusted ible in the photoconvertible protein and reversible in the pho depending on analysis conditions by those of ordinary skill in toswitchable protein. the art. To precisely measure the change in diffusion coeffi 0043. The GFP type includes enhanced green fluorescent cient at a single cell level, 1,000 or more, for example, 5,000 protein (EGFP), Emerald, Superfolder GFP, Azami green or more, preferably, 10,000 or more, more preferably 12,000 mWasabi, TagGFP. AcGFP, T-sapphire, muKG, Clover, or more, still more preferably, 15,000 or more, and most mNeon Green, etc., the blue fluorescent protein (BFP) type preferably 20,000 or more target membrane proteins may be includes enhanced blue fluorescent protein (EBFP), EBFP2, tracked, but the present invention is not limited thereto. After AZurite, mTagBFP mKalama1, Sirius, etc., the cyan type ward, each diffusion coefficient is obtained by the trajectory includes enhanced cyan fluorescent protein (ECFP), mono obtained from each target membrane protein, and these val meric ECFP (mECFP), Cerulean, mTurquoise, mTurquoise2, ues are averaged, thereby obtaining a more accurate diffusion CyPet, TagCFP. mTFP1 (Teal), SCFP3A, monomeric coefficient. For example, when the trajectory of the target Midoriishi Cyan, etc., the YFP type includes enhanced yellow membrane protein that is maintained in a steady state and has fluorescent protein (EYFP), Topaz, Benus, mCitrine, YPet, a length of 5 or more continuous images (in order to obtain TagYFP, PhiYFP, mBanana, SYFP2, etc., the RFP type MSD in which N is 5 or more in Equation 1) is tracked in one includes mRuby, mRuby2, mapple, mStrawberry, mRFP1, cell during measurement, diffusion coefficient values with mCherry, mRaspberry, dKeima-Tandem (monomeric ver respect to about 12,000 or more trajectories may be obtained sion), HcRed-Tandem (monomeric version), mPlum, etc., the within 2 minutes (detecting about 24,000 images), and the Far-red type includes mKate2 mNeptune, etc., the Near-IR resultant values may be averaged to ensure an error of the includes TagRFP657 IFP1.4, etc., the photoactivatable pro diffusion coefficients within 2% at a single cell level. tein includes PA-GFP. PAmCherry 1, PaTagRFP, etc., the pho 0040. When comparing before and after the treatment with toconvertible protein includes PS-CFP2, mEos2, mEos3.2. the candidate material, the size of target membrane protein is PSmOrange, etc., and the photoswitchable protein includes increased due to binding of the candidate material, and the Dronpa, etc. diffusion coefficient of the target membrane protein is 0044) For example, the fluorescent protein may include at reduced, thus the binding status of the target membrane pro least one selected from the group consisting of enhanced tein with the candidate material, a ratio of the target mem green fluorescent protein (EGFP), Emerald, Superfolder brane proteins binding to the candidate material among the GFP. Azami green mWasabi, TagGFP. AcGFP, T-sapphire, total target membrane proteins, the molecular weight of the mUKG, Clover, mNeonGreen, enhanced blue fluorescent candidate material can be analyzed using the degree of a protein (EBFP), EBFP2, AZurite, mTagEFP, mKalama1, change in the diffusion coefficient. Therefore, when there is a Sirius, enhanced cyan fluorescent protein (ECFP), mono change in diffusion coefficient, the candidate material may be meric ECFP (mECFP), Cerulean, mTurquoise, mTurquoise2, determined as a ligand binding to the target membrane pro CyPet, TagCFP. mTFP1 (Teal), SCFP3A, monomeric tein. For example, when the change in diffusion coefficient Midoriishi Cyan, enhanced yellow fluorescent protein (%) obtained from Equation 3 is, for example, at least 2% to (EYFP), Topaz, Benus, mCitrine, YPet, TagYFP, PhiYFP, US 2016/0116468 A1 Apr. 28, 2016

mBanana, SYFP2, mRuby, mRuby2, mApple, mStrawberry, 0049 Any type of probe to which a fluorescent material is mRFP1, mCherry, mRaspberry, dKeima-Tandem (mono conjugated and which can specifically bind to the target mem meric version), HcRed-Tandem (monomeric version), brane protein can be used. However, the probe specifically mPlum, mKate2, mNeptune, mKate2, mNeptune, binding to the target membrane protein that does not affect an TagRFP657, IFP1.4, PA-GFP, PAmCherry 1, PaTagRFP, PS activity of the target membrane protein may be selected. CFP2, mEos2, mEos3.2. PSmOrange and Dronpa. Although not limited thereto, in one exemplary embodiment, 0045. The fusion protein refers to a covalent complex the probe may be an antibody, an aptamer, or a non-antibody formed by genetic fusion or a chemical bond between the protein scaffold. The antibody includes a Fab antibody, a target membrane protein and the fluorescent protein. The single chain variable fragment, and a single-domain antibody 'genetic fusion” refers to a linear connection made by a as well as an antibody having a full sequence. A trajectory can covalent bond through genetic expression of a DNA sequence be tracked through the binding with the target membrane encoding the fusion protein. The formation of the fusion protein after a fluorescent material is conjuaged to the anti protein may be performed by a known method, for example, body having a full sequence, but since the degree of a change genetic recombination technology. in diffusion coefficient is reduced in the binding with a dif ferent bindable partner due to the size of a full antibody, a 0046. The cell expressing the fusion protein may be pre marker that maintains specificity of the antibody and has a pared by transformation with an expression vector. A known size as Small as possible may be preferable. Also, as well as expression vector developed to express the fusion protein the antibody, the aptamer or the non-antibody protein scaf may be suitably selected. For example, the expression vector fold, conjugated to a fluorescent probe, may be useful for the may be a plasmid vector, a virus or a cosmid vector. Host cells method. The non-antibody protein scaffold is made by find expressing the fusion protein may be prepared by a chemical ing and optimizing a scaffold specifically binding to the target treatment method such as a calcium phosphate method or protein among random protein scaffolds, not based on an calcium chloride/rubidium chloride method described in the antibody and is, for example, a material such as an affibody. literature (Sambrook, J., et al., Molecular Cloning, A Labo 0050. A fluorescent material binding to the probe is not ratory Manual (Second edition), Cold Spring Harbor Labo particularly limited. In one exemplary embodiment, the fluo ratory, 1.74, 1989), electroporation, electroinjection or PEG, rescent material may be an organic fluorescent dye, and the or a transformation method such as a method using a gun organic fluorescent dye that can be used in the present inven or virus transfection. For example, the cell expressing the tion may be selected from Atto 488, Alexa Flour 488, Dy505. fusion protein may be prepared by transfecting a host cell Rhodamine 123, Atto 520, Dy 530, ATTO 532, Alexa Fluor with an expression vector containing the fusion protein of the 532, Fluorescein, FITC, Cy2, Cy3B, Alexa Flour 568, target membrane protein and the fluorescent protein using TAMRA, Cy3, Cy3.5, SNAP-Cell TMR-Star, Atto 565, Atto lipofectamine, fugene or metafectin. 590, Alexa Fluor 647, Cy5, Atto 647, Atto 647N, Dyomics 0047. The cells expressing the fusion protein of the target 654, Atto 655, TMP-ATTO 655, Atto 680, Cy5.5, Atto 680, membrane protein and the fluorescent protein may be animal, Alexa Fluor 680, Atto 700, Alexa Fluor 700, DyLight 750, plant, yeast and bacterial cells. For example, the cells may be Cy7, Alexa Flour 750, Atto 740, Alexa Flour 790, and IRDye human embryonic kidney (HEK) cells, HEK293 cells (ATCC 800 CW. Other than this, any organic fluorescent dye corre CRL 1573), 3T3-L1 cells (ATCC CL 173), C6 cells (ATCC sponding to a photoSwitchable fluorescent dye may be used. CCL 107), CHO (Chinese hamster ovary) cells (ATCC 0051 Meanwhile, the method of the present invention CCL61), CHOK1 cells (ATCC CCL 61), NIH/3T3 (NIH may further include obtaining diffusion coefficients of the Swiss mouse embryo) cells (ATCC CRL 1658), BHK (baby target membrane protein after treatment with the candidate hamster kidney) cells, COS1 cells (ATCCCRL 1650), COS7 material at several points of time, and analyzing a change in cells (ATCC CRL 1651), HaCaT cells, HeLa cells (ATCC diffusion coefficient of at least one target membrane protein CCL 2), HeLa S3 cells (ATCCCCL 2.2), HepG2 cells (ATCC obtained thereby over time. For example, a dissociation con HB 8065), HL-60 cells (ATCC CCL 240), HUV-EC-C cells stant between the target membrane protein and the candidate (ATCCCRL 1730), Jurkat cells (ATCCTIB 152), K-562 cells material may be analyzed by detecting the fluorescence of the (ATCC CCL 243), L6 cells (ATCC CRL 1458), MCF7 cells target membrane protein moving in a flexible membrane State (ATCC HTP 22), MDCK cells (ATCC CCL 34), NIH/3T3 over time. cells (ATCC CRL 1658), RAW264.7 cells (ATCC TIB 71), 0.052 Also, the analysis method may further include ana RBL-1 cells (ATCC CRL 1378), SH-SY5Y cells (ATCC lyzing the change in diffusion coefficient of the target mem CRL 2266), or U-937 cells (ATCC CRL 1593.2). brane protein over time to determine membrane protein-spe 0048. In still another exemplary embodiment of the cific endocytosis. The endocytosis of the target membrane present invention, the motion of the membrane protein in the protein may be determined from a tracked trajectory based on cell membrane may be detected with a fluorescent material the diffusion coefficient. During the endocytosis, the trajec conjugated probe specifically binding to the target membrane tory of the membrane protein particle has a confined motion protein. A method of detecting the change in diffusion coef which shows the motion of particles in a limited space having ficient using the fusion protein of the membrane protein and a size of about 120 nm, which is the size of a clathrin-coated the fluorescent protein requires preparing cells which express particle (CCP), which is one of the endocytotic organelles, at the fusion protein whose diffusion coefficient will be a certain position while they freely diffuse. Therefore, the detected, and has a disadvantage in that a type of the cells pattern of changing the trajectory of the particle from the free capable of expressing the fusion protein depending on a type diffusion into the confined motion may be considered as the of the target membrane protein is limited. However, since the endocytosis. Since the endocytosis may be used as an impor method using the fluorescent material-conjugated probe does tant factor for determining an effect of a target-specific drug, not have Such a limitation and employs any cells expressing when membrane protein-specific endocytosis is found by the the target membrane protein, it can be usefully applied. treatment of the candidate material to target membrane pro US 2016/0116468 A1 Apr. 28, 2016

tein, it can be expected that the candidate material has an target membrane protein and a candidate material specifically effect as a target-specific drug. binding thereto without labeling a ligand, and therefore, a 0053 According to the analysis method, the process of positional and quantitative information about the binding forming the complex of the target membrane protein and the between the membrane protein and the candidate material candidate material may be analyzed by analyzing the change may be directly and precisely measured. Due to Such an in diffusion coefficient of the target membrane protein over effect, the method may be applied in various applications time. For example, the process of forming a complex of the Such as the measurement of a dissociation constant, a mutant target membrane protein and various candidate materials may study, complex formation, and signal transduction. Further be analyzed over time by analyzing the change in diffusion more, the method is also expected to explore various mem coefficient of the fusion protein by repeatedly adding the brane proteins and candidate materials, which are still same or different candidate materials. unknown. 0054 Abinding aspect in a single cell may be analyzed by the above method. When a conventional art such as a bio DESCRIPTION OF DRAWINGS chemical method, more specifically, immunoprecipitation is used, about ten million cells are needed, but the method of the 0.058 FIG. 1 is a schematic diagram of a system that can present invention can be used at a single-cell level, and there only specifically observe a target protein among various fore, the binding aspect to the target membrane protein may membrane proteins present in a cell membrane by labeling be analyzed only with a very small amount of the candidate the target membrane protein with a fluorescent protein, under materials. the environment of which binding of aligand makes diffusion 0055. In another aspect, the present invention provides a of the target protein slow. method of screening a drug for a target membrane protein, the 0059 FIG. 2 is a diagram illustrating a process of tracking method including: analyzing a binding aspect between a can the motion of a target membrane protein of one cell attached didate material and a target membrane protein by the above to around coverslip by TIRF and spatio-temporally averaging described method; and determining a drug specifically acting the tracking result to analyze the diffusion coefficient of the on the target membrane protein among the candidate materi target membrane protein in a single cell. als using the binding aspect. In the screening method, the 0060 FIG. 3 illustrates randomly selected trajectories of membrane protein, the candidate material, a diffusion coeffi PMT-mEos3.2 and EGFR-mEos3.2 before and after treat cient, and the detection of the motion of the membrane pro ment with cetuximab to the host cells expressing each protein tein in a cell membrane is as described above. respectively, in which PMT has almost no change in length of 0056. The membrane protein is a main target in drug the trajectory before and after the treatment with cetuximab, development, and 50% or more of all of the drugs developed but EGFR shows absolutely shorter length of the trajectory now target the membrane protein (Overington, J. P. How after the treatment with cetuximab. many drug targets are there?, Nature reviews. Drug discovery. 0061 FIG. 4 illustrates the means and standard deviations 5, 993-996, 2006). For example, isoprenaline, which is a of diffusion coefficients of PMT and EGFR before and after therapeutic agent for heart block or bradycardia, is a drug the treatment with cetuximab for 10 cycles according to targeting a b-, and insulin, which is a Example 1.5. diabetes therapeutic agent, is known as a drug targeting an 0062 FIG. 5 illustrates the quantitatively summarized insulin receptor (Peter Imming. et al. Drugs, their targets and results of FIG. 3. the nature and number of drug target. Nature reviews. Drug discovery, 2006). Also, various developed anticancer agents 0063 FIG. 6 illustrates that B2-AR statically is not bound are targeting the membrane protein. For example, the ErbB with a Gi-protein and thus has no difference even by the family plays an important role in , and the importance treatment with Pertussis toxin (PTX), but as FZD1 is bound on the development of an anticancer agent targeting the ErbB with a Gi-protein, when treated with PTX, ADP of the Gi family is already known (Eric K. Rowinsky. THE ERBB protein is ribosylated, thereby inhibiting the binding with FAMILY: Targets for Therapeutic Development Against Can FZD1. cer and Therapeutic Strategies Using Monoclonal Antibodies 0064 FIG. 7 illustrates changes in diffusion coefficients of and Tyrosine Kinase Inhibitors. Annu Rev. Med. 55, 433-57, PMT, FZD1 and B2-AR before and after PMT. FZD1 and 2004). For example, cetuximab, which is a colon cancer B2-AR are treated with PTX according to Example 2.4. therapeutic agent, is a drug targeting the ErbB family. There 0065 FIG. 8 illustrates absolute values of averaged diffu fore, research on binding between the membrane protein and sion coefficients of PMT. B2-AR, EGFR according to a ligand may be applied to develop a drug specifically acting Example 2.4. on the target membrane protein, and particularly, may be used 0.066 FIG. 9 illustrates changes of diffusion coefficients as a method of screening a drug against the target membrane of EGFR measured in different environments according to protein. For example, after cells expressing the target mem Example 3, in which batchi 1, batchi2 and batch #3 represent brane protein are treated with the candidate material, when a changes in diffusion coefficient of EGFR in COS7 cells pre change in diffusion coefficient (%) is in more than an error pared in three different batches, and batchi3-1 and batchi3-2 range of, for example, 2 to 5% at least, by Equation 3, the are in the same batch, but represent changes in diffusion candidate material may be determined as a drug specifically coefficient of EGFR measured at different glasses. acting on the target membrane protein. 0067 FIG. 10 illustrates changes in diffusion coefficient of EGFR in HEK293, HELA and CHO-K1 cells according to Advantageous Effects Examples 4 to 6. 0057. A method of analyzing a binding aspect according 0068 FIG. 11 is a diagram illustrating antibodies capable to the present invention is a method of accurately, sensitively, of binding to EGFR at different sites, for example, cetuximab, rapidly and easily analyzing the binding aspect between the mAb 199.12, mAb 528 and mAb R-1. US 2016/0116468 A1 Apr. 28, 2016

0069 FIG. 12 illustrates western blotting results accord I0088 FIG.31 illustrates a specific binding aspect between ing to Comparative Example 1, in which NT represents a the endogenous EGFR and cetuximab. non-treatment state, and EGF--f- represents treatment/non I0089 FIG. 32 is the quantitative results of the means and treatment of EGF. standard deviations of diffusion coefficients changed by treat 0070 FIG. 13 illustrates changes in diffusion coefficients ment of an immunoglobulin antibody and cetuximab in five of PMT and EGFR according to Example 7. different cell lines. 0071 FIG. 14 illustrates cetuximab specifically binding to 0090 FIG.33 is images of mouse primary cells classified EGFR, and trastuzumab specifically binding to ErbB2. by marker expressions of cell types in lung cell tissues and 0072 FIG. 15 illustrates flow cytometry results according autofluorescence levels. to Comparative Example 2, in which, particularly, the X axis 0091 FIG. 34 illustrates the changes in diffusion coeffi represents the intensity of a fluorescent signal, and the y axis cient depending on a binding aspect of cetuximab to endog represents a relative ratio of the cell population having a enous EGFRs present in four types of cells selected in FIG. specific level of the intensity of a fluorescent signal. 33, which are BASCs, endothelial cells, AT2 cells, and Clara 0073 FIG. 16 is a graph illustrating changes in diffusion cells. coefficients before and after PMT and EGFR are treated with cetuximab, trastuzumab, and an anti-actin antibody accord BEST MODE ing to Example 8. 0092. Hereinafter, the present invention will be described 0074 FIG. 17 is diagrams illustrating that EGFR is bound in further detail with reference to the following examples. with four different ligands having different molecular However, the following examples are merely provided to weights at the same binding position. illustrate the present invention, and the scope of the present 0075 FIG. 18 illustrates changes in diffusion coefficient of EGFR before and after PMT and EGFR are treated with invention is not limited to the following examples. cetuximab, cetuximab Fab, cetuximab F(ab')2, and EGFR specific binding aptamer according to Example 9. EXAMPLES 0076 FIG. 19 illustrates that the molecular weight of the Example 1 ligand and the change in diffusion coefficient have a linear relationship according to Example 9. Measurement of Target Membrane Protein: EGFR 0077 FIG. 20 is a diagram illustrating that EGFR-bound cetuximab increases in number as the concentration of cetux (0093. 1.1 Plasmid for Preparing EGFR-mEos3.2 Fusion Protein imab is increased according to Example 10.1. (0094 EGFRWT (Addgene plasmid #11011) used in the 0078 FIG. 21 illustrates reduction in diffusion coefficient experiment was provided by Dr. Matthew Meyerson, and of EGFR as the concentration of cetuximab is increased from mEos3.2 fluorescent protein was provided by Dr. Tao Xu. The 0.02 to 1.28 ug/ml in a single cell. given fluorescent protein was subcloned from pBGFP-N1/ 007.9 FIG. 22 is a graph of a dissociation constant and mEos3.2 to form pcDNA3.1/mEos3.2-his in order to facili cooperativity obtained by calculating the ratio of EGFR tate binding with a membrane protein. To prepare a protein in bound cetuximab by increasing the concentration of cetux which the membrane protein EGFR WT (SEQ ID NO: 1) is imab. linked with the fluorescent protein mEos3.2 (SEQID NO:2), 0080 FIG. 23 illustrates that EGFR wild-type (WT, wild pcDNA3.1/EGFR WT-mEos3.2-His was constructed using type) can be bound with both of cetuximab and an mAb R-1 restriction enzymes such as Xbal (SEQID NO:3, TCTAGA) antibody, but a mutant, EGFRVIII, is bound with only cetux and NotI (SEQID NO: 4, GCGGCCGC). imab, and not an mAb R-1 antibody. (0095) 1.2 Plasmid for Preparing PMT-mEos3.2 Fusion 0081 FIG. 24 illustrates that PMT is not bound with two Protein antibodies such as cetuximab and mAb R-1, and thus has no (0096. A plasmidin which PMT is linked with a fluorescent reduction in diffusion coefficient of PMT, but EGFRVIII is protein was prepared to be used as a control group for not bound with mAb R-1 but bound with cetuximab and thus Example 1.1. DNA (SEQ ID NO: 5) corresponding to PMT shows different change in diffusion coefficient. was inserted into the subcloned pcDNA3.1/mEos3.2-his 0082 FIG. 25 is a graph of a dissociation constant and using restriction enzymes such as Xbal (SEQID NO:3) and cooperativity obtained by calculating a ratio of EGFRVIII NotI (SEQID NO: 4), thereby constructing pcDNA3.1/PMT bound cetuximab by the method described in Example 10.1. mEos3.2-His. 0083 FIG. 26 illustrates that cetuximab, a secondary anti (0097. 1.3 Preparation of Host Cells for Expressing Fusion body, and a tertiary antibody that bind to EGFR in turn. Protein 0084 FIG. 27 illustrates that the diffusion coefficient of (0098 COS7 cells provided by ATCC were grown in EGFR is gradually reduced when EGFR is treated with the DMEM (Dulbecco's Modified Eagle Medium, Lonza) under cetuximab, secondary antibody, and tertiary antibody in turn. conditions of 37° C., 5% CO, and 95% humidity, with 10% 0085 FIG. 28 illustrates the changes in diffusion coeffi FBS (Gibco), and COS7 cells were transfected with each of cient of EGFR when the EGFR is sequentially treated with the the plasmids constructed in Examples 1.1 and 1.2 using Lipo cetuximab, tertiary antibody, and secondary antibody. fectamine 2000 (Invitrogen) according to the manufacturers I0086 FIG. 29 illustrates a process of detecting endog instruction. To reduce the influence of protein overexpres enous EGFR by conjugating a photoswitchable fluorescent Sion, cells showing especially low expression level were organic dye, Alexa Fluor 647, to a Fab fragment capable of sorted from the cells grown for 24 hours after transfection specifically binding to the endogenous EGFR. through flow cytometry. In detail, the transfected cells were I0087 FIG.30 illustrates that the binding of mAb 199.12 to washed with PBS to remove debris, detached from a bottom the endogenous EGFR does not interfere with the binding of using a detaching buffer, and collected and strained with a cell cetuximab. strainer (40 um, BD Bioscience). Afterward, the cells show US 2016/01 16468 A1 Apr. 28, 2016 ing low expression level were specifically sorted using a obtained in Example 1.3 was treated with 20 ug/ml of cetux MoFloTMXDP cell sorter (Beckman Coulter). imab, and the changes of diffusion were observed before and 0099. The sorted cells were seeded on a 25-mm-diameter after the treatment by the method described in Example 1.4. coverslip. Before use, the coverslip was prepared by proce The results are shown in FIG. 3. dures of sonication in acetone at 42° C. for 30 minutes to 0107 FIG. 3 illustrates randomly selected trajectories in minimize autofluorescence of the coverslip, washing with the host cells expressing PMT-mEos3.2 and the host cells distilled water three times, sonication again in 1% hydrofluo expressing EGFR-mEos3.2 before and after treatment with ric acid at 42°C. for 10 minutes, washing with distilled water cetuximab, in which PMT has almost no change in length of about 15 to 20 times to completely remove hydrofluoric acid, the trajectory before and after the treatment with cetuximab, and finally soaking in 100% ethanol under UV light for 30 but EGFR shows an obviously shorter length of the trajectory minutes to be sterilized. As a result, the coverslips on which after the treatment with cetuximab. Non-treatment (NT) host cells expressing the EGFR-mEos fusion protein and host shown in FIG.3 represents a state in which nothing is applied cells expressing the PMT-mEos fusion protein were seeded to an external environment, and cetuximab represents the were obtained. trajectories measured after the treatment with cetuximab. In 0100) 1.4. Identification of Membrane Protein to which each image, the enlarged image represents the trajectories of Ligand Did not Bind several fluorescent molecules randomly extracted from a cor 0101 The coverslip on which the cells were seeded as responding region. It can be seen that in non-treated NT (left). described in Example 1.3 was placed onto a microscope with both of the PMT and EGFR membrane proteins are relatively a live cell chamber (37°C., 5% CO) and an electron multi freely diffused. However, it can be seen that, in the cetuximab plying charge coupled device (EM-CCD) and observed the (right), the motion patterns are different. Since PMT do not cells at a single cell level. The microscope, the Olympus IX71 bind with cetuximab, there is no significant change in the model, is designed based on total internal reflection fluores length of the trajectory of the fluorescent molecule. However, cence (TIRF), and can be used to observe only fluorescent since EGFR directly binds with cetuximab, it can be observed molecules at a plasma membrane of a cell closely attached that the length of the trajectory of the fluorescent molecule (within about 200 nm from a coverslip) to a round coverslip. becomes obviously shorter. mEos3.2 fluorescent proteins were randomly converted from I0108. To prove that the result of the change shown in FIG. agreen form into a red from by cleaving their backbone using 3 is not caused by a certain factor generated during the experi 405 nm laser, and an image of the randomly converted red ment, but by EGFR and an antibody binding thereto, changes form mEos3.2 fluorescent protein was obtained by an elec of other factors occurring before and after the treatment with tron multiplying charge coupled device (EM-CCD), ixon3 cetuximab were observed. In each experiment, a total of 897, manufactured by Andor Technology, which can detect a 4,000 images were obtained, trajectories are obtained from signal emitted from a single fluorescent molecule using 561 these images, diffusion coefficients of PMT and EGFR were nm laser. calculated by Equations 1 and 2, and then the mean and 0102) The fluorescent images were sequentially obtained standard deviation of the diffusion coefficients are shown in at regular intervals of time (about 150ms), and a centroid was FIG. 4. found from the signal of a single molecule present on each 0109 FIG. 4 illustrates the means and standard deviations image through centroid fitting, and the moving distance per of diffusion coefficients of PMT and EGFR before and after unit time and the trajectory of the fluorescent protein were the treatment with cetuximab for 10 cycles, and it can be seen determined by using multiple tracking between consecutive from FIG. 4 that the diffusion coefficient of EGFR is changed images, and a mean square displacement (MSD) was calcu by cetuximab, and not by other imaging factors during mea lated by Equation 1. Surement. (0103 FIG. 1 is a schematic diagram of a system that can 0110 FIG. 5 illustrates the quantitatively summarized specifically observe a target protein among many membrane results of FIG. 3, which show changes in diffusion coeffi proteins present in a cell membrane by labeling the target cients (%) of PMT and EGFR after treatment with cetuximab, membrane protein with a fluorescent protein, under an envi compared with the diffusion coefficients of non-treated PMT ronment in which binding of a ligand makes diffusion of the and EGFR. While PMT had almost no change in diffusion target protein slow. coefficient before and after the treatment with cetuximab, 0104 FIG. 2 is a diagram illustrating a process of tracking EGFR showed a change in diffusion coefficient of about 40% the motion of a target membrane protein in a single cell after the treatment with cetuximab. In detail, the diffusion attached to a round coverslip by TIRF and spatio-temporally coefficients may be obtained by Equations 1 and 2, and when averaging the tracking result to analyze the diffusion coeffi a ratio of the change in diffusion coefficient is calculated by cient of the target membrane protein in a single cell. A ligand Equation 3, provided that diffusion coefficients obtained treatment gave a different environment in a single cell, and the under different conditions were D and D. PMT has almost trajectory of a single target membrane protein over time was the same D and D values, thereby having a very small acquired. To obtain statistical significance, the above proce change in diffusion coefficient, but the D value of EGFR is dures were repeated, and the values obtained therefrom were about 60% of the D value thereof, and thus it can be seen that spatio-temporally averaged, thereby overcoming heterogene the diffusion coefficient is reduced by about 40%. ity of the cell itself, and enabled to be compared with diffu 0111. As shown in FIGS. 1 to 5, an increase in size caused sion of the target membrane protein measured under different by the binding between the target membrane protein and the conditions. ligand leads to reduction in diffusion coefficient of the target 0105 1.5. Identification of Ligand-Bound Target Mem membrane protein, and a degree of the reduction can be brane Protein quantified. It could be confirmed that PMT that cannot bind to 0106 Each group of the host cells expressing PMT cetuximab had no significant change in diffusion coefficient mEos3.2 and the host cells expressing EGFR-mEos3.2 before and after the treatment with cetuximab, but EGFR was US 2016/0116468 A1 Apr. 28, 2016

reduced in diffusion coefficient after the treatment with Example 3 cetuximab, and the reduction degree was about 40%. Measurement of Target Membrane Protein in Example 2 Different Environments 0.124. To prove technical reproducibility of the method Measurement of Target Membrane Protein: GPCR described in Example 1, three groups of transfected COS7 cells were prepared on different days (batchi 1, batchi 2, and 0112 2.1 Plasmid for Preparing fB2-AR-mEos3.2 Fusion batchi3), two groups of transfected COS7 cells (batchi3-1 Protein and batchi3-2) were prepared independently but on the same 0113 B2-AR (Addgene plasmid #14697) used in the day, and then coverslips on which these batches were seeded experiment was provided by Dr. Robert Lefkowitz, and were prepared. The host cells were treated with cetuximab by |B2-AR DNA (SEQID NO: 6) was inserted into the pcDNA3. the same method as described in Example 1.4, and then a 1/mEos3.2-his subcloned in Example 1.1 using restriction change in diffusion coefficient of EGFR was measured, and enzymes Such as BamHI and Xbal, thereby constructing the results are shown in FIG. 9. pcDNA3.1/B2-AR-mEos3.2-His. 0.125 FIG. 9 illustrates the changes in diffusion coefficient 0114 2.2 Plasmid for Preparing FZD1-mEos3 Fusion of EGFR measured by the above-described method, in which Protein batchi 1, batchi 2, and batchi3 represent the changes in dif 0115 FZD1 (Addgene plasmid #16819) used in the fusion coefficient of EGFR in the COS7 cells prepared in experiment was provided by Dr. Randall Moon, and FZD1 three different batches, and batchi3-1 and batchi3-2 repre DNA (SEQ ID NO: 7) was inserted into the pcDNA3.1/ sent changes in diffusion coefficient of EGFR measured at mEos3.2-his Subcloned in Example 1.1 using restriction different glasses, but in the same batch. enzymes Such as BamHI and Xbal, thereby constructing I0126. As shown in FIG. 9, all of the batches prepared on pcDNA3.1/FZD1-mEos3.2-His. different days and the batches independently prepared on the same day showed no difference in change in diffusion coef 0116 2.3 Preparation of Host Cells for Expressing Fusion ficient of EGFR, and it was confirmed by ANOVA analysis Protein that there is no statistical difference between the batches 0117 For an experiment to inactivate GPCR, COS7 cells (p=0.768). were transfected with J32-AR-mEos3.2. FZD1-mEos3.2. and PMT-mEos3.2 (refer to Example 1.2) by the same Examples 4-6 method as described in Example 1.3, and each group of the transfected COS7 cells were starved in 1% serum (minimum Identification of Target Membrane Protein (Cell condition)-containing medium for 16 hours. Also, for a per Lines: HEK293, HELA, and CHO-K1) tussis toxin (PTX) experiment, prior to measurement of a I0127. To demonstrate that target membrane proteins can membrane protein, each group of the transfected COS7 cells be measured in other cell lines than COS7 cells, host cells were treated with PTX at a concentration of 100 ng/ml for 6 expressing PMT-mEos3.2 and host cells expressing EGFR hours. mEos3.2 were prepared using HEK293, HELA, and CHO 0118 2.4 Identification of Ligand-Bound Target Mem K1 cells provided by ATCC instead of COS7 cells used in brane Protein Example 1.3 by the method described in Example 1.4, and 0119) Diffusion coefficients of PMT. FZD1 and B2-AR changes in diffusion coefficients of EGFR and PMT were were measured by the same method described in Example measured by treating each group of the host cells with cetux 1.5, and the results are shown in FIGS. 7 and 8. imab by the method described in Example 1.5, and the results are shown in FIG. 10. The results obtained using HEK293, 0120 FIG. 6 illustrates that B2-AR basically do not bind HELA, and CHO-K1 cells as host cells were described in with a Gi-protein, but as FZD1 bind with a Gi-protein, when Examples 4 to 6, respectively. treated with PTX, ADP of the Gi-protein is ribosylated, thereby inhibiting the binding with FZD1. According to the I0128 FIG. 10 illustrates changes in diffusion coefficient above-described mechanism, regardless of the presence of of EGFR in HEK293, HELA, and CHO-K1 by the above PTX, B2-AR has no change in diffusion coefficient, and FZD described method, and like the results obtained with COS7 increases in diffusion coefficient after the treatment with cells, in all cell lines, a reduced change indiffusion coefficient PTX. of EGFR caused by cetuximab was able to be measured. 0121 FIG. 7 illustrates changes in diffusion coefficients of Comparative Example 1 PMT, FZD1, B2-AR before and after treatment with PTX. 0122 FIG. 8 shows absolute values of mean diffusion Measurement of Binding with Ligand: Western coefficients of PMT. B2-AR, and EGFR, in which since Blotting B2-AR is faster than EGFR, the binding to B2-AR is more I0129. As a comparative example to analyze the binding sensitively measured than EGFR. between a target membrane protein and a ligand, western 0123. As shown in FIGS. 7 and 8, the method can be even blotting was carried out. applied to GPCR without limitation to EGFR. This means 0.130 Ligand mAb 199.12 was purchased from Invitro that it can be applied to other membrane proteins, and also can gen, ligand mAb 528/mAb R-1 was purchased from Santa detect intracellular binding as well as extracellular binding. Cruz, anti-pEGFR (Y1068) antibody was purchased from Also, FIGS. 7 and 8 show that the method can detect the Invitrogen, and anti-total EGFR and anti-pErk1/2 antibodies binding between EGFR and cetuximab, and the dissociation were purchased from Signaling, and an anti-actin antibody of FZD1 from the Gi-protein. was purchased from MP biomedical. US 2016/0116468 A1 Apr. 28, 2016

0131 COS7 cells were lysed with radioimmunoprecipita expressing EGFR-mEos3.2 were treated with the corre tion (RIPA) buffer. The cell lysate was loaded in a 6 to 16% sponding antibodies were calculated by Equation 3, and the gradient SDS-PAGE gel to allow proteins to migrate in an results are shown in FIG. 13. electrical field. The proteins separated in the SDS-PAGE gel 0.138 FIG. 13 illustrates changes in diffusion coefficient, were transferred to a nitrocellulose membrane, treated with a in which PMT do not specifically bind with cetuximab, mab primary antibody at 4°C. overnight, and then treated with a 199.12, mab 528 and mAb R-1, and thus there is no change secondary antibody tagged with horseradish peroxidase in diffusion coefficient before and after the treatment with the (HRP) or IRDye800CW (Li-COR) at room temperature for 1 antibodies, but as EGFR specifically binds with all of cetux hour. The presence and amount of the proteins were detected imab, mAb 199.12, mAb 528, and mAb R-1, the diffusion using chemiluminescence (ECL system, Pierce) or infra-red coefficient is changed by about 40% before and after the fluorescence (Odyssey system, Li-COR), and the results are treatment. shown in FIG. 12. (0.139. As shown in FIG. 13, PMT has no change in diffu (0132 FIG. 11 illustrates cetuximab, mAb 199.12, mAb sion coefficient before and after the treatment with the anti 528 and mAb R-1 which can bind to EGFR at different sites. bodies, but EGFR is changed in diffusion coefficient by about 0.133 FIG. 12 illustrates western blotting results, in which 40% before and after the treatment with the antibodies. This NT represents no treatment with an antibody, and EGF--/- shows that the change in diffusion coefficient measured by the represents treatment/non-treatment of EGF. As shown in FIG. above-described method is influenced by direct binding, and 12, cetuximab and mAb 528 prevent phosphorylation of is capable of being measured regardless of binding sites. EGFR, and also prevent phosphorylation of Erk1/2 in a dif These advantageous characteristics cannot be obtained using ferent manner, and mAb 199.12 and mAb R-1 do not effec fluorescence resonance energy transfer (FRET) and protein tively prevent the phosphorylation of EGFR. In detail, EGFR fragment complementation assay (PCA). is activated by EGF, and plGFR may be detected using an antibody that can specifically bind to the activated EGFR. Comparative Example 2 0134 Since EGFR is not activated without EGF, it was not detected using pEGFR antibody, but in an EGF-existing envi Molecular Specificity Distinction Test: Flow ronment, EGFR is activated by EGF, and the activated EGFR Cytometry was detected using the pEGFR antibody, and therefore the same results as described are shown. pErk1/2 is a main 0140. As a comparative example to analyze a molecular marker to confirm downstream signaling during signal trans specificity distinction test between a target membrane protein duction, which maintains a minimal signal transduction in the and a ligand, flow cytometry was carried out. EGF-free environment, and increases in the amount in the EGF-existing environment, and therefore it can be seen that 0141 Alexa Fluor 647 fluorescent molecules were conju the downstream signal transduction is effectively performed. gated to immunoglobulins (anti-Human (A21445)/anti Mouse(A21235). Invitrogen), cetuximab, trastuzumab 0135 Actin is one of the main proteins composing a cell, (Roche), and an anti-actin antibody (691001), MP Biomedi which serves as a marker to confirm that a similar amount of cals) to be employed for flow cytometry, and the flow cytom proteins are used in each experiment when several western etry was performed using Gallios manufactured by BD bio blotting analyzes are performed. science. In detail, COS7 cells were treated with the Alexa 0136. When EGF, and cetuximab or mAb 528 are present Fluor 647-conjugated antibody to induce binding with a at the same time, EGF binding site on EGFR is interfered with membrane protein, and a fluorescent signal attached to the cetuximab and mAb 528, and as a result, it inhibits the EGFR surface of the cell was observed by flow cytometry, and the activation, whose tendency is seen through western blotting. results are shown in FIG. 15. However, mAb 199.12 and mAb R-1 do not show the ten dency of inhibiting EGFR activation in the EGF-existing 0.142 FIG. 14 illustrates cetuximab specifically binding to environment through the western blotting result, and thus it EGFR, and trastuzumab specifically binding to ErbB2. can be noted that mAb 199.12 and mAb R-1 do not have an 0.143 FIG. 15 illustrates flow cytometry results, in which influence on the EGFR activation caused by EGF. the X axis represents the intensity of a fluorescent signal, and the y axis represents a relative ratio of the cell populations Example 7 having a specific level of intensity of a fluorescent signal. That is, as the graph goes to the right, the fluorescent signal becomes stronger, which is interpreted that more fluorescent Measurement of Bonding with Ligand dyes are attached to the cell surface. 0137 The host cells expressing PMT-mEos3.2 and the 0144. As shown in FIG. 15, it can be seen that cetuximab host cells expressing EGFR-mEos3.2. which were obtained or trastuzumab is relatively well attached to the cell surface in Example 1.3, were treated with cetuximab, mAb 199.12, (the immunoglobulin or actin serves as a control that does not mAb 528, and mAb R-1, and diffusion coefficients were specifically bind to a membrane protein present on the cell measured by the method described in Example 1.5. In detail, Surface), but this approach cannot show to which membrane images were obtained in a non-treatment state (about 10 protein molecules cetuximab or trastuzumab binds. That is, cycles), and after each type of the antibodies was mixed with the flow cytometry can reveal the presence or absence of a the cell culture medium up to the final concentration of 20 corresponding membrane protein using a fluorescent mol ug/ml to induce binding of the antibody to a membrane pro ecule-labeled antibody for each of EGFR and ErbB2 endog tein present on the cell Surface, images were obtained (about enously expressed in COS7 cells, but cannot reveal that, 10 cycles). Changes of diffusion coefficient before and after especially, cetuximab and trastuzumab specifically bind to the host cells expressing PMT-mEos3.2 and the host cells EGFR and ErbB2, respectively. US 2016/0116468 A1 Apr. 28, 2016

Example 8 binding aptamer), and changes in diffusion coefficient of EGFR were measured before and after the treatment by the Molecular Specificity Distinction Test same method as described in Example 1.5, and the results are shown in FIGS. 18 and 19. (0145. Diffusion coefficients of PMT and EGFR were mea 0151 FIG. 17 is a diagram illustrating the bindings sured for the host cells expressing PMT-mEos3.2 and the host between the four different ligands with different molecular cells expressing EGFR-mEos3.2 obtained in Example 1.3 by weights and EGFR, which are made at the same binding site. the method described in Example 1.5, and in detail, each 0152 FIG. 18 illustrates changes in diffusion coefficients group of the host cells were treated with cetuximab, trastu of EGFR and PMT before and after the host cells expressing Zumab or an anti-actin antibody, and a change in diffusion PMT-mEos3.2 and the host cells expressing EGFR-mEos3.2 coefficient was measured by Equation 3 using the values of were treated with cetuximab, cetuximab Fab, cetuximab the diffusion coefficients before and after the treatment with F(ab'), and EGFR-specific binding aptamer. the antibody. The results are shown in FIG. 16. (O153 FIG. 19 illustrates that the molecular weight of a 0146 FIG. 16 is a graph illustrating changes in diffusion ligand and the change in diffusion coefficient have a linear coefficients before and after the host cells expressing PMT relationship. mEos3.2 and the host cells expressing EGFR-mEos3.2 were 0154 As shown in FIGS. 17 and 19, it was confirmed that treated with the cetuximab, trastuzumab, and anti-actin anti as the molecular weight of a ligand was increased as 17 kDa, body, which were measured by the above-described method. 50 kDa, 90 kDa, and 150 kDa, the change in diffusion coef 0147 As shown in FIG. 16, when the diffusion coeffi ficient of EGFR was also increased by about 7%, 16%, 27%, cients of PMT and EGFR were measured by the above-de and 38%, and it can be noted that such relationship is linear. scribed method, PMT had no change in diffusion coefficient In detail, when the diffusion coefficients were calculated by by the treatment with the cetuximab, trastuzumab, or anti Equations 1 and 2, the molecular weight of a corresponding actin antibody, but it was able to be seen that EGFR was ligand can be estimated using the fact that the diffusion coef changed in diffusion coefficient due to cetuximab. In detail, ficient becomes Smaller as the molecular weight of the ligand PMT has almost uniform diffusion coefficients regardless of is increased. the treatment of the antibodies as shown in the previous experimental results, and thus has a very Small change in Example 10 diffusion coefficient. However, since cetuximab specifically binds to EGFR, EGFR has a change in diffusion coefficient, Measurement of Dissociation Constant but it does not show any changes in diffusion coefficient by trastuzumab and the anti-actin antibody which do not specifi 0.155 10.1 Measurement of Dissociation Constant of cally bind to EGFR. EGFR-WT 0148 Compared to the flow cytometry results obtained in 0156 When there is insufficient treatment with cetux Comparative Example 2, the flow cytometry shows that imab, EGFRs are classified into two states: a cetuximab cetuximab and trastuzumab bind to membrane proteins bound EGFR state and an unbound EGFR state. If these two present on the cell surface, but cannot determine whether this different EGFR states exist at the same time, a ratio of the two binding is specific. However, when the diffusion coefficient is groups can be estimated by measuring a diffusion coefficient measured by the method of the present invention, it can be of EGFR by the method described in Example 1. Since the seen that EGFR is specifically bound with cetuximab, not ligand-bound and unbound groups have different sizes, a trastuzumab. difference between the diffusion coefficients of EGFR in these groups can be distinguished by the method described in Example 9 Example 1, and therefore the dissociation constant between EGFR and cetuximab in a single living cell can be quantita Sensitivity to Molecular Weight of Measuring tively measured using this ratio. Method of Target Membrane Protein 0157 To demonstrate it, a diffusion coefficient of EGFR was measured in each environment in which the COS7 cells 0149 When a target membrane proteinbinds with a ligand expressing EGFR-mEos3.2 prepared in Example 1.3 was whose molecular weight is not exactly known, as well as the treated with cetuximab at concentrations increased gradually ligand whose molecular weight is known Such as cetuximab, from 0.02, 0.04, 0.08, 0.16, 0.32, 0.64 up to 1.28ug/ml, which to prove that information about a molecular weight of the is the saturation level. The obtained diffusion coefficients of ligand which has not been known can be estimated from a EGFR were converted into bound/unbound ratios between degree of change in diffusion coefficient, materials having EGFR and cetuximab using Equations 4 to 8. total four different molecular weights (17 kDa to 150 kDa) 0158. In the following Equations 4 to 8, U and B are such as cetuximab (150kDa), cetuximab Fab (50kDa), cetux random variables of a diffusion coefficient of ligand-unbound imab F(ab') (90 kDa)) and an EGFR-specific binding or ligand-bound target membrane proteins, respectively. For a aptamer (17 kDa) were bound with EGFR, and then an experi simple model, it is assumed that there is no conversion ment of observing changes in diffusion coefficient of EGFR between the two states, and the concentration of a ligand is was performed. represented as c. 0150 Cetuximab was purchased from Merck Serono, cetuximab Fab and F(ab')2 were prepared using preparation Equation 4 kits (Pierce, 44685 and 44688), and the EGFR-specific bind 10159) M is the diffusion coefficient of a target mem ingaptamer was made using SELEX. As shown in Example brane protein when the concentration of a ligand is c, 1.5, COS7 cells transfected with EGFR-mEos3.2 were (0160 Oc is the ratio of ligand-unbound membrane pro treated with each of the four prepared ligands (cetuximab, teins to the total target membrane proteins, when the concen cetuximab Fab, cetuximab F(ab'), and the EGFR-specific tration of a ligand in one cell is c, and US 2016/0116468 A1 Apr. 28, 2016

(0161 Bc is the ratio of ligand-bound target membrane imab, and it can be confirmed that the dissociation constant proteins to the total target membrane proteins, when the con between EGFR and cetuximab in a single living cell can be centration of a ligand in one cell is c, and Clc +fc is 1. quantitatively measured using this ratio. 0162 U is the diffusion coefficient of a ligand-unbound 0.178 10.2 Measurement of Dissociation Constant of target membrane protein, Mutant Form, EGFRVIII 0163 B is the diffusion coefficient of a ligand-bound tar 0179 The binding between a ligand and a fluorescence get membrane protein. labeled target membrane protein can only be observed by the method described in Example 1. In other words, even in an Equation 5 environment in which a variety of membrane proteins are (0164] E(M) is the mean diffusion coefficient of a tar mixed, only the fluorescence-labeled target membrane pro get membrane protein when the concentration of a ligand is c, tein can be observed. 0.165 E(U) is the mean diffusion coefficient of a ligand 0180. To demonstrate this, an experiment was performed unbound target membrane protein, by a method of treating EGFRVIII (SEQID NO: 8), which is 0166 E(B) is the mean diffusion coefficient of a ligand a mutant form of EGFR from which a part of an extracellular bound target membrane protein. domain is deleted, compared to EGFR, with cetuximab and 0167. Therefore, the ratio of a ligand-bound membrane mAb R-1 antibody. In detail, transfected cells were prepared protein leads to the following Equation 6. by the method described in Example 1, except that, instead of EGFR-mEos3.2. EGFRVIII-mEos3.2 was expressed in COS7 cells, and PMT-expressed COS7 cells were prepared by the same method as used above. Also, the experiment was Equation 6 performed with mAb R-1 that binds to EGFR but does not bind to EGFRVIII by the same method as described in (0168 D is the diffusion coefficient of a target mem Example 10.1, and the results are shown in FIG. 24. brane protein when the ligand concentration in one cell is c. 0181 FIG. 23 illustrates that EGFRWT can bind to both Therefore, the dissociation constant (K) is defined as Equa cetuximab and the mab R-1 antibody, but the mutant form, tion 7. EGFRVIII, only binds with cetuximab, not mAb R-1. FIG. 24 Equation 7 illustrates that since PMT does not have the binding of two 0169 L is the concentration of a ligand, antibodies such as cetuximab and mAb R-1, a reduced diffu 0170 R is the concentration of a target membrane pro sion coefficient of PMT is not shown, but EGFRVIII binds tein (receptor), with cetuximab, but not with mAb R-1. 0182. As shown in FIG. 24, it can be confirmed that a 0171 ILR is the concentration of the binding complex of reduced diffusion coefficient of EGFRVIII due to cetuximab a ligand and a target membrane protein. is obvious, but reduced diffusion coefficient of EGFRVIII due 0172 Equation 8 may be obtained by Equations 6 and 7. to mAb R-1 is insignificant. This means that EGFR VIII, (DC-)-Doo)/(Dec.--Doo)-C(Ki-C) Equation 8 which is a mutant made by deleting a mAb R-1 binding site 0173 Cooperativity as well as a dissociation constant from EGFRWT, binds to cetuximab, but not to mAb R-1. between EGFR and cetuximab was confirmed by Equations 4 0183. The dissociation constant between EGFR VIII and to 8 from the bound/unbound ratio between EGFR and cetux cetuximab was measured by a dose-dependent test, and in imab at a single cell level. In addition, it was confirmed that detail, the diffusion coefficient of EGFRVIII was measured by the dissociation constant measured as described above and the method described in Example 1.4 while the concentration the cooperativity values obtained from a scatchard plot were of cetuximab was increased as 0.02, 0.04, 0.08, 0.16, 0.32, similar to those obtained by an actual experiment carried out 0.64, 1.28 g/ml by the same method as described in Example in vitro, and the detail results are shown in FIGS. 20 to 22. 1.4, the dissociation constant between EGFRVIII and cetux 0.174 FIG. 20 is a diagram illustrating that cetuximab imab was measured as described in Example 10.1 using the binding to EGFR increases in number as the concentration of ratio of binding with cetuximab and analyzed by plotting a cetuximab is increased. scatchard plot, and then the results are shown in FIG. 25. (0175 FIG. 21 illustrates the reduction of a diffusion coef 0.184 FIG. 25 illustrates the dissociation constant and ficient of EGFR as the concentration of cetuximab in a single cooperativity between EGFRVIII and cetuximab using the one cell is increased from 0.02 to 1.28 ug/ml. A reduced value method described in Example 10.1. As shown in FIG. 25, it of the diffusion coefficient of EGFR was normalized by the was able to be confirmed from the scatchard plot that there is value in non-treatment condition. no cooperativity in the binding between EGFRVIII and cetux 0176 FIG. 22 is a graph illustrating a ratio of cetuximab imab. binding to EGFR as the concentration of cetuximab is 0185. As shown in FIGS. 20 to 25, the method can obtain increased based on the result of FIG. 21, and thereby the the dissociation constant and cooperativity of a ligand spe dissociation constant between EGFR and cetuximab can be cifically binding to a target membrane protein to be observed calculated. Also, the inner graph of FIG.22 shows that there in a live cell. is no cooperativity in the binding of cetuximab according to the scatchard plot. Example 11 0177. As shown in FIGS. 20 to 22, when the diffusion Analysis of Process of Forming Complex of Target coefficient of EGFR was measured by the above-described method, the size of the ligand-bound or ligand-unbound Membrane Protein group is not the same as each other, and therefore a difference 0186. When the diffusion coefficients of target membrane between the diffusion coefficients of EGFR can be distin proteins including EGFR are measured by the method guished by the degree of binding between EGFR and cetux described in Example 1, the size of a molecule becomes larger US 2016/0116468 A1 Apr. 28, 2016 during the formation of a complex of the target membrane bodies targeting EGFR. A target protein may be regulated by proteins, and is directly connected to the diffusion coefficient. antibody binding, but mAb 199.12 is known as an antibody Therefore, the process of forming the complex may be known insignificantly affecting the activity of EGFR. based on the diffusion coefficient. 0195 12.1 Preparation of Fab Fragment 0187 Host cells expressing EGFR-mEos3.2 prepared in 0.196 Fab fragments were generated using a Fab prepara Example 1.3 were treated with cetuximab, a secondary anti tion kit (Pierce, 44685). First, a resin conjugated with a pro body (goat anti-human immunoglobulin (IgG) antibody (81 teinase, papain, was mixed with an immunoglobulin antibody 7100), invitrogen) and a tertiary antibody (rabbit anti-goat in a buffer effective for an enzyme reaction to occur and the immunoglobulin (A10537), invitrogen) in turn, and in order reaction was conducted at 37° C. for 5 to 6 hours. After the to figure out the process of forming the complex, the cetux reaction was completed, the antibody that is cleaved into two imab, secondary antibody and tertiary antibody that sequen Fab fragments and an Fc fragment and the papain resin were tially bind to EGFR as shown in FIG. 26 are treated in turn to centrifuged to separate only a Supernatant. A protein A-con measure the diffusion coefficient of EGFR by the method jugated resin and the Supernatant were mixed, and then cen described in Example 1.5. The results are shown in FIG. 27. trifuged again, thereby obtaining a purified Fab fragment. 0188 FIG. 26 illustrates sequentially binding of the cetux imab, secondary antibody and tertiary antibody to the host 0.197 12.2 Binding Between Antibody and Organic Fluo cells expressing EGFR-mEos3.2, and FIG. 27 illustrates the rescent Dye diffusion coefficients of EGFR measured in the order of treat 0198 Antibodies were conjugated with an organic fluo ing the host cells expressing EGFR-mEos3.2 with the cetux rescent dye, Alexa Fluor 647, using an Alexa Fluor R 647 imab, secondary antibody and tertiary antibody. antibody labeling kit (Invitrogen). The antibodies were added 0189 As shown in FIG. 27, when the host cells expressing in a 0.1M sodium bicarbonate solution, and mixed with a EGFR-mEos3.2 were sequentially treated with the cetux fluorescent dye to which functional groups such as a Succin imab, secondary antibody and tertiary antibody, the diffusion imidyl ester or tetrafluorophenyl ester was tagged to allow a coefficients of EGFR also tended to be reduced in the same reaction for 1 hour, thereby forming a stable complex order. The size of the final complex formed by binding the between the fluorescent dye and a primary amine group in the cetuximab, secondary antibody and tertiary antibody to the antibody. It were loaded on a purifying resin to bind remain target membrane protein was about 600 kDa or more, and the ing non-binding fluorescent dyes to the resin, and then cen reduction of the diffusion coefficient of EGFR caused by trifuged to obtain a Supernatant, from which only the antibod binding the tertiary antibody to the cetuximab and secondary ies conjugated with the fluorescent dyes were isolated. antibody complex was about 12%. As a result, it can be seen 0199. 12.3 Preparation of Coverslip and Cell Seeding that Such a complex forming process can also be applied to a 0200 Intact cells without any treatment were seeded on a larger complex having a size of 600 kDa or more. 25-mm or 18-mm-diameter coverslip. These coverslips were (0190. Contrarily, diffusion coefficients of EGFR were prepared by procedures of Sonication in acetone at 42°C. for measured by sequentially treating the host cells expressing 30 minutes to minimize autofluorescence of the coverslip, EGFR-mEos3.2 with the cetuximab, tertiary antibody and washing with distilled water three times, Sonication again in secondary antibody by the same method as described in 1% hydrofluoric acid at 42°C. for 10 minutes, washing with Example 1.5, and the results are shown in FIG. 28. distilled water about 15 to 20 times to completely remove (0191 FIG. 28 illustrates the diffusion coefficients of hydrofluoric acid, and finally soaking in 100% ethanol and EGFR measured by sequentially treating EGFR with the exposure to UV light for 30 minutes to sterilize. cetuximab, secondary antibody and tertiary antibody. 0.192 As shown in FIG. 28, when the EGFR was sequen 0201 When the cells were seeded onto the prepared cov tially treated with the cetuximab, tertiary antibody and sec erslip, the coverslip was treated with a Surface coating mate ondary antibody, a complex was formed as soon as the sec rial to facilitate cells to be attached to the coverslip. As the ondary antibody binds to it, and therefore at this moment, a Surface coating material that can be used herein, collagen, dramatic change in diffusion coefficient of EGFR, for fibronectin, gelatin or poly-L-lysine may be used. example, about 30%, was shown. (0202 12.4 Fixation of the Cell-Seeded Coverslip onto (0193 Comparing FIGS. 27 and 28, when EGFR was Microscope treated with the cetuximab and then the tertiary antibody, the 0203 The coverslip on which cells were seeded as diffusion coefficient of EGFR was not changed, but the sec described in Example 12.3 was placed onto a microscope ondary antibody first bound with the tertiary antibody, and with a live cell chamber (37 C, 5% CO) and an electron then bound with the cetuximab, and therefore a much larger multiplying charge coupled device (EM-CCD) to observe the reduction in diffusion coefficient of EGFR was shown. Such cells at a single cell level. The microscope, Olympus IX71 specificity shows that the above-described method may model, is designed based on total internal reflection fluores become an effective means to study the process of forming the cence (TIRF), and therefore can observe only fluorescent complex of target membrane proteins. molecules at a plasma membrane of a cell closely attached (within about 200 nm from a coverslip) to the round coverslip. Example 12 0204 12.5 Treatment with Organic Fluorescent Dye Binding Antibody Observation of Target Endogenous Membrane 0205 Endogenous membrane proteins expressed in the Protein: Endogenous EGFR Expressed in COS7 cells seeded as described in Example 12.4 were specifically Cells labeled with fluorescent dye-conjugated Fabs prepared in 0194 EGFR-specific binding was observed using a mouse Examples 12.1 and 12.2. A degree of labeling the target (monoclonal) Anti-Human Epidermal Growth Factor Recep proteins may be regulated by changing the concentration and tor (mAb 199.12 (Invitrogen)) among various types of anti treatment time of the fluorescent dye-conjugated Fab. US 2016/0116468 A1 Apr. 28, 2016

0206. After the endogenous protein was sufficiently body is inhibited by the high concentration of cetuximab to labeled, unbound fluorescent dye-conjugated Fabs present in which a fluorescent dye is not conjugated, and therefore a the solution were removed by washing with a growth medium reduced fluorescent signal was detected. 2 to 3 times. 0214 FIG. 31 illustrates data obtained by observing a 0207 To induce photoswitching and inhibit photobleach specific binding aspect between the endogenous EGFR and ing of an organic fluorescent dye under the condition in which cetuximab. Fab of the mab 199.12 antibody specifically cells were alive, 1 mM f-mercaptoethylamine (MEA), 0.2 binding to EGFR was purified, labeled with Alexa Fluor 647 u/ml protocatechuic acid (PCA) and 2.5 mM protocat fluorescent dye, and then added to a single living COS7 cell to echuate-3,4-dioxygenase (PCD) were added to a cell culture bind to the endogenous EGFR expressed in the cell. A diffu medium. sion coefficient was obtained before the treatment with an 0208 12.6 Observation of Endogenous Membrane Pro immunoglobulin antibody and cetuximab, and a diffusion tein to which Ligand and Interaction Partner Did not Bind coefficient changed by the binding between the EGFR and the 0209. The fluorescence-labeled endogenous EGFR pre immunoglobulin antibody or cetuximab was calculated after pared in Example 12.5 was illuminated with 642 nm laser to the treatment with the immunoglobulin antibody and cetux temporarily turn off the fluorescence of Alexa Fluor 647 to imab. It can be observed that, since the immunoglobulin make it possible to observe at a single molecule level, the antibody does not specifically bind to EGFR, there is almost electron state of the fluorescent molecule was changed with no change in diffusion coefficient, but since cetuximab spe 405 nm laser to allow to randomly turn on the fluorescence, cifically binds to EGFR, the diffusion coefficient is changed and then fluorescent images were sequentially obtained with by about 40%. 642 nm laser at regular intervals of time (about 50 ms) by an 0215 12.8 Identification of Endogenous EGFRs electron multiplying charge coupled device (EM-CCD) Expressed in Various Cell Lines ixon3 897, manufactured by Andor Technology, which can 0216) To observe the change in diffusion coefficient of detect a signal transmitted from a single fluorescent molecule. endogenous EGFRs expressed in various cell lines by cetux A signal of the single molecule present on each image was imab, COS7, A431, MDA-MB-231, BT20, HCC827 cell found, and then the images were compared to each other, lines were prepared as described in Example 12.4, endog thereby tracking and measuring the moving distance and enous EGFRs expressed in the cells were labeled with fluo trajectory of the fluorescent protein per unit time, and a mean rescent dye using Fab fragments as described in Example square displacement (MSD) was calculated by Equation 1. 12.5, and a diffusion coefficient under a non-treatment con 0210 FIG. 29 illustrates a process of detecting endog dition was measured as described in Example 12.6. After enous EGFR by conjugating the photoswitchable fluorescent ward, the same cells as previously used were treated with an organic dye, Alexa Fluor 647, to the Fab fragment that can immunoglobulin antibody or cetuximab, and diffusion coef specifically bind to the endogenous EGFR (left). This can be ficients were measured by the same method as described observed because, actually, Alexa Fluor 647 is photoactivated above. Changes in diffusion coefficient were calculated by when binding to EGFR of a cell and emits a fluorescent signal Equation 3, compared to that under the non-treatment condi as shown in the middle image. As the position information tion, and the results are shown in FIG. 32. and trajectory of the fluorescent signal emitted from each 0217 FIG. 32 illustrates the quantitative results of means single molecule are tracked down until the fluorescent signal and standard deviations of diffusion coefficients changed by is photobleached, a diffusion coefficient may be obtained. the immunoglobulin antibody and cetuximab binding to five The right image illustrates the trajectories of tracked 10,000 different cell lines. Compared with the diffusion coefficient of single molecules. the endogenous EGFR in each cell under the non-treatment 0211 12.7 Observation of Endogenous Membrane Pro condition, there were almost no changes in diffusion coeffi tein to which Ligand and Interaction Partner Bind cient before and after the immunoglobulin antibody was 0212. The cells in which the endogenous EGFR was treated, but after the treatment with cetuximab, all cells were labeled with the fluorescent dye, prepared in Example 12.5, changed in diffusion coefficient by about 40%. Specifically, were treated with 20 g/ml of cetuximab. As shown in FIG. the diffusion coefficients may be obtained by Equations 1 and 30, it was confirmed that fluorescent dye-conjugated mAb 2, and provided that the diffusion coefficients under different 199.12 and cetuximab have different EGFR-specific binding conditions are D and D, a ratio of the change in diffusion sites, they do not affect each other. The cells before and after coefficient is calculated by Equation 3. With the treatment of treatment with cetuximab were observed by the method the immunoglobulin antibody, D and D values are almost described in Example 12.6, diffusion coefficients before and the same, which means that the change in diffusion coefficient after the treatment were calculated by Equations 1 and 2, and is very small. However, with the treatment of the cetuximab, means and standard deviations thereof are shown in FIG. 31. D value is about 60% of D value, which means that the 0213 FIG. 30 illustrates the result of confirming that the diffusion coefficient is reduced by about 40%. bindings of mAb 199.12 and cetuximab to the endogenous Example 13 EGFRs are not influenced by each other. 1 g/ml each of cetuximab and mAb 199.12 antibody, which were labeled Observation of Endogenous Membrane Protein in with a fluorescent dye, and 100 lug/ml of cetuximab, which Primary Cell Originating from Animal Model: was not labeled with a fluorescent dye, were mixed to allow binding to the endogenous EGFR expressed in the COS7 cells Observation of Endogenous EGFRs in Various Types for 30 minutes, and a fluorescent signal depending on a bind of Primary Cells Isolated from Mouse's Lung ing degree was detected by flow cytometry. The flow cytom 0218 13.1 Preparation of Cell Marker-Specific Antibody etry results show that the binding degree of the mab 199.12 to which Fluorescent Dye is Conjugated antibody is not interfered even with the 100-fold higher con 0219. Since, unlike a cell line, cells purified from tissues centration cetuximab, but the binding of the cetuximab anti include a mixture of various types of cells, cells were labeled US 2016/0116468 A1 Apr. 28, 2016

with primary antibodies specific to various cell membrane Bender Kim, et al., Cell, Vol. 121,823-835, Jun. 17, 2005), markers (Sca-1, PECAM) and fluorescent dye-conjugated four different types of cells that are mixed in lung cell groups secondary antibodies binding to the primary antibodies to separated from one mouse model were classified, and then a determine the type of a cell to be observed. Here, to enable diffusion coefficient of endogenous EGFR in each cell type observation using lasers with different wavelengths, a type of was measured, and a change in diffusion coefficient by cetux the fluorescent marker was different depending on the marker imab was measured. (Pacific blue, Alexa Fluor 488, Dylight 549). 0224 FIG. 33 illustrates images of the mouse primary cells classified by markers for the cell types in the lung cell 0220 13.2 Purification of Primary Cells from Mouse's tissue and autofluorescence levels. DIC is an image showing Lung Tissues and Cell Seeding an original cell shape, and all of Sca-1, Pecam and autofluo 0221) A mixed liquid of 0.5 mg/ml Collagenase type II rescence are fluorescent images. Celli1 in which Sca-1 was (Worthington) and 1 mg/ml Collagenase type IV (Worthing very highly expressed and Pecam was rarely expressed was ton) was injected into a lung tissue of a 8-week-old C57BL/6 classified as a Bronchio-alveolar stem cell (BASC), and mouse model through a respiratory tract to dissolve the tissue, celli2 in which all of Sca-1 and Pecam were expressed was and the separatedlung tissue was put into a collagenase mixed able to be classified as an endothelial cell. Celli3 showed very liquid at 37°C. for 30 minutes. After neutralization of colla low expression of Sca-1 and Pecam but very high autofluo genase using PBS, and the tissue sample was centrifuged to rescence at a 488 nm channel, and therefore it is considered as obtain only cells, which was strained with a cell strainer (BD an AT2 cell, and celliiA in which Sca-1 and Pecam were rarely falcon) to separate single cells. The separated cells were expressed and the autofluorescence level was very low is grown in a growth medium treated with a fibroblast growth considered as a Clara cell. inhibitor for 24 hours, and seeded on an 18-mm-diameter 0225 FIG. 34 illustrates the results of observing changes coverslip and fixed on a microscope by the methods described in diffusion coefficient depending on patterns of binding in Examples 12.3 and 12.4. cetuximab to endogenous EGFRs expressed in the four types 0222 13.3 Classification of Primary Cells by Cell Label of the cells classified in FIG.33, such as the BASC, endothe ing lial cell, AT2, and Clara cells. It was seen that the EGFRs 0223) The fixed primary cells of murine lung were treated expressed in the BASC cell and the Clara cell show a great with a pacific blue fluorescent dye-conjugated antibody change in diffusion coefficient due to cetuximab, for against Sca-1 protein (Invitrogen), a Dylight 549-conjugated example, by about 40%, but the endothelial cell and the AT2 secondary antibody which binds to primary antibody against cell show the change in diffusion coefficient by less than 10%. Pecam protein (Santa cruz), and an Alexa Fluor 647 fluores 0226. It would be understood by those of ordinary skill in cent dye-conjugated anti-EGFR protein antibody. Using the art that the above descriptions of the present invention are lasers with different wavelengths suitable for respective fluo exemplary, and the exemplary embodiments disclosed herein rescent dyes (405, 488, 561, and 633 nm), the autofluores can be easily modified into other specific forms without cence level, protein expression level and motion of each pro changing the technical spirit or essential features of the tein were observed by the method described in Example 12.6. present invention. Therefore, it should be interpreted that the Based on the results, compared with markers for the previ exemplary embodiments described above are exemplary in ously known types of cells in the lung cell tissue (Carla F. all aspects, and are not limitative.

SEQUENCE LISTING

<16 Os NUMBER OF SEO ID NOS: 8

<21 Os SEQ ID NO 1 &211s LENGTH: 3630 &212s. TYPE: DNA <213> ORGANISM: Homo sapiens

<4 OOs SEQUENCE: 1 atgcgaccct Cogggacggc cggggcagcg ct cotggcgc tigctggctgc gctctgc.ccg 60

gcgagtcggg Ctctggagga aaagaaagtt to Caaggca cagtaacaa gCtcacgcag 12O ttgggcactt ttgaagat.ca ttittcticago: ct coagagga tigttcaataa citgtgaggtg 18O

gtc.cttggga atttggaaat tacctatgtg cagaggaatt atgat ctitt c cttcttaaag 24 O

accatcCagg aggtggctgg ttatgtcCt c attgcc.ctica acacagtgga gcaattic ct 3 OO

ttggaaaacc togcagat cat cagaggaaat atgtact acg aaaatticcita toccittagca 360

gtc.ttatcta actatgatgc aaataaaacc ggactgaagg agctgcc.cat gagaaattta 42O

Caggaaatcc ticatggcgc cgtgcggitt C agcaacaac C ctgc cctgtg caacgtggag 48O

agcatcCagt gg.cggga cat agt cagcagt gactittctica gcaa.catgtc gatggactitc 54 O

Cagaaccacc tdggcagctg ccaaaagtgt gatccaa.gct gtcc caatgg gagctgctgg 6 OO

US 2016/0116468 A1 Apr. 28, 2016 22

- Continued gcttgcattgatagaaatgg gctgcaaagc tigt cc catca aggaaga cag Cttcttgcag 24 OO cgatacagct Cagaccc.cac aggcgc.cttg actgaggaca gCatagacga Cacct tcct c 246 O Ccagtgcctgaatacatalaa C cagt ccgtt cccaaaaggc cc.gctggct C ttgcagaat 252O cctgtctato acaatcagcc tictdalacc cc gcgcc cagoa gag acccaca citaccaggac 2580 c cccacagca citgcagtggg caa.ccc.cgag tat ct caa.ca citgtc.ca.gcc cacctgttgtc 264 O alacagcac at t cacagc cc tic cc actgg gcc.ca.gaaag gcagccacca aattagcctg 27 OO gacaac cctd actaccagca ggacittctitt cocaaggaag ccaa.gc.caaa toggcatctitt 276 O aagggcticca cagctgaaaa to agaatac Ctaagggtc. c9ccacaaag cagtgaattit 282O attggagca 2829

1. A method of analyzing a binding aspect between a can particle, N is the total number of a pair of coordinates in didate material and a target membrane protein in a living cell, one trajectory for the target membrane protein particle, the method comprising: (x, y) are the coordinates of the target membrane pro obtaining diffusion coefficients of a target membrane pro tein particle at an in"-numbered position in one trajec tein before and after treatment with a candidate material tory, (x, y) are the coordinates of the target membrane in a living cell expressing the target membrane protein, protein particle at the start point in one trajectory, (X, and analyzing a change in diffusion coefficient of at least y) are the coordinates of the target membrane protein one target membrane protein obtained thereby. particle at the end point in one trajectory, and (X, A, 2. The method of claim 1, wherein the analysis of the yA) are the coordinates of the target membrane protein binding aspect between the target membrane protein and the particle at an n+A-numbered position in one trajectory, candidate material includes the analysis of binding between however, n+A is the same as or smaller than N; and the target membrane protein and the candidate material, a MSD(A)=4DA Equation 2 ratio of target membrane proteins that the candidate material binds to among the total target membrane proteins, a molecu where D is the diffusion coefficient, and A is the step size lar weight of the candidate material bound to the target mem between coordinates of the target membrane protein brane protein, a dissociation constant between the target particle. membrane protein and the candidate material, or a process of 7. The method of claim 1, wherein the change in the diffu forming a complex between the target membrane protein and sion coefficient is obtained by Equation 3: the candidate material. Change in diffusion coefficient (%)=100* |1-(D/ 3. The method of claim 1, wherein the target membrane D2) Equation 3 protein is selected from the group consisting of an integral where D is the diffusion coefficient of the target mem membrane protein, a peripheral membrane protein, a trans brane protein at a concentration c 1 of the candidate membrane protein, a membrane glycoprotein and a lipid material in a peripheral environment of the cell at a anchored membrane protein. single cell level, and D is the diffusion coefficient of 4. The method of claim 1, wherein the candidate material the target membrane protein at a concentration c2 of the includes at least one selected from the group consisting of a candidate material in a peripheral environment of the compound, a nucleic acid, a saccharide, a carbohydrate, a cell at a single cell level. lipid, a peptide, and a protein. 8. The method of claim 7, wherein, when the change in the 5. The method of claim 1, wherein the diffusion coefficient diffusion coefficient (%) obtained by Equation 3 is 2% or of the target membrane protein is obtained by detecting the more, the candidate material is determined as a ligand binding motion of the membrane protein in a cell membrane by single to the target membrane protein. particle tracking (SPT). 9. The method of claim 5, wherein the detection of the 6. The method of claim 5, wherein the diffusion coefficient movement of the membrane protein in the cell membrane is is obtained by Equations 1 and 2: performed by detecting fluorescence signal of a fluorescent protein in a cell expressing a fusion protein of the target membrane protein and the fluorescent protein. 1 NA Equation 1 10. The method of claim 9, wherein the fluorescent protein MSD(A) = X(x,-A-X, + y, A -y, includes one or more selected from the group consisting of a =l green fluorescent protein (GFP) type, a blue fluorescent pro tein (BFP) type, a cyan type, a yellow fluorescent protein where A is the step size between coordinates of the target (YFP) type, a red fluorescent protein (RFP) type, an orange membrane protein particle, wherein A is a natural num type, a far-red type, a near-IR, a photoactivatable protein, a ber, MSD(A) is the mean square displacement of a target photoconvertible protein, and a photoswitchable protein. membrane protein particle with respect to the step size 11. The method of claim 9, wherein the fluorescent protein between coordinates of the target membrane protein includes one or more selected from enhanced green fluores US 2016/0116468 A1 Apr. 28, 2016

cent protein (EGFP), Emerald, Superfolder GFP. azami green 15. The method of claim 13, wherein the fluorescent mate mWasabi, TagGFP. AcGFP, T-sapphire, muKG, Clover, rial is an organic fluorescent dye. mNeonGreen, enhanced blue fluorescent protein (EBFP), 16. The method of claim 15, wherein the organic fluores EBFP2, AZurite, mTagBFP mKalama1, Sirius, enhanced cent dye is selected from Atto 488, Alexa Flour 488, Dy505, cyan fluorescent protein (ECFP), monomeric ECFP Rhodamine 123, Atto 520, Dy 530, ATTO 532, Alexa Fluor (mECFP), Cerulean, mTurquoise, mTurquoise2, CyPet, 532, Fluorescein, FITC, Cy2, Cy3B, Alexa Flour 568, TagCFP. mTFP1 (Teal), SCFP3A, monomeric Midorishi TAMRA, Cy3, Cy3.5, SNAP-Cell TMR-Star, Atto 565, Atto Cyan, enhanced yellow fluorescent protein (EYFP), Topaz, 590, Alexa Fluor 647, Cy5, Atto 647, Atto 647N, Dyomics Benus, mCitrine, YPet, TagYFP, PhiYFP mBanana, SYFP2, 654, Atto 655, TMP-ATTO 655, Atto 680, Cy5.5, Atto 680, mRuby, mRuby2, mApple, mStrawberry, mRFP1, mCherry, Alexa Fluor 680, Atto 700, Alexa Fluor 700, DyLight 750, mRaspberry, dKeima-Tandem (monomeric version), HcRed Cy7, Alexa Flour 750, Atto 740, Alexa Flour 790, and IRDye Tandem (monomeric version), mPlum, mKate2, mNeptune, 800 CW. mKate2, mNeptune, TagRFP657, IFP1.4, PA-GFP, PAm 17. The method of claim 1, further comprising: Cherry 1, PaTagRFP, PS-CFP2, mEos2, mEos3.2. PSmOr measuring the diffusion coefficient of the target membrane ange and Dronpa. protein after the treatment with the candidate material at 12. The method of claim 1, wherein the cell expressing the several time points, and analyzing a change in a diffu fusion protein of the target membrane protein and the fluo sion coefficient of at least one target membrane protein rescent protein is selected from the group consisting of a obtained thereby over time. human embryonic kidney (HEK) cell, HEK 293 cell, 3T3-L1 18. The method of claim 17, further comprising: cell, C6 cell, Chinese hamster ovary (CHO) cell, CHOK1 determining membrane protein-specific endocytosis by cell, NIH/3T3 cell, baby hamster kidney (BHK) cell, COS1 analyzing the change in the diffusion coefficient of the cell, COS7 cell, HaCaT cell, HeLa cell, HeLa S3 cell, HepG2 target membrane protein over time. cell, HL-60 cell, HUV-EC-C cell, Jurkat cell, K-562 cell, L6 19. A method of screening a drug for a target membrane cell, MCF7 cell, MDCK cell, NIH/3T3 cell, RAW264.7 cell, protein, comprising: RBL-1 cell, SH-SY5Y cell and U-937 cell. analyzing a binding aspect between a candidate material 13. The method of claim 5, wherein the detection of the and a target membrane protein according to any one of movement of the membrane protein in the cell membrane is claims 1 to 18; and performed with a fluorescent material-conjugated probe spe determining a drug specifically acting on the target mem cifically binding to the target membrane protein. brane protein among the candidate materials using the 14. The method of claim 13, wherein the probe is an anti binding aspect. body, an aptamer, or a non-antibody protein scaffold.