ﬁcacy of anti-ron antibody zt/g4 drug maytansinoid...

Cancer Therapy: Preclinical

Efficacy of Anti-RON Antibody Zt/g4–Drug MaytansinoidConjugation (Anti-RON ADC) as a Novel Therapeutics forTargeted Colorectal Cancer Therapy

Liang Feng1,2, Hang-Ping Yao1, Wei Wang3, Yong-Qing Zhou4, Jianwei Zhou5, Ruiwen Zhang3, andMing-Hai Wang1,2

AbstractPurpose: The receptor tyrosine kinase RON is critical in epithelial tumorigenesis and a drug target for

cancer therapy. Here, we report the development and therapeutic efficacy of a novel anti-RON antibody

Zt/g4–maytansinoid (DM1) conjugates for targeted colorectal cancer (CRC) therapy.

ExperimentalDesign:Zt/g4 (IgG1a/k)was conjugated toDM1via thioether linkage to formZt/g4–DM1

with a drug-antibody ratio of 4:1. CRC cell lines expressing different levels of RON were tested in vitro to

determine Zt/g4–DM1-induced RON endocytosis, cell-cycle arrest, and cytotoxicity. Efficacy of Zt/g4–DM1

in vivo was evaluated in mouse xenograft CRC tumor model.

Results: Zt/g4–DM1 rapidly induced RON endocytosis, arrested cell cycle at G2–M phase, reduced cell

viability, and caused massive cell death within 72 hours. In mouse xenograft CRCmodels, Zt/g4–DM1 at a

single dose of 20mg/kg bodyweight effectively delayedCRC cell-mediated tumor growth up to 20 days. In a

multiple dose-ranging study with a five injection regimen, Zt/g4–DM1 inhibited more than 90% tumor

growth at doses of 7, 10, and 15mg/kg body weight. The minimal dose achieving 50% of tumor inhibition

was approximately 5.0 mg/kg. The prepared Zt/g4–DM1 is stable at 37�C for up to 30 days. At 60 mg/kg,

Zt/g4–DM1 had a moderate toxicity in vivo with an average of 12% reduction in mouse body weight.

Conclusion: Zt/g4–DM1 is highly effective in targeted inhibition of CRC cell-derived tumor growth

in mouse xenograft models. This work provides the basis for development of humanized Zt/g4–DM1 for

RON-targeted CRC therapy in the future. Clin Cancer Res; 20(23); 6045–58. �2014 AACR.

IntroductionThe RON receptor tyrosine kinase, a member of the MET

proto-oncogene family (1, 2), has been implicated in epi-thelial tumorigenesis (3). Overexpression of RON exists invarious primary tumors, including colorectal, breast, andpancreatic cancers (4–10). In colorectal cancers (CRC),RON is overexpressed in more than 50% of cases (4, 5).Aberrant RON expression also results in generation of

oncogenic and constitutively active RON variants such asROND160 (3, 5). The consequence of these abnormalities isthe activation of various intracellular signaling pathwaysthat facilitate CRC cell growth, invasion, and chemoresis-tance (3). Overexpression of RON in CRC also has prog-nostic value in predicting patient survival and clinical out-comes (11). Thus, aberrant RON expression is a pathogenicfeature in CRC cells, which contributes to tumorigenicphenotype and malignant progression (3–5, 11–13).

The high frequency of CRC RON overexpression andthe dependency of CRC cells on RON signaling for growthprovide the rationale to target RON for therapy. Tyrosinekinase inhibitors (TKI) such as foretinib (14), BMS-777607 (15), and MK-2461 (16) that target RON andMET are currently under clinical trials (www.clinicaltrials.gov). Therapeutic monoclonal antibodies (TMA) specificto RON such as IMC-41A10, narnatumab (clinical trialID: NCT01119456), and Zt/f2 also have been evaluatedin preclinical models (17, 18). Results indicate that tar-geted inhibition of RON has a therapeutic effect ontumors mediated by colon, breast, and pancreatic cancercells in animal models (17–19). However, efficacy islimited to only about 40% to 50% (17–19). Completeinhibition of tumor growth by a single RON-targeted TKIor TMA has not been observed (14–19). Thus, there is an

1State Key Laboratory for Diagnosis and Treatment of Infectious Diseases,First Hospital of Zhejiang University School of Medicine, Zhejiang, China.2Department of Biomedical Sciences, Texas Tech University HealthSciences Center School of Pharmacy, Amarillo, Texas. 3Department ofPharmaceutical Sciences, Texas Tech University Health Sciences CenterSchool of Pharmacy, Amarillo, Texas. 4Department of Neurosurgery, FirstHospital of Zhejiang University School of Medicine, Zhejiang, China.5Department of Molecular Cell Biology and Toxicology, Nanjing MedicalUniversity School of Public Health, Jiangsu, China.

Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).

L. Feng and H.-P. Yao contributed equally to this article.

Corresponding Author: Ming-Hai Wang, TTUHSC School of Pharmacy,1406 Coulter Street, Amarillo, TX 79106. Phone: 806-414-9214; Fax: 806-356-4770; E-mail: [email protected]

doi: 10.1158/1078-0432.CCR-14-0898

�2014 American Association for Cancer Research.

ClinicalCancer

Research

www.aacrjournals.org 6045

on May 23, 2018. © 2014 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst October 7, 2014; DOI: 10.1158/1078-0432.CCR-14-0898

http://clincancerres.aacrjournals.org/

urgent need to develop and improve the efficacy of RON-targeted therapeutics.

One highly attractive strategy to enhance efficacy is totarget RON for cytotoxic drug delivery. First, RON is pref-erentially expressed in cancer cells withminimal expressionin corresponding normal epithelial cells (4–10). Also, RONis not expressed in fibroblasts, endothelial cells, and bloodleukocytes (1, 4, 7, 20). Such expressionpattern is crucial forachieving themaximaldrugdeliverywithmanageable safetyprofiles. Second,RON-specificmAbs such asZt/g4 andZt/f2rapidly induceRON internalization by cancer cells (21–24).This process requires a transient RON phosphorylation,which is essential for receptor endocytosis (21–24). Finally,anti-RON mAb-directed drug delivery, which exertsincreased cytotoxicity against cancer cells, has been provenin experimental CRC therapy (21–24). Considering theadvanced technology used in antibody–drug conjugates(ADC) for targeted cancer therapy (25), the developmentof anti-RON ADC is a promising strategy for RON-targetedtherapy. This approach should also overcome the short-comings in TKI- or TMA-targeted therapies that depend onRON signaling for the growth and survival of cancer cells.

The present study evaluates a novel anti-RON ADC forCRC therapy. Our goal is to determine whether RON-directed delivery of highly potent drug in the form of ADCis effective in inhibiting tumor growth in mouse xenograftCRC models. ADC is a combination of target-specific anti-body, highly potent compound, versatile chemical linker,and controlled drug payload. The development of anti-RON ADC provides a rational approach to evaluate theefficacy of RON-targeted therapy. To this end, we selected amouse mAb Zt/g4 highly specific to the RON extracellularsequences as the drug carrier. Zt/g4 was conjugated to

maytansinoid known as DM1 through nonreduciblethioether linkage (24). The efficacy of anti-RON Zt/g4 ADCwas evaluatedusing in vitro and in vivomodels. Theobtainedresults provide the foundation for future development ofhumanized anti-RON ADC for targeted cancer therapy.

Materials and Methods

Cell lines and reagentsCRC cell lines DLD1, LoVo, HCT116, HT29, and SW620

were from ATCC and authenticated in 2010 with cytogen-esis. HT29-luc2 andHCT116-luc2 cells expressing the fireflyluciferase gene-2 were from PerkinElmer and authenticatedin 2011 with DNA profiling and cytogenesis. Mouse anti-RON mAbs Zt/g4, Zt/c1, and rabbit IgG antibody to theRON C-terminal peptide were used as previously described(2). Goat anti-mouse IgG labeled with FITC or rhodaminewas from Jackson ImmunoResearch. Maytansinoid (DM1)and N-succinimidyl-4-[maleimidomethyl]-cyclohexanecarboxylate (SMCC) were from Concortis.

Conjugation of anti-RON mAb with DM1 throughthioether linkage

Conjugation was performed according to a protocol toachieve a drug-antibody ratio (DAR) at 4:1 (26–28). Briefly,Zt/g4 at 10mg/mLwasmixedwith 10mmol/L SMCC-DM1in a conjugation buffer to form Zt/g4–SMCC-DM1 (desig-nated as Zt/g4–DM1). The anti-RON mAb Zt/c1 also wasconjugated with SMCC-DM1 to form Zt/c1–DM1. We alsoprepared the control ADC by conjugating normal mouseIgG (CmIgG) with SMCC-DM1 to form CmIgG-DM1 asdescribed above. All conjugates were purified using a PC10Sephadex G25 column, sterilized through a 0.22 mmol/Lfilter, and stored at 4�C.

Analysis of Zt/g4–DM1 conjugation and its stabilityThe conjugation of DM1 to Zt/g4 was verified by hydro-

phobic interaction chromatography (HIC) using a VarianProstar 210 Quaternary HPLC system coupled with a TSKbutyl-NPR 4.6 � 3,5 column (Tosoh Biosciences; ref. 29).The average DARs were calculated from the integrated areasof theDAR species. Thismethod also was used to determinethe stability of Zt/g4–DM1 at 37�C.

Assay for cell surface RON expressionCell surface RON was quantitatively determined by the

immunofluorescence assay using QIFKIT reagents fromDAKO. Cells (1 � 106 cells per mL in PBS) were treatedwith Zt/g4 at saturating concentrations followed by incu-bation in parallel with the QIFIKIT beads and goat F(ab’)2F0479. After establishing a calibration curve, the number ofRON receptor on the cell surface was then determined byinterpolation following the manufacturer’s instruction.

Western blot analysis of RON expressionCellular proteins (50 mg per sample) were separated in an

8% SDS-PAGE under reduced conditions. Western blottingof RON expression was performed as previously described

Translational RelevanceAberrant RON expression is a pathogenic factor con-

tributing to epithelial tumorigenesis. However, thera-peutic antibodies or tyrosine kinase inhibitors targetingRON for cancer therapy have shown very limited effica-cy. Thus, there is a need to develop RON-targeted ther-apeutics with improved efficacy. Here, we describe anovel therapeutics in the form of anti-RON antibodyZt/g4–drug maytansinoid conjugates (Zt/g4–DM1) fortargeted cancer therapy. Zt/g4–DM1 retains its intrinsicactivity that induces RON endocytosis, resulting in cell-cycle arrest, reduced cell viability, andmassive cell death.In mouse xenograft tumor models, Zt/g4–DM1 displaysa strong efficacy and a long-lasting effect on colorectalcancer (CRC) cell-derived tumors with a favorable safetyprofile. Thus, targeted CRC therapy can be significantlyimproved by anti-RON antibody–drug conjugates(ADC), which should have broad implications for treat-ment of various types of cancers. In this sense, Zt/g4–DM1 represents a promising and novel ADC and war-rants for further clinical development.

Feng et al.

Clin Cancer Res; 20(23) December 1, 2014 Clinical Cancer Research6046




(2). Membranes also were reprobed with anti-actin anti-body to ensure equal sample loading.

Detection of internalized RONCells at 1 � 105 cells per well in a 6-well plate were

treated with 5 mg/mL Zt/g4 or Zt/g4–DM1 for various

times followed by goat anti-mouse IgG coupled with FITCor rhodamine. Nuclear DNAs were stained with 40,6-diamidino-2-phenylindole (DAPI). Immunofluorescencewas observed under an Olympus BK71 microscopeequipped with DUS/fluorescent apparatus as previouslydescribed (30).

DM1:3 to 4 per IgG

Linker-thioether-

Anti-RON Zt/g4-DM1(mIgG1)

-LysNH2 (random)

A

0 2 4 6 8 10 12 14 16 18 20 22

100

80

60

40

20

0

Mea

sure

men

t at A

280

nm

Elution time (min)

P1:

4.2

5%P

2:16

.69%

P3:

11.

23%

P4:

39.

05%

P5:

25.

39%

P6:

2.4

3%

P7/

P8:

0.4

5%

P0:

0.5

1%

BDAR: 3.724P2–P5: 92.36%

100

100

100

100

0

0

0

50

50

50

50

0

0 2 4 6 8 10 12 14 16 18 20 22

Mea

sure

men

t at A

280

nm

Elution time (min)

P0P1

P2

P3P4 P5

P6 P7/P8

Day 0, DAR: 3.724

Day 7, DAR: 3.670

Day 14, DAR: 3.612

Day 21, DAR: 3.607

C

0

50

100

Day 30, DAR: 3.484

Figure 1. Generation andcharacterization of anti-RON ADCZt/g4–DM1. A, schematicrepresentation of Zt/g4–DM1structure: Zt/g4 is a mouse mAbspecific to the RON sema domain(18). DM1 was conjugated to Zt/g4by nonreducible thioether linkage(SMCC) through lysine residues inthe antibody molecule. B, HICanalysis of the number of DM1conjugated to Zt/g4: Individual Zt/g4–DM1swith different numbers ofDM1 (0–8) are marked as P0 to P8.C, stability of Zt/g4–DM1. Zt/g4–DM1 was kept at 37�C for 30 days.Samples analyzed at differenttime-points with the average DARswere shown.

Anti-RON ADC for Targeted Colorectal Cancer Therapy

www.aacrjournals.org Clin Cancer Res; 20(23) December 1, 2014 6047




Cell viability and death assaysCell viability 72 hours after Zt/g4–DM1 treatment was

determined by theMTT assay (22). Viable or dead cells weredetermined by the Trypan blue exclusion assay. A total of900 cells were counted from three individual wells to reachthe percentages of dead cells.

Analysis of cell cycleHT29, HCT116, and SW620 cells (1� 106 cells per dish)

were incubated at 37�C with 5 mg/mL Zt/g4–DM1 for 24hours, labeled with propidium iodide, and then analyzed

by an Accuri Flow Cytometer. Cell-cycle changes weredetermined by measuring DNA contents as previouslydescribed (30).

Mouse xenograft CRC model and anti-RON ADCtreatment

All experiments on mice were approved by the institu-tional animal care committee. Female athymic nudemice at6 weeks of age (Taconic) were injected with 5 � 106

HT29-Luc2, HCT116-luc2, or SW620 cells in the subcuta-neous space of the right flank as previously described

Fluorescence intensity

200

0

400

600

103 104 105

HT29

300

0

600

HCT116

300

0

600

SW620

103 104 105 103 104 105

Rel

ativ

e ce

ll nu

mbe

r

IgG

CT

L

Zt/g

4-D

M1

Zt/g4

Zt/c

1&Z

t/c1-

DM

1

IgG

CT

L

Zt/c

1 &

Zt/c

1-D

M1

Zt/g

4 &

Zt/g

4-D

M1

IgG

CT

L

Zt/c

1&Z

t/c1-

DM

1Z

t/g4

& Z

t/g4-

DM

1

Fluorescence intensity Fluorescence intensity

B

HCT116:15,005 ±116

SW620:11,265 ± 2,006

DLD1:4,480 ± 352

LoVo:0 ± 0

HT29:1,8793 ± 278

Rel

ativ

e ce

ll nu

mbe

r

Relative immunofluorescence intensity

A

0

20

40

60

80

100 SW620IC50:

: 12.30 h: 18.56 h

HT29 IC50:

: 11.02 h: 19.41 h

0

20

40

60

80

100

0 6 12 24 48 0 6 12 24 48 0 6 12 24 48

Incubation time of Anti-RON ADCs (h)

RO

N r

ecep

tor

on c

ell s

urfa

ce (

%)

Zt/g4-DM1Zt/c1-DM1

C

0

20

40

60

80

100HCT116IC50:

: 12.26 h: 19.11

Figure 2. Binding and induction ofRON endocytosis by Zt/g4–DM1 inCRC cells. A, levels of RONexpression by different CRC celllines: Five CRC cells lines(1 � 106 cells/ml) in PBS wereincubated at 4�C with 5 mg/mL Zt/g4 for 60 minutes. Isotope-matched mouse IgG was used asthe control. Cell surface RON wasquantitatively determined by theimmunofluorescence assay usingQIFKIT reagents from DAKO asdetailed in Materials and Methods.B, binding of Zt/g4–DM1 to humanCRC cell lines: HCT116, HT29, andSW620 cells (1 � 105 cells) wereincubated with 5 mg Zt/g4–DM1 orZt/c1–DM1. Free Zt/g4 and Zt/c1were used as the control.Fluorescence intensity fromindividual sampleswas determinedby flow-cytometric analysis. C,kinetic reduction of cell surfaceRON: HCT116, HT29, and SW620cells (1 � 106 cells per dish) weretreated at 37�C with 5 mg/mL of Zt/g4–DM1, collected at different timepoints,washedwith acidic buffer toeliminate cell surface-bound IgG (),and then incubated with 1 mg/mLof anti-RON mAb 2F2.Immunofluorescence wasanalyzed by flow cytometer usingFITC-coupled anti-mouse IgG.Immunofluorescence from cellstreated with Zt/g4–DM1 or Zt/c1–DM1 at 4�C was set as 100%.Internalization efficiency wascalculated as the time required toachieve 50% cell surface RONreduction. (Continued on thefollowing page.)

Feng et al.





(18, 31). Mice were randomized into different groups (5mice per group). Treatment began when all tumors hadreached an average bioluminescence of approximately1 � 107 (for HT29- and HCT116-luc2 cells) or a meantumor volume of approximately 100 mm3 (for SW620cells). The single-dose group received a tail vein injectionof 20 mg/kg Zt/g4–DM1 in 0.1 mL PBS followed by obser-vation for 28 days. The multidose study was performed bytreating mice with Zt/g4–DM1 at 1, 3, 7, 10, and 15 mg/kgevery 4 days for a total of five injections. Bioluminescencefrom individual tumors was measured every 4 days usingCaliper IVIS image system (PerkinElmer). Tumor volumesfrom SW620-derived tumors were measured according to aformula: V ¼ pi/6 � 1.58 � (length � width)3/2 (18, 31).Animals were euthanized when tumor volumes exceeded2,000 mm3 or if tumors became necrotic or ulceratedthrough the skin.

In vivo toxicity studiesAcute toxicity with MTD was determined in Balb/C mice

(4 mice per dose) by a single tail vein injection of Zt/g4–DM1 at 20, 40, and 60 mg/kg body weight. Toxicityassociated with different therapeutic doses was evaluated

in athymic nude mice bearing HT29 tumor xenograft (5mice per dose). Mice were observed for about 30 days.Toxicity was assessed by observing mouse behavior, weightloss, and survival.

Statistical analysisGraphPad Prism 6 software was used for statistical anal-

ysis. Results are shown as mean � SD. The data betweencontrol and experimental groups were compared usingStudent t test. Statistical differences at P < 0.05 were con-sidered significant.

ResultsCharacterization of anti-RON ADC Zt/g4–DM1

Zt/g4 was selected as a lead ADC candidate due to itsability to induce RON internalization in various cancercells (Supplementary Table S1; refs. 21–23, 32). Zt/g4only recognizes human RON but not mouse RON homo-logue (32) and by itself has no tumor agonistic effect invivo (18). Structures of Zt/g4–DM1 are shown in Fig. 1A.A total of 250 mg Zt/g4 was conjugated to DM1 withconditions to achieve an average DAR of 4:1. Our

0 6 12 24 36 0 6 12 24 36 Treatment (h)

Zt/g4 Zt/g4-DM1

pro-RONRON β-chain

pro-RONΔ160RONΔ160 β-chain

β-Actin

β-Actin

β-Actin

RON β-chainpro-RON

SW

620

HT

29

HC

T-1

16

D

% R

ON

β-c

hain

Treatment of Zt/g4 or Zt/g4-DM1 (h)

EHCT116, IC50:

: 14.86 h: 19.46

HT29, IC50:: 9.86 h: 6.82

SW620, IC50:: 19.87 h: 17.94

Zt/g

4-D

M1

DAPI +RON

37°C, 6 h

4° C, 6 h

2C1-

DM

1

37° C, 6 h

4 °C, 6 h

37°C, 6 h

4°C, 6 h

Cm

G- D

M1

DAPI RON LAMP1DAPI +LAMP1

RON +LAMP1

DAPI + RON +LAMP1

F

Figure 2. (Continued. ) D, RON reduction analysis byWestern blotting: cellular proteins (50 mg per lane) from cells treated with 5 mg/mL of Zt/g4 or Zt/g4–DM1for various times were separated in an 8% SDS-PAGE under reduced conditions and transferred to the membrane. RON was detected by rabbitanti-RON antibody followed enhanced chemiluminescent reagents. The same membrane was reprobed for actin as the loading control. E, quantitativemeasurement of RON expression: The intensity of individual RON-b chains was determined by densitometric analysis. Internalization efficiency wascalculated as the time required to achieve a 50% RON reduction. F, immunofluorescent localization of cytoplasmic RON: HT29 cells (1 � 105 cells perchamber) were treated at 4�C or 37�C with 5 mg/mL of Zt/g4–DM1 or Zt/c1–DM1 for 6 hours followed by FITC-coupled anti-mouse IgG. After cell fixation,immunofluorescence was detected using the BK70 Olympus microscope equipped with a fluorescence apparatus. LAMP1 was used as a marker for proteincytoplasmic localization. DAPI was used to stain nuclear DNA.






selection of this ratio was based on published observa-tions of trastuzumab-emtansine (T-DM1) in which oneIgGmolecule coupling with four DM1molecules achievesmaximal therapeutic efficacy (26, 33). HIC analysisrevealed average DARs of Zt/g4–DM1 at 3.724 (Fig.

1B). The percentages of conjugates with different DARsfrom the integrated areas of the conjugates also weredetermined (Fig. 1B and Supplementary Table S2). Themajor peak accounting for 39.05%was peak 4 with a DARof 4:1. The prepared Zt/g4–DM1 with DARs at 5:1, 4:1,

Rel

ativ

e ce

ll co

unt

Relative fluorescence intensity

Zt/g4-DM1: 0 h 3 h 6 h 12 h 24 h

HCT116

HT29

SW620

A

0 3 6 9 12 15 0 3 6 9 12 15 0 3 6 9 12 15

HT29

24 h 48 h 72 h

0 3 6 9 12 15 0 3 6 9 12 15 0 3 6 9 12 15

0 3 6 9 12 15 0 3 6 9 12 15 0 3 6 9 12 15

CTL-DM12C1-DM11G4-DM1

HCT116

SW620

Concentrations of antibody in ADC (µg/mL)

0

20

40

60

80

100

0

20

40

60

80

100

0

20

40

60

80

100

% C

ell v

iab

ility

% C

ell v

iab

ility

% C

ell v

iab

ility

B

Figure 3. Effect of Zt/g4–DM1 onCRCcell cycle, survival, and death.A, changes in cell cycles: threeCRC cell lines (1 � 106 cells perdish) were treated at 37�C with5 mg/mL of Zt/g4–DM1 for varioustimes, collected, stained withpropidium iodide, and thenanalyzed by flow cytometer aspreviously described (32). B,reduction of cell viability: threeCRC cell lines (5,000 cells per wellin a 96-well plate in triplicate) weretreated with different amountsof Zt/g4–DM1 for 24, 48, and72 hours. Cell viability wasdetermined by the MTS assay.(Continued on the following page.)

Feng et al.





3:1, and 2:1 accounted for more than 92% of the totalconjugates. DARs for Zt/c1–DM1 and CmIg-DM1 were3.91 and 4.01, respectively.The stability of Zt/g4–DM1 was determined by incubat-

ing the conjugates in vitro at 37�C for 30 days. DAR changeswere measured by HIC from different time-points. Zt/g4–DM1 seems to be stable at 37�C for up to 30 days (Fig. 1Cand Supplementary Table S2). At day 30, it has an averageDARof 3.484, which represents only a 6.4% reduction fromtheDARof3.724 at day0. Themajor changes appeared tobepeak 4 and peak 5, which were reduced from 39.05% to32.72% for peak 4 and 25.39% to 20.06% for peak 5,respectively. Thus, the prepared Zt/g4–DM1 has a suitableDAR and is relatively stable at 37�C.

Effect of Zt/g4–DM1 on induction of RON endocytosisby CRC cellsWe selected CRC cell lines LoVo, DLD1, HT29, HCT116,

and SW620 expressing variable levels of RON as the cellularmodel (5, 13). RON signaling is implicated in growth,survival, and invasion in HT29, HCT116, and SW620cells (). We first determined the number of RON receptorsexpressed on CRC cell surfaces by the QIFKIT fluorescence-based quantitative method (Fig. 2A). The calculatedRON molecules on the surface of a single CRC cell was

18,793� 278 for HT29, 15,005� 115.62 for HCT116, and11,265 � 2,006 for SW620 cells, respectively. DLD1 hasabout 4,480 � 347 specific-binding sites per cell. Specificbindingwas not observed in LoVo cells.We then studied thebinding capacity of Zt/g4–DM1 to RON by flow-cytometricanalysis. We found no difference in binding intensitybetween free Zt/g4 and Zt/g4–DM1 in all three CRC celllines tested (Fig. 2B). This suggests that the conjugationdoesnot impair the Zt/g4 binding capability.

We next studied Zt/g4–DM1-induced RON endocyto-sis, a process essential for delivering DM1 into CRC cells.Zt/g4–DM1 causes a progressive reduction of cell surfaceRON in a time-dependent manner in all three CRC celllines tested (Fig. 2C). Less than 20% of RON remained onthe cell surface after a 48-hour treatment. The timerequired for Zt/g4–DM1 to induce 50% RON reduction(internalization efficacy) was at 12.26, 11.02, and 12.30hour for HCT116, HT29, and SW620 cells, respectively. Incontrast, the time required for Zt/c1–DM1-induced 50%RON reduction in HCT116, HT29, and SW620 was at19.11, 19.41, and 18.65 hour, respectively. Thus, Zt/g4–DM1 is more efficient and potent in induction of RONendocytosis.

We performed Western blotting to verify the effect of Zt/g4–DM1 on RON expression (Fig. 2D). Both pro-RON andmature RON (indicated by RON-b chain) were progressive-ly reduced in all three CRC cell lines tested. Zt/g4–DM1waseffective in reducingmature RON expression, which resideson the cell surface. Less than 20% of the RON-b chain wasdetected 36 hours after Zt/g4–DM1 treatment. The kineticreduction of mature RON was quite different among threecell lines (Fig. 2E). However, the patterns of Zt/g4–DM1-induced RON reduction were comparable with those offree Zt/g4-induced RON reduction, suggesting that theconjugation does not impair the ability of Zt/g4–DM1 toinduce RON endocytosis.

We further confirmed Zt/g4–DM1-induced RON endo-cytosis by immunofluorescence analysis of cytoplasmicRON using HT29 cells as the model (Fig. 2F). Cells stainedfor lysosomal-associated membrane protein 1 (LAMP1)were used as a marker for colocalization of internalizedRON. At 4�C, RON is detected on the cell surface. Theintracellular localization of internalized RON occurred at37�C after Zt/g4–DM1 treatment. Also, the cytoplasmicRONwas colocalized with LAMP1 in HT29 cells, indicatingthat internalized RON resides within lysosomes. In con-trast, RON endocytosis was minimal in cells treated withCmIgG-DM1. Colocalization of RON with LAMP1 was notobserved in these cells. Thus, results fromFig. 2demonstratethat Zt/g4–DM1 is effective in induction of RON endocy-tosis by CRC cells.

Effect of Zt/g4–DM1 on CRC cell cycle, growth, anddeath

DM1 acts on microtubules to cause cell-cycle arrest atG2–M phase followed by cell death (27, 34, 35). Zt/g4intracellular delivery of DM1 results in cell-cycle changes.The changes in cell-cycle profile were observed as early as 3

HC

T11

6H

T29

SW

620

0 0.95 1.85 3.75 7.50 15.0

Zt/g4-DM1 (µg/mL):

100

80

60

40

20

00 0.25 0.45 0.95 1.85 3.75 7.50 15.0

Cel

l dea

th (

%)

Concentrations of Zt/g4-DM1 (µg/mL)

HT29: 5.92 µg/mLHCT116: 6.16 µg/mLSW620: 7.15 µg/mL

IC50:D

C

Figure 3. (Continued. ) C, increased cell death: cells were treated withdifferent amount of Zt/g4–DM1 for 72 hours. Morphologic changes wereobserved under the Olympus BK-41 inverted microscope andphotographed. Images showing cell death are presented. D, cell deathpercentages were determined by the Trypan blue exclusion method. TheIC50 values for cell viability or death at 72 hours from individual groupswere calculated using the GraphPad Prism 6 software. Results shownhere are from one of three experiments with similar results.






hours after addition of Zt/g4–DM1, featuring a significantreduction in G0–G1 phase, a decrease in S phase, and adramatic increase in G2–M phase (Fig. 3A). These changeswere present in all three CRC cell lines tested. Quantitativemeasurement of cell-cycle changes at 24 hours is shown inSupplementary Table S3. CmIgG-DM1 treatment had min-imal effect on cell cycles compared with those from theZt/g4–DM1-treated cells. Thus, Zt/g4-targeted delivery ofDM1 affects cell cycles in CRC cells.

We next studied the effect of Zt/g4–DM1 on cell viability.Sensitivity of CRC cells to free DM1 was shown in Supple-mentary Fig. S1 with IC50 values at 4.1 nmol/L for HCT116,4.4 nmol/L for HT29, and 3.2 nmol/L for SW620 cells,which suggests high sensitivity to DM1. We then treatedcells with Zt/g4–DM1. A significant reduction in cell via-bility was observed in a time and dose-dependent manner(Fig. 3B). The IC50 value of Zt/g4–DM1 at 72 hours was1.64mg/mL forHT29, 2.16mg/mL forHCT116, and4.03mg/mL for SW620 cells, respectively. The effect of Zt/c1–DM1was relativelyweakwith IC50 values at 6.26mg/mL forHT29,4.64 mg/mL for HCT116, and 4.36 mg/mL for SW620 cells,respectively. Both Zt/g4–DM1 and Zt/c1–DM1 had noeffect on RON-negative LoVo cells. DLD1 cells showed aslight reduction in cell viability with IC50 value at 20.36 mg/mL (Supplementary Fig. S2). This suggests that anti-RONADC is ineffective in CRC cells expressing low levels ofRON (below 5,000 sites per cell). A comparison of theZt/g4–DM1 efficacy among four CRC cell lines with thedifferent number of RON receptor per cells is shown inSupplementary Fig. S3. Thus, Zt/g4–DM1 is more efficientthan Zt/c1–DM1 in reducing viability of CRC cells expres-sing high levels of RON.

Morphologic observation indicated a massive celldeath 72 hours after cells were exposed to Zt/g4–DM1(Fig. 3C). More than 50% cell death was observed 72hours after cells were treated with 7.5 mg/mL Zt/g4–DM1(Fig. 3D). The IC50 value ranged at 5 to 7 mg/mL in allthree CRC cell lines tested. We also counted viable cells 72hours after incubation of 1� 104 CRC cells per well in thepresence of Zt/g4–DM1. Zt/g4–DM1 treatment results ina significant reduction in the number of viable cells(Supplementary Fig. S4). Thus, Zt/g4–DM1 not onlycauses cell-cycle arrest and reduces cell viability, but alsoreduces viable cell numbers and induces massive CRC celldeath.

Therapeutic activity of Zt/g4–DM1 in mouse xenografttumor model

We first determined the efficacy of a single dose of Zt/g4–DM1 at 20mg/kg body weight on tumors derived fromHCT116, HT29, and SW620 cells. Tumor growth byHCT116-luc2 and HT29-luc2 cells was measured by bio-luminescence emitted from tumor cells. SW620-mediatedtumors were evaluated by tumor volume (18, 34). A singledose of Zt/g4–DM1 at 20 mg/kg is sufficient to delaytumor growth caused by all three CRC cell lines (Fig. 4Aand B). This time-dependent inhibition was statisticallysignificant. Images of tumors obtained at day 16 are shown

in Fig. 4C. More than 95% inhibition, measured by aver-age bioluminescence intensity, was achieved in both HT29andHCT116 tumormodels. We observed similar results inmice bearing SW620 tumors. In this case, an average 82%inhibition in tumor volume was documented (Fig. 4C).Tumor regrowth was observed at day 20 and thereafter. Anaccelerated phase was observed from day 24 to 28 (Fig. 4Aand B). It is known that mouse IgG1 has a half-life ofapproximately 6 days in vivo (36). Thus, our results indi-cate that maintenance of Zt/g4–DM1 at about 5 mg/mL invivo is required to delay tumor growth (SupplementaryFig. S5). Nevertheless, by measuring the average tumorweight at day 28, we still found a significant delay in tumorgrowth in the single-dose experiment. The inhibition ratewas 50.98% for HT29, 58.0% for HCT116, and 61.9% forSW620 tumors, respectively (Fig. 4D). Thus, a single doseof 20 mg/kg Zt/g4–DM1 is effective and displays long-lasting activity in inhibition of tumor growth initiated byall three CRC cell lines.

We then selected the HT29-Luc2 xenograft tumor modelfor the dose-ranging study.Micewere injectedwith differentdoses of Zt/g4–DM1 once every 4 days for a total of fiveinjections. Zt/g4–DM1 at 1 or 3 mg/kg showed no inhibi-tion of tumor growth (Fig. 5A). Significant Inhibition wasobserved inmice treated with 7mg/kg Zt/g4–DM1 after thethird injection. In this case, more than 80% inhibition,calculated by the average photon emission, was obtainedfrom day 19 to day 43. The efficacy was more prominent inmice treated with 10 and 15 mg/kg Zt/g4–DM1. In bothcases, tumor growth was dramatically delayed after thesecond injection. Repeated injections at both doses kepttumor growth at minimal levels during the entire period oftherapy. By analyzing the average photons at day 31, theIC50 dose for this multidose study was 5.01 mg/kg bodyweight (Fig. 5B). Images of tumors from different groups atday 31 are shown in Fig. 5C. In mice treated withZt/g4–DM1 at 7, 10, and 15 mg/kg, inhibition was in adose-dependent manner. More than 95% inhibition inmice treated with 10 and 15 mg/kg Zt/g4–DM1 wasachieved compared with that of control mice (Fig. 5C). Wealso compared the average tumor weight from the controlmice and themice treatedwith 15mg/kg Zt/g4–DM1 at day31 to determine the rate of inhibition. A 90% inhibition ataverage tumor weight was observed (Fig. 5D). Tumors werecollected at day 33 (for 1 and 3 mg/kg groups) and day 43(for 7 and 10 mg/kg groups) and compared with tumorsfromcontrol group. Significant inhibitionwas still observedfor mice treated with 7 and 10mg/kg Zt/g4–DM1. Thus, Zt/g4–DM1 at the regimens of 7, 10, 15 mg/kg Q 4 days � 5with a total dose of 35, 50, and 75mg, respectively, is highlyeffective in delaying HT29 cell-mediated tumor growth inmouse xenograft models.

To determine whether cell death occurs in xenografttumors, HT29 cell-derived tumor samples collected at day31 from both control and 15 mg/kg-treated mice wereprocessed for histologic analyses. Analysis by hematoxylinand eosin (H&E) staining revealed cell death in differentregions in all Zt/g4–DM1-treated tumors but not in

Feng et al.





control samples (Fig. 5E). An average percentage of deadareas in a tumor mass were 65% � 7.4. Western blotanalysis using cell lysates from tumor samples also

showed that RON expression in Zt/g4–DM1-treatedtumors (16.44% � 5.75) was dramatically reduced com-pared with that in control samples (100% � 15.56; Fig.

0

500

450

350

400

300

250

200

150

100

50

Tum

or g

row

th b

y ph

oton

/s (

10–8

)

HT29 controlHT29 Zt/g4-DM1HCT116 controlHCT116 Zt/g4-DM1

0 4 8 12 16 20 24 28

Days after tumor cell inoculation

Zt/g

4-D

M1

inje

ctio

n

A B

C

D

Tum

or g

row

th b

y vo

lum

e (m

m3 )

SW620 controlSW620 Zt/g4-DM1

0 4 8 12 16 20 24 28


Zt/g

4-D

M1

inje

ctio

n

HT

29H

CT

116

SW

620

Effect of Zt/g4-DM1 on tumor growth at day 16

ControlTW (g): 0.51 ± 0.22 Inhibition (%): 0.00Zt/g4-DM1 TW (g): 0.25 ± 0.09 Inhibition (%): 50.98

Zt/g4-DM1 TW (g): 0.28 ± 0.11Inhibition (%): 58.8

ControlTW (g): 0.68 ± 0.13 Inhibition (%): 0.00

ControlTW (g): 0.63 ± 0.28 Inhibition (%): 0.00Zt/g4-DM1 TW (g): 0.24 ± 0.07 Inhibition (%): 61.9

HT

29H

CT

116

SW

620

Effect of Zt/g4-DM1 on tumor weight at day 28

HT29 ControlPt/s: 1.79 x1010 ± 0.22 Inhibition (%): 0.00HT29 Zt/g4-DM1 Pt/s: 5.92 x 108 ± 3.91Inhibition (%): 96.67

ControlPt/s: 1.56 x1010 ± 1.66 Inhibition (%): 0.00Zt/g4-DM1 Pt/s: 6.23 x 108 ± 1.66 Inhibition (%): 96.0

ControlTV (mm3): 623.84 ± 112.02 Inhibition (%): 0.00Zt/g4-DM1 TV (mm3): 109.78 x 61.39 Inhibition (%): 82.40

0

400

800

1,200

1,600

2,000

2,400

Figure 4. Therapeutic effect of asingle dose Zt/g4–DM1 on CRCcell-derived tumors: athymic nudemice (five mice per group) weresubcutaneously inoculated with 5� 106 HCT116, HT29, and SW620cells followed by injection of 20mg/kg Zt/g4–DM1 through tailvein. A, tumor growth from HT29-luc2 or HCT116-luc2 cells wasdetermined by measuring averagephoton intensity. B, SW620-derived tumor growth wasmonitored by measuring tumorvolume. C, tumor images withphoton emission or calipermeasurement at day 16 arepresented. The scale from minimalto maximal is set at 300 to 35,000photons per second. Thepercentages of inhibition werecalculated from the averagephoton emission (for HT29 andHCT116 cells) or tumor volume (forSW620 cells). D, individual tumorsfrom different groups wereweighed at day 28. Thepercentages of inhibition werecalculated by a formula: (averagetumor weight from Zt/g4–DM1treated group/average tumorweight from control mice)� 100%.






0 mg/kg1 mg/kg3 mg/kg7 mg/kg10 mg/kg15 mg/kg


150

Zt/g

4-D

M1

inje

ctio

n

Tum

or

gro

wth

by

ph

oto

n/s

(10

9 )

A

C

D

200

250

100

50

0

0 5 10 15 20 25 30 35 40 45

0

0.5

1.0

1.5

2.0

2.5

3.0

Mea

n t

um

or

wei

gh

t (g

)

0 3 6 9 12 15

Doses of Zt/g4-DM1 (mg/kg)

IC50: 5.01 mg/kg B

2.90 3.21 2.58 1.92 1.55

0.61 0.39 0.06 0.05 0.02

ControlTW (g): 2.43 ±,0.69 Inhibition (%): 0.00

Zt/g4-DM1, 15 mg/kg:TW (g): 0.23 ± 0.26, Inhibition (%): 90.71

2.82 2.20 1.77 1.13 1.06

2.52 2.48 1.03 0.89 0.83

Zt/g4-DM1,1 mg/kgTW (g): 1.80 ± 0.74Inhibition (%): 26.15

Zt/g4-DM1, 3 mg/kgTW (g): 1.55 ± 0.87,Inhibition (%): 36.27

1.68 0.95 0.94 0.55 0.29

1.58 0.55 0.54 0.16 0.07

Zt/g4-DM1, 7 mg/kg:TW (g): 0.88 ± 0.53Inhibition (%): 63.73

Zt/g4-DM1, 10 mg/kg: TW (g): 0.58 ± 0.60Inhibition (%): 76.56

31

31

35

35

43

43

Tumor weight by grams

7

15

3

10

HT29 tumor images by photon intensity

Photons/s:

30,000

20,000

10,000

Counts

Color scaleMin = 600Max = 35,000

1.31 × 1011 ± 0.52 Inhibition (%): 0.00

0

Zt/g4-DM1mg/kg

1Photons/s: 1.45 × 1011 ± 0.62 Inhibition (%): -11.15

Photons/s: 7.81 × 1010 ± 3.71 Inhibition (%): 40.30




Days

Figure 5. Evaluation of differentdoses of Zt/g4–DM1 on tumorgrowth and RON expression. A,effect of multi-dose of Zt/g4–DM1on tumor growth was tested inHT29 cell-induced tumors. Tumor-bearing mice were treated withdifferent doses of Zt/g4–DM1every 4 days for a total of fiveinjections ( ). Tumor growth wasdetermined by the averagebioluminescence intensity. B, anIC50 value based on the averagebioluminescence intensity fromindividual groups at day 31 wascalculated using GraphPad Prism6 software. C, bioluminescenceimages of individual tumors fromeach group at day 31 are shown.The percentages of inhibition werecalculated from the averagephoton emission. The color scalefrom minimal to maximal is set at300 to35,000photonsper second.D, individual tumors from differentgroups were collected andweighed at day 31, 35, and 43,respectively. The percentages ofinhibition were calculated asdetailed in Fig. 4C. (Continued onthe following page.)


Feng et al.




5F). Thus, Zt/g4–DM1 causes cells death in CRC xenografttumors, which is associated with elimination of CRC cellsoverexpressing RON.

Toxic effect of Zt/g4–DM1 on miceThree experiments using two different types of mice were

performed to study Zt/g4–DM1 on animal behavior andbody weight. The first experiment addressed the impact of

multidoses of Zt/g4–DM1. Athymic nude mice wereinjected five times with 1, 3, 5, 7, 10, 15 mg/kg of Zt/g4–DM1 and monitored every 4 days for a total period of 31days. All mice behaved normally during the entire obser-vational period. The average body weight of experimentalgroups was comparable with that of control mice with nodifferences (Fig. 6A). The second experiment observed theeffect of a single dose of Zt/g4–DM1 at 20 mg/kg in nude

Control Enlarged Zt/g4-DM1,15 mg/kg Enlarged

tumor massdead region

dead regiontumor mass

E

F Control Zt/g4-DM1,15mg/kg

pro-RONRON β-chain

β-actin

T1 T2 T3 T4 T5 T6 T7 T8 T9 T10

Figure 5. (Continued. ) E, samples ofHT29 cell-derived xenograft tumorsfrom both control and 15 mg/kgZt/g4–DM1-treated mice at day 31were processed for histologicexamination. Analysis by H&Estaining reveals cell death indifferent regions in Zt/g4–DM1-treated tumors but not in controlsamples. F, Western blot analysis ofRON expression in tumor samplesfrom both control and 15 mg/kgZt/g4–DM1-treated mice.Densitometry analysis wasperformed to determine the levels ofRON expression.

90

95

100

105

110

115

120

0 4 8 12 16 20 2490

95

100

105

110

115

120

125

130

Ch

ang

es o

f b

od

y w

eig

ht

(%

)

90

95

100

105

110

115

120

125

130

0 5 10 15 20 25 30 35 40 45 0 4 8 12 16 20 24 28

Days after cell inoculation

A B C Nude mice Nude mice Balb/c mice

HT29 controlHT29 Zt/g4-DM1HCT116 controlHCT116 Zt/g4-DM1SW620 controlSW620 Zt/g4-DM1

Days after cell inoculation Days after Zt/g4-DM1 injection

Zt/g4-DM1: 20 mg/kgHT29 + Zt/g4-DM1

0 mg/kg1 mg/kg3 mg/kg7 mg/kg10 mg/kg15 mg/kg

20 mg/kg40 mg/kg60 mg/kg

HT29 + Zt/g4-DM1

125

130

Figure 6. Toxicity of Zt/g4–DM1 in vivo. Body weight was measured every 4 days during the period of Zt/g4–DM1 treatment. A, effect of multiple dosesof Zt/g4–DM1 on mouse body weight was determined by administration of Zt/g4–DM1 at 1, 3, 7, 10, 15 mg/kg every 4 days with a total of 5 injections. Micewere weighed and monitored for a total of 31 days. B, effect of a single dose of Zt/g4 at 20 mg/kg on mouse body weight was determined using micebearing HT29, HCT116, or SW620-derived tumors. Body weight was monitored up to 28 days. C, effect of high doses of Zt/g4–DM1 on mouse body weightwas analyzed by tail vein injection at 20, 40, and 60 mg/kg to Balb/c mice. Mice were euthanized at day 21. In all cases, the average body weight ofmice before Zt/g4–DM1 injection was 19.8 � 3.6 grams (5 mice per group) and set as 100%.






mice bearing tumors derived from HT29, HCT116, andSW620 cells. No changes in behavior or body weight wereobserved (Fig. 6B). The third involved a single-dose injec-tion of Zt/g4–DM1 at 20, 40, and 60 mg/kg in Balb/c micemonitored for 24 days (Fig. 6C). Moderate distress wasobserved in mice administered with 60 mg/kg Zt/g4–DM1.Also, a moderate reduction of about 6% body weight wasobserved within the first 4 days after 60 mg/kg Zt/g4–DM1injection. Although the average body weight from thisgroup of mice slowly recovered during the observationperiod, the overall average remained lower than that ofcontrol mice with a 19% difference compared with that ofcontrol mice at day 24. Thus, Zt/g4–DM1 at the multiple-dose regimen seemed to be well tolerated. However, asingle-dose of Zt/g4–DM1 at 60mg/kg showed a toxic effecton mouse behavior and body weight.

DiscussionThis report is about the development of anti-RON ADC

Zt/g4–DM1 for targeted cancer therapy.We showed that Zt/g4–DM1 retains its specificity to RON after conjugationwithDM1. The conjugateswere stable at 37�Cwithminimaldissociation ofDM1 fromantibody. Binding of Zt/g4–DM1to CRC cells causes a rapid endocytosis of cell surface RON.Internalized Zt/g4–DM1 results in cell-cycle arrest in G2–Mphase, followed by cell viability reduction, and massivecell death. Studies from mouse xenograft tumor modelsconfirmed that a single dose of Zt/g4–DM1 at 20 mg/kg issufficient to inhibit tumor growth with a long-lasting effectup to 20 days. The multiple dose-ranging studies furtherdemonstrated that the therapeutic regimen at 7, 10, 15mg/kg, every 4 days � 5 with a total dose of 35, 50, and75mg, respectively, displays strong efficacy in tumor growthinhibition. Furthermore, we showed that Zt/g4–DM1 atdoses up to 40mg/kg has no toxic effect onmouse behavioror bodyweight. Thus, Zt/g4–DM1 is anovel biotherapeuticswith enhanced efficacy for RON-targeted cancer therapy.Humanization of Zt/g4 is currently under way.

Zt/g4 was conjugated to DM1 at appropriate DARsthrough the thioether linkage (25, 26, 33). Consistentwith previous reports (26, 33, 27), Zt/g4–DM1 has afavorable conjugation profile. Most conjugates haveDARs ranging from 2:1 to 5:1 with the major peak at4:1. Such a profile is the typical pattern of ADCs using thethioether linkage technology (29). Zt/g4–DM1 is relative-ly stable. Incubation of Zt/g4–DM1 at 37�C for 30 daysresulted in only 6.5% reduction in DARs of DM1. Thesedata are consistent with previous reports showing thatantibodies conjugated with DM1 through thioether link-age are highly stable both in vitro and in vivo under variousconditions (29, 31). Although we did not test the stabilityof Zt/g4–DM1 under in vivo conditions, it is expected thatthe conjugates have a similar stability profile due to thesimilar conjugation method (29, 31). The efficacy of invivo studies using a single dose of Zt/g4–DM1 at 20 mg/kgseems to support this notion. In this case, a single injec-tion is sufficient to inhibit tumor growth for almost

3 weeks, implying that Zt/g4–DM1 is relatively stablein vivo to exert a long-lasting effect. Clearly, the use ofthioether linkage provides the practical basis for futuredevelopment of humanized Zt/g4–DM1.

The selection of Zt/g4 as the leading candidate for DM1conjugation is based on its unique features. Zt/g4 is amAb highly specific and sensitive to RON, and recognizesan epitope in the RON sema domain (32). The bindingof Zt/g4 to RON results in a rapid and efficient RONinternalization process. The internalized RON coloca-lizes with LAMP1, suggesting that the endocytosis couldbe mediated through a clathrin-dependent pathway (37).Significantly, more than of 80% of cell surface RONis internalized within 48 hours after addition of Zt/g4–DM1. In the case of HT29 cells expressing approximately18,800 RON molecules per cell, it translates into 15,000RON receptors that are internalized within 48 hours. Thisis equivalent to 60,000 DM1 molecules within a singlecell, sufficient to cause cell-cycle arrest. It is noticed thatthe kinetics of RON internalization among three CRC celllines are quite different after addition of Zt/g4–DM1,suggesting the importance of the rate of endocytosis inregulating efficacy of Zt/g4–DM1. Clearly, Zt/g4-inducedRON endocytosis facilitates intracellular delivery of DM1to exert cytotoxic activity. Moreover, Zt/g4 has no ago-nistic activities in CRC cells expressing RON (18).

The action of DM1 delivered through Zt/g4 was clearlydisplayed in CRC cells. First, we showed by flow-cyto-metric analysis that the delivery of DM1 results in cell-cycle arrest in G2–M phase, which is a feature of DM1 thatimpairs microtubule dynamics (35). This effect wasobserved as early as 3 hours after addition of Zt/g4–DM1,which is characterized by progressive reduction of the G1

phase and the accumulation of cells at the G2–M phase.Second, we observed that targeted delivery of DM1 pro-gressively decreases cell viability. More than 80% reduc-tion in cell viability 72 h after treatment was achievedamong the three CRC cell lines tested. Finally, we docu-mented a massive cell death in Zt/g4–DM1-treated CRCcells in a dose-dependent manner with IC50 values in therange of 5 to 7 mg/mL Zt/g4–DM1. These evidences suggestthat DM1 is effectively delivered by Zt/g4 through atargeted pathway, which results in cell-cycle arrest, viabil-ity reduction, and cell death.

Results from mouse xenograft CRC models prove thatZt/g4–DM1 is highly efficient in inhibition of tumorgrowth. This conclusion is supported by mouse modelsusing two treatment regimens. The single-dose therapyusing 20 mg/kg Zt/g4–DM1 was designed to determinewhether this dose is sufficient to inhibit tumor growthand, if so how long the effect will last. Indeed, Zt/g4–DM1 at 20 mg/kg was highly effective in delaying xeno-graft tumor growth with a long-lasting effect of almost 2weeks. It is known that mouse IgG1 has a half-life ofapproximately 6 days in vivo (36). Administration of 20mg/kg Zt/g4–DM1 should allow us to monitor its efficacyin a four half-life cycle within 24 days. The obtainedresults confirmed that the efficacy of Zt/g4–DM1 lasts up

Feng et al.





to 12 days without signs of tumor regrowth (from day 4 today 16 as shown in Fig. 4A). By calculation, the amount ofZt/g4–DM1 in vivo required to inhibit tumor growth isabout 5 mg/kg (Supplementary Fig. S5). In other words, adose of 5 mg/kg Zt/g4–DM1 maintains a balance betweentumor growth and inhibition.The multiple dose-ranging studies were designed to

determine the minimum dose required to inhibit xeno-graft tumor growth. Zt/g4–DM1 at 7 mg/kg in the regimenof Q 4 days � 5 with a total dose of 35 mg/kg achieves asignificant inhibition. An increase of Zt/g4–DM1 up to 10and 15 mg/kg in a similar regimen results in a superiortherapeutic index. In both cases, the total amount ofZt/g4–DM1 was at 50 and 75 mg/kg, respectively. Theseresults provide us the IC50 value of 5.01 mg/kg (calculatedaccording to the repeated Zt/g4–DM1 administration andthe estimated antibody half-life), which is consistent withthe estimated values of 5 mg/kg from the single-dosestudy. Thus, results from multiple dose regimens shouldhelp us to design an optimal treatment strategy for thepotential use of humanized Zt/g4–DM1 in future clinicalsettings.Analysis of the toxic profile in two types of mice indicates

that Zt/g4–DM1 is relatively safe at therapeutic doses withminimal impact on animal’s behavior and body weight.Because Zt/g4 does not recognizemouse RON, the observedlow toxicity suggest a very limited dissociation of the Zt/g4–DM1 conjugates in vivo. However, a single dose of Zt/g4–DM1at 60mg/kghas anegative impact onmousehighlight-ed by an average of 6% to 19% reduction of body weightduring the entire period of study. This suggests that duringthe administration of multiple doses of Zt/g4–DM1, theaccumulated Zt/g4–DM1 in vivo should not exceed the 60

mg/kg limitation. This dose limitation should be a valuablereference for the use of humanized Zt/g4–DM1 in humansubjects in the future.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: H.-P. Yao, Y.-Q. Zhou, J. Zhou, R. Zhang,M.-H. WangDevelopment of methodology: H.-P. Yao, Y.-Q. Zhou, M.-H. WangAcquisitionofdata (provided animals, acquired andmanagedpatients,provided facilities, etc.): W. Wang, M.-H. WangAnalysis and interpretation of data (e.g., statistical analysis, biosta-tistics, computational analysis): L. Feng, H.-P. Yao, M.-H. WangWriting, review, and/or revision of the manuscript: Y.-Q. Zhou, J. Zhou,R. Zhang, M.-H. WangAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): M.-H. WangStudy supervision: W. Wang, M.-H. Wang

AcknowledgmentsThe authors thank Susan Denney (TTUHSC School of Pharmacy in

Amarillo) for editing the manuscript.

Grant SupportThis work was supported by NIH grant R01CA91980, funds from

Amarillo Area Foundation, SKLDTID subproject #2011ZZ01 from FirstHospital of Zhejiang University School of Medicine, and ZhejiangMajor Medical Health & Sciences Technology Foundation Projects#WKJ-ZJ-13 and #2014C33204. R. Zhang was supported by NIH grantR01CA112029.

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received April 11, 2014; revised August 26, 2014; accepted September 15,2014; published OnlineFirst October 7, 2014.

References1. Ronsin C, Muscatelli F, Mattei MG, Breathnach R. A novel putative

receptor protein tyrosine kinase of the met family. Oncogene 1993;8:1195–202.

2. WangMH, Ronsin C, GesnelMC, Coupey L, Skeel A, Leonard EJ, et al.Identification of the ron gene product as the receptor for the humanmacrophage stimulating protein. Science 1994;266:117–9.

3. Yao HP, Zhou YQ, Zhang R,WangMH.MSP-RON signaling in cancer:pathogenesis and therapeutic potential. Nat Rev Cancer 2013;13:466–81.

4. WangMH, LeeW, Luo YL,WeisMT, Yao HP. Altered expression of theRON receptor tyrosine kinase in various epithelial cancers and itscontribution to tumorigenic phenotypes in thyroid cancer cells.J Pathol 2007;213:402–11.

5. Zhou YQ, He C, Chen YQ, Wang D, Wang MH. Altered expression ofthe RON receptor tyrosine kinase in primary human colorectal ade-nocarcinomas: generation of different splicing RON variants and theironcogenic potential. Oncogene 2003;22:186–97.

6. Yao HP, Zhuang CM, Zhou YQ, Zeng JY, Zhang RW, Wang MH.Oncogenic variant RON160 expression in breast cancer and its poten-tial as a therapeutic target by small molecule tyrosine kinase inhibitor.Curr Cancer Drug Targets 2013;13:686–97.

7. Thomas RM, Toney K, Fenoglio-Preiser C, Revelo-Penafiel MP, Hin-gorani SR, Tuveson DA, et al. The RON receptor tyrosine kinasemediates oncogenic phenotypes in pancreatic cancer cells and isincreasingly expressed during pancreatic cancer progression. CancerRes 2007;67:6075–82.

8. MaggioraP,MarchioS,StellaMC,GiaiM,BelfioreA,DeBortoliM, et al.Overexpression of the RON gene in human breast carcinoma. Onco-gene 1998;16:2927–33.

9. Lee WY, Chen HH, Chow NH, Su WC, Lin PW, Guo HR. Prognosticsignificance of co-expression of RON and MET receptors in node-negative breast cancer patients. Clin Cancer Res 2005;11:2222–8.

10. Welm AL, Sneddon JB, Taylor C, Nuyten DS, van de Vijver MJ,Hasegawa BH, et al. The macrophage-stimulating protein pathwaypromotes metastasis in a mousemodel for breast cancer and predictspoor prognosis in humans. Proc Natl Acad Sci USA 2007;104:7570–5.

11. Park YL, Lee GH, Kim KY, Myung E, Kim JS, Myung DS, et al.Expression of RON in colorectal cancer and its relationships withtumor cell behavior and prognosis. Tumori 2012;98:652–62.

12. Wang J, Rajput A, Kan JL, Rose R, Liu XQ, Kuropatwinski K, et al.Knockdown of Ron kinase inhibits mutant phosphatidylinositol 3-kinase and reduces metastasis in human colon carcinoma. J BiolChem 2009;284:10912–22.

13. Xu XM, Wang D, Shen Q, Chen YQ, Wang MH. RNA-mediated genesilencing of the RON receptor tyrosine kinase alters oncogenic phe-notypes of human colorectal carcinoma cells. Oncogene 2004;23:8464–74.

14. Qian F, Engst S, Yamaguchi K, Yu P, Won KA, Mock L, et al. Inhibitionof tumor cell growth, invasion, and metastasis by EXEL-2880 (XL880,GSK1363089), a novel inhibitor of HGF and VEGF receptor tyrosinekinases. Cancer Res 2009;69:8009–16.






15. Schroeder GM, An Y, Cai ZW, Chen XT, Clark C, Cornelius LA, et al.Discovery of N-(4-(2-amino-3-chloropyridin-4-yloxy)-3-fluorophenyl)-4-ethoxy-1-(4-fluorophenyl)-2-oxo-1,2-dihydropyridine-3-carboxa-mide (BMS-777607), a selective and orally efficacious inhibitor of theMet kinase superfamily. J Med Chem 2009;52:1251–4.

16. Pan BS, Chan GK, Chenard M, Chi A, Davis LJ, Deshmukh SV,et al. MK-2461, a novel multitargeted kinase inhibitor, preferen-tially inhibits the activated c-Met receptor. Cancer Res 2010;70:1524–33.

17. O'Toole JM1, Rabenau KE, Burns K, Lu D, Mangalampalli V, BalderesP, et al. Therapeutic implications of a human neutralizing antibody tothe macrophage-stimulating protein receptor tyrosine kinase (RON), ac-MET family member. Cancer Res 2006;66:9162–70.

18. Yao HP, Zhou YQ, Ma Q, Guin S, Padhye SS, Zhang RW, et al. Themonoclonal antibody Zt/f2 targeting RON receptor tyrosine kinase aspotential therapeutics against tumor growth-mediated by colon can-cer cells. Mol Cancer 2011;10:82–93.

19. Eyob H, Ekiz HA, Derose YS, Waltz SE, Williams MA, Welm AL.Inhibition of ron kinase blocks conversion of micrometastases to overtmetastases by boosting antitumor immunity. Cancer Discov 2013;3:751–60.

20. Kauder SE, Santell L, Mai E, Wright LY, Luis E, N'Diaye EN, et al.Functional consequences of the macrophage stimulating protein689C inflammatory bowel disease risk allele. PLoS ONE 2013;8:e83958–67.

21. GuinS,YaoHP,WangMH.RON receptor tyrosine kinase asa target fordelivery of chemodrugs by antibody directed pathway for cancer cellcytotoxicity. Mol Pharm 2010;7:386–97.

22. Guin S,MaQ, PadhyeS, ZhouYQ, YaoHP,WangMH. Targeting acutehypoxic cancer cells by doxorubicin-immunoliposomes directed bymonoclonal antibodies specific to RON receptor tyrosine kinase.Cancer Chemother Pharmacol 2011;67:1073–83

23. Li Z, Yao H, Guin S, Padhye SS, Zhou YQ, Wang MH. Monoclonalantibody (mAb)-induced down-regulation of RON receptor tyrosinekinase diminishes tumorigenic activities of colon cancer cells. Int JOncol 2010;37:473–82.

24. Padhye SS, Guin S, Yao HP, Zhou YQ, Zhang R, Wang MH. Sustainedexpression of the RON receptor tyrosine kinase by pancreatic cancerstem cells as a potential targeting moiety for antibody-directed che-motherapeutics. Mol Pharm 2011;8:2310–9.

25. Sievers EL, Senter PD. Antibody-drug conjugates in cancer therapy.Annu Rev Med 2013;64:15–29.

26. Ducry L, Stump B. Antibody-drug conjugates: linking cytotoxic pay-loads to monoclonal antibodies. Bioconjug Chem 2010;21:5–13.

27. Lewis Phillips GD, Li G, Dugger DL, Crocker LM, Parsons KL, Mai E,et al. Targeting HER2-positive breast cancer with trastuzumab-DM1,an antibody-cytotoxic drug conjugate. Cancer Res 2008;68:9280–90.

28. Brun MP, Gauzy-Lazo L. Protocols for lysine conjugation. MethodsMol Biol 2013;1045:173–87.

29. Junutula JR, Flagella KM, GrahamRA, Parsons KL, Ha E, RaabH, et al.Engineered thio-trastuzumab-DM1 conjugate with an improved ther-apeutic index to target human epidermal growth factor receptor 2-positive breast cancer. Clin Cancer Res 2010;16:4769–78.

30. Sharma S, Zeng JY, Zhuang CM, Zhou YQ, Yao HP, Hu X, et al. Small-molecule inhibitor BMS-777607 induces breast cancer cell polyploidywith increased resistance to cytotoxic chemotherapy agents. MolCancer Ther 2013;12:725–36.

31. Jumbe NL, Xin Y, Leipold DD, Crocker L, Dugger D, Mai E, et al.Modeling the efficacy of trastuzumab-DM1, an antibody drug conju-gate in mice. J Pharmacokinet Pharmacodyn 2010;37:221–42.

32. Yao HP, Luo YL, Feng L, Cheng LF, Lu Y, Li W, et al. Agonisticmonoclonal antibodies potentiate tumorigenic and invasive activitiesof splicing variant of the RON receptor tyrosine kinase. Cancer BiolTher 2006;5:1179–86.

33. Teicher BA, Chari RV. Antibody conjugate therapeutics: challengesand potential. Clin Cancer Res 2011;17:6389–97.

34. Liu C, Tadayoni BM, Bourret LA, Mattocks KM, Derr SM, WiddisonWC, et al. Eradication of large colon tumor xenografts by targeteddelivery of maytansinoids. Proc Natl Acad Sci USA 1996;93:8618–23.

35. Lopus M, Oroudjev E, Wilson L, Wilhelm S, Widdison W, Chari R, et al.Maytansine and cellular metabolites of antibody-maytansinoid con-jugates strongly suppress microtubule dynamics by binding to micro-tubules. Mol Cancer Ther 2010;9:2689–99.

36. Vieira P, Rajewsky K. The half-lives of serum immunoglobulins in adultmice. Eur J Immunol 1988;18:313–6.

37. Henriksen L, Grandal MV, Knudsen SL, van Deurs B, Grøvdal LM.Internalization mechanisms of the epidermal growth factor rec-eptor after activation with different ligands. PLoS ONE 2013;8:e58148–58.


Feng et al.




2014;20:6045-6058. Published OnlineFirst October 7, 2014.Clin Cancer Res Liang Feng, Hang-Ping Yao, Wei Wang, et al. Colorectal Cancer TherapyConjugation (Anti-RON ADC) as a Novel Therapeutics for Targeted

Drug Maytansinoid−Efficacy of Anti-RON Antibody Zt/g4

Updated version

10.1158/1078-0432.CCR-14-0898doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2014/10/08/1078-0432.CCR-14-0898.DC1

Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/20/23/6045.full#ref-list-1

This article cites 37 articles, 15 of which you can access for free at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/20/23/6045To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-14-0898

http://clincancerres.aacrjournals.org/content/suppl/2014/10/08/1078-0432.CCR-14-0898.DC1

http://clincancerres.aacrjournals.org/content/20/23/6045.full#ref-list-1

http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/20/23/6045


ﬁcacy of anti-ron antibody zt/g4 drug maytansinoid...

Documents