A blocking monoclonal antibody to endothelial-leukocyte adhesion molecule-1 (ELAM1)

Vol.

171, No.

August

BIOCHEMICAL

1, 1990

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

Pages

31, 1990

348-353

A BLOCKING MONOCLONAL ANTIBODY TO ENDOTHELIAL-LEUKOCYTE ADHESION MOLECULE-l (ELAMl) Christopher Benjamin, Irene Dougas, Gloria Margaret Rosa, Barbara Newman, Laurelee Catherine Iiession, Susan Goelz, Kathy Biogen, Received

Inc.,

July

11,

14 Cambridge

Chi-Rosso, Stefan Luhowskyj, Osborn, Cornelia Vassallo, McCarthy, and Roy Lobb’

Center,

Cambridge,

MA 02142

1990

ELAMl is a leukocyte adhesion molecule induced on human umbilical vein endothelial cells (HWECs) by inflammatory cytokines. Balb/C mice were immunized with COS cells transiently expressing cell-surface ELAMl after transfection with ELAMl cDNA. After fusion, ELAMl-specific monoclonal antibodies (Mabs) were identified by selective adhesion to ELAMlexpressing, but not control, CHO cells, and to cytokine-treated but not One Mab, designated BBll, untreated HlJVECs. binds to and immunoprecipitates ELAHl expressed on HWECs, COS and CHO cells. BBll blocks the interaction of ELAMl with human PMN, the human myelomonocytic cell line HL60, and the human colon carcinoma line HT29. 0 1990 Academic Press, rnc.

ELAMl

is

an induced

adhesion growth

of adhesion factor,

at sites

of

vascular

inflammation

bed,

We have

molecule

called expressed

Mab

to

transfected the

cell

molecules

1 Author 0006-291X/90 Copyright All rights

ELAMl,

surface. where

to It

lectins, ELAMl

expressed

in

is

the

one of

epidermal is present the

normal

in PMN extravasation

expression

functional method

to

method

cDNAs are

to whom correspondence

after should available

should

$1.50

0 1990 by Academic Press, Inc. of reproduction in any form reserved.

cloning

at

clones clone

(4).

a novel

EUVECs,

of ELAMl, We have lymphocyte

which

we

of specific Mabs to characterize In this report we describe a

cDNA, and transiently

This

(1,Z). (2).

role

on cytokine-treated

generated

ELAMl

not

direct

to select

the lack problematic.

is

with

face

the

However,

proteins

is

an important

powerful

expressed

VCAMl (5).

contributes

proteins

but

described this

that

(l-3).

to HL60 cells

exploited

adhesion

plays

--in vivo

recently

adhesion

already

regulatory

in vivo,

and likely sites

protein

HWECs --in vitro related to mammalian

molecules

and complement

inflammatory

ing

cell

of PMN to cytokine-treated

a family

using

endothelial

348

immunization expressing be applicable but

Mabs are

be addressed.

with ELAMl to other lacking.

have the block-

COS cells protein on cell

sur-

Vol.

171,

No.

MATERIALS

1, 1990

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

AND METHODS

HWECs were isolated and subcultured as described (4) and Cells. used for assays at passages 3-5. HL60, COS7, CHO and HT29 cells were maintained, and human PMN isolated, as described (4,6). The generation of stable ELAMl-expressing CHO cell lines will be detailed elsewhere. Briefly, the ELAMl cDNA (4) was inserted into vector pBG341JOD, which casette contains an expression for the dihydrofolate reductase cDNA CHO . DHFR- cells were transfected with pBG341JOD.ELAM by gene. electroporation (6), methotrexate selection performed, and clones expressing sufficient ELAMl to bind HL60 cells detected directly by adhesion assay. Two cell lines, CH0338 and CH0660, were used in these studies. Immunoprecipitation. All cells were labelled with [35S]-cysteine and extracted with Triton Xl00 as described for HWECs (4). Cell sunernatants were precleared for 2-3 hr at 4’C with protein‘A:Sepharose IgG, incubated overnight at 4OC with protein A-sepharose-BBll, the beads washed 5 times with PBS pH 7.2, 0.5% Tween 20, 0.05% SDS, 0.1% BSA, and 0.02% sodium azide, boiled in 2% sample buffer, and the eluate run on a 4-20% SDS gradient gel. N-glycanase treatment was performed as described (2). Adhesion Assays. HWECs were grown in 48 well cluster plates as described (4). Control CHO cells, or CHO338 or CHO660 cells, were grown in 48 well or 96 well plates, and used within 2 days of reaching confluence. HL60 and HT29 cells were labelled, and adhesion assays performed, as described (4). For inhibition assays, the appropriate target cells were preincubated for 30 min at room temperature with Moabs at 10 ug/ml, cells added to washed monolayers, and the assays performed as above. Moabs 60.3 and 4B9, to inhibitory epitopes on CD18 and VCAMl, respectively, were the gift of Dr. John Harlan. Hybridoma Generation. Eight to ten week old female Balb/cJ mice (The Jackson Laboratory, Bar Harbor, ME) were immunized i.p. (0.5 ml) every two weeks for eight weeks with 2-5 million COS cells, and bled from the tail vein 7-10 days after each boost. COS cells were transfected with ELAMl DNA by electroporation as described (4), and 72 hr after transfection, were harvested with 5 mM EDTA in Dulbecco’s PBS (GIBCO Laboratories, Grand Island, NY) 1% BSA, by incubation for 30 min. at 37OC. The cells were collected and washed (200 x g) once with DPBS, 5 mM EDTA, and then twice with DPBS without EDTA. The cells were resuspended in a small volume of DPBS, counted in trypan blue and scored for viability. Cells used for immunization were always greater than 95% viable. Sera were analyzed for antibodies which blocked HL60 adhesion to 4 hour IL-l activated HWECs. One mouse was selected, boosted with COS ELAM-1 cells, and spleen cells were fused to P3X63Ag8.653 cells as described (7). Hybrid cells were selected in HAT supplemented medium and grown initially in the presence of a commercial cloning factor (IGEN, Rockville, MD). Culture supernatants were screened for differential binding to a stable transfectant of ELAMl in CHO cells vs untransfected CHO cells, using radiolabeled goat anti-mouse immunoglobulin (New England Nuclear, Boston, MA). Specificity was corroborated with an ELISA using IL-l activated HWECs, and selected cultures examined for inhibition of adhesion. Positive cultures were subcloned at limiting dilution. Monoclonality was confirmed by subclass analysis using goat immunoglobulin coated plates and a kit from Hyclone anti-mouse Laboratories (Logan, Utah). BBll ascites was raised in pristane primed The monoclonal antibody was purified on protein A coupled Balb/c mice. to sepharose 4B (Zymed, CA). The ascites was diluted 1:3 into 1.5M pH 8.9 and passed over a 10 ml column of protein A glycine, 3H NaCl, pre-equilibrated in the same buffer. Bound antibody was eluted with (0.05f-l) acetate saline buffer, pH 3.7 and dialyzed against PBS pH 7.2. 349

Vol.

BIOCHEMICAL

AND BIOPHYSICAL

an ELAMl-selective

Mab.

171, No. 1, 1990

RESEARCH COMMUNICATIONS

RESULTS AND DISCUSSION Generation

of

at biweekly

intervals

with

COS

Anti-ELAMl-secreting

hybridomas

binding

not

to CH0338 but

tives

were

also

detected

HWECs,

using

tently

scored

in each

assay.

Originally at

and adhesion for

was

CHO.JOD cells control and

with

COS cells

(not

referred

are

proteins

N-glycanase

1B).

BBll

with

proteins

of (Fig.

the

generates a

published

BBll.

Kd

of

cells

core

was

control

control

HWECs

ELAMl behave

but

not

similarly

immunoprecipitates HWECs

96 1B).

BBll

and 100 Kd from

not

of about

about (Fig.

(2).

BBll Specificity

not

Mab BBll

adhesion

specificity

but

ELAMl-expressing

protein COS

IgM,

CH0660

subclones

not

consis-

the

ELAMl

expressing

a protein

results

130

Kd

80 Kd (Fig. from

These

also

(Fig. ELAMl-

data

are

in

immunoprecipitates

CH0660 but

not

control

CHO

1B). of

cell

surface

and functional

adhesion

as

but

in

and

but

the

120 Kd from

control

about

number

expression cytokine

not

of

posi-

BBll,

subclones.

with

transiently

hereafter

110 Kd and

assays,

well,

IgGZb

the IgG2b

All

ELAMl. screen

inhibited

cytokine-treated

shown).

treatment

but

agreement

A

with

immunoprecipitates

expressing

with

expressing as a primary

7 subclones.

reactivity

COS cells

to

of about

1B).

cells

lA),

and

One

containing

by

immunized

to cytokine-treated

and also

yielding

resided

confirmed

(Fig.

lA),

clone

were

In parallel

assay.

assay

dilution

inhibition

ELAMl

(Fig.

a mixed

using

cells.

ELISA

binding

limiting

selected

by binding

a standard

mice

transiently

were

CHO control

readily

control

subcloned

cells

Balb/c

assay

molecules

receptors

ELAMl

for

IL6

(9),

molecules without (4),

have using VCAMl

and GMCSF (10).

now been specific (5),

cloned

by direct

Mabs,

including

and ICAMZ (8), Our

results

and the

show

that

B

A CHO -g 0.2 c 8 z :: ::c

_Lr HUVEC

0.1

b :: a C

JO0 660

-

TNF

Figure 1. A) BBll binds ELAMl: control CHO cells (JOD) or ELAMl- expressing CHO cells (660), or HUVECs with or without TNF treatment (10 ng/ml,4h), were incubated &th Mab BBll (10 ug/ml, 60',23V), washed, incubated with alkaline phosphatase-coupled anti-mouse antibody (1:5000,60',23YZ), washed and developed using standard protocols. B) BBll immunoprecipitates ELAHl: HUVBCs expressing ELAHl (TNF, 10 ng/ml,4h), either N-glycanase treated (lane 1) or untreated (lane 2): COS cells expressing EUMl (lane 3) or control COS cells (lane 4); CH6 cells expressing ELAHl (lane 5) or control CHO cells (lane 6).

350

Vol.

171, No. 1, 1990

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

Control CHOLJOD) Control IgG2b BBll (lOpg/ml) BBII (Ipg/ml) BBII (0.1pg/ml) I I cells/mm

0

I 2 2 ( x 10-31

Figure 2. Mab BBll inhibits HL60 adhesion to ELAMl: CHO cells expressing ELAMl were preincubated with Mab BBll (10 ug/m1,30’,23’C) and the of HL60 cells to adhere as compared to control CHO cells (JOD) ability was assessed (see methods).

immunization

of

CDMB expression specific

mice

with

vector

Mabs,

even

is

COS cells

adequate

though

sion

pathway

expression, to

HL60

at a concentration CH0660

adhesion.

However, binding

a second another

its while

BBll,

even

to TNF-treated

adhesion adhesion

pathway.

used

adhesion

cell

expressed

ELAMl

As shown

blocks doses,

found

of the (4).

(Fig.3),

completely

on

by

only HL60s

has adhe-

direct

in Figure

HL60

Mab has no effect

adheon HL60

partially

suggesting that

of

HL60

function

at optimal

We have

line

ELAMl

IgG2b

the

generation

clone

lOug/ml

into

(11).

to

a control HWECs

molecule

it

cloned

subsequent

transient

the characterization

of

cells,

the is

genes

The promyelocytic

We have

and to examine

2, Mab BBll sion

(1,Z).

for

expression

of HL60 Adhesion. -Inhibition proved particularly useful for

expressing

inhibits

the existence also

cytokine-treated

bind

to

VCAMl,

HUVECs (5).

I

II TNF. 4h

W’

VCAMl assessed.

or Mab

HL60 24h) 4B9

adhesion to or untreated, as indicated

TNF.24h

HWECs: HWECs, either were preincubated (each 10 ug/m1,30’,23”C)

351

treated with BBll and

of

with TNF (10 and/or antiHL60 adhesion

Vol.

171, No. 1, 1990

BIOCHEMICAL

AND BIOPHYSICAL

Adhesion in

to VCAMl is fully blocked with 4B9 blocks conjunction

treated pathway

the VCAMl

pathway

predominates

binding

PMN

epitope

to

HWECs

partially

in

recently

published

HT29 binding

As

shown

in

60.3

abolishes

with

HWECs.

These

using

different

established

Recent

studies

from

to cytokine-treated

colon

carcinoma

course

reported

to HWECs.

lines

in adhesion

data

Figure

are

4, BBll

also

completely,

in agreement

anti-ELAMl

and

that

HT29

lines, with

cells

have

tumors

HWECs

cell

consistent

solid

with

anti-CD18

(14,15).

including

that

increased

HT29 adhesion

published

information

In summary, cloned

HT29,

bound

should

surface a

(14). expressed

to ELAMl

adhesion

the

role

be a general

and cells

method

Mab to the

ELAMl.

ELAMl

COS

In particular,

blocking

protein of

ELAMl-mediated, with

proteins.

potent

cell

all

a

series

to HWECs with

adhesion

selectively

4h,

Ue

in COS

recently

This

both

--in vitro

transiently

for

the generation have

and --in

by

help adhesion

in evaluating and

transen-

vivo.

‘02 0.4

I 0.2

0

kgZl&/ml

PMN

l.1g/m1,30’,23”k) or treated

with

“E E . =* u” 3 II

0

TNF

-

+-I-++

MA6

-

-

60.3

-++ 6611 Both

and HT29 adhesion

-

-

BBll

to HUVECs: HUVECs, either treated with 4h) or untreated, were preincubated with BBll (10 where indicated, and adhesion of PMN (either untreated Mab 60.3,50 ug/m1,20’,23°C) or HT29 cells was assessed.

352

this

human endothelial

greatly cell

of Mabs to

generated

cytokine-induced

Mab should

expressing

2- lo

x b 2 4

a

have

the increased showing that

confirming

we

in leukocyte/endothelial

migration,

1 8 Q t

of

increase

(16).

immunization

cDNAs

new cell

is

a number

In particular,

ELAMl-mediated adhere

shown

show a selective

cells (4). Figure 4 shows that BBll abolishes completely binding of HT29 cells to HWECs treated with TNF for

dothelial

ELAMl

24h when

(13).

human cell

means

the

and at

the

(4). in conjunction

information

when n=4),

which recognizes a blocking 60.3, binding of PMN to cytokine-

Mab

assays

and

treatment, by BBll,

3).

inhibits

to cytokine-treated

antibodies

time

HWECs. our

inhibits,

PMN binding

of

(Fig.

on CD18, partially

treated

by the anti-VCAMl Mab 4B9 (12). BBll AL60 binding completely to cytokine-

cytokine at 4h post (60-9011: inhibition

HWECs, both predominates

RESEARCH COMMUNICATIONS

Vol.

171, No. 1, 1990

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

REFERENCES 1. Bevilacqua, M.P., Pober, J.S., Mendrick, D.L., Cotran, R.S., and Jr. (1987) Proc. Natl. Acad. Sci. U.S.A. 84, 9238Gimbrone, M.A., 9242. 2. Bevilacqua, M.P., Stengelin, S., Gimbrone, M.A., Jr., and Seed, B. (1989) Science 243, 1160-1165. R.S., and Pober, J. (1988) in: Endothelial Cell Biology, 3. Cotran, eds. Simionescu, N. and Simionescu, M. (Plenum, New York), pp.335347. 4. Hession, C., Osborn, L., 5. 6. 7. 8. 9.

Goff, D., Chi-Rosso, G., Vassallo, C., Pasek, M., Pittack, C., Tizard, R., Goelz, S., McCarthy, K., Hopple, S., and Lobb, R. (1990) Proc. Natl. Acad. Sci. U.S.A. 87, 1673-1677. Osborn, L., Hession, C., Tizard, R., Vassallo, C., Luhowskyj, S., Chi-Rosso, G., and Lobb, R. (1989) Cell 59, 1203-1211. Barsoum, J. (1990) DNA. In Press. Lerner, R.A. (1981) Yale J. Biol. Med. 54, 387-402. Staunton, D., Dustin, M., and Springer, T. (1989) Nature 339, 61-64. Yamasaki, K., Taga, T., Hirata, Y., Yawata, H., Kawanishi, Y., Seed, B Taniguchi, T., Hirano, T., and Kishimoto, T. (1988) Science 241,

8&828. 10. Gearing, 3667-3676.

D., King, J.,

Gough, N., and Nicola,

N. (1989) EMBOJ. 12,

11. Seed, B., and Aruffo, A. (1987) Proc. Natl. Acad. Sci. U.S.A. 84, 3365-3369. 12. Carlos, T.M., Schwartz, B.R., Kovach, N.L., Yee, E., Rosa, M., Osborn, L., Chi-Rosso, G., Newman,B., Lobb, R., and Harlan, J.H. (1990) Blood. In Press. 13. Luscinskas, F.W., Brock, A.F., Arnaout, M.A., and Gimbrone, M.A., Jr. (1989) J. Immunol. 142, 2257-2263. 14. Dejana, E., Bertocchi, F., Bortolami, M.C., Regonisi, A., Tonta, A., Brevario, F., and Giavazzi, R. (1988) J. Clin. Invest. 82, 14661470.

15. Rice, G.E., Gimbrone, M.A., Jr., and Bevilacqua, M.P. (1988) Amer. J. Pathol. 133, 204-210. 16. Rice, G.E., and Bevilacqua, M.P. (1989) Science 246, 1303-1305.

353