INFECTION AND IMMUNITY, July 1988, p. 1807-18130019-9567/88/071807-07$02.00/0Copyright © 1988, American Society for Microbiology

Production and Characterization of Monoclonal Antibodies to theProtective Antigen Component of Bacillus anthracis Toxin

STEPHEN F. LITTLE,* STEPHEN H. LEPPLA, AND ELSA CORAt

U.S. Army Medical Research Institute of Infectious Diseases, Fort Detrick, Frederick, Maryland 21701-5011

Received 8 February 1988/Accepted 15 April 1988

Thirty-six monoclonal antibodies to the protective antigen protein of Bacillus anthracis exotoxin have beencharacterized for affinity, antibody subtype, competitive binding to antigenic regions, and ability to neutralizelethal and edema toxin activities. At least 23 antigenic regions were detected on protective antigen by a

blocking, enzyme-linked immunosorbent assay. Two clones, 3B6 and 14B7, competed for a single antigenicregion and neutralized the activity of both the lethal toxin in vivo (Fisher 344 rat) and the edema toxin in vitro(CHO cells). These two antibodies blocked the binding of _251-labeled protective antigen to FRL-103 cells. Ourresults support the proposal that binding of protective antigen to cell receptors is required for expression oftoxicity.

Two plasmid-encoded virulence factors, a capsule andexotoxin (10, 17, 28), have been described for Bacillusanthracis. The exotoxin is composed of three proteins:protective antigen (PA), edema factor (EF), and lethal factor(LF). Protective antigen induces immunity to infection withB. anthracis and is the major component in the commerciallyprepared, cell-free U.S. and British vaccines (11, 16, 26).Toxic activity is expressed only when PA is combined withEF or LF. A combination ofPA with EF or LF forms edematoxin or lethal toxin, respectively. The edematous lesionscaused by edema toxin (7, 23, 25) have been attributed to thecalmodulin-dependent, adenylate cyclase activity of EF (14,15). Lethal toxin causes severe pulmonary edema and deathin Fisher 344 rats (2, 7), is rapidly cytolytic for macrophages(8), and inhibits growth of certain cell lines (15). No enzy-matic activity has been identified for LF. Both the lethal andedema toxins inhibit the release of superoxide anion (02-)from polymorphonuclear leukocytes induced by bacterialproducts (32).With our current knowledge of receptor mechanisms,

early studies on anthrax toxin (14, 18) are now seen to beconsistent with the hypothesis that PA must bind to recep-tors to permit subsequent binding of LF or EF. PA may alsocontain a common receptor site(s) for LF and EF, as

demonstrated by competition between LF and EF in toxicityassays (6, 14, 24).

This report describes the preparation and characterizationof monoclonal antibodies to PA. Two monoclonal antibodieswere obtained which neutralize lethal and edema toxinactivity by inhibiting binding of PA to cell receptors.

MATERIALS AND METHODS

B. anthracis antigens. Purified PA, LF, and EF antigenswere prepared as previously described (14). Each proteinwas at least 95% pure, as determined by gel electrophoresis,and contained <0.1% contamination by other toxin compo-nents, as judged from toxicity assays.

Immunizations. Female BALB/c mice were immunized

* Corresponding author.t Present address: Department of Biochemistry and Nutrition,

University of Puerto Rico, San Juan.

with purified PA by a different procedure for each fusionperformed. Immunized mice used as spleen donors demon-strated high titers by enzyme-linked immunosorbent assay(ELISA) for PA-reactive antibodies. See Table 1 for immu-nization schedules and procedures.

Hybridomnas. Spleen cells from immunized mice werefused with either logarithmically growing SP2/0-Agl4 mye-loma cells (5) for fusions PAI, PAII, PA2II, PA2III andPA2IV or P3x-63-Ag8-653 myeloma cells for fusions PAIII

TABLE 1. Immunization of female BALB/c mice withpurified PA

Fusion Immunization PA (Lg) Diluent Injection schedule'Immunation P(g (0.5 Ml)b

I i.m. 20 FCA Day 0i.m. 20 FIA 17 and 29 weeksi.v. 50 Saline 40 weeks (3 days PF)

II i.m. 20 FCA Day 0i.m. 20 FIA 29 weeksi.v. 50 Saline 61 weeks (3 days PF)

III i.m. 1 FCA Day 0i.m. 10 FIA 3 and 10 weeksi.m. 10 IFA 35 weeks (3 days PF)

IV"l i.p. 50 FCA 15 and 8 days PFi.p. 5 Saline 3 days PFi.p. and i.v. 2.5 Saline 2 and 1 days PF

211 i.p. 50 FCA Day 0 and 7i.p. 5 Saline Day 12i.p. and i.v. 2.5 Saline Day 13 and 14i.v. 20 Saline 8 weeksi.v. 100 Saline 25 weeks (day 3 PF)

2111 i.m. 10 FCA Day 0i.m. 10 FA 4 weeksi.v. 100 Saline 16 weeks (day 3 PF)

21V i.m. 10 FCA Day 0i.m. 10 FIA 4 weeksi.p. 200 Saline 16 weeks (day 3 PF)i.p. 100 Saline Day 2 PFi.p. 100 Saline Day 1 PF

"im., Intramuscular.b FCA, Freund complete adjuvant; FIA, Freund incomplete adjuvant.'PF, Prefusion."Method of Cianfriglia et al. (4).

1807

Vol. 56, No. 7

on February 21, 2021 by guest

http://iai.asm.org/

Dow

nloaded from

1808 LITTLE ET AL.

and PAIV. The latter two fusions were performed by Hazel-ton Biotechnology Corp., Vienna, Va. Hybridoma cultureswere screened by the ELISA method described below.Positive hybridomas were subcloned twice by limiting dilu-tion, and one to four clones for each hybridoma wereexpanded in vitro. Ascites from each hybridoma were pro-duced by inoculating 1 x 106 hybridoma cells intraperitone-ally (i.p.) into BALB/c female mice 2 to 8 weeks after i.p.injection of 0.5 ml of pristane (2,6,10,14-tetramethyl penta-decane; Sigma Chemical Co., St. Louis, Mo.). The ascitesfrom each clone were pooled, clarified by centrifugation(2,000 x g for 20 min), and stored at -70°C.ELISA. The ELISA previously described (16) was used

with a few changes. (i) Dilutions were made in phosphate-buffered saline (PBS) with 0.5% gelatin and 0.05% Tween 20.(ii) Plates were incubated for 2 h at 37°C with rabbitantiserum to mouse immunoglobulin G (IgG), 1gM, and IgA(Calbiochem-Behring, La Jolla, Calif.), diluted 1:400 prior toaddition of horseradish peroxidase conjugated to staphylo-coccal protein A (Sigma). (iii) Horseradish peroxidase-pro-tein A was incubated for 30 min at room temperature. (iv)Wells with absorbance values >0.200 nm were selected aspositive wells.

Determination of antibody subclass. An ELISA-based sub-type kit (Boehringer Mannheim Biochemicals, Indianapolis,Ind.) was used to determine immunoglobulin subtype andsubclass specificity.

Affinities of MAbs. The apparent affinity of each purifiedmonoclonal antibody (MAb) for PA was estimated from theslope of the ELISA-binding curve (31) by using the ELISAdescribed above instead of a radioimmunoassay (29). Goatantibody to mouse IgG and IgM conjugated to horseradishperoxidase (Kirkegaard and Perry, Gaithersburg, Md.) at 1:1,000 (0.5 jig/ml) was substituted for the rabbit antiserum tomouse IgG, IgM, and IgA and horseradish peroxidase-protein A reagents in the ELISA assay described above.Incubation was done for 2 h at 37°C.

Purification of MAb. Ascitic fluids (2 to 3 ml) from IgGmonoclones were dialyzed against 20 mM Tris hydrochloride(pH 8.0), centrifuged (10,000 x g, 30 min), and pumped ontoa 14-ml column of DEAE-Affi-Gel Blue (Bio-Rad Laborato-ries, Rockville Centre, N.Y.). Elution was done with agradient of 60 ml each of 20 mM Tris hydrochloride (pH 8.0)and 20 mM Tris hydrochloride (pH 8.0) with 100 mM NaCl.Fractions (1.0 ml) were assayed by ELISA. The purified IgGwas precipitated with 50% saturated (NH4)2SO4 (pH 7.5).The pellet was suspended in water, extensively dialyzedagainst 10 mM Tris hydrochloride (pH 8.0), and frozen at-700C.

Ascitic fluids containing IgM MAb were purified by theaddition of (NH4)2SO4 (pH 7.5) to 45% saturation. After 1 h,the material was centrifuged at 10,000 x g for 30 min, thepellet was resuspended in 20 mM Tris hydrochloride (pH8.0), and the precipitation was repeated to 40% saturation.The precipitate was dissolved in and dialyzed extensivelyagainst PBS (pH 8.4) at 4°C and frozen at -70°C.

Determination of immunoglobulin concentration. Radialimmunodiffusion plates were prepared with rabbit anti-mouse IgG (Miles Scientific, Naperville, Ill.) or rabbit anti-mouse IgM (Kirkegaard and Perry) in 0.5% agarose. Immu-noglobulin concentrations were calculated from either IgGor IgM (Sigma) standard curves after incubation for 18 to 20h (IgG) or 48 h (IgM) in a moist chamber at 23°C.

IEF. The monoclonal nature of each cell line was identifiedby isoelectric focusing (IEF). Purified immunoglobulins fromIgG clones and marker proteins (IEF pl calibration kit

[Pharmacia-LKB Biotechnology, Inc., Piscataway, N.J.] orprestained IEF standards [Bio-Rad]) were analyzed on pre-cast polyacrylamide gels (pH 3.5 to 9.5; Pharamicia-LKBBiotechnology, Inc.) according to the directions of themanufacturer. Protein bands were visualized by Coomassieblue stain.

In vivo neutralization. Neutralization of lethal toxin wasassayed in Fisher 344 male rats (225 to 250 g) by intravenous(i.v.) injection of mixtures of 40 jig of PA, 8 p.g of LF, and 4mg of protein from each monoclone after incubation at 37°Cfor 1 h. Protein concentrations were estimated by using therelationship of 1 A280 = 1 mg of protein per ml. Only one ratper monoclone was tested initially. Ascites fluids that pro-tected the rats or demonstrated a delayed time to death wereretested at lower concentrations until no protection wasobserved. Four rats were tested for each ascitic fluid thatneutralized the toxin. Rats were observed for 2 days.

In vitro neutralization assay. In the presence of edematoxin, Chinese hamster ovary (CHO) cells undergo an elon-gation in response to the adenylate cyclase activity of EF

TABLE 2. Affinities and subtypes of 37 PA MAb-producinghybridoma clones raised in seven separate fusions

Hybridoma Affinity Antibodyclone (,ig) subtype

PAI

PAII

PAIII

PAIV

PA2I1

PA2111

PA21V

IG7-2-12D3-3-12D5-1-13B6-1-13D2-1-13F3-2-23F10-1-16C3-1-1

1C5-1-12D3-1-12G4-1-13C5-1-17C3-1-1

6B7-1-1

lF2-1-1

2F7-1-12F9-1-13D11-1-11OD2-1-1lOG4-1-114B7-1-114C7-1-116E11-2-118C2-1-118G8-2-120C5-1-1

2B8-1-18C8-1-19B4-2-19EII-1-1

3E2-1-14F3-2-15B3-2-15G3-1-16F3-1-116E5-1-116E9-1-1

0.0640.0330.0480.5210.0790.0700.0440.054

0.1150.0540.1440.2100.098

IgGlIgGlIgGlIgGlIgGIIgGlIgG2bIgGl

IgGlIgGlIgGlIgGlIgGl

IgGl

IgGl

IgG2bIgGlIgG2aIgGlIgGlIgGlIgGlIgGlIgGlIgGlIgGl

IgG2algG2aIgG2aIgGl

IgMIgMlgMIgG2bIgMIgMlgM

275

0.047

200.0700.1030.0360.0470.0690.1070.0510.3990.4210.092

0.3770.3890.22820

2.94.74.7

378.96.30.945

INFECT. IMMUN.

on February 21, 2021 by guest

http://iai.asm.org/

Dow

nloaded from

MONOCLONAL ANTIBODIES TO PROTECTIVE ANTIGEN 1809

CATHODE (-)APi Std 18C2 268 1002 3D11 2F9 3F3

930

:,~~~~~~. :

08.458.15

7.35

6.856.55

5.85

5.20

4.55

5G3 16E11 14C7 14B7 20CS 9E11 18C2 18G8 1G7 3D2 3C5

!L4FW-"

*.F.- IIw['P Ar _ > _

.....

'._-

'iN

BPt Std 3C5 6B7 6B7 6C3 2F7 3F1O 3C5 8C8 984 2D3-1 1C5 3B6 1OG4 1 F2 7C3 2G4 205 2D3 Std

8.658.45 .W _

8.15..

a

I

4.55 -

3. 50. _i

ANODE (+)FIG. 1. IEF patterns of purified IgG from PA MAbs separated on polyacrylamide IEF gels. Std, Standard.

(14). CHO cells, cultured in Eagle minimal essential mediumcontaining nonessential amino acids supplemented with 25mM HEPES (N-2-hydroxyethylpiperazine-N'-2-ethanesul-fonic acid), 0.05 mg of gentamicin per ml, and 10% fetalbovine serum, were plated thinly (3 x 104 to 6 x 104 cells perml) in 96-well, flat-bottom, cell culture plates and incubatedovernight before use. MAbs were serially diluted in separate96-well plates containing 100 ful of toxin (100 ng of EF and 50ng of PA per ml) in cell culture medium. After 1 h at 37°C, 75iul of the toxin-antibody mixture was transferred into thecorresponding wells of a CHO cell plate. The final amountsper well were 8 ng ofEF and 4 ng of PA. After incubation for3 h at 37°C, the wells were inspected visually to determinethe smallest amount of MAb that inhibited CHO cell elon-gation.

Biotinylation of purified MAb. Purified MAbs were biotin-ylated by a procedure modified from that of Stahli et al. (22).Briefly, MAbs were adjusted to a concentration of 1 mg ofIgG per ml after dialysis in 0.1 M NaHCO3-0.15 M NaCl,buffered to pH 8.5. N-Hydroxysuccinimidyl biotin (1 mg/mlin dimethyl sulfoxide; Sigma) was added (0.1 ml per ml ofMAb) and incubated for 3 h at room temperature. Biotinyl-ated MAbs were then dialyzed in 10 mM Tris hydrochloride(pH 8.0) and frozen at -70°C. For use, glycerol was added to50% and vials were held at -20°C.

Competitive binding assay. Competition between MAbs fora single epitope was measured by ELISA, using each puri-fied MAb to compete for the binding of a limiting concentra-tion of biotinylated MAb. Each blocking, nonbiotinylatedMAb (MAb1) was used at a concentration which, in the

S d

7.35

6.85 m -.6.55 _o

d-W4*_

.i5.85

5.20 _ UIr!

VOL. 56, 1988

I_ _qm

I "'MiL. .......

FrW - ..--W W.w-.II

4MMMENENOW

.a

- -iZ& "anom-

--w

i on February 21, 2021 by guest

http://iai.asm.org/

Dow

nloaded from

1810 LITTLE ET AL.

CATHODE (-)P1 Std 386 14B7 14C7 1487+ 14C7 3B6+14B7 3B6+14C7 Std

9.60

8.80 _

_IF

7.00 E

._;;

.&_::.::s_

_fiM_

:'_.ix1 _.:

s._:

*. :.. t:. _''*:: : : .. :: j_.. ._::.: ..*iF'?*d 3-

t§::::::tiw3w':iiF X @84_#i: _..i F.e . .. ,te. 1 -.... _

6.00 'P

ANODE (+)FIG. 2. IEF patterns of purified IgG from MAbs 3B6, 14B7, and 14C7 individually and in mixtures separated on polyacrylamide IEF gels.

Std, Standard.

regular ELISA (described above), gave an A405 value >2.0,while not giving significant, nonspecific A405 values oncontrol plates lacking PA. After 50 ll of MAb1 was incu-bated for 1 h at 37°C, 50 pul of biotinylated purified MAb wasadded to each well (MAb2), without removing unboundMAb1, and the plate was incubated for 2 h at 37°C. Biotincontrol wells contained 50 pul of biotinylated, purified MAband 50 ,ul of PBS with 0.5% gelatin-0.05% Tween 20. Wellswere washed four times with PBS-0.05% Tween 20. Avidinconjugated to horseradish peroxidase (Sigma) and diluted to1:1,000 (1 pug of avidin-horseradish peroxidase per ml) was

ASCITES (i g PROTEIN)

FIG. 3. Survival of Fisher 344 rats injected i.v. with 40 [ig of PAand 8 jig of LF. Survival groups consist of four guinea pigs each. Allother groups consist of one to two guinea pigs.

added to each well (100 plI per well); plates were incubatedfor 1 h at 37°C. For color development, wells were washedsix times with PBS-0.05% Tween 20, and 100 ,ul of ABTSreagent [2,2'-azino-bis(3-ethylbenzthiazolinesulfonic acid);Sigma], prepared as 1 mg of reagent per ml in 0.1 M sodiumcitrate buffer (pH 4.0) and 0.003% hydrogen peroxide, wasadded to each well. The reaction was stopped after 15 min byadding 50 [lI of 10% sodium dodecyl sulfate per well.Absorbance was read on a Dynatech Microelisa AutoReader MR580 (Dynatech Instruments, Inc., Chantilly, Va.)at a wavelength of 405 nm. Dilutions for MAb2 were selectedto give a minimum adsorption value of 0.300 for biotincontrol wells. The observed A405 values were used to calcu-late the percentage that MAb1 blocked the binding of thebiotinylated MAb2 by the following formula: % blocking = (1- test well A405/control well A405) x 100.

Selected MAbs were also tested in a competitive bindingassay similar to the procedure described above, except thatMAb1 was assayed on the ELISA microtiter plate prior tothe addition of MAb2.

Binding of '251-labeled PA to FRL-103 cells. FRL-103 cells(fetal Rhesus lung, ATCC CCL-160), plated in 24-well cellculture dishes in Eagle minimal essential medium containingnonessential amino acids and supplemented with 10% fetalbovine serum, were grown to confluence. The monolayerwas washed, and 0.40 ml of H199 medium, supplementedwith 25 mM HEPES and 2% fetal bovine serum and contain-ing 0.2 ,ug of 125I-labeled PA per ml (2.4 x 107 cpm/pug) andindividual ascites MAb (5 pLg of immunoglobulin per ml), wasadded to each well. Labeled PA, provided by A. M. Fried-lander, was prepared by Bolton-Hunter reagent and retainedbiological activity. After incubation for 8 h at 0°C, the

.::!::U:̂ 9 41;x}i

5.10

4.65

..... ..

INFECT. IMMUN.

Ilzl-""as:: ::: .:.-

on February 21, 2021 by guest

http://iai.asm.org/

Dow

nloaded from

MONOCLONAL ANTIBODIES TO PROTECTIVE ANTIGEN 1811

monolayers were washed four times with Hanks balancedsalt solution, dissolved in 1.0 ml of 0.1 M NaOH, andcounted. Controls containing 30 ,ug of nonradioactive PA perml averaged 740 cpm, and wells with no competing MAbaveraged 9,000 cpm.

RESULTS AND DISCUSSION

This report describes the isolation and partial characteri-zation of 37 MAbs against the PA component of the B.anthracis toxin. All PA MAbs were specific for PA and didnot bind to LF or, EF, as determined by ELISA. Table 2 liststhe 37 hybridoma clones and their apparent affinities andimmunoglobulin subtypes. Five separate fusions (PAI, PAII,PA2II, PA2III, and PA2IV) with the SP2/0 myeloma cell lineresulted in 35 clones. Fusion experiments PAIII and PAIV,in which the P3x-63-Ag8-653 myeloma cell line was used,resulted in only one clone each. Designations of clones inthis report consist of the first three alphanumeric charactersfor each clone except for PA2II 2D3-1-1, designated 2D3-1 todistinguish it from PAI 2D3-3-1 (2D3).

Thirty-one MAbs were typed as IgG clones (24 IgGl, 4IgG2a, and 3 IgG2b subtype clo-nes), and six were typed asIgM clones (Table 2). All 37 clones contained the kappa lightchain. We have no explanation why three of four clones infusion PA2III were of the IgG2a subtype. The six IgM classhybridomas obtained from fusion PA2IV may have resultedfrom the i.p. and i.v. boosters administered on days -2 and-1. MAbs can be placed into one of three groups on thebasis of their relative affinities (31): high affinity (-0.2 [Lg),moderate affinity (.0.2 to s1 ,ug), and low affinity (>1 ,ug).The IgGl subtype MAb comprised 18 of 20 clones producinghigh-affinity MAbs. The remaining 17 clones produced mod-erate- and low-affinity MAbs.IEF of purified immunoglobulins from IgG hybridomas on

polyacrylamide gels (Fig. 1A and B) demonstrated bandingpattern and pl value differences among the MAbs. OnlyMAbs 14B7 and 14C7 appeared to have identical pIs and IEFpatterns (Fig. 2). These two MAbs were located in adjacentwells of the original 96-well fusion plate, and contaminationfrom one well into another probably occurred. MAb 3B6,which was functionally similar to 14B7 and 14C7 (discussedbelow), was different from them chemically, as shown byrunning these MAbs individually and in mixtures on IEF gels(Fig. 2). These results indicate that 36 individual PA hybrid-omas were obtained. We were unable to resolve adequatelythe IEF patterns of the IgM clones by using agarose IEF.

In vivo neutralization studies showed a significant delay oftime to death (2D3 and 2D5) or survival (3B6, 14B7, and14C7) of Fisher 344 male rats after i.v. injection of mixturesof MAb ascites and PA plus LF (Fig. 3). Lethal toxin (40 jigof PA and 8 ,ug of LF) was neutralized by 4, 0.4, and 0.04 mgof 3B6, 14C7, and 14B7 ascites protein, respectively. Neu-tralization was confirmed with three more rats protected foreach concentration of MAb ascites. The apparently greaterprotection of 14B7 over 14C7 (Fig. 3) can be explained by the2.3-fold-higher immunoglobulin concentration (data notshown) of the 14B7 ascites. The ascites used from these twohybridoma clones were produced at different times.We observed neutralization of edema toxin (4 ng of PA

and 8 ng of EF) activity in CHO cells in vitro with 8, 1, and5 ,ug of purified 3B6, 14B7, and 14C7 immunoglobulin,respectively. A few other MAbs demonstrated an inhibitionof edema toxin activity at undiluted or slightly dilutedconcentrations. Although this assay, which uses a visual,subjective end point, was not performed in a manner to yield

TABLE 3. Determinants on the PA molecule identifiedby blocking ELISA

Determinant Monoclone(s) Determinant Monoclone

1 1G7, 2D3, 2D5, 3D2, 12 18C23C5, 2F9, 10D2, 10G4 13 18G8

2 3B6, 14B7, 14C7 14 2B83 3F3, 3F10, 6C3 15 8C84 1CS, 2D3-1 16 9E115 2G4, 20C5 17 3E26 7C3 18 4F37 6B7 29 5B38 1F2 20 5G39 2F7 21 6F310 3D11, 9B4 22 16E511 16E11 23 16E9

highly precise, quantitative data, as would be obtained bymeasurement of cyclic AMP (14), the assay did demonstratethat these three MAbs inhibited the cellular elongationresponse of CHO cells in the presence of edema toxin.

Analysis of epitope diversity among MAbs has beendescribed by others who used various competitive-bindingassays (1, 3, 12, 13, 19, 20, 21, 27, 30, 31). A competitive-binding ELISA, with biotinylated MAb and fixed concentra-tions of unlabeled competing MAb, allowed us to designateat least 23 different antigenic regions on the PA molecule(Table 3). The distinctive binding features of 3B6 and 14B7were further demonstrated in a competitive-binding ELISAwith biotinylated 3B6 or 14B7 and various concentrations ofseveral representative, unlabeled, competing MAbs (Fig. 4and 5). The data from the competitive-binding experimentsalso indicated that 3B6, 14B7, and 14C7 recognized the sameantigenic region on the PA molecule. These preliminaryresults will aid us in the study of the various binding regionsof the PA molecule.Monoclonal antibodies 3B6, 14B7, and 14C7 inhibited

binding of 125I-labeled PA to the receptor site on FRL-103cells by 97.5, 95.1, and 96.5%, respectively, whereas the

120

R 100

80

-JLLC 60-J

IU-

CDC- 40

z

H:20co

0O.01 0.1 1 10 100

UNLABELED MAb Ig (Mug/ml)FIG. 4. Competitive antibody-binding assay with biotinylated

and nonlabeled MAbs. The binding of biotinylated 3B6 was mea-sured in the presence of different concentrations of competing,unlabeled MAbs: 14B7 (-), 3B6 (V), 3F3 (x), 9E11 (A), lCS (*),and 2D3 (+).

VOL. 56, 1988

on February 21, 2021 by guest

http://iai.asm.org/

Dow

nloaded from

1812 LITTLE ET AL.

120

R 100

M 80uJ

Lim 60-J

LL-

CDCD 40z

C

H 20co

o 110.01 0.1 1 10 100

UNLABELED MAb Ig (/, g/ml)

FIG. 5. Competitive antibody-binding assay with biotinylatedand nonlabeled MAbs. The binding of biotinylated 14B7 was mea-sured in the presence of different concentrations of competing,unlabeled MAbs: 14B7 (U), 3B6 (V), 2B8 (x), 6B7 (A), 2F9 (*), and16E9 (+).

values for the controls and the other MAbs ranged from 0 to49.2%. The antigenic region of PA recognized by 3B6, 14B7,and 14C7 appears to be the cell receptor-binding site(s), thusexplaining the neutralizing ability of these three clonesagainst the activity of either lethal or edema toxin. Theseresults support previous proposals that binding of PA to itscell receptor is required prior to expression of lethal oredema toxin activity. It is probable then that PA acts as theactive B fragment containing receptor-binding activity andthat LF and EF act as active A fragments containingbiological activity in the usual A-B enzyme-binding structureof intracellular bacterial protein toxins described by Gill (9).

In summary, we have developed and characterized 36different MAbs against the PA component of anthrax toxin.Two of these MAbs were able to neutralize the lethal andedema toxin activities by inhibiting the binding of PA to cellreceptors, thus providing supporting evidence that initialbinding of PA to cell receptors is required for toxin activity.

ACKNOWLEDGMENTS

We thank C. Osterling for preparation and maintenance of thehybridomas and R. Adams for technical assistance.

LITERATURE CITED

1. Adam, E., I. Nasz, A. Lengyel, J. Erdei, and J. Fachet. 1987.Determination of different antigenic sites on the adenovirushexon using monoclonal antibodies. Arch. Virol. 93:261-271.

2. Beall, F. A., M. J. Taylor, and C. B. Thorne. 1962. Rapid lethaleffect in rats of a third component found upon fractionating thetoxin of Bacillus anthracis. J. Bacteriol. 83:1274-1280.

3. Carbone, F. R., and Y. Paterson. 1985. Monoclonal antibodiesto horse cytochrome C expressing four distinct idiotypes dis-tribute among two sites on the native protein. J. Immunol.135:2609-2616.

4. Cianfriglia, M., D. Armellini, A. Massone, and M. Mariani.1983. Simple immunization protocol for high frequency produc-tion of soluble antigen-specific hybridomas. Hybridoma 2:451-457.

5. Earley, E. M., and M. C. Osterling. 1985. Fusion of mouse-mouse cells to produce hybridomas secreting monoclonal anti-body. J. Tissue Culture Methods 9:141-146.

6. Ezzell, J. W., B. E. Ivins, and S. H. Leppla. 1984. Immunoelec-trophoretic analysis, toxicity, and kinetics of in vitro productionof the protective antigen and lethal factor components ofBacillus anthracis toxin. Infect. Immun. 45:761-767.

7. Fish, D. C., B. G. Mahlandt, J. P. Dobbs, and R. E. Lincoln.1968. Purification and properties of in vitro-produced anthraxtoxin components. J. Bacteriol. 95:907-918.

8. Friedlander, A. M. 1986. Macrophages are sensitive to anthraxlethal toxin through an acid-dependent process. J. Biol. Chem.261:7123-7126.

9. Gill, D. M. 1978. Seven toxic peptides that cross cell mem-branes, p. 291-332. In J. Jeljaszewicz and T. Wadstrom (ed.),Bacterial toxins and cell membranes. Academic Press, Inc.,New York.

10. Green, B. D., L. Battisti, T. M. Koehler, C. B. Thorne, and B. E.Ivins. 1985. Demonstration of a capsule plasmid in Bacillusanthracis. Infect. Immun. 49:291-297.

11. Hambleton, P., J. A. Carman, and J. Melling. 1984. Anthrax: thedisease in relation to vaccines. Vaccine 2:125-132.

12. Huber-Lukac, M., F. Jaquet, P. Luethy, R. Huetter, and D. G.Braun. 1986. Characterization of monoclonal antibodies to acrystal protein of Bacillus thuringiensis subsp. kurstaki. Infect.Immun. 54:228-232.

13. Huber-Lukac, M., P. Luthy, and D. G. Braun. 1983. Specificitiesof monoclonal antibodies against the activated delta-endotoxinof Bacillus thuringiensis var. thuringiensis. Infect. Immun.40:608-612.

14. Leppla, S. H. 1982. Anthrax toxin edema factor: a bacterialadenylate cyclase that increases cyclic AMP concentrations ineukaryotic cells. Proc. Natl. Acad. Sci. USA 79:3162-3166.

15. Leppla, S. H., B. E. Ivins, and J. W. Ezzell, Jr. 1985. Anthraxtoxin, p. 63-66. In L. Leive, P. F. Bonventre, J. A. Morello, S.Schlesinger, S. D. Silver, H. C. Wu (ed.), Microbiology-85.American Society for Microbiology, Washington, D.C.

16. Little, S. F., and G. B. Knudson. 1986. Comparative efficacy ofBacillus anthracis live spore vaccine and protective antigenvaccine against anthrax in the guinea pig. Infect. Immun.52:509-512.

17. Mikesell, P., B. E. Ivins, J. D. Ristroph, and T. M. Dreier. 1983.Evidence for plasmid-mediated toxin production in Bacillusanthracis. Infect. Immun. 39:371-376.

18. Molnar, D. M., and R. A. Altenbern. 1963. Alterations in thebiological activity of protective antigen of Bacillus anthracistoxin. Proc. Soc. Exp. Biol. Med. 114:294-297.

19. Ray, K., J. Vockley, and H. Harris. 1984. Epitopes of humanintestinal alkaline phosphatases, defined by monoclonal anti-bodies. FEBS Lett. 174:294-299.

20. Sheppard, A. J., D. Cussell, and M. Hughes. 1984. Productionand characterization of monoclonal antibodies to tetanus toxin.Infect. Immun. 43:710-714.

21. Smith-Gill, S. J., T. B. Lavoie, and C. R. Mainhart. 1984.Antigenic regions defined by monoclonal antibodies correspondto structural domains of avian lysozyme. J. Immunol. 133:384-393.

22. Stahli, C., V. Miggiano, J. Stocker, T. Staehelin, P. Haring, andB. Takacs. 1983. Distinction of epitopes by monoclonal antibod-ies. Methods Enzymol. 92:242-253.

23. Stanley, J. L., K. Sargeant, and H. Smith. 1960. Purification offactors I and II of the anthrax toxin produced in vivo. J. Gen.Microbiol. 22:206-218.

24. Stanley, J. L., and H. Smith. 1961. Purification of factor I andrecognition of a third factor of the anthrax toxin. J. Gen.Microbiol. 26:49-66.

25. Thorne, C. B., D. M. Molnar, and R. E. Strange. 1960. Produc-tion of toxin in vitro by Bacillus anthracis and its separation intotwo components. J. Bacteriol. 79:450-455.

26. Turnbull, P. C. B., M. G. Broster, J. A. Carman, R. J. Manchee,and J. Melling. 1986. Development of antibodies to protectiveantigen and lethal factor components of anthrax toxin in humansand guinea pigs and their relevance to protective immunity.

INFECT. IMMUN.

on February 21, 2021 by guest

http://iai.asm.org/

Dow

nloaded from

MONOCLONAL ANTIBODIES TO PROTECTIVE ANTIGEN 1813

Infect. Immun. 52:356-363.27. Tzartos, S. J., D. E. Rand, B. L. Einarson, and J. M. Lindstrom.

1981. Mapping of surface structures of electrophorus acetylcho-line receptor using monoclonal antibodies. J. Biol. Chem.256:8635-8645.

28. Uchida, I., R. Sekizaki, K. Hashimoto, and N. Terakado. 1985.Association of the encapsulation of Bacillus anthracis with a 60megadalton plasmid. J. Gen. Microbiol. 131:363-367.

29. Van Heyningen, V., D. J. H. Brock, and S. Van Heyningen. 1983.A simple method for ranking the affinities of monoclonal anti-bodies. J. Immunol. Methods 62:147-153.

30. Volk, W. A., B. Bizzini, R. M. Snyder, E. Bernhard, and R. R.Wagner. 1984. Neutralization of tetanus toxin by distinct mono-clonal antibodies binding to multiple epitopes on the toxinmolecule. Infect. Immun. 45:604-609.

31. Wnek, A. P., R. J. Strouse, and B. A. McClane. 1985. Produc-tion and characterization of monoclonal antibodies against Clos-tridium perfringens type A enterotoxin. Infect. Immun. 50:442-448.

32. Wright, G. G., and G. L. Mandell. 1986. Anthrax toxin blockspriming of neutrophils by lipopolysaccharide and by muramyldipeptide. J. Exp. Med. 164:1700-1709.

VOL. 56, 1988

on February 21, 2021 by guest

http://iai.asm.org/

Dow

nloaded from