Humoral Immune Response to the Class 3 Outer Membrane Protein ...

3 downloads 108 Views 2MB Size Report
Fredriksen, A. Haistensen, E. Holten, A.-K. Lindbak, H. N0k- leby, E. Rosenqvist, L. K. Solberg, ... John Wiley & Sons, Chichester,. England. 5. Blake, M. S., C. M. ...
INFEcrION AND IMMUNITY, Nov. 1993, p. 4734-4742

Vol. 61, No. 11

0019-9567/93/114734-09$02.00/0 Copyright © 1993, American Society for Microbiology

Humoral Immune Response to the Class 3 Outer Membrane Protein during the Course of Meningococcal Disease HILDE-KARI GUTTORMSEN,l* LEE M. WETZLER,2 ARE NAESS1 Medical Department B, Haukeland Hospital, University of Bergen, N-5021 Bergen, Norway, 1 and The Maxwell Finland Laboratory for Infectious Diseases, Boston City Hospital, Boston, Massachusetts 021182 Received 1 June 1993/Returned for modification 28 July 1993/Accepted 20 August 1993

We have determined the amounts of specific anti-class 3 outer membrane protein antibodies of immunoglobulin G (IgG), IgM, and IgA isotypes in patient sera during the course of meningococcal disease by using purified class 3 protein as the sensitizing antigen in an enzyme-linked immunosorbent assay. The class 3 protein was obtained from a variant of strain 44/76 (B:15:P1.7,16) lacking class 1 and class 4 outer membrane proteins. Serum samples from 25 patients with systemic meningococcal disease caused by organisms of various serotypes were collected during the course of disease. Seven of these patients had been immunized with a meningococcal outer membrane vesicle vaccine made from strain 44/76 prior to disease. An increase in specific anti-class 3 (type 15) outer membrane protein IgG antibodies was demonstrated in 22 of 25 patients (88%), regardless of the serotype of the infecting strain. This indicates that the specific anti-class 3 antibodies were reacting in part with epitopes not determined by the monoclonal antibodies used for serotyping. A considerable heterogeneity in antibody levels and IgG subclass response was seen. Most patients had low levels of anti-class 3 antibodies during the acute illness, with antibodies peaking during the second week of disease and returning to near baseline in sera collected 6 to 12 months after the onset of the disease. The majority of the specific anti-class 3 IgG antibodies were shown to bind to surface-exposed epitopes on the whole bacteria and to belong to IgGl and IgG3. The highest anti-class 3 IgG peak levels were seen in patients infected with strains of the homologous serotype after vaccination with the meningococcal outer membrane vesicle vaccine, suggesting an anamnestic response. However, these patients were not protected from meningococcal disease after immunization.

All meningococcal strains express either class 2 or class 3 protein as the predominant protein of the outer membrane. The ongoing epidemic of Neisseria meningitidis disease in Norway has been dominated by serogroup B strains containing outer membrane class 3 (serotype 15) protein (10). The class 2 and 3 proteins function as outer membrane porins (4), determine the serotype specificity, and are important immunodeterminants in humans after infection (19, 22). Anti-class 2 and 3 monoclonal antibodies have been shown to be bactericidal in vitro (6, 23, 25) and to protect animals in infectivity models (6, 25). Previously, it has been difficult to obtain purified class 3 protein without contamination by class 1 and class 4 outer membrane proteins. Different approaches have been used to try to quantify the amount of anti-class 2 and 3 antibodies in human sera, including the use of human sera as inhibitors of monoclonal antibodies that bind to meningococcal outer membrane proteins (21) and immunoblotting (19, 30). We were able to purify class 3 protein by detergent solubilization and chromatographic purification from a variant of strain 44/76 (B:15:P1.7,16) lacking both the class 1 and class 4 proteins in its outer membrane (3, 17, 29, 33). The amount of specific immunoglobulin G (IgG), IgM, IgA, and IgG subclass antibodies in serum recognizing class 3 during the course of systemic group B meningococcal disease was determined by using this purified class 3 protein as the sensitizing antigen in an enzyme-linked immunosorbent assay (ELISA). Knowledge about the immunogenicity of individual meningococcal proteins might be important for the development of a vaccine

*

against group B meningococci because of the low immunogenicity of the group B capsular polysaccharide (34). (Part of this work was presented at the Eighth International Pathogenic Neisseria Conference in Cuernavaca, Mexico, 1992 [12].) MATERIALS AND METHODS Patient population. We included serum samples from 25 patients with systemic meningococcal disease who had been admitted to the adult Infectious Diseases Division at Haukeland Hospital, Bergen, Norway, between November 1989 and December 1992. The diagnosis was based on clinical symptoms and findings together with positive cultures of N. meningitidis in cerebrospinal fluid and/or blood (Table 1). The median age of the patients was 19.4 years (range, 14 to 49 years), and all patients had normal complement activity as judged by the hemolytic activity of complement. Six of the patients were immunized twice in 1988 with a meningococcal outer membrane vesicle vaccine (MenBV) prepared from strain 44/76 (B:15:P1.7,16) as a part of the Norwegian meningococcal protection trial (2), one additional patient who originally received placebo vaccine in the same trial received one vaccine dose 4 weeks prior to her disease, and one patient originally receiving placebo was immunized during his convalescence. Informed consent was obtained from the participants. The Revised Helsinki Declaration was followed in the conduct of the trial. Meningococcal experimental strains. A variant of strain 44/76 (B:15:P1.7,16) lacking the class 1 protein (44/76A1, termed HI5) was kindly provided by Jan T. Poolman (29). Strain 44/76A1 was transformed with a plasmid construct of the class 4 gene (rpmM) with the structural portion substi-

Corresponding author. 4734

VOL. 61, 1993

IMMUNE RESPONSE TO MENINGOCOCCAL CLASS 3 PROTEIN

4735

TABLE 1. Patient meningococcal isolates and immunization statusa Patient no.

Bacteria isolated fromb:

Serogroup:serotype:serosubtype:immunotypec

MenBV immunization status

1 2 3 4 5 6 7

CSF CSF NP, B NP, B CSF NP, CSF B CSF NP, CSF NP, B, CSF NP, CSF CSF NP, B NP B NP, B, CSF NP, CSF NP, CSF NP, B, CSF B CSF NP, B B, CSF NP, B, CSF NP, B NP, B, CSF B, CSF

B:15:P1.7,16:L3,7,9 B:15:P1.7,16:L3,7,9 B:15:P1.7,16:L3,7,9 B:15:P1.2,7:L3,7,9 B:15:P1.2,7:L3,7,9 B:15:P1.2:L3,7,9 B:15:NT:L3,7,9 B:15:P1.2:L1,8,10 B:15:P1.12:L3,7,9 B:15:P1.12:L3,7,9 B:15:P1.12:L3,7,9 B:15:P1.12:L3,7,9 B:15:P1.12:L3,7,9 B:15:P1.12:L3,7,9 B:15:P1.12:L3,7,9 + L1,8,10 B:4:P1.14:L:3,7,9 B:NT:NT:L3,7,9 B:NT:P1.2:L3,7,9 B:NT:P1.12:L3,7,9 B:2b:P1.6:L3,7,9 + Li B:2b:NT:L1,8,10 W135:NT:NT:NIT C:2a:P1.2:L,3,7,9 C:2a:P1.2:L,3,7,9 C:2a:P1.2:L,3,7,9 C:2a:P1.2:L,3,7,9 C:2a:P1.2:L,3,7,9

Not immunized Not immunized Not immunized Not immunizedd Not immunized Not immunized Immunized, one dose 4 weeks prior to disease

8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

Immunized in 1988, Immunized in 1988, Immunized in 1988, Immunized in 1988, Not immunized Not immunized

3 years prior to disease 3 years prior to disease 3 years prior to disease 1 year prior to disease

Not immunized Not immunized Not immunized Not immunized Not immunized Not immunized Not immunized Not immunized Immunized in 1988, 3 1/2 years prior to disease Not immunized Not immunized Immunized in 1988, 3 1/2 years prior to disease

a Seven of the patients were immunized prior to their disease with a meningococcal outer membrane vesicle vaccine prepared from strain 44/76 (B:15:P1.7,16) as part of the Norwegian meningococcal protection trial (2). b

NP, nasopharynx; B, blood; CSF, cerebrospinal fluid. c The serogrouping, serotyping, and immunotyping were done by Andrew J. Fox, Public Health Laboratory, Manchester, United Kingdom (1, 16, 20). NT, not typeable; NIT, not immunotypeable. d This patient was immunized twice during his convalescence (weeks 5 and 9 after admission to the hospital).

tuted with an antibiotic resistance gene by the method described by Klugman et al. (17). We thus obtained a strain lacking both outer membrane protein class 1 and class 4 (44/76A1A4), from which we isolated the class 3 protein. Meningococcal porin isolation. Class 3 outer membrane protein was obtained from strain 44/76A1A4 by detergent solubilization and chromatographic purification (3, 33). This variant of strain 44/76 was used to ensure that only one type of porin was present in the preparation. There was no contamination with other outer membrane proteins as demonstrated by gel electrophoresis and Western blot (immunoblot) (18). Lipopolysaccharide contamination was less than 0.01% as judged by gel electrophoresis and silver staining

(28). ELISA determinations. The amount of serum antibody recognizing the class 3 outer membrane protein was determined in serum collected from each patient during his hospital stay and at subsequent intervals after hospitalization. ELISA was performed with class 3 as the sensitizing antigen (9) and secondary antibodies and/or conjugates from Sigma Chemical Co., St. Louis, Mo. Plates were gently mixed during antigen coating and all incubations. Optimal antigen and conjugate concentrations were determined by checkerboard titration for each isotype and subclass assay performed (14). Microtiter plates (Immuno Plates; Maxisorb; Nunc, Roskilde, Denmark) were sensitized by adding 0.1 ml of class 3 outer membrane protein in 0.1 M sodium carbonate buffer (pH 9.8) with 0.02% azide per well. The plates were incubated overnight at room temperature and washed five times with 0.12 M sodium acetate-0.154 M NaCl-0.05% Brij

35 (pH 7.5). The patient sera were diluted in 10 mM Tris (pH 8.2)-0.150 M NaCI-0.05% Brij 35-0.02% azide (incubation buffer), and 0.1 ml of the prediluted serum was added to the plates and serially diluted twofold in incubation buffer. After 2 h of incubation at room temperature, the plates were washed as described before and reincubated with appropriate alkaline phosphatase-conjugated secondary antibody in

incubation buffer for 2 h at room temperature. After washing, 0.1 ml of p-nitrophenyl phosphate (Sigma phosphatase 104; 1 mg/ml) in 10% diethanolamine (pH 9.8)-3 mM MgCl2 was added, and the plates were incubated at 37°C. After 1 h of incubation, the A405 was determined with a Titertek Multiskan MCC photometer (Flow Laboratories, Inc., McLean, Va.). This will be referred to as the standard ELISA protocol. A significant increase in anti-class 3 antibodies in serum is defined as a fivefold or higher increase in anti-class 3 antibody titers or levels. IgA and IgM titrations. For IgA and IgM titrations, the standard protocol using an antigen concentration of 2 ,ug/ml for the IgA ELISA and 1 ,ug/ml for the IgM ELISA was followed. Alkaline phosphatase-conjugated goat anti-human IgA and goat anti-human IgM were used at a dilution of 1:1,000. The antibody titers are reported as the reciprocal dilutions that gave an A4,5 of 1.0. Maximal optical density (OD) values less than 1 have been set equal to 1 in the

figures. IgG quantitation. For IgG quantitation, the standard protocol using an antigen concentration of 2 ,ug/ml and alkaline phosphatase-conjugated goat anti-human IgG at a dilution of 1:1,000 as the secondary antibody was followed. The amount

4736

GUTIORMSEN ET AL.

of anti-class 3-specific IgG in each serum was determined by comparing the A405 of the patient serum with a standard curve calculated from a separate ELISA that used antihuman IgG (Fab-specific) coated onto microtiter wells and known concentrations of an IgG standard (Sigma) (32). Appropriate blocking experiments to demonstrate the specificity of the reactions (i.e., preincubation of serum samples with soluble class 3 protein blocked the ELISA) were performed (data not shown). An IgG standard ELISA was included in each experiment. Results are reported as micrograms of specific anti-class 3 IgG antibodies per milliliter. IgG subclass titration. Serum anti-class 3 antibodies of the different IgG subclasses were determined by the standard protocol (antigen concentration of 2 ,ug/ml) but with mouse anti-human IgGl, IgG2, IgG3, and IgG4 as the secondary antibodies and alkaline phosphatase-conjugated goat antimouse IgG at a dilution of 1:5,000 as the conjugate. The anti-human IgG subclass antibodies were used at concentrations that gave identical absorption at 405 nm when they were used as antigens and goat anti-mouse IgG was used as the conjugate. Test sera, secondary antibodies, and conjugates were all incubated on the microtiter plates for 2 h at room temperature. The ELISA values are reported as the dilutions that gave an A405 of 1.0 after 1 h of incubation. ELISA absorptions. For ELISA absorptions, the method described by Wetzler et al. (32) was used. Briefly, N. meningitidis 44/76 (2) was grown to the logarithmic growth phase in proteose peptone broth after overnight growth on gonococcal typing agar (5, 27), adjusted to an OD at 600 nm equal to 1.0, and diluted twofold across a V-bottom microtiter plate. Patient sera were added to the plates at a dilution that would elicit an ELISA OD that was half the maximum value (i.e., an OD at 405 nm equal to 1.0). After 2 h, the organisms were centrifuged, and 0.1-ml aliquots were transferred to class 3-sensitized ELISA plates. The remainder of the ELISA was performed by the standard protocol. Controls. Serum samples from one patient with a pneumococcal meningitis and one patient with varicella zoster meningoencephalitis were used as control sera.

RESULTS Serum IgG response. The amount of IgG antibodies recognizing the class 3 protein (type 15) in patient sera at day 1, day 7, day 14, and week 6 is shown in Fig. 1. Most patients had low values of anti-class 3 IgG antibodies on admission to the hospital. An increase in anti-class 3 IgG antibody levels was demonstrated in all but three patients (22 of 25, 88%). The anti-class 3 IgG antibodies peaked during the second week of disease, were still markedly elevated at day 14 and week 6, and returned to near baseline in sera collected 6 to 12 months after the onset of the disease. The three patients that showed no increase in anti-class 3 IgG antibodies had not been immunized prior to the disease (Fig. 2). One of them had a high level of specific anti-class 3 IgG on admission to the hospital (22 ,ug/ml) and was infected with a homologous class 3 strain, whereas the other two showed a low anti-class 3 IgG level on admission to the hospital and were infected with a class 2 strain (type 2b) and a nontypeable strain, respectively. The highest anti-class 3 IgG peak levels (119.2 to 635.0 ,ug/ml) were seen in patients immunized with the meningococcal vesicle vaccine prior to infection with class 3 (type 15) strains. Intermediate anti-class 3 IgG peak levels (18.1 to 52.1 ,ug/ml) were seen in patients immunized prior to infection with class 2 or class 3 strains and in nonimmunized patients infected with class 2 or class

INFECr. IMMUN.

a

I

IUV -=

lllI

1111111

,inn

1ml

I t-|

§

' ,,~~A

1111'

IlTffF--AF

-

,,4:

'.x 5 1,41 1'.. 4 -_, ,.,-

Day 7

I~~~~~~ 0, 1 01.

0,01

10

1

100

b 100 0,1~~~~~~~~~~

Day 14

0,01

0-01

C

1

0 1

10

100

1000

xl1

100 10

Week 6

l_l lm

1

mm 0,01

Iu lil11

.H

111

l111111

.:.:

-

O,tDl

0,1

1

10

100

Day 1 FIG. 1. Amount of specific serum anti-class 3 IgG antibodies (in micrograms per milliliter) in 25 patients with systemic meningococcal disease. All panels show values obtained in samples from day 1 along thex axis. (a) Values obtained at day 1 versus day 7; (b) values obtained at day 1 versus day 14; (c) values obtained at day 1 versus week 6. The bold diagonal line represents no increase (x1) in anti-class 3 IgG levels. The thin diagonal line represents a fivefold increase in anti-class 3 IgG antibodies.

3 strains. The lowest anti-class 3 IgG peak values (1 to 16.2 ,ug/ml) were seen in nonimmunized patients, five of whom were infected with strains of homologous class 3 and seven of whom were infected with class 2 or nontypeable strains. Anti-class 3 IgG antibody profiles during the course of meningococcal disease are shown for some of the patients in Fig. 3. Serum IgM response. The amount of IgM antibodies recognizing the class 3 protein (type 15) in patient sera at day 1, day 7, day 14, and week 6 is shown in Fig. 4. Most patients (19 of 25, 76%) had low anti-class 3 IgM titers on admission to the hospital. A significant increase in anti-class 3 IgM

IMMUNE RESPONSE TO MENINGOCOCCAL CLASS 3 PROTEIN

VOL. 61, 1993

a

1000-

x5

100-

x1

4737

10Day 7

111111

.Z

1 0,1

0,01 0,01

b

0,1

100

10

1 Day 1 C

1000-

1000-

100-

100

10-

10

Day 7

Day 7 1

I111111

01>rn 0,01 - z

0,01

0,1

1

10

1

Oil

I1 11111 100

0,01 L, 0,01

I11111

1 1111

.,-

..!!

0,1

1

Day 1

10

100

Day 1

e

d 1000-

1000-

IJi1 LJ. _LIII

.

100-

In-

I1I111111

&III III -1111

. = = = = t .-

= = =

in

Day 7

Day 7

1

1 0,1 1 -.41

0,01 0,01

0,01

01

1

10

100

Day 1

V;1-

0,01

I111111 01

1 1

11111111 1

11

11111

11111111

1111111

11111

10

100

1

Day FIG. 2. Amount of specific serum anti-class 3 (type 15) IgG antibodies (in micrograms per milliliter) in 25 patients with systemic meningococcal disease related to the major porin (class 2 or 3) of the infecting strain and the meningococcal outer membrane vesicle vaccine status of the patients prior to disease. All panels show values obtained in samples from day 1 versus day 7. (a) All patients; (b) patients infected with class 3 strains immunized prior to disease; (c) patients infected with class 2 strains immunized prior to disease; (d) nonimmunized patients infected with class 3 strains; (e) nonimmunized patients infected with class 2 or nontypeable strains. The bold diagonal line represents no increase (xl) in anti-class 3 IgG levels. The thin diagonal line represents a fivefold increase in anti-class 3 IgG antibodies.

antibodies was demonstrated in all but four patients (21 of 25, 84%). The anti-class 3 IgM antibodies peaked during the second week of disease and were increased at week 6 for 17 of the patients (17 of 25, 68%). Three of the patients had elevated IgM titers (1/50 to 1/350) 6 months after the disease, and one of these patients was a carrier of a meningococcal

strain (NG:NT:P1.14:L3,7,9) at that time. Anti-class 3 IgM titers did not increase in four patients: two of these had high titers on admission to the hospital while the other two never showed measurable titers. Serum IgA response. The amount of IgA antibodies recognizing the class 3 protein (type 15) in patient sera at day 1 and

4738

GUTTORMSEN ET AL. a

INFECr. IMMUN.

500 400 300

pg/ml 200-

1000-

Dl D2 D3 D4 D5 D6 D7 D9 D10 D13 D16 W6 M6 b

100-

I

PATIENT NO. 22

75pg/ml 50-

_ 0-...... Dl c

D2 D3 D4 D5 D6 D9 D16 D19 W4 W6 M8

100 PATIENT NO. 24

75pg/mi 5025-

D

Dl d

25

D

D2 D3 D4 D6 D7 D8 D9 D13 D16 D23 W6 W10 M5

PATIENT NO. 2

\

20 15 pg/mi

10 5 0 Dl

D2 D3 D4 D5 D6 D7 D8 D9 D10 Dll D12 D13 D15 W6 M6 M12

FIG. 3. Anti-class 3 IgG antibody profiles during the course of meningococcal disease. Abbreviations: D, day; W, week; M, month. (a) Patient 9, immunized with meningococcal outer membrane vesicle vaccine status prior to infection with a homologous class 3 strain; (b) patient 22, immunized prior to infection with a class 2 strain; (c) patient 24, nonimmunized patient infected with a class 2 strain; (d) patient 2, nonimmunized patient infected with a homologous class 3 strain.

day 7 are shown in Fig. 5. Most patients (20 of 25, 80%) had low anti-class 3 IgA titers on admission to the hospital. An increase in anti-class 3 IgA antibodies was demonstrated in all but six patients (19 of 25, 76%). The anti-class 3 IgA antibodies peaked during the second week of disease, and only five of the patients (5/25, 20%) had increased IgA titers at week 6 (data not shown). Anti-class 3 IgA titers did not increase in sera from 6 of 25 patients (76%), 1 patient had a high IgA titer on admission to the hospital, while the others did not show measurable titers. Serum IgG subclass response. An increase in IgGl response was seen in 19 of the 22 patients that showed an

increase in anti-class 3 IgG antibodies (86%), whereas an IgG2 response was demonstrated in 8 patients (8 of 22, 36%), and an IgG3 response was seen in 15 patients (15 of 22, 68%). None of the patients showed an increase in anti-class 3 IgG4 antibodies. An increase in three IgG subclasses was demonstrated in 5 of the patients (5 of 22, 22.5%), 11 patients (11 of 22, 50%) showed an increase in two IgG subclasses, whereas 5 of the patients (5 of 22, 22.5%) showed an IgG response restricted to one subclass. For the 16 patients demonstrating an increase in more than one IgG subclass, 12 patients (12 of 16, 75%) had highest titers for IgGl and 2 patients (2 of 16, 12.5%) had highest titers for IgG2 and IgG3, respectively.

IMMUNE RESPONSE TO MENINGOCOCCAL CLASS 3 PROTEIN

VOL. 61, 1993

a

10000

x5 xl

10000-

x5

1000

4739

-_

_I1111

'=__,.

,,11

Day 7

100

Day 7

TI

1n

1003_ 0_

-

.4

1 000

100

10

1

_

_

10

1

__

-

100

11

.:

1000

10000

Day 1

b

l

I

111 -1-1-1 ,- -

Day 14

- -

x5

Sx

1

-111

FIG. 5. Specific serum anti-class 3 IgA antibody titers in 25 patients with systemic meningococcal disease given as reciprocal dilutions of sera at an OD equal to 1; maximum OD less than 1 is set as 1. The panel shows values obtained in samples from day 1 versus day 7. The bold diagonal line represents no increase (xl) in anti-class 3 IgG levels. The thin diagonal line represents a fivefold

increase in anti-class 3 IgG antibodies.

recognized surface-exposed epitopes 100 C

10000

Week 6

1000 x5

1000

x

of the

protein.

All

tested contained a significant percentage of antibodies that bound to exposed epitopes on the meningo-

patient

1

sera

cocci (Fig. 6). Controls. The control patients, one patient with pneumococcal meningitis and one with varicella zoster meningoencephalitis, showed no increase in anti-class 3 antibodies of any isotype or subclass tested (data not shown).

DISCUSSION

loo

In the present study, we have been able to quantitate specific anti-class 3 (type 15) outer membrane protein antibodies of IgG, IgM, and IgA isotypes in patient sera during

10

the course of

1

10

100

1000

Day 1 FIG. 4. Specific serum anti-class 3 IgM antibody titers in 25 patients with systemic meningococcal diseasige given as reciprocal dilutions of sera at an OD equal to 1; maximunn OD less than 1 is set as 1. All panels show values obtained in samlples fromday 1 along the x axis. (a) Values obtained in samples fror (b) values obtained in samples from day 1 ver sus day obtained in samples from day 1 versus week 6. The bod diagonal line represents no increase (xl) in anti-class 2 IgG levels. The thin diagonal line represents a fivefold increase in anti-class 3 IgG antibodies.

14ers(c)s vdalyues

Two of the five patients with an increase in anti-class 3 IgG antibodies restricted to one IgG subclas;s showed an IgG2 response, whereas the other three patienits showed an IgGl response. The anti-class 3 IgGl and IgG33 antibodies peaked during the second week of disease, as diid total anti-class 3 IgG antibodies. The IgG2 antibodies sh owed peak titers 6 weeks after the onset of the disease for siix of the patients (6 of 9, 67%) and earlier for three patients I(3 of 9, 33%). Inhibition ELISA. Whole-organism absorption studies were done for some of the immune serum samples to determine the percentage of anti-class 3 antibodies that

meningococcal

disease. We used

purified class

3 protein as the sensitizing antigen in the ELISAs. The specificity of the antibody response towards class 3 was certain because the protein was purified from a variant of the meningococcal strain 44/76 lacking both the class 1 and class 4 protein in its outer membrane, and gel electrophoresis and immunoblotting showed that the purified class 3 protein was not

contaminated with other

outer

membrane

components.

The sera, collected from the patients daily during their hospital stay and at subsequent intervals after hospitalization, allowed us to follow the antibody response during the course of meningococcal disease. The sera were collected 1 to 4 years after the Norwegian meningococcal group B

vaccine protection trial, which provided us with a unique opportunity to compare the immune response in immunized versus nonimmunized patients with meningococcal disease. Previously, several investigators tried to determine the presence of anti-class 2 or 3 meningococcal outer membrane protein antibodies in patient and vaccinee sera. Immunoblotting studies have shown that meningococcal disease and vaccination with meningococcal group B polysaccharideprotein complex vaccines elicit an antibody response against several major outer membrane proteins (19, 22, 30). Another study using human sera as inhibitors of monoclonal antibody binding to meningococcal outer membrane proteins in a radioimmunoassay (HIMSPRIA) demonstrated seroconversions of 91, 57, and 44% to the P1.2, 2a, and P5.1 epitopes, respectively, after immunization with a meningococcal

4740

GUTTORMSEN ET AL.

IN-FECT. IMMUN.

a (a

0 ._

:0 0

Co

0 0

-e

C 0.)

0

L.0

0)

a1

10

100

Reciprocal Dilution of Organisms b Co

~0 .0

6

-0

c

CU

.0 .0 o

.0

a0 0~

1

10

100

Reciprocal Dilution of Organisms FIG. 6. Percentage of anti-class 3 IgG antibodies left in patient sera after absorption with meningococcal strain 44/76 (class 3, serotype 15). (a) Patients 3 (A), 6 (A), and 10 (0) infected with homologous class 3 (serotype 15) strains; (b) patient 14 (A) infected with a heterologous class 3 strain (serotype 4) and patients 21 (A) and 22 (0) infected with class 2 (serotype 2a) strains.

B polysaccharide-protein complex vaccine (21). These studies are difficult to interpret, partly because it has been shown that the results obtained by immunoblotting vary depending on the outer membrane preparation and transfer buffer (8), and the proteins on Western blots are seldom in native configuration. Furthermore, HIMSPRIA will only detect human polyclonal antibodies reacting with the epitopes defined by the specific monoclonal antibodies. In this study, an increase in anti-class 3 IgG antibodies was demonstrated in all but three patients (22 of 25) regardless of the serotype of the infecting meningococcal strains. The three nonresponders had not been immunized prior to disease. However, one of these patients who showed no increase in anti-class 3 IgG antibodies during the course of his disease had high levels of anti-class 3 IgG antibodies on admission to the hospital (22 ,ug/ml). This patient (patient 4) was the only patient in our study with a benign meningococcemia upon arrival to the hospital, and this might indicate group

that the high level of anti-class 3 IgG antibodies offered some kind of protection. The highest anti-class 3 IgG levels were seen in patients immunized with an outer membrane vesicle vaccine containing the homologous class 3 protein and then infected with a meningococcal strain containing the same class 3 protein (serotype 15), suggesting an anamnestic response. However, some of the nonimmunized patients who acquired meningococcal disease with class 2 or nontypeable strains had higher anti-class 3 IgG antibody levels than nonimmunized patients who were infected with strains containing the homologous class 3 protein. This finding indicates that the anti-class 3 IgG antibodies induced during the course of meningococcal disease react at least partly with epitopes different from those defined by the serotyping monoclonal antibodies. This is in accordance with the findings of others (15, 21, 24) and an earlier study by ourselves (11). Furthermore, the majority of the specific anti-class 3 IgG antibodies induced in patients infected with class 2 or nontypeable strains were shown to recognize surface-exposed epitopes of the class 3 protein on the whole bacteria (Fig. 6). The last two findings indicate that these crossreacting antibodies recognize surface-exposed epitopes common to class 3 and other outer membrane proteins (i.e., class 1 and class 2). There was considerable heterogeneity in the amounts of anti-class 3 antibodies and IgG subclass response and some heterogeneity in the immunoglobulin isotype elicited during the course of meningococcal disease. All patients who were immunized prior to disease showed an increase in anti-class 3 antibodies of IgA, IgM, and IgG isotypes. However, of the nonimmunized patients, two failed to induce any increase in anti-class 3 antibodies during the course of the infection, whereas the rest showed an increase in at least one isotype. The extensive heterogeneity in the immune response demonstrated in this study concurs with the findings of others (21, 22, 30, 31). The increase in specific anti-class 3 IgG antibodies during the course of meningococcal disease was primarily of IgGl and IgG3 subclasses. Parallel results with other bacterial proteins have been shown (13). The four human IgG subclasses have different abilities to trigger effector function activities, i.e., complement activation and binding to Fc receptors (7). IgGl and IgG3 are more effective than IgG2 and IgG4, and the anti-porin IgG antibodies induced in these patients thus belong to the subclasses with the greatest ability to stimulate serum bactericidal and/or opsonic activity. Other investigators have shown similar results by immunoblotting or ELISA techniques by using outer membrane vesicles as the antigen preparations (26, 31). The anti-class 3 IgGl and IgG3 antibodies peaked on the same day during the second week of disease as did the total IgG in the individual patients studied, whereas the majority of the patients showing an IgG2 response demonstrated peak levels 6 weeks after hospitalization. The reason for this finding is not known. A majority of the anti-class 3 IgG antibodies elicited during meningococcal disease recognized epitopes exposed on the bacteria as demonstrated by whole-cell-absorption ELISAs (Fig. 6). This might indicate, along with the IgG subclass profile, that the anti-class 3 IgG antibodies in these patients have potential functional activity and could possibly be protective. Unfortunately, this study does not allow us to make any conclusions regarding overall protectivity. It can only be stated that for the patients developing disease after being immunized, effective protection was obviously not induced by the vaccine, even though high amounts of

VOL. 61, 1993

IMMUNE RESPONSE TO MENINGOCOCCAL CLASS 3 PROTEIN

anti-class 3 IgG were induced during the course of the disease. The levels of anti-class 3 IgM and IgG antibodies peaked at approximately the same time during the second week of disease regardless of whether the patients were immunized prior to disease. However, the amounts of anti-class 3 IgG antibodies were greater, and the anti-class 3 IgG antibodies tended to persist longer in the sera of the patients that were immunized prior to disease. These findings indicate that in our study the actual levels of IgG antibodies and their duration were a better indicator of primary versus secondary antibody response than the timing of the IgM and the IgG

peaks. In summary, we have been able to demonstrate an inspecific anti-class 3 (type 15) outer membrane protein antibodies in patient sera during the course of meningococcal disease. Specific anti-class 3 antibodies were present in the majority of the serum samples from our patients, but a considerable heterogeneity in the levels of the antibodies and in the IgG subclass response was seen. The specific anti-class 3 antibodies were reacting in part with epitopes not defined by the serotyping monoclonal antibodies. Furthermore, the anti-class 3 IgG antibodies were shown to bind to surface-exposed epitopes on the whole bacteria, and the IgG antibodies belonged to the subclasses with the greatest abilities to stimulate bactericidal and/or opsonic activity. Patients who were immunized with a class 3-containing meningococcal vesicle vaccine prior to their infection with homologous class 3 strains showed the highest IgG antibody levels, indicating a priming of their immune system, but they were not protected by the immunization from infection with meningococcal class 3 strains. crease in

ACKNOWLEDGMENTS This work was supported by Ninas Minnefond, Oslo, Norway, and grant AI-00680 from the National Institutes of Health. N. meningitidis H15 was kindly provided by Jan T. Poolman, Unit of Bacterial Vaccine Development and Pathogenesis Research, National Institute of Public Health and Environmental Protection, Bilthoven, The Netherlands. The serotyping and the immunotyping were done by Andrew J. Fox, The National Institute of Public Health, Birmingham, England. The complement activity (50% hemolytic complement) was measured by the Department of Immunology, Haukeland Hospital, Bergen, Norway. We would like to thank Claus Ola Solberg for support and discussions concerning this work and Milan S. Blake for helpful advice. We thank Steinar Sornes for excellent technical assistance. REFERENCES 1. Abdillahi, H., and J. T. Poolman. 1987. Whole-cell ELISA for typing Neisseria meningiidis with monoclonal antibodies. FEMS Microbiol. Lett. 48:367-371. 2. Bjune, G., E. A. H0iby, J. K. Gr0nnesby, 0. Amesen, J. H. Fredriksen, A. Haistensen, E. Holten, A.-K. Lindbak, H. N0kleby, E. Rosenqvist, L. K. Solberg, 0. Closs, J. Eng, L. 0. Fr0holm, A. Lystad, L. S. Bakketeig, and B. Hareide. 1991. Effect of outer membrane vesicle vaccine against group B meningococcal disease in Norway. Lancet 338:1093-1096. 3. Blake, M. S., and E. C. Gotschlich. 1982. Purification and partial characterization of the major outer membrane proteins of Neisseria gonorrhoeae. Infect. Immun. 36:277-283. 4. Blake, M. S., and E. C. Gotschlich. 1987. Functional and immunological properties of pathogenic neisserial surface proteins, p. 377-400. In M. Inouye (ed.), Bacterial outer membranes as model systems. John Wiley & Sons, Chichester,

England.

5. Blake, M. S., C. M. MacDonald, and K. P. Klugman. 1989. Colony morphology of piliated Neisseria meningitidis. J. Exp. Med. 170:1727-1736.

4741

6. Brodeur, B. R., Y. Larose, P. Tsang, J. Hamel, F. Ashton, and A. Ryan. 1985. Protection against infection with Neisseria meningitidis group B serotype 2b by passive immunization with serotype-specific monoclonal antibody. Infect. Immun. 50:510516. 7. Burton, D. R., L. Gregory, and R. Jefferis. 1986. Aspects of the molecular structure of IgG subclasses. Monogr. Allergy. 19:735. 8. Davies, R. L., R. A. Wall, and S. P. Borriello. 1990. Comparison of methods for the analysis of outer membrane antigens of Neisseria meningitidis by Western blotting. J. Immunol. Methods 134:215-225. 9. Engvali, E., and P. Perlmann. 1972. Enzyme-linked immunosorbent assay, ELISA III. Quantitation of specific antibodies by enzyme-labeled anti-immunoglobulin in antigen-coated tubes. J. Immunol. 109:129-135. 10. Fr0holm, 0. L., E. Holten, J. T. Poolman, and W. D. Zollinger. 1985. Typing of Norwegian meningococcal isolates and possible implications for serogroup B vaccination, p. 541-545. In G. K. Schoolnik (ed.), The pathogenic neisseriae. American Society for Microbiology, Washington, D.C. 11. Guttormsen, H.-K., R Bjerknes, A. Naess, V. Lehmann, A. Halstensen, S. S0mes, and C. 0. Solberg. 1992. Cross-reacting serum opsonins in patients with meningococcal disease. Infect. Immun. 60:2777-2783. 12. Guttormsen, H.-K, L. M. Wetzler, and A. Naess. 1992. Anticlass 3 IgG antibodies in meningococcal disease, abstr. III-27, p. 231. Eighth Int. Neisseria Conf., Cuernavaca, Mexico. 13. Hammarstr0m, L., and C. I. E. Smith. 1986. IgG subclasses in bacterial infections. Monogr. Allergy 19:122-133. 14. Harlow, E., and D. Lane. 1988. Antibodies: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 15. Harthug, S., E. Rosenqvist, E. A. H0iby, T. W. Gedde-Dahl, and L. 0. Fr0holm. 1986. Antibody response in group B meningococcal disease determined by enzyme-linked immunosorbent assay with serotype 15 outer membrane antigen. J. Clin. Microbiol. 24:947-953. 16. Jones, D. M., R. Borrow, A. J. Fox, S. Gray, K. A. Cartwright, and J. T. Poolman. 1992. The lipooligosaccharide immunotype as a virulence determinant in Neisseria meningitidis. Microb. Pathog. 13:219-224. 17. Klugman, K., E. C. Gotschlich, and M. S. Blake. 1989. Sequence of the structural gene (rpmM) for the class 4 outer membrane protein of Neisseria meningitidis, homology of the protein to gonococcal protein III and Escherichia coli ompA, and construction of meningococcal strains that lack class 4 protein. Infect. Immun. 57:2066-2071. 18. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227:680-685. 19. Maeland, J. A., and E. Wedege. 1989. Serum antibodies to cross-reactive Neisseria outer membrane antigens in healthy persons and patients with meningococcal disease. APMIS (C) 97:774-780. 20. Mandrell, R. E., and W. D. Zollinger. 1977. Lipopolysaccharide serotyping of Neisseria meningitidis by hemagglutination inhibition. Infect. Immun. 16:471-475. 21. Mandrell, R. E., and W. D. Zollinger. 1989. Human immune response to meningococcal outer membrane protein epitopes after natural infection or vaccination. Infect. Immun. 57:15901598. 22. Poolman, J. T., C. T. P. Hopman, and H. C. Zanen. 1983. Immunogenicity of meningococcal antigens as detected in patient sera. Infect. Immun. 40:398-406. 23. Poolman, J. T., F. B. Wientjes, C. T. P. Hopman, and H. C. Zanen. 1985. Influence of the length of lipopolysaccharide molecules on the surface exposure of class 1 and class 2 outer membrane proteins of Neisseria meningitidis 2996 (B:2b:P1.2), p. 562-570. In G. K. Schoolnik (ed.), The pathogenic neisseria. American Society for Microbiology, Washington, D.C. 24. Rosenqvist, E., S. Harthug, L. 0. Fr0holm, E. A. H0iby, K. B0vre, and W. D. ZoLinger. 1988. Antibody response to serogroup B meningococcal outer membrane antigens after vaccina-

4742

GUTTORMSEN ET AL.

tion and infection. J. Clin. Microbiol. 26:1543-1548. 25. Saukkonen, K., M. Leinonen, H. Abdiliahi, and J. T. Poolman. 1989. Comparative evaluation of potential components for group B meDingococcal vaccine by passive protection in the infant model and in vitro bactericidal assay. Vaccine 7:325328. 26. Sjursen, H., E. Wedege,,E. Rosenqvist, A. Naess, A. Haistensen, R. Matre, and C. 0. Solberg. 1990. IgG subclass antibodies to serogroup B meningococcal outer membrane antigens following infection and vaccination. APMIS 98:1061-1069. 27. Swanson, J. 1978. Studies on gonococcus infections. XII. Colony color and opacity variants of gonococci. Infect. Immun. 19:320-331. 28. Tsai, C.-M. 1986. The analysis of lipopolysaccharide (endotoxin) in meningococcal polysaccharide vaccines by silver staining following SDS-polyacrylamide gel electrophoresis. J. Biol. Stand. 14:25-33. 29. Tommassen, J., P. Vermej, M. Struyve, R. Benz, and J. T. Poolman. 1990. Isolation of Neisseria meningitidis mutants deficient in class 1 (PorA) and class 3 (PorB) outer membrane

INFECr. IMMUN. proteins. Infect. Immun. 58:1355-1359. 30. Wedege, E., and 0. L. Froholm. 1986. Human antibody response to a group B serotype 2a meningococcal vaccine determined by immunoblotting. Infect. Immun. 51:571-578. 31. Wedege, E., and T. E. Michaelsen. 1987. Human immunoglobulin G subclass immune response to outer membrane antigens in meningococcal group B vaccine. J. Clin. Microbiol. 25:13491353. 32. Wetzler, L. M., M. S. Blake, K. Barry, and E. C. Gotschlich. 1992. Gonococcal porin vaccine evaluation: comparison of por proteosomes, liposomes, and blebs isolated from rnnp deletion mutants. J. Infect. Dis. 166:551-555. 33. Wetzler, L. M., M. S. Blake, and E. C. Gotschlich. 1988. Characterization and specificity of antibodies to protein I of Neisseria gonorrhoeae produced by injection with various protein I-adjuvant preparations. J. Exp. Med. 168:1883-1897. 34. Wyle, F. A., M. S. Artenstein, B. L. Brandt, E. C. Tramont, D. L. Kasper, P. L. Alteri, S. L. Berman, and J. P. Lowenthal. 1972. Immunogenic response of man to group B meningococcal polysaccharide veccines. J. Infect. Dis. 126:514-522.