A 200 kDa protein is associated with haemagglutinating isolates of

Moraxella (Branhamella) catarrhalis

Margaret Fitzgeralda, Riona Mulcahy

b, Susan Murphy

b, Conor Keane

c,

Davis Coakleyb, Thomas Scott

a;

*

aDepartment of Biological Sciences, Dublin Institute of Technology, Kevin St., Dublin 8, Ireland

bMercer's Institute for Research on Ageing, St. James's Hospital, Dublin 8, Ireland

cDepartment of Clinical Microbiology, St. James's Hospital, Dublin 8, Ireland

Received 2 May 1997; accepted 5 May 1997

Abstract

Moraxella catarrhalis adheres to human erythrocytes by means of a proteinaceous, trypsin sensitive, heat modifiable

haemagglutinin. A 200 kDa protein was found to be associated with haemagglutinating isolates of M. catarrhalis. This protein

was present on all haemagglutinating isolates (n=17), but was absent on the non-haemagglutinating isolates (n=23) examined.

This protein demonstrated heat-modifiable properties in sodium dodecyl sulfate and was degraded by trypsin. Immunoblot

assays with polyclonal antiserum indicated that the 200 kDa protein was associated exclusively with haemagglutinating isolates

and antibodies to this protein did not recognise epitopes on non-haemagglutinating isolates. This protein, which appears to be

a surface expressed protein may be a haemagglutinin of M. catarrhalis.

Keywords: Moraxella (Branhamella) catarrhalis ; Haemagglutination; Surface protein; Heat modi¢ability; Immunoblotting

1. Introduction

In recent years Moraxella (Branhamella) catarrha-

lis has been increasingly recognised as an important

human pathogen [1,2]. Previously dismissed as a

commensal of the oro-nasopharyngeal tract, it is

now reported as the third most common isolate after

Haemophilus in£uenzae and Streptococcus pneumo-

niae as the causative agent in lower respiratory tract

infections such as bronchitis and pneumonia in the

elderly, especially in those with underlying pulmo-

nary disease [3,4]. M. catarrhalis has been sporadi-

cally implicated in invasive infections including sep-

ticaemia in both the immunocompetent [5] and the

immunocompromised patient [6].

The recent recognition of M. catarrhalis as an im-

portant upper and lower respiratory tract pathogen

has prompted researchers to investigate the possible

virulence mechanisms employed by the organism in

order to understand the pathogenesis of M. catarrha-

lis infection. Bacterial adherence is believed to be the

initial event in most infectious processes [7]. There

have been a number of studies on the adherence of

M. catarrhalis to human oropharyngeal cells and

human erythrocytes [8^11]. The adhesive processes

0928-8244 / 97 / $17.00 ß 1997 Federation of European Microbiological Societies. Published by Elsevier Science B.V.

PII S 0 9 2 8 - 8 2 4 4 ( 9 7 ) 0 0 0 4 1 - 2

FEMSIM 766 27-8-97

* Corresponding author. Tel. : +353 (1) 4024747;

Fax: +353 (1) 4024995.

Abbreviations: PBS A, Dulbecco A phosphate buffered saline;

NOG, N-octyl-L-D-glucopyranoside; TBS, Tris buffered saline

FEMS Immunology and Medical Microbiology 18 (1997) 209^216

involved remain unresolved. It has been established

that the haemagglutinin is a trypsin sensitive protein

[11,12] and is associated with tracheal cell adherence

[11]. Two phenotypes of the haemagglutinin have

been identi¢ed, phenotype I agglutinates human er-

ythrocytes only, while phenotype II agglutinates hu-

man and rabbit erythrocytes [12]. A recent study has

reported that a signi¢cantly higher percentage of M.

catarrhalis isolates recovered from infected elderly

patients, with respiratory tract infection, haemagglu-

tinated human erythrocytes when compared with the

haemagglutinating activity of isolates recovered from

colonised healthy elderly people, suggesting that the

haemagglutinin may play a role in the infectious

process in respiratory tract infection in the elderly

population [13].

In this present study 40 isolates of M. catarrhalis

were examined by SDS-PAGE and a 200 kDa pro-

tein, which was found to be associated with haemag-

glutinating isolates was investigated.

2. Materials and methods

2.1. Bacterial isolates and growth conditions

Forty isolates of M. catarrhalis were examined.

These isolates were from a collection acquired during

a study of the epidemiological aspects of M. catar-

rhalis infection and colonisation [13]. Isolates were

identi¢ed according to the criteria previously out-

lined [14] and were stored at 370³C using the Protect

Bead System (Technical Service Consultants). Prior

to all tests isolates were grown on Columbia blood

agar at 37³C for 18 h, unless otherwise stated.

2.2. Haemagglutination

The ability of the 40 isolates to agglutinate human

group O red blood cells and rabbit erythrocytes was

determined by a micotitre method [12].

2.3. SDS-PAGE of crude whole-cell protein

preparations

Whole cell preparations of 17 haemagglutinating

and the 23 non-haemagglutinating isolates of M. ca-

tarrhalis were analysed by SDS-PAGE.

Bacterial suspensions were centrifuged in a micro-

fuge at 10 000Ug for 5 min. The pellet was boiled in

200 Wl sample bu¡er containing 0.065 M Tris, 2% (w/

v) SDS, 5% (v/v) L-mercaptoethanol, 10% (w/v) glyc-

erol and 0.001% (w/v) bromophenol blue. The result-

ing protein extracts were resolved by SDS-PAGE on

10% polyacrylamide gels by the technique of

Laemmli [15]. The resolved proteins were visualised

by staining with Coomassie brilliant blue.

Isolate K48 showed variability on repeated hae-

magglutination testing. Separate colonies were sub-

cultured and three phenotypes were identi¢ed. These

were titled W, X and Y and gave titres of 4, 2 and 0

against human erythrocytes and 32, 16 and 1 against

rabbit erythrocytes. These phenotypes were tested by

SDS-PAGE, using standardised suspensions.

Bacterial suspensions of phenotype I and pheno-

type II haemagglutinating isolates (B4, K29, S109,

S407 and S580) were treated with an equal volume

of trypsin solution (2 mg ml

31) in PBS A at 37³C for

2 h. Following washing in PBS A the bacterial pellets

were prepared for SDS-PAGE analysis as described

above. Control preparations were treated identically,

using PBS A instead of trypsin solution.

2.4. Heat modi¢ability of the 200 kDa protein

The heat modi¢ability of the 200 kDa protein was

examined by incubating crude whole-cell protein ex-

tracts of isolate S407 in SDS-PAGE sample bu¡er at

various temperatures ranging from 20³C to 100³C

for 5 min, before resolving the preparations by

SDS-PAGE on 5% polyacrylamide gels.

2.5. Isolation of outer membranes

The outer membranes of two haemagglutinating

isolates (S407 and S580) and two non-haemaggluti-

nating isolates (K16 and S89) were obtained using

the broth culture supernatant technique previously

described [16], with some slight modi¢cations. Brain

Heart Infusion broth (100 ml) was inoculated with

several colonies of bacteria from a Columbia blood

agar plate and incubated at 37³C for 24 h, with

shaking. The cells were centrifuged at 10 000Ug for

15 min at 4³C. The supernatant fraction was col-

lected and centrifuged again at 10 000Ug for 15

min at 4³C. The resulting supernatant fraction was

FEMSIM 766 27-8-97

M. Fitzgerald et al. / FEMS Immunology and Medical Microbiology 18 (1997) 209^216210

then centrifuged at 100 000Ug for 90 min at 4³C.

The pellet was suspended in sample bu¡er and the

proteins present in these outer membrane vesicles

were resolved by SDS-PAGE using 10% polyacryl-

amide gels.

These blebs consist of outer membrane vesicles

that can be observed by electron microscopy; they

contain OMPs and lipooligosaccharide and are free

of cytoplasmic membrane contamination [17].

2.6. Preparation of anti-serum

Polyclonal antiserum against whole cells was ob-

tained by immunisation of a rabbit with a formalin

treated suspension of isolate S407. In addition, poly-

clonal antiserum was prepared against a protein ex-

tract of isolate S407, obtained by ion-exchange chro-

matography. Following elution of the proteins from

the ion-exchange column, fractions containing pro-

tein were analysed by SDS-PAGE. The fractions

containing the 200 kDa protein were pooled, dia-

lysed with several changes of sterile water at 4³C

for 48 h and used in the production of antisera. In

the preparation of each anti-serum a New Zealand

white rabbit was given three injections, subcutane-

ously, of the whole cell preparation or the protein

extract (800 Wg protein per dose) in Freund's com-

plete adjuvant on day 0, and in Freund's incomplete

adjuvant on days 14 and 35.

Following immunization, antisera were obtained

on day 49 and the immunoglobulin fraction of the

antiserum was separated by the method of Harboe

and Ingild [18]. Antisera were adsorbed by using (i) a

cocktail of non-haemagglutinating isolates (21, 25,

K2, K16, S22, S68, S75, S76, S89) and (ii) a cocktail

of haemagglutinating isolates (B4, K38, K44, K51,

S109, S274, S407, S540, S580).

2.7. Immunoblot analysis

Both adsorbed and unadsorbed antisera were

studied using immunoblot analysis. Western immu-

noblots were performed, with polyclonal anti-serum

to whole cell preparation of isolate S407, by electro-

phoretic transfer of crude protein extracts of hae-

magglutinating isolates (B4 and S407) and a non-

haemagglutinating isolate (isolate 21) of M. catarrha-

lis resolved by SDS-PAGE to nitrocellulose, employ-

ing the technique described by Bollag and Edelstein

[19]. The membranes were blocked in 2% (w/v) skim

milk (Difco) prepared in Tris bu¡ered saline (TBS)

for 1 h. Following washing, the membrane was

probed with the polyclonal antiserum as the primary

antibody. The membranes were washed in TBS, in-

cubated in 1:1000 dilution of swine anti-rabbit,

horseradish peroxidase-conjugated antibody for 2 h,

washed again and developed in a solution of 4-

chloro-1-naphthol for 10 min.

Antiserum to the protein extract of isolate S407

which had been adsorbed with the cocktail of non-

haemagglutinating isolates (described above) was ex-

amined in more detail using immunoblot analysis by

probing against a selection of haemagglutinating iso-

lates (B4, K29, K38, K48, S64, S109, S407, S540 and

S580) and non-haemagglutinating isolates (21, K1,

K3, K5, K10, S68 and S89) of M. catarrhalis.

3. Results

3.1. Haemagglutination and SDS-PAGE analysis

Seventeen of the 40 isolates examined agglutinated

human erythrocytes, while the remaining 23 isolates

were non-haemagglutinating. Of the 17 haemaggluti-

nating isolates, 10 isolates haemagglutinated human

erythrocytes (phenotype I), while 7 isolates haemag-

glutinated both human and rabbit erythrocytes (phe-

notype II). The protein pro¢les obtained following

the resolution of crude protein extracts of the 17

haemagglutinating and the 23 non-haemagglutinat-

ing isolates of M. catarrhalis by SDS-PAGE demon-

strated the presence of a 200 kDa protein in all prep-

arations of haemagglutinating isolates which was

absent in all preparations of non-haemagglutinating

isolates examined (Fig. 1). The three phenotypes of

K48 showed di¡erences in the amount of the 200

kDa protein that was present. The strength of the

200 kDa band was related to the haemagglutinating

activity of the phenotypes (Fig. 2). The trypsin-

treated haemagglutinating strains showed no 200

kDa protein. The apparent molecular mass (Mr) of

the 200 kDa protein varied slightly between some

isolates from 185^205 kDa. However, the Mr of

the `200 kDa' of a particular isolate was repeatedly

the same when di¡erent preparations of the same

FEMSIM 766 27-8-97

M. Fitzgerald et al. / FEMS Immunology and Medical Microbiology 18 (1997) 209^216 211

isolate were compared. In addition, no apparent dif-

ference in protein pro¢le was noted when phenotype

I isolates were compared with phenotype II isolates,

thus inter-strain variability in Mr was unrelated to

haemagglutination phenotype.

The 200 kDa protein demonstrated heat-modi¢-

ability, it migrated as a protein of higher Mr when

solubilised at temperatures less than 80³C. Determi-

nation of the Mr of this protein proved di¤cult, but

it appeared to be greater than 400 kDa.

OMP preparations were obtained using the broth

culture supernatant technique. The puri¢ed outer

membranes of two haemagglutinating (S407 and

S580) and of two non-haemagglutinating (K16 and

FEMSIM 766 27-8-97

Fig. 2. SDS-PAGE analysis of phenotypes W, X and Y of isolate K48. Lane 3, phenotype W (titre against rabbit erythrocytes 32 and hu-

man erythrocytes 4); lane 2, phenotype X (titre against rabbit erythrocytes 16 and human erythrocytes 2) and lane 4, phenotype Y (titre

against rabbit erythrocytes 1 and human erythrocytes 0). The 200 kDa protein is marked `s '. Proteins were stained with Coomassie bril-

liant blue on a 10% polyacrylamide gel. Mr markers are shown in lane 1.

Fig. 1. SDS-PAGE analysis of crude whole-cell protein extracts of M. catarrhalis. Lanes 2^6, haemagglutinating isolates and lanes 7^11,

non-haemagglutinating isolates. The presence of a 200 kDa protein in the extracts of haemagglutinating isolates is indicated by `s '. The

10% polyacrylamide gel was stained with Coomassie brilliant blue. Mr markers are shown in lane 1.

M. Fitzgerald et al. / FEMS Immunology and Medical Microbiology 18 (1997) 209^216212

S89) M. catarrhalis isolates were resolved by SDS-

PAGE. The protein band patterns were similar to

those obtained by Bartos and Murphy [16] and Mur-

phy and Loeb [17] with the eight major outer mem-

brane proteins (A^H) being present in all prepara-

tions. However, in the haemagglutinating isolates

FEMSIM 766 27-8-97

Fig. 4. Western blot analysis of crude whole-cell protein preparations of M. catarrhalis. Haemagglutinating isolate S407 (lane 1) and iso-

late B4 (lane 2) and non-haemagglutinating isolate 21 (lane 3) probed with unadsorbed polyclonal antiserum (left), polyclonal antiserum

adsorbed with a cocktail of non-haemagglutinating isolates (middle) and polyclonal antiserum adsorbed with a cocktail of haemagglutinat-

ing isolates (right). The 200 kDa protein is indicated by `s '.

Fig. 3. SDS-PAGE analysis of outer membrane preparation of isolate S407 (bottom panel) and isolate S 580 (top panel). In (bottom)

upper lane, outer membrane bleb preparation; middle lane, whole cell preparation. In (top) middle lane, outer membrane bleb prepara-

tion; bottom lane, whole cell preparation. The 200 kDa protein is marked `s '. Proteins are stained with 1% Coomassie blue on a 10%

polyacrylamide gel. Mr markers are shown in the bottom lane in the bottom panel and in the top lane in the top panel.

M. Fitzgerald et al. / FEMS Immunology and Medical Microbiology 18 (1997) 209^216 213

examined (S407 and S580), the 200 kDa protein ob-

served in the crude protein whole cell extracts was

also evident, albeit weakly in the outer membrane

preparations. (Fig. 3). This band was not found in

the outer membrane preparations of the non-hae-

magglutinating isolates (K16 and S89).

3.2. Western immunoblotting

The polyclonal antiserum to a whole cell prepara-

tion of M. catarrhalis (isolate S407) was analysed in

immunoblot assays (Fig. 4a). When this antiserum

was adsorbed with a cocktail of non-haemagglutinat-

ing isolates and examined by immunoblotting analy-

sis, the adsorbed antiserum recognised epitopes on

the 200 kDa protein on the two haemagglutinating

isolates studied (S407 and B4), while antibodies rec-

ognising epitopes on other proteins had been signi¢-

cantly adsorbed out (Fig. 4b). Adsorption of the

antiserum with a cocktail of haemagglutinating iso-

lates resulted in the signi¢cant removal of antibodies

recognising the 200 kDa protein (Fig. 4c).

To determine if antiserum adsorbed with a cock-

tail of non-haemagglutinating isolates, which con-

tained antibodies predominantly to the 200 kDa pro-

tein, demonstrated reactivity with a larger selection

of M. catarrhalis isolates, a total of 16 isolates were

examined by immunoblot analysis, using antiserum

to the protein extract of M. catarrhalis (isolate 407).

This adsorbed antiserum recognised epitopes on the

200 kDa protein in all nine haemagglutinating iso-

lates examined (B4, K29, K38, K48, S64, S109, S407,

S540 and S580), but failed to detect a 200 kDa pro-

tein in all seven non-haemagglutinating isolates (21,

K1, K3, K5, K10, S68 and S89) of M. catarrhalis

studied (Fig. 5).

4. Discussion

M. catarrhalis is now a well recognised cause of

upper and lower respiratory tract infections in chil-

dren and in the elderly, respectively. Due to the in-

creased awareness of the pathogenic role played by

M. catarrhalis in infection, much work has evolved

around establishing the virulence factors employed

by the organism. Adherence of bacteria to host cells

is generally considered the initial step in the patho-

FEMSIM 766 27-8-97

Fig. 5. Western blot analysis of crude whole-cell protein preparations of haemagglutinating and non-haemagglutinating isolates of M. ca-

tarrhalis probed with polyclonal antiserum adsorbed with non-haemagglutinating isolates of M. catarrhalis. The haemagglutinating isolates

were S407 (lane 1), B4 (lane 2), S109 (lane 3), S580 (lane 4), K38 (lane 5), S64 (lane 6); the non-haemagglutinating isolates were S68

(lane 7), K1 (lane 8), K3 (lane 9), K5 (lane 10) and K10 (lane 11). The 200 kDa protein is indicated by `s '.

M. Fitzgerald et al. / FEMS Immunology and Medical Microbiology 18 (1997) 209^216214

genesis of many infections [7]. It has been proposed

that the adherence of M. catarrhalis to oropharyn-

geal cells and human erythrocytes may be mediated

by ¢mbriae [8]. While Ahmed et al. [9] report an

association between nasopharyngeal adherence of

strains and haemagglutination, this association is

tenuous as a number of strains, which were highly

¢mbriated, were non adherent to nasopharyngeal

cells and were non-haemagglutinating. Similarly, Ri-

kitomi et al. [10] have observed no signi¢cant di¡er-

ence in the adhesive capacity to oropharyngeal cells

or in haemagglutination between ¢mbriated and

non-¢mbriated strains of M. catarrhalis. In the latter

study, it was also noted that no correlation between

haemagglutination and oropharyngeal cell adherence

was evident. In an ongoing study no correlation be-

tween haemagglutination and adherence to HEp-2

cells by M. catarrhalis has also been observed (M.

Fitzgerald, unpublished observations). Thus it would

appear that haemagglutination and epithelial cell ad-

herence are mediated by separate adhesins and that

non-¢mbrial adhesins, may be involved in haemag-

glutination.

Recently, a correlation was found between the

haemagglutination titre of isolates of M. catarrhalis

and their ability to adhere to tracheal epithelial cells

[11]. This ¢nding may be of importance as the or-

ganism is commonly associated with tracheobronchi-

tis in the elderly [20]. Furthermore, the fact that 75%

of strains isolated from the sputum of the elderly

with respiratory tract infections were haemaggluti-

nating in comparison to 15% of the strains isolated

from healthy elderly individuals would suggest a role

for the haemagglutinin in the aetiology of M. catar-

rhalis infection in the elderly [13].

A recent study has established that M. catarrhalis

has at least two haemagglutination phenotypes [12].

Phenotype I isolates haemagglutinate human eryth-

rocytes, while phenotype II isolates haemagglutinate

human and rabbit erythrocytes. Haemagglutination

by both phenotypes was mediated by trypsin-sensi-

tive proteins that were heat-labile at 70³C. The ob-

servation that haemagglutination is mediated by pro-

teinaceous structures has also been reported by

others [11]. To date no study characterising these

haemagglutinins has been documented.

In this present study it was established that a 200

kDa protein is exclusively associated with haemag-

glutinating isolates of M. catarrhalis. There was

some variability of the Mr of this protein from one

isolate to another, but the Mr was consistent for

each isolate. This protein was found on both hae-

magglutinating phenotypes and there was no corre-

lation between Mr and phenotypes. In addition, it

was established that in the SDS-PAGE of the three

phenotypes of K48, the density of the 200 kDa band

correlated closely with the haemagglutinating titre of

the phenotype. SDS-PAGE analysis of the bleb prep-

arations of haemagglutinating isolates demonstrated

the presence of the 200 kDa protein, indicating that

it may be an outer membrane protein. However,

while it was identi¢ed in the outer membrane

vesicles, the relative amount was not greater than

the lysate, thus the possibility exists that this protein

may not be an outer membrane protein, but may be

a surface protein bound to the membrane. Evidence

that the 200 kDa protein was surface expressed was

provided by the fact that: (a) antibodies recognising

epitopes on this protein were removed following ad-

sorption of the antiserum with haemagglutinating

isolates, as demonstrated by immunoblot analysis

and (b) ongoing studies have shown that antibodies

to the 200 kDa protein are present in the serum of

children and adults when analysed by immunoblot-

ting, indicating that the protein is expressed in vivo

(R. Mulcahy, unpublished observations).

Electron microscopy studies have shown that hae-

magglutinating strains have an outer ¢brillar coat

which is removed by trypsin treatment and adher-

ence to red blood cells appears to be mediated by

this outer ¢brillar coat [12]. Haemagglutination ap-

pears to be mediated by a trypsin sensitive, heat

sensitive, protein haemagglutinin. The 200 kDa pro-

tein is heat sensitive and trypsin sensitive and be-

cause of the association between the 200 kDa protein

and haemagglutinating strains it is possible that this

protein may be a haemagglutinin. Further work is

currently underway to investigate this.

Acknowledgments

We thank Professor Cyril Smyth for his useful

discussion and his critical comments on this manu-

script. This work was supported by a grant to M.F.

from the Health Research Board of Ireland.

FEMSIM 766 27-8-97

M. Fitzgerald et al. / FEMS Immunology and Medical Microbiology 18 (1997) 209^216 215

References

[1] Catlin, B.W. (1990) Branhamella catarrhalis : an organism

gaining respect as a pathogen. Clin. Microbiol. Rev. 3, 293^

320.

[2] Murphy, T.F. (1996) Branhamella catarrhalis : epidemiology,

surface antigenic structure, and immune response. Microbiol.

Rev. 60, 267^279.

[3] Carr, B., Walsh, J.B., Coakley, D., Mulvihill, E. and Keane,

C. (1991) Prospective hospital study of community acquired

lower respiratory tract infection in the elderly. Respir. Med.

85, 185^187.

[4] Murphy, T.F. and Sethi, S. (1992) Bacterial infection in

chronic obstructive pulmonary disease. Am. Rev. Respir.

Dis. 146, 1067^1083.

[5] Meyer, G.A., Shope, T.R., Waecker, N.J. and Lanningham,

F.H. (1995) Moraxella (Branhamella) catarrhalis bacteremia

in children. Clin. Pediatr. 34, 146^150.

[6] Wallace, M.R. and Old¢eld III, E.C. (1990) Moraxella (Bran-

hamella) catarrhalis bacteremia. Arch. Intern. Med. 150,

1332^1334.

[7] Beachey, E.H., Giampapa, C.S. and Abraham, S.N. (1988)

Bacterial adherence. Adhesin receptor-mediated attachment

of pathogenic bacteria to mucosal surfaces. Am. Rev. Respir.

Dis. 138, S45^S48.

[8] Ahmed, K. (1992) Fimbriae of Branhamella catarrhalis as pos-

sible mediators of adherence to pharyngeal cells. APMIS 100,

1066^1072.

[9] Ahmed, K., Rikitomi, N. and Matsumoto, K. (1992) Fimbria-

tion, haemagglutination and adherence properties of fresh

clinical isolates of Branhamella catarrhalis. Microbiol. Immu-

nol. 36, 1009^1017.

[10] Rikitomi, N., Andersson, B., Matsumoto, K., Lindstedt, R.

and Svanborg, C. (1991) Mechanism of adherence of Morax-

ella (Branhamella) catarrhalis. Scand. J. Infect. Dis. 23, 559^

567.

[11] Kellens, J., Persoons, M., Vaneechoutte, M., van Tiel, F. and

Stobberingh, E. (1995) Evidence for lectin-mediated adherence

of Moraxella catarrhalis. Infection 23, 37^41.

[12] Fitzgerald, M., Murphy, S., Mulcahy, R., Keane, C., Coakley,

D. and Scott, T. (1996) Haemagglutination properties of Mor-

axella (Branhamella) catarrhalis. Br. J. Biomed. Sci. 53, 257^

262.

[13] Murphy, S., Fitzgerald, M., Mulcahy, R., Keane, C., Coakley,

D. and Scott, T. (1997) Studies on haemagglutination and

serum resistance status of strains of M. catarrhalis isolated

from the elderly. Gerontology (in press).

[14] Doern, G.V. and Morse, S.A. (1980) Branhamella (Neisseria)

catarrhalis : criteria for laboratory identi¢cation. J. Clin. Mi-

crobiol. 11, 193^195.

[15] Laemmli, U.K. (1970) Cleavage of structural proteins during

the assembly of the head of bacteriophage T4. Nature 227,

680^685.

[16] Bartos, L.C. and Murphy, T.F. (1988) Comparison of the

outer membrane proteins of 50 strains of Branhamella catar-

rhalis. J. Infect. Dis. 158, 761^765.

[17] Murphy, T.F. and Loeb, M.R. (1989) Isolation of the outer

membrane of Branhamella catarrhalis. Microb. Pathog. 6,

159^174.

[18] Harboe, N.M.G. and Ingild, A. (1983) Immunization, isola-

tion of immunoglobulins and antibody titre determination.

Scand. J. Immunol. 17, 345^351.

[19] Bollag, D.M. and Edelstein, S.J. (1992) Protein Methods. Wi-

ley-Liss, New York.

[20] Soto-Hernandez, J.L., Holtsclaw-Berk, S., Harvill, L.M. and

Berk, S.L. (1989) Phenotypic characteristics of Branhamella

catarrhalis strains. J. Clin. Microbiol. 27, 903^908.

FEMSIM 766 27-8-97

M. Fitzgerald et al. / FEMS Immunology and Medical Microbiology 18 (1997) 209^216216