The genus Bordetella as defined by Lopez
(1952) consists of the three organisms Bordetella
pertussis, Bordetella parapertussis and Bordetella
bronchiseptica. These species were formerly
classified in the genus Haemophilus, but their
inclusion in a new genus was suggested on the
grounds of cultural and antigenic differences from
other Haemophilus species.
Bord. pertussis was first isolated by Bordet
and Gengou (1906) from children suffering from the
disease known as whooping cough. Eldering and
ILendrick (1937) isolated a second organism, Bord.
hparapertussis, which was also a cause of whooping
cough, although the infection was generally of a
mild type. The third member of the genus, Bord.
bronchiseptica, has frequently been found in
respiratory -tract infections of rabbits, guinea
igs, cats and dogs and has also been isolated from
a few cases of whooping cough (Brown, 1926; Medical
esearch Council, 1951). The isolation of these
species has never been easy, even using complex
edia rendered selective by the addition of
penicillin, and few diagnostic laboratories obtain
igh percentage isolations of Bordetella strains
from cases of whooping cough. Problems are also
encountered in the laboratory culture of Bord.
ertussis. This organism is difficult to grow and maintain in a virulent state, but the other
two species are rriore easily grown and can be
cultured on most,simple laboratory media. Cultur
problems are also reflected in vaccine production.
Whooping cough in young children causes high
morbidity and since the discovery of the causal
organisms, attempts have been made to develop
suitable prophylactic agents. Whole cell
vaccines are commonly used but possess many
disadvantages, the most serious being the
occasional occurrence of encephalopathy following
vaccination. L1uch research has been carried out
to obtain a satisfactory prophylactic material,
without such disadvantages.
The present work is an attempt to prepare a
medium giving optimal growth of virulent Bord.
pertussis, and, using this medium, to obtain a
better knowledge of the chemical and immunological
properties of Bord. pertussis antigens. An
attempt is also made to clarify some of the
antigenic relationships within the genus, using
modern techniques of serological analysis such as
gel diffusion and immunoelectrophoresis.
A historical introduction to the work is
presented, to show the difficulties which earlier
workers have encountered with this genus and to
give an outline of our present knowledge.
1). A liquid medium supporting growth of virulent
Bordetella pertussis has been described. It
contains casamino acids, nicotinamide, glutathione
tris buffer and inorganic salts as well as an
anion exchange resin. The medium is reproducable,
giving good yields of bacteria and is also cheap
and easy to prepare. After the resin has been
removed by decantation or filtration, the medium
contains only dialysable components. Cells
grown in this medium produce haemagglutinin and
protective antigen. The liquid medium can be
solidified by the addition of agar and then
compares favourably with Bordet- Gengou medium
for the growth of Bordetella species.
2). The protective activity of Bordetella pertussis
cells grown in tris -resin medium was assayed by
intracerebral challenge in mice and was shown to
reside in the cell wall. The activity was not
destroyed by treatment of the cell wall with
trypsin, lipase or detergent, although these
removed adherant cytoplasmic material and
trypsin destroyed the histamine -sensitising
factor. The lipopolysaccharide moiety of the
cell wall had no major role in protection.
3). The lipopolysaccharides from Bordetella
pertussis, Bordetella parapertussis and
Bordetella bronchiseptica showed some chemical
similarities, containing hexosamine, heptose and
hexose in varying proportions. The lipopolysacch-
-arides were species specific, but that isolated
from Bord.pertussis was present in virulent and
avirulent strains.
4). Agglutination tests showed some similarities
between strains of the same Bordetella species.
Precipitation tests in agar gel revealed a large
number of antigens common to all three species.
The only antigens which could be identified were
the species specific lipopolysaccharides. Two
antigens were found to be common to the
Bordetella and Haemophilus genera.
5). Soluble haemagglutinins were prepared from
Bord. Pertussis and Bord. parapertussis. They
were reversibly inactivated by oxygen and all
activity was rapidly lost on dialysis. They were
destroyed by the proteolytic enzyme trypsin. An
inhibitor to haemagglutination was found in
cooked meat medium, and an inhibitory material,
removable by treatment with neuraminidase, was
found on ox erythrocytes. Most red cell species
were agglutinated to the same titre.
6). Two previously unknown pigments have been
isolated from Bordetella species. A blue pigment,
for which the name azurin is suggested, has been
obtained from virulent strains of all three
species. Azurin is a protein of low molecular
weight, approximately 16,500, containing copper
(0.156%). It has absorption maxima at 280 mμ and
625 mμ in the oxidised form, but the peak in the
visible wavelengths disappears on reduction with
cysteine or thioglycollate. Azurin is autoxidisable.
A cytochrome of the c type has been isolated
from Bord.pertussis and Bord. bronchiseptica.
It has absorption maxima at 408 mµ in the
oxidised form and at 416 mµ, 522 mμ. and 550 mp,
in the reduced form. The method used for the
purification of the bacterial cytochrome could
also be applied to mammalian cytochrome c.
