Microbiology (1 997), 143, 55-61
Printed in Great Britain
1.2 National Institute of
Vacci nology1 and
Bacteriologyz, PO Box
4404 Torshov, N-0403
3 Biotechnology Centre o f
Oslo and Department of
Biochern istry, University
of Oslo, Gaustadalleen
21, N-0371 Oslo, Norway
Monoclonal antibodies against Streptococcus
pneumoniae detect epitopes on eubacterial
ribosomal proteins L7/L12 and on streptococcal
elongation factor Ts
Jan Kolberg,’ E. Arne Haib~,~ Rodrigo Lopez3 and Knut Sletten3
Author for correspondence : Jan Kolberg. Tel : + 47 2 204 2660. Fax : + 47 2 235 3605.
Two monoclonal antibodies (mAbs) designated 144,H-3 (IgG2a) and 218,C-5
(IgM) were produced after immunization of mice with two different heat-
treated and sonicated pneumococcal strains. Western blotting, with solubilized
proteins from different bacterial genera and from mammalian lymphocytes,
showed that both mAbs reacted with a protein of approximately 12 kDa in all
66 strains of eubacteria examined, representing 27 different species. The
12 kDa protein was isolated by immunoaff inity chromatography. Subsequent
preparative Western blotting enabled N-terminal amino acid sequence analysis
by microsequencing. A high degree of amino acid sequence similarity with
eubacterial ribosomal proteins L7/L12 was demonstrated. One of the mAbs
(144,H-3) also cross-reacted in Western blotting with a 43 kDa protein, but only
from streptococci. The 43 kDa protein carrying the common streptococcal
epitope was isolated and sequenced in the N-terminal region. A high degree of
amino acid sequence identity was found to elongation factor Ts from
Keywords : monoclonal antibodies, eubacteria, Streptococcus pneumoniae, ribosomal
protein L7/L12, elongation factor Ts
Infections caused by Streptococcus pneumoniae still
remain a major cause of morbidity and mortality
throughout the world, in particular among infants and
elderly people. Capsular polysaccharides are essential
virulence determinants and are used for classification of
pneumococci into 90 types (Lund & Henrichsen, 1978;
Henrichsen, 1995). This organism also has a common
cell wall polysaccharide of which phosphorylcholine is
the immunologically dominant epitope. When mice are
immunized with pneumococci, one of the most promi-
nent antibody responses is to this phosphorylcholine
determinant (McDaniel et al., 1986). This may be the
reason why few reports on monoclonal antibodies
Abbreviations: DOC, sodium deoxycholate; EF-Ts, elongation factor Ts.
The EMBL accession numbers for the amino acid sequences of ribosomal
L7/L12 protein from Streptococcus pneumoniae and Neisseria meningitidis
reported in this paper are P80714 and P80716, respectively. The accession
number for the sequence of the elongation factor from 5. pneumoniae is
P807 1 5.
(mAbs) against pneumococcal capsular polysaccharides
and protein antigens are published. This observation
may also explain why only a few proteins from this
Gram-positive bacterium have been characterized
(Tuomanen et al., 1995), even though proteins may
contribute to the pathogenesis of pneumococcal in-
fections, Furthermore, increased knowledge of common
pneumococcal proteins might also be useful in the
development of new vaccines.
We have immunized mice with different clinical isolates
of S. pneumoniae. After fusion of spleen cells with
myeloma cells the supernatant fluids were first screened
with the strain used for immunization and purified cell
wall polysaccharides. Antibodies that reacted with the
bacterial strain used as immunogen, but not with the cell
wall polysaccharides, were then tested by Western
blotting to select those reacting with epitopes expressed
on proteins. From these experiments we report two
mAbs that recognize highly conserved epitopes on
bacterial L7/L12 ribosomal proteins. The L7/L12 ribo-
somal proteins are among the most investigated com-
ponents of the prokaryotic ribosomes, but there has, up
J. KOLBERG and OTHERS
to now, only been one report on mAbs against these
proteins (Sommer et al., 1985). The mAbs cross-reacted
with L7/L12 ribosomal proteins of other eubacteria,
indicating a common conserved epitope as the reactive
Bacterial strains. The bacteria examined are described in
Table 1. Most of the strains were human clinical isolates, but
some were obtained from the American Type Culture Col-
lection (ATCC). Pneumococcal strains of the different cap-
sular types included in the 23-valent vaccine (Pneumovax ;
Merck, Sharp & Dohme) were typed by the capsular reaction
test with rabbit antisera purchased from Statens Seruminstitut
(Copenhagen, Denmark). Differentiation of types within
groups was carried out at Statens Seruminstitut. Statens
Seruminstitut also supplied us with Streptococcus suis type 8,
strain 14636, and Escherichia coli strain U5/91. Myco-
bacterium leprae was obtained from Dr M. J. Colston
(National Institute for Medical Research, London, UK) .
Mycobacterium bovis, strain Myc 14323, was received from
Rij ksinstituut voor Volksgezondheid en Mileuhygiene
(Bilthoven, The Netherlands). Archaeoglobus fulgidus strain
VC-16 (Steiter, 1988) was obtained from Dr Torleif Lien
(Institute of Microbiology, University of Bergen, Bergen,
Polysaccharides. Pneumococcal capsular polysaccharides
were obtained from ATCC. Purified cell wall polysaccharides
from S. pneurnoniae were a gift from Dr Jprrgen Henrichsen
(Statens Seruminstitut, Copenhagen).
Production of mAbs 144,H-3 and 218,C-5. Heat-treated
(30 min at 60 "C), sonicated pneumococcal strains of capsular
types 8 and 23F were used as immunogens. The type-8-derived
antigen was inoculated into BALB/c mice and resulted in mAb
144,H-3. mAb 218,C-5 was generated by immunization of
C57B1 mice with the antigen from the type 23F strain.
Six-week-old BALB/c and C57Bl mice were immunized with
sonicated bacterial suspensions containing 50 pg protein in
0.25 ml PBS mixed with 0.25 ml Freund's incomplete adjuvant
followed by a booster injection 2 weeks later with the same
mixture. Four months later, and 7 d before fusion, one
BALB/c mouse was injected with the above mentioned
immunogen, followed by 50 pg protein in PBS on days 4 and 3
prior to fusion. Four weeks after the primary injections and 3
d prior to fusion, one C57BI mouse was injected intravenously
with 50 pg protein in PBS. All injections, except the last one to
the C57Bl mouse, were given intraperitoneally. Spleen cells
from the two mouse strains were fused with NSO myeloma
cells by standard methods (Fazekas de St Groth & Scheidegger,
1980; Lane, 1985). Mouse splenocytes were used as feeder
Hybridoma supernatant fluids were screened in an ELISA (see
below) against the immunizing pneumococcal strain and
against cell wall polysaccharides. Those which reacted only
with sonicated and heat-treated pneumococci (30 min at
60 "C) were checked by Western blotting with bacterial
proteins from the immunizing strain and from Haemophilus
infiuenzae, originally used as a negative control. Hybridoma
cells were cloned by limiting dilution with Hybridoma
Enhancing Supplement (Sigma) instead of feeder cells. Hybri-
doma cells were expanded and cultured in the peritoneal
cavities of Pristane-primed mice to obtain ascitic fluid.
Isotyping of mAbs in hybridoma supernatant fluids was
performed in an ELISA with heat-treated (30 min at 60 "C)
and sonicated pneumococci as coating antigen (see below)
using a kit from Zymed Laboratories.
Protein assay. For determination of protein concentrations,
aliquots of the bacteria were dissolved in 0.5 M NaOH. The
Lowry method was used, with BSA as standard.
ELISA. Flat-bottomed microtitre plates (MaxiSorp, Nunc)
were coated overnight at 4 "C with heat-treated (30 min at
60 "C) and sonicated pneumococci (25 pg protein ml-l) in PBS
with 0.02% sodium azide, 50 pl per well. Hybridoma super-
natant fluids were added in 50 pl aliquots and incubated for
2 h at 37 "C. Alkaline-phosphatase-conjugated goat anti-
mouse Ig (Sigma) was used at a 1/2000 dilution in PBS
containing 3 ' / o BSA. p-Nitrophenyl phosphate was used as the
substrate [l mg ml-' in 10% diethanolamine buffer (pH 9.8)
containing 5 mM MgCl,]. PBS plus Tween 20 (0.05%) was
used for the washes between each step (Kolberg et al., 1992).
SDSPAGE and irnmunoblotting. The bacterial suspensions in
PBS were boiled for 5 min with sample buffer containing 2-
mercaptoethanol (Laemmli, 1970). In the case of the myco-
bacteria, the bacteria were first sonicated for 20 min in iced
water and then boiled as for the other samples. Samples
containing 7 pg protein were placed in each well formed by a
15-tooth comb in a Bio-Rad Mini-Protean slab cell apparatus.
SDS-PAGE was performed with stacking and separating gels
containing 4 and 15 '/o (w/v) acrylamide, respectively. The
separated proteins were electrotransferred to a nitrocellulose
membrane (pore size 0.2 pm). To prevent non-specific binding
of proteins, the membranes were incubated for 30 min in a
blocking buffer consisting of 3 '/o BSA in PBS. The mAbs were
used as hybridoma supernatant fluids diluted (range 1/1&
1/1000) in blocking buffer. As secondary antibody, peroxi-
dase-conjugated rabbit anti-mouse Ig was used (Dakopatts,
dilution 1/1000). The immunostaining was performed with 3-
amino-9-ethylcarbazole and hydrogen peroxide in sodium
acetate buffer (pH 5.0). Rainbow protein molecular mass
markers were obtained from Amersham. The protein content
of the different organisms and cells, electroblotted to nitro-
cellulose membranes after SDS-PAGE, was controlled by
protein staining with Amido black.
Cross-inhibition studies of mAb epitope specificity. Nitrocel-
lulose membrane strips containing S. pneumoniae proteins
separated by SDS-PAGE were blocked with 3% BSA in PBS
(see above) and then incubated for 1.5 h at room temperature
with various concentrations of mAb 144,H-3 (isotype IgG2a)
using hybridoma supernatant fluids diluted in blocking buffer
(range 1/10-1/1000). The strips were then incubated with a
1/1000 dilution of mAb 218,C-5 (isotype IgM) followed by a
biotinylated anti-mouse IgM antibody from a kit used for
isotyping (Zymed Laboratories). Peroxidase-conjugated strep-
tavidin was then added and antibody binding visualized as
described above. Conversely, the inhibition of the binding of
mAb 144,H-3 (IgG2a) by various concentrations of mAb
218,C-5 was examined by using a biotinylated anti-mouse
IgG2a antibody. Controls included incubations of strips with
the blocking buffer instead of the mAb used for inhibition.
The specificity of the isotype specific antibodies was also
checked in this assay.
Dot-blot assay. Heat-treated bacteria (30 min at 60 "C) from
the stationary phase of growth were spotted onto nitro-
cellulose as previously described (Kolberg et al., 1992). Some
experiments were also performed with live, exponential phase
pneumococci (determined by optical density measurements
during growth). The primary antibodies were applied in the
Ribosomal protein L7/L12 epitopes in eubacteria
form of diluted hybridoma supernatant fluids and bound
mAbs were detected with peroxidase-conjugated anti-anti-
bodies (see above).
Solubilization of bacterial proteins. Bacteria from the station-
ary phase of growth were harvested by centrifugation, washed
three times in PBS and killed by heat-treatment at 100 "C for
5 min. The suspension was then sonicated, and nucleic acids
were precipitated by addition of ethanol at 4 "C to a
concentration of 20 '/o (v/v). Proteins were precipitated by
increasing the ethanol concentration to 80 '/o. The ethanol
precipitations were omitted in later experiments. The bacteria
were solubilized in 0-5 '/o (w/v) sodium deoxycholate
(DOC)(Merck) in 0.05 M Tris/HCl buffer containing 2 mM
EDTA (pH 8.6) for about 1 h at room temperature. Non-
solubilized materials were removed by centrifugation (10 min,
4 "C, 20000 8).
lmmunoaff inity purification of bacterial proteins and micro-
sequencing. mAb 144,H-3 was purified from ascitic fluid by
affinity chromatography on Protein A-Sepharose (Pharmacia
Biotech). The mAb was then coupled to CNBr-activated
Sepharose according to the recommendations by the manu-
facturer (Pharmacia Biotech). The gel contained about 4 mg
protein ml-'. The DOC extract was first run through a
Sepharose 6B column to remove non-specifically binding
proteins, and then through the column with immobilized mAb
144,H-3. After washing with the buffer used for application,
bound proteins were eluted with 0.1 M triethylamine in
distilled water (pH 11.5). The fractions were neutralized with
acetic acid, dialysed against 0.01 M Tris/HCl buffer (pH 8.6)
and concentrated by vacuum filtration. The eluted proteins
were separated by SDS-PAGE using a comb for preparative
electrophoresis and electroblotted onto a PVDF membrane
(pore size 0.45 pm, Millipore). Rainbow molecular mass
markers were used in the reference well of the comb. Strips
were cut off and probed with mAb 144,H-3. The rest of the
membrane was stained with 0.025 ' / o Coomassie brilliant blue
in methanol. After destaining in 50% (v/v) methanol, bands
reacting with the mAb were identified by comparison with the
molecular mass markers. The strips were incubated with mAb
144,H-3 and then cut out for N-terminal amino acid micro-
sequencing. Automatic Edman degradation was performed on
an Applied Biosystems model 477A equipped with an on-line
120A PTH amino acid analyser.
Sequence alignment. Searches in the SWISS-PROT protein
sequence databank (Bairoch & Boeckmann, 1991) and se-
quence alignment were done using the Wisconsin Sequence
Analyses Package (Devereux et al., 1984). Prediction of
antigenic epitopes was determined according to Kolaskar &
Tongaonkar (1990) using the European Molecular Biology
Network on-line EGCG-8.0 package.
mAb 144,H-3 reacts with a 12 kDa eubacterial protein
and also with a 43 kDa streptococcal protein in
Dot-blotting showed that 60 different heat-treated
pneumococcal strains representing all capsular types
included in the 23-valent vaccine reacted with this
antibody (isotype IgG2a) (data not shown). Live pneu-
mococci did not react with mAb 144,H-3.
Thirty of the pneumococcal strains were also analysed
by Western blotting against mAb 144,H-3. All showed
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Fig. 1 . Western blots with bacteria probed with mAb 144,H-3
(a) or 218,C-5 (b). Lanes 1 and 2, two different strains of
Streptococcus pneumoniae; lane 3, Streptococcus mitis; lane 4,
Streptococcus sanguis; lane 5, Escherichia coli; lane 6,
Haemophilus influenzae; lane 7, Micrococcus luteus; lane 8,
Mycobacterium bovis; lane 9, Archaeoglobus fulgidus.
two reactive bands, of 12 and 43 kDa (Fig. 1). The
hybridoma cells had been recloned and the clones
examined in Western blotting to ensure that the cultures
did not contain two hybridomas producing two anti-
bodies of the same isotype directed against different
proteins. The cross-reactivity of mAb 144,H-3 with
bacteria other than pneumococci was investigated by
Western blotting (Fig. 1, Table 1). For all the 12
examined streptococcal species, mAb 144,H-3 reacted
with both the 12 and 43 kDa proteins. In contrast, in the
case of the other 15 non-streptococcal eubacterial
species, both Gram-positive and Gram-negative, mAb
144,H-3 only reacted with the 12 kDa protein by
Western blotting (Table 1). The two examined myco-
bacterial species showed much weaker reactions than
the other bacteria (M. bovis is shown in Fig. 1, lane 8).
In addition to reacting with the 12 kDa protein of E.
coli, mAb 144,H-3 also showed a weak cross-reaction
with a 30 kDa protein (Fig. 1, lane 5). Some minor
differences in the mobilities of the reactive proteins were
seen between the analysed organisms (Fig. 1). The only
archaeobacterium examined (Table 1) did not react with
mAb 144,H-3 (Fig. 1, lane 9). The epitope on the 12 kDa
protein recognized by mAb 144,H-3 was expressed in all
examined eubacteria, whereas the epitope on the 43 kDa
protein was common only to all the tested streptococci.
Amino acid sequence similarities of eubacterial
12 kDa protein with L7/L12 ribosomal proteins and of
streptococcal 43 kDa protein with elongation factor
Bacterial proteins from DOC extracts were purified by
immunoaffinity chromatography with the mAb bound
to CNBr-Sepharose. The eluted proteins were run in
preparative SDS-PAGE, and the separated proteins were
electrotransferred to a PVDF membrane. In the case of
S. pneumoniae strain 456/94 (type 6B), strips bearing
the 12 and 43 kDa proteins were cut off and the N-
terminal amino acid sequences of these proteins were
determined. The N-terminal amino acid sequence of the
J. KOLBERG and OTHERS
Table 1. Proteins from different organisms detected by mAb 144,H-3 in Western blots
Organisms/cells Western blotting reactivity
S. pneumoniae (n = 30)
S. suis type 8 SIS 14636
S. mitis (n = 3)
S. bovis 971/92
S. salivarius 56/93
S. milleri (n = 2)
S. sanguis (n = 3)
S. pyogenes ATCC 12353
S. agalactiae 1010/90
S. equisimilis C 74
S. mutans 604/92
S. zooepidemicus 44/93
Staphylococcus aureus ATCC 12598
Haemophilus influenzae (n = 5)
Neisseria meningitidis (n = 2)
Escherichia coli U5/41
Salmonella typhimurium 1406 strain 9
Brucella melitensis 16M
Actinobacillus actinomycetemcomitans ATCC 33384
Listeria monocytogenes 4a 52-14
Micrococcus luteus FH-Ba 2029
Bacillus subtilis ATCC 6633
Bacillus stearothermophilus ATCC 7953
Enterococcus faecalis ATCC 19433
Enterococcus faecium ATCC 19434
Mycobacterium bovis BCG
Archaeoglobus fulgidus VC-16
Mouse splenocytes (n = 1)
Human lymphocytes (n = 1)
"Some additional bands were seen in this region. They were also detected with mAbs against irrelevant
t Weaker reactions than for the other bacteria.
12 kDa protein from Neisseria meningitidis strain 44/76
(group B) was also obtained.
These Western blot analyses and the protein staining of
the PVDF membrane showed two additional bands
corresponding to proteins of 25 and 50 kDa, respect-
ively. These were assumed to be light and heavy chains
from leakage of mAb 144,H-3 from the affinity column.
When the sequence of the first 42 amino acids of the
12 kDa protein from S. pneumoniae strain 456/94 was
compared to sequences in the SWISS-PROT databank,
significant sequence similarity was found with bacterial
ribosomal proteins. The identity in the first 42 residues
was 86% to Bacillus subtilis ribosomal protein B-L9,
which corresponds to E. coli ribosomal protein L7/L12
(Fig. 2). In contrast, in the case of the 12 kDa protein of
N. meningitidis, amino acid residues in positions 1-17
were found to be identical in only two positions with the
pneumococcal 12 kDa protein (Fig. 2). However, the
next three amino acids of the meningococcal protein
(18-20) were identical to those of S. pneumoniae.
Comparison of the amino acid sequences of 16 eubac-
terial L7/L12 ribosomal proteins available in the SWISS-
PROT databank showed conserved sequences in several
regions relative to the amino acid sequence of the
Ribosomal protein L7/L12 epitopes in eubacteria
12 kD8 protein
A L N I E N I I A E I K E A S I L E L N D L V K A I E E E F G V T A A A P V A V A A
............................................................................................................ . ...............................................................
Fig. 2 . Comparison of the N-terminal amino acid sequences of Streptococcus pneumoniae (Strpn) 12 and 43 kDa proteins
from strain 456/94 and Neisseria meningitidis (Nesme) 12 kDa protein from strain 44/76 with Bacillus subtilis (Bacsu)
ribosomal protein B-L9, which corresponds to Escherichia coli L7/L12 (Itoh & Wittmann-Liebold, 1978) and EF-Ts from E.
coli (Ecoli) (An eta/., 1981). The proteins were isolated from DOC extracts of the bacteria by affinity chromatography on a
column with mAb 144,H-3 bound to CNBr-Sepharose. The eluted proteins were separated by SDS-PAGE and
electrotransferred onto a PVDF membrane. A strip of the membrane was cut off and probed with mAb 144,H-3. The rest
of the membrane was stained with Coomassie brilliant blue. Strips containing the bands with the 12 and 43 kDa proteins
were cut off and subjected to amino acid microsequencing. Blanks (-) indicate amino acid residues identical to those of S.
pneumoniae at the corresponding positions.
- - S I T
- - U A D
- - S I T
- S S I T
- - N S N
- - S O T
- - N S N
- T T E S
T K A -
L - A -
L K A -
L - A -
L - A C
L - A -
L E A -
L E - -
L K A -
L E C F
L V E -
L K C F
L T C -
L T C -
L T C -
G E D V
A G A -
- C C -
- - - -
G D R P
F E V V
F D V I
Fig. 3 . Alignment of the amino acid sequences of L7/L12 ribosomal proteins from Escherichia coli (rl7-ecoli), Brucella
abortus (rl7_bruab), Salmonella typhimurium (rl7_salty), Bacillus stearothermophilus (rl7_bacst), Bacillus subtilis (rl7-
bacsu), Micrococcus luteus (rl7-miclu), Mycobacterium bovis (rl7_mycbo), Mycobacterium leprae (rl 7_mycle),
Desulfovibrio vulgaris (rl7_desvm), fiberobacter africanum (rl7_libaf), Pseudomonas putida (rl7_psepu), Citrus greening
disease-associated bacterium-li ke organism (rl7_cgdab), Streptomyces coelicolor (rl7_strco), Streptomyces griseus (rl7-
strgr), Streptomyces antibioticus (rl7-strat) and Chlamydia trachomatis (rl7-chltr). The amino acid sequences are from the
SWISS-PROT databank (Bairoch & Boeckmann, 1991). Proteins are numbered relative to rl7-ecoli. Blanks (-) represent
gaps in alignment to obtain the best identity/similarity in the sequence. The asterisk indicates lle69.
L7/L12 ribosomal protein of E. coli (Fig, 3). Of these 16
eubacterial species, the first eight listed in Fig. 3 were
included in the present study. All 16 sequences were
analysed for antigenic epitopes using the predictive
method of Kolaskar & Tongaonkar (1990). The
strongest common predicted epitope for all 16 of the
L7/L12 amino acid sequences was observed in the
consensus sequence comprising amino acid residues
65-73 of the E. coli ribosomal protein L7/L12 at and
around residue Ile69.
The N-terminal sequence of the 43 kDa protein of
pneumococcal strain 456/94 reactive with mAb 144,H-
3 is given in Fig. 2. A sequence identity of 78 % with E.
coli elongation factor Ts (EF-Ts) was found.
mAb 218,C-5 reacts with ribosomal L7/L12 proteins,
but not with the same epitope as mAb 1WH-3
While characterizing mAb 144,H-3, a second mAb that
bound the 12 kDa protein was found. This was desig-
nated 218,C-5 (IgM). mAb 218,C-.5, unlike mAb 144,H-
3, did not react with the 43 kDa pneumococcal protein
with amino acid sequence identity to the E. coli EF-Ts
(Fig. 1). However, the dot-blot patterns of the two mAbs
against different Gram-positive and Gram-negative
eubacteria were similar. To show that the epitopes for
mAbs 144,H-3 and 218,C-5 were located on the same
12 kDa protein, Western blotting was performed with
pneumococcal proteins obtained by immunoaffinity
chromatography with a mAbl44,H-3-CNBr-Sepharose
J. KOLBERG and OTHERS
column. mAb 218,C-5 was also found to react with the
12 kDa protein (data not shown).
The inhibition studies using nitrocellulose strips con-
taining separated pneumococcal proteins showed that
mAb 144,H-3 did not inhibit the binding of mAb 218,C-
5. A weak inhibiton of the binding of mAb 144,H-3 was
seen with the highest concentrations of mAb 218,C-5
(data not shown).
N-terminal amino acid sequencing of the pneumococcal
12 kDa protein reacting with mAb 144,H-3 showed a
high degree of amino acid sequence identity with Bacillus
subtilis ribosomal protein B-L9 (Fig. 2). The molecular
mass of this protein is 12633 Da (Itoh & Wittmann-
Liebold, 1978), which is in agreement with our es-
timation for the pneumococcal analogue. The B. subtilis
B-L9 protein corresponds to E. coli ribosomal proteins
L7/L12 (Itoh & Wittmann-Liebold, 1978). The L7/L12
proteins are essential in bacterial ribosomes for proper
functioning of several elongation factors involved in
protein synthesis (for a review see Liljas, 1990). Ribo-
somal proteins L7/L12 are encoded by the same gene
and differ only by acetylation of the N-terminus of the
L12 protein, which then becomes the L7 protein
(Terhorst et al., 1973).
Data on the nucleotide sequences of ribosomal protein
genes from bacteria, or on the amino acid sequences of
ribosomal proteins, are limited to a few micro-
organisms. Ribosomal proteins belonging to the L7/L12
family share strong amino acid sequence identity within
the eubacteria (Liao & Dennis, 1994). Alignment of the
complete amino acid sequences available for L7/L12
ribosomal proteins from 16 different bacterial species in
the SWISS-PROT databank is shown in Fig. 3. The C-
terminus contains the highly conserved regions, whereas
there is less sequence conservation at the N-terminus.
Our finding that the amino acid residues of the N.
meningitidis 12 kDa protein in this region were quite
different from those of the pneumococcall2 kDa protein
support this observation. Moreover, even though no
sequences for the S. pneumoniae L7/L12 proteins are
available, the fact that the three amino acids identified at
amino acid positions 18-20 were identical to the
pneumococcal protein suggests possible homology be-
tween the N. meningitidis and S. pneumoniae proteins.
The mAb 144,H-3 seems to detect a conserved epitope
within eubacterial L7/L12 proteins. When the available
amino acid sequences were analysed for antigenicity,
several regions displayed strong antigenic properties.
The region around residue Ile69, which in Chlamydia
trachomatis is replaced by Leu (Fig. 3), gave the highest
values, indicating that the epitope is most probably
located within this helix.
The L7/L12 proteins in M. bovis and M. leprae showed
much weaker Western blot reactivity than the other
eubacteria. The sequence comparison showed sequence
identity in the putative epitope. It might be that these
slowly growing organisms contain less ribosomal pro-
teins than the other eubacteria.
The only examined archaeobacterium showed a lack of
reactivity with the mAbs. This was expected because the
archaeal equivalent L7/L12 proteins are more closely
related in both amino acid sequence and size to the
human P-proteins than to their E. coli counterparts
(Casiano & Traut, 1994).
A second mAb, designated 218,C-5, also recognized the
L7/L12 protein. This mAb did not cross-react with the
streptococcal 43 kDa protein, indicating interaction
with different epitopes. This was supported by com-
petition studies. The epitopes are most likely closely
located because the IgM antibody (218,C-5) weakly
inhibited the binding of the IgG2a antibody (144,H-3).
N-terminal sequencing of the 43 kDa protein from S.
pneumoniae reacting with mAb 144,H-3 showed 78%
identity with E. coli EF-Ts. E. coli EF-Ts contains 283
amino acid residues (An et al., 1981), giving a molecular
mass of about 30 kDa. Western blots performed with E.
coli showed a very faint band in the 30 kDa region that
could be EF-Ts (Fig. 1). A shift in molecular mass to
43 kDa was found for the strongly reacting analogue in
all examined strains of S. pneumoniae and other
streptococci. There are no reports on amino acid
sequences for eubacterial EF-Ts proteins other than E.
coli. However, the complete amino acid sequence of
Thermus thermophilus EF-Ts was recently reported by
Blank et al. (1996). EF-Ts from T. thermophilus is
considerably shorter than EF-Ts from E. coli, differing
in size by 86 amino acids.
Our work with mAb 144,H-3 indicates similarities
between ribosomal protein L7/L12 and streptococcal
ET-Ts, a factor which is essential for the elongation of
the polypeptide chain during protein synthesis. The two
proteins recognized by mAb 144,H-3 are both com-
ponents of the translational machinery and should
normally be inaccessible to antibodies. This is consistent
with our finding that the antigens in living pneumococci
were not recognized by mAb 144,H-3 in the dot-blot
It is difficult to speculate about the relationship between
the conserved antigenic determinant on the L7/L12
protein and the cross-reacting epitope on streptococcal
EF-Ts because we do not know the amino acid sequence
of the streptococcal protein.
The mAb 144,H-3 may be a potentially valuable reagent
in studying the streptococcal EF-Ts. The data from the
bacteria analysed so far indicate that this epitope on EF-
TS from streptococci might represent a taxonomic
marker. Interestingly, in the case of the two enterococcal
species examined, mAb 144,H-3 did not react with the
43 kDa protein. The enterococci were formerly classified
as streptococci, but the genus Enterococcus is now
generally accepted (Facklam & Sahm, 1995). Our
observations appear to support this new classification.
Ribosomal protein L7/L12 epitopes in eubacteria
The immunological importance of ribosomal prepara-
tions of M. bovis and other intracellular micro-organ-
isms such as Brucella abortus has been emphasized by
different investigators, and the L7/L12 proteins have
been identified as important antigens (Tantimavanich et
al., 1993; Bachrach et al., 1994; Oliveira et al., 1994).
Ribosomal preparations have been used as vaccines in
experimental animals (Normier et al., 1992) and have
also been examined for use as vaccines for humans
(Michel et al., 1978; Bknk et al., 1993). The mAbs
described here could possibly be useful tools in the
characterization and standardization of ribosomal
Anne Lise Heist@, Gro Lermark, Torunn Marigird and
Gunnhild R ~ d a l provided excellent technical assistance. Part
of this work was financed by the Norwegian Research Council.
An, G., Bendiak, D. S., Mamelak, L. A. & Friesen, 1. D. (1981).
Organization and nucleotide sequence of a new ribosomal operon
in Escherichia coli containing the genes for ribosomal protein S2
and elongation factor Ts. Nucleic Acids Res 9,4163-4172.
Bachrach, G., Banal, M., Bardenstein, S., Hoida, G., Genizi, A. &
Bercovier, H. (1994). Brucella ribosomal protein L7/L12 is a
major component in the antigenicity of brucellin INRA for
delayed-type hypersensitivity in Brucella-sensitized guinea pigs.
Znfect Zmmun 62,5361-5366.
Bairoch, A. & Boeckmann, B. (1991). The SWISS-PROT protein
sequence data bank. Nucleic Acids Res 19, 2247-2249.
Bend, M. C., Kahl, L., Perruchet, A. M., Hermes, H., Mosges, M.,
Normier, G., Binz, H. & Faure, G. C. (1993). Bacterial lysates and
ribosomes as inducers of specific immune responses: a com-
parative study. Scand J Zmmunol38,496-498.
Blank, J., Nock, S., Kreutzer, R. & Sprinzl, M. (1996). Elongation
factor Ts from Thermus thermophilus. Overproduction in
Escherichia coli, quaternary structure and interaction with
elongation factor Tu. Eur J Biochem 236,222-227.
Casiano, C. A. & Traut, R. R. (1994). Stoichiometry of Sulfolobus
ribosomal protein L12e in 50s subunits determined by quantifi-
cation of immunoblots. Biochem Biophys Res Commun 203,
Devereux, J., Haeberli, P. & Smithies, 0. (1984). A comprehensive
set of sequence analysis programs for the VAX. Nucleic Acids Res
Facklam, R. R. & Sahm, D. F. (1995). Enterococcus. In Manual of
Clinical Microbiology, 6th edn, pp. 308-314. Edited by P. R.
Murray, E. J. Baron, M. A. Pfaller, F. C. Tenover & R. H.
Yother. Washington, DC : American Society for Microbiology.
Fazekas de St Groth, 5. & Scheidegger, D. (1980). Production of
monoclonal antibodies : strategy and tactics. J Zmmunol Methods
Henrichsen, J. (1995). Six newly recognized types of Streptococcus
pneumoniae. J Clin Microbiol33, 2759-2762.
Itoh, T. & Wittmann-Liebold, B. (1978). The primary structure of
Bacillus subtilis acid ribosomal protein B-L9 and its comparison
with Escherichia coli proteins L7/L12. FEBS Lett 96, 392-394.
Kolaskar, A. 5. & Tongaonkar, P. C. (1990). A semi-empirical
method for prediction of antigenic determinants on protein
antigens. FEBS Lett 276, 172-174.
Kolberg, J., Aaberge, 1. S . , Jantzen, E., Lsvik, M., Lermark, G. &
Steen, T. (1992). Murine monoclonal antibodies against pneumo-
coccal capsular polysaccharide types 4, 8, 22F and 19A/19F.
Laemmli, U. K. (1970). Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature 227, 680-685.
Lane, R. D. (1985). A short-duration polyethylene glycol fusion
technique for increasing production of monoclonal anti body-
secreting hybridomas. ] Zmmunol Methods 81, 223-228.
Liao, D. & Dennis, P. P. (1994). Molecular phylogenies based on
ribosomal protein L11, L1, L10, and L12 sequences. J Mol Evol
Liljas, A. (1990). Some structural aspects of elongation. In The
Ribosome: Structure, Function 6 Evolution, pp. 309-317. Edited
by W. E. Hill, A. Dahlberg, R. A. Garrett, P. B. Moore, D.
Schlessinger & J. R. Warner. Washington, DC : American Society
Lund, E. & Henrichsen, J. (1978). Laboratory diagnosis, serology
and epidemiology of Streptococcus pneumoniae. Methods Micro-
biol 12, 241-262.
McDaniel, L. 5. & Briles, D. E. (1986). Monoclonal antibodies
against surface components of Streptococcus pneumoniae. In
Monoclonal Antibodies Against Bacteria, pp. 143-164. Edited by
A. J. L. Macario & E. C. de Macario. Orlando, FL: Academic
Michel, F. B., Dussourd d'Hinterland, L., Bousquet, J., Pinel, A. M.
& Normier, G. (1978). Immuno-stimulation by a ribosomal vaccine
associated with bacterial cell wall adjuvant in humans. Znfect
Normier, G., Pinel, A. M., Dussourd d'Hinterland, L., Wigzell, H. &
Binz, H. (1992). Ribosomes as carriers for antigenic determinants
of the surface of micro-organisms. Dev Biol Stand 77,79-85.
Oliveira, S.C., Zhu, Y. & Splitter, G.A. (1994). Recombinant
L7/L12 ribosomal protein and y-irradiated Brucella abortus
induce a T-helper 1 subset response from murine CD4+ T cells.
Sommer, A., Etchison, J. R., Gavino, G., Zecherle, N., Casiano, C. &
Traut, R. R. (1985). Preparation and characterization of two
monoclonal antibodies against different epitopes in Escherichia
coli ribosomal protein L7/L12. J Biol Chem 260,65224527.
Steiter, K. 0. (1988). Archaeoglobus fulgidus gen. nov.: a new
taxon of extremely thermophilic archaebacteria. Syst Appl
Microbiol 10, 172-173.
Tantimavanich, S., Nagai, S., Nomaguchi, H., Kinomoto, M.,
Ohara, N. & Yamada, T. (1993). Immunological properties of
ribosomal proteins from Mycobacterium bovis BCG. Znfect
Terhorst, C., Moller, W., Laursen, R. & Wittmann-Liebold, B.
(1973). The primary structure of an acidic protein from 50-S
ribosomes of Escherichia coli which is involved in GTP hydrolysis
dependent on elongation factors G and T. Eur J Biochem 34,
Tuomanen, E. I., Austrian, R. & Masure, H. R. (1995). Pathogenesis
of pneumococcal infection. N Eng J Med 332, 1280-1284.
Received 22 April 1996; revised 22 August 1996; accepted 1 1 September