OrIGINAL PAPErS - Advances in Clinical and Experimental Medicine

OrIGINAL PAPErS - Advances in Clinical and Experimental Medicine

original papers
Adv Clin Exp Med 2012, 21, 3, 289–299
ISSN 1899–5276
© Copyright by Wroclaw Medical University
Dorota Tichaczek-Goska1, Danuta Witkowska2, Agnieszka Cisowska1,
Stanisław Jankowski1, Andrzej B. Hendrich1
The Bactericidal Activity of Normal Human Serum
Against Enterobacteriaceae Rods with Lipopolysaccharides
Possessing O-Antigens Composed of Mannan*
Bakteriobójcza aktywność normalnej surowicy ludzkiej
wobec szczepów z rodziny Enterobacteriaceae z lipopolisacharydami
zawierającymi O-antygeny typu mannanowego
1
2
Department of Biology and Medical Parasitology, Wroclaw Medical University, Wroclaw, Poland
Institute of Immunology and Experimental Therapy, Polish Academy of Sciences, Wroclaw, Poland
Abstract
Background. The susceptibility of bacteria to the bactericidal activity of serum depends on the structure and organization of the bacterial outer membrane. It is known that the structure of the O-specific polysaccharide chain of
lipopolysaccharide (LPS) plays an important role in the resistance of bacteria to host immune defenses.
Objectives. The susceptibility of rods belonging to Enterobacteriaceae family to the bactericidal activity of the normal human serum (NHS) was examined. The mechanisms of complement activation were also investigated.
Material and Methods. The study was carried out on 15 strains containing LPSs with O-specific polysaccharides
composed of mannan, belonging to the following species: Citrobacter freundii, C. werkmanii, C. braakii, C. youngae, Hafnia alvei, Escherichia coli and Klebsiella pneumoniae. The levels of C3 and C4 complement components,
IgG and IgM immunoglobulin in NHS were examined using specific antibodies. The bactericidal activity of NHS
and its preparations (HS50/20, HSMgEGTA) was determined. LPSs from E. coli O8 strains were analyzed by polyacrylamide gel electrophoresis (PAGE) in the presence of sodium dodecyl sulphate (SDS).
Results. Eleven strains were sensitive to NHS bactericidal activity, and four were resistant. Only one group of
strains was fully susceptible to NHS action. In three other groups, both sensitive and resistant strains were found.
The majority of the strains remained susceptible to NHS activity irrespective of which pathway of serum activity
was blocked. All E. coli O8 strains contained smooth-type LPSs.
Conclusions. Strains belonging to the same serotype showed variable susceptibility to the bactericidal action of
normal human serum. Two mechanisms of the bactericidal activity of NHS have been identified (Adv Clin Exp
Med 2012, 21, 3, 289–299).
Key words: mannan, O-antigen, human serum, bactericidal effect, complement system.
Streszczenie
Wprowadzenie. Podatność bakterii na bakteriobójcze działanie surowicy zależy od struktury i organizacji błony
zewnętrznej drobnoustroju. Wiadomo, że struktura polisacharydowego łańcucha O-swoistego odgrywa istotną
rolę w oporności bakterii na działanie układu odpornościowego organizmu gospodarza.
Cel pracy. Zbadano podatność pałeczek należących do rodziny Enterobacteriaceae na bakteriobójcze działanie
normalnej surowicy ludzkiej (NSL). Określono również mechanizm aktywacji układu dopełniacza oraz jego rolę
w procesie lizy komórek przez NSL.
Materiał i metody. Badaniami objęto grupę 15 szczepów mających O-antygeny typu mannanowego, należących do
gatunków: Citrobacter freundii, C. werkmanii, C. braakii, C. youngae, Hafnia alvei, Escherichia coli i Klebsiella pneumoniae. Oznaczano stężenia składowych dopełniacza C3 i C4 oraz przeciwciał klasy IgG i IgM w NSL. Określano
* This research was supported by a grant from Wroclaw Medical University, Wroclaw, Poland (protocol number
1706).
290
D. Tichaczek-Goska et al.
bakteriobójcze działanie NSL oraz jej preparatów (SL50/20, SLMgEGTA) w stosunku do testowanych pałeczek.
Wyizolowano i charakteryzowano elektroforetycznie (SDS-PAGE) lipopolisacharydy (LPS) szczepów E. coli O8.
Wyniki. Jedenaście szczepów bakterii było podatnych, a cztery oporne na działanie NSL. Tylko jedna grupa pałeczek była wrażliwa na działanie NSL. W pozostałych grupach drobnoustrojów, mimo że struktury O-antygenowe
były identyczne, odnotowano zarówno podatne, jak i niewrażliwe na NSL szczepy. Większość bakterii nie zmieniła swojego wzoru podatności na surowicę, niezależnie od tego, którą z dróg układu dopełniacza inaktywowano.
Wszystkie szczepy E. coli O8 miały LPS typu gładkiego.
Wnioski. Szczepy bakteryjne należące do jednego gatunku i serotypu wykazują zróżnicowanie w podatności na
lityczne działanie białek układu dopełniacza. Odnotowano dwa mechanizmy bakteriobójczego działania NSL
wobec testowanych pałeczek (Adv Clin Exp Med 2012, 21, 3, 289–299).
Słowa kluczowe: mannan, O-antygen, ludzka surowica, działanie bakteriobójcze, układ dopełniacza.
The outer membrane is the first layer of the cell
envelope of Gram-negative bacteria. The internal
part is built of phospholipids, and the external part
is built mainly of lipopolysaccharide (LPS) [1]. LPS
contributes greatly to the structural integrity of the
bacterial cell envelope. It is one of the virulence
factors of bacteria. Lipopolysaccharides of some
Enterobacteriaceae rods from the genera Citrobacter, Hafnia, Escherichia and Klebsiella possess the
O-specific polysaccharide moieties consisting of
mannan or including mannose [2–5].
Mannan is a polymer of monosaccharide mannose. It is a storage product and component of many
plant cells, and a major cell wall glycoprotein of
Candida spp. [6]. The mycobacterial cell wall components lipomannan and lipoarabinomannan have
been described as virulence factors of Mycobacterium tuberculosis [7]. The biological importance of
mannan as a component of bacterial structures is
relatively poorly described. It is known that mannan strongly activates the complement system and
the body’s immune response [8, 9].
As Podschun and Ullmann wrote, “As opportunistic pathogens, Klebsiella spp. primarily attack immunocompromised individuals who
are hospitalized and suffer from severe underlying diseases” [10]. Hafnia alvei strains have been
shown to cause respiratory tract infections, diarrhea and gastroenteritis [11]. Gupta et al. pointed
out: “Citrobacter species cause a wide spectrum of
infections in the urinary tract, blood, … peritoneum and several other normally sterile sites, most
frequently in hospitalized and immunocompromised patients” [12]. Escherichia coli strains can
cause urinary tract infections, and severe diarrhea
in children and adults [13]. Escherichia coli and
Klebsiella spp. bacilli are the most common bacteria responsible for severe infections in humans,
mainly in small children, e.g. in neonatal intensive
care units [14]. Rods belonging to the Enterobacteriaceae family, especially those resistant to wide
range of antibiotics, are an important epidemiological problem in Poland and elsewhere [15].
The bactericidal activity of serum is a natural
barrier, and is a very important factor protect-
ing the macroorganism against various infections caused by Gram-negative bacteria (among
others) [16]. It is known that the structure of
the O-specific polysaccharide chain of LPS plays
an important role in the resistance of bacteria to
host immune defenses, especially to complement
protein deposition and complement lytic activity [17–19]. There are three routes by which the
complement system can be activated: the classical
pathway, the alternative pathway and the lectin
pathway. The lectin pathway is activated when
mannan-binding lectin (MBL) present in the host
serum attaches to mannose or to some other sugar residues (N-acetylglucosamine, fucose) on the
pathogen surface. This activates the MBL- associated serine proteases MASP-1 and MASP-2, and
the latter cleaves C2 and C4 to generate a C3 convertase [20]. MBL also associates with MASP-3
and MAp19, although the function of them is still
unknown. Recent literature also describes other
venues of complement activation: the “extrinsic
protease” pathway, which involves direct cleavage
of C3 and C5 by non-complement proteins such as
kallikrein or thrombin [21] and the so-called “C2
bypass” pathway, entailing direct cleavage of C3 by
MBL/MASP-2 [22].
A number of authors have emphasized the importance of LPS structure on the susceptibility of
bacteria to serum bactericidal activity. These studies were mainly focused on the role of sialic acid in
this phenomenon [18, 23, 24]. Since mannan structures are another component that is rarely found
in the LPS of some bacterial strains, the authors of
the present article deemed it worthwhile to investigate how these structures affect the susceptibility
of Gram-negative bacteria from the Enterobacteriaceae family to serum bactericidal activity.
Thus, the aim of this study was to compare
the susceptibility of selected Enterobacteriaceae
strains with LPSs possessing O-antigens composed
of mannan to the bactericidal activity of normal
human serum (NHS). The authors also identified
which of the complement activation pathways
plays a crucial role in the bactericidal activity of
NHS against the examined strains.
291
Human Serum Activity Against Bacteria with Mannan O-antigens
Material and Methods
Bacterial Strains
The study was carried out on 15 Enterobacteriaceae strains that contain LPSs with mannan Ospecific polysaccharides, which have been well described in world literature [2–5, 25–29] (Table 1).
The bacterial strains were provided by the Polish
Collection of Microorganisms (PCM) in Wroclaw
and by the Veterinary Research Institute in Pulawy (IW). The species affiliation of the examined
strains was confirmed using the API-20E identification test kit (BioMérieux, Warsaw, Poland).
Serum
As described in a previous publication, normal
human serum (NHS) was obtained from healthy
volunteers untreated with any antimicrobial drug.
The samples of NHS were collected, pooled and
kept frozen at –70°C in 0.25 ml portions. A suitable volume of serum was thawed immediately before the experiment and used only once [18].
Determination of C3, C4, IgG
and IgM Levels in NHS
The levels of C3 and C4 complement components, IgG and IgM immunoglobulin were determined using specific antibodies. Nutrient agar
plates with anti-C3, anti-C4, anti-IgG and antiIgM monospecific polyclonal antibodies (MEGATRADING, Gliwice, Poland) were used. The assay
was carried out according to the instructions provided by the supplier.
Treatment of Sera
The alternative pathway of complement activation was blocked by incubating the NHS samples
at 50°C for 20 min (HS50/20) [30]. The classical
and lectin complement pathways were inhibited
by using EGTA (SIGMA-ALDRICH, USA) and
MgCl2 (HSMgEGTA). Complement activation by
these pathways requires Ca2+ and hence may be
inhibited by EGTA, which preferentially chelates
Ca2+. The final concentration of EGTA and MgCl2
in the serum was 10 mM. EGTA solution was prepared according to Fine et al. [31]. NHS samples
decomplemented by heating at 56°C for 30 min
(HS56/30) were used as a control [19]. In HS56/30
serum all complement activation pathways were
inhibited.
Table 1. The polysaccharide repeating units of O-antigens composed of mannan.among rods of the Enterobacteriaceae family
Tabela 1. Wzory podjednostek polisacharydowych O-antygenów typu mannanowego w obrębie pałeczek z rodziny
Enterobacteriaceae
Group
(Grupa)
Strain
(Szczep)
Structure of the O-antigen repeating unit
(Struktura powtarzającej sie podjednostki O-antygenowej)
References
(Literatura)
I
C. freundii O23
→4)-α-D-Manp-(1→2)-α-D-Manp-(1→2)-β-D-Manp-(1→3)-α-DGalpNAc-(1→
[25]
II
C. youngae O1
→4)-α-D-Rhap-(1→3)-β-D-Manp-(1→4)-β-D-Manp-(1→
α-D-Ribf-(1→4)┘
[3]
III
E. coli O8
K. pneumoniae O5
α-D-Manp3Me→[3)-β-D-Manp-(1→2)-α-D-Manp-(1→2)-α-DManp-(1→]n
[2]
IV
E. coli O9
K. pneumoniae O3
H. alvei PCMa 1223
→3)-α-D-Manp-(1→3)-α-D-Manp-(1→2)-α-D-Manp-(1→2)-α-DManp-(1→2)-α-D-Manp-(1→
[5, 26, 27]
Other
strains
C. braakii O7
→3)-α-D-Manp-(1→3)-α-D-Manp-(1→2)-α-D-Manp-(1→2)-α-DManp-(1→2)-α-D-Manp-(1→
α-D-Glcp-(1→3)┘
[28]
C. werkmanii O21
→6)-α-D-Manp3Ac-(1→2)-α-D-Manp-(1→2)-α-D-Manp-(1→3)-α-DGlcpNAc-(1→α-D-Glcp-(1→3)┘
[29]
H. alvei PCM 1204
→3)-α-D-Manp-(1→2)-α-D-Manp-(1→3)-β-D-GlcNAcp-(1→2)-β-DQui3N(Fo)p-(1→3)-α-D-GalNAcp-(1→4)-α-D-GlcAp(1→
[4]
a PCM
a PCM
– Polish Collection of Microorganisms (also applies to Tables 2–5).
– Polska Kolekcja Mikroorganizmów (dotyczy także tabel 2–5).
292
D. Tichaczek-Goska et al.
Bactericidal Activity of NHS
The bactericidal activity of NHS was determined as described by Cisowska et al. [18]. Briefly,
the strains were grown overnight, and then bacterial cells were transferred to fresh nutrient broth
(BIOMED, Warsaw, Poland) and incubated at 37°C
for 30 min. After incubation the bacterial cells were
centrifuged (4000 rpm for 20 min). Then the bacterial suspensions were added to NHS, HS50/20,
HSMgEGTA and HS56/30. To obtain various serum concentrations (25%, 50% and 75%), the NHS
samples were diluted (v/v) with PBS. Then the bacteria with sera were incubated in a water bath at
37°C. After 0, 60 and 180 min, samples were collected, diluted and cultured on nutrient agar plates
(BIOMED, Warsaw, Poland) for 18 h at 37°C. The
number of colony forming units (c.f.u.) at time
0 was taken as 100%. Strains that showed a survival rate above 100% after 180 min of incubation
in sera were regarded as resistant. The mean values
from three separate experiments were calculated.
Isolation and SDS-PAGE
Electrophoresis of LPS
The LPS of E. coli O8 strains (IW 728, IW 729,
IW 928) were obtained using the phenol-water
method described by Westphal and Jann [32] and
analyzed by sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) according
to Laemmli [33]. Briefly, LPS samples were mixed
with sample buffer containing 4% SDS and boiled
for 5 min; 2.5 µl (1 µg) portions were applied to the
gel. LPS bands were visualized after electrophoresis by the silver staining method [34]. Smoothtype LPS from E. coli O111 strain (Sigma-Aldrich,
Poznan, Poland) was used as a control reference in
the process of electrophoretic separation [35].
Results
The levels of C3, C4, IgG and IgM in NHS were
found to be: 1.15 mg/ml (standard: 0.9–1.8 mg/ml),
0.11 mg/ml (standard: 0.1–1.8 mg/ml), 16.3 mg/ml
(standard: 7.0–16.0 mg/ml) and 1.77 mg/ml (standard: 0.4–2.3 mg/ml), respectively [36].
The current study focused on bacterial strains
for which the O-polysaccharide chain structures
have already been determined and published (appropriate references are given in Table 1). An
analysis of these structures showed that some of
the studied strains possess the same formula of
O-antigen repeating units. Therefore, the tested
strains were divided into four groups (see Table 1).
Strains possessing other O-antigen subunit structures were considered separately.
In the first part of experiments the susceptibility of all the studied strains to 50% NHS was
determined. The results of these experiments are
summarized in Table 2. The susceptibility to NHS
among the examined strains was diverse: Eleven
strains were sensitive to the bactericidal action of
50% NHS, and four were resistant. It is worth noting that among the four groups of strains only one
was fully susceptible to the bactericidal action of
50% NHS, while in the three other groups both
sensitive and resistant strains were found. To more accurately characterize the susceptibility of the studied bacterial strains to NHS, experiments in which the bacteria were treated with
25% or 75% NHS were also performed. Strains
sensitive to 50% NHS were incubated with 25%
NHS, while the ones that were resistant to 50%
NHS were treated with 75% NHS. The results of
these experiments are shown in Tables 3 and 4,
respectively.
As seen in Table 3, most of the strains that
were sensitive to 50% NHS were also sensitive to
25% NHS. Only the strains belonging to the Group
III became resistant to the reduced concentration
of NHS.
The strains resistant to 50% NHS remained resistant to 75% NHS (Table 4). The exception was
one of the strains belonging to Group III, which
became susceptible to 75% NHS. Comparing Tables 3 and 4 it is noticeable that the susceptibility
to different NHS concentrations changed exclusively among the strains belonging to Group III.
The characterization of the susceptibility of the
tested bacterial strains to NHS was followed by experiments attempting to determine which complement activation pathway is crucial for the NHS activity against the bacteria. The 11 strains that were
susceptible to 50% NHS were chosen for this phase
of the study. Inspection of the data in Table 5, which
presents the results of this experiment, allowed the
authors to conclude that the majority of the strains
(i.e. groups I, II, IV and others) remained sensitive
to 50% NHS activity irrespective of which pathway
of serum activity was blocked. The exception to this
general rule were the strains in Group III, in which
blocking the classical and lectin pathways of complement activation did not change the susceptibility
of those bacteria, while inhibition of the alternative
complement pathway led to the strains becoming
resistant to 50% NHS (Table 5). The percentage
of surviving bacteria in HS50/20 was 814.8% for
K. pneumonia O5 PCM 57; 1041.7% for E. coli O8
IW 729; and 1880.0% in the case of E. coli O8 IW
728. Bacteria incubated in HSMgEGTA remained
sensitive to the NHS preparation.
1.1·103
1.1·106
5.0·105
1.7·107
4.0·106
1.5·107
PCM 1554
6.3·107
2.3·107
C. werkmanii O21
5.2·107
6.3·107
PCM 1532
1.5·106
1.5·107
C. braakii O7
3.4·106
2.1·107
IW 928
PCM 1204
6.6·107
1.7·107
IW 729
H. alvei
8.3·106
2.5·107
IW 728
– Collection of Microorganisms of Veterinary Research Institute (also applies to Tables 3–5).
– Kolekcja Mikroorganizów Instytutu Weterynaryjnego w Puławach (dotyczy także tabel 3–5).
– colony forming units (also applies to Tables 3 and 4).
– jednostki formujące kolonie (dotyczy także tabel 3 i 4).
dpodatność:
s – sensitive strain; r – resistant strain (also applies to Tables 3–5).
s – szczepy wrażliwe; r – szczepy oporne (dotyczy także tabel 3–5).
dsusceptibility:
cIW
cIW
bc.f.u.
bc.f.u.
Other strains
(Inne szczepy)
2.0·106
1.9·107
PCM 1506
PCM 180
2.1·107
1.1·107
PCM 1493
E. coli O9
1.7·102
6.0·106
PCM 1492
PCM 1223
2.3·101
1.1·107
H. alvei
1.2·102
8.6·106
PCM 2352
PCM 11
7.0·102
5.8·106
PCM 1556
K. pneumoniae O3
E. coli O8
III
IV
C. youngae O1
II
c.f.u.
c.f.u.b
PCM 57
C. freundii O23
I
60
3.3 ± 0.8
27.5 ± 5.4
0.01 ± 0.003
273.9 ± 77.3
82.5 ± 40.8
10.0 ± 7.3
16.2 ± 5.7
388.2 ± 240.2
33.2 ± 31.8
10.5 ± 1.1
190.9 ± 66.5
0.003 ± 0.001
0.0002 ± 0.0001
0.001 ± 0.001
0.012 ± 0.01
percent survival#± SD
Time of incubation [min] (Czas inkubacji [min])
0
K. pneumoniae O5
Strain
(Szczep)
Group
(Grupa)
Tabela 2. Bakteriobójcza aktywność 50% NSL wobec bakterii z O-antygenami typu mannanowego
Table 2. The bactericidal activity of 50% NHS against bacteria with O-antigens composed of mannan
2.0·104
1.4·104
1.9·102
4.4·108
5.7·108
8.2·105
7.8·106
1.0·108
7.2·105
3.2·105
1.0·108
7.0·100
7.0·100
1.0·101
7.0·100
c.f.u.
180
0.1 ± 0.03
0.4 ± 0.03
0.001 ± 0.0001
1913.0 ± 360.9
904.8 ± 75.6
5.5 ± 3.3
37.1 ± 7.5
588.2 ± 174.6
2.9 ± 2.7
1.7 ± 0.9
909.1 ± 101.1
0.0001 ± 0.00000
0.0001 ± 0.00001
0.0001 ± 0.0001
0.0001 ± 0.00004
percent survival#± SD
s
s
s
r
r
s
s
r
s
s
r
s
s
s
s
Susceptibilityd#
(Podatność)
Human Serum Activity Against Bacteria with Mannan O-antigens
293
1.6·106
1.9·106
8.3·106
1.5·107
PCM 1506
IW 928
PCM 1223
PCM 180
Strain (Szczep)
C. youngae O1
E. coli O8
H. alvei
E. coli O9
Group
(Grupa)
II
III
IV
12.7 ± 4.3
19.2 ± 4.1
0.03 ± 0.002
34.7 ± 4.1
19.7 ± 0.8
161.3 ± 25.0
23.8 ± 14.5
0.005 ± 0.001
0.03 ± 0.02
0.0008 ± 0.0003
0.0002 ± 0.0001
percent survival ± SD
60
c.f.u.
1.0·107
6.0·106
4.2·107
3.0·107
0
c.f.u.
3.2·106
1.2·107
1.3·107
1.1·107
272.7 ± 11.6
323.1 ± 42.8
50.0 ± 17.4
312.5 ± 280.8
percent survival ± SD
Time of incubation [min] (Czas inkubacji [min])
Tabela 4. Bakteriobójcza aktywność 75% NSL wobec bakterii z O-antygenami typu mannanowego
Table 4. Bactericidal activity of 75% NHS against bacteria with O-antigens composed of mannan
PCM 1554
3.8·103
1.4·107
C. werkmanii O21
5.2·106
1.5·107
PCM 1532
6.9·106
3.5·107
IW 729
C. braakii O7
1.0·107
6.2·106
IW 728
PCM 1204
3.1·106
1.3·107
PCM 1493
H. alvei
2.6·102
5.4·106
PCM 1492
Other strains
(Inne szczepy)
2.3·103
8.8·106
PCM 11
1.3·102
1.6·107
PCM 2352
K. pneumoniae O3
E. coli O8
III
1.3·101
7.1·106
PCM 1556
IV
C. youngae O1
II
c.f.u.
c.f.u
PCM 57
C. freundii O23
I
60
Time of incubation [min] (Czas inkubacji [min])
0
K. pneumoniae O5
Strain (Szczep)
Group
(Grupa)
Tabela 3. Bakteriobójcza aktywność 25% NSL wobec bakterii z O-antygenami typu mannanowego
Table 3. Bactericidal activity of 25% NHS against bacteria with O-antigens composed of mannan
4.1·108
1.0·108
7.5·106
6.1·106
c.f.u.
180
3.4·106
3.9·106
3.6·102
2.1·106
1.2·108
2.0·107
3.4·107
1.4·103
1.1·102
1.0·101
7.0·100
c.f.u.
180
3727.3 ± 756.3
769.2 ± 172.6
62.5 ± 17.1
190.6 ± 123.0
percent survival ± SD
22.7 ± 4.2
47.0 ± 16.0
0.003 ± 0.001
14.0 ± 2.3
342.9 ± 78.7
322.6 ± 33.7
261.5 ± 7.8
0.03 ± 0.01
0.001 ± 0.01
0.0001 ± 0.0000
0.0001 ± 0.0000
percent survival ± SD
r
r
s
r
Susceptibility
(Podatność)
s
s
s
s
r
r
r
s
s
s
s
Susceptibility
(Podatność)
294
D. Tichaczek-Goska et al.
295
Human Serum Activity Against Bacteria with Mannan O-antigens
Table 5. The mechanism of the NHS bactericidal activity against bacteria with O-antigens composed of mannan
Tabela 5. Mechanizm bakteriobójczej aktywności NSL wobec bakterii z O-antygenami typu mannanowego
Group
(Grupa)
Strain
(Szczep)
Percent of surviving bacteria after 180 min of incubation (±SD)
(Odsetek przeżywających bakterii po 180 min inkubacji (±SD))
HS50/20e (SL50/20)
I
PCM 1556
0.0001 ± 0.00003
s
0.01 ± 0.004
s
PCM 2352
0.1 ± 0.03
s
0.2 ± 0.2
s
PCM 1492
0.006 ± 0.004
s
0.03 ± 0.01
s
PCM 1493
1.1 ± 0.2
s
0.04 ± 0.02
s
IW 728
1880.0 ± 521.2
r
0.7 ± 0.2
s
IW 729
1041.7 ± 187.6
r
1.2 ± 0.9
s
K. pneumoniae O5
PCM 57
814.8 ± 210.8
r
39.3 ± 9.2
s
IV
K. pneumoniae O3
PCM 11
5.5 ± 2.0
s
10.8 ± 3.5
s
Other
strains
(Inne
szczepy)
H. alvei
PCM 1204
2.0 ± 1.1
s
0.01 ± 0.001
s
C. braakii O7
PCM 1532
0.4 ± 0.1
s
0.4 ± 0.01
s
C. werkmanii O21
PCM 1554
2.6 ± 1.0
s
0.4 ± 0.02
s
II
III
e
C. freundii O23
HSMgEGTAf (SLMgEGTA)
C. youngae O1
E. coli O8
HS50/20 – human serum with inactivated alternative complement pathway.
– ludzka surowica z inaktywowaną alternatywną drogą układu dopełniacza.
eSL50/20
fHSMgEGTA
fSLMgEGTA
– human serum with inactivated classical and lectin complement pathways.
– ludzka surowica z inaktywowaną klasyczną i lektynową drogą układu dopełniacza
O-antigen
O-antygen
core and lipid A
rdzeń i lipid A
Fig. 1. Silver-stained SDS–PAGE of the LPS from E. coli O8 strains
Ryc. 1. Obraz elektroforetyczny LPS ze szczepów E. coli O8
296
D. Tichaczek-Goska et al.
Heat inactivation of NHS (HS56/30) completely abolished its bactericidal activity against all
examined strains. The percent of surviving bacterial cells, depending on the strain, ranged from
1111.1% to 10181.8%.
The data concerning the susceptibility of bacteria to NHS and the mechanism of the bactericidal activity of NHS showed that the strains in
Group III significantly differ from the others. This
inspired the authors to isolate the LPSs from three
E. coli O8 strains in Group III. The LPSs were purified and subjected to SDS-PAGE. The electrophoretic separation of LPS, when observed by silver staining, exhibited a profile with characteristic
ladder-like bands, which unambiguously indicates
the presence of the O-antigen part of the LPS. As
shown in Figure 1, all the tested E. coli O8 strains
(IW 728, IW 729 and IW 928), as well as the control E. coli O111 strain [35], possess such ladderlike bands and, consequentl,y smooth-type LPSs.
Discussion
The impact of LPS on the susceptibility of bacteria to the bactericidal activity of serum has been
investigated by many scientists [17–19, 23]. These
studies include, among other things, analyses of
the length of LPS [37] and its composition [23,
24], but up to now there have been no reported
analyses of the role of mannan in the susceptibility of bacterial strains to the bactericidal activity of
serum. Therefore, the aim of the present study was
to check and elucidate the susceptibility of bacterial strains with the same O-antigen repeating unit
structures to NHS bactericidal activity. The authors also decided to study and compare which of
the complement activation pathways plays a key
role in serum bactericidal activity against these
strains.
Considering the results of the tests with 50%
NHS, it is not possible to state that there is a close
correlation between the structure of O-antigen
mannan repeating units and the tested strains’
susceptibility to NHS. As shown in Table 2, different strains with the same type of O-antigen varied
in their susceptibility to 50% NHS. In three studied groups of bacteria both sensitive and resistant
strains were found.
Looking only at the results obtained for 50%
NHS, one might draw the (not fully reasonable)
conclusion, that the structure of O-antigen mannan
does not affect bacterial susceptibility to NHS at all.
The reliability of such a hypothesis can be refuted,
however, when one considers the results obtained
for NHS concentrations other than 50%t. In particular, a comparison of the behavior of the strains
in Group III with other groups suggests that some
kind of association might be found between the
O-antigen mannan structure and the strains’ susceptibility to NHS. With the exception of the Group
III, the strains studied kept the same susceptibility pattern to 25% and 75% serum concentrations
(Tables 3 and 4). Moreover, in additional experiments (data not shown), the strains in Groups I,
II and IV also showed the same susceptibility pattern when analyzed with 12.5% (those susceptible
to 25% NHS) and 87.5% (those resistant to 75%
NHS) concentrations of NHS. Only the strains in
Group III “changed” their behavior when the NHS
concentration was altered: The strains that were
sensitive to 50% NHS were resistant to a 25% concentration, while the resistant strains at 50% were
sensitive when exposed to a 75% serum concentration. These results were unlike the results observed
in all of the other groups of bacteria tested. As
discussed below, the exceptional behavior of the
strains in Group III seems to be confirmed by the
results of the experiments aimed at elucidating the
complement activation pathways.
The discovery of the fact that bacterial strains
with the same O-antigen differ in their susceptibility to various NHS concentrations led the authors
to investigate which of the complement activation
pathways plays a key role in serum bactericidal activity against the tested strains.
The biological significance of mannan as
a component of the O-specific chain is relatively
well documented. Its importance in the activation
of the immune response has been demonstrated by
many scientific scientists [8, 9, 37, 38]. However,
its role in the resistance of bacteria to serum bactericidal activity has not yet been well described.
It is known that E. coli O8 LPS has a linear
mannose homopolysaccharide chain and that
the chemical structure of its O-specific chain are
identical to that of K. pneumoniae O5. The K.
pneumoniae O3 LPS and LPSs from E. coli O9
and H. alvei PCM 1223 strains also possess mannose homopolysaccharide [2, 5, 26, 27], as shown
in Table 1. It could therefore be assumed that the
main mechanism for serum bactericidal activity
against these strains is based on activation of the
lectin pathway, which is initiated by the binding of
MBL to carbohydrates, e.g. mannose, on microbial
surfaces.
The results presented in Table 5 show that
such a mechanism of complement system activation could be significant in the cases of the strains
in Groups I, II, IV and the three strains not belonging to any of the groups. Blocking the classical and lectin complement activation pathways
(HSMgEGTA), as well as the alternative pathway
(HS50/20), did not affect the susceptibility of these
Human Serum Activity Against Bacteria with Mannan O-antigens
strains to serum activity. For these strains, all
pathways of complement activation are probably
equivalent. They were activated independently of
each other, and this activation was sufficient to
mediate bacterial lysis.
As shown in studies carried out on bacteria
with mannose-rich O-antigens [9], LPSs possessing the mannose homopolysaccharide as an Ospecific chain (K. pneumoniae O3, K. pneumoniae
O5, E. coli O8, and E. coli O9) strongly activated
the complement system, and there was no noticeable difference in the intensity of the anticomplementary activity among these LPSs. However, LPSs
possessing heteropolysaccharide moiety showed
much lower activity. The ability of LPS from K.
pneumoniae O3 to activate human complement
was more than 100 times higher than that of LPS
from E. coli O111, E. coli O55 or Salmonella enteritidis, strains without mannose in the O-antigens.
Moreover, Yokochi et al. [9] established that K.
pneumoniae O3 LPS probably activates the complement system by using the alternative pathway.
For the strains in Group III, inactivation of
the alternative complement pathway (by using
HS50/20 serum preparation) led to these strains
becoming resistant to serum. This result clearly
demonstrates the crucial role of the alternative
route in the mechanism of serum bactericidal activity against strains in Group III. This observation
seems to be important, since these strains possess
mannose-containing homopolysaccharides as the
O-antigens.
In contrast to the results presented above, Jiang et al. [38] showed that LPSs possessing a mannose-rich polysaccharide structure (isolated from
K. pneumoniae O3, K. pneumoniae O5, E. coli O8,
and E. coli O9 strains) activate the complement system via the lectin pathway. Other reports, however,
emphasize the role of the alternative pathway or
demonstrate the equal role of all three pathways of
complement activation (ie, the classical, lectin and
alternative pathways). Schweinle et al. [8] showed
that MBL enhances complement deposition via
the alternative pathway on Salmonella montevideo
297
strains possessing a mannose-rich LPS and results
in serum killing these organisms, which are resistant to complement lysis in the absence of MBL.
They emphasized that MBL and the complement
system cooperate in the host organism’s first line
of defense. On the other hand, a study carried out
by Fernandez-Prada et al. [39] showed that both
the classical and lectin complement pathways are
involved in killing Brucella abortus and Brucella
melitensis. This result is interesting, because these
strains do not possess mannose in the O-antigens.
Thus, in light of the results of the current study
and those obtained in other laboratories, one can
conclude that the presence of the mannose Oantigen homopolymer is not a sufficient factor to
determine lectin complement pathway activation.
The differences in susceptibility to NHS and
complement pathway activation among the E. coli
O8 rods in Group III led the current authors to
isolate the LPSs from these strains. Contrary to
expectations, all strains contained smooth-type
LPS, regardless of whether they were derived
from serum-sensitive or serum-resistant strains.
This means that although the presence of mannan
structures does not directly correlate with the susceptibility of bacteria to NHS, mannan O-antigen
structures can in some cases affect the mechanism
of serum bactericidal activity against certain bacterial strains.
It should be noted that mannan is a substance
rarely found in structures of Gram-negative bacteria. Therefore, the small number of strains and the
wide diversity of species used in this study does
not allow clear conclusions to be formed regarding
the importance of mannan in protecting bacteria
against the host’s defense mechanisms. It should
be also borne in mind that the bactericidal activity
of serum is a very complex phenomenon, and still
not fully understood. Many components of the
bacterial membrane are involved in this process,
but LPS plays the most important role. Therefore,
it seems advisable carry out further studies to check
how mannan-containing LPSs isolated from the
E. coli O8 strains in Group III affect this activity.
References
  [1] Raetz CRH: Biochemistry of endotoxin. Annu Rev Biochem 1990, 59, 129–170.
  [2] Jansson P, Lonngren J, Widmalm G: Structural studies of the O-antigen polysaccharides of Klebsiella pneumoniae
O5 and Escherichia coli O8. Carbohydr Res 1985, 145, 59–66.
  [3] Kocharova NA, Mieszała M, Zatonsky GV, Staniszewska M, Shashkov AS, Gamian A, Knirel YA: Structure of
the O-specific polysaccharide of Citrobacter O1 containing an α-D-ribofuranosyl group. Carbohydr Res 2004, 339,
321–325.
  [4] Katzenellenbogen E, Romanowska E, Kocharova NA, Shashkov AS, Knirel YA, Kochetkov NK: Structure of the
polysaccharide of Hafnia alvei 1204 containing 3,6-dideoxy-3-formamido-D-glucose. Carbohydr Res 1995, 273,
187–195.
  [5] Prehm P, Jann B, Jann K: The O9 antigen of Escherichia coli. Structure of the polysaccharide chain. Eur J Biochem
1976, 67, 53–56.
298
D. Tichaczek-Goska et al.
  [6] Nelson RD, Shibata N, Podzorski RP, Herron MJ: Candida mannan: chemistry, suppression of cell-mediated
immunity, and possible mechanisms of action. Clin Microbiol Rev 1991, 4, 1–19.
  [7] Dao DN, Kremer L, Guerardel Y, Molano A, Jacobs WR, Porcelli SA, Briken V: Mycobacterium tuberculosis
lipomannan induces apoptosis and interleukin-12 production in macrophages. Infect Immun 2004, 72, 2067–
2074.
  [8] Schweinle JE, Ezekowitz RA, Tenner AJ, Kuhlman M, Joiner KA: Human mannose-binding protein activates
the alternative complement pathway and enhances serum bactericidal activity on a mannose-rich isolate of Salmonella. J Clin Invest 1989, 84, 1821–1829.
  [9] Yokochi T, Inoue Y, Kimura Y, Kato N: Strong interaction of lipopolysaccharides possessing the mannose homopolysaccharides with complement and its relation to adjuvant action. J Immunol 1990, 144, 3106–3110.
[10] Podschun R, Ullmann U: Klebsiella spp. as nosocomial pathogens: epidemiology, taxonomy, typing methods, and
pathogenicity factors. Clin Microbiol Rev 1998, 11, 589–603.
[11] Ramos A, Damaso D: Extraintestinal infection due to Hafnia alvei. Eur J Clin Microbiol Infect Dis 2000, 19,
708–710.
[12] Gupta N, Yadav A, Choudhary U, Arora D: Citrobacter bacteremia in a tertiary care hospital. Scand J Infect Dis
2003, 35, 765–768.
[13] Farmer JJ. III. Enterobacteriaceae: introduction and identification. In: Murray PR, Baron EJ, Pfaller MA, Tenover
FC, Yolken RH. Eds. Manual of clinical microbiology. Washington D.C.: American Society for Microbiology 1999,
pp. 442–458.
[14] Castro B, Montesinos I, Fuster-Jorge P, Delgado T, Miguel-Gómez MA, Sierra A: Epidemiology of Enterobacteriaceae causing bloodstream infections in neonatal intensive care unit patients. Enferm Infecc Microbiol Clin
2010, 28, 227–232.
[15] Empel J, Baraniak A, Literacka E, Mrówka A, Fiett J, Sadowy E, Hryniewicz W, Gniadkowski M; Beta-PL
Study Group. Molecular survey of beta-lactamases conferring resistance to newer beta-lactams in Enterobacteriaceae isolates from Polish hospitals. Antimicrob Agents Chemother 2008, 52, 2449–2454.
[16] Frank MM: Complement deficiencies. Pediatr Clin North Am 2000, 47, 1339–1354.
[17] Kłak M, Jankowski S: The influence of lipopolysaccharides isolated from Enterobacteriaceae strains on the bactericidal activity of normal cord serum. Adv Clin Exp Med 2010, 19, 57–64.
[18] Cisowska A, Bugla-Płoskońska G, Tichaczek-Goska D, Doroszkiewicz W, Jankowski S: The susceptibility of
Escherichia coli strains with sialic acid-containing lipopolysaccharides or capsules to the bactericidal action of
normal human serum. 7th conference: Molecular biology in diagnostics of infectious diseases and biotechnology.
Warsaw: SGGW, 2004, 41–47.
[19] Jankowski S, Rowiński S, Cisowska A, Gamian A: The sensitivity of Hafnia alvei strains to the bactericidal effect
of serum. FEMS Immunol Med Microbiol 1996, 13, 59–64.
[20] Fujita T, Matsushita M, Endo Y: The lectin-complement pathway – its role in innate immunity and evolution.
Immunol Rev 2004, 198, 185–202.
[21] Markiewski MM, Lambris JD: The role of complement in inflammatory diseases from behind the scenes into the
spotlight. Am J Pathol 2007, 171, 715–727.
[22] Atkinson JP, Frank MM: Bypassing complement: evolutionary lessons and future implications. J Clin Invest 2006,
116, 1215–1218.
[23] Bugla-Płoskońska G, Doroszkiewicz W: Bactericidal activity of normal bovine serum (NBS) directed against
some Enterobacteriaceae with sialic acid-containing lipopolysaccharides (LPS) as a component of cell wall. Pol
J Microbiol 2006, 55, 169–174.
[24] Mielnik G, Gamian A, Doroszkiewicz W: Bactericidal activity of normal cord serum (NCS) against Gram-negative
rods with sialic acid-containing lipopolysaccharides (LPS). FEMS Immunol Med Microbiol 2001, 31, 169–173.
[25] Katzenellenbogen E, Ekiel I, Romanowska E: The structure of the O-specific polysaccharide chain from Citrobacter O23-lipopolysaccharide. Carbohydr Res 1988, 179, 349–357.
[26] Curvall M, Lindberg B, Lonngren J, Nimmich W: Structural studies on the Klebsiella O group 3 lipopolysaccharide. Acta Chem Scand 1973, 27, 2645–2649.
[27] Katzenellenbogen E, Kocharova NA, Zatonsky G.V Kübler-Kiełb J, Gamian A, Shashkov AS, Knirel YA, Romanowska E: Structural and serological studies on Hafnia alvei O-specific polysaccharide of alpha-D-mannan
type isolated from the lipopolysaccharide of strain PCM 1223. FEMS Immunol Med Microbiol 2001, 30, 223–
227.
[28] Kocharova NA, Zatonsky GV, Bystrova OV Shashkov AS, Knirel YA, Kholodkova EV, Stanislavsky ES: Structure of the O-specific polysaccharide of Citrobacter braakii O7a, 3b, 1c. Carbohydr Res 2001, 333, 335–338.
[29] Knirel YA, Kocharova NA, Bystrova OV, Katzenellenbogen E, Gamian A: Structures and serology of the
O-specific polysaccharides of bacteria of the genus Citrobacter. Arch Immunol Ther Exp 2002, 50, 379–391.
[30] Eidinger D, Bello E, Mates A: The heterocytotoxicity of human serum. I. Activation of the alternative complement pathway by heterologous target cells. Cell Immunol 1977, 29, 174–186.
[31] Fine DP, Marney SR, Colley DG, Sergent JS, Des Prez RM: C3 shunt activation in human serum chelated with
EGTA. J Immunol 1972, 109, 807–809.
[32] Westphal O, Jann K: Bacterial lipopolysaccharides: extraction with phenol-water and further application of the
procedure. Meth Carbohydr Chem 1965, 5, 83–91.
[33] Laemmli UK: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970,
227, 680–685.
Human Serum Activity Against Bacteria with Mannan O-antigens
299
[34] Tsai CM, Frasch CE: A sensitive silver stain for detecting lipopolysaccharides in polyacrylamide gels. Anal Biochem 1982, 119, 115–119.
[35] Gupta RK, Egan W, Bryla DA, Robbins JB, Szu SC: Comparative immunogenicity of conjugates composed of
Escherichia coli O111 O-specific polysaccharide, prepared by treatment with acetic acid or hydrazine, bound to
tetanus toxoid by two synthetic schemes. Infect Immun 1995, 63, 2805–2810.
[36] Herold G: Medycyna wewnętrzna 4th ed. Warsaw: PZWL, 2006, pp. 1131–1142.
[37] Murray GL, Attridge SR, Morona R: Altering the length of the lipopolysaccharide O-antigen has an impact on
the interaction of Salmonella enterica serovar Typhimurium with macrophages and complement. J Bacteriol 2006,
188, 2735–2739.
[38] Jiang G, Sugiyama T, Kato Y, Koide N, Yokochi T: Binding of mannose-binding protein to Klebsiella O3 lipopolysaccharide possessing the mannose homopolysaccharide as the O-specific polysaccharide and its relation to
complement activation. Infect Immun 1995, 63, 2537–2540.
[39] Fernandez-Prada CM, Nikolich M, Vemulapalli R Sriranganathan N, Boyle SM, Schurig GG, Hadfield TL,
Hoover DL: Deletion of wboA enhances activation of the lectin pathway of complement in Brucella abortus and
Brucella melitensis. Infect Immun 2001, 69, 4407–4410.
Address for correspondence:
Dorota Tichaczek-Goska
Department of Biology and Medical Parasitology
Wroclaw Medical University
Mikulicza-Radeckiego 9
50-367 Wrocław
Poland
Tel.: +48 71 784 1523
Email: [email protected]
Received: 5.09.2011
Revised: 25.10.2011
Accepted: 6.06.2012