OUP user menu

Protective effect of β-glucan against systemic Streptococcus pneumoniae infection in mice

Geir Hetland, Naohito Ohno, Ingeborg S. Aaberge, Martinus Løvik
DOI: http://dx.doi.org/10.1111/j.1574-695X.2000.tb01420.x 111-116 First published online: 1 February 2000


The antimicrobial effect of soluble β-1,3-d-glucan from Sclerotinia sclerotiorum (SSG) was examined in mice experimentally infected intraperitoneally (i.p.) with Streptococcus pneumoniae serotypes 4 and 6B. SSG was administered i.p. either 3 days before challenge or 3–48 h after challenge. The number of bacteria in blood samples and the mouse survival rates were recorded. Pre-challenge SSG administration protected dose-dependently against both S. pneumoniae type 4 and 6B infections. SSG injected 24 h post-challenge had a curative effect against type 6B but not type 4 pneumococcal infection. The data demonstrate that SSG administered systemically protects against pneumococcal infection in mice.

  • β-1,3-d-Glucan
  • Antimicrobial effect
  • Streptococcus pneumoniae

1 Introduction

β-1,3-d-Glucan is a polyglucose and a major structural component of the cell wall of yeasts and fungi. It is a potent non-specific stimulator of the reticuloendothelial system [1] and phagocytic defense mechanisms in macrophages [2]. β-Glucans stimulate protective activities against bacterial [3,4], viral [5], protozoal [5] and fungal [6] infections. Recently, we found that the polyglucose had a protective effect against Mycobacterium bovis BCG infection in mice as well [7]. β-Glucans also stimulate tumoricidal activity in polymorphonuclear leukocytes (PMN), macrophages and natural killer (NK) cells [810]. These effects are mediated via the lectin binding site for β-glucan in complement receptor 3 (CR3) (CD11b/CD18) on mononuclear phagocytes [11], PMN and NK cells [12].

SSG is a soluble β-1,3-d-glucan obtained from the culture broth of the fungus Sclerotinia sclerotiorum IFO 9395 [13]. It is highly branched and branches at every other main chain glucosyl unit [14], has a high molecular mass (>5×106) and is gel-forming and more viscous than other β-1,3-d-glucans [15]. SSG has proven potent antitumor effects when administered systemically in mice [13,15,16]. In Japan β-1,3-d-glucans from mushrooms, e.g. lentinan, have been used to treat patients with cancer [17].

Streptococcus pneumoniae is a Gram-positive diplococcus that causes potentially lethal diseases like septicemia and meningitis and also less serious infections like otitis media and sinusitis. It has a polysaccharide capsule, which is its most important virulence factor and upon which the classification of pneumococci into 90 different serotypes is based [18]. Among these, serotype 4 is highly virulent in mice in contrast to serotype 6B, which has a more protracted course of infection in the animals [19].

One major public health concern lately has been the increased occurrence of antibiotic-resistant bacteria, like multiresistant S. pneumoniae and methicillin-resistant Staphylococcus aureus. Efforts should therefore be made to find good alternative preventive and curative agents. Since β-glucans protect against infection with Gram-positive bacteria like Staphylococcus aureus[3], but their effect against encapsulated bacteria like streptococci has not been examined, the aim of the present paper was to study whether SSG protects against S. pneumoniae infection in mice. We examined the effect of SSG against the moderately virulent type 6B, as well as against serotype 4, which causes a rapidly progressive infection.

2 Materials and methods

2.1 Mice

All animal experiments were approved by the local representative of the National Animal Research Committee and were performed in accordance with laws and regulations published by the National Ministry of Agriculture. Female, inbred, specific pathogen-free NIH/OlaHsd mice (Harlan Olac Ltd., UK) were 6 weeks of age at arrival and were rested 1 week before entering the experiments.

2.2 β-Glucan

A soluble, highly branched and gel-forming β-1,3-d-glucan with MW>5×106 obtained from the culture broth of the fungus S. sclerotiorum IFO 9395 (SSG) [1315] was used.

2.3 Bacteria

The strain of S. pneumoniae serotype 4 was a human isolate sent to our Institute for serotyping [18]. A strain of S. pneumoniae serotype 6B was kindly supplied by Dr. Jan Poolman, RIVM, The Netherlands. The pneumococcal strains were stored and prepared for challenge as described [19].

2.4 Mouse blood sampling and quantification of colony-forming units (cfu) in blood

Blood samples were obtained from the distal part of the lateral femoral vein and cultured as described [20]. Briefly, the peripheral blood (25 µl) was serially diluted 10-fold in Todd-Hewitt broth (Difco Laboratories, Detroit, MI, USA), and 25 µl of diluted blood was plated onto blood agar plates, which were incubated at 37°C with 5% CO2. After 18 h the colonies were counted.

2.5 Determination of antipneumococcal antibodies

IgG and IgM antibodies to pneumococcal polysaccharide serotype 6B were determined as previously described [20].

2.6 Examination of cross-reactivity between pneumococci serotype 6B and β-glucan

Sera were obtained from mice before and 3, 6 and 14 days after i.p. immunization with 12 µg of pneumococcal 6B polysaccharide (ATCC), 106 cfu of heat-killed (56°C, 1 h) S. pneumoniae serotype 6B, or 200 µg of SSG. Two enzyme immunoassays were performed, with either pneumococcal polysaccharide 6B or SSG as coat. Both assays were incubated with sera from the three groups of mice (n=8) and the anti-IgM conjugate described above.

2.7 Experimental procedure

Three sets of experiments were performed with 8–9 animals in each treatment group (Table 1). The volume of phosphate-buffered saline (PBS) or SSG diluted in PBS injected was 0.4 ml. All animals were bled at the times indicated in the figures and the blood seeded onto plates. Surviving animals were inspected each day and moribund mice were killed by cervical dislocation. Experiment one was terminated by heart puncturing of surviving animals under CO2 anesthesia to obtain serum for determination of anti-pneumococcal antibodies.

View this table:
Table 1

Experimental protocol for i.p. β-glucan SSG treatment of NIH/OlaHsd mice infected with pneumococci serotype 6B or 4

A: Experiments 1 and 3*: Pre-challenge SSG treatment
GroupDay −3Day 0Day 14/Day 9*
UntreatedPn6B ×106 cfu/Pn4 5×101 cfu*Termination
Pre-PBSPBSPn6B ×106 cfu/Pn4 5×101 cfuTermination
Pre-SSG 4 µgSSG LPn6B ×106 cfu/Pn4 5×101 cfuTermination
Pre-SSG 40 µgSSG mPn6B ×106 cfu/Pn4 5×101 cfuTermination
Pre-SSG 200 µgSSG hPn6B ×106 cfu/Pn4 5×101 cfuTermination
B: Experiment 2: Post-challenge SSG treatment
GroupDay 0, 0 hDay 0, 3 hDay 1Day 2Day 9
UntreatedPn6B ×106 cfuTermination
Post-PBS×3Pn6B ×106 cfuPBSPBSPBSTermination
Post-SSG 4 µg×1Pn6B ×106 cfuPBSSSG LPBSTermination
Post-SSG 200 µg ×1Pn6B ×106cfuPBSSSG hPBSTermination
Post-SSG 200 µg×2Pn6B ×106 cfuPBSSSG hSSG hTermination
Post-SSG 200 µg×3Pn6B ×106 cfuSSG hSSG hSSG hTermination
  • Abbreviations: L (low dose), m (median dose), h (high dose).

2.8 Statistics

Non-parametric statistics: Mann-Whitney U and Spearman's rank correlation coefficient (ρ), were used throughout. P values below 0.05 were considered significant.

3 Results

3.1 Effect of β-glucan SSG on S. pneumoniae in peripheral blood

Mice were given PBS or 4–200 µg of SSG i.p. 72 h before challenge with S. pneumoniae serotype 6B (Table 1A). After 24 h post-challenge bacteremia increased sharply only in untreated and PBS-treated animals (Fig. 1). Bacteremia levels were lower in mice receiving 200 µg of SSG than in control groups during the experiment (P=0.035). Animals that received 40 µg of SSG had less bacteremia than PBS-treated or untreated controls after 48 (P=0.03) and 72 h (P=0.34 and P=0.006). Mice given 4 µg of SSG had lower cfu levels than untreated mice 72 h post-challenge only (P=0.01).

Figure 1

Number of pneumococci serotype 6B cfu in peripheral blood from NIH/Ola Hsd female mice pre-treated with PBS or SSG (low dose (L), median dose (m), high dose (h)) i.p. 3 days before challenge with 106 cfu pneumococci type 6B i.p. (see Table 1A). The animals were bled at the intervals indicated, the samples plated and number of cfu counted. (Note log scale for cfu ml−1.) Dead animals are depicted with 1×109 cfu. Data points represent median values from eight animals and indicate lower cfu levels in SSG-treated mice.

Next, 4 µg or 200 µg of SSG or PBS was given i.p. once or repeatedly after 6B challenge (Table 1B). After 24 h bacteremia levels declined for SSG groups only and then increased, except for mice given a single high SSG dose (P=0.03 compared with the PBS group) (Fig. 2). Then mice were challenged with the more virulent S. pneumoniae serotype 4 (Table 1A). Those given the highest SSG dose pre-challenge had less bacteremia than untreated controls after 12, 24 and 48 h (P=0.02) (Fig. 3). Mice receiving this SSG dose 1 day post-challenge also had lower bacteremia than untreated animals after 24 and 48 h (P=0.002).

Figure 2

Number of pneumococci serotype 6B cfu in peripheral blood from mice treated with PBS or SSG (low or high dose) 7, 5 and/or 3 days after challenge with serotype 6B (see Table 1B which also contains an untreated group) and blood sampled and plated as indicated. Data points are median numbers of cfu from eight mice.

Figure 3

Number of pneumococci serotype 4 cfu in peripheral blood from NIH/Ola Hsd female mice pre-treated with PBS or SSG (low dose (L), median dose (m), high dose (h)) i.p. 3 days before or 1 day (h) after challenge with 50 cfu pneumococci type 4 i.p. The data points represent median values of eight mice and demonstrate less effect of SSG against serotype 4 compared with type 6B (Fig. 1).

There was a strongly negative (ρ=-0.88) and statistically significant (P=0.02) correlation between the number of cfu obtained and the dose of SSG given in the first but not the second experiment. In the last one a similar but less pronounced result was found (ρ=-0.68, P=0.14).

3.2 Effect of SSG on survival of mice infected with S. pneumoniae

After 3 days half of the controls in experiment one above were alive compared with =70% of SSG-treated animals (Fig. 4). At day 14 still 40–50% of mice given 4 or 200 µg of SSG had survived compared with 10% of PBS-treated (P=0.005) and none of the untreated (P=0.003) mice. Mice given 40 µg of SSG had a higher survival rate than untreated (P=0.01), but not PBS-treated animals (P=0.21).

Figure 4

Survival rates (median values) of mice (see Fig. 1) pre-treated with PBS or SSG (low dose (Ld), median dose (md), high dose (hd)) i.p. 3 days before challenge with pneumococci serotype 6B i.p. Data points represent median values from eight animals.

One high dose of SSG 24 h post-challenge was the most effective treatment for surviving the infection in the next experiment (P=0.02 compared with PBS controls). For the other SSG treatments there was a similar tendency (Fig. 5). Administration of 200 µg SSG pre-serotype 4 challenge gave a higher survival rate than no treatment (P<0.02) (Fig. 6). Post-challenge SSG had no effect (data not shown).

Figure 5

Survival rates (median values) of mice (see Fig. 2) treated with PBS or SSG (low or high dose) 7, 5 and/or 3 days after challenge with pneumococci serotype 6B. Data points are median numbers of cfu from eight mice.

Figure 6

Survival rates (median values) of mice (see Fig. 3) pre-treated with PBS or SSG (low dose (L), median dose (m), high dose (h)) i.p. 3 days before challenge with pneumococci serotype 4 i.p. The data points represent median values of eight mice and demonstrate less effect of SSG against serotype 4 compared with type 6B (Fig. 1).

3.3 Effect of SSG on antipneumococcal antibody formation

Sera from mice surviving experiment one were examined for IgG and IgM antipneumococcal serotype 6B antibodies. There was no IgG response, but a positive correlation between anti-6B IgM levels (median values: 11 741, 23 143, 112 758 and 95 485 U ml−1) and SSG dose injected (0, 4, 40 and 200 µg, respectively) (ρ=0.69, P=0.04). Possible cross-reactive polysaccharide epitopes on S. pneumoniae type 6B and SSG were searched for, but not found (data not shown).

4 Discussion

Our results show that the β-glucan SSG protects against S. pneumoniae serotypes 4 and 6B infection in mice when administered i.p. 3 days pre-challenge. A positive effect was observed against type 6B even at the low concentration of 0.2 mg kg−1 of SSG (4 µg mouse−1) given in a single dose. SSG also seems to have a curative effect against S. pneumoniae 6B infection because it reduced the death rate of mice injected with the agent 1 day post-challenge. The concentration needed was a single dose of 10 mg kg−1 of SSG (200 µg mouse−1). The findings compare well with the antitumor activity of SSG observed with repeated i.p. administration of 25–200 µg mouse−1 after tumor inoculation [16].

The present i.p. model for S. pneumoniae infection in mice is well established for systemic infection with a reproducible bacteremia [19]. In this model serotype 6B results in a more protracted infection in susceptible mice than does the more virulent serotype 4 [19]. Hence, it is not surprising that β-glucan is more effective against type 6B infection. The more positive effect of SSG against pneumococcal infection when administered pre-challenge is probably due to priming of the monocyte-macrophage system for microbial attack. However, it is difficult to explain why a single SSG dose was effective post-challenge when multiple dosages were not (Figs. 2 and 5). Endotoxin is another well-known stimulator of macrophages [21]. It protects against S. pneumoniae 6B infection in the present model as well (data not shown). However, we rule out the possible influence of endotoxin in the SSG preparations that contained <0.00001% (w/w) endotoxin [22] and were dissolved in sterile and pyrogen-free PBS before use.

The suggested mechanisms for protection of SSG against pneumococcal infection are the following: since SSG is a large molecule (molecular mass >5×106) it could cross-bind CR3 and enhance the activation state of monocytes [11], PMN and NK cells [12]. β-Glucans may also activate macrophages [23] via activation of the alternative complement pathway [11]. However, modulation by soluble β-1,3-glucan of the induction cytokines during sepsis, resulting in an overall decrease in host mortality [24], may be a more important mechanism. The reason for the suggested positive correlation between SSG and the protective (unpublished data) anti-type 6B IgM antibodies may be that SSG prolongs the animals' lives and enables them to generate a humoral response or may be due to a humoral effect of β-glucan per se. This could be unspecific polyclonal stimulation or adjuvant effect of SSG, which apparently is not caused by cross-reactivity between polysaccharide epitopes on S. pneumoniae 6B and SSG.

Pneumococci given i.p. to mice result in a systemic pneumococcal infection [19], which constitutes a major problem in individuals at risk, e.g. splenectomized patients. Although it is not fully understood why splenectomized individuals are at risk for developing fulminant pneumococcal infection [25], the spleen, in addition to being a ‘phagocytic filter’, plays a role in induction and regulation of the antibody response to pneumococcal polysaccharide antigens [26]. Since we found a positive correlation between levels of anti-6B pneumococcal antibodies and SSG and most of the SSG injected i.p. is transported to the spleen and liver [27], the spleen may be an important organ for the protective effect of β-glucan against pneumococcal infection. However, further studies on splenectomized mice in the present model are warranted to study this aspect.

In the present situation with increased occurrence of antibiotic-resistant bacterial infections an important application for β-glucans could be as alternatives or supplements to prophylactic antibiotics for elective surgery and splenectomized patients or treatment of localized infections, e.g. otitis media, in which the role of antibiotics is unclear. Also, increased intracellular killing of bacteria by β-glucan-activated phagocytes may be preferable to the extracellular bactericidal effect of antibiotics and the resulting increased circulating levels of toxins and other bacterial compounds.


We thank Else-Carin Groeng and Rita Bente Leikvold for excellent technical assistance, and Dr. Harald G. Wiker for helpful discussions.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
View Abstract