Protective Effect of Plantago major L. Pectin Polysaccharide against Systemic Streptococcus pneumoniae Infection in Mice


Dr GeirHetland Section of Environmental Immunology, Department for Environmental Medicine, National Institute of Public Health, PO Box 4404 Torshov, N-0403 Oslo, Norway. E-mail:


The antibacterial effect of a soluble pectin polysaccharide, PMII, isolated from the leaves of Plantago major, was examined in inbred NIH/OlaHsd and Fox Chase SCID mice experimentally infected with Streptococcus pneumoniae serotype 6B. Serotype 6B is known to give a more protracted infection when injected intraperitoneally into susceptible mice than more virulent serotypes like type 4. PMII was administered i.p. either once 3 days before challenge or once to thrice from 3 to 48 h after challenge. The number of bacteria in blood and the mouse survival rate were recorded. Pre-challenge administration of PMII and also lipopolysaccharide (LPS), included as a control, gave a dose-dependent protective effect against S. pneumoniae type 6B infection. However, injection of PMII after establishment of the infection in NIH/OlaHsd mice had no effect. The data demonstrate that, firstly, the polysaccharide fraction PMII from P. major protects against pneumococcal infection in mice when administered systemically prechallenge, and secondly that the protective effect is owing to stimulation of the innate and not the adaptive immune system.


The leaves of Plantago major have traditionally been used as a remedy for wound-healing in many countries. Lithander [ 1] reported prophylactic effects for mammary cancer in mice of an aqueous extract of the leaves of the plant, that was attributed to its possible immuno-stimulatory activity. A P. major leaf-water extract did not show any direct in vitro antibiotic activity against S. aureus or E. coli [ 2]. Others [ 3] reported that pectin polysaccharides from P. major were therapeutic for ulcers in rats. A similar effect of a water extract was later observed by Yesilada et al. [ 4]. In addition, the aqueous extract of dried leaves given orally to mice has been shown to have anti-inflammatory and analgesic activities related to inhibition of prostaglandin synthesis [ 5]. The active compounds were not well characterized.

Recently, biologically active polysaccharides have been isolated from P. major [ 6]. One of the leaf polysaccharides, PMII, is a highly purified pectin polysaccharide with a molecular weight of 46–48 kDa. PMII contains both homogalacturonan and ramified regions, and it has a strong complement-fixing activity [ 7] that varies highly between individual normal sera [T. E. Michaelsen, manuscript submitted]. Interestingly, heteroxylan polysaccharides isolated from the seeds of P. major, also show effect on the complement system [ 8]. Bacterial endotoxin (e.g. lipopolysaccharide (LPS)) is a well-known stimulator of macrophages [ 9], promoter of differentiation of pro-myelocytic cells [ 10] and activator of complement [ 11]. Because it is also a possible contaminator of all biological materials, LPS was included as a control in our experiments.

Streptococcus pneumoniae is a gram-positive diplococcus that causes potentially lethal diseases like septicemia and meningitis and also localized infections like otitis media and sinusitis. It has a polysaccharide capsule, that is its most important virulence factor and the basis for classification of pneumococci into 90 different serotypes [ 12]. Among these, serotype 6B is of intermediate virulence in mice and has a relatively protracted course of infection in the animals compared with certain other serotypes like type 4 [ 13].

Lately, a major public health concern is the increased occurrence of antibiotic-resistant bacteria, like multiresistant S. pneumoniae. Efforts are therefore made to find good alternative preventive and curative agents for local and systemic use. Previously, it has been shown that polysaccharides like β-glucans protect against various bacterial infections [ 14]. Recently, we found that β-glucan polysaccharides protect against Bacille Calmette-Guerin (BCG) [ 15] and S. pneumoniae 6B infection in mice as well [ 16]. The aim of the present paper was to study whether PMII from P. major shows similar protection as β-glucan against S. pneumoniae infection.


Isolation of PMII

Leaves of P. major were pretreated with 80% ethanol, dried and extracted with water at 50 °C. Polysaccharide fraction PMII (46–48 kDa) was isolated by ion-exchange chromatography as described [ 6]. The Endospecy test (Seikasaku Sogyo Co. Ltd, Tokyo, Japan) indicated an LPS activity similar to a contamination of 0.06% of LPS in PMII [ 17].


E. coli 055:B5 endotoxin (lot 5L2660) was purchased from BioWhittaker, Walkersville, ML, USA.


All animal experiments were approved by the local representative for the National Animal Research Committee and were performed in accordance with standards published by the Norwegian Ministry of Agriculture. Female inbred, specific pathogen-free NIH/OlaHsd mice (Harlan Olac Ltd, London, UK), male inbred Fox Chase SCID mice on Balb/c background (Gl. Bomholt gård Ltd, Ry, Denmark) or control male Balb/c mice (Bomholt), were 6 weeks of age at arrival and were rested 1 week before entering the experiments. The SCID mice were screened for spontaneous immunoglobulin (Ig) production (‘leakiness’) by analyzing levels of total IgG [ 18] and only mice with < 5 μg IgG/ml were used.


A strain of Streptococcus pneumoniae serotype 6B was kindly supplied by Dr Jan Poolman, RIVM, The Netherlands. It was kept frozen and prepared for challenge as described [ 13].

Mouse blood sampling

Blood samples were obtained from the distal part of the lateral femoral vein and cultured as described [ 19]. At the end of some experiments, blood for antipneumococcal type 6B IgG and IgM determination [ 19] was obtained from surviving mice by heart puncture under anesthesia, and the serum stored at − 20 °C.

Quantification of colony forming units (CFU) in blood

Peripheral venous blood (25 μl) was serially diluted 10-fold in Todd-Hewitt (TH) broth, 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 visually.

Experimental procedure

Four sets of experiments were performed with 7–9 animals in each treatment group ( Table 1, Fig. 2 legend). The volume of phosphate-buffered saline (PBS) or PMII diluted in PBS, was 0.4 ml. All animals were bled at the time points indicated in Figs. 1. 23 and the blood seeded onto plates. Surviving animals were inspected each day and moribund mice were sacrificed by cervical dislocation. Experiment one and three were terminated by heart puncture of surviving mice under CO2 anesthesia to obtain serum for determination of antipneumococcal antibodies.

Table 1.  A. Experimental protocol for PMII or LPS treatment i.p. of NIH/OlaHsd, Fox Case SCID and Balb/c mice infected with pneumococci serotype 6B. Pre-challenge treatment (all animals*) * SCID and Balb/c mice were only given 1.2 mg of PMIIThumbnail image of
Figure . 2.

The experiment is similar to that depicted in Fig. 1, except for the treatment with the lower concentrations of PMII (Ld: 12 μg, LLd: 1.2 μg or LLLd: 0.12 μg) or lower concentrations of LPS (Ld: 120 ng or LLd: 12 ng). The data points represent (A) median numbers of CFU and (B) survival rates of eight mice.

Figure . 1.

(A) Colony-forming units (CFU) in peripheral blood from NIH/OlaHsd female mice pretreated with phosphate-buffered saline (PBS), the pectin PMII (low dose (Ld): 12 μg, median dose (md): 120 μg, high dose (hd): 1200 μg), or E. coli LPS (1.2 μg) i.p. 3 days before challenge with 106 pneumococci 6B i.p. (see Table 1A). The animals were bled at the intervals indicated, the samples plated and number of CFU counted. The data points represent median values from eight animals. (Number of CFU in dead animals was defined as 1 × 109/ml of blood). (B) Survival rates of the same animals.

Figure . 3.

Mice were treated with PBS, PMII (Ld: 12 μg or Hd: 1200 μg) or LPS (1.2 μg) 7, 5 and/or 3 days after challenge with pneumococci 6B (see Table 1B). Blood was sampled and plated as indicated. Data points are (a) median numbers of CFU and (B) survival rates of eight mice.


Non-parametric statistics; Mann-Whithey-U or Spearman's rank correlation test, were used throughout. P-values below 0.05 were considered significant.


Effect of prechallenge PMII and LPS i.p. on bacteremia and survival of NIH/OlaHsd mice infected with S. pneumoniae

Mice were given 12–1200 μg of PMII, 1.2 μg of LPS, or PBS i.p. 72 h before challenge with S. pneumoniae serotype 6B ( Table 1A). After 24 h postchallenge only bacteremia levels for PBS-treated mice increased sharply until the animals succumbed after 72 h ( Fig. 1A,B). The bacteremia for LPS-treated mice reached lethal levels after 96 h, whereas that for animals given PMII either rose moderately (low dose) or declined further before increasing to PBS control levels at day 9 ( Fig. 1A). For animals that received 12 μg or 120 μg of PMII there were significantly lower levels of CFU compared with those for PBS at each time point between 24 and 96 h (P ≤ 0.03). For mice given 1.2 mg of PMII or 1.2 μg of LPS such a difference was only found after 48 and 72 h (P ≤ 0.03).

To determine the lowest effective dose, 12 μg to 120 ng of PMII or 120 and 12 ng of LPS was given before type 6B challenge. After 24 h there were significantly lower bacteremia levels compared with the PBS controls only for mice that received 12 μg of PMII (P = 0.006) ( Fig. 2A). Moreover, in contrast with LPS (ρ = − 0.77, P = 0.2), a more negative (ρ = − 0.89) and statistically significant (P = 0.02) correlation was found between the dose of PMII given and the number of cfu in these experiments.

Figure 1(B) shows the survival rate of the mice in the first experiment ( Fig. 1A) above. After 3 days none of the PBS-treated animals were alive compared with ≥ 50% of the PMII- and LPS-treated ones and the survival rate of the latter two groups was also significantly higher (P < 0.01). In the second experiment with lower doses there were no statistically significant differences between the survival rates ( Fig. 2B). The lowest protective doses in experiment one and two of PMII and LPS against S. pneumoniae infection were 12.0 μg and 1.2 μg, respectively, determined as significantly reduced bacteremia and increased survival rate compared with PBS treatment.

Effect of postchallenge PMII and LPS i.p. in NIH/OlaHsd mice infected with S. pneumoniae

In the third experiment PMII, LPS or PBS were injected once or repeatedly after serotype 6B challenge ( Table 1B). However, there were no statistically significant differences in bacteremia levels ( Fig. 3A). The most effective treatment for surviving the infection was one injection of 12 μg of PMII after 24 h ( Fig. 3B), which resulted in a higher survival rate than each of the LPS treatments (P = 0.01), but not the PBS treatment (P = 0.68) ( Fig. 3B).

Table 2. Table 1B. Experimental protocol for PMII or LPS treatment i.p. of NIH/OlaHsd, Fox Case SCID and Balb/c mice infected with pneumococci serotype 6B. Post-challenge treatment (NIH/OlaHsd mice) Abbreviations: Ld (low dose), md (median dose), hd (high dose), h (hour).Thumbnail image of

Effect of prechallenge PMII and LPS i.p. in Fox Chase SCID and Balb/c mice infected with S. pneumoniae

In order to discern the possible immunological effector mechanisms a fourth experiment similar to the first one, but only with PBS, 1.2 mg of PMII or 1.2 μg of LPS, was performed with SCID and control Balb/c mice. SCID mice treated with PMII had significantly lower bacteremia than those treated with PBS (P = 0.04) or LPS (P = 0.01) ( Table 2A). The survival rate was also higher for PMII than PBS (P = 0.003) or LPS (P = 0.01) treated SCID mice with, respectively, 38%, 17% and 14% of the animals surviving day 14 ( Table 2B). There was no statistically significant difference between the LPS and PBS SCID groups with regard to bacteremia or survival (P = 0.8 and P = 0.1, respectively). The Balb/c mice were more susceptible to serotype 6B infection. After 4 days, only 13% of PMII- and LPS-treated animals survived compared with none of the PBS controls (P = 0.002) ( Table 2C). Hence, calculations of possible differences in bacteremia were not done.

Table 3. Table 2A. Bacteremia (CFU median numbers) and survival of male SCID mice infected with pneumococci serotype 6B after prechallenge treatment with PMII or LPS i.p. * P = 0.04 and P = 0.01 compared with PBS and LPS, respectively. Thumbnail image of
Table 4. Table 2B. Survival (%) of male control Balb/c mice infected with pneumococci serotype 6B after prechallenge treatment with PMII or LPS i.p. * P = 0.003 and P = 0.01 compared with PBS and LPS, respectively. * P =0.1 compared with PBS. Thumbnail image of
Table 5. Table 2C. Survival (%) of male control Balb/c mice infected with pneumococci serotype 6B after prechallenge treatment with PMII or LPS i.p. * P = 0.002 compared with PBS. Thumbnail image of


The present i.p. model for pneumococcal infection is well established as a model for systemic infection that gives reproducible bacteremia [ 13]. Our results show that the polysaccharide PMII protects against S. pneumoniae serotype 6B infection in the mouse model when administered i.p. 3 days before challenge. A protective effect was observed against type 6B at the concentration of 0.48 mg/kg of PMII (12 μg/mouse) given in a single dose. On the other hand, PMII given i.p. had no curative effect against S. pneumoniae 6B infection when given postchallenge.

LPS is ubiquitous and often contaminates biological preparations. This could possibly be the case for the present PMII preparation, in which an LPS activity similar to a contamination of 0.06% of the material was indicated [ 17]. However, because it is unknown whether this represents a proper LPS contamination or a very weak LPS-mimicking effect of PMII, LPS controls were included in all experiments. The data indicate that an LPS concentration of at least 1.2 μg was needed to significantly increase the survival rate of the infected non-SCID mice when given prechallenge. Thus, the putative LPS contamination of 0.06% of PMII can not explain the positive effect found with the 120 μg or 12 μg doses of PMII. This shows that PMII per se had a protective effect against pneumococcal infection in mice, which is also supported by the strongly negative and significant correlation found between the number of CFU and the dose given of PMII. It is unknown whether the effects of the putative LPS contaminant and PMII could be additive or synergistic.

In order to examine whether the spleen is important for the protective effect of PMII against pneumococcal infection [ 20], serum anti6B pneumococcal antibodies were measured in animals surviving the experiments. However, no correlation was found between levels of anti6B pneumococcal IgM or IgG antibodies and the dose of PMII given (data not shown). This agrees with the results obtained with SCID mice, which shows that the underlying mechanism for the observed PMII effect probably is PMII stimulation of the innate and not the adaptive immune system. Furthermore, in contrast to the findings with NIH/OlaHsd and Balb/c animals, LPS did not significantly alter the bacteremia or survival rate of the SCID mice. This may indicate that the effect observed with LPS, in contrast with PMII, is also dependent on an intact adaptive immune system.

CD14 is the cellular receptor on monocytic cells and granulocytes for LPS complexed to LPS-binding protein [ 21]. However, the functional mechanism for PMII, besides the findings referred to above, is unknown. It lacks mannose residues [ 7] and probably does not bind to the cellular-mannose receptor [ 22]. The pectin contains some glucose, but this is considered a contaminant [ 6]. It is therefore also unlikely that PMII functions by binding to the lectin site in CR3 on leukocytes similar to β-glucan [ 23], which is also protective in this pneumococcal infection model [ 16]. However, PMII has complement-fixing activities [ 7] and is recently shown to be a complement activator (T.E. Michaelsen, manuscript submitted). PMII might then initiate an inflammatory process that facilitates the protection against pneumococcal infection.

The S. pneumoniae experiments were performed with serotype 6B because it results in a more protracted infection in susceptible mice than does the more virulent serotype 4 [ 13]. Thus, putative protective effects of PMII against S. pneumoniae infection would be easier to detect when utilizing serotype 6B. Normally, opsonization and phagocytosis by macrophages and polymorphonuclear leukocytes (PMN) eliminate pneumococci and the cells should be more effective after addition of a stimulator. Therefore, the more pronounced effect of PMII against pneumococcal infection when administered pre compared with postchallenge, is not surprising because in the former situation the monocyte-macrophage system is already activated and primed for microbial attack either through direct monocyte activation [ 6] or complement activation by PMII [T.E. Michaelsen, manuscript submitted].

Future applications for PMII and other anti-infectious polysaccharides could be as alternatives or supplements to prophylactic antibiotics for patients subjected to elective surgery or treatment with antibiotics for infections (e.g. with pneumococci 6B).


In conclusion, the observed antipneumococcal effect of PMII polysaccharide extract from P. major L. is mediated via the innate immune system. The effect is probably independent of the putatively positive effect of the suspected LPS contaminant and shows promise for extracts from the plant as anti-infection substances.


We thank Rita Bente Leikvold for excellent technical assistance.