Introduction

Mucositis is a debilitating side effect of chemotherapy differentially affecting all regions of the gastrointestinal tract constituted with differing rates of proliferation and differentiation in the small and large intestine [1]. Methotrexate (MTX) is a common chemotherapeutic agent known to induce intestinal mucositis. MTX is an anti-metabolite that exerts its cytotoxic effect through the inhibition of folate metabolism by a down-regulatory effect on the enzyme dihydrofolate reductase, resulting in an inhibition of DNA synthesis [2]. Moreover, MTX-induced mucositis in rodent models produces similar intestinal mucosal injury to that seen in humans [2] and is associated with the decreased proliferation of enterocytes and disruption to intestinal permeability [3, 4].

Currently, an effective agent for the amelioration of chemotherapy-induced mucositis is not clinically available. Such a product would be highly beneficial for cancer patients undergoing chemotherapy. Recently, probiotics have emerged as a potential form of therapy for gastrointestinal diseases by maintaining intestinal barrier function and health [5]. Probiotics are defined as bacteria or bacterial products that have a significant healthy benefit to the host [6]. Probiotic bacteria, such as lactobacilli and bifidobacteria, have been reported to enhance intestinal epithelial barrier function and have demonstrated preventative or treatment properties in human diseases, such as rotavirus-induced diarrhoea [7] and inflammatory bowel disease [8]. Additionally, prevention of antibiotic-induced diarrhoea in the small intestine has been achieved following administration of either Lactobacillus GG or the yeast Saccharomyces boulardii [9]. In the context of chemotherapy, Lactobacillus species have been reported to improve intestinal microflora by re-establishing intestinal microflora and reducing bacterial translocation post-MTX administration [3]. However, effects of commercially available probiotics on intestinal integrity and barrier function have not been described in a chemotherapy-induced intestinal damage model. This study aimed to evaluate two commercially available yoghurt products, containing known probiotics, in rats with MTX-induced intestinal mucositis.

Materials and Methods

Animals

Thirty-two adult male Sprague Dawley rats (90.8 ± 1.3 g) were housed individually in Tecniplast® metabolism cages in the Animal Care Facility of the Women’s and Children’s Hospital (WCH), Children, Youth and Women’s Health Service. The study was approved by the Animal Ethics Committees of the University of Adelaide and WCH Children, Youth and Women’s Health Service, Adelaide, South Australia (Australia), and followed the National Health and Medical Research Council (Australia) Code of Practice for Animal Care and the use of animals for scientific purposes (1997).

Rats were allocated to four groups (n = 8 rats/group), provided with an 18% casein-based diet [10] and allowed access to water ad libitum for the duration of the study. Rats were injected with MTX (Lederle Laboratories, Baulkham Hills, NSW, Australia) or saline (controls) subcutaneously (2.5 mg/kg) at day 0 once daily for 3 days and were sacrificed 5 days after the first MTX injection. Rats received one of the following: saline + water gavage (group 1), MTX + saline (group 2), MTX + cow’s milk yoghurt (CY, group 3), or MTX + sheep’s milk yoghurt (SY, group 4). Gavages were delivered twice daily (1 ml/gavage) for 7 days prior to MTX injection and for the remainder of the trial. The CY was fermented with Lactobacillus johnsonii, and the SY was a Chr Hansen starter culture (YC-180) containing Lactobacillus bulgaricus and Streptococcus thermophilus at concentrations of 107 organisms/l. Small intestinal permeability (IP) was determined on days −7, −1, 2, and 5 of the study. Rats were killed via cervical dislocation under CO2 anaesthesia. The abdomen was opened via a midline incision and the stomach, duodenum, small and large intestine were excised and weighed. Gut contents were removed and placed on an ice-cold slab, and small intestinal segments were determined as previously described by Howarth [11]. Intestinal tissues were collected for determinations of small intestinal biochemical brush-border enzyme activity, and sections were collected for histological assessment and were placed in methacarn fixative.

Small Intestinal Permeability

Pre-test urine samples were collected prior to commencement of permeability testing. Following an overnight fast, rats were gavaged with a 2-ml oral solution containing 13.7 mg lactulose (L; Dupholac®, Solvay-Duphar, B.V., Holand), 7.2 mg l-rhamnose (R; SIGMA-Aldrich, Germany) and 7.2 mg mannitol (M; SIGMA-Aldrich, Germany) at approximately 0900 h. Water was permitted throughout testing to ensure urine production, and food was permitted 5 h after the sugar gavage. All subsequent urine voided for the following 24 h was collected, measured and stored at −20°C for later analysis. Urine samples were diluted according to urine volume, desalted and washed, centrifuged and the resultant supernatant filtered twice through a 0.2-μm filter (Acrodisc ®; Gelman Sciences, Ann Arbor, MI).

L, R, and M sugar excretion (permeability) were determined using a modified method of high-performance liquid chromatography (HPLC; Spectra Physics SP 8810; San Jose, CA) [12], using a mobile phase of degassed 70% acetonitrile (Merck, Darmstadt, Germany). The HPLC was equipped with a refractive index detector (LC 1240 RI Detector; GBC Scientific Equipment, VIC, Australia) and a linear chart recorder (Linear 1200, Alltech Associates, Deerfield, IL). Total percentage urinary excretion of the individual sugar probes were calculated, and results were expressed as the percentage of L/R or L/M ratios to eliminate confounding factors such as gastric emptying, intestinal transit and renal clearance [12].

Histology

After 2 h methacarn-fixed samples were transferred to ethanol for 48 h and then embedded in paraffin wax and cut transversely into 2-μm sections. Fixed sections were then stained with haematoxylin and eosin and were scored by a blinded observer using a light microscope (Olympus, BH-2, Tokyo, Japan) utilizing a previously reported severity scoring index [11]. In brief, sections of the duodenum, proximal jejunum and distal ileum were examined using a semi-quantitative histological assessment of intestinal damage to obtain an overall damage/severity score. A total of 11 previously determined criteria were examined in each section, graded from 0 (normal) to 3 (maximal damage), where the maximal damage score was 33.

In Vitro Sucrase Activity

Duodenum, jejunum, and ileum sections (4 cm) were prepared for sucrase and lactase activity according to the first two steps in Shirazi-Beechey et al. [13]. Briefly, the brush-border membrane containing the disaccharidase enzymes were isolated via hypo-osmolar shock followed by centrifugation. Aliquots were stored in liquid nitrogen prior to being thawed and assayed using a previously described method by Dahlqvist [14]. This assay involves the addition of a known quantity of the substrate to the enzyme for 30 min combined with glucose oxidase to determine the amount of glucose liberated over the 30 min. Sucrase and lactase activity were then related to the protein concentration of the enzyme in the analysed sample. Sucrase and lactase activities were expressed as μmol of substrate hydrolysed at 37°C at pH 6.0/mg protein/h.

Statistical Analyses

Statistical analysis for semi-quantitative histological scoring of intestinal damage was determined using a Kruskal-Wallis non-parametric ANOVA with a Dunn’s post-hoc test, expressed as mean (range). For remaining data, statistical analyses were carried out using a one-way ANOVA with a Tukey’s post-hoc test, and data expressed as mean ± SEM. Differences were considered significant if P < 0.05.

Results

Semi-quantitative histological assessment in the duodenum showed that MTX-induced significant intestinal damage (no treatment) compared to both control and MTX + SY rats (P < 0.05; Table 1), whereas MTX + CY was not significantly different to saline or MTX controls. In the jejunum and ileum, regardless of treatment, histological severity in all groups receiving MTX was significantly higher compared to saline controls (P < 0.02; Table 1). MTX + SY rats had a significantly reduced severity in the duodenum score compared to MTX control (P < 0.05). Ileal histological assessment also revealed that MTX + CY and MTX + SY groups had significantly higher scores compared to control rats and were similar to results obtained from MTX rats.

Table 1 Semi-quantitative histological assessment of MTX-induced intestinal damage in different regions of the small intestine
Full size table

The small intestinal permeability, or L/M ratio, in MTX-treated rats was significantly elevated on days +3 and +6 when compared to rats treated with MTX + CY or MTX + SY, where P < 0.05 (Fig. 1), indicative of impaired mucosal barrier function. No significant difference was observed in intestinal permeability at any of the remaining time points. With respect to brush-border enzyme activity, sucrase activity was significantly lower in the duodenum, jejunum, and ileum for treatment groups MTX, MTX + CY, and MTX + SY when compared to control rats, P < 0.05 (Fig. 2). Minor protection was achieved in the jejunum sucrase activity of rats treated with MTX + SY, which was significantly higher compared to MTX-treated control rats (P < 0.05). No intestinal protection was evident from the results of small intestinal lactase activity, where rats treated with MX + SY or MTX + CY produced similar lactase levels to that observed in MTX control rats, which was significantly elevated compared to normal controls (P < 0.05; Fig. 3).

Fig. 1
figure 1

Urinary lactulose/mannitol excretion ratio in MTX-treated controls (MTX) and MTX-treated animals gavaged with either bovine yoghurt (MTX + CY) or sheep yoghurt (MTX + SY). *MTX + CY and MTX + SY was significantly less than MTX (P < 0.025). β, Day + 6 was greater than Day + 3 for all groups (P < 0.05). δ, MTX at Day + 6 was significantly greater than baseline (P < 0.05). γ, MTX + CY rats were significantly less than MTX rats on Day + 6 (P < 0.025)

Full size image
Fig. 2
figure 2

Small intestinal sucrase activity of untreated (control), MTX-treated controls (MTX) and MTX-treated animals gavaged with either bovine yoghurt (MTX + CY) or sheep yoghurt (MTX + SY). *Sucrase activity was significantly less than control rats (P < 0.005). δ, MTX + SY sucrase activity was significantly greater than MTX rats (P < 0.03)

Full size image
Fig. 3
figure 3

Small intestinal lactase activity of untreated (control), MTX-treated controls (MTX) and MTX-treated animals gavaged with either cow’s yoghurt (MTX + CY) or sheep yoghurt (MTX + SY). *Control rats were significantly less than all treatment groups (P < 0.05)

Full size image

Discussion

This study investigated small intestinal changes after administration of yoghurts containing probiotics during MTX-induced mucositis. We concluded that sheep yoghurt containing Lactobacillus bulgaricus and Streptococcus thermophilus was capable of maintaining the mucosal barrier and improving intestinal permeability, brush border sucrase and lactase activity, and tissue architecture following MTX-induced small intestinal mucositis. Although the cows yoghurt containing Lactobacillus johnsonii was equally as effective at maintaining intestinal integrity, it was not as effective in restoring functionality as evidenced by brush border enzyme activity.

A reduction in brush border enzyme activity has been used as an indicator of damage in the small intestinal epithelium [15]. Sheep yoghurt increased jejunal sucrase activity in MTX-treated rats, but this increase was minimal when compared to jejunal sucrase activity observed in saline control rats, indicating only a partial protection from MTX-induced small intestinal damage. Interestingly, in this study it was observed that lactase activity was increased 5 days post MTX administration. Lactase concentrations are highest during early stages of development of the intestinal epithelium [16, 17]. Therefore, the elevation in lactase activity apparent following MTX and probiotic-containing yoghurts may have been related to an early stage of mucosal repair. Sucrase activity has been linked closely to enterocyte and villus repair following MTX treatment [18]. Therefore, the minimal rise in sucrase activity observed in rats receiving cow’s yoghurt may have been indicative of mucosal repair.

A recent study by Tooley et al. [18] illustrated that female dark agouti rats treated with 109 cfu/ml of Streptococcus thermophilus partially attenuated MTX-induced small intestinal mucositis, with protection most profound in the proximal jejunum. Interestingly, a decrease by a factor of ten in the cfu of Streptococcus thermophilus yielded no protection to the small intestine, and specifically the proximal jejunum. These findings were confirmed with the 13C-sucrose breath test as a biomarker of mucosal damage and histological parameters. These findings suggest that higher doses of the studied probiotics in cow and sheep yoghurt may have led to enhanced repair and recovery following chemotherapy.

This study provides preliminary evidence that sheep yoghurt containing Lactobacillus bulcaricus and Streptococcus thermophilus may be a useful natural prophylactic treatment to minimize the deleterious effects of chemotherapy-induced intestinal mucositis. Further studies are required to define the optimal probiotic species, or combination, required to produce a readily available efficacious probiotic yoghurt formulation for patients with cancer undergoing chemotherapy regimens.