The synergistic tumor growth-inhibitory effect of probiotic Lactobacillus on transgenic mouse model of pancreatic cancer treated with gemcitabine

Pancreatic cancer is one of the most lethal and chemo-resistant cancers worldwide. Growing evidence supports the theory that the gut microbiota plays an essential role in modulating the host response to anti-cancer therapy. The present study aimed to explore the effect of probiotics as an adjuvant during chemotherapy for pancreatic cancer. An LSL-KrasG12D/−-Pdx-1-Cre mouse model of pancreatic ductal adenocarcinoma (PDAC) was created to study the effects of using four-week multi-strain probiotics (Lactobacillus paracasei GMNL-133 and Lactobacillus reuteri GMNL-89) as an adjuvant therapy for controlling cancer progression. At 12 weeks of age, pancreatitis was induced in the mice by two intraperitoneal injection with caerulein (25 μg/kg 2 days apart). Over the next 4 weeks the mice were treated with intraperitoneal injections of gemcitabine in combination with the oral administration of probiotics. The pancreas was then harvested for analysis. Following caerulein treatment, the pancreases of the LSL-KrasG12D/−-Pdx-1-Cre transgenic mice exhibited more extensive pancreatic intraepithelial neoplasia (PanIN) formation. Combined treatment with gemcitabine and probiotics revealed a lower grade of PanIN formation and a decrease in the expression of vimentin and Ki-67. Mice that received gemcitabine in combination with probiotics had lower aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels. Notably, the use of high-dose probiotics alone without gemcitabine also had an inhibitory effect on PanIN changes and serum liver enzyme elevation. These findings suggest that probiotics are able to make standard chemotherapy more effective and could help improve the patient’s tolerance of chemotherapy.

Blood biochemistry and histopathology. For the histological studies, the animals were sacrificed 30 days after the beginning of treatment. After anesthetization, the mice were disinfected on their ventral side with 75% ethanol 15 . Body weight was recorded for each mouse before sacrifice and pancreas weight was recorded right after they were sacrificed. To study the biochemical effects of caerulein, gemcitabine and probiotics on blood cells and biochemistry, the blood was collected after sacrifice by cardiac puncture and immediately sent to Axel Biotechnology Inc., Taiwan for blood cell counting and biochemical analysis. Pancreas tissue was aseptically removed from the mice, and a small piece of tissue was fixed with 10% neutral buffered formalin (TONYAR Biotech. Inc. Taiwan) for 24-48 h at 4 °C. The fixed tissues were trimmed to an appropriate size, embedded in paraffin, sectioned and then stained with Hematoxylin and Eosin (H&E) 16,17 . Vimentin (Vimentin (D21H3) XP Rabbit mAb, Cell Signaling Tech. # 5741) and Ki67 (Anti-Ki67 antibody KO tested, Abcam; ab15580) expression were detected using immunohistochemical staining methods in mouse pancreatic tissue. Immunohistochemical staining was performed using the UltraVision Quanto Detection System kit (Thermo Fisher Scientific Inc., Fremont, CA, USA) according to the manufacturer's instructions 18 .
Statistical analysis. Statistical software SPSS version 10.1.3C (SPSS Inc., Chicago, IL, USA) for Windows was used for recording data and analyzing results. Student's t-test was used for the comparison of two means. Fisher's exact test for categorical data was used as appropriate. A P-value < 0.05 was considered to indicate a statistically significant difference.

Results
Creation of pancreatic cancer animal models. The creation of pancreatic cancer animal models and the treatment process are shown in Fig. 1. Pancreas histology comparisons of non-treated wild-type mice (n = 3), non-treated KC transgenic mice (n = 4), and caerulein-treated KC transgenic mice (n = 5) are shown in Fig. 2. H&E staining of pancreas sections revealed a reactive duct with enlarged nuclei (arrowhead) in KC transgenic mice (Fig. 2b,c). Vimentin-positive stains around the pancreas duct (arrow) and its surroundings were significantly found in non-treated KC and caerulein-treated mice (Fig. 2e,f) in comparison to the non-treated wildtype mice. Ki-67 staining was also distributed in the pancreas of non-treated KC and caerulein-treated KC mice (Fig. 2h,i). Histological evidence indicates that the caerulein induced KC transgenic mouse model appropriately imitates PanIN.
Effect of different dosages of gemcitabine. Figure 3 shows the pancreas histology of KC transgenic mice treated with caerulein combined with different doses of gemcitabine (20, 50, 100, or 200 mg/kg). H&E staining of the pancreatic sections revealed that caerulein-treated KC transgenic mice who also received gemcitabine developed a lower grading of PanIN lesions, compared with those who received caerulein alone. A lower expression of EMT was also shown in vimentin staining, and a lower Ki-67 expression in Ki-67 staining as well (Fig. 3). The 50 mg/kg and 100 mg/kg gemcitabine treated groups showed superior outcomes based on their histological responses.
Administration of dual-strain probiotics. After caerulein treatment, the probiotics were administered by oral gavage and combined with gemcitabine at a dosage of 20 mg/kg (n = 3), 50 mg/kg (n = 3), 100 mg/kg (n = 3) or 200 mg/kg (n = 4). The group treated with gemcitabine and probiotics (GMNL-133: GMNL-89 in a 1:1/10 ratio) showed a lower grade of PanIN lesions in the H&E staining of pancreatic sections compared with those mice only receiving gemcitabine treatment (Fig. 3). Combined treatment with gemcitabine and probiotics also led to lower vimentin and Ki-67 staining (Fig. 3). To understand if this outcome was related to the proportion of probiotic used, GMNL-133: GMNL-89 was used at a ratio of 1:1 with the simultaneous use of gemcitabine (200 mg/kg). However, pancreas histology revealed that the progression of PanIN lesions become more severe compared with when GMNL-133: GMNL-89 was administered in a 1:1/10 ratio (Fig. 4). When we used  www.nature.com/scientificreports/

Discussion
In the past two decades, research into inflammation and the pathogenesis of cancer has demonstrated the tumorpromoting effect of immune cells (mainly the innate immune system) on tumor development. Studies have shown that inflammation is associated with multiple stages of carcinogenesis and can supply bioactive molecules to the inflammatory tumor microenvironment, including growth factors, survival factors, proangiogenic factors, www.nature.com/scientificreports/ induction signals for EMT, and extracellular matrix-modifying enzymes, which are known to promote cancer invasion and metastasis 19 . Conversely, we now recognize that the immune system also plays a crucial role in host immune surveillance, which inhibits carcinogenesis by identifying and destroying nascent transformed cells 20 . It was generally considered that the dual roles of inflammatory cytokines were associated with cancer suppression and progression 21 . PDAC is the most common and often lethal form of pancreatic cancer, and it is also considered to be typical of an inflammation-driven cancer 22 . As mentioned above, controlling or reversing the cancer  www.nature.com/scientificreports/ inflammatory microenvironment is likely to be one feasible approach for pancreatic cancer treatment. In both animal experiments and human trials, probiotics produce anti-inflammatory metabolites and have been proven to actively facilitate resolution of inflammation in various inflammatory and autoimmune diseases, including ulcerative colitis, rheumatoid arthritis, systemic lupus erythematosus and multiple sclerosis 23 . Research studies have confirmed that the progression of pancreatic cancer is directly related to acute pancreatitis and chronic pancreatitis 24 . To clarify the potential role of probiotics in the treatment of pancreatic cancer, we conducted this animal trial using Kras-driven GEMM of PDAC.
Interactions between intestinal cells and gut microorganisms play a crucial role in maintaining host health. In recent years, the potential anti-cancer properties of various probiotics have been evaluated in gastrointestinal neoplasms such as gastric cancer, colorectal cancer and liver cancer 25 . Several potential mechanisms involved in the anti-carcinogenic effects of probiotics have been suggested, including changes in the intestinal microbiota, immunomodulation, reduction of inflammation, inhibitory effects on carcinogenesis, improvement of nutrient absorption, production of antitumorigenic compounds, production of short chain fatty acids and inhibition of pathogen replication 23 . Accumulating evidence indicates that gut microbial dysbiosis is associated with the pathogenesis of acute and chronic pancreatitis 26 . Microcirculatory alterations, splanchnic vasoconstriction and ischemia-reperfusion damage contribute to intestinal damage and increase the permeability of the intestinal barrier, which facilitates the translocation of bacteria and toxic compounds from the gut to the blood during acute pancreatitis. This exacerbates inflammation of the pancreas and can subsequently lead to fibrosis or necrosis. In addition, microbes contributing to carcinogenesis of PDAC and modulating tumor response to therapy 27 . For PDAC, a recent study shows that human intestinal microbiome represents about 25% of the intratumoral microbiome, but is absent from the adjacent normal tissue. They also observe that the levels of CD8 + T cells and activated CD8 + T cells have significantly increased in tumors from mice received fecal microbial transplantation from long-term survivors 28 . Sethi et al. demonstrated that Th1 (IFNγ + CD4 + CD3 + ) and Tc1(IFNγ + CD8 + CD3 + ) cells in the TME were increased, and pro-tumor IL17a (IL17a + CD3 + ) and IL10 (IL10 + CD4 + CD3 + ) were decreased after gut microbiome depletion by oral antibiotics 29 . These findings suggest that there is microbial cross-talk between gut microbiome and PDAC microbiome, and influences the host immune response and tumor progression. Direct manipulation of the intestinal microbiota may be a potential strategy for PDAC treatment. A meta-analysis of six randomized controlled trials indicated that there is insufficient evidence to recommend routine use of probiotics in patients with severe acute pancreatitis 30 . However, various probiotic strains and therapeutic dosages are likely to present heterogeneous conclusions. In contrast, the findings of the current study provide supportive evidence that four weeks of Lactobacillus paracasei GMNL-133 and Lactobacillus reuteri GMNL-89 treatment are able to inhibit PDAC progression in mice regardless of whether they are used as a part of a combination therapy or as a high dose-monotherapy. However, this dose-response relationship is insufficient to extrapolate because so far only limited evidence has suggested that more probiotic bacteria yield a greater benefit 31 . It is worth including additional higher doses as endpoints in future studies.
Lactobacillus paracasei are gram-positive, facultatively heterofermentative lactic acid bacteria. They are found in normal human and animal intestinal flora and are utilized in dairy product fermentation or as a probiotic supplement 32 . Lactobacillus paracasei GMNL-133 was chosen for this trial because it has been proven to be effective in children with asthma (Th2-driven diseases). It has been shown to inhibit Th2 cytokine production and modulate the Th1/Th2 immune balance by increasing IFN-γ levels 33 . Previous research has shown that Th2 responses exhibit tumor-promoting effects in PDAC. The Th2 immune phenotype can be reversed into the pre-existing Th1 immune phenotype through the induction of IFN-γ, interleukin (IL)-12, or IL-27, which subsequently causes tumor-suppressive effects 34,35 . Konduri et al. reported that adjuvant Th1 dendritic cell vaccination combined with standard gemcitabine chemotherapy can provide durable protection against PDAC 36 . Furthermore, IL-6 is known to have pro-tumorigenic activity in the progression of PDAC 37 . Previous studies have shown that high serum IL-6 levels are linked with PDAC and are associated with a worse prognosis. In animal studies, the use of IL-6 signaling inhibitors reduced tumor growth in PDAC 37 . Supplementation with Lactobacillus reuteri GMNL-89, the second probiotic used in the current study, can cause antioxidant activity and reduced IL-6 levels in animal models 38 . Furthermore, probiotic strains that exhibit potential anti-carcinogenic properties Table 3. Comparison of biochemical values in KC transgenic mice after treatment with caerulein (Cae) and different doses of probiotics. ALT, alanine aminotransferase; AST, aspartate aminotransferase; BUN, blood urea nitrogen; Cae + Prob (1:1/10) 1/3x, caerulein(25 μg/kg) + probiotics (GMNL133:GMNL89 = 1:1/10, 1/3 times); Cae + Prob (1:1/10) 10x, caerulein (25 μg/kg) + probiotics(GMNL133: GMNL89 = 1:1/10, 10 times); Cr, creatinine; KC mice, LSL-Kras G12D/− -Pdx-1-Cre transgenic mice; Prob, probiotics. Data are presented as mean ± standard deviation (SD). a P≦0.05 differences between Cae + Prob (1:1/10) 1/3 × and Cae-+ Prob (1:1/10) 10x. b P≦0.05 differences between Non-treated and Cae + Prob (1:1/10) 1/3x. www.nature.com/scientificreports/ may directly bind with mutagenic heterocyclic amines (HCAs), thus reducing intestinal HCA absorption, and eliminating HCAs through fecal residues [39][40][41] . Some studies support the theory that multi-strain probiotics are more beneficial than single-strain probiotics 42 . In the current study we increased the proportion of GMNL-89 probiotic administered but there was no beneficial effect. From this we speculate that the intake of higher doses of GMNL-133 probiotics may be more helpful in improving PDAC. This study supports the theory that high doses of probiotics produce better inhibition of cancer progression. However, there is currently insufficient evidence to support the additive or synergistic effects between these two probiotics in vitro for PDAC. Gemcitabine is a chemotherapy agent widely used to treat various types of cancer, including non-small cell lung cancer, bladder cancer, breast cancer, and pancreatic cancer 43 . Accumulating evidence has shown that the therapeutic tolerability of gemcitabine is highly dose schedule dependent 44 . The most frequently reported side effects include bone marrow suppression, skin rash, altered liver function tests, flu-like syndromes and fever 45 . Following treatment with gemcitabine combined with probiotics, we observed a notable change in liver enzymes. Gemcitabine caused abnormalities in liver function tests for AST and ALT were reported in 67% and 68%, of cases, respectively 46 . In the present study, a general elevation in liver enzymes was noted after treatment with gemcitabine. However, low levels of liver enzymes were noted following the use of a combination probiotic therapy, particularly when a high-dose probiotic treatment was used. High-dose probiotics alone not only inhibited the progression of PDAC but also had a beneficial effect on liver enzyme levels. In addition, a slight decrease in white blood cells and an increase in platelet levels were observed in mice receiving gemcitabine treatment in the present study. Although thrombocytopenia is a well-known side effect of gemcitabine, thrombocytosis has also been reported 46 .
This study provides some important implications for the care of pancreatic cancer patients. The effects of probiotics are strain-specific, and the same probiotic in different doses may exert different modulating effects. As the effect of traditional chemotherapy is still not satisfactory despite decades of use, the addition of probiotics as an adjuvant or combination therapy should be considered as a potential treatment method. However, there were several limitations to this study. First, although GEMM PDAC is remarkably similar to human PDAC, the etiology is not exactly the same. Human PDAC commonly occurs in elderly patients and develops as a single neoplastic focus, but most of the tumors in GEMM are multi-focal 47 . Second, given the complexity of human tumor development, the mouse model may be too simplified. Third, cross-talk between host and gut microbiota is regarded as host-specific, thus the effects seen in the mouse model may not be observed in humans. Finally, disparity in experimental results may occur in different mouse models. Randomized control trials in humans are needed to determine the role of probiotics in cancer treatment.
In conclusion, the current study suggests that probiotics together with chemotherapy could help decrease the side effects of chemotherapy and simultaneously achieve better treatment outcomes.

Data availability
The data that support the findings of this study are available on request from the corresponding author. There are restrictions to the availability of mice due to material transfer agreement. www.nature.com/scientificreports/ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.