In the current ruminant production systems, large amounts of high-concentrate (HC) diets are fed to goats or dairy cows to maximise the yield of meats and milk(Reference Kennelly, Robinson and Khorasani1). Although these short-term feeding regimens might be effective to support energy requirements, the excessive amounts of non-structural fermentable carbohydrates lead to a lot of negative influences as the accumulation of SCFA and microbial translocation in rumen and hindgut(Reference Gaebel, Bell and Martens2,Reference Plaizier, Krause and Gozho3) . Moreover, the decline of pH in the rumen and gastrointestinal tract likely results in the lysis of gram-negative bacteria and the release of lipopolysaccharide (LPS) that translocate into the blood circulation system, which enhances mucosal damage of the hindgut(Reference Khafipour, Krause and Plaizier4). Under normal physiological conditions, only a small amount of LPS penetrates the gastrointestinal epithelial barrier by a special immune mechanism as a consequence of an endocytotic pathway(Reference Andersen, Hesselholt and Jarlov5,Reference Drewe, Beglinger and Fricker6) . In contrast, under physiological stresses caused by endotoxin and cytokines, the barrier properties of tight junctions (TJ) can be provoked, causing increased epithelial permeability and shift of LPS(Reference Gareau, Silva and Perdue7).
Accumulation evidence shows that the large intestine is invaded easily by gut LPS translocation in ruminants. Furthermore, not only multiple approaches of transcellular transport but the monolayer intestinal mucosa is also more prone to damage by low pH resulting in an increase in the paracellular permeability compared with the rumen epithelial structure(Reference Khiaosa-Ard and Zebeli8). Ruminants can experience hindgut acidosis if the rumen cannot maintain physiological degradability, increasing the flow rate of substrates for gastrointestinal (caecum and colon) microbial fermentation(Reference Gressley, Hall and Armentano9). Bertok (2004) testified that bile acids cause degradation of LPS in the small intestine(Reference Bertok10). LPS is more easily degraded in the small intestine due to the detoxification by intestinal alkaline phosphatase and antimicrobial peptide(Reference Ribeiro, Xu and Klein11–Reference Hersoug, Møller and Loft13). Moreover, the animal’s colon or caecum has a higher permeability to macromolecules than the small intestine, probably connected with a peculiar paracellular pathway(Reference McKie, Zammit and Naftalin14). Thus, the large intestine is more LPS susceptible than the small intestine. After translocation from the rumen to the hindgut, LPS interacts with LPS-binding protein (LBP), which markedly enhances LPS activity and augments the production of pro-inflammatory cytokines(Reference Tobias, Mathison and Ulevitch15). Subsequently, the LPS–LBP conjugates were transferred to cell surfaces and interact with toll-like receptors 4 (TLR4), and then NF-κB were activated, which resulting in a range of cascade immune responses(Reference Guha and Mackman16).
As a water-soluble vitamin, thiamine plays an important role not only in energy metabolism but also in the regulation of barrier function(Reference Bubber, Ke and Gibson17,Reference Ma, Zhang and Zhang18) . Studies have shown the fact that an increase of dietary non-fibre carbohydrate levels (HC diet) can decrease the content of thiamine in the rumen and may cause thiamine deficiency(Reference Schwab, Schwab and Shaver19,Reference Pan, Nan and Yang20) . Feeding on long-term HC diets caused a low ruminal pH and microbial activity(Reference Mao, Huo and Liu21) and may, therefore, affect thiamine production(Reference Dabak and Gul22). Pan et al. (2017) revealed that excessive feeding of HC diets caused disturbance of bacterial community associated with thiamine metabolism, resulting in thiamine deficiency(Reference Pan, Xue and Nan23). Our prior studies have confirmed that exogenous thiamine supplementation not only relieved inflammation in ruminal epithelium via regulating the NF-KB pathway but promoted epithelial development in goats(Reference Ma, Zhang and Zhang18,Reference Ma, Wang and Elmhadi24) . However, the literature revealed no data regarding thiamine regulation of inflammation and intestinal integrity in the colon during a long period of feeding HC diet. Thus, we hypothesised that thiamine supplementation could protect colonic integrity via modulating mucosal inflammation injury. The aim of this study focuses on the evaluation of the anti-inflammatory properties of thiamine and enzyme activities related to thiamine function in colonic mucosa during long-term HC diet feeding.
Methods
Ethics statement
Animal care and procedures were under the Chinese guidelines for animal welfare and approved by the Guidelines for the Ethics Committee of Yangzhou University (SXXY 2015-0054).
Animals and experimental design
The experimental sample size was determined according to the prior studies on the function of thiamine during a long period of HC feeding(Reference Ma, Zhang and Zhang18,Reference Ma, Wang and Elmhadi24,Reference Ma, Wang and Zhang25) . We expanded the sample size in the study to obtain more valid information. Twenty-four female Boer goats (body weight = 35·62 (sem 2·4) kg, body condition score = 3·15 (sem 0·14), where 0 = emaciated and 5 = obese)(Reference Russel, Doney and Gunn26) bought from Lingtang, a village in Jiangsu province, were used in this study. During the 2-week adaptation period, all trial animals were housed in separate pens (1·5 × 1·5 m) and receiving ad libitum the same diet (concentrate:forage = 30:70). At the end of the dietary adaptation period, goats (dry period) in parity 1 or 2 were randomly allocated to three groups (complete randomised design). The first group were fed a low-concentrate diet comprising 70 % forage and 30 % mixed concentrate (CON, n 8), the second group were offered a HC diet containing 70 % mixed concentrate and 30 % forage (HC, n 8), while the last group received a a HC diet supplemented with 200 mg of thiamine/kg DMI (HCT, n 8). Based on NRC standard (2007)(27), the diets components and nutrient compositions were formulated to meet the nutrient and energy requirements of 35-kg Boer goats (online supplementary Table S1). The feedstuff thiamine content was determined using thiamine measurement kits (TSZ Biological Trade Co. Ltd), according to the manufacturer protocols. The dietary supplementation dose of thiamine was confirmed according to our previous research(Reference Zhang, Peng and Zhao28). For daily thiamine supplementation method, 100 g of concentrate diet mixed with thiamine (half of daily thiamine total intake) (T283819, thiamine hydrochloride, purity≥99 %, Aladdin) was firstly fed to each animal. Then, the remaining ration was fed until the goats had consumed the entire 100 g of concentrate to ensure full intake of thiamine. The goats were fed the respective diets twice daily at 07.00 h and 18.00 h for 12 weeks and had free access to freshwater during the experimental period. All goats were dewormed with 0·2 mg of ivermectin/kg (I8411, Solarbio) of BW for controlling gastrointestinal parasites once a week during the entire experiment. Moreover, diets and orts of goats were tracked and gathered daily. At the end of the experimental period, all goats were euthanised with an intravenous injection of sodium pentobarbital (200 mg/kg BM, provided by Experiment Management Center of Yangzhou University).
Sample collection and analysis
On the last day of weeks 10, 11 and 12, approximately 30 ml of representative rumen fluid sample was collected via the oral cannula at 0, 1, 2, 4, 6, 8, 10 and 12 h after feeding to measure the pH value by a high-accuracy portable pH metre (Testo 205, Testo AG) as described by Gozho et al. (2005)(Reference Gozho, Plaizier and Krause29). The pH value of each sample was measured three times to avoid anthropogenic errors. After death, approximately 50 ml of colonic digesta from the proximal colon was collected, following by the immediate determination of pH values using a portable pH metre (Testo 205, Testo AG). Colonic digesta samples were divided into three portions. One portion was mixed with an equal amount of physiological saline (0·90 % wt/vol of NaCl), immediately centrifuged at 10 000 × g for 15 min at 4 °C, and the supernatants were taken and stored at −20°C until subsequent analyses for VFA according to previous studies(Reference Ma, Zhang and Zhang18,Reference Pan, Xue and Nan23,Reference Ma, Wang and Zhang25,Reference Zhang, Peng and Zhao28,Reference Li, Khafipour and Krause30) . The second portion of each of the colon samples was transferred into pyrogen-free tubes and processed for LPS concentration analyses using Rogers et al. (1985) method, and a lower measurement limit of LPS as 0·1 endotoxin units/ml(Reference Rogers, Moore and Cohen31). The last portion of the colon samples was used to determine the concentration of thiamine and lactate using a similar method described by Ma et al. (2021b)(Reference Ma, Zhang and Zhang18). Within 5 min after slaughtering, representative colonic epithelium fragments were collected from the same position from each animal and the muscular layer was stripped using blunt dissection. Then immediately, they were washed three times in precooled PBS buffer, followed by snap-freezing in liquid N2 and preservation under −80 °C until further analysis. In addition, 2 cm of the colonic fragments were fixed in 4 % paraformaldehyde (Sigma) for histomorphometric microscopic analysis.
Histological analysis
Samples of the intestinal wall of the colonic mucosa were collected for histological observation, fixing in 4 % formaldehyde buffered solution (Beyotime Institute of Biotechnology), embedded in paraffin and then sectioned. Specimens were measured for injury after haematoxylin–eosin staining as described by Yue et al. (2012)(Reference Yue, Ma and Zhao32). A scoring criterion was adopted for determining histological damage as described previously(Reference Wu, Vallance and Boyer33).
Immunohistochemistry of colonictight junction analysis
The expression and distribution of the intestinal wall of the colonic mucosa TJ proteins were conducted using immunohistochemistry. The following antibodies were used in the immunohistochemistry assay: claudin-1 (diluted at 1:100, ab15098; Abcam), claudin-4 (diluted at 1:100, ab210796; Abcam), occludin (diluted at 1:100, ab167161; Abcam) and ZO-1 (diluted at 1:100, ab214228; Abcam). Histological sections were prepared by embedding in paraffin, based on the method of histological analysis. Then sections were incubated with antibodies and dyed with hematein for light microscope observation. Image-Pro Plus v.6.0 software (Media Cybernetics) was used to choose the same brown colour as the consistent criterion for estimating all photos. Each photo was measured to acquire the cumulative optical density of each image.
Enzyme activities assay of pyruvate dehydrogenase, α-ketoglutarate dehydrogenase and lactate dehydrogenase in colon tissue
The pyruvate dehydrogenase (PDH), α-ketoglutarate dehydrogenase (α-KGDH) and lactate dehydrogenase (LDH) enzyme activities in the colonic tissue were measured using commercial high-precision kits (Jiehuigao biological technology Co., Ltd), according to the supplier protocols. The absorbance of all samples was read at 605 nm and 340 nm, using a multi-function microplate reader (FLx800, Bio-Tek Instruments, Inc.). All trial tissue samples were normalised to total protein concentration by the BCA kits (Beyotime Biotechnology Institute).
Assay of matrix metalloproteinase activity in colon tissue
The matrix metalloproteinase (MMP)-2 and MMP-9 activities were determined using gelatin zymography. Total protein was extracted from 100 mg of a ground colon tissue sample based on a previous method(Reference Lee, Choi and Na34). Briefly, 60 µg of total protein was utilised to electrophoresis with a 12 % polyacrylamide gel containing 1 % gelatin for 90 min. Subsequently, the gel was incubated with a renaturation buffer and a developing buffer at 37 °C, respectively. Then, the gel was stained with 0·5 % Coomassie blue R-250 for the following analysis. The Quantity One software (Bio-Rad Laboratories Inc.) was used to analyse the relative zymographic intensity of each sample(Reference Zhang, Peng and Zhao28). All experiments were repetitively conducted thrice, and all values were expressed as fold changes compared with the CON group.
Caspase-3 and caspase-8 activities assay in colon tissue
Caspase-3 and caspase-8 enzyme activities in the colonic mucosa tissue were determined using a caspase activity Assay Kit (Nanjing Jiancheng Bioengineering Institute) according to the manufacturer instructions. In brief, 7-amino-4-trifluoromethylcoumarin (AFC) was used to label the caspase molecule in each sample, then fluorescence value were detected by an automatic fluorescence microplate reader (FLx800, Bio-Tek Instruments, Inc.) at 405 nm(Reference Tao, Duanmu and Dong35). All values were expressed as fold changes relative to the CON group.
Total RNA extraction and real-time quantitative PCR
Total RNA for all samples were extracted from approximately 100 mg of colonic tissues by using total RNAiso Plus Kit V2 (Vazyme Biotech Co., Ltd) according to the manufacturer’s protocols. Total RNA was quantified using a Nanodrop (NC2000, Thermo Fisher), and RNA integrity number (RIN) was measured on a Bioanalyzer (Bioanalyzer2100, Agilent). The experiment of electrophoresis with a 1·2 % (wt/vol) denaturing agarose gel was conducted to measure the integrity of RNA, and clear 28S and 18S RNA bands were confirmed in each sample before being used in the additional trials. Subsequently, 2 µg of total RNA was reverse-transcribed into cDNA using a PrimeScript RT reagent Kit (Vazyme Biotech Co., Ltd) based on the manufacturer’s guidance. The primers information of reference and target genes were displayed in Table 1 (Sangon Biotech). Primers were designed to span exon-exon junctions, where possible and were assessed for amplification efficiency (calculated as −1 + 10(−1/slope) × 100) using a serial 10-fold dilution of pooled cDNA using a serial dilution of pooled cDNA. Melt curves for primers were checked to verify the presence of a single product and avoid dimer formation. Based on Gao et al. (2013)’s method in the screening of reference genes, RefFinder (http://www.leonxie.com/referencegene.php), including Normfinder, geNorm and the comparative ΔCT method, was used to select the first-rank reference gene (ACTB, glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and hypoxanthine phosphoribosyltransferase (HPRT)) by determining the candidate genes’ ranking(Reference Gao, Zhang and Wang36). A lower gene geomean of ranking value means a higher expression stability. The order of gene expression stability, from most to least stable, was HPRT1, ACTB and GAPDH. Therefore, GAPDH was chosen as the reference gene to normalise mRNA expression. The quantitative reverse transcription PCR experiments were performed as described by Ma et al. (2021b) using SYBR green plus reagent kit (Vazyme Biotech Co., Ltd)(Reference Ma, Wang and Elmhadi24). Final reference genes were used to standardise target gene abundance using the 2−ΔΔCt method (Livak and Schmittgen, 2001)(Reference Livak and Schmittgen37). All reactions were performed in triplicate.
Primers for quantitative real-time PCR
CXCL, C-X-C motif chemokine ligand; MMP-9, matrix metalloproteinase 9; Bcl-2, B-cell lymphoma/leukaemia 2; Bax, Bcl-2-associated X protein; ZO-1, zonula occludens-1; SLC25A2, solute carrier family 19, member 2; SLC19A3, solute carrier family 19; SLC25A19, mitochondrial thiamine pyrophosphate transporter; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; HPRT1, hypoxanthine phosphoribosyltransferase 1; F, forward; R, reverse.
* Efficiency = –1 + 10(–1/slope) × 100.
Western blotting analysis
The ground colonic tissue samples (about 100 mg) were pyrolysed with RIPA lysis and extraction buffer (Solabio Biotech Co., Ltd) for 10 min. Colonic lysates were centrifuged at 12 000 × g at 4 °C for 30 min and then the supernatant was collected. Total protein concentrations were measured in triplicate using an Enhanced BCA Protein Detection Kit (Vazyme Biotech Co., Ltd). The protein samples were denatured at 100 °C for 5 min, and then 60 mg of protein was run on a 12 % SDS-PAGE gel electrophoresis at 60 V for 30 min, then 100 V for 2 h and then transferred to a nitrocellulose membrane (Millipore). These membranes were blocked with 5 % BSA (Beyotime Biotechnology) for 1 h at 25 °C, and then incubated overnight at 4 °C with primary antibodies, including anti-ZO-1 (ab214228, diluted at 1:1000, Abcam), anti-occludin (ab167161, diluted at 1:1000, Abcam), anti-claudin-1 (ab15098, diluted at 1:1000, Abcam) and anti-β-actin (diluted at 1:1500, Santa Cruz). β-Actin was used as a loading control for standardisation of target protein by Western blotting. After washing step with Tris-Buffered Saline Tween (TBST), membranes were incubated with the secondary antibody horseradish peroxidase (HRP-conjugated goat anti-rabbit IgG, diluted at 1:1000, Beyotime) for 45 min at room temperature. Then, enhanced chemiluminescence Plus kit (Vazyme Biotech Co., Ltd) was used to visualise the protein bands. The signals were captured using a LAS4000 imaging system (GE Healthcare Bio-Sciences AB) and converted to grey values by Image-Pro Plus 6.0 (Media Cybernetics Inc.) software. The protein expression was showed as fold change relative to the average value of the CON(Reference Hnasko and Hnasko38).
Statistical analysis
Shapiro–Wilk’s and Levene’s tests were used to test for normality and homogeneity of variances, respectively. The general linear model repeated measures was repetitively performed to analyse ruminal pH by IBM SPSS 21.0 statistics (SPSS Inc.). For each goat, pH value resluts for the last day of consecutive weeks 10, 11 and 12 were averaged before analysis(Reference Zhang, Meng and Gao39). Other data were analysed using one-way ANOVA with Dunnett’s post-test(Reference Zhang, Meng and Gao39,Reference Pan, Yang and Beckers40) . The fixed effect of parity, which was included in the original models and was not significant (P > 0·05), was excluded from the final model, in which only treatment was the fixed effect. The data were considered statistically significant at P < 0·05.
Results
Changes in the ruminal pH
Compared with the CON group, the HC group had (P < 0·05) lower ruminal pH value (P < 0·05; Fig. 1), and pH was below 6·0 within 12 h after feeding. The HCT group had a significantly higher pH value than the HC groups from 0 to 12 h (P < 0·05) after the feeding, and ruminal pH was above 6·0 after 4 h of feeding.
Ruminal pH values for goats fed the low-concentrate diet (CON), high-concentrate diet (HC) and HC diet supplemented with 200 mg of thiamine/kg of DMI (HCT). Data are shown as means values with their standard errors. Means with different letters (a–c) are significantly different (P < 0·05) at 0–12 h.
SCFA, lipopolysaccharide, thiamine and lactate concentrations and pH in colonic digesta
Compared with the CON group, the HC group showed a markedly decrease (Table 2; P < 0·05) in pH value and thiamine concentration of colonic digesta, but a markedly increase (P < 0·05) in free LPS, lactate, acetate, propionate, butyrate and total SCFA concentrations. On the contrary, the pH value and thiamine concentration of colonic digesta were significantly increased (P < 0·05) in the HCT group compared with the HC goats, but free LPS, lactate, acetate, propionate, butyrate and total SCFA concentrations were significantly decreased (P < 0·05).
Effects of dietary thiamine supplementation in colonic digesta parameters, contents of thiamine and lipopolysaccharide of goats fed with a high-concentrate diet at the time of slaughter
(means values with their standard errors)*
CON, control; HC, high-concentrate diet; HC, high-concentrate diet supplemented with 200 mg of thiamine/kg of DMI; LPS, lipopolysaccharide; EU, endotoxin unit; TSCFA, total SCFA.
* Data are expressed as means values with their standard errors. Mean values with different superscripts are significantly as determined by Tukey’s test (P < 0 05).
n 8 goats/treatment.
Morphological analysis in colon tissue
Haematoxylin–eosin staining showed desquamation and severe cellular damage that was observed in the colon epithelium of the HC group, whereas the CON and HCT goats exhibited structural integrity of the epithelial cell morphology (Fig. 2(a)–(c)). Meanwhile, the colonic epithelial damage score was significantly higher in the HC group compared with the CON and HCT group (P < 0·05).
Representative histology sections with haematoxylin–eosin staining of the colon in goats from the low-concentrate diet (CON), high-concentrate diet (HC) and high-concentrate diet with thiamine (HCT) group. (a) The colonic mucosa epithelium in the CON group was intact and showed no disruption (n 8; scale bar = 50 μm). (b) In the HC treatment, desquamation and severe cellular damage were observed in the colonic mucosa epithelium (n 8; scale bar = 50 μm). (c) The colonic mucosa epithelium showed slight damage in the HCT treatment (n 8; scale bar = 50 μm). (d) Colonic epithelial injury score. Arrow indicates the degree of injury in the colonic mucosa epithelium. The mean values in columns without a common superscript letter differ (P < 0·05).
Matrix metalloproteinase-, caspase- and thiamine-dependent enzyme activities in the colonic mucosa tissues
As shown in Table 3, the HC goats showed significantly higher levels of MMP-2, MMP-9, caspase-3, caspase-8 and LDH enzyme activities (P < 0·05) and lower levels of PDH and α-KGDH enzyme activities (P < 0·05) in the colonic mucosal tissue than that in the CON goats. Moreover, compared with the HC goats, the HCT group showed a markedly increase (P < 0·05) in PDH and α-KGDH enzyme activities and a markedly decrease (P < 0·05) in MMP-2, MMP-9, caspase-3, caspase-8 and LDH enzyme activities.
Effects of dietary thiamine supplementation on enzyme activities in the colonic mucosa of goats fed with a high-concentrate diet
(means values with their standard errors)*
CON, control; HC, high-concentrate diet; HCT, high-concentrate diet supplemented with 200 mg of thiamine/kg of DMI; MMP-2, matrix metalloproteinase 2; PDH, pyruvate dehydrogenase; α-KGDH, α-ketoglutarate dehydrogenase; LDH, lactate dehydrogenase.
* Data are expressed as means values with their standard errors. Mean values with different superscripts are significantly as determined by Tukey’s test (P < 0 05).
n 8 goats/treatment.
Immunohistochemistry of colonic epithelium tight junction
To investigate the differences in colonic epithelium barrier with different diet treatments, we examined the expression and distribution of TJ (zonula occludens-1, occludin, claudin-1 and claudin-4). The HC diet treatment showed significantly downregulated protein expression levels (P < 0·05; Fig. 3) in occludens-1 (ZO-1), occludin, claudin-1 and claudin-4 that in the CON goats, whereas the HCT had a significant increase (P < 0·05) in protein expression levels of ZO-1, occludin, claudin-1 and claudin-4.
Effects of dietary thiamine supplementation on the expression and distribution of tight junction proteins in colonic mucosa epithelium. (a) Immunohistochemistry results of occludin, claudin-1, claudin-4 and ZO-1 in colonic mucosa epithelium. (b) Positive rate analysis. All scale bars show 50 µm (n 8 goats/treatment). CON, low-concentrate diet; HC, high-concentrate diet; HCT, high-concentrate diet supplemented with 200 mg of thiamine/kg of DM intake. The mean values in columns without a common superscript letter differ (P < 0·05).
Gene expression in colon tissue
The relative mRNA expression of SLC19A2, SLC19A3, SLC25A19, Bcl-2, occludin, claudin-1, claudin-4 and ZO-1 were lower (P < 0·05; Table 4) and NF-κB, IL-1β, IL-6, IL-10, CXCL-10, CXCL-13, MMP-9, MMP-13 and Bax were higher (P < 0·05) in the HC group than that in the CON group. Compared with the HC treatment, dietary thiamine supplementation (HCT group) reversed this change for the expression of the above genes (P < 0·05).
Effect of dietary thiamine on the expression of genes in the colonic mucosal tissue of goats
(means values with their standard errors)*
CON, control; HC, high-concentrate diet; HCT, high-concentrate diet supplemented with 200 mg of thiamine/kg of DMI; SLC25A2, solute carrier family 19, member 2; SLC19A3, solute carrier family 19; SLC25A19, mitochondrial thiamine pyrophosphate transporter; CXCL, C-X-C motif chemokine ligand; MMP, matrix metalloproteinase; Bcl-2, B-cell lymphoma/leukaemia 2; Bax, Bcl-2 associated X protein; ZO-1, zonula occludens-1.
* Data are expressed as means values with their standard errors. Mean values with different superscripts are significantly as determined by Tukey’s test (P < 0 05).
n 8 goats/treatment.
Protein expression in colon tissue
Western blotting analysis showed significantly higher p65 protein levels (P < 0·05; Fig. 4) and markedly lower claudin-1 protein levels (P < 0·05) than those in the CON group, whereas the expression levels of p65 were lower (P < 0·05) and claudin-1 were higher (P < 0·05) in the HCT group compared with the HC group.
Effects of dietary thiamine supplementation on barrier function and inflammatory key protein expression. (a) Western blot analysis of p65. (b) Western blot analysis of claudin-1. Results of protein levels expressed as arbitrary units relative to GAPDH protein, fold change of (c) p65 and (d) claudin-1 protein content in the colonic mucosa. CON, low-concentrate diet; HC, high-concentrate diet; HCT, high-concentrate diet supplemented with 200 mg of thiamine/kg of DM intake; GAPDH, glyceraldehyde-3-phosphate dehydrogenase. Values are mean ± sem. The mean values in columns without a common superscript letter differ (P < 0·05).
Discussion
This study was designed to investigate the potential regulatory effects of thiamine in simulated prolonged feeding of excess HC diets (extreme conditions); thus, we chose a diet with a higher concentrate to forage ratio (concentrate:forage = 30:70). The study can provide some guidances for intensive production in ruminants through the overfeeding of HC diets. Feeding the HC diets to ruminants causes a high risk to damage the hindgut epithelial structure(Reference Chang, Ma and Zhang41,Reference Samo, Malhi and Kachiwal42) . Prior studies demonstrated that dietary thiamine supplementation alleviates the HC diets-induced oxidative stress, apoptosis and promotes rumen epithelial development in goats(Reference Ma, Zhang and Zhang18,Reference Ma, Wang and Elmhadi24) . Therefore, we focused on the regulation role of thiamine on colonic integrity and mucosal inflammation injury. Our current results showed that long-term feeding of the HC diets can cause a decrease in the pH values, the SCFA and thiamine concentration and an increase in LPS concentration in the colon, while thiamine supplementation reversed these changes. These data revealed that thiamine addition improved microbial fermentation and the hindgut environment. Plaizier et al. (2012) uncovered that the fluid flows out of the rumen into the omasum and subsequently into the hindgut contains LPS(Reference Plaizier, Khafipour and Li43). Therefore, it is expected that part of the LPS found in the large intestine is derived from gram-negative bacteria of the rumen. This decline in ruminal pH (pH value has been below 5·6 for a long time) can cause the lysis of gram-negative bacteria and release LPS that translocate into the bloodstream and hindgut(Reference Khafipour, Krause and Plaizier4). In the current study, dietary thiamine supplementation elevated pH value compared with the HC groups from 0 to 12 h after feeding, and ruminal pH was above 6·0 after 4 h of feeding. Thiamine is essential for the growth of most bacteria strains of Ruminococcus, such as Ruminococcus albus (cellulolytic bacteria) and Ruminococcus flavefasciens (cellulolytic bacteria)(Reference Pan, Xue and Nan23,Reference Bryant and Robinson44) . Moreover, Wetzels et al. (2016) demonstrated that exogenous thiamine supplementation facilitates higher pH, which helped in the proliferation of Succinivibrio (Reference Wetzels, Mann and Metzler-Zebeli45). Although the exact mechanism of inhibition of HC diet-induced LPS production via thiamine supplementation is unclear, these data may show more valid evidence for us to study the protective effect of thiamine on colonic integrity.
The thiamine concentration in colonic digesta under different dietary concentrate diet levels has rarely been reported. In the present study, the thiamine concentrations in colonic digesta of three groups were 5·47, 1·13 and 2·56 μg/l, respectively. Dabak and Gul et al. (2004) demonstrated that long-term HC diet feeding decreased thiamine concentration in the rumen due to the reduction in the microbial thiamine synthesis and the degradation by thiaminase under lower pH(Reference Dabak and Gul22). Thus, we supposed that the decrease of thiamine concentration in the colon was associated with thiaminase and microbial community. The absorption of thiamine is mediated by carriers(Reference Zhao and Goldman46–Reference Zhu, Fang and Subramanian47). In this study, the HC diet reduced the gene expression of SLC19A2, SLC19A3 and SLC25A19 in the colonic epithelium. The transport carriers of thiamine were pH-sensitive(Reference Arun, Edmondnson and Goldman48), and the LPS and pro-inflammatory cytokines could also downregulate the expressions of SLC19A2 and SLC19A3(Reference Zhu, Fang and Subramanian47). The current research observed that dietary thiamine supplementation increased the gene expression of transport carriers of thiamine. Our previous results have demonstrated that excessive feeding of the HC diet leading to a decrease in thiamine-related enzyme activities such as PDH, α-KGDH and transketolase in rumen epithelium tissues(Reference Ma, Wang and Elmhadi24). In order to investigate whether a similar effect would occur in the colonic mucosa under the stimulation of LPS, we measured the PDH, α-KGDH and LDH enzyme activities in the colonic mucosal epithelium and obtained the same results as the previous study in rumen epithelium, while the supplementation of exogenous thiamine reverse the change. The one possible reason was that exogenous supplementation of thiamine maintains the colon homoeostasis by higher pH and lower LPS, which stimulated the expression of thiamine transport carriers in colonic epithelium. In addition, supplementary thiamine promoted carbohydrate metabolism and production of energy(Reference Falder, Silla and Phillips49), while energy is also a crucial factor for the active transport of thiamine(Reference Hoyumpa, Nichols and Schenke50).
In intensive production, injury of epithelial barrier function in the hindgut is repeatedly diagnosed in ruminants which are continually fed a HC diet. In contrast to the multilayered structure cellular structure of the ruminal epithelium, the colon only is composed of a monolayer structure(Reference Graham and Simmons51). Thus, the integrity of the colon is more easily attacked by a high concentration of LPS(Reference Li, Khafipour and Krause52). Besides, the acidity in the gastrointestinal tract also plays a vital role in maintaining the integrity of the epithelial barrier(Reference Tao, Duanmu and Dong53). The histological analysis showed that dietary thiamine supplementation during long-term the HC diets feeding protects the colonic barrier function, which may be explained that the thiamine supplementation decreases the accumulation of SCFA, lactate and LPS in the colon and weakens the intense stimulation for the colonic epithelial barrier.
As highly dynamic structures, TJ proteins dedicate to the maintenance of physical barrier function(Reference Ma, Zhang and Zhang18). Moreover, TJ also play a crucial role in promoting cell–cell interactions and stabilising the transcellular pathways(Reference Steed, Balda and Matter54). Our immunohistochemistry data showed that dietary thiamine supplementation promoted the colon epithelial barrier integrity in comparison with the HC group as confirmed by the markedly changed expression and distribution of TJ in colon epithelium (occludin, claudin-1, claudin-4 and ZO-1). The expression of the TJ gene and protein were also in accord with the data results of immunohistochemistry. Several stimulating factors including lower pH and cumulative toxic compounds from bacterial metabolism have been reported to involve in the process of TJ inhibition caused by feeding the HC diet(Reference Lan, Lagadic-Gossmann and Lemaire55). Occludin protein is responsible for the transport of the macromolecule(Reference Al-Sadi, Khatib and Guo56), with ZO-1 as an organising component of the TJ link occludin to the cortical actin cytoskeleton(Reference Fanning, Jameson and Jesaitis57). Thus, the coordination of occludin and ZO-1 is of profound significance for the integrity of the epithelium and the transport of nutrients. Claudins consist of several gene families, which have the role to enhance cell proliferation(Reference Tsukita, Amazaki and Katsuno58). Dietary thiamine supplementation promotes an increase in pH and a decrease in LPS levels in colon, which may be one of the reasons for the protection of the colon integrity. Although our previous studies have confirmed that thiamine can promote the development of rumen epithelium(Reference Ma, Wang and Elmhadi24), the exact action mechanism of thiamine for the promotion of TJ expression is still unclear.
The HC diet-induced LPS triggers the injury and death of intestinal mucosal epithelial cells involved in apoptotic signal transduction pathway, which impaired mucosal barrier function, leading to epithelial leakage(Reference Droin and Green59). The caspases family play a vital role during the initiation and effector stages of cell apoptotic among the multiple molecules involved in the apoptotic programme(Reference Kuida, Zheng and Na60). By regulation of apoptotic components, cytochrome C forms complex, then the apoptosome which via caspase 9 triggers executioner caspase 3 and eventually induces cell death(Reference Bender, Fitzgerald and Tait61). The caspase 8 regulates and controls ligand binding triggers apoptosis in the extrinsic pathway(Reference Samo, Malhi and Kachiwal42). The Bcl-2 families of proteins, including Bax (pro-apoptotic) and Bcl-2 (anti-apoptotic), are vital regulators in the early phase of the apoptotic pathway. The change in the expression of the Bcl2/Bax ratio would affect dramatically the apoptotic rate and would eventually alter the phenotypic status of the cell(Reference Droin and Green59). Samo et al. (2020) revealed that long-term HC diet feeding can induce epithelial injury and apoptosis in the colon. In the present study(Reference Samo, Malhi and Kachiwal42), Bax mRNA levels were significantly decreased, whereas the Bcl-2 mRNA level and the Bcl2/bax ratio were significantly increased in HCT goats compared with that in the HC goats. Moreover, the results of enzyme activity analysis in colonic mucosa showed that the HCT group had a markedly decrease in caspase-3 and caspase-8 enzyme activity compared with the only HC diet feeding, which suggests that thiamine contributes to the attenuation of the HC diet-induced apoptosis through the regulation of the Bax/Bcl-2 ratio and the caspase pathway.
Alleviation of inflammation in the colonic mucosa can improve the integrity of the colon. Zhang et al. (2020) confirmed that the effects of thiamine supplementation on local inflammation in the ruminal epithelium could be associated with a reduced ruminal LPS content(Reference Zhang, Peng and Zhao28). In addition, SCFA have also been demonstrated to involve leucocyte migration and modulate the production of certain cytokines such as TNF-α and IL-6(Reference Vinolo, Rodrigues and Nachbar62). Therefore, we analysed the mRNA expression levels of NF-κB and other related inflammatory genes, and the results suggested that the gene expression levels of NF-κB, IL-1β, IL-6 and IL-10 were lower in the HCT treatment than in the HC goats. Western blot data showed the decline of NF-κB-associated proteins p65 in HCT goats. Long-term exposure of rumen epithelial cells to LPS significantly upregulates the expression of chemokines and induces acute inflammatory responses(Reference Kent-Dennis, Aschenbach and Griebel63). These chemokines are crucial chemo-attractants for the recruitment of leucocytes during inflammatory responses. Thus, the mRNA expression levels of CXCL-10 and CXCL-13 were also lower in the HCT goats than in HC group, though the decrease of CXCL-10 mRNA expression was not significant. These results revealed that thiamine regulates the expression of inflammatory cytokines and chemokines in the colonic mucosa of goats, which may be attributed to the inhibitory effect of thiamine-supplemented HC diet compared with long-term HC diet to induce inflammation. Prior research has also suggested that NF-κB signal pathway can damage barrier function by motivating the release of inflammatory factors(Reference Fang, Chen and Hu64), while the fact has been confirmed that thiamine alleviates inflammation and protects barrier function by regulating the NF-κB/TLR4 pathway(Reference Ma, Zhang and Zhang18,Reference Zhang, Peng and Zhao28) .
In this study, we also measured the expression of MMP2, MMP9 and MMP13 related to defence functions in the gastrointestinal tract. Thiamine caused a significant down-regulation in mRNA expression of MMP-9 and MMP-13 in colonic mucosa epithelium during long-term HC diet feeding. Furthermore, the enzyme activities of MMP2 and MMP9 were assessed using gelatin zymography, and our results also demonstrated that dietary thiamine supplementation decreased the activation of MMP2 and MMP9 enzymes. The TJ along with the extracellular matrix are the dominating substrates for MMP(Reference Kruger, Saffarzadeh and Weber65). Furthermore, Wu and Schmid-Schonbein (2011) found that NF-κB involves in the activation of MMP(Reference Wu and Schmid-Schonbein66). As a result, thiamine supplementation can downregulate the expression of MMP that may be associated with alleviation of inflammation.
In general, the results from this study demonstrate that colonic inflammation will be induced in goats fed the HC diet for 12 weeks that is reflected in desquamation and severe cellular damage in colon epithelium. However, thiamine, as a water-soluble vitamin, reversed the adverse effects and decreased the cells apoptosis by suppressing the NF-κB activation and the production of inflammatory cytokines. Thus, these findings provide a new strategy that thiamine could potentially serve as a dietary supplementation to contribute to alleviating the negative effects of the trigger of the HC diet.
Acknowledgements
The authors are grateful to all the members of Hongrong Wang’s laboratory that contributed to the sample determinations. Thanks also extended to the University of Western Australia, Queen Elizabeth II Medical Centre and Xinjiang Academy of Agricultural and Reclamation Sciences for the cooperation that made this paper possible.
This work was supported by the Project of the National Natural Science Foundation of China (No. 31872988, Beijing), the Excellent Doctoral Dissertation Fund of Yangzhou University and International Academic Exchange Fund for Graduate Students of Yangzhou University.
Y. M. and H. R. W. designed the research. Y. M., C. W., X. L. G., F. Y. L. and H. Z. conducted the research. Y. M. and E. M. analysed the data. Y. M. wrote the paper. E. M. and X. L. G. revised the language of this article. Y. M. and H. R. W. had primary responsibility for the final content. All authors read and approved the final manuscript.
The authors declare no conflict of interest.
Supplementary material
For supplementary material accompanying this paper visit https://doi.org/10.1017/S0007114522000174
