Direct Effect of Lipopolysaccharide and Histamine on Permeability Barrier of Rumen Epithelium

Background: Disruption of the ruminal epithelium barrier occurs during subacute ruminal acidosis due to low pH, hyper-osmolality, and increased concentrations of lipopolysaccharide and histamine in ruminal uid. However, the individual roles of lipopolysaccharide and histamine in the process of ruminal epithelium barriers disruption are not clear. The objective of the present investigation was to evaluate the direct effect of lipopolysaccharide and histamine on barrier function of the ruminal epithelium. Results: Compared with control (CON), lipopolysaccharide (HIS) increased the short-circuit current (Isc) (88.2%, P = 0.0022), transepithelial conductance (Gt) (29.7%, P = 0.0564) and the permeability of uorescein 5(6)-isothiocyanate (FITC) (1.04-fold, P = 0.0047) of ruminal epithelium. The apparent permeability of LPS was 1.81-fold higher than HIS (P = 0.0005). The mRNA abundance of OCLN in ruminal epithelium was decreased by HIS (1.1-fold, P = 0.0473). Conclusions: The results of the present study suggested that histamine plays a direct role in the disruption of ruminal epithelium barrier function while lipopolysaccharide without acidic pH has no signicant effect on the permeability of rumen tissues.


Background
Factors contributing to the RE barrier function include the epimural microbiome, a continuously shedding mechanical barrier of highly keratinized cells, and tight-cell junctions, desmosomes, and gap junctions in the more basal cell strata (Graham and Simmons, 2005). It is well known that ruminal conditions that occur during SARA disrupt the RE barrier function, but a complete understanding of factors in uencing the response are still limiting. Past research has reported that exposure to acidic and hyperosmotic conditions independently (Penner et al., 2010) and additively (Penner et al., 2010;Wilson et al., 2012) increase permeability to paracellular permeability markers such as mannitol. In vivo, however, increased concentrations of LPS, short chain fatty acids (SCFA), and biogenic amines (e.g. histamine) occur concurrently with decreased pH and hyperosmotic conditions with SARA (Liu et  indicating that the direct effect of pH on RE barrier function is as not strong as originally thought and that factors affecting cell function may be critical to the process. As such, research is needed to understand the contribution of multifaceted effects occurring in vivo (Aschenbach et al., 2019).
Endotoxins such as LPS and biogenic amines such as histamine (HIS) are two important microbial metabolites found in the rumen during SARA. Zhao et al. (2018) reported that SARA increased the concentration of LPS in ruminal uid and blood, and reported that LPS exposure upregulated the expression of in ammatory pathways and production of proin ammatory cytokines in RE. The ability of RE cells to initiate a pro-in ammatory response when exposed to LPS was further con rmed by Kent-Dennis et al. (2020) and warrants further investigation as in ammation may compromise epithelial barrier function.
In addition to LPS, Pilachai et al. (2012) con rmed a negative relationship between ruminal HIS concentration and ruminal pH under high concentrate diets. Aschenbach et al. (1998) rst showed that application of HIS in relevant dosages (10 and 100 µM) impairs differentiation of RE cells in culture and that when exposed to acidic pH (pH 4.5), histamine translocation across the ruminal epithelium was markedly increased . Sun et al. (2017) suggested that HIS could activate an in ammatory pathway in cultured RE cells via NF-κB. Moreover, it has been suggested that low ruminal pH and the presence of Gram Negative Bacilli (GNB) products (e.g. cell-free LPS, HIS) negatively affects the expression of RE cadherins leading to increased permeability of the RE (Zebeli and Metzler-Zebeli, 2012). However, the direct effect of the HIS and LPS on barrier function of RE in vivo or ex vivo are still not clear. Therefore, the objective of the present study was to evaluate the direct effect of LPS and HIS on the RE barrier function measured ex vivo.

Treatments and Design
Three treatments [CON (control), LPS (lipopolysaccharide), HIS (histamine)] were compared using a complete randomized design. The 3 treatments were allocated to 6 chambers with 2 technical replicates for each treatment. The experiment was repeated 8 times (n = 8), each with tissue sourced from a different steer.

Ruminal Epithelia Sample Collection and Preparation
Ruminal tissue samples for this experiment were collected from a commercial abattoir (Beijing, China). A 15 cm 2 area of ruminal tissue from the ventral sac were excised from healthy feedlot Simmental steers within 10 min after slaughter (n = 8). Tissues were then washed in a pre-heated buffer solution at 37 to 39℃ adjusted to pH 7.4 using HCl (6 mol/L) or NaOH (6 mol/L) as needed. The buffer solution consisted of CaCl 2 ×2H2O (1. and Na-butyrate (5.0 mM). The osmolarity of the buffer solution was 315.2 mOsmol/L. To maintain the activity of the tissues and enable respiration of the tissues during the transportation, the rumen tissue was stored in an insulated container with 37 to 39℃ buffer solutions continuously gassed with carbogen (95% O 2 /5% CO 2 ) until processing and mounting the tissues in the Ussing chamber (KINGTECH, China).
Once at the laboratory, the serosal and muscular layers were separated gently by hand. The maximum possible brous tissue was removed from the mucosa without injuring the tissue. Rumen tissue samples collected from a single animal were run on an individual day. At the time of slaughter, rumen whole digesta was collected and squeezed through 4 layers of muslin cloth to separate rumen uid from the solid digesta. Following straining, pH was measured and samples were collected to determine the concentration of NH 3 N and VFA as previously described (Gao et al., 2017). The results were shown in Table 1. The time from slaughter to mounting of tissues in the chambers was approximately 50 min. The isolated epithelium was cut into 6 pieces approximating 2 cm 2 and mounted between two-halves of the Ussing chamber with an exposed area of 1.27 cm 2 each, and then the clamps were assembled between the lucite chambers. A silicon washer, placed on each side of the Ussing chamber was used to prevent edge damage of the tissue. The buffer solution (pH 7.4) was re-added (10 mL to each side) and the airlift were reconnected after the clamps were assembled. Tissues were allowed to equilibrate for 15 min under open-circuit conditions, then the instrument was switched to short-circuit conditions. During the study, transepithelial short-circuit current (Isc), as a measure of net ion transport, and transepithelial conductance (Gt) were continuously recorded with the aid of an automatic computer-controlled voltage-clamp device (voltage/current clamp, VCC MC6 Plus, KINGTECH, China). Tissues were randomly assigned to 1 of 3 treatments with 2 technical replicates within from each steer. For all chambers, 8 µL FITC ( nal concentration 0.2 mM, Sigma-Aldrich China Ltd.) was added to the mucosal side as a permeability marker. Treatments included a negative control (no further addition), a treatment where HIS was added to the mucosal side to achieve a nal concentration of 20 µM (8 µL histamine, Sigma-Aldrich China Ltd.), or a treatment where LPS was added to the mucosal side to achieve a nal concentration of 1 µg/mL (10 µL from E. coli B:055, Sigma-Aldrich China Ltd.). Samples (100 µL) were taken from the mucosal side of the chambers immediately after FITC, HIS, and LPS were added to detect the initial concentrations. Additional samples were collected at 20, 40, 60, and 80 min from the serosal compartment of the chambers for detection of permeability of FITC, LPS, and HIS (2 samples per time per chamber, one for FITC detection, and the other one for LPS and histamine detection). All the samples were collected using pyrogen free tubes. The uorescence intensity of the FITC samples was measured after dilution in water 1:5 using a uorescence spectrometer (Tecan In nite 200 Pro, Tecan, Austria). The samples for LPS and HIS detection were stored at -20℃ until analysis. At 100 min, the tissues mounted in the chambers were collected and stored in liquid nitrogen until RNA extraction.

Determination of Histamine
The concentration of HIS in the serosal and mucosal side of the chambers were determined using commercial Elisa kit (MLBio, Shanghai, China). The detection was performed following the manufacturer's instructions. In principle, the wells of the microtiter plates were pre-coated with the antibody (anti-histamine). Samples, standard, and horseradish peroxidase (HRP) conjugated with antihistamine were successively added, incubated, removed, and plates were washed thoroughly. The substrate tetramethylbenzidine (TMB) was used for color development, which was converted to blue by HRP catalysis and nally to yellow by acid. There was a positive correlation between the color depth and the histamine concentration in the samples.  Where: dQ/dt = transport rate (µg/min); C 0 = initial concentration in the donor chamber (µg/ml); A = exposed surface area of the membrane (cm 2 ).
The data were subjected to statistical analysis using MIXED PROC in SAS 9.4 (SAS Institute Inc., Cary, NC). Apparent permeability coe cient of FITC, HIS, and LPS from mucosal to serosal side, and electrophysiological parameters (Isc and Gt) were analyzed utilizing the follow model: Y = µ + b k + π l + φ m + πφ ml + ε kml , Where µ = overall mean; b k = the random effect of the kth tissue; π l = the xed effect of the lth sampling time (20 min intervals); φ m = the xed effect of the mth treatment; πφ ml = the interaction effect of the mth treatment by the lth time; ε ijk = the random error and is ~ N(0, σ b 2 ). The model included tissue as repeated effect with compound symmetry as a covariance structure as it produced the least Akaike's and Bayesian Information Criterion values.
Gene expression data were analyzed with treatment as xed effects and tissue as repeated effect with compound symmetry as a covariance structure. Differences between treatments were determined by least square means methods using the PDIFF option and considered signi cant if P ≤ 0.05 and a tendency if 0.05 < P ≤ 0.10.

Electrophysiological Parameters of LPS and Histamine Treated Rumen Epithelium
HIS increased the Isc of rumen tissue when compared with CON (Fig. 1A, 88.2%, P = 0.0022). The Isc of tissue exposed to LPS was not different relative to the CON. Gt of HIS tended to higher than CON (Fig. 1B,  29.7%, P = 0.056).

Permeation of FITC, LPS, and Histamine of Rumen Epithelium
Adding LPS to the mucosal side did not affect the Papp of the RE to FITC and FITC ux rate across RE, but HIS increased Papp (P = 0.017, 1.04-fold) and the ux rate (P = 0.0223, 71.45%) of rumen tissue to FITC compared with CON ( Fig. 2A and Fig. 2B). The Papp of LPS from the mucosal-to-serosal side was 1.81-fold greater than that of HIS (Fig. 2D, P = 0.0005). As shown in Fig. 2C, the ux rate of HIS was 56.17 pmol/(cm 2 ×h); the ux rate of LPS was 12.71 EU/(cm 2 ×h).

Relative mRNA Abundance of Genes Associated with Tight Junction
As shown in Fig. 3A, OCLN expression was less for HIS compared with CON (1.1-fold, P = 0.0473), while there were no differences in the expression of CLDN1, CLDN4, and TJP1.

Disscusion
Subacute ruminal acidosis (SARA) is considered a major animal health and welfare issue in intensive ruminant production systems (Plaizier et al., 2008). While initially the focus was on low pH, it is now recognized that outcomes arising from SARA initiated by low digesta pH, increased SCFA concentration, and hyperosmolarity ( Thus, there is a growing body of research suggesting that changes occurring concurrently with decreased ruminal pH during SARA may damage RE barrier function. However, the speci c role of individual components is not clear. In the present study, the Isc of the RE was increased by HIS compared with CON, which is similar to the changes of physiological parameters of RE during SARA where increased Isc and Gt were observed (Klevenhusen et al., 2013). However, these results differ from  where no effect of HIS was detected on Isc.  illustrated a very e cient intraepithelial catabolism at a mucosal pH of 7.4. The catabolism of histamine seems to comprise a complex enzymatic pathway initiated by the diamine oxidase enzyme (DAO) (Sjaastad, 1967;Dickinson and Huber, 1972). Sun et al. (2017) demonstrated that histamine can activate of the NF-κB in ammatory pathway and upregulate the expressions of the in ammatory cytokines (TNF-α, IL-6, and IL-1β), a then induce the in ammatory response in bovine rumen epithelial cells. Thus, both e cient intraepithelial catabolism of histamine and induced in ammatory response in rumen epithelium seemingly indicated an induced metabolism in rumen epithelium which may be partly account for the increased Isc of RE under histamine.
The apparent permeability of HIS and LPS in RE were compared and the results showed that the Papp of HIS was less than that for LPS.  suggested that, at a mucosal pH of 7.4, permeability of the ruminal epithelium to histamine was very low. In addition, their study also demonstrated a very e cient intraepithelial catabolism of histamine (mucosal to serosal direction, 98.7%) at mucosal pH 7.4 and a signi cant secretory mechanisms from serosal to mucosal side . Thus, their results in vitro approach established that the intact ruminal epithelium is a very effective barrier to luminal histamine . LPS is thought to enter circulation by transport across the intestinal epithelium via not only paracellular pathways through the openings of intestinal tight junctions between two epithelial cells, but also by a transcellular pathway through lipid raft membrane domains involving receptor-mediated endocytosis (Berg, 1995;Drewe et al., 2001;Mani et al., 2012). Transcellular passive transportation is the predominate pathway of LPS absorption by intestinal mucosa (Drewe et al., 2001) and speci c transport system for LPS was observed in colonic epithelial cells (Tomita et al., 2004). Furthermore, signi cantly increased translocation of LPS from the mucosal to the serosal side of rumen tissues under the presence of mucosal side LPS was observed by Emmanuel et al. (2007). Thus, in the present study, the higher Papp of LPS than HIS indicated that LPS seemingly can more easily pass through the gastrointestinal tract than HIS.
Supporting past research, we did not observe an effect of HIS on Gt under incubation conditions with a pH of 7.4 . While Gt was not affected, HIS increased Papp and ux rate to FITC suggesting a direct role of HIS on altering RE permeability.  also reported that HIS receptors are broadly distributed and evidence of their localization within smooth muscle of the rumen (Ohga and Taneike, 1978) causing cessation of rumen contractions, increased vascular blood ow, and increased vascular permeability. In addition, HIS has been reported to increase permeability of the intestinal tract in rabbits (Kingham and Loehry, 1976;Miller et al., 1992) and permeability of HIS to cross the RE barrier increases with exposure to acidic pH ; however, when measured in vivo under anesthesia, permeability of the rumen to HIS was low (Kay and Sjaastad, 1974). Overall, the results of the present study support previous research that HIS exposure may have a causative role in reducing the barrier function of the RE. In addition, the mRNA abundance of OCLN, one of the tight junctions, was downregulated in HIS compared with CON (Fig. 3). Epithelial barrier function is primarily dependent on tight junction (TJ) proteins that limit paracellular permeation (Marchiando et al., 2010). A variation in epithelial permeability can be related to a change in the general abundance of TJ proteins, including the localization and the interaction of the proteins (Markov et al., 2015(Markov et al., , 2017 Availability of data and materials The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.