ABSORPTION AND METABOLISM OF VOLATILE FATTY ACIDS BY RUMEN AND OMASUM

Volatile fatty acids (VFA) absorption and metabolic capacity of rumen and omasum were compared, in vitro. Fragments of rumen wall and omasum laminae were taken from eight adult crossbred bovines. An isolated fragment of the mucosa was fitted in a tissue diffusion chamber. Valeric acid and CrEDTA were added to ruminal fluid and placed on the mucosal side and buffer solution was placed on the serosal side. Fractional absorption rates were measured by exponential VFA:Cr ratio decay over time. Metabolism rate was determined as the difference between VFA absorbed and VFA which appeared on the serosal side over time. Mitotic index was higher in omasum (0.52%) than in rumen epithelium (0.28%). VFA fractional absorption rate was higher in omasum (4.6%/h.cm2) than in rumen (0.4%/h.cm2). Acetate, propionate, butyrate, and valerate showed similar fractional absorption rates in both fragments. Percentage of metabolized acetate and propionate was lower than butyrate and valerate in both stomach compartments. In the rumen, individual VFA metabolism rates were similar (mean of 7.7 μmol/h.cm2), but in the omasum, valerate (90.0 μmol/h.cm2) was more metabolized than butyrate (59.6 μmol/h.cm2), propionate (69.8 μmol/h.cm2) and acetate (51.7 μmol/h.cm2). Correlation between VFA metabolism and mitotic index was positive in the rumen and in the omasum. In conclusion, VFA metabolism and absorption potential per surface of the omasum is higher than that of the rumen. Variations on rumen and omasum absorption capacities occur in the same way, and there are indications that factors capable of stimulating rumen wall proliferation are similarly capable of stimulating omasum walls.


INTRODUCTION
Absorptive surfaces of reticulorumen (DIRKSEN et al. 1984) and probably that of omasum (BALDWIN et al. 2004) are directly related to VFA absorption capacity.The understanding of omasal physiology seems to be as important in the control of current digestive disturbances as the understanding of ruminal physiology.
Despite the absorptive surface of reticulorumen (7.7 m 2 ) being higher than that of the omasum (2.1 m 2 ) (DANIEL et al. 2006), and absorption and metabolism potential of the rumen being well documented, the potential of absorption and metabolism of the omasum and any comparison of these parameters among organs are poorly understood.The purpose of this work was to compare, in vitro, VFA absorption and the metabolism capacity of rumen and omasum.

MATERIAL AND METHODS
Eight adult crossbred bovines, of different weights and ages, of both sexes, coming from a commercial slaughterhouse, were allocated to a completely randomized block design where each animal was considered one block.Animals were slaughtered by exsanguinations after stunning, and the forestomach was removed from abdominal cavity 5 to 10 minutes later.One portion of rumen wall was taken from the ventral sac (Recessus ruminis) and one portion of omasal laminae (Laminae omasi) were cut and placed in a Krebs-Ringer bicarbonate buffer solution (Sigma-Aldrich, Saint Louis, Missouri, USA) at 38 o C with pH adjusted to 7.4, and immediately transported to the laboratory.In the laboratory, ruminal mucosa was isolated by removing serosa and muscular layers, and omasal mucosa was carefully separated.Mucosal sheets were cut into circles (3 cm diameter) and inserted between halfchambers of a tissue diffusion chamber with an inner aperture of 4.91 cm 2 (Indústria e Comércio Estanhof Ltda, Lavras/MG, Brazil).Bathing solutions on both sides of the chamber were kept circulating by a mini air compressor and maintained at 38 o C in water-jacketed reservoirs.Absorption assays were started 30 to 40 minutes after slaughter.
Previously, about 320 mL of rumen fluid was collected from a fistulated cow, fed on tropical pasture and 2.5 kg d -1 of a ground corn and soybean meal based commercial concentrate.Fermentation was stopped by the addition of 6.5 mL of 50% sulfuric acid.Forty milliliter aliquots of rumen fluid were frozen at -20 o C. For each animal one aliquot was thawed before incubation.Valeric acid (ALLEN et al., 2000) and CrEDTA solution (BINNERTS et al., 1968) were added to rumen fluid until reachi ng concent rations of 25 mM and 1 mM, respectively.PH was adjusted for 6.8 with addition of 50% NaOH solution and the final solutions of rumen fluid plus Valeric acid and CrEDTA were placed on mucosal side of the chamber fitted with rumen or omasum fragments.The Krebs-Ringer bicarbonate buffer solution with pH adjusted for 7.4 was used on the serosal sides of the chamber.All solutions were heated at 38 o C before incubation and this temperature was maintained during incubation.Initial pH of mucosal fluid was the same for rumen and omasum because of their similarity in vivo (EDRISE et al. 1986).
Samples (300 μL) of mucosal and serosal fluids were obtained immediately after the assay was started (time zero) and at 25, 50, 75, 100, and 125 minutes thereafter.Mucosal pH was determined at each sampling time and serosal pH was determined at the start and end of the incubation.The samples were mixed with 100 μL of 10% sulfuric acid solution and immediately frozen at -20 o C for subsequent analyses of VFA and chromium.
VFA and chromium concentrations were determined in supernatant obtained after centrifugation at 8,855 g for 15 minutes at room temperature.Samples were analyzed for VFA by Gas Liquid Chromatography (CP-3800 Gas Chromatograph Varian, with flame ionization detector, Varian Chromatography Systems, California, USA) using a capillary column with nitroterephthalic acid-modified phase, chemically bonded polyethylene glycol, 25 m x 0.25 mm I.D. and 0.2 μm of film thickness (CP-Wax 58 (FFAP) CB, Varian Analytical Instruments, California, USA), and N 2 as carrier gas.Oven temperature was kept at 65 o C for 30 s, then increased to 125 o C, at a rate of 20 o C/minute, and then increased again to 170 o C, at a rate of 50 o C/minute.Total time of analysis was 4.9 minutes.Chromium concentration was determined by atomic absorption spectrophotometry (SpectrAA -100 Varian, Varian Australia Pty LTD, Victoria, Australia).
Exponential VFA:Cr decay ratio over time on the mucosal side was used to estimate fractional absorption rate of VFA (RESENDE JÚNIOR et al. 2006).Since the major objective of this study was to compare compartments, only net absorption rates were determined.It was not accessed mucosal-serosal and serosal-mucosal unidirectional transport fluxes.VFA metabolism was estimated as the difference between VFA absorbed and that appearing in serosal fluid.
Rumen surface area was measured with the aid of a scanner (Scanjet 4C, Hewlett Packard).The number of papillae was counted and twelve papillae were randomly sectioned and digitalized for surface determination through the image analysis software UTHSCSA Image Tool, version 3.00 (Free Software, The University of Texas Health Science Center in San Antonio).Total fragment surface area was the sum of papillae surface plus wall surface minus surface of the base of papillae, assumed as 0.002 cm 2 per papilla (DANIEL et al. 2006).Omasum absorptive surface area was considered as constant (4.91 cm 2 -chamber orifice), since omasum papillae were not take into account, because of their very small size.
To determine mitotic index, 5 μm sections of histologically prepared fragment were stained with hematoxylin and eosin, and to measure thickness of epithelium layers, 5 μ m sections were stained with Masson's Tricromic.Mitotic index of cells of epithelium basal layer, epithelium thickness, keratin layer thickness, and non-keratin layer thickness were determined with an optic microscope (Ernst Leitz Wetzlar Nr. 438895, Germany) at 400 magnification.Mitotic index was determined by counting all mitotic nucleus, which were expressed as percentage of total visible nucleus.Percentage of cells undergoing mitosis was the average of five independent evaluators.
Morphological variables were compared using the MIXED procedure (LITTELL et al., 1998) of SAS (SAS INSTITUTE 1999), according to the following model: y =μ+α +β +ε ; where μ: overall mean; i α : random effect of animal (i = 1 to 8); j β : fixed compartment effect (j = rumen or omasum); and ij ε : residual error assumed independently and identically distributed in a normal distribution with mean zero and variance 2 σ .Fractional absorption and metabolism rates in both compartments (rumen and omasum) were analyzed as a split plot arrangement, using the MIXED procedure of SAS, according to the following model: y =μ+α +β +αβ +γ +βγ +ε ; where μ : overall mean; i α : random animal effect (i = 1 to 8); j β : fixed compartment effect (j = rumen or omasum); ij αβ : interaction between animal and compartment effect (error term used for testing compartment effect); k γ : fixed VFA A effect (k = acetate, propionate, butyrate or valerate); jk βγ : interaction between compartment and acid effect; and ijk ε : residual.For pH analysis the time collection effect was added to the model.
Pearson correlation coefficients among variables within compartments were performed with CORR procedure of SAS.Linear regressions among measurements of both forestomach compartments were performed with the REG procedure of SAS.

RESULTS AND DISCUSSION
Mitotic index (Table 1) indicates a faster cell proliferation in omasum than in rumen epithelium.Rumen absorptive surface area, however, was higher than that of the omasum, mostly due to ruminal papillae.Despite systemic regulation on cell proliferation (SAKATA et al. 1980;SHEN et al. 2004), the local stimulus affects epithelial cells dynamics (SAKATA;TAMATE 1976;GÁLIFI et al. 1986).Thus, higher surface:digesta mass ratio in omasum (DANIEL et al. 2006) could maximize VFA local stimulation on cell proliferation, since large amounts of VFA pass from reticulorumen to omasum incorporated to the ruminal fluid phase (RESENDE JÚNIOR et al. 2006).Omasum mitotic index was highly correlated with rumen mitotic index (Figure 1), indicating that stimulating factors for rumen wall proliferation may be the same for omasum wall.
Total thickness, thickness of keratin layer, and thickness of non-keratin layers of the rumen and omasum epithelium were similar (Table 1) and highly correlated with each other (Figure 2).Given that the epithelium dimension is the result of cellular synthesis and deletion (TAMATE; FELL 1977), higher omasum epithelium cell proliferation might have been compensated by higher cell loss, since larger surface:digesta mass ratio and dehydration of content in the omasum could increase abrasive effects on the epithelium.DIJKSTRA et al. 1993).Due to the addition of valeric acid to the mucosal fluid, its concentration was higher than that of butyrate and similar to that of propionate, which does not occur in physiologic conditions.Utilization of valeric acid as a marker of ruminal VFA clearance, in vivo, was proposed by Allen et al. (2000) and validated by Resende Júnior et al. ( 2006), but published data about its use in vitro is scarce.
VFA fractional absorption rate by surface of the omasum epithelium was 11 times greater than that of the rumen (Table 2).This finding demonstrates the higher absorptive potential of omasum, even considering the fact that the absorptive surface of reticulorumen is approximately 4 times higher than that of omasum (DANIEL et al. 2006).The reason for this high magnitude of difference could be related to absorption mechanisms in the organ wall.In the rumen, VFA are absorbed under dissociated (through bicarbonate exchange) and nondissociated forms (lipophilic and highly permeable) (GÄBEL et al. 2002).According to VFA pKa is minor than 4.9, over 95% of the VFA should be in ionized form at the ruminal pH around 6 to 7. In the omasum, VFA seems to be absorbed predominantly under protonated form (ALI et al. 2006).Furthermore, unlike the rumen, the omasum absorbs bicarbonate, through chloride exchange (ALI, et al. 2006).
Correlation between rumen and omasum VFA fractional absorption rates was positive (Figure 3), indicating that variation in the absorptive capacity of those organs is unidirectional.Thus, the use of nutritional strategies to prevent digestive disturbances related to ruminal VFA accumulation could to improve the absorption capacity of the whole forestomach.
VFA metabolism in rumen epithelium, as a proportion of absorbed VFA, was higher than in omasum epithelium (Table 2), probably because of its larger mass of active metabolic cells.Since thicknesses of non-keratin layers of the rumen and omasum epithelia were similar (Table 1), the absorptive surface can be used as an indicator of active metabolic cells mass.
Thickness of the rumen epithelium keratin layer (r 2 = 0.76; P = 0.03) and the mitotic index (r 2 = 0.68; P = 0.06) were positively correlated with rumen metabolism rate.In the omasum, correlation between VFA absorption and mitotic index was positive (r 2 = 0.80; P = 0.02).Total and individual VFA concentrations did not differ among solutions used for rumen and omasum fragments (Table 1), which excludes the possibility of Omasum mitotic index (%) Omasum epithelium ( m) Absorption and metabolism of volatile fatty... 97 Ciênc. agrotec., Lavras, v. 36, n. 1, p. 93-99, jan./fev., 2012   Non-physiologic concentrations of valerate on the mucosal side of both rumen and omasum may have affected butyrate metabolism (KRISTENSEN;HARMON 2005).Kristensen and Harmon (2005) observed that the portal appearance of the butyrate that disappeared from the rumen increased from 25% to 52% when ruminal concentration of valerate increased from 1.2 to 8.0 mmol kg -1 .Moreover, the same authors observed that the increase in ruminal butyrate concentration (4 to 36 mmol kg -1 ) resulted in an increase of portal recovery of butyrate (18 to 52%) and valerate (16 to 54%), reinforcing the theory of competing metabolism routes in the ruminal epithelium between those two VFA (KRISTENSEN;HARMON 2004).
Increase of mucosal fluid pH and its reduction in serosal fluid (Figure 4) reflected VFA transfer from the mucosal to serosal direction.Rumen and omasum mucosal pH have increased during incubation, but there was a tendency (P = 0.08) of interaction between compartment and time.Rumen mucosal pH rose faster than omasum mucosal pH.This may have happened because of bicarbonate secretion by rumen epithelium and bicarbonate absorption by omasum epithelium, seeing as VFA absorption rates were similar among compartments.pH of mucosal and serosal fluid of rumen and omasum fragments incubated for 2.08 hours in a tissue diffusion chamber.P = 0.30 for compartment effect; P < 0.01 for solution effect; P = 0.17 for interaction between compartment and solution effect; P < 0.01 for time effect; P = 0.27 for interaction between compartment and time effect; P < 0.01 for interaction between solution and time effect; P = 0.16 for interaction between compartment, solution and time effect.

Figure 4 -
Figure4-Means of initial (black bar) and final (white bar) pH of mucosal and serosal fluid of rumen and omasum fragments incubated for 2.08 hours in a tissue diffusion chamber.P = 0.30 for compartment effect; P < 0.01 for solution effect; P = 0.17 for interaction between compartment and solution effect; P < 0.01 for time effect; P = 0.27 for interaction between compartment and time effect; P < 0.01 for interaction between solution and time effect; P = 0.16 for interaction between compartment, solution and time effect.

Table 1 -
Absorptive surface area, mitotic index, epithelium layers thickness and mucosal fluid VFA concentration in rumen and omasum.

Table 2 -
Fractional absorption and metabolism rates of volatile fatty acids in bovine rumen and omasum fragments incubated for 2.08 hours in a tissue diffusion chamber.Comp: compartment effect; Acid: Acid effect; Comp*Acid: Interaction between compartment and acid effect. 2 acetate; 3 P: propionate; 4 B: butyrate; 5 V: valerate.6Standarderror of the mean. 1