--$\\~.\\‘u-“r .q—.--_ . . . 7V ‘ ,.. I! . I l . ~ ~ I .J r I ‘. , ‘ . i -- INVESTIGATION OF SOME FACTORS 1 f? . _ 'FFECTING VOLATILE FATTY ACID . TRANSFER ACROSS RUMEN EPITHELIW IN AN IN VITRO SYSTEM - ' ; * A- '. fj 5,! 3. JOHN WALTIIR BELL.“ '31-" *‘ ‘ . H... . -. an} ,. M .. ...-.'.,-.‘.-._ .. .- 1.7...3. , . ~..p:.-.- «I r. . ~ . an 3‘- 4/ '1‘!” .7; - qvw-r 4;. 'gta p» a?" c" ,_,‘ ... .. .. . 'tr- alrr- Va peso! owovu-vm ‘ I. ‘1 t . . A o—«vr -.......... ’fvvbu'n— ’4_....:_wl::..v.,n . ’_ ... .. ___.,, - '. ' . ‘zf '....".. 1.2m ... A .I...'.4 ..:3:r.-.'; vvna‘ '51:,“ . A . *Worw~v’ ”1:...“ yum-ta. ‘ "or ‘4 d T ,. um/pyu '~v' w v.» ”d | I .. .. .-‘ 3Z..::..,.., .. .r. w... 1’.“ 22:33:31.1. m." 3&3”. - * .— . r u now- as}; N ovnnva-vut "' -v-a~ " .~-- p. ”Iv _ .1, ." . .:g :MW.‘ n- Mu”...ny u ..u....-,‘ . -N1'r'fi'. u W‘W-d-r .. ._:: ' .m” "arr? Ely-:3. T'HF‘fUS I “f ‘ , -—a , ”My” 1‘ This is to certify that the thesis entitled Investigation of Some Factors Affecting Volatile Fatty Acid Transfer Across Rumen Epithelium in An lg Vitro System presented by John Walter Bell has been accepted towards fulfillment of the requirements for Ph.D. Dairy degree in Major professm/ (/ Date September 14, 1970 0-169 I I I ABSTRACT INVESTIGATION OF SOME FACTORS AFFECTING VOLATILE FATTY ACID TRANSFER ACROSS RUMEN EPITHELIUM IN AN 1;; VITRO SYSTEM BY John Walter Bell An in Vitro system for studying transfer across rumen epithelium is described and results of changing parameters are given. Although the constant maximum rate of volatile fatty acid transfer was not measured, due to inclusion of the lag time, this lag time of serosal ap- pearance was reduced by higher fatty acid levels and by decreased pH, and was independent of fatty acid chain length. Among fatty acids, serosal accumulation at two hours is a measure of relative transfer rate. Parameters investigated and results included: 1) Mucosal concentration and fatty acid chain length on serosal accumulation. Sample regression co— efficients were: 2) 3) 4) John Walter Bell 0.301 X - 0.179 Acetate: y 0.454 X - 0.250 Propionate: y 0.166 X - 0.048 Butyrate: y There was a significant departure from linear regression of serosal accumulation on mucosal concentration of acetate and propionate. Decrease in pH to 4.0 increased (P < 0.01) serosal accumulation of acetate and prOpionate. HCl caused a further increase (P < 0.01) in acetate accumula- tion relative to H2504 as the adjusting medium. Serosal glucose and mucosal HCN did not affect acetate accumulation and serosal butyrate reduced acetate accumulation in one instance. Both serosal glucose and serosal butyrate reduced propionate accumulation at some concentrations. Back transfer moved a larger percentage of C14- butyrate carbon than mucosal-to-serosal transfer of unlabeled butyrate. The amount of back trans- fer was not enough to affect other results. 2 John Walter Bell 5) There was less serosal accumulation with biopsy tissue than with slaughter tissue. These results are consistent with the thesis that volatile fatty acid transfer is a passive process, modi- fied by epithelial metabolism. INVESTIGATION OF SOME FACTORS AFFECTING VOLATILE FATTY ACID TRANSFER ACROSS RUMEN EPITHELIUM IN AN IN_VTTRO SYSTEM BY John Walter Bell A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Dairy College of Agriculture and Natural Resources 1970 6697/49 DEDICAT ION This volume is respectfully dedicated to my parents and teachers who taught by example that worthwhile things must be sought diligently. ACKNOWLEDGEMENTS The author wishes to express appreciation to Dr. R. S. Emery for his guidance during this program. Appreciation is also extended to guidance committee members, Dr. J. W. Thomas, Dr. E. M. Convey, Dr. R. L. Anderson, and Dr. L. F. Wolterink for their suggestions and critical reading of this manuscript. The help of Mrs. Dorothy Carpenter, Mrs. Judy Olney, and George Blank in the laboratory was greatly appreciated. The secretarial help of Mrs. Elaine Kibbey was also of great value. My thanks to Dr. C. A. Lassiter and the Depart— ment of Dairy for financial support during this period of study. Finally my grateful appreciation to my wife, children, and family for their constant support and en- couragement. Special thanks are due to Mr. and Mrs. David Savidge for timely financial and moral support. iii TABLE OF CONTENTS Page ACKNOWLEDGEMENTS. . . . . . . . . . . . . . . . . . iii LIST OF TABLES. . . . . . . . . . . . . . . . . . . vi LIST OF FIGURES . . . . . . . . . . . . . . . . . . Vii INTRODUCTION. . . . . . . . . . . . . . . . . . . . 1 REVIEW OF LITERATURE. . . . . . . . . . . . . . . . 3 Methods of Studying Rumen Transfer: General Considerations. . . . . . . . . . . . . . . . 4 Pool Size and Heterogeneity . . . . . . . . 4 Inlets to the Rumen . . . . . . . . . . . . 5 Outlets from the Rumen. . . . . . . . . . . 9 Isotopic Methods. . . . . . . . . . . . . . 14 The Rumen Pouch Method. . . . . . . . . . . 14 Perfused Rumen Techniques . . . . . . . . . 15 .lfl Vitro Methods of Studying Rumen Absorption. . . . . . . . . . . . . . . . 16 Factors Affecting Volatile Fatty Acid Disap- pearance from the Rumen . . . . . . . . . . . 16 Influence of Volatile Fatty Acid Chain Length on Their Disappearance from the Rumen . . . . . . . . . . . . . . . . . . 17 Effect of Volatile Fatty Acid Concentra- tion on Disappearance from the Rumen. . . 20 Effect of Hydrogen Ion Concentration on Volatile Fatty Acid Disappearance from the Rumen . . . . . . . . . . . . . . . . 23 iv TABLE OF CONTENTS (cont.) Page Factors Influencing Appearance of the Volatile Fatty Acids in the Portal Vein. . . . . . . . 28 Effect of Volatile Fatty Acid Chain Length on Appearance in the Portal Blood . . . . 29 Influence of Rumen Concentration on Volatile Fatty Acid Appearance in the Portal Blood. . . . . . . . . . . . . . 31 Influence of Rumen pH Value on Volatile Fatty Acid Appearance in the Portal Blood . . . . . . . . . . . . . . . . . . 32 Summary of Literature . . . . . . . . . . . 33 METHODS AND MATERIALS . . . . . . . . . . . . . . . 35 RESULTS AND DISCUSSION. . . . . . . . . . . . . . . 43 Effects of Changing Mucosal Concentrations on Serosal Accumulations . . . . . . . . . . . . 43 Comparison of Slaughter- Versus BiOpsy— Obtained Tissues. . . . . . . . . . . . . 58 Effect of Adding Glucose to Serosal Fluid on Serosal Accumulation of Acetate and Pro— pionate . . . . . . . . . . . . . . . . . . . 60 Effect of Serosal Butyrate. . . . . . . . . 62 Effect of Sodium Cyanide on Serosal Accumula- tion of Acetate . . . . . . . . . . . . 65 Effect of Changing pH on Acetate and Propionate Accumulation in Serosal Fluid. . . 67 Study of Serosal to Mucosal Transfer Using Cl4-Sodium Butyrate . . . . . . . . . . . . . 73 Time Sequences for Serosal Accumulations. . . . 75 CONCLUSIONS . . . . . . . . . . . . . . . . . . . . 82 APPENDIX I. . . . . . . . . . . . . . . . . . . . . 85 APPENDIX II . . . . . . . . . . . . . . . . . . . . 86 REFERENCES. . . . . . . . . . . . . . . . . . . . . 88 Table 10. 11. 12. LIST OF TABLES Page EFFECT OF MUCOSAL CONCENTRATION ON SEROSAL ACCUMULATION OF ACETATE . . . . . . . . . . 44 EFFECT OF MUCOSAL CONCENTRATION ON SEROSAL ACCUMULATION OF PROPIONATE. . . . . . . . . 44 EFFECT OF MUCOSAL CONCENTRATION ON SEROSAL ACCUMULATION OF BUTYRATE. . . . . . . . . . 45 DIFFERENCES IN MEANS OF SEROSAL ACCUMULA- TIONS OF THE VOLATILE FATTY ACIDS . . . . . 48 ANALYSIS OF SUMS OF SQUARES FOR GROUP MEANS . 55 BIOPSY VERSUS SLAUGHTER MATERIAL. . . . . . . S9 EFFECT OF ADDING GLUCOSE TO SEROSAL FLUID ON SEROSAL ACCUMULATION. . . . . . . . . . . . 62 EFFECT OF SEROSAL BUTYRATE ON 2-HOUR SEROSAL ACCUMULATION OF ACETATE AND PROPIONATE. . . 64 EFFECT OF CHANGING MUCOSAL HYDROGEN ION CONCENTRATION ON SEROSAL ACETATE AND PRO- PIONATE ACCUMULATION. . . . . . . . . . . . 68 SEROSAL TO MUCOSAL BUTYRATE TRANSFER. . . . . 74 TIME SEQUENCE FOR ACETATE ACCUMULATION. . . . 76 TIME SEQUENCE FOR PROPIONATE ACCUMULATION . . 77 Vi LIST OF F IGURES Figure Page 1. Serosal Accumulation of Volatile Fatty Acids as a Function of Mucosal Concentration. . . . . . . . . . . . . . . 47 2. Time Sequence of Acetate Serosal Accumulations. . . . . . . . . . . . . . . 78 3. Time Sequence of Propionate Serosal 79 Accumulations. . . . . . . . . . . . . . . vii INTRODUCT ION The ruminant's ability to increase the nutritional value of carbohydrate and protein sources for the human pOpulation is both historic fact and future promise. The studies of the past one hundred years have defined many.of the processes by which the rumen microbiota accomplish this task. As knowledge accumulates concerning the amounts and variety of materials leaving the rumen through the rumen epithelia, its importance as an agent between the microbiota and host becomes apparent. Further study of epithelial contribution and function is necessary and no method of studying this tissue is completely satisfac- tory under all conditions. A method of study less depen- dent upon host and microbiota variation would be useful for many purposes and further adaptation of ig_vitro methods approaches this goal. This project was initiated with three purposes: 1) The development of a simple ig_vitro system for studying transfer across the rumen epithelium. 1 2) 3) The comparison of responses of this system with responses of in vivo and other in vitro methods to such experimental parameters as might be im— posed. The examination of the processes by which mate- rials are transferred across the rumen epithelium. REV IEW OF L IT ERATURE In many respects the rumen may be considered as a pool. This is especially so when studying absorption from the rumen. The stream enters through the esophagus and exits via the omasal orifice. In addition, materials enter as secretions and other exits exist as absorption through the rumen wall or eructed gasses. Interactions between the pool inlets and outlets cause the volume of the rumen pool to constantly change. The pool composition is also constantly changing, due to internal production and consumption of metabolites and possibly differential absorption. This review of literature is divided into three sections: 1. Methods of evaluating passage of materials into and out of the rumen, including in vivo, perfused rumen, and in vitro methods. 2. Factors affecting the rate of disappearance of volatile fatty acids from the rumen. 3 3. Factors affecting appearance of volatile fatty acids in portal blood. Methods of Studying Rumen Transfer: General Considerations Pool Size and Heterogeneity Under some circumstances the problems of variation in volume and the heterogeneity of the contents may be studied concurrently. Rumen volume may be determined by removing the rumen contents and measuring the volume and rumen composition by mixing and sampling while the con— tents are removed. The materials may then be returned to the rumen or a material of known composition substituted. The disadvantages of these methods include disruption of rumen fermentation, restriction of studies to animals with rumen fistulas, and the impossibility of performing rapid time sequence eXperiments because of the volume of mate- rials. Marker techniques afford a second means of deter- mining rumen volume, a means whose validity is dependent upon the uniformity with which the marker is distributed throughout the rumen. Under natural conditions this dis— tribution is sometimes incomplete because of time limits and natural layering of rumen contents. Pumps may be used to mix the materials more rapidly, but are largely inef- fective when long hay or similar roughage is present. Johnson (28) included a discussion of available marker materials and analytical techniques in his review of rumen study procedures. Inlets to the Rumen l) Esophagus. The principle avenue of material inflow to the rumen is the eSOphagus, which carries both food and saliva. Both amount and content of food may be controlled. The common method is to withhold food en- tirely, or if some intake is essential, the animal may be trained to consume all of a known amount within a speci- fied time limit. As long as the food is relatively homo- geneous, animals wiIl not discriminate against any por— tion. However, if the ration contains long roughage as well as concentrate, the animal may refuse part of the roughage or the smaller particles may separate by gravity and knowledge of composition as well as amount of refused feed is necessary and not always accurate. Saliva, as a secretion of the animal, is not easily controlled. Various stimuli have been shown to affect not only the amounts of saliva secreted, but also its composition. Saliva contains large amounts of sodium, potassium, and magnesium bicarbonates. Ordinarily only the sodium and magnesium ions are of interest, because of the large amounts of potassium in ruminant rations. With increasing interest in non—protein-nitrogen sources and their metabolism, the urea content of saliva and its pool kinetics are of interest, the mode of entrance being a basic part of these studies. Saliva is an effective buf- fering agent (55), due in part to the bicarbonates, and many rumen reactions are influenced by pH value. Because of these considerations, there are three choices: 1) Ignore the contribution of the saliva, 2) drain off the saliva, or 3) drain, sample, and measure the saliva, and return it to the rumen. If course two or three is followed, and saliva is to be drained, the tech- niques available include the esophageal cannula (28) and the esophageal catheter (18, 50). The former may be forced out of the fistula, the latter adds stress at the time of the experiment by the stimulation of the cardia during the experiment. With these methods, saliva volume and constituents may be determined and the saliva returned to the animal. If the saliva is not returned to the animal, it must be replaced by a solution of equivalent buffering capacity because of the sensitivity of rumen fermentation to pH value changes. 2) The Rumen Epithelium. The materials passing through the epithelium into the rumen may be divided into three classes: organic, inorganic, and water. Water pas- sage into the rumen resulting from differences in osmotic pressure may play a role in controlling the concentrations of materials in the reticulorumen. Murray et al. (35) suggest that changes in osmotic pressure and water passage influence the amounts of rumen contents and influence the outflow of rumen fluid to the omasum. warner and Stacy (60) noted that postwprandial increases in rumen content osmotic pressure was accompanied by signs of haemoconcen— tration, indicating a transfer of water from the blood into the alimentary tract, including the rumen. Stacy and Brook (44) concluded that the renal response of sheep to feeding was partly due to water withdrawal from the blood into the rumen. The work of Parthasarathy and Phillipson (36) indicated that, in anaesthetized sheep with a surgically isolated rumen, hypertonic solutions within the rumen gained water as a result of water move— ment across the rumen wall. This dilution of rumen con- tents as a result of an increase in the osmotic pressure could serve to obscure increases in concentration if con- current rumen volumes are not determined. Annison (lb) reported that blood volatile fatty acids entered the rumen to bring final volatile fatty acid concentrations of originally volatile fatty acid-free solutions close to that of blood. This would probably be true of all small organic molecules. However, especially in the inorganic classifica- tion, not all assumed secretions via the rumen epithelium have been substantiated. Smith et a1. (41) measured se— cretion of isotopic phosphorus into rumen contents, using assumed salivary secretion rates, and reported that the amount of phosphorus transferred was greater than could be accounted for by salivary secretion. Their alternative explanation was that substantial amounts of phosphate must enter the rumen via its epithelium. However, this would require that phosphate move against both concentration and potential gradients through the rumen epithelium. According to Hyden (26) and Ash and Dobson (7), rumen epithelium is only slightly permeable to the phosphate ion in either direction. With revised secretion rates Annison and Lewis (3b) concluded that endogenous phos- phorus in the rumen came from saliva. These examples serve to illustrate both the dif- ficulty and the importance of being able to measure the contributions of the saliva and rumen epithelium to rumen contents. Outlets from the Rumen In the intact animal, there are three outlets from the rumen, the eSOphagus, reticulo-omasal opening, and the rumen epithelium. l) The eSOphagus functions as an outlet for gasses from the rumen. Loss of fermented materials from the rumen as carbon dioxide and methane usually constitute five to seven percent of the gross energy consumed (34). Negative pressure applied to the gas space at the top of 10 the rumen, via a rumen fistula, with measurement of the amount and concentration of materials in the effluent is one means of measuring these losses. 2) The reticulo—omasal Opening is the exit for all solids and part of the liquids leaving the rumen. Methods of dealing with omasal outflow are usually varia- tions of three approaches: 1) Disregard this outflow, 2) Close the opening, assuming that changes in rumen and animal metabolism will not interfere with the experiment, or 3) Use a marker to measure omasal outflow. The choice is made on the essentiality of knowing how much material leaves the rumen Via this route, the acuteness of the experiment, and the effect of blocking the reticulo—omasal opening on the metabolism of the animal. The first ap- proach was used by Barcroft et a1. (8) in their pioneering studies which demonstrated that the volatile fatty acids were absorbed from the rumen. For the purpose of their investigation, it was only necessary to demonstrate the appearance of volatile fatty acids in the rumen vein. The second method, which they also used, requires ligation of the neck of the omasum and is unusable for many investiga- 'tions because of the extensive surgery required. Other ll disadvantages include the unphysiological state caused by the disturbance of the digestive organs, anesthesia, and the inability to use the animal more than once, pre- venting inter-treatment comparisons. The third approach to measuring omasal passage is to replace the rumen contents with a known amount of liquid containing a known amount of a marker. Johnson (28) discussed several of these substances and their limitations as well as references where detailed protocols for their usage may be found. Use of markers for this purpose is predicated upon a strict relationship between the marker and the material under investigation. Ideally as a quantity of material being studied leaves the rumen via the reticulo-omasal opening, it is accompanied by a proportional amount of marker. If, however, there is a departure from strict proportionality caused by a unilat— eral loss or gain of either study material or marker, errors will be introduced. A restriction to marker use is the absorption onto the wall of the rumen and upon particles of ingesta in the rumen. Another area of concern is the behavior of the replacement fluid. In most biological solutions water is the solvent. Rumen fluid is no exception, and 12 consequently, most rumen fluid replacements use water as the solvent. Water, however, has the ability to diffuse across the rumen epithelium in response to osmotic pres- sure gradients, and experiments which depend on material: marker ratios to evaluate absorption require careful study to detect water flux. If absorption in the entire fore-stomach is to be considered, the omasal-abomasal fistula described by Kameaka and Morimoto (29) might be considered. A slightly different approach would be the re-entrant duodenal fis— tula of Conver, McGilliard, and Huffman (l6). 3) The rumen epithelium is the most important exit from the rumen in terms of dry matter disappearance. Measuring transfer through the rumen epithelium may be done directly or indirectly. Direct methods of measuring rumen epithelium transfer involves measurement of arteriovenous differences in concentration and blood flow rate of the material being studied. Measuring rumen absorption by comparison of rumen or portal vein blood concentrations with concentra- tions in arterial blood is surgically difficult (7). In contrast to the abomasum, the rumen is drained by more than one vein (40). These veins discharge their contents 13 at various points into the portal vein, which contains blood draining from the more distal portions of the di- gestive tract. There are two alternatives, both less than satisfactory, measuring less than the complete rumen contribution or including contributions from more distal portions of the digestive tract. A further limitation of measuring arteriovenous differences is that, without a measure of blood flow, no quantification of materials absorbed is possible. Another disadvantage to the direct method is the selective metabolism of the rumen epithelium. Sutherland (49) cites a number of reports to indicate that rumen epithelium, both in vitro and ig_yiyg, metabolizes butyric acid to ketone bodies, and under natural conditions, little butyrate appears in the portal flow (2). These reports indicated limited metabolism of acetate in the rumen wall, most being carried unchanged to the liver, and limited metabolism of propionic acid to lactic acid. The indirect approach is described by Sutton et a1. (50) using young calves. Absorption was calculated as the amount disappearing from the rumen that could not be accounted for by other exits. In this case the only 14 exit was the omasal orifice, and disappearance via this route was calculated using polyethylene glycol as a marker. Isotopic Methods Another way of studying rumen absorption is the use of isotopes. Isotopes are especially valuable when used in combination With other techniques, which can help to solve the problem of recirculation of materials from the rumen to the blood and back. Discussions of their uses, advantages, and limitations, as well as specific applications, are given by Annison (la) and Cook (17). The Rumen Pouch Method A method to alleviate some of the difficulties encountered in the incompletely blocked off rumen inlets and outlets is described by Tsuda (51, 52). In this method a portion of the rumen is surgically closed to make a blind pouch opening to the outside through a fis- tula. In measuring the relationship between rumen motil- ity and volatile fatty acid absorption, the amounts 15 ~absorbed were expressed as percent of initial concentra- tion (53). Absorption rate was dependent upon dilution and in later work, Tsuda (54) found that the direction of water flux depended upon the concentration of materials in the rumen. Perfused Rumen Techniques In further efforts to control rumen inlets and outlets and remove other variables, perfused rumen tech- niques, such as that described by Brown (13) have been employed. These methods use a rumen excised from the animal, suspended so that material may be placed in it, and perfused with the collected, oxygenated blood of the donor animal. These preparations have many of the same limitations as acute experiments, but do have the advan- tage of recirculating blood without metabolism, so that slowly absorbed materials may be concentrated. However, the low blood flow rate, lack of liver and kidney func- tion to maintain normal metabolite levels, and gradual anoxia make the perfused rumen a method used primarily for special uses. 16 In Vitro Methods of Studying Rumen Absorption The requirements and limitations of lfl.Y£YQ me- thodology have caused some investigators to turn to in vitro methods. Early in vitro devices for studying in— testinal absorption in both ruminants and non-ruminants are described by Wilson (62). A device first used by Ussing and Zerhan (57) to study transport of sodium in frog skin and more recently used by Stevens (45, 46, 47, 48) for the study of rumen epithelial transport is a more elaborate version of these early approaches. Hird and Weidemann (25) described a similar device and used it to study transport and metabolism of butyrate by rumen epi— thelium. Both of these systems suspend rumen epithelium between two chambers containing fluids representing the blood and rumen contents. Factors Affecting Volatile Fatty Acid Disappearance from the Rumen The volatile fatty acids have been shown to be the main source of energy for the ruminant (3a) and the main product of cellulose digestion. In this group of 17 substances there is a range of properties and conditions that affect their absorption and utilization. In this section the properties and conditions that affect rumen disappearance are discussed. Influence of Volatile Fatty Acid Chain Length on Their Disappearance from the Rumen Danielli et al. (19) first reported that volatile fatty acids disappear from an acid medium in the order butyric > prOpionic > acetic. Gray (22) confirmed that propionate was absorbed more rapidly than acetate. Gray and Pilgrim (24) combined the results of two methods, absorption as measured by disappearance from the rumen, and volatile fatty acid production as measured by in vitro fermentation studies, and suggested that prOpionate was absorbed from the rumen more rapidly than acetate and butyrate. This conclusion is dependent upon the corre- Spondence of in yiyg and ig_vitro volatile fatty acid production rates. Johnson (27) used goats, and found the order of disappearance to be butyrate > propionate > ace- tate. These reports indicate that in yiyg, the longer the volatile fatty acid chain, the more rapidly it is 18 absorbed. This would indicate that the size of the mole- cule p§£.§§,is not the limiting factor in rumen absorp- tion. If that were the case, the smaller molecule, the shorter chain, would be absorbed faster. Stevens and Stettler (47) compared acetate and butyrate in their lg vitro preparations and found that butyrate disappeared two to three times as rapidly at both pH 6.4 and 7.4. They later (48) proposed an active transport system for acetate in the blood to lumen direc- tion. This transport was not of such magnitude as to account for the difference in net acetate and butyrate transfer rates. Because of their common substrates, the volatile fatty acids, comparisons might be made between the rumen epithelium and the intestinal mucosa. A distinction should be made, however, between the dual substrates of the intestinal mucosa where both long and short chain fatty acids are available in the intestinal contents, and the rumen epithelium, because of the paucity of free long chain fatty acids in rumen contents. Bloom et al. (12) established the difference between the transfer of short Chain fatty acids which appear free in the portal flow, l9 and the long chain acids (>C10) which appear mainly in the lymph in the form of triglycerides. Senior (39) wrote one review of work in the latter field. Absorption of the volatile fatty acids is thought to involve a combination of passive penetration and a fatty acid permease in the intestine. Micelle formation, proposed as a limiting factor in the decreasing capacity of fatty acids to be absorbed at higher chain lengths by Carr, as quoted by Coe (14), would be limited with the short chain fatty acids and passive penetration would be determined either by diffusion through aqueous channels or by transfer across lipid membranes. Benzenezwater partition coefficients are used as criteria of lipid:water solubility. The benzene:water partition coefficients would predict that passive pene- tration rates through lipid membranes should increase nearly three hundred times with an increase in chain length from two to six carbon atoms. Barry and Smyth (9) observed only a 20 percent increase in rate of ab- sorption ig_yiyg_as chain length increased from two to six carbon atoms. This would imply that passive transfer via lipid membranes is of minor importance for acids 20 below C6° If passive tranSport across lipid membranes may be based on the simple oil:water partition coeffi- cients, the failure to find increases in transfer corre- sponding to the increases in oil:water partition coeffi- cients indicates that (1) active transfer tends to in- crease with decreasing chain length, or (2) passive transfer by aqueous pathways is a major factor. In a report by Coe and Coe (15) and expanded by Coe (14) the hypothesis is advanced that the vapor pres- sure of the solute may be an important factor in the rate of penetration through biological membranes. According to their equations, passive transfer rate through a lipid membrane could actually decrease with increasing chain length. On this basis, it would be difficult to distin- guish between aqueous and lipid pathways. Effect of Volatile Fatty Acid Concentration on Disappearance from the Rumen Pfander and Phillipson (38) used the emptied, isolated rumen technique described by Danielli et al. (19) but modified it so that a mixture of volatile fatty acids was continually added to the rumen, approximating the 21 constant production found under most normal conditions. By adjusting the concentration of the acids in the mix- ture added, the concentrations in the rumen were held constant. The rate of infusion thus became the index of absorption rate at normal concentrations. The solution introduced into the rumen was Krebs—Ringer bicarbonate solution with sodium salts of acetic, propionic, and butyric acids added to give concentrations of 63, 21, and 15 m-moles per liter respectively. A similar solution containing different proportions of the three acids was added to the rumen. The actual losses from the rumen, in order of magnitude, were acetic > butyric > proprionic. A comparison of the rations added and the rations in the rumen at the beginning of the eXperiment indicated that the long chain acids disappeared at higher rates relative to their concentration when compared to acetic acid. Annison, Hill, and Lewis (2) added 50 m-moles of the sodium salts of the volatile fatty acids to the rumen of a similar preparation at one-hour intervals for five hours, then added 100 m-moles one hour later. Fatty acid concentrations were compared 30 minutes after addition of the sodium salts. Doubling the amount added resulted in 22 slight increases in the rumen concentration of acetic and propionic acids and a pronounced increase in the concen- tration of butyric acid. Smyth's laboratory (9, 42) studied the transport of short-chain fatty acids ;g_vitro and found that higher concentrations of fatty acids suppressed active transport. They reported an increase in amounts of volatile fatty acids lost from the mucosal side of intestinal epithelium with increasing mucosal concentrations until a maximum was reached. With further increases in concentration, there were decreases in amounts lost from the mucosal fluid. These maxima were observed at lower concentrations as the length of the volatile fatty acid carbon chain in- creased. The maximum mucosal losses were, in m-molar initial concentrations, acetic, 60: propionic, 40; buty— ric, 20: valeric, 20: and hexanoic, 10. The serosal gain maxima, in the same order, were observed at 40, 40, 20, 20, and 10 m—molar initial concentrations. After reaching maxima, further increases in mucosal concentration some— times reduced transport by almost one half. They were unable to observe this suppression of transport in vivo. 23 A possible parallel to suppression of epithelial transport may be noted in the work of Annison et al. (2) where the addition of sodium butyrate caused a large in- crease in rumen butyrate concentration when compared to the increase when equal amounts of acetate and propionate were added. Another aspect of concentration effect on volatile fatty acid disappearance from the rumen is the interaction Of ionic form of the volatile fatty acid and hydrogen ion concentration. When determining the effect of concentra- tions under conditions whereby changes in the relative proportion of the ionic forms will be changed, the con— centrations and proportions of the form must be specified. These factors and conditions will be discussed in the section on effect of pH. Effect of Hydrogen Ion Concentration on Volatile Fatty Acid Disappearance from the Rumen Change in rumen pH value may affect rumen volatile fatty acid absorption three ways: 1) Physical changes in rumen epithelium brought about by the action of the 24 hydrogen ions, 2) Changes in the proportions of the ionic forms of the volatile fatty acids, and 3) Changes in the animal‘s metabolism caused by the altered acid—base bal- ance of the blood, such as increased rumen and portal vein blood flow. At normal rumen pH values, physical changes in rumen epithelium caused by changes in pH value have not been shown to be detrimental to rumen absorption. A con- dition known as parakeratosis has been shown to be due to feeding rations high in readily fermentable carbohydrates, with a subsequent reduction in rumen pH value. It has not been shown that the changes in the rumen epithelium are due to changes in rumen pH value E§£.§2 or that the changes in rumen epithelium interfere with rumen absorp— tion. Discussion of the second way hydrogen ion concen— tration affects absorption of volatile fatty acid must begin with the consideration of the changes in prOportion of the two ionic forms, anions and free acid, that exist in the rumen at normal rumen pH values. Changes in con- centration of total volatile fatty acid may be slight in relation to the changes in relative concentrations of the 25 two forms caused by changes in pH value. At pH 7.0 only about 0.5 percent of the total acid is in the unionized form. Danielli §t_al.'s (19) original experiments on the effect of pH value on volatile fatty acid absorption led them to propose the hypothesis that the rumen wall is composed of water-filled channels in a fatty membrane. Dobson et_§l,'s (21) detailed study of the histological organization of the rumen epithelium does not show the existence of water-filled channels. Lavker g£_§1. (31), using an electron microscope, found the stratum corneum and stratum granulosum to be quite similar to the cellular arrangement of the epidermis. Lindhé and Sperber cited by Annison (la) demonstrated with the electron microsc0pe the presence of intercellular spaces of about 0.1“. In addi- tion to physical evidence, experimental evidence may be used to show the validity of the water-filled channels:lipid membrane model. Danielli §t_al, (l9) concluded that at pH 7.5 only the fatty acid anion was absorbed, that there was no unionized fatty acid absorbed, and that the fatty acid anion was accompanied by roughly a molecular equiva- lent amount of sodium. This loss of the fatty acid anion 26 is via the water-filled channels, the diameter of which are effectively large in proportion to the size of the butyrate molecule. They speculated that these channels are probably in the intercellular cement of the rumen epithelium. After further studies with the rumen con- tents at pH 5.8, they found a large loss of unionized acid from the rumen in addition to the fatty acid anion loss. The theory advanced was, that though some of the associated acid loss was via the water—filled channels, the greater part of it was via the lipoidal membrane surrounding the channels. As a consequence of the rela- tive importance assigned to these two means of transfer and differences in dissociation constants, at alkaline pH values the volatile fatty acids would be lost in this order: acetate > propionate > butyrate. At acid pH values the order would be reversed: butyrate > propionate > acetate. These differences would come as a result of the differential solubilities of the different acids in aqueous and lipid media as well as the differences in pKa' The results of Danielli §£_§1. (19) were challenged by Gray (22, 23) and Gray and Pilgrim (24) who could find no significant losses from the rumen at alkaline pH values. 27 They attributed the results of Danielli's group to the less sensitive analytical techniques employed. Gray found a 96 percent recovery in the rumen after six hours with his own technique, and 89 and 109 percent recovery with Danielli et al.'s technique. Ash (6), using acetate, propionate, and butyrate buffers, found changes occurring at pH 3.6 to 4.0 that indicated rapid absorption of free acid. Later, Ash and Dobson (7) investigated this problem from a different cause:effect relationship. Their investigations of effect of absorption upon acidity of the rumen contents indicated that about one half of the acetic acid was absorbed from the rumen in the unionized form and one half as the anion, even though only about 0.5 percent of the free acid is unionized at a neutral pH. Because the concentration of free acid is low, below one m-mole per liter, they sur- mised that a boundary within the epithelium must be much more permeable to the free acid than to the anion. Be- cause the proportion of acid is quite small at neutrality, Danielli et al. (19) had assumed that they were studying the permeability of the rumen wall to the anion alone. The results reported by Ash and Dobson (above) render this 28 assumption unjustified. These observations indicated that as larger amounts of free acid became available, that is, as the pH approached the pKa of a particular acid, the amounts of that volatile fatty acid passively absorbed would increase. Stevens and Stettler (46) increased the proportion of unionized form tenfold by decreasing pH and compared results from these trials with the results obtained by increasing the total volatile fatty acid concentration threefold. Both changes caused a doubling of acetate disappearance but only a 1.5 fold increase in butyrate disappearance. These results indicate that disappearance is not strictly proportional to amount of free or union— ized acid available. Factors Influencing Appearance of the Volatile Fatty Acids in the Portal Vein Appearance of material in the portal vein is the second observable event of a complex sequence of events in absorption of material from the rumen. Prior to its appearance in the portal blood flow, the material, or its precursors, must first have disappeared from the rumen 29 and any factor that affects rumen disappearance will af- fect appearance of the material in the portal blood flow. This section will deal only with those items that affect the processes that occur subsequent to disappearance from the rumen. Effect of Volatile Fatty Acid Chain_;pngth on Appearance in the Portal Blood Both volatile fatty acid chain length and rumen epithelium metabolism exert significant effects on appear- ance of the volatile fatty acids in portal blood. These effects were first discovered with ip_yiyp procedures and later confirmed ip vitro. Barcroft, McAnally, and Phillipson (8) studied the relative amounts of volatile fatty acids in the rumen veins, and reported that absorption was inversely related to chain length, a result confirmed by Danielli et al. (19) and Masson and Phillipson (33). Masson and Phillip- son measured both portal concentrations and disappearance from the rumen and reported that volatile fatty acid con- centrations in the rumen vein blood were not proportionate to the amounts of the acids leaving the rumen. 30 Kiddle, Marshall, and Phillipson (30) reported that rumen vein blood contained proportionately more ace- tate and less butyrate than the rumen contents, even when the rumen vein concentrations were corrected for arterial amounts of acetate. Pennington (37), using observations on the change in prOportions of butyrate and acetate as a basis for further work, has shown rumen epithelium to selectively metabolize butyrate. Bensadoun et al. (10) reported that the ratio of acetate to propionate in the portal blood of fed sheep was greater than the ratio of the two acids in the rumen fluid, this result indicating that propionate is more extensively metabolized than ace- tate in the rumen epithelium. Stevens and Stettler (47), using an ip vitro sys— tem, found less butyrate transported than either acetate or pr0pionate when all were added to the mucosal side in equimolar amounts and reported molar percentage serosal accumulations of 44, 40, and 16 for acetate, propionate, and butyrate, respectively. In an earlier paper, Stevens and Stettler (50) compared mucosal losses and serosal accumulations of acetate and butyrate and found nearly three times as much butyrate as acetate left the mucosal 31 fluid, but the relative amounts accumulating on the ser- osal side were almost reversed. Most of these differences were reflected in the amounts of ketone bodies recovered. Hird and Weidemann (25) found that rumen epithe— lium.ip vitro releases B-hydroxybutyrate preferentially on the blood or serosal side, regardless of whether it was exposed to the mucosal or serosal side. This uni- directional effect has not been shown in the metabolism of other volatile fatty acids. Influence of Rumen Concentration on Volatile Fatty AcidpAppearance in the Portal Blood The influence of rumen volatile fatty acid concen- tration on volatile fatty acid appearance and concentra- tion in the portal blood was first investigated by Masson and Phillipson (33), who found that acetic acid passes into the blood more rapidly at lower concentration than did either pr0pionic or butyric acids. The work of Anni- son et al. (2) indicated the close dependence of the vola- tile fatty acid concentration in the portal blood on the volatile fatty acid concentration in the rumen. 32 Stevens and Stettler (46) increased the volatile fatty acid concentration from 30 to 90 m-moles per liter and with acetate both the serosal gain and the mucosal losses made approximately the same increases. With butyrate, the mucosal losses made only about half the relative increase, while the serosal gain was somewhat more than the relative or threefold increase. Influence of Rumen pH Value on Volatile Fatty Acid Appearancepin the_§prtal Blood Dobson and Phillipson (20) reported that increased concentration of volatile fatty acids, as well as decreas— ing rumen pH value with volatile fatty acids present, caused an increase in rumen blood supply. An increased blood supply would effectively increase the concentration gradient between rumen contents and blood. Two later groups of investigators (ll, 59) found that both portal blood flow and rumen volatile fatty acid concentration attained maximum levels three to nine hours after feeding. The differences in mucosal disappearance resulting from changing proportion of ionic form versus increasing total concentration found by Stevens and Stettler (46) 33 were not paralleled directly in serosal accumulations. Instead they found both increased total acid and in- creased unionized acid resulted in three to fourfold increases in serosal butyrate accumulation, whereas acetate serosal accumulation did not increase as much with the increased unionized form as it did with in— creased total concentration. Increased disappearance and serosal accumulation at lower pH values could re— sult from either the increased transepithelial concen- tration gradient of the unionized form or a greater tissue permeability to this form. Spmmary of Literature Problems that must be considered with.ip_yiyp studies of rumen absorption include variable pool size, heterogeneity of rumen contents, control or replacement of saliva, rumen gasses, outflow through the omasal ori— fice, water movement through the rumen wall, and epi- thelial metabolism. Techniques for measuring absorption from the rumen use either disappearance from the rumen or appearance in the rumen veins or the portal vein as criteria of 34 absorption. As both have disadvantages, the technique used will depend upon the purpose of the experiment. In order to gain more control over some problems of rumen absorption, some experimenters have turned to rumen pouch, isotopic, perfused rumen, and ip vitro methods. Absorption from the rumen may be divided into two observable events, disappearance from the rumen and ap- pearance in the rumen or portal vein blood. Disappearance from the rumen is dependent upon volatile fatty acid chain length and concentration, and hydrogen ion concentration. In general disappearance is inversely proportional to chain length, increases with concentration, and increases with decreased pH value. These same factors, modified by selective epi- thelial metabolism, affect appearance in the rumen or portal vein blood. METHODS AND MATERIALS In order to study the transfer of material across the rumen epithelium with as simple a system as possible, an ip_yip£p system was developed and used. A forerunner of this device was described by Vidacs, Ward, and Nagy (58) and used to study transfer from rumen fluid to blood plasma. The device consisted of two Corningl 6780-glass ring seal joints, 2 cm in diameter, with synthetic rubber O—rings as seals. Each joint was bent and when two were held together by Corning #32430 clamps, they formed a U-tube with the rumen epithelium clamped across the base as a diaphragm, with an area of 3.14 cm2. The upper ends of the tubes were stoppered with #3 rubber stoppers, through which passed three polyethylene P.E. 2402 tubes. One of these carried gas in, extending close to the epi- thelial diaphragm to insure both saturation of the fluid 1 . . Corning Glass Works, Corning, New York. 2 Clay—Adams, Inc., New York, New York. 35 36 with the gas and mixing of the fluid. A second, which did not reach the surface of the solution, served as a vent for the escaping gas. The third tube, also extending close to the epithelium, terminated above the stopper with a lS-gauge hypodermic needle which served as a connector for sampling. The assembled U—tubes with their clamps were held upright in a wire basket. The wire basket was then sub- merged in a water bath at 39°C. The general experimental procedure was as follows: Krebs-Ringer bicarbonate (Ap- pendix I) was brought to 38°C in an insulated container and equilibrated for 15 minutes with a 95 percent oxygen- 5 percent carbon dioxide gas mixture at the laboratory. Simultaneously with the departure of one worker for the tissue, another proceeded to prepare the various bathing fluids according to the experimental protocol. For these fluids, Krebs—Ringer bicarbonate was used for both the fluid bathing the lumen (mucosal side) and the muscle (serosal side) sides of the tissue. Additions of mate- rials studied and pH adjustments were made at this time. The fluids were placed in glass and sealed with plastic sheeting to reduce gas exchange pending arrival of the tissue. 37 Sections of rumen wall, approximately 15 cm square, were removed from the ventral sac of the rumen as soon as it would not interfere with abattoir operation. Samples were selected with papillae from ca .25 to 1.0 cm in length. The rumen wall sections were rinsed once with warm tap water, then placed in an insulated container of oxygenated Krebs-Ringer bicarbonate for transporting back to the laboratory. When tissue was obtained from the Large Animal Clinic, a non-fasted animal from the Michigan State Uni- versity dairy herd with a functional rumen was used. A laparotomy was performed to remove a section of the caudo- dorsal blind sac of the rumen. Upon excision of the tissue, it was taken immediately to the laboratory. When tissue from either source arrived at the laboratory, the transporting fluid was gased with the oxygen-carbon dioxide mixture for three minutes, and then the tissue was rinsed twice in freshly oxygenated Krebs- Ringer bicarbonate. The epithelium was then dissected free of the muscle layers, using as little tension as possible. The epithelium was kept moist by dissecting under oxygenated Krebs-Ringer bicarbonate. 38 The epithelium was cut into squares large enough to cover the end of the glass joint and clamped in place. The fluids, 40 milliliters each, were added to the appro- priate reservoir on either side of the tissue. Fluids on the serosal side were immediately gassed with the oxygen- carbon dioxide mixture. After all U-tubes were positioned and oxygenated, the mucosal fluid was gassed with carbon dioxide. Both the carbon dioxide and the oxygen-carbon dioxide mixtures were first passed through water to sat- urate them in an attempt to reduce loss of fluid by evap- oration. The time elapsed between stunning the animal in the abattoir and oxygenation of the epithelial tissue was approximately one hour. The time elapsing between exci- sion of the rumen wall by biopsy and oxygenation was about 20 minutes. However, the blood supply had been interfered with to some extent for up to 15 minutes prior to exci— sion. Zero—time was designated as five minutes after oxygenation was initiated in the last of the U-tubes. Amounts of material in samples taken at this time were subtracted from final amounts to adjust for materials 39 adhering on membrane surface, or to confirm the presence of tears or cuts in the epithelium, etc. In the usual experimental procedure, the fluids on either side of the epithelium were gassed for the two-hour experimental period, samples being withdrawn at intervals for anal— ysis. Tissues were used for only one two—hour experi- mental period. The tissues were incubated in groups of ten ali- quots from one animal. All treatments to be compared were represented with the ten aliquots and treatments were replicated if possible. Hydrogen ion activity was measured as pH units using either a Beckman Model G3 pH meter or a Coleman Medallion4 pH meter. Samples were analyzed for volatile fatty acids by gas—liquid chromatography. A three M1 sample acidified to pH 2 to 3 was injected into a 200°C injection port connected to a 2.75 meter length X 2 mm teflon tube packed 3Beckman Instrument, Inc., Fullerton, California. 4Coleman Instrument Co., Maywood, Illinois. 40 with 10 percent FFAP5 on Chromasorb6 W, DMCS7 Acid washed, 80/100 mesh. A flame ionization detector, burning hydro- gen and air, ionized and quantified the volatile fatty acids as they left the column in the nitrogen carrier gas. The oven containing the column was operated at 140°C. Two instrument systems were used during the course of the work. The first system consisted of an Aerograph8 gas chromatograph coupled to a Sargent9 recorder, using either a one or two mv range plug. An Aerograph steam generator1 was used with this system. The alternate system was 5FFAP is the Varian Aerograph designation for Carbowax 20M treated with 2-nitroterephthalic acid. 6Chromasorb "W“ is a white diatomaceous earth material which was flux—calcined with about three percent sodium carbonate and has a surface area of one to 3.5 mZ/g. 7DMCS-treated Chromasorb is a Chromasorb support that has been coated with dimethyldichlorosilane to re- duce the surface active sites of the diatomaceous earth material. 8Hy Fi Model 600, Wilkins Instrument and Re- search, Inc., now Varian Aerograph, Walnut Creek, Cali- fornia. 9Model SRL, E. H. Sargent & Co., Chicago. 10 Model 675, Wilkins Instrument and Research, Inc., now Varian Aerograph, Walnut Creek, California. 41 ll . composed of an Aerograph oven, With flame detector con- 12 . nected to a Beckman electrometer. This system used a 13 . recorder With one mv full-scale travel. A locally fabricated gas saturator was used to saturate the ni— trogen carrier gas with water. The results were quan- tified by measuring the peak heights and comparing them to a standard curve constructed from standards injected into the gas chromatograph. Peak heights were compared only when all parameters were the same (machine, date, temperature of oven, injector, and room, instrument set— ting, and operator). Amounts of volatile fatty acids added, changes in pH, glucose additions, and other experimental variables will be detailed with the individual experiments. Back transfer was checked in a series of experi— - . l4 . . ments With C -sod1um butyrate. In these experiments 0.2 uCi of 1- Cl4-sodium butyrate in the normal 40 m1. of 11Model 550, Wilkins Instrument and Research, Inc., now Varian Aerograph, Walnut Creek, California. 2Beckman Instrument, Inc., Fullerton, California. 13 . Brown Electronic Model Y153X18(V)-II-III-(ll8)— (V)10, Brown Instrument Div., Minneapolis-Honeywell Reg. Co., Philadelphia, Pa. 42 Krebs-Ringer bicarbonate were placed on the serosal side of the tissue. Unlabeled sodium-butyrate in concentra- tions of 10, 20, 30, and 40 mM were added to the mucosal side. Tubes were incubated as described earlier. At the end of the two-hour period, the contents were treated as in other experiments except that one milliliter samples from serosal and mucosal fluid were counted by liquid scintillation. The samples were added to 15 m1 of Wer- bin's solution (61), equilibrated at 32°F., and counted three times at 20 minutes each in a 720 System liquid scintillation counter.14 Then 1046 dpm of standard C14- benzoic acid was added to each vial and the vials were recounted three times at 20 minutes each from which count- ing efficiency was calculated. The results were analyzed according to standard methods as outlined by Lewis (32) and Snedecor (43). 4Nuclear-Chicago, Division of Nuclear—Chicago Corporation, Des Plaines, Illinois. RESULTS AND DISCUSSION Effects of Changing Mucosal Concentrations on Serosal Accumulations The first question to be investigated was the in- fluence of mucosal concentration of volatile fatty acids on their serosal accumulation. To study this effect so- dium salts of volatile fatty acids were added to the mu- cosal solution to a final concentration of 20, 40, 60, or 80 mM in the case of acetate and 10, 20, 30, or 40 mM for propionate and butyrate. After a two—hour experimental period, samples were removed from the U—tubes, acidified, and volatile fatty acids quantified. Volatile fatty acids other than the one added to the mucosal fluid were not detected. In conjunction with other trials, three dif- ferent tissues were incubated without mucosal fatty acid and no volatile fatty acids were detected in the serosal fluid at the end of two hours. Results for the incubations with acetate, pro- pionate, and butyrate are given in Tables 1, 2, and 3, respectively. The three tissues incubated with acetate 43 44 TABLE 1 EFFECT OF MUCOSAL CONCENTRATION ON SEROSAL ACCUMULATION OF ACETATE Mucosal concentration Serosal Confidence Sodium A etate a umulation Interval c CC (P < 0.05) mM n -- ----- p-moles/Q hr. ------- 20 6 8.4 11.4 40 7 11.1 10.4 60 8 15.5 11.4 80 8 26-5 14.5 TABLE 2 EFFECT OF MUCOSAL CONCENTRATION ON SEROSAL ACCUMULATION OF PROPIONATE Mucosal concentration Serosal Confidence Sodium Pro ionate accumulation Interval P (P < 0.05) mM n ------- u-moles/2 hr. ------- 10 8 4.3 11.2 20 7 7.2 10.9 30 6 11.6 11,9 40 8 19.2 13.4 45 TABLE 3 EFFECT OF MUCOSAL CONCENTRATION ON SEROSAL ACCUMULATION OF BUTYRATE Mucosal concentration . Serosal a m lation Sodium butyrate CCU u mM ------- p-moles/2 hr. ——————— Trial A Trial B 10 1.28, 1.24 0.63 20 2.30. 3.52 1.16 30 6.60, 6.70 2.80 40 4.50 2.91 responded uniformly so that results from all three tissues could be combined. Similarly, the results from incubation of three different tissues with propionate could also be combined. In contrast to the results with acetate and propionate, the two tissues used with butyrate differed by twofold. Nothing was known of the treatment of the animals prior to slaughter, differences in pre—slaughter conditioning may have contributed to the differences in serosal accumulations. Fasting would be expected to lower the reserves of metabolizable material within the rumen and the rumen epithelium, causing the epithelium to retain more butyrate and release more of it as ketone bodies. Armstrong and his colleagues (4, 5) reported that a 46 mixture of acids was absorbed more rapidly in sheep pre- viously fasted four days than in fed sheep. Because these workers used disappearance from the rumen fluid as cri- teria of absorption, they were not able to define the time when volatile fatty acids started appearing in the blood. The graph (Figure 1) of the serosal accumulations of the volatile fatty acids as a function of their mucosal concentrations has four features to discuss. They are: l) The differences in magnitude or amounts of the serosal accumulations of the volatile fatty acids. 2) The differences in slopes or sample regression lines of the different volatile fatty acids. 3) The departure of the sample regression lines from linear regression, indicating changing amounts of accumulation with increasing mucosal concentra— tions. 4) The intersection of the lines representing pro- pionate and acetate accumulations. The first feature, the differences in magnitude, was examined by testing the differences between the mean 47 12‘ Serosal Accumulation u—moles/Q hours l—‘ p. Fig. l.-—Serosal Accumulation of Volatile Fatty Acids as i r v ? 10 ; 20 3o 40 so 60 70 so Initial Mucosal Concentration, mM a Function of Mucosal Concentration. 48 accumulation of the volatile fatty acids at the same mucosal concentrations according to the method given by Lewis (32). The results are shown in Table 4. TABLE 4 DIFFERENCES IN MEANS OF SEROSAL ACCUMULATIONS OF THE VOLATILE FATTY ACIDS Level at , Mucosal . Ac1ds Compared . which means Concentrations . are different mM P < Acetate vs. Propionate 20 .10 Acetate vs. Propionate 40 .01 Acetate vs. Butyrate A 40 .01 Propionate vs. Butyrate A 10 .10 Propionate vs. Butyrate A 20 .01 Propionate vs. Butyrate A 30 .01 Lower serosal accumulation with increasing carbon chain length at 20 u-mole/ml mucosal concentrations may indicate differences in relative rates of metabolism of the volatile fatty acids. At lower mucosal concentrations, butyrate accumulates in the serosal fluid at the lowest rate. At 20 p—moles/ml mucosal concentration, these 49 comparisons indicate a much greater transfer relative to concentration with the shorter chain lengths in agreement with Barcroft et a1. (8). Kiddle et al. (30) found that the rise in concen- tration of butyrate in the blood draining the rumen was significantly less than the decline in rumen concentration of butyrate. In recent ip_vitro work, Stevens and Stettler (46) found that of 16 u-moles of butyrate lost from the mucosal solution per cm2 per 2.5—hour experimen- tal period, only about 6 percent reached the serosal solu- tion as butyrate, with about 50 percent being released as ketone bodies, chiefly acetone. Hird and Weidemann (25) also found a large proportion of butyrate metabolized to ketones with ip_vitro preparations. A similar case seems to be true for propionate, though to a lesser extent. Stevens and Stettler (46) found that as much as 85 percent of the butyrate lost from the mucosal bath was metabolized, but only 60 to 70 per- cent of the propionate lost from the mucosal fluid was metabolized. About 88 percent of the butyrate appeared as ketones, but none of the propionate reappeared as ketones. Lower tissue metabolism could account for the 50 higher relative transfer of propionate than butyrate at the lower mucosal concentrations. The higher relative transfer of acetate at the 20 p-moles/ml mucosal concen- trations would be a reflection of the lower tissue metab- olism of this fatty acid (Fig. l). The second feature is the different slopes of the lines representing the serosal accumulations of the dif- ferent acids. When the serosal accumulation values are used in a least—squares formula, the line is described by the equation: y = mx + b. The lines representing the volatile fatty acid accumula- tions are represented by the following equations: Acetate y = 0.3013 X — 0.1788 +0.88 Propionate y = 0.454 X — 2.495 +0.93 Butyrate A y = 0.1655 X - 0.0484 +0.79 Butyrate B y = 0.1005 X — 0.306 +0.52 The slope, or sample regression coefficient, de- scribes the serosal accumulations as a function of in- creasing X, or mucosal concentration. As the slopes of these lines are all positive, they indicate that with 51 increasing mucosal concentration, more of the volatile fatty acids will accumulate in the serosal reservoir. The magnitude of the term indicates the rate of change with increasing mucosal concentration. The value of 0.454 for the prOpionate slope indicates that propionate accumulations will increase more rapidly with increasing concentrations than either acetate or butyrate. The values for the correlation coefficient, "r" for acetate, propionate, and butyrate A indicate that the regression coefficients for these acids are significantly different from zero. Further analysis indicates a significant (P < 0.01) difference between the acetate and propionate regression coefficients. Due in part to the small number of values, the butyrate A regression coefficient could not be distinguished from those of acetate and propionate. These results seem to confirm the findings of Pfander and Phillipson (38) in that increasing mucosal concentrations increased the transfer of the longer chain acids to serosal solutions more than it did the shorter chain acids, at least in the comparison of acetate and propionate. The present study does not completely par- allel Pfander and Phillipson because appearance was 52 measured, not disappearance. The smaller sample regres- sion coefficient with the increase in chain length can also be attributed to a change in the proportions of material passing through the water—filled pores and the lipid membrane proposed by Danielli (19). The amount passing through the water—filled pores may indeed change only in proportion to the larger amounts available, but the greater lipid solubility of the longer chain acids may allow a larger amount to pass through the lipid mem— brane resulting in higher serosal accumulations of the longer chain acids. Annison et a1. (2) found that dif— ferences in portal concentration were greater in the longer chain volatile fatty acids than with acetate with equivalent rumen concentration changes. When the changes in portal concentration are divided by the changes in ruminal Concentrations necessary to produce them, there is a progression of 0.70, 0.86, to 1.00 for acetate, pro- pionate, and butyrate, respectively. The butyrate accumulations were smaller, both in amount and increase with concentration. The smaller amount probably represents greater metabolism by the epithelium. The smaller regression coefficient indicates 53 less dependence upon concentration, showing that the me— tabolic pathways are using enough of the available butyrate so that serosal accumulations are almost independent of concentration. If the decrease in serosal butyrate ac- cumulations at 40 m-molar initial mucosal concentrations is real, this would be a parallel to the inhibition of active transport of volatile fatty acid observed by Barry and Smyth (9). Stevens and Stettler (47) found that butyrate disappeared ip_vitro from the mucosal fluid two to three times as fast as acetate at pH 7.4. However, they found only about 23 percent of the acetate disappear— ing from the mucosal side appeared as volatile fatty acid in the serosal fluid. They also found somewhat more than three times as much acetate appearing as the volatile fatty acid, indicating greater epithelial metabolism with butyrate. Despite the general inadvisability of extrapolat- ing regression values beyond the range of values covered by the data, it is interesting to note that the b or in- tercept term has a negative coefficient. Though a nega— tive concentration is meaningless, it does point out the probability that at low concentrations all the available 54 volatile fatty acid could be used for epithelial metab— olism. The intercepts were tested statistically and were not significantly different from each other or from zero. The third feature on Figure 1 is the increasing slope with increasing mucosal concentrations. To test the significance of this deviation from linear regression, the sum of squares for group means was partitioned ac- cording to the method of Snedecor (43b) into two parts, one attributed to linear regression, the other to devia- tions from linear regression. When the F-ratios were tested, they indicated more than random sampling departure from linear regression. These calculations are shown in Table 5. The significant F ratios due to concentration and deviations from linear regression indicate that, in addi- tion to a significant effect of concentration, there is at least one other factor influencing serosal accumula- tions. As a strict semi—permeable membrane would only be affected by concentration, other explanations are needed. Of the other explanations, three are most likely: 1) Changes in permeability of the aqueous layer to the anion. 55 TABLE 5 ANALYSIS OF SUMS OF SQUARES FOR GROUP MEANSl Source of Variation Degrees Of Mean Freedom Square Acetate Concentration 3 478.0 Linear Regression 1 Deviations from linear regression 2 69.52 Error 25 9.48 F ratio 69.52/9.48 = 7.33** Propionate Concentration 3 343 Linear Regression 1 Deviations from linear regression 2 185.83 Error 25 6.01 F ratio 185.83/6.01 = 30.92** **P < 0.01. 1According to Snedecor (43b). 2) Changes in permeability of the lipid layer to the unionized form. 3) Saturation of the volatile fatty acid utilizing pathways in the epithelial cell, resulting in more passage across to the serosal reservoir. 56 Saturation of the metabolic pathways in the epi- thelial cell is the more plausable explanation. Propio- nate is more likely to be metabolized within the epithe- lium than acetate and also propionate had a greater departure from linear regression in the range of mucosal concentrations studied. Due in part to the smaller number of values, the test for the two butyrate series did not show a significant departure from linear regression. An attempt to reduce the departure from linear regression by using the logarithms of the accumulation values and testing them by the same method was unsuccess— ful in reducing the deviations from linear regression. The fourth feature of Figure l is the intersection of the lines representing serosal accumulations of acetate and prOpionate. The lines intersect possibly as a conse- quence of greater epithelial metabolism of propionate compared to acetate, causing propionate to have a lower accumulation at 20 mM mucosal concentration and the larger value of the regression coefficient of rate of accumula- tion. Apparently epithelial metabolic pathways for pro- pionate are saturated at 40 p—moles/ml the next direct comparison, and the mean for prOpionate accumulation is 57 greater (P < 0.01) than the acetate accumulation mean. This could also be a consequence of the higher lipid solubility of propionate. This changing of relative amounts transferred at different concentrations may ex— plain some of the difference existing in the literature. Different values in the literature can be ex- plained by this crossing of serosal accumulation lines. For example, Masson and Phillipson (33) found that ace- tate passed into the blood stream in larger quantities than did the other acids at equally low concentrations. Annison et al. (2) found more rapid increases in portal concentrations with increased ruminal butyrate concentra- tions. Stevens and Stettler (46) found that at higher mucosal concentrations of butyrate an increase in mucosal concentration resulted in a larger than equivalent in- crease in serosal accumulation, while under the same con- ditions, acetate serosal accumulations did not respond to the same extent. All these examples may be taken as indicative of the change in mucosal accumulation brought about by a smaller proportion of longer chain acids being metabolized in the tissues at higher mucosal concentra- tions and the increased lipid solubility of the longer chain acids. 58 Comparison of Slaughter— Versus Biopsy-Obtained Tissue In order to test the viability of the tissues used a comparison of responses of tissue obtained by rumen biopsy with those from slaughter material was made. Rumen biopsies were done on animals leaving the Michigan State University dairy research herd. Experimental conditions were similar for biopsy and slaughter material. Two-hour accumulations with the experimental parameters are shown in Table 6. The differences between the means of the two tissue sources were calculated and tested according to Lewis (32) and the results shown in Table 6. The significant or highly significant differences between the biopsy and slaughter material, with the slaughter material having a higher accumulation in all cases at pH 7.4 indicates the possibility that the epithelium from the biopsy material is metabolizing more of the available volatile fatty acid. No explanation is offered for the results with acetate at 4.0. Stevens (45) compared slaughter material with biopsy material and found the electrical properties favored the biopsy material. As much of the experimental procedures 59 .Aao.o v av mocmummmae .mam11 .Amo.o v as mocmuwmmae .mamt 116H.6 Ame om.ma Aoav oa.mH ow mumcoamoum m.a a 166.4 lav om.a Aoav HH.6 om mumcoaaoua m.a m os.omn Ame 66.6m Aoav om.m6 om mumomua 6.4 N 1166.6 Aoav ov.mH Aoav ea.6 om mumumom m.s H IIIIIIIIIIIII Moon m\mmHoEI1IIIIIIIIIIIII SE coaumuu meOHm HoufimsmHm meOHm Icmocoo ¢m> tumormsmam am> ma aways mUCCHCMMHQ coaumHsfisoom Homonom HGUOEMH mm H wwOUSS. AflHMMBmE MMBEOD¢Am mDmmm> MmmOHm o mqm48 60 used in his experiments required the improved electrical properties, only biopsy material was used in further ex— periments. There was no comparison of slaughter material and biopsy material for other properties. The biopsies for his work were taken through a rumen fistula, removing only a small area each time which permitted using rumen tissue from one animal for a series of experiments. Re- peated operations were not utilized in this experiment. Effect of Adding Glucose to Serosal Fluid on Serosal Accumulation of Acetate and Propionate This series of experiments investigated the ef- fects of adding 29.6 mM glucose to the serosal fluid when acetate and propionate were added to the mucosal fluid at 20, 40, 60, and 80 mM and 10, 20, 30, and 40 mM final con- centration for acetate and propionate, respectively. The aim of this series of experiments was to determine if: 1) Glucose would supply energy to enhance volatile fatty acid transfer. 61 2) Glucose would spare or reduce the amounts of fatty acids metabolized, allowing more serosal accumula- tion. 3) Increasing the osmolarity of the serosal fluid would affect serosal accumulation. Accumulations reported in Table 7 show that glu- cose tended to inhibit serosal accumulation of volatile fatty acids, although differences between means were not significant. Whether transfer of volatile fatty acids is carrier mediated or not, the addition of glucose to the serosal fluid did not enhance transfer. Thus increased blood glucose would not enhance fatty acid transfer. This may be a protective device, whereby fatty acid ab— . sorption is not dependent upon glucose, a source of energy in somewhat short supply. In answer to the second question, glucose did not spare volatile fatty acid metabolism, as evidenced by no significant differences in the means. The lack of signif- icant differences also negates the third possibility (ef— fects due to osmolarity) eSpecially as no differences were observed in fluid volumes. 62 TABLE 7 EFFECT OF ADDING GLUCOSE TO SEROSAL FLUID ON SEROSAL ACCUMULATION Serosal . , Difference Mucosal Accumulation . Glucose— Concentration Control Glucose Control mM u-moles/Z hrs. Sodium Acetate: 20 6.47 10.40 —3.93 40 11.11 11.51 —0.40 60 13.39 17.85 -4.46 80 26.98 25.17 +1.81 Sodium Propionate: 10 5.06 3.56 +1.50 20 6.70 11.01 -4.31* 30 10.25 13.12 -2.78 40 13.35 21.75 —8.40** *(P < 0.05). **(P < 0.01). Effect of Serosal Butyrate In order to investigate the effect of butyrate on the accumulation of the other volatile fatty acids, 0.8 mM sodium butyrate was added to the serosal side. The com- parisons were made with various experimental parameters on the mucosal side, including change in pH, concentrations 63 of volatile fatty acids, different volatile fatty acids, and combinations of acids. The results of these comparisons are presented in Table 8. Serosal butyrate had no effect on acetate or propionate accumulation at pH 7.4. Serosal butyrate in- hibited both acetate and propionate accumulation at pH 4. In the intact animal there is normally little butyrate circulating beyond the liver. The liver removes almost all of the butyrate appearing in the portal vein. Hird and Weidemann (25) found that sheets of rumen epithelium, ip_vitro, synthesized greater amounts of ketone bodies when butyrate was presented to the muscle or serosal side than when presented to the mucosal side of their prepara— tion. They suggested that either the muscle side is more permeable to butyrate or that synthesis of ketone bodies is Spatially closer to that surface. If the acceptance of acetate into the mucosal surface of the cell was a carrier mediated process or a process dependent upon the metabolism of the cell, the presence of butyrate at the more permeable serosal surface might cause the cell not to accept acetate as rapidly and it would not be avail- able to the serosal side. 64 TABLE 8 EFFECT OF SEROSAL BUTYRATE ON 2-HOUR SEROSAL ACCUMULATION OF ACETATE AND PROPIONATE With Volatile Mucosal Butyrate Control Difference Fatty Concen- pH Butyrate- Acid tration Accumu- Accumu- Control lation lation n mM ------- n-moles/2 hours ——————— HAC 4 80 7 4 13.5 14.0 - .5 HAC 4 80 4.0 34.6 76.4 -4l.8** HAc 4 80 4.0 33.6 54.0 —20.4 HPr 2 40 7.4 15.4 12.4 + 3.0 HPr 4 80 4.0 26.3 51.6 -25.3** HAC/HPr 2 80 HAC 4.0 32.4 44.6 -12.2 HPr 4.0 28.8 24.6 + 4.2 **(P < 0.01). The effect of serosal butyrate upon the transfer of propionate at pH 7.4 is easier to explain, though the difference only approaches significance. Here the mar— ginal amount of propionate available to the cell is spared by the serosal butyrate. 65 Effect of Sodium Cyanide on Serosal Accumulation of Acetate In an attempt to determine if the processes by which the volatile fatty acids were transferred were ac- tive or passive, 10-3 molar sodium cyanide was added to the serosal side of the membrane in a group of five tubes which were compared with a group of five tubes Without sodium cyanide. The fluid on the mucosal side contained 80 mM sodium acetate in Krebs-Ringer Bicarbonate, pH 7.4, and the tubes were incubated for two hours. The serosal acetate accumulation averaged 25.7 p-moles/Q hr. NaCN, 24.6 p-moles/2 hr. without NaCN, a difference of 1.1 u-mole, which is not significant. There are three possibilities which may be used to ex— plain the results: 1) If an energy dependent process for acetate trans- port Operates in the cell, it is not oxygen de— pendent. 2) Enough energy is stored as chemical intermediates to support transport in the absence of oxygen. 66 3) Serosal accumulation is the result of passive transport through the rumen epithelium. The first two possibilities are not probable. Glycolysis, the normal metabolic pathway when oxygen is not available, would not have any normal substrate in the chemically defined media used. The possibility of a sufficient supply of energy to support transfer as chemical intermediates is unlikely, if not impossible. Because of the wide range of enzymes inhibited by cyanide, Specifically those in the cytochrome chain, the rumen epithelium is unlikely to be so different as to have a system unaffected by cyanide. Hird and Weidemann (30) found that oxygen stimulated conversion of butyrate to B-hydroxybutyrate in rumen epithelium. If an active system for mucosal to serosal transport of acetate exists in the rumen epithelium, its effects were small. In view of the small increase in serosal accumu— lation with cyanide treatment, cyanide might have blocked the mechanism proposed by Stevens and Stettler (48) for active acetate transport from the blood to rumen. Less back transfer could account for the larger net serosal accumulation. 67 Effect of ChangingppH on Acetate and Propionate Accumulation in Serosal Fluid To investigate the effect of changing the propor- tions of ionized and unionized acid in the mucosal fluid, trials were conducted at different pH values. The pH of the mucosal solution was adjusted to a predetermined level by adding sulfuric or hydrochloric acids. These acids were added to the buffer with the volatile fatty acids, before exposing the fluid to the membrane. The results are shown in Table 9. Changing the pH from 7.4 to 6.0 with sulfuric acid did not cause a significant change in serosal acetate accumulation. Further adjustment of the pH from 6.0 to 4.0 with sulfuric acid caused a significant (P < 0.01) increase in serosal acetate accumulation, despite the fact that there was no significant difference in accumu- lation between the control tissues. The use of hydrochloric acid to alter the pH also resulted in a significant (P < 0.01) increase in serosal accumulation relative to the control. Comparing sulfuric and hydrochloric acids as adjusting agents, there was a larger increase (P < 0.01) in serosal accumulation with 68 TABLE 9 EFFECT OF CHANGING MUCOSAL HYDROGEN ION CONCENTRATION ON SEROSAL ACETATE AND PROPIONATE ACCUMULATION . . Serosal Accumulationl Difference Adjusting . A id pH Adjusted- C Adjusted Control2 Control - ---------- u-moleS/Q hr. -------- -~-- With 80 mM sodium acetate on mucosal side: H2804 6.0 12.95 14.18 - 1.23 H2304 4.0 34.6 13.77 20.83**3 HCl 4.0 65.4 6.57 58.83?”3 With 80 mM sodium propionate on mucosal side: H2804 4.0 71.2 13.79 51.41** n = 5 2 . . . Krebs-Ringer Bicarbonate at pH 7.4 on mucosal Side. 3Significant (P < 0.01) difference between increases caused by H2S04'and HCl. **Significantly different (P < 0.01). the hydrochloric acid, despite the significantly (P < 0.01) smaller accumulation in control tissues. The difference in serosal accumulation in the com- parison between hydrochloric acid and sulfuric acid as the pH adjusting acid is evidently due to differing 69 actions of the chloride and sulfate ions on the epithe- lium. The mechanism of this action is not presently known. Propionate incubated tissues responded to pH change in much the same way. When tissues exposed to 80 mM sodium propionate adjusted to pH 4.0 with sulfuric acid were compared with tissues under the same conditions at pH 7.4, there was a significantly (P < 0.01) larger serosal accumulation at the lower pH value. Changing pH has at least three ways of affecting serosal accumulation or transfer by rumen epithelium. The first of these is the direct effect of the hydrogen ions on the rumen epithelium. The second is the change in molecular or ionic Species of the weak electrolytes in the rumen. The third is an inhibition of epithelial me- tabolism, with a larger amount passing into the serosal fluid as a result. Between pH 7.4 and 6.0, any effect of changing hydrogen ion concentration on the epithelium itself must be slight. However, pH 4.0 is beyond the range of normal pH values for most tissues. The change in prOportion of molecules existing as the two Species of volatile fatty acid brought about 70 by changing the pH from 7.4 to either pH 6.0 or pH 4.0 is extensive. Free acid increases 25 times from pH 7.4 to pH 6.0 and 100 times from pH 6.0 to 4.0. The changes in serosal accumulation were not of this magnitude. The question is not so much whether serosal ac- cumulation is 25 to 100 times as much but Whether it in— creases as much as regression equations on Figure 1 would predict. If we may assume that the regression equation for total acid is applicable to free acid, then the in- crease in free acid caused by reducing the pH is 100 times. The sample regression coefficient, 0.301, times the 100 increase in free acid would be 30, which is less than the 58.8 difference for hydrochloric acid adjusted tissue, but more than the difference for the sulfuric acid adjusted tissue. However, the observed differences were within the precision of the prediction equation. Ash (6) used acetate, propionate, and butyrate buffers and found changes occurring at pH 3.6 to 4.0 that indicated rapid absorption of the free acid. Ash and Dobson (7) found about one half of the acetic acid ab- sorbed from the rumen as the unionized form and one half as the anion, even though only about 0.5 percent of the 71 total acid is unionized at a neutral pH. Because the concentration of unionized acid was low, less than a m— mole per liter, they surmised that a boundary within the epithelium must be much more permeable to the free acid than to the anion. These results render unjustified Danielli et a1.'s (19) assumption that because the pro- portion of free acid is reduced to small prOportion at neutral pH, only the anion leaves the rumen. Ash and Dobson (7) indicated that as the pH of the rumen fluid approached the pKa of a particular acid, the amounts of that volatile fatty acid passively transferred would in— crease. Stevens and Stettler (46) increased the concen- tration of the unionized forms of the fatty acids acetate and butyrate tenfold by lowering the pH from 7.4 to 6.4 and found that this change affected the two acids differ- ently. Acetate transport increased 1.5 times under these conditions, while butyrate transport increased three- to fourfold. When these results are compared with the re— sults of changing the amount of sodium salt of the par- ticular fatty acid added to mucosal solution, acetate accumulation increased in proportion to the rise in total 72 mucosal acid, but when the proportion of unionized form was increased acetate transfer did not increase in pro- portion to the increase in unionized form. A sidelight is the close comparison between the amounts of acetate and propionate accumulating (113.8 M“ moleS/2 hr.) under the control conditions of 80 mM at pH 7.4 (Table 9). This was considerably less propionate accumulation than would be predicted by extrapolation from Fig. l which may indicate that propionate, as well as butyrate, inhibits serosal accumulation when added to the mucosal fluid at higher concentrations. However, acetate accumulation in Table 9 was slightly less than predicted from Fig. l. The differences in serosal accumulations for con- trol tissues, both in this series of experiments and in others where control tissues were necessary, have at least two possible explanations. The first and more probable is the differences in papilla length and epithelial thickness. The longer the papilla or the thicker the epithelia, the longer distance and the more barriers volatile fatty acids must traverse to appear as serosal accumulations. The second reason, somewhat related, is 73 differences in tissue utilization of volatile fatty acid. Here again, more tissue or greater depth of tissue would metabolize more volatile fatty acid. Study of Serosal to Mucosal Transfer Using C14-Sodium Bupyrate To study the amount of back transfer, tubes with 0.2 uCi as 13 u-moles of sodium butyrate—l-14C in the serosal fluid were compared to tubes without serosal sodium butyrate. The mucosal amounts, as shown in Table 10, were 10, 20, 30, or 40 mM unlabeled sodium butyrate. At the end of the two—hour incubation period, samples were taken for liquid scintillation counting as well as for chromatographic analysis. Addition of 0.325 mM sodium butyrate to the ser— osal side had no effect on the serosal accumulation of butyrate. This is in contrast to the effect of adding 0.8 mM sodium butyrate on the accumulation of acetate and propionate (Table 8). While the back diffusion as measured by the radioactivity was small, always less than one percent, it was larger than accumulation of butyrate in the 74 .mocmHCMMHU ma CSHMHSUHMO 6 .mocmHonmm mom pmuomnHOUm .zmo em.mm u pesoumxomme .mUoHnmm mCHucsoo mUSCHEIhucm3u .m mo mmmum>4w .HoHuumo 6cm mumumusn ConQmH mm ©6666 me0811 MH Rom CCDUSHHOON . o o H Q o n m H o.mH Hh.o vHM6 N.6v mm.N 06 OH N.NH 06.0 Hmm6 H.5N mm.m 06 m m.mm 06.0 6¢H6 m.mm H6.N 0m 6 o.vm hm.o 5006 N.mm 6m.H ON 5 m.mH mm.o N666 m.mm 56.0 0H 6 Hm.N 06 m om.m om v 66.6 om m 6H.H om N N6.o OH H as; Q; . A: «\moHOSIa :6 osmmHB Homoumm Homoosz coHumuucmocoo 6 ma mconDMMHQ mcoHumHsEsoo¢ mumnmusm meEmm Homm Hmmoumm omchumm v.MHE\EmO umz Homoosz H OH mnmdB mmmmz C ox N t 1 n .L 1 GI J j Fig. 30 60 90 120 minutes 3.--Time Sequence of Propionate Serosal Accumulation. Mucosal Source Conc. pH Acid mM Slau 40 7.4 Biopsy 40 7.4 Slau 80 4.0 H SO 80 l) The fact that the fatty acid ion or molecule, as the case may he, must traverse two membrane sur- faces and the distance between them. 2) Any metabolic uses for the fatty acid going through the cells themselves are first satisfied before the excess is available to pass through the serosal membrane. Lowering the pH of the mucosal fluid increased the total amount transferred and also caused transfer to occur with less time lag. This earlier accumulation with lower pH could reflect larger amounts of unionized fatty acid available to the cell or an inhibition of the cell metabolic processes or both. The serosal accumulations are not direct measures of transport rate in the usual enzyme kinetic sense of showing maximum reaction rates. The second feature is the failure to reach equil— ibrium or a steady state. Normally a graph of reaction products as a function of time consists of three phases: 1) lag, 2) rate, and 3) steady state. In measuring and comparing two hour accumulations we are measuring at least 81 one and probably at least two of these phases. In the absence of longer than two—hour experimental periods, there is no eVidence that the steady state phase has been reached. Further work is indicated to locate these curves on the usual reaction products curve. CONCLUSIONS DeSpite the apparent failure to measure the con- stant maximum rate of fatty acid transfer due to the time lag, this lag was reduced at higher mucosal concentrations and lower pH values but was independent of fatty acid chain length. Comparisons among fatty acids at two hours are a measure of their relative transfer rates. Within these limitations, the following conclusions are valid. Passage of the volatile fatty acids through the rumen epithelium was influenced by chain length. The shorter chain length acids accumulated in the serosal fluid at two hours in larger amounts at lower mucosal concentrations, except for acetate vs.pr0pionate at 20 mM. At higher mucosal concentrations propionate accumulated in larger amounts than did acetate. Butyrate accumulation was not appreciably affected by mucosal concentration in the ranges used. These results are consistent with the premise that the volatile fatty acids are transported by a non-active process modified by the metabolism of the 82 83 epithelial cells. The metabolic processes become satur- ated, allowing larger amounts to accumulate. With higher mucosal concentrations more fatty acids accumulate in the serosal fluid possibly because the epithelial metabolic pathways become saturated. In a comparison of tissue obtained by rumen biopsy and tissue obtained from the abattoir, the slaughter tissue allowed significantly larger serosal accumulation, a result consistent with larger amounts used in metabolic pathways in the tissue obtained by biopsy. Serosal glucose had no effect on acetate accumu- lation at two hours and significantly decreased propionate accumulation. Glucose would not be a normal source of energy for the cell and it would not spare volatile fatty acid metabolism by the epithelial cells, or add to its transfer. Sodium butyrate added to the serosal fluid signif- icantly reduced both acetate and propionate accumulation for the one mucosal concentration and did not increase accumulation with either acetate or propionate. Sodium cyanide in the serosal fluid did not cause a significant difference in acetate accumulation. The non-Significant increase may have been due to either a wizllll 1 III... I all ll Ill-l .t..l.. . I I I I 84 reduction in acetate metabolism or a reduction in a serosal to mucosal active transport. Adjusting the pH to 4.0 significantly increased two-hour serosal accumulation of both acetate and pro— pionate. The increase was larger when hydrochloric acid was used as the adjusting agent than when sulfuric acid was used. The increase in serosal accumulation with the change in pH is probably due to the change in proportion of anion and unionized forms but this does not eXplain the change in accumulation caused by the change in ad- justing agent. Serosal to mucosal transfer measured with C14— butyrate moved a larger percentage of carbon than did mucosal to serosal transfer. Back diffusion did not affect other results be- cause of the small amounts available in the serosal fluid. APPENDIX I KREBS-RINGER BICARBONATE (56) SOLUTIONS: Parts 1. 0.90%.NaCl (0.154M) 100 2. 1.15% KCl (0.154M) 4 3. 1.22% CaCl2 (0.11M) 3 4. 2.11%KH2PO4 (0.154M) l 5. 3.82%MgSO4 ° 7H20 (0.154M) 1 6. 1.30%INaHCO3 (0.154M) 21 The solutions were combined in the above propor- tions, gassed for 10 minutes with 95% 02-5%CO2 and kept cold and stoppered until used. Fresh solutions were pre- pared for each day's experiments. 85 APPENDIX II 14 EFFICIENCY DATA FOR COUNTING C SODIUM BUTYRATE CPM/ml. Efficiency Sample Sample From Percentage * Background Standard Standard Solution 1** 51.53 679.83 628.30 60.06 2** 53.15 635.92 582.77 55.70 Mucosal 6 74.86 687.56 612.80 58.58 7 77.60 666.28 588.68 56.28 8 105.06 683.01 577.95 55.25 9 79.48 658.76 579.28 55.38 10 98.58 681.60 583.02 55.74 Serosal 6 6604.3 7308.2 703.9 67.29 7 6059.3 6769.7 710.4 67.92 8 6198.5 6906.3 707.8 67.67 9 6303.5 6943.9 640.4 61.22 10 6368.0 7043.2 677.2 64.74 *1046 DPM added in standard. **Solution 1 and Solution 2 were Krebs-Ringer Bicarbonate plus scintillation mixture and served as background. 86 VITA John Walter Bell was born in Cave Springs, Ar- kansas, October 9, 1936. He received his elementary and secondary schooling at Pea Ridge, Arkansas, graduating in 1954. He received an Associate in Science in Agriculture from Arkansas Polytechnic College in 1956, and a Bachelor of Science in Agriculture in Dairy Science from the Uni- versity of Arkansas in 1958. He entered the U.S. Army in 1959 as an officer and served two years. He re-entered the University of Arkansas and received the M.S. degree in Animal Nutrition in 1963 doing his research under the guidance of Dr. 0. T. Stallcup. He was a graduate re- search assistant in the Department of Dairy, Michigan State University, from 1962 to 1967. He is presently Assistant Professor of Agriculture at Western Illinois University, Macomb, and will receive the Ph.D. degree in 1970. He is a member of National Association of College Teachers of Agriculture, American Dairy Science Associa— tion, American Society of Animal Science, Alpha Zeta, Gamma Sigma Delta, ODK, and an associate member of Sigma Xi. 87 la. 1b. 3a. 3b. REFERENCES Annison, E. F. in Digestive Physiology in the Rumi- nant. Page 185. Edited by R. W. Dougherty, et al. Butterworths, Washington. 1965. Ibid., page 189. Annison, E. F., K. J. Hill, and D. Lewis. Studies on the Portal Blood of Sheep. 2. Absorption of Volatile Fatty Acids from the Rumen of the Sheep. Biochem. J. 66:592, 1957. Annison, E. F., and D. Lewis. Metabolism in the Rumen. Page 59. Methuen, London. 1959. Ibid., page 137. Armstrong, D. G., and K. L. Blaxter. The Heat Incre— ment of Steam Volatile Fatty Acids in Fasting Sheep. Brit. J. Nutr. 11:247, 1957. Armstrong, D. G., K. L. Blaxter, and N. McGraham. The Heat Increment of Mixtures of Steam Volatile Fatty Acids in Fasting Sheep. Brit. J. Nutr. 11:392, 1957. Ash, R. W. Inhibition of Reticulo-Rumen Contractions by Acid. J. Physiol. 133:75p, 1956. Ash, R. W., and A. Dobson. The Effect of Absorption on the Acidity of Rumen Contents. J. Physiol. 169:39, 1963. Barcroft, J., R. A. McAnally, and A. T. Phillipson. Absorption of Volatile Fatty Acids from the Ali~ mentary Tract of the Sheep and Other Animals. J. Exp. Biol. 20:120, 1944. 88 10. 11. 12. 13. 14. 15. 16. 17. 89 Barry, R. J. C., and D. H. Smyth. Transfer of Short- Chain Fatty Acids by the Intestine. J. Physiol. 152:48, 1960. Bensadoun, A., O. L. Paladines, and J. T. Reid. Effect of Level of Intake and Physical Form of the Diet on Plasma, Glucose Concentration, and Volatile Fatty Acid Absorption in Ruminants. J. Dairy Sci. 45:1203, 1962. Bensadoun, A., and J. T. Reid. Estimation of Portal Blood Flow in Ruminants: Effect of Feeding, Fast— ing, and Anesthesia. J. Dairy Sci. 45:540, 1962. Bloom, B., I. L. Chalkoff, and W. O. Reinhardt. In- testinal Lymph as a Pathway for Transport of Ab- sorbed Fatty Acids of Different Chain Lengths. Am. J. Physiol. 166:451, 1951. Brown, R. E., C. L. Davis, J. R. Staubus, and W. 0. Nelson. Production and Absorption of Volatile Fatty Acids in the Perfused Rumen. J. Dairy Sci. 43:1788, 1960. Coe, E. L. Prediction of the Relative Passive— Transfer Rates of Short-Chain Fatty Acids Across Lipid Membranes. Nature 211:80, 1966. Coe, E. L, and M. H. Coe. A Hypothesis Relating Oil:Water Partition Coefficients and Vapor Pres— sures of Nonelectrolytes to their Penetration Rates through Biological Membranes. J. Theoret— ical Biology 8:327, 1965. Conver, H. G., D. McGilliard, and C. F. Huffman. Bovine Re-entrant Duodenal Fistula Studies. I. Establishment of Re-entrant Duodenal Fistula. J} Animal Sci. 16:692, 1957. 14 . . . Cook, R. M. Use of C to Study Utilization of Sub- strates in Ruminants. J. Dairy Sci. 49:1018, 1966. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 90 Cunningham, H. M. Inflatable Plug for Reticulo— Omasal Orifice and Equipment for Saliva Collection from Fistulated Steers. Can..1 Animal Science 38:84, 1958. Danielli, J. F., M. W. S. Hitchcock, R. A. Marshall, and A. T. Phillipson. The Mechanism of Absorption from the Rumen as Exemplified by the Behaviour of Acetic, Propionic, and Butyric Acids. J. Expl. Biol. 22:75, 1945. Dobson, A., and A. T. Phillipson. The Influence of the Contents of the Rumen and of Adrenaline Upon Its Blood Supply. J. Physiol. 133:76p, 1956. Dobson, M. J., W. B. Brown, A. Dobson, and A. T. Phillipson. A Histological Study of the Organi— zation of the Rumen Epithelium of Sheep. Quart. J. Expl. Physiol. 41:247, 1956. Gray, F. V. The Absorption of Volatile Fatty Acids from the Rumen. J. Expl. Biol. 24:1, 1947. Gray, F. V. The Absorption of Volatile Fatty Acids from the Rumen. II. The Influence of pH on Ab— sorption. J. EXpl. Biol. 25:135, 1948. Gray, F. V., and A. F. Pilgrim. Fermentation in the Rumen of the Sheep. II. The Production and Ab- sorption of Volatile Fatty Acids During the Fer- mentation of Wheaten Hay and Lucerne Hay in the Rumen. J. Expl. Biol. 28:83, 1951. Hird, F. J. R., and M. J. Weidemann. TranSport and Metabolism of Butyrate by Isolated Rumen Epithel— ium. Biochem. J. 92:585, 1964. Hydén, S. Observation on the Absorption of Inorganic Ions from the Reticulo-Rumen of the Sheep. Kungl. LanterBgsk. Annlr. 27:273, 1961. Johnson, R. B. The Effects of the Administration of Acetic, Propionic, and N-Butyric Acids Upon the Blood Glucose and Ketone-Body Levels of Goats. Cornell Vet. 45:273, 1955. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 91 Johnson, R. R. Techniques and Procedures for lg Vitro and In_Vivo Rumen Studies. J. Animal Sci. 25:855. 1966. Kameaka, K. and H. Morimoto. Extent of Digestion in the Rumen-Reticulum-omasum of Goats. J. Dairy Sci. 42:1187, 1959. Kiddle, P., R. A. Marshall, and A. T. Phillipson. A Comparison of the Mixture of Acetic, Propionic, and Butyric Acids in the Rumen and in the Blood Leaving the Rumen. J. Physiol. 113:207, 1951. Lavker, R., J. F. Dickey, and W. Chalupa. Fine Structural Observations on the Stratum Germina- tivum from Bovine Fore~Stomach. Fine Structural Observations on the Stratum Corneum and Stratum Granulosum from Bovine Fore-Stomach. J. Dairy Sci. 50:997, 1967 (Abs.). Lewis, A. E. Biostatistics. Reinhold Publishing Corporation, New York. 1966. Page 59. Masson, M. J., and A. T. Phillipson. The Absorption of Acetate, Propionate, and Butyrate from the Rumen of Sheep. J. Physiol. 113:189, 1951. Maynard, L. A., and J. K. Loosli. Animal Nutrition. Page 290. 4th Edition. McGraw—Hill Book Company, Inc. New York. 1956. Murray, M. G., R. S. Reid, and T. M. Sutherland. The Rate of Passage of Digesta from the Reticulo— Rumen of Sheep. J. Physiol. 164:26p (Abs.), 1962. Parthasarathy, D., and A. T. Phillipson. The Move- ment of Potassium, Sodium Chloride, and Water Across the Rumen Epithelium of Sheep. J. Physiol. 121:452, 1953. Pennington, R. J. The Metabolism of Short Chain Fatty Acids in the Sheep. I. The Fatty Acid Utilization and Ketone Body Production by Rumen Epithelium and Other Tissues. Biochem. J. 51:251, 1952. 38. 39. 40. 41. 42. 43a. 43b. 44. 45. 46. 47. 48. 92 Pfander, W. H., and A. T. Phillipson. The Rates of Absorption of Acetic, Propionic, and NoButyric Acids. J. Physiol. 122:102, 1953. Senior, J. R. Intestinal Absorption of Fats. J. Lipid Res. 5:495, 1964. Sisson, S., and J. D. Grossman. The Anatomy of Do- mestic Animals. Page 726. 4th Edition. W. B. Saunders Company, Philadelphia. 1953. Smith, A. H., M. Kleiber, A. L. Black, and C. F. Blaxter. Transfer of Phosphate in the Digestive Tract. J. Nutr. 57:507, 1955. Smyth, D. H., and C. B. Taylor. Intestinal Transfer of Short-Chain Fatty Acids Ip_Vitro. J. Physiol. 141:73, 1958. Snedecor, G. W. Statistical Methods. 5th Edition. The Iowa State University Press, Ames. 1956. Page 87. Ibid., page 455. Stacy, B. D., and A. H. Brook. The Renal Response of Sheep to Feeding. Aust. J. agric. Res. 15:289, 1964. Stevens, C. E. Transport of Sodium and Chloride by the Isolated Rumen Epithelium. Am. J. Physiol. 205:1099, 1964. Stevens, C. E., and B. K. Stettler. Factors Affect- ing the Transport of Volatile Fatty Acids Across Rumen Epithelium. Am. J. Physiol. 210:365, 1966. Stevens, C. E., and B. K. Stettler. Transport of Fatty Acid Mixtures Across Rumen Epithelium. Am. J. Physiol. 211:264, 1966. Stevens, C. B., and B. K. Stettler. Evidence for Active Transport of Acetate Across Bovine Rumen Epithelium. Am. J. Physiol. 213:1335, 1967. 49. 50. 51. 52. 53. 54. 55. 56. 57. 93 Sutherland, T. M. in Progress in Nutrition and Allied Sciences. Page 159. Edited by D. P. Cutherbert- son. Oliver and Boyd, Edinburgh. 1963. Sutton, J. D., A. D. McGilliard, and N. L. Jacobson. Technique for Determining the Ability of the Reticulo—Rumen of Young Calves to Absorb Volatile Fatty Acids at Controlled pH. J. Dairy Sci. 45: 1357, 1962. Tsuda, T. Studies on the Absorption from the Rumen. I. Surgical Operation of Forming the Miniature Rumen. Tohoku J. Agric. Res. 7:231, 1957. Tsuda, T. Studies on the Absorption from the Rumen. II. Absorption of Several Organic Substances from the Miniature Rumen of the Goat. Tohoku J3 Agric. Res. 7:241, 1957. Tsuda, T. The Relationship Between Rumen Motility and the Rate of Acetic Acid Absorption from the Rumen by the Use of the Rumen Pouch Method. Tohoku J} Agric. Res. 13:29, 1962. Tsuda, T. Absorption of Water from the Rumen by the Use of Rumen Pouch Method. Tohoku J) Agric. Res. 15:83. 1964. Turner, A. W., and E. V. Hodgets. Buffer Systems in the Rumen of the Sheep. Aust. J. Agric. Res. 6:125, 1955. Umbreit, W. W., R. H. Burris, and J. F. Stauffer. Manometric Techniques. 4th Edition, 1964. Bur- gess Publishing Company, Minneapolis 15, Minne- sota, page 132. Ussing, H. H., and K. Zerhan. Active Transport of Sodium as the Source of Electric Current in the Short-Circuited, Isolated Frog Skin. Acta Phys- iol. Scand. 23:110, 1951. 58. 59. 60. 61. 62. 94 Vidacs, G., G. M. Ward, and G. Nagy. lp_Vitro Studies of Volatile Fatty Acids Transfer Through the Rumen Epithelium Using a New Instrument. J. Dairy Sci. 45:691, 1962 (abs.). Waldern, D. E., V. L. Johnson, and T. H. Blosser. Cardiac Output in the Bovine and Relationship to Rumen and Portal Volatile Fatty Acid Concentra- tion. J. Dairy Sci. 46:327, 1963. Warner, A. C. I., and B. D. Stacy. Solutes in the Rumen of the Sheep. Quart. J. Expl. Physiol. 50:169, 1965. Werbin, H., I. L. Chaikoff, and M. R. Imada. Rapid Sensitive Method of Determining H3-Water in Body Fluids by Liquid Scintillation Spectrometry. Proc. Soc. Expl. Biol. Med. 102:8, 1959. Wilson, T. H. Intestinal Absorption. W. B. Saunders Company, Philadelphia. 1962. 2 8 A m E“ “I.“ “u V: ”I6 5 u R H” E I“ ”0 m3 “0 lll.l3 mg “2 m1 m3