SIMULTANEOUS ISOLATION AND CHARACTERIZATION FROM BOVINE SMALL INTESTINE' J. W. Wilson and K. E. Webb, Jc2 Virginia Polytechnic Institute and State Universit? Blacksburg 24061 ABSTRACT
Purified brush border and basolateral membraneswere isolated from homogenized intestinal enterocytes of Holstein steers by divalent cation precipitation followed by differential and sucrose density gradicnt centrifugation. Alkaline phosphatase and NaiK adenosine triphosphatase served as marker enzymes for the brush border and basolateral membranes, respectively. The brush border and basolateral membrane fractions were enriched 5.1- and 10.1-fold, respectively, over the cellular homogenate. Electron micrographs, obtained with transmission electron microscopy, confirmed the vesicular nature of the membranes and revealed that basolateral membrane vesicles generally were smaller and more irregular in shape than brush border membrane vesicles. The vesicular nature of isolated membrane preparations was confirmed with osmotic activity experiments. Enrichment of brush border and basolateral membrane fractions compared to the initial homogenate and the vesicular configurationof both preparationsindicate that the isolated brush border and basolateralmembrane preparationswere suitable models for evaluating nutrient transport properties of bovine small intestine. The number of transport expcrimentspossible per animal using the membrane vesicle technique is many times more efficient than some other in vitro techniques (i.e, intestinal rings or everted sacs). (Key Words: Brush Border, Basolateral, Bovidae, Vesicles, Small Tntestine, Enterocytes.) J. Anim Sci. 1990.68583-590 Introduction
The basis for the majority of information on the mechanisms of nutrient transport has arisen through the use of in vitro techniques. These in vitro techniques have included intestinal rings, everted sacs, ligated intestinal segments and isolated mucosal cells (Fisher and Parsons, 1949; Agar et al., 1953; Wilson and Wiseman, 1959; Kimmich, 1975;Phillips et al., 1976).These techniques have limitations in that cellular metabolism is occurring and that the experimenter
'Appreciation is expressed to A l i a Lukes for assisin manuscript preparation and to D. R. Notter, T. Caceci and Bruce Welch for technical assistance. 2Send reprint request to this author. 3Dept. of A i m . Sci. Received October 24, 1988. Accepted June 5,1989. tance
cannot regulate the intracellular substrate and electrolyte concenmtion gradients. The recent use of isolated membrane vesicles has overcome some of these problems and has given researchers the ability to characterize nutrient transport systems at both the luminal and serosal side of the mucosal lining. Isolated brush border (BB) and basolateral (BL) membrane vesicles have been used to evaluate the transport properties of the small intestine and kidney of several nonruminant species (Douglas et al., 1972; Hopfer et al., 1973; Im et al., 1980; Ganapathy et al., 1981; Ling et al., 1981). Only recently have the transport properties of intestinal BB tissue of ruminant species been evaluated with the membrane vesicle technique (Kaunitz and Wright, 1984; Moe et al., 1985; Crooker and Clark, 1986). Isolation of intestinal BL membranes from ruminant species has not been reported to date. Characterization of
583
Downloaded from https://academic.oup.com/jas/article-abstract/68/2/583/4704042 by Iowa State University user on 24 January 2019
OF BRUSH BORDER AND BASOLATERAL MEMBRANE VESICLES
584
WILSON AND WEBB
Materials and Methods
Animals. Holstein steers weighing an average of 400 kg were used as the tissue donors for the experiments. Steers were fed to gain .9 kg/d. The diet (as-fed basis) consisted of 50%ground corn, 30% orchardgrass hay, 13.3%soybean meal, 5 % molasses, .42% defluorinated rock phosphate, .78% limestone and .5% trace mineral salt4 and was available until 12 h before slaughter. A total of 14 steers were used to develop various aspects of h e vesicle isolation procedures. Tissue Preparation. Steers were mechanically stunned and then were killed by exsanguination. Viscera were removed from steers within about 12 min of stunning. The small intestine was cut free from the mesentery beginning 2 m distal of the pyloric valve and proceeding distally to 2 m proximal of the ileocecal junction. The intestine was cut into l-m segments and flushed free of digesta with buffer containing 300 mM mannitol and 12 mM tris base; pH was adjusted to 7.4 with HCl (mannitol buffer). Tissue was maintained at 4°C throughout the procedure unless otherwise specified. The intestinal segments then were everted and incubated in buffer containing 1 mg hyaluronidase/ml, 1mg bovine serum albumin/ml, 120mM NaC1,20 mM tris base, 1 mM MgC12 and 3 mM K 2 m 4 (PH was adjusted to 7.4 with HCl) for 20 min at 37T. Following incubation, the intestinal segments were placed in a porcelain pan seated on a bed of crushed ice. The mucosal lining then was harvested by scraping the intestinal segments with a glass slide. The use of hyaluronidase incubation
?race mineral salt contained a minimum of 97.0% NaCI. 232% Fe, 225% Mn, .2%Zn, . l % Mg, .04% S, .M3% Cu, .007%Co and .007% I. 'Biofreezr, Forma Scientific, Marietta, OH. 6Kit 246. Sigma Chemical Co., St. Louis, MO. 'Pierce Coomassie Blue Protein Reagent, Pierce Chemical, Rockford, IL. 'PT 10/35polytron, Brinkman Instruments, Westburg, NY.
with the isolated intestinal segments was implemented to separate mucus from the luminal lining and to facilitate the harvesting of enterocytes (A. J. Moe, personal communication). Residual hyaluronidase was removed from enterocytes by twice resuspending cells in equal volumes of mannitol buffer and centrifuging at 4,500 x g for 12 min. All centrifugations were performed at 4'C in refrigerated centrifuges. Isolated enterocytes were divided into 3- to 4-g aliquots, placed in whirl-pac bags and flash-frozen in liquid N2. Enterocytes were stored at -8O'C in a ultralow temperature freeze2 for future use. Marker Enzymes. Alkaline phosphatase (EC 3.1.3.1) was the marker for BB membranes. Activity of alkaline phosphatase was determined with an enzyme assay kit6. Activity was expressed on a per milligram of protein basis. Sodium-potassium adenosine triphosphatase (EC3.6.1.3) (NalK ATPase) was the marker for BL membranes. The assay used to measure Na/K ATPase activity was the procedure of Fugita et al. (197 1). Liberated orthophosphate concentration was determined spectrophotometrically by the method of Eibel and Lands (1969). All samples were analyzed in triplicate. The activity of Na/K ATPase was obtained by subtracting ouabainsensitive ATPase activity from total ATPase activity, which was measured in the absence of ouabain. Activity of NaK ATPase was expressed on a microgram of orthophosphateper milligram of protein basis. Protein was assayed using a Coomassie Blue reagent (Bradford, 1976)7. Tissue Fractionation. The enterocytes were suspended at a concentration of 3 g of tissue per 24 ml of buffer containing 5 mM MgC12,150 mM mannitol, 10 mM tris base, 30 mM succinate, 5 mM potassium phosphate and .l mM MnC12; pH was adjusted to 7.4 with NaOH (mannitoi-succinate buffer) and homogenized with a polytron*, equipped with a 20-mmdiameter probe, for 15 s at 13,200 rpm. For each preparation, 36 g of mucosal tissue were used. The 36 g of tissue was a compositeof 12 g of tissue from three different steers. The BB and BL membranes then were isolated by differential and density gradient centrifugations (Figure 1). The homogenate was incubated for 30 min with gentle stirring to allow Mg++aggregation of internal (i.e., endoplasmic reticulum, lysosomes and mitochondria) and BL membranes (Schmitz et al., 1975; Kessler et al., 1978). After incubation, the homogenate was cenmfuged at 8,700 x g for 12 min. Through the use of marker enzymes, the resulting pellet (Pa) was found to contain a majority of the BL mem-
Downloaded from https://academic.oup.com/jas/article-abstract/68/2/583/4704042 by Iowa State University user on 24 January 2019
the transport properties of both BB and BL membranes will aid our understanding of the mechanisms involved in transport of organic solutes across intestinal epithelia, from the lumen to the circulation. These experimentswere conductedto develop a simple and rapid procedure for the isolation of BL membrane vesicles in conjunction with BB membrane vesicles utilizing the same starting tissue.
BRUSH BORDER AND BASOLATERAL MEMBRANE VESICLES
branes, whereas BB membranes remained primarily in the supernatant fluid (Sa). Brush border membranes were enriched further by applying the double precipitation technique of Orsenigoet al. (1985). This technique consisted of first harvesting BB membranes from Sa by centrifuging at 3 1,000x g for 15 min. The resulting pellet (Pb) then was resuspended in mannitol-succinate buffer with 12 strokes of a kflon-glass homogenizer' (.0889 to .1143 mm clearance) and was incubated for 30 min with mild agitation. This second Mg++precipitationfacilitated the removal of BL membrane contamination.The suspension was centrifuged at 8,700 x g for 12 rnin to precipitate aggregate membranes into a pellet (Pc). The supernatant fluid (S,) then was centrifuged at 31,000 x g for 15 min to harvest the BB membranes in the pellet (Pd). The BL membrane fraction (P,) was resuspended in a buffer containing 272 mM mannitol, 20 mM 4-(2-hydroxyethyl)-I-piperazineethanesulfonic acid (HEPES) and 2 mM MgClz, pH 7.4
9Kontes Scientific Glassware, Vineland. NJ. "nternational Equipment Company, Needham Ilei hts, MA. hNunc cryotubes, Vangard International Inc.. N e p tune, NJ.
(mannitol-transportbuffer) with 12 strokes of a teflon-glass homogenizer. The suspension then was centrifugedat 750 x g for 15 min, precipitating the internal membranes into a pellet (PI). During the initial phases of the development of this procedure in our laboratory, it was determined that the internal membranes (Le., endoplasmic reticulum, lysosomesand mitochondria) were of greater density than the BB and BL membranes and were mostly associated with PI. Moe et al. (1985) and Crooker and Clark (1986) reported similar observations. The supernatant fluid (SI) then was centrifuged at 8,700 x g for 12 min, yielding the BL membrane fraction in the pellet (P2). Brush border and BL membrane fractions (Pd and P2, respectively) were resuspended in mannitol-transport buffer to a protein concentration of 2 to 3 mg protein/ml and applied to a sucrose gradient of 27 and 31% sucrose (wuwt). Sucrose solutions were prepared with buffer containing 4 mM MgC12 and 4 mM HEPES (pH 7.4, adjusted with NH40H). Approximately 5 ml of membrane suspension were applied to density gradients made up of 3.5 ml27% sucrose solution layered on 3.5 ml31% sucrose solution and centrifuged at 105,000 x g for 90 min in a IEC model M-60 ultracentrifuge". The resulting bands then were collected by aspirating the membrane band and underlying sucrose layer. Bands were placed either in cryovials" and frozen in liquid N2 or diluted to 40 ml with mannitol-transport buffer and centrifuged at 105,000 x g for 60 min. The resulting washed pellets were resuspended in mannitol transport buffer and used for either marker enzyme analysis or transport experiments. The membranesoriginallyfrozen in liquid NZwere stored at -80°C in an ultralow ternpcrature freezer for future transport experiments. Moe et al. (1985) and Crooker and Clark (1986) found no difference between the transport prop erties of fresh and frozen membrane preparations. Transport Assay. Transport experiments were conducted with the micro-filtrationtechniques of Murer et al. (1974) and Kimmich (1975). Prior to the start of the transport assay, two to three cryovials of sucrose-membranesuspension were removed from -8o'C storage,thawed in warm tap water and then suspended in 10volumes of mannitol-transport buffer and centrifuged for 60 rnin at 105,000 x g to wash sucrose from the membranes. The pellet was resuspended in a volume of mannitol-transport buffer that yielded 2 to 3 mg protein/ml.
Downloaded from https://academic.oup.com/jas/article-abstract/68/2/583/4704042 by Iowa State University user on 24 January 2019
Figure 1. Scheme for the isolation of bovine brush border and basolateral membrane vesicles. S, P and B correspond to supernatant fluids, pellets and bands, r e sptively. The letter and number subscripts differentiated membrane fractions of the brush border and basolateral membrane isolation schemes, respectively.
585
586
WILSON A N D WEBB
Microscopy. Both fresh and frozen mucosal scrapings were evaluated with a differential interferencecontrast microscope". After obtaining a sampleof fresh mucosal scrapings,a subsample was placed in a whirl pac bag and flash-frozen in liquid N2. Frozen tissue was thawed at room temperature. Wet mounts of tissue from fresh and frozen origin were fitted with coverslips and placed under an oil immersion lens for evaluation. Total magnification was 1,250. Electron Microscopy. Brush border and BIZ membrane vesicle preparations that had been stored at -8O'C were separated from sucrose solution and resuspended in mannitol transport buffer as previously described. The suspension was centrifuged at 31,000 x g for 15 min. The resulting pellet was fixed in a mixture of 5.0% glutaraldehyde, 3.0% formaldehyde and 2.5% picric acid, buffered with .1M sodium cacodylate at pH 7.4. It was washed in buffer and post-fixed for 1 h in 1.0% osmium teroxide in the same vehicle. The post-fixed pellet was washed again then dehydrated in increasingly concentratedcthanol solutions and embedded in epoxy resin. U1trathin sections were cut'8 and stained with lead citrate (Reynolds, 1963)and uranyl acetate (Watson, 1958) and examined in a JEOl 1oOCX-I1 transmission electron micro~cope'~. Results and Discussion
The technique utilized in the simultaneous isolation of BB and BL membranes from the same initial homogenate relied on the differences in the surface charge densities that exist bctwcen BB and BL membranes (Schmiti! et al., 1975; Kessler et al., 1978).The BB membranes have a surface charge density that allows for the incorporation of both of the positive charges of divalent cations, such as Mg". This property minimizes the aggregation of BB membrane fragments that would result from Mg"' crosslinking. In contrast, the BL membrane surface charge density can accommodateonly one of the positive charges of a divalent cation, thus facilitating cross-linking between two fragments and l b r n e t t e A, Precision Systems Inc., Sudbury. MA. subsequent precipitation of BL membrane frag13New England Nuclear, Boston, MA. ments. This difference in surface membrane 141-IAWP.45 pW pore si7~,Millipore Corporation, properties allowed for separation of BL from BB Bedford, MA. membrane fractions by differential ccnh'ifuga"LS-271, National Diagnostics, Manville, NJ. 16Model SR7500, Beckman Instruments, Palo Alto, tion (Figure 1).The resulting mcnibrancfractions CA. were purified further with the use of discontinu17LeitzOrtholux 2, Bunton Instrument Company Inc., ous sucrose gradients. Rockville, MD. The use of a sucrose gradient consisting of 27 18UltrotomeNova, LKB, Swcden. and 31% sucrose bands proved most suitable for 19100CX Stem, JEOL Ltd.. Tokyo, Japan.
Downloaded from https://academic.oup.com/jas/article-abstract/68/2/583/4704042 by Iowa State University user on 24 January 2019
The composition of the intravesicular space was a function of the buffer used for the resuspension of the membranes. For the purposes of the present experiments, a Na-free buffer was required so that a Na gradient could be formed when Lhe vesicles were added to the reaction mixture containing Na. Mannitol-transport buffer was a suitable buffer for this purpose. Reaction vessels were prepared by adding 175 pl of transport buffer and 105 p1 of methionine solution to the vessel and incubating at 37'C. Transport was initiated by the addition of 70 pl of membrane solution to the reaction vessel. The transport buffer was prepared so that the final incubation medium contained 100 mM NaSCN, 2 mM MgC12, 10 mM HEPES and mannitol at concentrations of 25,50,100,200 and 300 mM. The pH was adjusted to 7.4 with HCl. The osmolalities of the incubation media were determined with an The methionine solution was prepared so that the final methionine concentration would be .1 mM, including 2.5 pCi of L-[3sS]-methionine'3(specific activity 1,151 Ci/mmol) per reaction vessel as the radiotracer. Reactions were terminated after 60 min by removing 100-pl aliquots of reaction mixture, placing it on a .45-pm nitrocellulose filter14 and applying vacuum. Filters then were washed with three 5-ml aliquots of ice-cold 150 mM KC1 solution. Three 100-pl aliquots were filtered for each reaction vessel with four reaction vessels per treatment. The filters were air-dried and placed in 20-ml scintillationvials and mixed with 9 ml of Ecoscint scintillation fluid". Radioactivity was determined by liquid scintillation counting16. Nonspecific radioactivity retention on the membrane filters was determinedby applying 30 p1 of substrate solution (equivalent to the radioactive methionine present in 100 p1 of reaction mixture) to the filter and then washing the filter with three 5-ml aliquots of 150 mM KC1. Filter nonspecific binding was subtracted from total activity of all data points prior to calculations of picomoles of substrate retention by vesicles.
587
BRUSI I BORDER AND BASOLATERAI, MEMBRANE VESICLES
TABLE 1. ENRICHMENTOF NM< ATPASE AND ALKALINE PHOSPIIATASE IN BRUSlI BORDER AND BASOLATERAL MEMBRAME FRACTIONS RELATIVE TO INITIAL HOMOGENATE Enzyme enrichment. Fractionb
Na/K ATPase
Ba
2.2 (3) 1.6 (.3) 1.4 ('7) 10.1 (3.5) 5.5 (2.9) 2.6 (1.7)
Bb RC
Bi B2 B3
Alkaline phosphatase
2.6 (.6) 5.1 (.6) 4.0 (.3) .5 (.1) 2.1 (6) 2.3 (2)
'Reported enrichments are means of five separate isolation experiments with SD in parentheses. Enrichment was determined by dividing the specific activity of enzyme in the membrane fraction by the specific activity of that enzyme in the homogenate. bBa, Bb and B, correspond to the membrane fractions isolated at 27 and 31 % sucrose and the pellet, respectively, from the sucrose density gradient centrifugation of the crude brush border membrane fractions. B1, B2 and B3 correspond to the bands recovered when the same sucrose density scheme was applied to the crude basolateral m e n brane fraction.
BL membrane vesicles appearcd to be consistently smaller and more irregular in shape and size than the BB membrane vesicles. This is similar to results obtained by Fugita et al. (1972) using isolated BB and BL membrane vesicles from rat intestines. The higher magnification of the membrane preparationsrevealed that the vesicles were formed by a well-preserved trilaminar plasma membrane. There appears to be a rougher outer texture (possibly microvilli) in the BB membranepreparation. Neither BB nor BLmembrane preparations had any recognizable organelle contamination &e., mitochondria). The vesicular nature of the membrane preparations also was tested experimentally by monitoring the osmotic responsiveness of the membrane preparation. The assumption made with this technique was that equilibrium uptake by vesicles would be proportional to the intravesicular space and that the intravesicular space will be determinedby the osmolality of the incubation buffer (Munck, 1966). Therefore, by monitoring the equilibrium accumulation of a substrate at various osmolalities by a membrane preparation, one can determine the configuration of that membrane preparation. If uptake decreases with increasing osmolality of incubation
Downloaded from https://academic.oup.com/jas/article-abstract/68/2/583/4704042 by Iowa State University user on 24 January 2019
both BB and BL membrane isolation. Several other sucrose gradientswere evaluated,including the higher-density sucrose gradients recommended by Moe et al. (1985) and Crooker and Clark (1986). Membrane vesicles were prepared with deliberate homogenization of the enterocytes with a plyuon followed by aresuspension of each subsequent pellet with a teflon-glass homogenizer. This process resulted in the most repeatable preparation of vesicles. Different homogenization procedures are likely to result in vesiclesof different sizes and, therefore,different buoyant densities, which would result in optimal separation on different sucrose gradients. The application of the membrane fractions from differential centrifugation to the 27% and 3 1% sucrose gradient and centrifugation at 105,000x g for 90 min resulted in the formation of two bands, one between the interface of the buffer and the 27% sucrose layer and another at the interface of the 27 and 31% sucrose layers. The third fraction formed was a pellet at the bottom of the 3 1% sucrose layer. When the BB membrane fraction (Pd) was applied to the sucrose density gradient, the three membrane fractions from top to bottom were referred to as B,, Bb and B,; the fractions resulting from cenuifugation of the BL membrane fraction Cpz) were referred to as B1, Bz and B3. The middle band (Bb) of the BB sucrose density gradient was found to have the highest BB enrichment (5.1-fold for alkaline phosphatase; Table 1). This enrichment is similar to that achieved by Moe et al. (1985) but is slightly lower than that of Crooker and Clark (1986). The fraction at the interface between the buffer and the 27% sucrose layer (BI) of the BL membrane density gradient was the most suitable for use as a model for the BL membrane in transport studies. This fraction had a 10.1-fold Na/K ATPase enrichment and only a .5-foId alkaline phosphatase cross-contamination (Table 1). This high ratio of Na/K ATPase:alkaline phosphatase, 20.2: 1, indicates that transport measurements by vesicles of B1 membrane fraction represent substrate uplake via BL membranes. In addition to being enriched with the marker enzyme, the membranes also must be of a vesicular nature, as opposed to globular or membrane sheets. The vesicular nature of the membrane preparations was confirmcd by both visual and experimental appraisals. Electron micrographs obtained with transmission electron microscopy showed that both BL and BB membrane preparations were of vesicularorienlation (Figure2). The
588
WILSON AND WEBB
buffer, then the membrane preparation is considered to be at least partially vesicular. If uptake is not affected by osmolality, the preparation is considcred to be nonvesicular. Methionine was the substrate utilized experimentally to test the vesicularity of BB and BL membrane preparations. Plotting equilibrium methionine accumulation by both BB and BL membrane preparation vs inverse osmolality revealed a linear relationship with a positive slope (Figure 3). The regression equations for BB and BL membrane preparations were Y = 140X + 276, R2 = .992 and Y = 171X + 222, R2 = .994, respectively. The linearity and positive slope of the equilibrium methionine accumulation vs the
inverse of osmolality indicate that methionine uptake was occurring into an osmotically active space, vesicles, with both BB and BL membrane preparations. Substrate accumulation was plolted against inverse osmolality to allow for the extrapolation of uptake data to infinite osmolality, Y-intercept, in order to predict the level of substrate that was present due to surface binding but not due to uptake into the vesicle. It is assumcd that at infinite osmolality there would be no intravesicular space (Faust et al., 1968; Eicholz et al., 1969). The BB and BL membrane surface binding, after 60 min of incubation, was determined lo be 276 and 222 pmoles of methionine/mg of protein,
Downloaded from https://academic.oup.com/jas/article-abstract/68/2/583/4704042 by Iowa State University user on 24 January 2019
Figure 2. Transmission electron micrographs of brush border and basolateral membrane preparations. A and B represent brush border membranes magnified 100,oooX and 19,oooX (before enlargement), respectively. C and D represent basolateral membranes magnified 3 6 , 0 0 0 ~and 29,oooX (before enlargement), respectively.
BRUSH BORDER AhD BASOLATERAL MEMBRANE VESICLES
z
1.000 7 bosoloteral
n r
brush border
0
800
600
400
1
i
Figure 3. Effect of osmolalities of incubation buffers on equilibrium (60 min) net fluxes of labeled L - m t h h by brush border (m) and basolateral (0) nine (100 membrane preparations. Osmolalities were adjusted by changing the concentrations of I)-mannitol. Data points are means of 12 observations.
respectively. Among the advantages of the membrane vesicle technique is the efficient use of animals. The scraping of intestinal segmentsresulted in mucosal tissue yields of an average of 1,556g per steer. This correspondsto a minimum of about 70 transport experiments per steer for each cell surface, BB and BL. Considerably fewer transport experiments can be performed per animal by conventional techniques such as the everted sac procedure of Phillips et al. (1976). The membrane vesicle technique thus is a far more efficient use of experimental animals. The large number of experiments pcr steer with the membrane vesicle technique is possible because mucosal scrapings can be frozen prior to BB and BL membrane separation. Freezing of mucosal scrapings was found to have no detrimental effects. In fact, enrichment of BL membranes was enhanced by freezing, as indicated by the increased Na/K ATPase:alkalinephosphatase ratio of 4.3 and 2.9 for frozen and fresh tissue preparations, respectively. These results are in agreement with the findingsof Crooker and Clark (1986).The advantage of frozen tissue may be due to freeze-fracturing of isolated enterocytes. Examination of previously frozen tissue with a differential interference contrast microscopeshowed uniform cell fragmentswith no intact cells. Therefore,homogenization of frozen tissue may result in membrane fragments of different sizes and composition than the same homogenization of
fresh tissue. Freeze-fracturing makes the use of homogenization techniques, such as N2-cavitation, impractical when frozen tissue is being used. The BB and BL membranes were found to be osmotically active and, therefore, of a vesicular nature. Thus, the BB and BL membrane vesicles are suitable tools for characterizing the transport propertiesof the bovine intestinal enterocyte.The BB membrane vesicles can be used to evaluate the transport properties regulating entry of nutrients into the enterocyte, and BL membrane vcsicles can be used to monitor transport properties leaving the enterocyte and entering the circulation. In both cases, the absence of cellular metabolism allows for the characterization of transport properties of metabolizable substrates. By using these membrane vesicles, investigators will be able to gain information leading to a better understanding of fundamental aspects of nutrient absorption and transport by ruminants. Furthermore, alterations in mucosal tissue function as a result of dietary or other factors might be explored, In the present paper, methioninewas the substrate used to evaluatethe quality of mcmbrane preparation. Membrane vesicles prepared by the above procedures could be utilized to study the absorption and transport of numcrous nutrients. By collecting enterocytes from discrete locations along the small intestine, all of the above could be evaluated with reference to site of intestinalactivity.The diversity of applications for these membrane vesicles makes this attractive as an alternative method for evaluating nutrient absorption and transport. Literature Cited Agar, W. T.,F.J.R. Ilird and G. S. Sidhu. 1954. The uptake of amino acids by the intestine. Biochim. Biophys. Acta 14:80. Bradford, M. M. 1976. A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Bioehem. 72248. Crooker, B. A. and J. II. Clark. 1986. Preparation of brush border membrane vesicles from fresh and frozen bovine intestine for nutrient uptake studies. J. Dairy Sci. 6958. Douglas, A. P., R. Kerley and K. I. lsselbacher. 1972. Preparation and characterization of the lateral and basal plasma membranes of the rat intestinal epithelial cell. Biochern J. 128:1329. Eibel. 11. and W. E. Lands. 1%9. A new, sensitive determination ofphosphate. Anal. Biochem. 3051. Eicholz, A., K.E. Howell and K.G. Crane. 1969. Studies on the organization of the brush border in intestinal epithelial cells. VI. Glucose binding to isolated intestinal brush borders and their subfractions. B i e
Downloaded from https://academic.oup.com/jas/article-abstract/68/2/583/4704042 by Iowa State University user on 24 January 2019
-.
589
590
WILSON AND WEBB
isolated intestinal epithelial cells and their use in studying intestinal transport. In: E. D. Korn (Ed.) Methods in Membrane Biology. Vol. 5. pp 175-219. Plenum Press, New York. Ling, K.Y., W. B.ImandR. G. Faust. 1981. Na+-independent sugar uptake by rat intestinal and rcnal brush border and basolateral membrane vesicles. I n t J. Biochem. 13:693. Moe, A. J., P.A. Pwius and C. E. Polan. 1985. Isolation and characterization of brush border mcnibranc vesicles from bovine small intestine. J. Nutr. 115:1173. Munck, B. G . 1966. Amino acid transport by small intestine of the rat. The existence and specificity of the transport mechanisms of imino acids and its relation to the transport of glycine. Uiochim. Ijiophys. Acta 120:97. Murer, M.,U. Hopfer, E. Kinne-Sauffran and K. Kinne. 1974. Glucose transport in isolated brush border and lateral-basal plasma membrane vesicles from intestinal epithelial cells. Biochim. Biophys. Acta 345:170. Orsenigo, M. N., M. Tosco, G. Esposito and A. Faelli. 1985. The basolateral membrane of rat entcrocyte: Its purification from brush border contamination. Anal. Biochem. 144577. Phillips, W. A., K. E. Webb and J. P. Fontenot. 1976. In vitroabsorptionofaminoacids by thesmallintestine of sheep. J. Anim Sci. 42:201. Reynolds, E. S . 1963. The use of lead citrate at high pII as an electron opaque stain in electron microscopy. J. Cell Uiol. 17:208. Schmitz, J., 11. Preiser, P. Maestracci, B. K. Ghosh, J. Cerda and P. K. Crane. 1975. Purification of the human intestinal brush border membrane. Uiochirn Biophys. Acta 323:98. Watson, M. L. 1958. Staining of tissue sections for electron microscopy with heavy metals. J. Biophys. L3iW chem. Cytol. 4:475. Wilson, T. €i.and G. Wiseman. 1959. The use of sacs of everted s m a l l intestine for the study of the transference of substances from the mucosal to t h e serosal surface. J. Physiol. (Lond.) 123:116.
Downloaded from https://academic.oup.com/jas/article-abstract/68/2/583/4704042 by Iowa State University user on 24 January 2019
chim Biophys. Acta 193:179. Faust, R. G., M. G. Leadbetter, R. K. Plenge and A. J. McCaslin. 1968. Active sugar transport by the small intestine. The effects of sugars, amino acids, hex* samines, sulfhydryl-reacting compounds and cations on preferential binding of D-glucose to tris disrupted brush borders. J. Gen. Physiol. 52:482. Fisher, R. B. and D. S. Parsons. 1949. A preparation of surviving rat small intestine for the study of absorp tion. J. Physiol. (Lond.) I10:36. Fugita, M., II. Matsui, L. Nagano and M. Nakao. 1971. Asymmetric dishibution of ouabain-sensitive ATPase activity in rat intestinal mucosa. Biochim. B i e phys. Acta 233:404. Fugita, M., €1. Ohta, K. Kawai, 11. Matsui and M. Nakao. 1972. Differential isolation of microvillous and basolateral plasma membranes from intestinal mucosa: Mutually exclusive distribution of digestive enzyme and ouabain-sensitive ATPase. Biochim Hiophys. Acta 274:336. Ganapathy, V., J. E Mendicino and F. II. Leibach. 1981. Effectofpapain treatmentondipeptide transportinto rabbit intestinal brush border vesicles. Life Sci. 29:2451. IIopfer, U., K. Nelson, J. Pemtto and 1. Issellbacher. 1973. Glucose transport in isolated brush border membrane from rat small intestine. J. Biol. Chem 248:25. Im, W. B., P. W. Misch, D. W. Powell and R. G. Faust 1980. Phenolphthalein and harmalineinduceddisturbances in the transport functions of isolated brush border and basolateral membrane vesicles from rat jejunum and kidney cortex. Biochem. Pharmacol. 29:2307. Kaunitz, J. D. andE. M. Wright. 1984. Kinetics of sodium D-glucose cotransport in bovineintestinal brush border vesicles. J. Membr. Biol. 79:41. Kessler. M., D. Acuto, C. Storell,li. Murer, M. Miller and G. Semenza. 1978. A modified procedure for the rapid preparation of efficiently transporting vesicles from small intestinal brush border membranes. B i e chim. Biophys. Acta 506:136. Kimmich, G. A. 1975. Preparation and characterizationof
Purified brush border and basolateral membraneswere isolated from homogenized intestinal enterocytes of Holstein steers by divalent cation precipitation followed by differential and sucrose density gradicnt centrifugation. Alkaline phosphatase and NaiK adenosine triphosphatase served as marker enzymes for the brush border and basolateral membranes, respectively. The brush border and basolateral membrane fractions were enriched 5.1- and 10.1-fold, respectively, over the cellular homogenate. Electron micrographs, obtained with transmission electron microscopy, confirmed the vesicular nature of the membranes and revealed that basolateral membrane vesicles generally were smaller and more irregular in shape than brush border membrane vesicles. The vesicular nature of isolated membrane preparations was confirmed with osmotic activity experiments. Enrichment of brush border and basolateral membrane fractions compared to the initial homogenate and the vesicular configurationof both preparationsindicate that the isolated brush border and basolateralmembrane preparationswere suitable models for evaluating nutrient transport properties of bovine small intestine. The number of transport expcrimentspossible per animal using the membrane vesicle technique is many times more efficient than some other in vitro techniques (i.e, intestinal rings or everted sacs). (Key Words: Brush Border, Basolateral, Bovidae, Vesicles, Small Tntestine, Enterocytes.) J. Anim Sci. 1990.68583-590 Introduction
The basis for the majority of information on the mechanisms of nutrient transport has arisen through the use of in vitro techniques. These in vitro techniques have included intestinal rings, everted sacs, ligated intestinal segments and isolated mucosal cells (Fisher and Parsons, 1949; Agar et al., 1953; Wilson and Wiseman, 1959; Kimmich, 1975;Phillips et al., 1976).These techniques have limitations in that cellular metabolism is occurring and that the experimenter
'Appreciation is expressed to A l i a Lukes for assisin manuscript preparation and to D. R. Notter, T. Caceci and Bruce Welch for technical assistance. 2Send reprint request to this author. 3Dept. of A i m . Sci. Received October 24, 1988. Accepted June 5,1989. tance
cannot regulate the intracellular substrate and electrolyte concenmtion gradients. The recent use of isolated membrane vesicles has overcome some of these problems and has given researchers the ability to characterize nutrient transport systems at both the luminal and serosal side of the mucosal lining. Isolated brush border (BB) and basolateral (BL) membrane vesicles have been used to evaluate the transport properties of the small intestine and kidney of several nonruminant species (Douglas et al., 1972; Hopfer et al., 1973; Im et al., 1980; Ganapathy et al., 1981; Ling et al., 1981). Only recently have the transport properties of intestinal BB tissue of ruminant species been evaluated with the membrane vesicle technique (Kaunitz and Wright, 1984; Moe et al., 1985; Crooker and Clark, 1986). Isolation of intestinal BL membranes from ruminant species has not been reported to date. Characterization of
583
Downloaded from https://academic.oup.com/jas/article-abstract/68/2/583/4704042 by Iowa State University user on 24 January 2019
OF BRUSH BORDER AND BASOLATERAL MEMBRANE VESICLES
584
WILSON AND WEBB
Materials and Methods
Animals. Holstein steers weighing an average of 400 kg were used as the tissue donors for the experiments. Steers were fed to gain .9 kg/d. The diet (as-fed basis) consisted of 50%ground corn, 30% orchardgrass hay, 13.3%soybean meal, 5 % molasses, .42% defluorinated rock phosphate, .78% limestone and .5% trace mineral salt4 and was available until 12 h before slaughter. A total of 14 steers were used to develop various aspects of h e vesicle isolation procedures. Tissue Preparation. Steers were mechanically stunned and then were killed by exsanguination. Viscera were removed from steers within about 12 min of stunning. The small intestine was cut free from the mesentery beginning 2 m distal of the pyloric valve and proceeding distally to 2 m proximal of the ileocecal junction. The intestine was cut into l-m segments and flushed free of digesta with buffer containing 300 mM mannitol and 12 mM tris base; pH was adjusted to 7.4 with HCl (mannitol buffer). Tissue was maintained at 4°C throughout the procedure unless otherwise specified. The intestinal segments then were everted and incubated in buffer containing 1 mg hyaluronidase/ml, 1mg bovine serum albumin/ml, 120mM NaC1,20 mM tris base, 1 mM MgC12 and 3 mM K 2 m 4 (PH was adjusted to 7.4 with HCl) for 20 min at 37T. Following incubation, the intestinal segments were placed in a porcelain pan seated on a bed of crushed ice. The mucosal lining then was harvested by scraping the intestinal segments with a glass slide. The use of hyaluronidase incubation
?race mineral salt contained a minimum of 97.0% NaCI. 232% Fe, 225% Mn, .2%Zn, . l % Mg, .04% S, .M3% Cu, .007%Co and .007% I. 'Biofreezr, Forma Scientific, Marietta, OH. 6Kit 246. Sigma Chemical Co., St. Louis, MO. 'Pierce Coomassie Blue Protein Reagent, Pierce Chemical, Rockford, IL. 'PT 10/35polytron, Brinkman Instruments, Westburg, NY.
with the isolated intestinal segments was implemented to separate mucus from the luminal lining and to facilitate the harvesting of enterocytes (A. J. Moe, personal communication). Residual hyaluronidase was removed from enterocytes by twice resuspending cells in equal volumes of mannitol buffer and centrifuging at 4,500 x g for 12 min. All centrifugations were performed at 4'C in refrigerated centrifuges. Isolated enterocytes were divided into 3- to 4-g aliquots, placed in whirl-pac bags and flash-frozen in liquid N2. Enterocytes were stored at -8O'C in a ultralow temperature freeze2 for future use. Marker Enzymes. Alkaline phosphatase (EC 3.1.3.1) was the marker for BB membranes. Activity of alkaline phosphatase was determined with an enzyme assay kit6. Activity was expressed on a per milligram of protein basis. Sodium-potassium adenosine triphosphatase (EC3.6.1.3) (NalK ATPase) was the marker for BL membranes. The assay used to measure Na/K ATPase activity was the procedure of Fugita et al. (197 1). Liberated orthophosphate concentration was determined spectrophotometrically by the method of Eibel and Lands (1969). All samples were analyzed in triplicate. The activity of Na/K ATPase was obtained by subtracting ouabainsensitive ATPase activity from total ATPase activity, which was measured in the absence of ouabain. Activity of NaK ATPase was expressed on a microgram of orthophosphateper milligram of protein basis. Protein was assayed using a Coomassie Blue reagent (Bradford, 1976)7. Tissue Fractionation. The enterocytes were suspended at a concentration of 3 g of tissue per 24 ml of buffer containing 5 mM MgC12,150 mM mannitol, 10 mM tris base, 30 mM succinate, 5 mM potassium phosphate and .l mM MnC12; pH was adjusted to 7.4 with NaOH (mannitoi-succinate buffer) and homogenized with a polytron*, equipped with a 20-mmdiameter probe, for 15 s at 13,200 rpm. For each preparation, 36 g of mucosal tissue were used. The 36 g of tissue was a compositeof 12 g of tissue from three different steers. The BB and BL membranes then were isolated by differential and density gradient centrifugations (Figure 1). The homogenate was incubated for 30 min with gentle stirring to allow Mg++aggregation of internal (i.e., endoplasmic reticulum, lysosomes and mitochondria) and BL membranes (Schmitz et al., 1975; Kessler et al., 1978). After incubation, the homogenate was cenmfuged at 8,700 x g for 12 min. Through the use of marker enzymes, the resulting pellet (Pa) was found to contain a majority of the BL mem-
Downloaded from https://academic.oup.com/jas/article-abstract/68/2/583/4704042 by Iowa State University user on 24 January 2019
the transport properties of both BB and BL membranes will aid our understanding of the mechanisms involved in transport of organic solutes across intestinal epithelia, from the lumen to the circulation. These experimentswere conductedto develop a simple and rapid procedure for the isolation of BL membrane vesicles in conjunction with BB membrane vesicles utilizing the same starting tissue.
BRUSH BORDER AND BASOLATERAL MEMBRANE VESICLES
branes, whereas BB membranes remained primarily in the supernatant fluid (Sa). Brush border membranes were enriched further by applying the double precipitation technique of Orsenigoet al. (1985). This technique consisted of first harvesting BB membranes from Sa by centrifuging at 3 1,000x g for 15 min. The resulting pellet (Pb) then was resuspended in mannitol-succinate buffer with 12 strokes of a kflon-glass homogenizer' (.0889 to .1143 mm clearance) and was incubated for 30 min with mild agitation. This second Mg++precipitationfacilitated the removal of BL membrane contamination.The suspension was centrifuged at 8,700 x g for 12 rnin to precipitate aggregate membranes into a pellet (Pc). The supernatant fluid (S,) then was centrifuged at 31,000 x g for 15 min to harvest the BB membranes in the pellet (Pd). The BL membrane fraction (P,) was resuspended in a buffer containing 272 mM mannitol, 20 mM 4-(2-hydroxyethyl)-I-piperazineethanesulfonic acid (HEPES) and 2 mM MgClz, pH 7.4
9Kontes Scientific Glassware, Vineland. NJ. "nternational Equipment Company, Needham Ilei hts, MA. hNunc cryotubes, Vangard International Inc.. N e p tune, NJ.
(mannitol-transportbuffer) with 12 strokes of a teflon-glass homogenizer. The suspension then was centrifugedat 750 x g for 15 min, precipitating the internal membranes into a pellet (PI). During the initial phases of the development of this procedure in our laboratory, it was determined that the internal membranes (Le., endoplasmic reticulum, lysosomesand mitochondria) were of greater density than the BB and BL membranes and were mostly associated with PI. Moe et al. (1985) and Crooker and Clark (1986) reported similar observations. The supernatant fluid (SI) then was centrifuged at 8,700 x g for 12 min, yielding the BL membrane fraction in the pellet (P2). Brush border and BL membrane fractions (Pd and P2, respectively) were resuspended in mannitol-transport buffer to a protein concentration of 2 to 3 mg protein/ml and applied to a sucrose gradient of 27 and 31% sucrose (wuwt). Sucrose solutions were prepared with buffer containing 4 mM MgC12 and 4 mM HEPES (pH 7.4, adjusted with NH40H). Approximately 5 ml of membrane suspension were applied to density gradients made up of 3.5 ml27% sucrose solution layered on 3.5 ml31% sucrose solution and centrifuged at 105,000 x g for 90 min in a IEC model M-60 ultracentrifuge". The resulting bands then were collected by aspirating the membrane band and underlying sucrose layer. Bands were placed either in cryovials" and frozen in liquid N2 or diluted to 40 ml with mannitol-transport buffer and centrifuged at 105,000 x g for 60 min. The resulting washed pellets were resuspended in mannitol transport buffer and used for either marker enzyme analysis or transport experiments. The membranesoriginallyfrozen in liquid NZwere stored at -80°C in an ultralow ternpcrature freezer for future transport experiments. Moe et al. (1985) and Crooker and Clark (1986) found no difference between the transport prop erties of fresh and frozen membrane preparations. Transport Assay. Transport experiments were conducted with the micro-filtrationtechniques of Murer et al. (1974) and Kimmich (1975). Prior to the start of the transport assay, two to three cryovials of sucrose-membranesuspension were removed from -8o'C storage,thawed in warm tap water and then suspended in 10volumes of mannitol-transport buffer and centrifuged for 60 rnin at 105,000 x g to wash sucrose from the membranes. The pellet was resuspended in a volume of mannitol-transport buffer that yielded 2 to 3 mg protein/ml.
Downloaded from https://academic.oup.com/jas/article-abstract/68/2/583/4704042 by Iowa State University user on 24 January 2019
Figure 1. Scheme for the isolation of bovine brush border and basolateral membrane vesicles. S, P and B correspond to supernatant fluids, pellets and bands, r e sptively. The letter and number subscripts differentiated membrane fractions of the brush border and basolateral membrane isolation schemes, respectively.
585
586
WILSON A N D WEBB
Microscopy. Both fresh and frozen mucosal scrapings were evaluated with a differential interferencecontrast microscope". After obtaining a sampleof fresh mucosal scrapings,a subsample was placed in a whirl pac bag and flash-frozen in liquid N2. Frozen tissue was thawed at room temperature. Wet mounts of tissue from fresh and frozen origin were fitted with coverslips and placed under an oil immersion lens for evaluation. Total magnification was 1,250. Electron Microscopy. Brush border and BIZ membrane vesicle preparations that had been stored at -8O'C were separated from sucrose solution and resuspended in mannitol transport buffer as previously described. The suspension was centrifuged at 31,000 x g for 15 min. The resulting pellet was fixed in a mixture of 5.0% glutaraldehyde, 3.0% formaldehyde and 2.5% picric acid, buffered with .1M sodium cacodylate at pH 7.4. It was washed in buffer and post-fixed for 1 h in 1.0% osmium teroxide in the same vehicle. The post-fixed pellet was washed again then dehydrated in increasingly concentratedcthanol solutions and embedded in epoxy resin. U1trathin sections were cut'8 and stained with lead citrate (Reynolds, 1963)and uranyl acetate (Watson, 1958) and examined in a JEOl 1oOCX-I1 transmission electron micro~cope'~. Results and Discussion
The technique utilized in the simultaneous isolation of BB and BL membranes from the same initial homogenate relied on the differences in the surface charge densities that exist bctwcen BB and BL membranes (Schmiti! et al., 1975; Kessler et al., 1978).The BB membranes have a surface charge density that allows for the incorporation of both of the positive charges of divalent cations, such as Mg". This property minimizes the aggregation of BB membrane fragments that would result from Mg"' crosslinking. In contrast, the BL membrane surface charge density can accommodateonly one of the positive charges of a divalent cation, thus facilitating cross-linking between two fragments and l b r n e t t e A, Precision Systems Inc., Sudbury. MA. subsequent precipitation of BL membrane frag13New England Nuclear, Boston, MA. ments. This difference in surface membrane 141-IAWP.45 pW pore si7~,Millipore Corporation, properties allowed for separation of BL from BB Bedford, MA. membrane fractions by differential ccnh'ifuga"LS-271, National Diagnostics, Manville, NJ. 16Model SR7500, Beckman Instruments, Palo Alto, tion (Figure 1).The resulting mcnibrancfractions CA. were purified further with the use of discontinu17LeitzOrtholux 2, Bunton Instrument Company Inc., ous sucrose gradients. Rockville, MD. The use of a sucrose gradient consisting of 27 18UltrotomeNova, LKB, Swcden. and 31% sucrose bands proved most suitable for 19100CX Stem, JEOL Ltd.. Tokyo, Japan.
Downloaded from https://academic.oup.com/jas/article-abstract/68/2/583/4704042 by Iowa State University user on 24 January 2019
The composition of the intravesicular space was a function of the buffer used for the resuspension of the membranes. For the purposes of the present experiments, a Na-free buffer was required so that a Na gradient could be formed when Lhe vesicles were added to the reaction mixture containing Na. Mannitol-transport buffer was a suitable buffer for this purpose. Reaction vessels were prepared by adding 175 pl of transport buffer and 105 p1 of methionine solution to the vessel and incubating at 37'C. Transport was initiated by the addition of 70 pl of membrane solution to the reaction vessel. The transport buffer was prepared so that the final incubation medium contained 100 mM NaSCN, 2 mM MgC12, 10 mM HEPES and mannitol at concentrations of 25,50,100,200 and 300 mM. The pH was adjusted to 7.4 with HCl. The osmolalities of the incubation media were determined with an The methionine solution was prepared so that the final methionine concentration would be .1 mM, including 2.5 pCi of L-[3sS]-methionine'3(specific activity 1,151 Ci/mmol) per reaction vessel as the radiotracer. Reactions were terminated after 60 min by removing 100-pl aliquots of reaction mixture, placing it on a .45-pm nitrocellulose filter14 and applying vacuum. Filters then were washed with three 5-ml aliquots of ice-cold 150 mM KC1 solution. Three 100-pl aliquots were filtered for each reaction vessel with four reaction vessels per treatment. The filters were air-dried and placed in 20-ml scintillationvials and mixed with 9 ml of Ecoscint scintillation fluid". Radioactivity was determined by liquid scintillation counting16. Nonspecific radioactivity retention on the membrane filters was determinedby applying 30 p1 of substrate solution (equivalent to the radioactive methionine present in 100 p1 of reaction mixture) to the filter and then washing the filter with three 5-ml aliquots of 150 mM KC1. Filter nonspecific binding was subtracted from total activity of all data points prior to calculations of picomoles of substrate retention by vesicles.
587
BRUSI I BORDER AND BASOLATERAI, MEMBRANE VESICLES
TABLE 1. ENRICHMENTOF NM< ATPASE AND ALKALINE PHOSPIIATASE IN BRUSlI BORDER AND BASOLATERAL MEMBRAME FRACTIONS RELATIVE TO INITIAL HOMOGENATE Enzyme enrichment. Fractionb
Na/K ATPase
Ba
2.2 (3) 1.6 (.3) 1.4 ('7) 10.1 (3.5) 5.5 (2.9) 2.6 (1.7)
Bb RC
Bi B2 B3
Alkaline phosphatase
2.6 (.6) 5.1 (.6) 4.0 (.3) .5 (.1) 2.1 (6) 2.3 (2)
'Reported enrichments are means of five separate isolation experiments with SD in parentheses. Enrichment was determined by dividing the specific activity of enzyme in the membrane fraction by the specific activity of that enzyme in the homogenate. bBa, Bb and B, correspond to the membrane fractions isolated at 27 and 31 % sucrose and the pellet, respectively, from the sucrose density gradient centrifugation of the crude brush border membrane fractions. B1, B2 and B3 correspond to the bands recovered when the same sucrose density scheme was applied to the crude basolateral m e n brane fraction.
BL membrane vesicles appearcd to be consistently smaller and more irregular in shape and size than the BB membrane vesicles. This is similar to results obtained by Fugita et al. (1972) using isolated BB and BL membrane vesicles from rat intestines. The higher magnification of the membrane preparationsrevealed that the vesicles were formed by a well-preserved trilaminar plasma membrane. There appears to be a rougher outer texture (possibly microvilli) in the BB membranepreparation. Neither BB nor BLmembrane preparations had any recognizable organelle contamination &e., mitochondria). The vesicular nature of the membrane preparations also was tested experimentally by monitoring the osmotic responsiveness of the membrane preparation. The assumption made with this technique was that equilibrium uptake by vesicles would be proportional to the intravesicular space and that the intravesicular space will be determinedby the osmolality of the incubation buffer (Munck, 1966). Therefore, by monitoring the equilibrium accumulation of a substrate at various osmolalities by a membrane preparation, one can determine the configuration of that membrane preparation. If uptake decreases with increasing osmolality of incubation
Downloaded from https://academic.oup.com/jas/article-abstract/68/2/583/4704042 by Iowa State University user on 24 January 2019
both BB and BL membrane isolation. Several other sucrose gradientswere evaluated,including the higher-density sucrose gradients recommended by Moe et al. (1985) and Crooker and Clark (1986). Membrane vesicles were prepared with deliberate homogenization of the enterocytes with a plyuon followed by aresuspension of each subsequent pellet with a teflon-glass homogenizer. This process resulted in the most repeatable preparation of vesicles. Different homogenization procedures are likely to result in vesiclesof different sizes and, therefore,different buoyant densities, which would result in optimal separation on different sucrose gradients. The application of the membrane fractions from differential centrifugation to the 27% and 3 1% sucrose gradient and centrifugation at 105,000x g for 90 min resulted in the formation of two bands, one between the interface of the buffer and the 27% sucrose layer and another at the interface of the 27 and 31% sucrose layers. The third fraction formed was a pellet at the bottom of the 3 1% sucrose layer. When the BB membrane fraction (Pd) was applied to the sucrose density gradient, the three membrane fractions from top to bottom were referred to as B,, Bb and B,; the fractions resulting from cenuifugation of the BL membrane fraction Cpz) were referred to as B1, Bz and B3. The middle band (Bb) of the BB sucrose density gradient was found to have the highest BB enrichment (5.1-fold for alkaline phosphatase; Table 1). This enrichment is similar to that achieved by Moe et al. (1985) but is slightly lower than that of Crooker and Clark (1986). The fraction at the interface between the buffer and the 27% sucrose layer (BI) of the BL membrane density gradient was the most suitable for use as a model for the BL membrane in transport studies. This fraction had a 10.1-fold Na/K ATPase enrichment and only a .5-foId alkaline phosphatase cross-contamination (Table 1). This high ratio of Na/K ATPase:alkaline phosphatase, 20.2: 1, indicates that transport measurements by vesicles of B1 membrane fraction represent substrate uplake via BL membranes. In addition to being enriched with the marker enzyme, the membranes also must be of a vesicular nature, as opposed to globular or membrane sheets. The vesicular nature of the membrane preparations was confirmcd by both visual and experimental appraisals. Electron micrographs obtained with transmission electron microscopy showed that both BL and BB membrane preparations were of vesicularorienlation (Figure2). The
588
WILSON AND WEBB
buffer, then the membrane preparation is considered to be at least partially vesicular. If uptake is not affected by osmolality, the preparation is considcred to be nonvesicular. Methionine was the substrate utilized experimentally to test the vesicularity of BB and BL membrane preparations. Plotting equilibrium methionine accumulation by both BB and BL membrane preparation vs inverse osmolality revealed a linear relationship with a positive slope (Figure 3). The regression equations for BB and BL membrane preparations were Y = 140X + 276, R2 = .992 and Y = 171X + 222, R2 = .994, respectively. The linearity and positive slope of the equilibrium methionine accumulation vs the
inverse of osmolality indicate that methionine uptake was occurring into an osmotically active space, vesicles, with both BB and BL membrane preparations. Substrate accumulation was plolted against inverse osmolality to allow for the extrapolation of uptake data to infinite osmolality, Y-intercept, in order to predict the level of substrate that was present due to surface binding but not due to uptake into the vesicle. It is assumcd that at infinite osmolality there would be no intravesicular space (Faust et al., 1968; Eicholz et al., 1969). The BB and BL membrane surface binding, after 60 min of incubation, was determined lo be 276 and 222 pmoles of methionine/mg of protein,
Downloaded from https://academic.oup.com/jas/article-abstract/68/2/583/4704042 by Iowa State University user on 24 January 2019
Figure 2. Transmission electron micrographs of brush border and basolateral membrane preparations. A and B represent brush border membranes magnified 100,oooX and 19,oooX (before enlargement), respectively. C and D represent basolateral membranes magnified 3 6 , 0 0 0 ~and 29,oooX (before enlargement), respectively.
BRUSH BORDER AhD BASOLATERAL MEMBRANE VESICLES
z
1.000 7 bosoloteral
n r
brush border
0
800
600
400
1
i
Figure 3. Effect of osmolalities of incubation buffers on equilibrium (60 min) net fluxes of labeled L - m t h h by brush border (m) and basolateral (0) nine (100 membrane preparations. Osmolalities were adjusted by changing the concentrations of I)-mannitol. Data points are means of 12 observations.
respectively. Among the advantages of the membrane vesicle technique is the efficient use of animals. The scraping of intestinal segmentsresulted in mucosal tissue yields of an average of 1,556g per steer. This correspondsto a minimum of about 70 transport experiments per steer for each cell surface, BB and BL. Considerably fewer transport experiments can be performed per animal by conventional techniques such as the everted sac procedure of Phillips et al. (1976). The membrane vesicle technique thus is a far more efficient use of experimental animals. The large number of experiments pcr steer with the membrane vesicle technique is possible because mucosal scrapings can be frozen prior to BB and BL membrane separation. Freezing of mucosal scrapings was found to have no detrimental effects. In fact, enrichment of BL membranes was enhanced by freezing, as indicated by the increased Na/K ATPase:alkalinephosphatase ratio of 4.3 and 2.9 for frozen and fresh tissue preparations, respectively. These results are in agreement with the findingsof Crooker and Clark (1986).The advantage of frozen tissue may be due to freeze-fracturing of isolated enterocytes. Examination of previously frozen tissue with a differential interference contrast microscopeshowed uniform cell fragmentswith no intact cells. Therefore,homogenization of frozen tissue may result in membrane fragments of different sizes and composition than the same homogenization of
fresh tissue. Freeze-fracturing makes the use of homogenization techniques, such as N2-cavitation, impractical when frozen tissue is being used. The BB and BL membranes were found to be osmotically active and, therefore, of a vesicular nature. Thus, the BB and BL membrane vesicles are suitable tools for characterizing the transport propertiesof the bovine intestinal enterocyte.The BB membrane vesicles can be used to evaluate the transport properties regulating entry of nutrients into the enterocyte, and BL membrane vcsicles can be used to monitor transport properties leaving the enterocyte and entering the circulation. In both cases, the absence of cellular metabolism allows for the characterization of transport properties of metabolizable substrates. By using these membrane vesicles, investigators will be able to gain information leading to a better understanding of fundamental aspects of nutrient absorption and transport by ruminants. Furthermore, alterations in mucosal tissue function as a result of dietary or other factors might be explored, In the present paper, methioninewas the substrate used to evaluatethe quality of mcmbrane preparation. Membrane vesicles prepared by the above procedures could be utilized to study the absorption and transport of numcrous nutrients. By collecting enterocytes from discrete locations along the small intestine, all of the above could be evaluated with reference to site of intestinalactivity.The diversity of applications for these membrane vesicles makes this attractive as an alternative method for evaluating nutrient absorption and transport. Literature Cited Agar, W. T.,F.J.R. Ilird and G. S. Sidhu. 1954. The uptake of amino acids by the intestine. Biochim. Biophys. Acta 14:80. Bradford, M. M. 1976. A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Bioehem. 72248. Crooker, B. A. and J. II. Clark. 1986. Preparation of brush border membrane vesicles from fresh and frozen bovine intestine for nutrient uptake studies. J. Dairy Sci. 6958. Douglas, A. P., R. Kerley and K. I. lsselbacher. 1972. Preparation and characterization of the lateral and basal plasma membranes of the rat intestinal epithelial cell. Biochern J. 128:1329. Eibel. 11. and W. E. Lands. 1%9. A new, sensitive determination ofphosphate. Anal. Biochem. 3051. Eicholz, A., K.E. Howell and K.G. Crane. 1969. Studies on the organization of the brush border in intestinal epithelial cells. VI. Glucose binding to isolated intestinal brush borders and their subfractions. B i e
Downloaded from https://academic.oup.com/jas/article-abstract/68/2/583/4704042 by Iowa State University user on 24 January 2019
-.
589
590
WILSON AND WEBB
isolated intestinal epithelial cells and their use in studying intestinal transport. In: E. D. Korn (Ed.) Methods in Membrane Biology. Vol. 5. pp 175-219. Plenum Press, New York. Ling, K.Y., W. B.ImandR. G. Faust. 1981. Na+-independent sugar uptake by rat intestinal and rcnal brush border and basolateral membrane vesicles. I n t J. Biochem. 13:693. Moe, A. J., P.A. Pwius and C. E. Polan. 1985. Isolation and characterization of brush border mcnibranc vesicles from bovine small intestine. J. Nutr. 115:1173. Munck, B. G . 1966. Amino acid transport by small intestine of the rat. The existence and specificity of the transport mechanisms of imino acids and its relation to the transport of glycine. Uiochim. Ijiophys. Acta 120:97. Murer, M.,U. Hopfer, E. Kinne-Sauffran and K. Kinne. 1974. Glucose transport in isolated brush border and lateral-basal plasma membrane vesicles from intestinal epithelial cells. Biochim. Biophys. Acta 345:170. Orsenigo, M. N., M. Tosco, G. Esposito and A. Faelli. 1985. The basolateral membrane of rat entcrocyte: Its purification from brush border contamination. Anal. Biochem. 144577. Phillips, W. A., K. E. Webb and J. P. Fontenot. 1976. In vitroabsorptionofaminoacids by thesmallintestine of sheep. J. Anim Sci. 42:201. Reynolds, E. S . 1963. The use of lead citrate at high pII as an electron opaque stain in electron microscopy. J. Cell Uiol. 17:208. Schmitz, J., 11. Preiser, P. Maestracci, B. K. Ghosh, J. Cerda and P. K. Crane. 1975. Purification of the human intestinal brush border membrane. Uiochirn Biophys. Acta 323:98. Watson, M. L. 1958. Staining of tissue sections for electron microscopy with heavy metals. J. Biophys. L3iW chem. Cytol. 4:475. Wilson, T. €i.and G. Wiseman. 1959. The use of sacs of everted s m a l l intestine for the study of the transference of substances from the mucosal to t h e serosal surface. J. Physiol. (Lond.) 123:116.
Downloaded from https://academic.oup.com/jas/article-abstract/68/2/583/4704042 by Iowa State University user on 24 January 2019
chim Biophys. Acta 193:179. Faust, R. G., M. G. Leadbetter, R. K. Plenge and A. J. McCaslin. 1968. Active sugar transport by the small intestine. The effects of sugars, amino acids, hex* samines, sulfhydryl-reacting compounds and cations on preferential binding of D-glucose to tris disrupted brush borders. J. Gen. Physiol. 52:482. Fisher, R. B. and D. S. Parsons. 1949. A preparation of surviving rat small intestine for the study of absorp tion. J. Physiol. (Lond.) I10:36. Fugita, M., II. Matsui, L. Nagano and M. Nakao. 1971. Asymmetric dishibution of ouabain-sensitive ATPase activity in rat intestinal mucosa. Biochim. B i e phys. Acta 233:404. Fugita, M., €1. Ohta, K. Kawai, 11. Matsui and M. Nakao. 1972. Differential isolation of microvillous and basolateral plasma membranes from intestinal mucosa: Mutually exclusive distribution of digestive enzyme and ouabain-sensitive ATPase. Biochim Hiophys. Acta 274:336. Ganapathy, V., J. E Mendicino and F. II. Leibach. 1981. Effectofpapain treatmentondipeptide transportinto rabbit intestinal brush border vesicles. Life Sci. 29:2451. IIopfer, U., K. Nelson, J. Pemtto and 1. Issellbacher. 1973. Glucose transport in isolated brush border membrane from rat small intestine. J. Biol. Chem 248:25. Im, W. B., P. W. Misch, D. W. Powell and R. G. Faust 1980. Phenolphthalein and harmalineinduceddisturbances in the transport functions of isolated brush border and basolateral membrane vesicles from rat jejunum and kidney cortex. Biochem. Pharmacol. 29:2307. Kaunitz, J. D. andE. M. Wright. 1984. Kinetics of sodium D-glucose cotransport in bovineintestinal brush border vesicles. J. Membr. Biol. 79:41. Kessler. M., D. Acuto, C. Storell,li. Murer, M. Miller and G. Semenza. 1978. A modified procedure for the rapid preparation of efficiently transporting vesicles from small intestinal brush border membranes. B i e chim. Biophys. Acta 506:136. Kimmich, G. A. 1975. Preparation and characterizationof
