Brain Microvessels Take up Large Neutral Amino ...

THE JOURNAL OF

BIOLOGICAL CHEMISTRY Vol. 258, No. 14, Issue of July 25, pp. 8949-8954, 1983 Printed in U.S. A

Brain Microvessels for Glutamine COOPERATIVE

ROLE

Take up Large Neutral Amino Acids in Exchange

OF Na’-DEPENDENT

AND

Na’-INDEPENDENT

SYSTEMS* (Received for publication,

Carlo Maria

Cangiano, Antonietta

Patrizia PatriziS,

Cardelli-CangianoS, Kim A. Brackette,

J. Howard James& Filippo Roberto Strom& and Josef

August 11, 1982)

Rossi-Fanelli, E. Fischer8

From the III Department of Internal Medicine, University of Rome, Wale dell’Universitti 37, 00161 Rome, Italy, the $Department of Biochemistry, University of Rome, Piazzak Aldo Moro, 00185 Rome, Italy, and the SDepartment of Surgery, Cincinnati General Hospital and University of Cincinnati Medical Center, 231 Bethesda Avenue, Cincinnati, Ohio 45267

creased L system activity in pathological conditions such as hyperammonemia and portal systemic shunting has recently been suggested (5, 6). It has also been hypothesized that the increased L system transport activity occurs through an increase of intracellular Gln content in the brain microvessels

(6, 7). The present paper demonstrates that increased intracellular Gln concentration affects the uptake by isolated brain microvessels of different categories of amino acids. MATERIALS

AND

METHODS

The following materials were obtained from New England Nuclear: L-[U-“Clleucine, 335 or 50 mCi/mmol; L-[U-%]tyrosine, 503 mCi/ mmol; MeAIB and BCH, 4.95 and 53 mCi/mmol, respectively; L-[U“Cllysine, 300 mCi/mmol; L-[4,5-3H]leucine, 33.7 Ci/mmol; [U-‘“Cl sucrose, 4.63 mCi/mmol; and Aquasol-2. HEPES, unlabeled amino acids, and all other chemicals were obtained from Sigma, from Merck (Darmstadt, West Germany), or from Fluka AG. Chemische Fabrik (Buchs, Switzerland). Isolation

of Brain

Microvessels-Microvessels

were

isolated

from

the gray matter of fresh bovine brain as described by Hjelle et al. (8) with minor modifications (7). Briefly, the gray matter was homogenized by hand in a buffer (l:l, w/v) containing 122 mM NaCI, 25 mM NaHC03, 10 mM glucose, 3 mM KCl, 1.4 mM Ca&, 1.2 mM MgSO+ 0.4 mM K,HP04, and 10 mM HEPES, pH 7.4, and aerated with 95% O2 + 5% CO,. The homogenate was poured on a nylon sieve (86-pm

pore size) and washed with a spray of ice-cold buffer. The material retained

The “blood-brain barrier” which regulates the restricted movements of sugars, amino acids, and other solutes between the blood and the brain has been tentatively localized in the cell membranes of brain capillaries (l-3). A polar distribution of the NAA’ transport systems has been suggested by in vitro and in uiuo experiments. Thus, blood-borne NAAs appear to cross the blood-brain barrier only by using the Na+-independent L system, while another NAA transport system, the Na’dependent A system, seems only to be present on the antiluminal side of the microvessels (4). The involvement of in* This work was supported in part by National Institutes of Health Grant AM 25638 and by the University of Rome “La Sapienza” (Progetto “Fenomeni di membrana”). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. This paper is dedicated to Professor Alessandro Rossi-Fanelli on the occasion of his 75th anniversary. ’ The abbreviations used are: NAA, neutral amino acid; HEPES, 4-(2.hydroxyethyl)-1-piperazineethanesulfonic acid; MeAIB, a-[~“C]methylaminoisobutyrate; BCH, 2-aminobicyclo[2,2,1]heptaoe.2. [“Clcarboxylic acid. 8949

on the sieve

was rehomogenized

and washed

again.

Isolated

vessels were immediately collected in the same buffer in a plastic tube and kept on ice until use. For some of the experiments, after the isolation step, microvessels were resuspended in Na+-free buffer in which NaCl and NaHC03 had been replaced, respectively, by choline chloride and KHC03. Enzyme Assays-The microvessels were subjected to homogenization in a Potter-Elvehjem motor-driven apparatus in the buffer appropriate to the subsequent assay. Alkaline phosphatase was measured usingp-nitrophenylphosphate as substrate in a reaction mixture containing 50 mM MgClz, 5 mM Ca&, 100 mM KCI, 5 mM pnitrophenylphosphate, 100 mM Tris-HCl, pH 9. The reaction was initiated by addition of 0.1 ml of microvessel homogenate in a final volume of 1 ml. The mixture was then incubated at 37 “C for 20 min

and the reaction stopped by addition of 2 ml of 1

N

NaOH and by

cooling in an ice bath. The insoluble material was then removed by centrifugation for 10 min at 3000 X g, and the absorbance at 420 nm determined for each sample and converted to micromoles. ml-’ using ap-nitrophenol standard (9). y-Glutamyl transpeptidase activity was determined using I-y-glutamyl-p-nitroanilide as substrate, according to Orlowski and Meister (10). 5’-Nucleotidase activity was determined in a reaction mixture containing 40 mM Verona1 buffer, pH 7.5,20 mM MnSO,, and 1 mM 5’-AMP in a final volume of 2 ml. The mixture was incubated 30 min at 37 “C in a shaking bath and the reaction stopped by addition of 2 ml of 20% trichloroacetic acid. Insoluble material was removed by centrifugation for 5 min at 3000 X g and aliquots of the supernatant used for Pi determination (11).

Downloaded from www.jbc.org by guest, on April 12, 2012

Some regulatory aspects of neutral amino acid transport were investigated in isolated brain microvessels, an in vitro model of the blood-brain barrier. Preloading of the microvessels with glutamine stimulated the subsequent uptake of other neutral amino acids by way of the Na+-independent L system, but had no effect on the uptake of either basic or acidic amino acids. Moreover, this stimulation was abolished when the loading step was carried out in the absence of Na+ ions or in the presence of a high concentration of a-methylaminoisobutyric acid, indicating that the microvessels were able to concentrate glutamine via the A system of amino acid transport. Since the presence of the A system of neutral amino acid transport has not been detected in studies of bloodbrain transport performed in viuo, the A system is probably associated with the antiluminal side of brain microvessels. Our results indicate, therefore, that the concentrative Na+-dependent A system and the exchanging Na+-independent L system can cooperate in the uptake of the large neutral hydrophobic amino acids. Such a cooperation may be relevant in the pathogenesis of some neurological disturbances such as hepatic encephalopathy, in which brain glutamine concentration is unusually high.

8950

Glutamine Uptake and Exchange Brain in

Microvessels I

The assaywas carried out in the presence or absence of Ni” ions (0.1 mM NiCI,) which selectively inhibit 5’-nucleotidase activity without affectingnonspecific phosphatases.Glutaminesynthetaseactivity was measured according to Wellner and Meister(12). Purity of the Microvessel Preparations-The isolated microvessel preparations appeared to consist essentially of branching capillary segments with somesmallarteriolesand venules. Neitherphasecontrast light microscopy nor scanning electron microscopy (after fixation in 2.5% glutaraldehyde and shadowing with gold) showed any contamination by nervous or glial cells (Fig. 1). The integrity and viability of the cells in the microvessel preparations were tested by the trypan blue exclusion method (8)which, a t variance with the findings reported by Williams et al. (13). indicated almost 100%cell viability. A consistently negligible uptake of [’4C]sucrosewas observed. If 1 pCi/ml was added tomicrovessel suspensions containing about1 mg/ ml of protein, less than 50 pCi/mg of protein hecame protein-bound within the first minute. The subsequentincrease over the entire 30min time course was less than 100 pCi/mg of protein. The microvessel preparations were found to be enriched, with respect tothe gray matter, in y-glutamyltranspeptidaseandin alkaline phosphatase (Table I). Detectable activities of 5’-nucleotidaseand of Glnsynthetase were alsopresent:theseenzymatic activities were not likely due to contamination by other cells, since these enzyme activities were still detectablefollowing extensive treatment with collagenase‘ as described by Williams et al. (13). Cln Preloading and Uptake Experiments-In order to increase their internal concentrationof Gln, isolated microvesselswere resuspended in Na+-containing buffer containing 20 mM Gln and incubated for 15 min a t 37 “C. They were then filtered on an 86-pm pore size nylon sieve, washed with 20 ml of cold buffer as a spray for 2 min, and protein/ml in resuspended a t a final concentration of1-1.5mgof warm Gln-free buffer containing the appropriatelabeled amino acid. As a control, other microvessels were incubated under the same conditions in Gln-free buffer. The uptake was either followed cumulatively for given time periods after addition of 0.65 pCi/ml of the

‘

P. Cardelli-Cangiano, M.A. Patrizi, F. Barberini, C. Cangiano, and R. Strom, unpublished data.

TABLE I Comparison of enzymatic activities in isolated brain microvessels and in gray matter Brain microvessels were obtained as outlined under “Materials and Methods.” Enzymatic activities are expressed as means + S.D. For the numberof preparations shown in parentheses, enzyme levels were determined in triplicate on each preparation by the procedures indicated under “Materials andMethods.” ~

Enzymatic activity

matter

Isolated microvessels

nmol. mg protein” .min”

5”Nucleotidase Alkaline phosphatase y-Glutamyl transpeptidase Glutamine synthetase

50+ 3 (n = 6) 18f I ( n = 5) 6 0.5 ( n = 6) 22 0.1 ( n = 3)

+ +

12+ 3 ( n = 6) 208 22 ( n = 15) 190 & 36 ( n = 16) 2.6 f 0.9 ( n = 6)

+

labeled amino acid or measured as a function of amino acid concentration after a fixed 2-min time interval. Preliminary experiments showed that the rate of uptake had a constant initial value for at least 3 min (Fig. 2). Portionsof 600 pl of the microvessel suspensions were withdrawn after thedesired incubation interval, pouredthrough a 44-pm pore size nylon sieve on a vacuum manifold, and washed three times with 5 ml of ice-cold buffer. The sieves with the retained microvessels were then placed in plastic tubes containing 1.8 ml of 1 N NaOH, left overnight at room temperature, and then subjected to sonication for 1 min. Portions were withdrawn for protein determination(14) using serumalbuminas a standard,and 0.5 ml was transferred to liquid scintillation counting vials containing 0.5 ml of 1 M HCI. After addition of 10 ml of Aquasol-2, the vials were counted in a Packard Tricarb liquid scintillation spectrometer. Nonspecific radioactivity due to the binding of the labeled amino acid to thenylon sieve was also tested in triplicate by omitting the microvessels; the radioactivity remaining on thesieve was constantly below 70 dpm. Kinetic Analysis-The initial (2 min) rateof uptake, when plotted


FIG. 1. Morphology of microvessels isolated from beef brain.Light microscopy (A ) (original X 624) and scanning electron microscopy ( R and C ) (originals, respectively, X 1,120 and X 3,220) indicate that small blood vessels appear as short tubular segments, practically free from nerve cell contamination; a few collagen fibers were occasionally found. For the scanning electron microscopy, the sample was suspended in a modified Karnovky’s fixative containing 2.5% glutaraldehyde a t 4 “C overnight. The fixative was removed from the microvessels by filtrations through a Nuclepore membrane of 5-pm pore size. The membrane was then dehydrated in alcohol and critical point dried with CO, using amyl acetate as the transitional solvent. The dried sample was coated with gold and examined in a JEOL JSM-35U scanningelectron microscope at 25 kV.

Glutamine Uptake and Exchangein Brain Microvessels

8951

75r

0

60

120

180

FIG.2. Time course of Leu uptake within the first 3 min. The data (means f S.D.) were obtained from three different experiments performed in triplicate. The shown linear regression represents the Leu uptake obtained after subtraction from the values observed a t 37 "C of the rates measured a t 0 "C (ie. nonspecific binding). At 0 "C the rates of Leu uptake measured at the same fixed intervals reported in the figure were (from 10 to 180s) 8.1 k 0.6, 9.4 k 0.5, 10 f 0.4, 11.3 rt_ 0.3, 15 f 0.6, and 16.5 f 0.8 pmol.mg of protein", respectively.

I

l2

t

Ki

15 TIME (rnln)

30

FIG.4. Relationship between intracellular Gln levels and Tyr uptake by brain microvessels. Top, 2-min Tyr uptake by brain microvessels measured a t fixed time intervals after preloading with Gin. After 20 min of preloadingwith(open bars) or without (solid bars) 20 mM cold Gln, themicrovessels were washed and rapidly resuspended in a warm (37 "C) Gln-free buffer. ["CITyrosine (1pCi/ ml) was added at thedifferent time intervals indicated on the abscissa, starting from theresuspension of microvessels in the Gln-freebuffer. Bottom, Gln levels in the microvessels (0)and in the medium (0) measured a t different times after the resuspension of the Gln-preloaded microvessels in the Gln-free buffer.

1-

[MeAIB]

1

0

I

I

0

200

[BCH]

pM

1 4 00

PM

FIG. 3. A Dixon plot ( l / u uersus ( I ) )showing the MeAIBinhibiting ( t o p ) and the BCH-inhibiting (bottom) effect on Gln uptake. The initial 2-min uptake of Gln a t 0.03 mM (0)and 0.1 mM (0) wastestedinfreshisolatedbrain microvessels. All the experiments were carried out in the presence of Na' assuming that MeAIB and BCH compete, respectively, for the A and L systemmediated Gln uptake. The results represent the averages of three determinations. The observed K, values were 175 and 275 p~ for MeAIB and BCH, respectively.

as a function of amino acid concentration, showed the presence of a saturable component superimposed ona nonsaturable one. As previously described (7), the latter was calculated from the slope of the linear portionof the uptakecurve between amino acid concentrations of 0.2 and 2.0 mM and then subtracted from the total uptake curve in order to estimate the steady state kinetic parametersof saturable transport. To this purpose the data are plotted in the S/u uersus S "Hanes plot," which allows straightforward statistical analysis, according to Wilkinson(15).Thedata were subjected tononlinear regression analysis toobtaintheoptimalestimate of thekinetic parameters and to evaluate the standard error impending on them. Amino Acid Ana2ysis"In order to measure the intracellular Gln concentration, the isolated microvessels were transferred after the preincubationstepto warm (37 " C ) Gln-free buffer andat fixed intervals thereafter filtered through a 44-pm pore nylon sieve on a vacuum manifold. The retained microvessels were then lysed in 1 ml of cold distilled water, usinga glass homogenizer with a motor-driven Teflonpestle. After proteindeterminationon a100-plaliquotof homogenate, 100 p1 of 40% (w/v) sulfosalicylic acid containing norleucine as internal standardwere added to the remaining homogenate. After centrifugation the supernatant was subjected to amino acid analysis in a Beckman 121 MB automatic amino acid analyzer using lithium citrate buffers,which allow the separate determination of Glu, Gln, and Asn. RESULTS

Gln Uptake by Isolated Brain Capillaries-Kinetic analysis of Gln transportby isolated brain microvessels was performed either in the presence of Na+ ionsor in Na+-freebuffer. Under both experimental conditions, V,,, had similar values of 900 20 and 877 & 15 nmol .mg of protein" .min", respectively, whereas the Na+-independent component was characterized


TIME mecI

in Microvessels Brain

Glutamine Exchange Uptake and

8952

R

20

- 150 P.

I \

I

-1

0

15

5

30

TIME ( m i n )

FIG. 6. Overshoot effect in the time course of [14C]Leu uptake by Gln-preloaded microvessels (0)and by microvessels After 20 min of preloading preincubated in Gln-free buffer (0). with 20 mM Gln, the microvessels were washed and resuspended in warm (37 "C) Gln-free buffer. Immediately after the resuspension, the labeled Leu was added and the time course uptake followed for 30 min. The datain the figure represent the averages + S.D. obtained from three different experiments, each performed in triplicate.

1

0

5

15

33 TIbIE(8ri~n)

FIG. 5. Hanes plot ( S / uuersus S ) of the saturable component microof Tyr uptake by both preloaded (0)and control (0) vessels. Isolated microvessels were assayed for Tyr uptake (2min at 37 "C) between the concentrations ranging from 25 to 1500 PM. After the subtraction of the nonsaturable component, which was comparable under the two experimental conditions, the obtained saturation curves were used to estimate theV,,, and K , values. Data shown are the averages of three different experiments f S.D.

FIG. 7. Effect of the addition of external Leu on Gln efflux from brain microvessels. Microvessels were preloaded with [3H] Gln (final concentration, 20 mM; specific activity, 3.37 Ci/mmol) for 20 min at 37 "C, rapidly washed, and then resuspended in warm Glnfree buffer (X) or in Gln-free buffer containing 0.01 mM (0) or 0.5 mM (0)cold Leu. At fixed intervals the isolated microvessels were rapidly washed and the internal radioactivity was measured. The points are the means of triplicate determinations +- S.D.

TABLEI1 Effect of incubation of the isolated microvessels in 20 mM Gln on the subsequent uptake of MeAIB and Lys Isolated microvessels were incubated with or without 20 mM Gln in Na+-containing buffer, washed with cold Gln-free buffer, and resuspended in warm (37 "C) Gln-free buffer containing Na+. Radiolabeled amino acid was added, the microvessels suspension mixed, and portion of the suspension was quickly withdrawn and washed as described under "Materials and Methods" (IO-s sample). Other samples were withdrawn at the indicated times. Uptake data areexpressed as the average f S.D. for three determinations. amino

Preincubation in

Radioactive acid

10 s

5 min

15 min

pmol. rng protein"

Buffer Buffer Buffer Buffer

+ Gln + Gln

MeAIB MeAIB LYS LYs

8.6 ? 0.1 7.4 -t 0.8 2.5 2 1.1 2.7 f 0.1

34.4 ? 3.8 28.446.3 ? 4.1 18.5 8.5 f 0.3 8.6 +- 0.3

42.5

+ 4.1 2

6.3

f 0.4 18.0 f 1.8

30 min

54.5 ? 5.3 56.5 5.4 27.5 f 1.3 27.3 f 2.5

+


by a lower affinity as compared to that observed in Na+. 3 containing buffer ( K , = 426 k 18 uersus 146 k 10 p ~ ) Fig. shows that both MeAIB and BCH, ordinarily specific substrates of A and L systems, respectively, are competitive inhibitors of Gln uptake, showing K, values of 175 and 275 p ~ respectively. , Gln appears therefore to be transported by the A system as well as by the L system. Effect of Increased IntracellularGlnon NAA UptakeIncubation of the microvessels with Gln,followed by removal of the Gln-containing medium, resulted in an over 2-fold increase in the rate of subsequent Tyr uptake. Subsequently therate of Tyruptake declined tocontrol values within approximately 15 min afterremoval of the microvessels from the Gln-containing medium, with a time course similar to that of the netescape of Gln from the same microvessels (Fig. 4). Kinetic analysis of the initial rate of uptake showed that Gln preloadingaffected only the maximalinflux (V,,,) of the saturable component of uptake (Fig. 5) since neither the K, for Tyr nor the nonsaturable component was changed. Preloading with Gln exerted similar effects on the uptake of other NAAs such as Leu and Trp,while the uptake of MeAIB and of Lys remained unaffected (Table 11). The relation between the Gln efflux and the increased uptake of NAA was confirmed by the demonstrationof an overshoot in the uptakeof the NAA (Fig. 6) as to beexpected inthepresence of a transstimulation phenomenon (10). This relation was confirmed by reciprocal stimulation of the rate of [3H]Gln net escapeobserved whenincreasingconcentrations of other NAAs were added to the suspension medium (Fig. 7).

in Microvessels Brain


8953

Identification of theTransport System Involved in Gln- MeAIB together with Gln during the preloading step commediated Stimulation of NAA Uptake-It was found that Gln- pletely abolished the stimulationby Gln of NAA uptake. mediated stimulation of NAA uptake was abolished if the In other studies, microvessels were incubated in the prespreloading step was performed in Na+-free medium (Fig. 8, ence of either MeAIB, Ala, Cys, or ser (20 mM each) and the top).Thisfindingunderlinestheimportance of the Na+- subsequentuptake of [14C]Leu wasmeasured (Table 111). dependent system in concentrating Gln inside microvessel Neither MeAIB nor Ala, Cys, and Ser were found to exert endothelial cells. On the other hand, the presence of BCH, any stimulatoryeffect on subsequentNAA uptake. Incubation together with Gln during the preloading step in a Na+-con- with Met (20 mM), however, strongly stimulated the subsetaining buffer, caused little or no interference with the sub- quent uptake of [14C]Leu, while incubation with BCH had sequent Gln-dependent stimulation of NAA uptake (Fig. 8, negligible stimulatory effect. bottom). Conversely, the presence of a high concentration of DISCUSSION

Our resultsshow that in brainmicrovessels a relation exists between the L system-mediated uptake of NAA and the A system-mediated increase of intracellular Gln concentration. Kinetic analysis of Gln uptake revealed that Gln can enter brain microvessel endothelial cells using two different transport systems:a Na+-dependent (most likely the A system) and a Na+-independent one (mostlikely the L system). Gln can thus be considered as a dual affinity NAA, while other NAAs such as Trp, Tyr, and Leu are known to utilize predominantly the L system (16-19). High intracellular concentration of Gln within brain microvessels stimulated uptake ' 5 of the otherNAAs. This effect appeared tobe proportional to theinternalGlnconcentration (Fig. 4)and wasdue to a stimulation of the rate of transport (Vm,,) by the L system rather than to a change of the K,,, of uptake. Furthermore, the transport activityof the A system asindicated by MeAIB uptake remained unaffected. The presence of a transstimulation effect of Gln on the uptake of NAA and the increased efflux of labeled intracellular Gln in the presence of high extracellular Leu suggest that a high intracellular Gln concentration acts on the uptake ofNAA essentially by an L system-mediated mechanism. The abilityof Gln to be taken up by the concentrative A system and also to be transported by the exchangingactivity of the L system is probably essential for the induction of this effect. The two Gln carrier systems appeared therefore to act in close cooperation.Significantlyhigher levels of Gln have I 0 15 30 indeed been detected in brainmicrovessels isolated from rats TIME ( m i n ) with a portocaval shunt (7), or in vitro, after incubation of FIG. 8. Two-min Leu uptake measured at fixed intervals the microvessel suspensions with NH: ions (20). High Gln after preloading with Gln of microvessels in four different concentrations have been found in the cerebrospinal fluid of experimental conditions. Top, with 20 mM Gln in a Na+-containing patients with hepatic encephalopathy (i.e. neurological disbuffer (open bars)and with 20 mM Gln in a Na+-free buffer (slashed bars). Bottom, with 20 mM Gln in the presence of BCH (open bars) turbances associated with severe liver disease) (21) as well as and with 20 mM Gln in the presence of 20 mM of MeAIB (slashed in animals subjected to surgical portocaval anastomosis (22bars). The time scale on the abscissae indicates the minutes elapsed 24). The asymmetric distribution of the NAA transport sysafter resuspension in Gln-free buffer and before the 2-min exposure tems found in the blood-brain barrier (i.e. A system present to the 14C-labeledLeu. Solid bars indicate, in both parts of the figure, only on the antiluminalside of brain microvessels), together the uptake by microvessels preincubated in Gln-free buffer (control). with the experimental findings mentionedabove, allow us to When not specified, Na+-containing buffer was used throughout the hypothesize that the increased rate of NAA transport across experiment. Data shown are the means + S.D. the blood-brain barrier is due atocooperation between the A and the L systems, the latter being actually responsible for TABLE111 NAA uptake. [14C]Leu 2-min uptake measured at fixed intervals following 20-min Theseresults, by assigningaprecise role to Gln-NAA preloading of the microvessels suspensions with different solutions containing 20 mM each of the amino acids listed exchange inregulatingtherate of NAA uptake by brain microvessels, support the hypothesis of an involvement of Time (min) Amino acid brain Gln synthesis in the pathogenesis of hepatic encepha15-170-2 8-10 3-5 28-30 lopathy (6, 20, 21).

fe l

pmol. mg protein".min"

MeAIB BCH Ala Ser CYS Met None

21 t- 2 18+3 1 91f 84 + 4 16 t- 3 1 8 f 13 2 k 13 3 k 1.2 35 f 5 27 3.4 2 4 f 6 2 6 + 3 25 f 2 2 22f 34 + 7 18+5 23 f 2 24 f 2 1 7f 4 . 5 25 t- 5 144 f 10" 106 10" 66 f 8" 30 k 10" 23 f 4 20 2 Significantly different from controls ( p < 0.001).

+

+

+

16+ 2 10 2.5 24 5 21 f 3.4 22 f 5 24 f 5 21 3

+ + +

Acknowledgments-The skilled technical assistance of Vincenzo Peresempio and Laura Edwards is gratefully acknowledged. REFERENCES 1. Oldendorf, W. H. (1971) Am. J. Physiol. 221, 1629-1639 2. Oldendorf, W. H., and Szabo, J. (1976) Am. J. Physiol. 2 3 0 , 9 4 98 3. Daniel, P. M., Pratt, 0. E., and Wilson, P. A. (1977) Proc. R. SOC.


-!25a

8954


Lond. B Biol. Sci. 196, 333-346 4. Betz, A. L., and Goldstein, G . W. (1978) Science (Wash. D. C.) 202,225-227 5. James, J. H., Escourrou, J., and Fischer, J. E. (1978) Science, (Wash. D. C . ) 2 0 0 , 1395-1397 6. James, J. H., Ziparo, V., Jeppsson, B., and Fischer, J. E. (1979) Lancet 2, 772-776 7. Cardelli-Cangiano, P., Cangiano, C., James, J. H., Jeppsson, B., Brenner, W., and Fischer, J. E. (1981) J. Neurochem. 36,627632 8. Hjelle, J . T., Baird-Lambert, J., Cardinale, G., Spector, S., and Udenfriend, S.(1978) Proc. Natl. Acad. Sci. U. S. A . 7 5 , 45444548 9. Torack, R. M., and Barrnett, R. J. (1964) J . Neuropathol. Exp. Neurol. 23,46-59 10. Orlowski, M., and Meister, A. (1965) J . Biol. Chem. 240, 338347 11. Gerlach, U., and Hiby, W. (1974) in Methods of Enzymatic Analysis (Bergmeyer, H. V., ed) Vol. 2, pp. 871-873, Academic Press, New York 12. Wellner, V. P., and Meister, A. (1966) Biochemistry 5,872-879 13. Williams, S.K., Gills, J. F., Matthews, M. A., Wagner, R. C., and

in Microvessels Brain Bitensky, M. W. (1980) J. Neurochen. 35,374-381 14. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J . Biol. Chem. 193, 265-275 15. Wilkinson, G. N. (1961) Biochem. J. 8 0 , 324-332 16. Christensen, H. N. (1975) Biological Transport, 2nd Ed., pp. 107460, Academic Press, New York 17. Oxender, D. L., and Christensen, H. N. (1963) J. Bid. Chem. 238,3686-3699 18. Christensen, H. N. (1979) Biochem. Phrmacol. 2 8 , 1989-1992 19. Christensen, H. N., andHandlogten,M.E. (1979) J . Neural Transm. Suppl. 15, 1-13 20. Cangiano, C., Cardelli Cangiano, P., James, J. H., and Fischer, J. E. (1980) Gastroenterology 78,1302 21. Cascino, A., Cangiano, C., Fiaccadori, F., Ghinelli, F., Merli, M., Pelosi, G., Riggio, O., Rossi-Fanelli, F., Sacchini, D., Stortoni, M., and Capocaccia, L. (1982) Dig. Dis. Sci. 27,828-832 22. Smith, A,, Rossi-Fanelli, F., Ziparo, V., James, J. H., Perelle, B. A,, and Fischer, J . E. (1978) Ann. Surg. 187, 343-350 23. Gjedde, A., Lockwood, A. H., Duffy, T. E., and Plum, F. (1979) Ann. Neurol. 3,325-330 24. Williams, A. H., Kyu, M. H., Fenton, J. C. B., and Cavanagh, J. B. (1972) J. Neurochem. 19, 1073-1077
