Acta Physiol Scand 1979, 107:161-167
A comparative study on the uptake and subsequent deca rboxylat ion of monoa mine precursors in cerebral microvessels J. E. HARDEBO, B. FALCK and CH. OWMAN Departments of Histology and Neurology. University of Lund, Sweden
HARDEBO, J . E . , FALCK, B. & OWMAN, CH.: A comparative study on the uptake and subsequent decarboxylation of monoamine precursors in cerebral microvessels. Acta Physiol Scand 1979, 107: 161-167. Received 3 April 1979. ISSN 0001-6772. Departments of Histology and Neurology, University of Lund, Sweden. The endothelial cells and pericytes of brain microvessels (capillaries and small veins) are equipped with an enzymatic barrier, impeding the passage of circulating amino acids, such as amine precursors, into the brain. The properties of this mechanism was studied in brain slices and isolated microvessels from various species including man and also fetal material, following incubation in dihydroxyphenylalanine (DOPA), 5-hydroxytryptophan (5-HTP) and dihydroxyphenylserine (DOPS). A stereospecific, energy-dependent uptake leading to accumulation in the brain microvessel walls was found in all species studied; this process was found to exist already prenatally. The capacity of decarboxylation, the second step in the trapping mechanism at the blood-brain interphase, showed considerable species variation. The enzyme was present also in fetal brain microvessels. Inhibition experiments provided support for the presence of monoamine oxidase, but absence of catechol-0methyl transferase, in the microvessel walls. Key vcord5: brain microvessels, aromatic aminoacid decarboxylase, monoamine oxidase,
blood-brain barrier
MATERIALS AND METHODS The microvessels (capillaries and small veins) in the central nervous system are capable of taking up and Animals. The majority of the study was performed on adult animals of either sex: 8 albino mice, 67 Spraguedecarboxylate monoamine precursors, such as L-3. Dawley rats, 8 hamsters, 6 guinea-pigs, 6 albino rabbits. 4 4-dihydroxyphenylalanine(L-DOPA) (Bertler et al. cats. 4 dogs, 4 pigs, 4 cows, 3 baboons and brain tissue 1964, 1966, Owman & Rosengren 1967, Bartholini from 5 humans (obtained during neurosurgical operaet al. 1971,Dowson & Laszlo 1971,Wade & Katztions). The laboratory animals had free access to standard pellet food and water. They were killed under light ether man 1975 a, b), L-5-hydroxytryptophan (L-5-HTP) (Bertler et al. 1964)and 3,4-dihydroxyphenylserine or nembutal anesthesia. In addition, fetuses of various age from rat (32 fetuses from 4 animals), rabbit (8 from 3 (DOPS) (Constantinidis et al. 1975).This presence animals) and human (in which pregnancy was interrupted of aromatic L-aminoacid decarboxylase (AAD) in at 20 and 22 weeks of gestation) was studied. Blood vesthe endothelial cells and pericytes of these vessels sels appear in the central nervous system of the rat on day 12-14 of gestation (Bar & Wolff 1972). in the rabbit at 1 1 constitute an enzymatic barrier, impeding the pasdays of gestation (Donahue 1964), and are present in husage of the monoamine precursors into the brain, as man fetuses at the 10th week of gestation (Povlishok et al. shown in the rat and mouse (Bertler et al. 1966). 1977). The present study was undertaken to investigate Incubation qf‘ tissue slices. One mm thin slices of the parietal cortex, caudate nucleus, cerebellum, spinal cord. the presence of the uptake mechanism for rnonoamine precursors, as well as the AAD activi- and heart from 4 animals of each species studied (from temporal cortex. frontal cortex or cerebellum of man) ty, in brain microvessels of various species includwere cut with a razor blade and transferred to incubation ing man. The appearance of these functions in the vials containing icecold Krebs-Ringer buffer solution. fetal brain was also studied. Slices were also taken from the brain hernispheres-prefI 1 795880 ~
162
J . E . H a r d e h o e t al.
erentially cortical tissue-f fetuses of various ages from rat, rabbit and man. The buffer solution had the following composition (mM): NaCl 118, KCI 4.5, CaClx2H,O 2.5, MgS04x7H,0 1. O , NaHCO, 25, KH,P04 1 .O to which was added 1 mglml glucose and 0.2 mglml ascorbic acid. The solution was continuously aerated with a mixture of 95 95 O2 and 5 % CO,, giving a pH of 7.4. The vials were placed in an incubation bath at 37°C for preincubation during 20 min in the presence of the decarboxylase inhibitors, carbidopa or benserazide ( lo-, M), followed by incubation during 20 min after the addition of the various monoamine precursors at a dose of 1, 10 or 100 pg/ml: L-DOPA, D-DOPA, (+I-erythro-DOPS, (-)-erythro-DOPS, (+)-threo-DOPS and L-5-HTP. In some of the incubations the monoamine oxidase inhibitor, nialamide ( M), or the catechol-0-methyltransferase inhibitor, U-0521 M), replaced carbidopa or benserazide. Control slices were incubated without addition of any drugs. The uptake of L-DOPA, L-5-HTP and (+)-erythroDOPS (in the presence of carbidopa) was further characterized in the presence of ouabain M), at low sodium and high potassium concentrations (in the buffer solution KCI substituted for all NaCl), under combined anoxia (continuous bubbling with 95 % N, and 5 % CO,) and absence of glucose, in the presence of dinitrophenol, and also at 0°C. Fluorescence microscopy. Following incubation, the various tissue pieces were frozen to the temperature of liquid nitrogen and further processed for fluorescent monoamine histochemistry according to the Falck-Hillarp method (Bjorklund, Falck & Owman 1972). The paraformaldehyde used in the histochemical procedure had previously been equilibrated with air of 70% humidity. The formaldehyde-induced, histochemically visible fluorophores of noradrenaline, adrenaline, dopamine, DOPA, and DOPS are indistinguishable since they have the same spectral characteristics and exhibit a green light under the optical conditions used (Bjorklund et al. 1972). 5-HTP and 5-hydroxytryptamine exhibit a yellow, rapidly fading light under these conditions (Bjorklund et al. 1972). When evaluating the changes in fluorescence intensity after incubations under various conditions, the material to be compared was always freeze-dried and further processed on the same occasion. Isolation of cerebral microvessels. Whole brains from rats were used. After removal they were kept in an icecold 67 mM phosphate buffer (pH 7.4) throughout the procedure. The meninges, including the pial membrane and its vessels, were carefully torn off and the choroid plexuses were removed. White matter, including the whole brain stem, was dissected away. Tissue obtained from 4-6 rats was chopped with a razor blade, and then gently further disrupted by some 10 strokes up and down through 3 different 1&20 ml plastic syringes equipped with a nylon net (pore size I OOO p m , 500 p m and 280 p m , respectively) glued to their open cut end. The material was subsequently homogenized by hand with a loosely fitting Teflon pestle in a smooth glass tube (0.1 mm clearance). The homogenate was washed through one nylon sieve with 225 p m pore size. The material remaining on the sieve, after extensive washing, was re-homogenized and re-sieved 3 4 times through the same, carefully
washed sieve. The tissue passing these sieves was collected and sieved through another sieve with 75 p m pore size. The material remaining on this sieve after washing was re-homogenized and re-sieved in the same, carefully washed sieve. The tissue fraction remaining on the last sieve, after extensive washing, consists of capillaries, venules and a few larger vessels, whereas most of the larger vessels and vessels branching off into clusters of small vessels, as well as a few small clusters of neurons and glia, remain on the 225 p m sieve (Hardebo et al. 1979). Neurons, glia, small microvessel segments, and subcellular fragments pass the 75 p m sieve. The capillary fraction is pure with regard to vessels and is only contaminated with a few glial endfeet, stuck onto the vessel wall (Hardeb0 et al. 1977), and possibly perivascular nerves. The yield of microvessels is less than 1/1OOO of the original wet weight of the grey matter. Brendel, Meezan & Carlson (1974) have shown that a mixed fraction of intracerebral vessels, prepared under similar conditions, is metabolically active. Incubation of isolated microvessels. The microvessel fractions were transferred to incubation vials containing ice-cold Krebs-Ringer buffer solution (composition as above). The vials were placed in an incubation bath and allowed to equilibrate for 20 min at 37°C in the presence of the decarboxylase inhibitor benserazide ( M) before incubation, which was started by adding ,H-L-DOPA to the solution. The tissue preparations were incubated for 15 min at a temperature of 37°C. The buffer solution was continuously aerated with 95% 0, and 5 % CO,. The incubation was terminated by transferring the tissue into icecold buffer solution. The microvessel fraction was collected by centrifugation at 4°C at 110 g and washed twice in isotope-free cold buffer for each 10 min. In order to further characterize the uptake, microvessel fractions were also incubated in the presence of ouabain (lOP M), and at 10°C. Measurement of radioactivity. After incubation, the microvessel samples were weighed and transferred to liquid scintillation vials. They were solubilized in 0.5 ml Soluene (Packard), and liquid scintillation counting was performed in 10 ml Instagel (Packard), as were appropriate samples (25 pl) of the incubation medium. Quench corrections were obtained according to conventional principles. Drugs. L-DOPA and D-DOPA (Sigma), (t)-threoDOPS, (+)-L- and (-)-D-erythro-DOPS (all gifts from Roche), L-5-HTP (Sigma), carbidopa (MSD), benserazide (Roche), ouabain (Sigma), nialamide (Pfizer), and 2,33H-L-DOPA (Radiochemical Centre, Amersham; 5 Ci/ mmol). Sraristics. Mean values were compared according to the Student's t-test for unpaired data.
RESULTS Fluorescence microscopy of the various brain regions from control slices incubated in buffer solution alone showed a dense network of delicate catecholamine-containing nerve terminals emitting
SF'OCKWYLM
Addition of drugs or modification of medium
TIM
T
L
Fluorescence intensity (arbitrary units1
0
(*)
+
*+
None Ouabain lo-' M
LOW Na', high K+ Anoxia without glucose Dinitrophenol 5 =lo-' M Incubation at O'C Nialamide 10" M
Fig. 1 a . Accumulation of L-DOPA in the walls of microvessels in rat parietal cortex slices pretreated with benserazide M and incubated in the presence of 10 pglml L-DOPA alone and after addition of drugs or modification of the medium (following decarboxylase inhibition by M benserazide). For comparison, the fluorescence intensity following pretreatment with nialamide instead of benserazide is shown (the accumulated fluorophore now consists of Fig. I b. Tissuelmedium ratio (r/M; mg tissue per pl medium) following 15 min incubation of isolated cerebral M benserazide) under control conditions microvessels with 3H-L-DOPA (and after decarboxylase inhibition by (37°C) and in the presence of ouabain M) or during hypothermia (0°C). Values are means k S . E . ; number of experiments within parenthesis. Control vs. experimental according to Student's I-test: *O.Ol
A comparative study on the uptake and subsequent deca rboxylat ion of monoa mine precursors in cerebral microvessels J. E. HARDEBO, B. FALCK and CH. OWMAN Departments of Histology and Neurology. University of Lund, Sweden
HARDEBO, J . E . , FALCK, B. & OWMAN, CH.: A comparative study on the uptake and subsequent decarboxylation of monoamine precursors in cerebral microvessels. Acta Physiol Scand 1979, 107: 161-167. Received 3 April 1979. ISSN 0001-6772. Departments of Histology and Neurology, University of Lund, Sweden. The endothelial cells and pericytes of brain microvessels (capillaries and small veins) are equipped with an enzymatic barrier, impeding the passage of circulating amino acids, such as amine precursors, into the brain. The properties of this mechanism was studied in brain slices and isolated microvessels from various species including man and also fetal material, following incubation in dihydroxyphenylalanine (DOPA), 5-hydroxytryptophan (5-HTP) and dihydroxyphenylserine (DOPS). A stereospecific, energy-dependent uptake leading to accumulation in the brain microvessel walls was found in all species studied; this process was found to exist already prenatally. The capacity of decarboxylation, the second step in the trapping mechanism at the blood-brain interphase, showed considerable species variation. The enzyme was present also in fetal brain microvessels. Inhibition experiments provided support for the presence of monoamine oxidase, but absence of catechol-0methyl transferase, in the microvessel walls. Key vcord5: brain microvessels, aromatic aminoacid decarboxylase, monoamine oxidase,
blood-brain barrier
MATERIALS AND METHODS The microvessels (capillaries and small veins) in the central nervous system are capable of taking up and Animals. The majority of the study was performed on adult animals of either sex: 8 albino mice, 67 Spraguedecarboxylate monoamine precursors, such as L-3. Dawley rats, 8 hamsters, 6 guinea-pigs, 6 albino rabbits. 4 4-dihydroxyphenylalanine(L-DOPA) (Bertler et al. cats. 4 dogs, 4 pigs, 4 cows, 3 baboons and brain tissue 1964, 1966, Owman & Rosengren 1967, Bartholini from 5 humans (obtained during neurosurgical operaet al. 1971,Dowson & Laszlo 1971,Wade & Katztions). The laboratory animals had free access to standard pellet food and water. They were killed under light ether man 1975 a, b), L-5-hydroxytryptophan (L-5-HTP) (Bertler et al. 1964)and 3,4-dihydroxyphenylserine or nembutal anesthesia. In addition, fetuses of various age from rat (32 fetuses from 4 animals), rabbit (8 from 3 (DOPS) (Constantinidis et al. 1975).This presence animals) and human (in which pregnancy was interrupted of aromatic L-aminoacid decarboxylase (AAD) in at 20 and 22 weeks of gestation) was studied. Blood vesthe endothelial cells and pericytes of these vessels sels appear in the central nervous system of the rat on day 12-14 of gestation (Bar & Wolff 1972). in the rabbit at 1 1 constitute an enzymatic barrier, impeding the pasdays of gestation (Donahue 1964), and are present in husage of the monoamine precursors into the brain, as man fetuses at the 10th week of gestation (Povlishok et al. shown in the rat and mouse (Bertler et al. 1966). 1977). The present study was undertaken to investigate Incubation qf‘ tissue slices. One mm thin slices of the parietal cortex, caudate nucleus, cerebellum, spinal cord. the presence of the uptake mechanism for rnonoamine precursors, as well as the AAD activi- and heart from 4 animals of each species studied (from temporal cortex. frontal cortex or cerebellum of man) ty, in brain microvessels of various species includwere cut with a razor blade and transferred to incubation ing man. The appearance of these functions in the vials containing icecold Krebs-Ringer buffer solution. fetal brain was also studied. Slices were also taken from the brain hernispheres-prefI 1 795880 ~
162
J . E . H a r d e h o e t al.
erentially cortical tissue-f fetuses of various ages from rat, rabbit and man. The buffer solution had the following composition (mM): NaCl 118, KCI 4.5, CaClx2H,O 2.5, MgS04x7H,0 1. O , NaHCO, 25, KH,P04 1 .O to which was added 1 mglml glucose and 0.2 mglml ascorbic acid. The solution was continuously aerated with a mixture of 95 95 O2 and 5 % CO,, giving a pH of 7.4. The vials were placed in an incubation bath at 37°C for preincubation during 20 min in the presence of the decarboxylase inhibitors, carbidopa or benserazide ( lo-, M), followed by incubation during 20 min after the addition of the various monoamine precursors at a dose of 1, 10 or 100 pg/ml: L-DOPA, D-DOPA, (+I-erythro-DOPS, (-)-erythro-DOPS, (+)-threo-DOPS and L-5-HTP. In some of the incubations the monoamine oxidase inhibitor, nialamide ( M), or the catechol-0-methyltransferase inhibitor, U-0521 M), replaced carbidopa or benserazide. Control slices were incubated without addition of any drugs. The uptake of L-DOPA, L-5-HTP and (+)-erythroDOPS (in the presence of carbidopa) was further characterized in the presence of ouabain M), at low sodium and high potassium concentrations (in the buffer solution KCI substituted for all NaCl), under combined anoxia (continuous bubbling with 95 % N, and 5 % CO,) and absence of glucose, in the presence of dinitrophenol, and also at 0°C. Fluorescence microscopy. Following incubation, the various tissue pieces were frozen to the temperature of liquid nitrogen and further processed for fluorescent monoamine histochemistry according to the Falck-Hillarp method (Bjorklund, Falck & Owman 1972). The paraformaldehyde used in the histochemical procedure had previously been equilibrated with air of 70% humidity. The formaldehyde-induced, histochemically visible fluorophores of noradrenaline, adrenaline, dopamine, DOPA, and DOPS are indistinguishable since they have the same spectral characteristics and exhibit a green light under the optical conditions used (Bjorklund et al. 1972). 5-HTP and 5-hydroxytryptamine exhibit a yellow, rapidly fading light under these conditions (Bjorklund et al. 1972). When evaluating the changes in fluorescence intensity after incubations under various conditions, the material to be compared was always freeze-dried and further processed on the same occasion. Isolation of cerebral microvessels. Whole brains from rats were used. After removal they were kept in an icecold 67 mM phosphate buffer (pH 7.4) throughout the procedure. The meninges, including the pial membrane and its vessels, were carefully torn off and the choroid plexuses were removed. White matter, including the whole brain stem, was dissected away. Tissue obtained from 4-6 rats was chopped with a razor blade, and then gently further disrupted by some 10 strokes up and down through 3 different 1&20 ml plastic syringes equipped with a nylon net (pore size I OOO p m , 500 p m and 280 p m , respectively) glued to their open cut end. The material was subsequently homogenized by hand with a loosely fitting Teflon pestle in a smooth glass tube (0.1 mm clearance). The homogenate was washed through one nylon sieve with 225 p m pore size. The material remaining on the sieve, after extensive washing, was re-homogenized and re-sieved 3 4 times through the same, carefully
washed sieve. The tissue passing these sieves was collected and sieved through another sieve with 75 p m pore size. The material remaining on this sieve after washing was re-homogenized and re-sieved in the same, carefully washed sieve. The tissue fraction remaining on the last sieve, after extensive washing, consists of capillaries, venules and a few larger vessels, whereas most of the larger vessels and vessels branching off into clusters of small vessels, as well as a few small clusters of neurons and glia, remain on the 225 p m sieve (Hardebo et al. 1979). Neurons, glia, small microvessel segments, and subcellular fragments pass the 75 p m sieve. The capillary fraction is pure with regard to vessels and is only contaminated with a few glial endfeet, stuck onto the vessel wall (Hardeb0 et al. 1977), and possibly perivascular nerves. The yield of microvessels is less than 1/1OOO of the original wet weight of the grey matter. Brendel, Meezan & Carlson (1974) have shown that a mixed fraction of intracerebral vessels, prepared under similar conditions, is metabolically active. Incubation of isolated microvessels. The microvessel fractions were transferred to incubation vials containing ice-cold Krebs-Ringer buffer solution (composition as above). The vials were placed in an incubation bath and allowed to equilibrate for 20 min at 37°C in the presence of the decarboxylase inhibitor benserazide ( M) before incubation, which was started by adding ,H-L-DOPA to the solution. The tissue preparations were incubated for 15 min at a temperature of 37°C. The buffer solution was continuously aerated with 95% 0, and 5 % CO,. The incubation was terminated by transferring the tissue into icecold buffer solution. The microvessel fraction was collected by centrifugation at 4°C at 110 g and washed twice in isotope-free cold buffer for each 10 min. In order to further characterize the uptake, microvessel fractions were also incubated in the presence of ouabain (lOP M), and at 10°C. Measurement of radioactivity. After incubation, the microvessel samples were weighed and transferred to liquid scintillation vials. They were solubilized in 0.5 ml Soluene (Packard), and liquid scintillation counting was performed in 10 ml Instagel (Packard), as were appropriate samples (25 pl) of the incubation medium. Quench corrections were obtained according to conventional principles. Drugs. L-DOPA and D-DOPA (Sigma), (t)-threoDOPS, (+)-L- and (-)-D-erythro-DOPS (all gifts from Roche), L-5-HTP (Sigma), carbidopa (MSD), benserazide (Roche), ouabain (Sigma), nialamide (Pfizer), and 2,33H-L-DOPA (Radiochemical Centre, Amersham; 5 Ci/ mmol). Sraristics. Mean values were compared according to the Student's t-test for unpaired data.
RESULTS Fluorescence microscopy of the various brain regions from control slices incubated in buffer solution alone showed a dense network of delicate catecholamine-containing nerve terminals emitting
SF'OCKWYLM
Addition of drugs or modification of medium
TIM
T
L
Fluorescence intensity (arbitrary units1
0
(*)
+
*+
None Ouabain lo-' M
LOW Na', high K+ Anoxia without glucose Dinitrophenol 5 =lo-' M Incubation at O'C Nialamide 10" M
Fig. 1 a . Accumulation of L-DOPA in the walls of microvessels in rat parietal cortex slices pretreated with benserazide M and incubated in the presence of 10 pglml L-DOPA alone and after addition of drugs or modification of the medium (following decarboxylase inhibition by M benserazide). For comparison, the fluorescence intensity following pretreatment with nialamide instead of benserazide is shown (the accumulated fluorophore now consists of Fig. I b. Tissuelmedium ratio (r/M; mg tissue per pl medium) following 15 min incubation of isolated cerebral M benserazide) under control conditions microvessels with 3H-L-DOPA (and after decarboxylase inhibition by (37°C) and in the presence of ouabain M) or during hypothermia (0°C). Values are means k S . E . ; number of experiments within parenthesis. Control vs. experimental according to Student's I-test: *O.Ol
