- Email: [email protected]
Changes in amino acid levels in rat plasma, cisternal cerebrospinal fluid, and brain tissue induced by intravenously infused arginine-vasopressin
Peptides, Vol. 16, No. 5, pp. 965-971, 1995 Copyright 0 1995 Elsevier Science Ltd Printed in the USA. All rights reserved 0196-9781/95 $9.50 + .OO
Pergamon
Changes in Amino Acid Levels in Rat Plasma, Cisternal Cerebrospinal Fluid, and Brain Tissue Induced by Intravenously Infused Arginine-Vasopressin ANDREAS REICHEL,*-f DAVID J. BEGLEYt’
AND ARMIN ERMISCH*
*Section of Biosciences, University of Leipzig, Talstrasse 33, 04103 Leipzig, Germany and fBiomedica1 Sciences Division, King’s College London, Strand, London WC2R 2L.S, UK
Received 20 December 1994 REICHEL, A., D. J. BEGLEY AND A. ERMISCH. Changes in amino acid levels in rat plasma, cisternal cerebrospinaljuid, and brain tissue induced by intravenously infused arginine-vasopressin. PEPTIDES 16(5) 965-971, 1995.-Circulating argininevasopressin (AVP) is known to reduce the blood-to-brain transfer of large neutral amino acids (AA). As a first step to examine whether the reduced uptake by brain endothelial cells is reflected in changes in large neutral amino acid levels of the extracellular fluid environment of cells within the nervous tissue, we measured the concentrations of amino acids in plasma, cerebrospinal fluid (CSF), and hippocampal tissue of rats before and after infusion of AVP (34 and 68 nglminkg, respectively) over the time period of 60 min. AA levels changed in all compartments investigated during both saline and AVP infusions. Whereas in the salineinfused controls changes in CSF AA levels paralleled those in plasma, this correlation was abolished by raising AVP concentrations. The effect of AVP was found to be i) dependent on the AA, ii) different with respect to direction and iii) magnitude of changes in AA levels, and iv) in some cases dose dependent. In summary, AVP infusion increased plasma levels of 10 AA, but decreased all 15 AA measured by some 30% in CSF. In contrast to CSF, levels of AA were slightly enhanced in the hippocampal tissue. The results are not solely explicable by a reduced blood-to-brain transfer of AA. We conclude that further mechanisms by which AVP affects the availability of AA to the brain may exist. The physiological significance of the findings might be related to brain osmoregulation, espec:ially in situations of stress. Amino acids
Blood-brain
barrier
CSF
Hippocampus
THERE is considerable evidence that arginine-vasopressin (AVP) induces alterations of amino acid (AA) transport at the blood-brain barrier (BBB) [(17) for review], and AVP administration in physiological concentrations leads to a decrease in the kinetic parameters of the blood-to-brain transfer of large neutral AA [e.g., leucine (6,8), phenylalanine (7,16), valine (32), methionine (9), and tyrosine: (33)]. These effects are not dependent on hemodynamic changes (6) but have been ascribed as a consequence of the occupation of AVP receptors (24) at the lmninal endothelial cell membrane of brain capillaries (15), which are presumably of the V, type (22). The effect is most pronounced in the hippocampus, where the highest density of AVP rece:ptors has been found (17). It has not yet been investigated whether the peptide simply induces a decrease of the uptake of large neutral AA into the cerebral capillary endothelium or whether the reduced uptake results in a depressed transport across the endothelial cells to the brain interstitial fluid, which, supplies brain cells with substances
Plasma
Vasopressin
essential for neuronal performance. Changes in the availability of AA within this compartment may influence the brain metabolism, (e.g., neurotransmitter synthesis). As a first step to investigate this problem, we infused saline or AVP for 1 h into the jugular vein and measured the concentration of 15 AA in samples from blood plasma, cerebrospinal fluid (CSF), and hippocampal tissue with a recently adapted HPLC method (2). The peptide infusion induced specific changes in AA levels in all compartments examined; the effects were dose dependent, and are likeiy to differ in their underlying causes. Changes in the CSF levels of AA may be interpreted as an effect induced by the peptide on AA transporters on the barriers separating blood and brain compartments (i.e., BBB and blood-CSF barrier) after occupation of AVP receptors. The data provide evidence that AVPinduced AA transport alterations from blood to brain are followed by changed concentrations of AA in brain interstitial fluids. Interestingly, AVP also altered CSF and brain tissue levels
’ Requests for reprints should be addressed to David Begley.
965
REICHEL, BEGLEY AND ERMISCH
of AA in a way likely to be independent on the BBB transport of AA. METHOD
Animals and Surgery Adult Wistar rats of both sexes, weighing 250-300 g, were maintained on a 12-h light-dark cycle and allowed food (Rat & Mouse No. 3 Breeding Diet, SDS, Witham, Essex, UK) and water ad lib. Following pentobarbitone anesthesia (Sagatal, BDH, 60 mg/kg, IP), a flexible polyethylene catheter was inserted into the right femoral artery. A second catheter, which was connected with an infusion pump, was inserted into the right jugular vein of the animal. Subsequently, implantation of a polyethylene catheter in the cisterna magna followed, as previously described (34). Briefly, rats were positioned in a stereotaxic frame and a burrhole formed as a 2-3-mm slot at the end of the external occipital crest with a dental drill. For implantation, each catheter was made rigid by the insertion of an enamelled copper wire into the lumen. A button, 6.5 mm from the end of the catheter, formed by gently heating the plastic, indicated the correct position within the cistema magna when touching the brain surface after insertion. Dental acrylic and a second burr-hole 5 mm lateral to midline helped to stabilize the system. To minimize animal-to-animal variation, the experiments were commenced at the same time each morning. All animal procedures were in strict accordance with the Home Office guidelines, and specifically licensed under the Animals (Scientific Procedures) Act of 1986. Plasma, CSF, and Tissue Sampling Immediately after surgery the CSF catheter was connected for sampling and a gentle negative pressure helped the CSF to flow. A CSF sample of 20-40 ~1 was withdrawn for analysis. Additionally, a 250 ~1 blood sample was withdrawn from the femoral artery and immediately centrifuged (5 mitt, 12,000 X g) to prepare the plasma sample. Postinfusion, a second CSF sample was taken from the cistema magna, and a final blood sample was taken directly from the right carotid artery and prepared as above. The animals were subsequently killed, the brain removed, and each hippocampus was dissected from the brain and homogenized.
X g for hippocampal tissue and plasma) samples were stored at -20°C until detection. The analysis of AA was based upon a two buffer HPLC system with fluorometric detection of precolumn derivatized primary AA with o-phthaldialdehyde (2,23). Concentrations of 15 AA were calculated for the compartments tested by referring to the detection response to AA standard solutions essentially as described elsewhere (2). Statistical Analysis Data are presented as mean values 2 SEM. Statistical significance of changes between AA levels before and after infusion were determined using two-tailed t-test for paired comparisons. Intergroup comparisons were tested for significance by one-way analysis of variance (ANOVA) of the A-values (i.e., AA level after infusions, minus AA level before infusions, at different doses of AVP) and Tukey’s test, when a significant effect was obtained. Relative changes of AA levels in plasma and CSF were tested for correlations using linear regression analysis. RESULTS
Infusion of Saline In plasma (Table l), all branched chain and basic AA rose by about half and other AA did not show the same within-group tendency (e.g., Glu levels decreased, but Asp levels increased). Although the majority of AA levels increased (12 out of 15), statistical significance was only achieved for Phe, Tyr, Val, Ser, and Tau. In CSF (Table 2), 12 AA levels rose, including all branched chain AA (by about 50%), aromatic AA (20%), and basic AA (20%). In contrast, levels of acidic AA decreased. Changes in CSF, however, correspond to changes in plasma because there were no significant differences, but a significant correlation between changes in AA levels of both compartments (r = 0.63, p < 0.05). In the hippocampus (Table 3), though there were no significant differences, again 11 AA levels rose, including branched chain, small neutral AA, and Tau, whereas other AA showed different within-group tendencies compared with controls. These findings are interpreted to suggest that anesthesia, surgical, and sampling procedures per se cause changes in basal plasma AA levels that are paralleled in both brain tissue and CSF.
AVP Infusion The AVP infusion was started after the initial CSF and plasma samples were taken. The infusion was carried out with an Harvard syringe pump over the time period of 60 min at 5 and 10 $/mm of a 1.6 /IM AVP solution. This produced an infusion rate of 34 and 68 ng AVP/min/kg rat, and significantly raised basal plasma levels of AVP (1.7 + 0.2 pg/ml) by approximately 2.5 times (4.4 ? 0.7 pg/ml) and 4.4 times (7.3 2 1.3 pg/ml), respectively, measured by radioimmunoassay (n = 4). In a control group the infusate contained isotonic saline. Here, plasma AVP levels were not significantly changed (1.8 2 0.2 pg/ml). It was determined, in a separate group of animals, that the above experimental rates of infusion of AVP did not elevate blood pressure. Biochemical Analysis Blood plasma and homogenized hippocampal tissue were deproteinized by adding a fourfold excess of methanol (4:1), and CSF samples by adding an equivalent volume of methanol (1: 1). After centrifuging (10 min, 5000 X g for CSF, and 5 min 10,000
Infusion of AVP Infusion of AVP produced changes in plasma (Table l), CSF (Table 2), and hippocampal tissue AA levels (Table 3), which do not correspond to changes caused by the procedure. In plasma, either AVP infusion increased the average levels of 11 AA (e.g., branched chain and basic AA rose significantly by 50% and 25%, respectively). Ala, Gln, and Glu decreased but the change did not reach statistical significance. In comparison with saline infusion, Glu was more strongly depressed when AVP was infused, whereas Tyr, Ser, and Gln decreased significantly despite their increased plasma levels after saline infusion. In CSF, all 15 AA decreased after AVP infusion with statistical significance for 13 AA, despite their enhanced plasma levels. Aromatic, branched chain AA, and Tau declined by about 25%, small neutral and acidic AA by about one-third. A correspondence of AA levels in CSF and plasma as found in the control animals was not observed, because infusion of 34 ng/min/kg AVP abolished the correlation found in saline-infused animals. In hippocampus, levels of 10 AA (mainly large neutral and small neutral AA) increased after AVP infusion, with significant
967
EFFECTS OF AVP IN PL.4SMA, CSF, AND BRAIN
TABLE 1 CONCENTRATIONS OF AA (PM) IN RAT PLASMA BEFORE AND AFTER AVP INFUSION S,alineInfusion Amino Acids
Phe Leu TY~ Trp Met Ile Val Ser Ala Gin Asp Gill LYs kg Tau
Before
32.43 83.68 19.26 95.91 50.40 47.27 80.16 95.42 375.42 396.65 4.67 34.86 324.17 113.31 17.55
After
+ 7.98 k 15.80 ‘-t 4.97 2 26.49 2 13.14 2 12.43 * 22.04 -c 8.67 2 99.23 2 94.87 2 0.90 2 6.03 t 61.73 2 54.4.8 + 6.24
45.08 172.62 26.42 91.97 55.64 79.35 135.16 138.92 309.21 474.74 6.18 31.84 547.71 153.87 23.57
2 2 + % 5 2 ‘+ 2 2 2 2 2 ? 5
5.22* 32.02 5.15* 19.94 1.68 4.43 15.94* 8.067 82.27 36.88 1.13 7.59 66.91 60.00 5.21-i.
Before
34.86 103.50 23.04 98.32 38.67 62.45 105.42 132.47 344.10 443.58 6.70 52.82 477.33 229.53 14.36
2 2 ? 2 2 2 2 z 2 + k k 2 + +
Values are given as mean k SEM, n = 5 (saline), n = 8 (34 ng/min/kg * p 6 0.05,-t p 6 0.01 (paired t-test) compared to before infusion.
differences only for Phe, Tyr, and Met. In contrast, concentrations of acidic and basic amino acids were lowered in response to the infusion, with significance only for Asp. Dose-Dependent
AVP (68 ng/min/kg)
AVP (34 nghinikg)
AVP Effects
In contrast to the variety of changes induced by AVP in comparison with control, the am~ount of AVP to be infused had less effect on AA levels. In plasma (Fig. l), despite slightly enhanced Phe levels after infusion of 68 nglminfkg AVP compared with half the infusion rate, there were no differences in AA levels between both infusion rates. In CSF (Fig. 2), some large neutral AA showed striking changes dependent upon the amount of AVP infused. Whereas all large neutral AA were depressed with the lower infusion rate (34 ng/min/kg AVP), the group subdivided into aromatic AA had the same percentage of decrease (about 25%) and the branched chain AA showed increased levels (about 13%) when twice the amount of AVP was infused (68 ng/min/kg). With exception of Ser, which also showed this characteristic, no other AA showed dose effects. In the hippocampus (Table 3), 6 AA levels were dependent upon the amount of circulating AVP. Leu, Tyr, Ile, Set-, Gln, and Tau levels were higher when 68 nglminikg AVP was infused than in the control group, but lower when half the amount of AVP was given, although this effect was only significant for Tyr. DISCUSSION
The aim of the present study was to investigate whether AVP, which reduces the extraction of AA at the BBB (17), also produces a corresponding alteration in free AA concentrations in CSF and brain. Basal levels of AA measured in plasma and CSF are in good agreement with previously published values [see (13) for review], attesting to the suitability of the recently adapted HPLC analysis for AA in these compartments (2). However it has to be kept in mind that all AA levels derive from animals anesthetized
3.00 7.15 1.77 9.48 2.96 5.65 12.31 7.45 30.09 43.31 0.72 4.33 26.1 I 24.54 2.11
34.42 173.29 22.62 99.19 51.89 84.12 147.86 142.98 248.69 362.66 6.53 36.37 555.25 283.30 15.97
+ z 2 2 2 ? 2 ? 2 2 t + 2 2 ”
4.20 12.18t 3.27 11.48 7.14* 6.31t 18.07$ 15.95 40.68t 37.53 1.10 5.28t 32.38t 41.88 2.89
AVP), n = 6 (68 ng/min/kg
After
Before
After
29.95 78.23 27.96 84.61 38.82 89.07 85.14 101.30 298.38 492.78 6.64 46.52 271.13 194.25 29.30
2 2 + + + 2 + 2 + 2 k 2 ? 2 5
3.51 3.84 3.73 4.18 3.19 11.96 17.45 19.12 6.60 33.89 0.64 2.67 17.42 18.03 3.32
37.39 150.60 21.54 78.83 46.82 149.87 167.86 123.91 295.18 413.46 7.45 36.93 375.23 260.91 26.60
k 2 + 5 2 k 2 2 2 ? k ? t ? %
3.87* 23.42.t 4.62 6.73 1.42 15.60t 13.281_ 15.38 25.57 32.17* 0.67 2.81-i 72.32 82.13.t 3.25
AVP).
with pentobarbitone, which is known to change plasma AA levels (20). Discrepancies in AA levels before the different infusions due to animal-to-animal and day-to-day variations were unavoidable. Therefore, measurements of AA levels before and after infusion were always made on the same animal so that each animal was its own control and a paired t-test was applicable. The infusion of AVP changed AA levels in plasma, CSF, and hippocampal tissue, though there was no general effect of the peptide. The complexity of alterations induced excludes a uniform mechanism underlying the observed diversity. In plasma 7 out of 15 AA showed significant changes after AVP infusion. Interestingly, branched chain and basic AA levels increased, whereas levels of the acidic AA Glu and the small neutral AA Ala and Gln decreased. AVP exerted no effect on aromatic AA and Tau levels in plasma. In CSF the infusion of AVP significantly reduced levels of 12 AA, where the mean reduction was about 30%. In contrast to CSF, most of the AA increased after AVP infusion in rat hippocampus; however, statistical significance was found only for the increases of Phe, Tyr, and Met levels and the decline in Asp. Some changes were found when saline was infused. The reason remains unclear, but may be ascribed to an activation of the sympathetic nervous system and/or a possible release of AVP as a response to the anesthetic or to taking the blood samples. In fact, AVP levels were found to be slightly increased after 1 h of saline infusion (about 7%). However, despite these changes in basal AA levels caused by the procedure itself, the effect of AVP on CSF levels of AA remains highly significant (Fig. 2). In control animals, increases in plasma AA levels were paralleled in CSF as AA concentrations in both compartments correlated significantly for saline-infused animals. AVP infusion, however, abolished the reflection of changes in plasma AA in their corresponding CSF levels as the correlation of AA concentrations be-
tween both compartments ceased. The results are interpreted in relation to the known ability of vasopressin to dowmegulate the activity of the L-system AA transporter at the BBB (17). Blood-borne AVP reduces the kinetic constants of the blood-to-brain transfer of several large neutral AA, as reported for Phe, Leu, Tyr, Met, and Val, without
968
REICHEL,
TABLE CONCENTRATIONS
OF AA (@f)
Phe Leu 5r Trp Met Ile Val SW Ala Gln Asp Glu LYS Ar8 Tau
Before
4.25 6.24 3.35 3.44 5.50 3.23 5.95 65.18 56.88 510.91 3.42 10.01 100.94 39.53 5.27
? 2 5 z 2 + ” ” + -t k ? 2 2 2
Before
5.04 2 0.59 10.23 k 1.57t 3.36 z 0.28 4.77 -c 0.95 5.50 t 0.36 5.51 + 0.8l.t 7.12 ” 0.8Ot 64.84 5 4.28 46.21 + 1.427 501.95 2 20.06 2.40 I 0.33 7.65 + 1.52 99.97 k 5.24 36.14? 1.69 5.43 2 1.00
0.14
0.47 1.08 3.46 2.14 24.10 0.94 2.26 14.02 1.95 1.17
2
AVP (34 nglmitig) After
0.26 0.90 0.26 0.50
AND ERMISCH
IN RAT CSF BEFORE AND AFTER AVP INFUSION
Saline Infusion Amino Acids
BEGLEY
3.74 k 9.02 2 2.02 t 1.58 t 4.57 ? 3.98 2 6.44 2 74.22 ? 75.20 + 567.53 2 4.25 + 14.24 k 66.54 k 52.97 2 2.15 2
AVP (68 ng/min/kg) After
0.63 1.76 0.42 0.14 1.07 0.76 0.36 8.20 14.24 90.69 0.64 0.61 5.44 9.79 0.42
3.13 2 0.70t 8.08 2 1.56* 1.49 -t 0.52t 1.08 k 0.20t 3.67 k 0.81* 3.10 2 0.27 4.46 k 0.15.t 57.80 2 8.42* 63.09 2 14.82* 362.36 2 76.01t 2.74 -c 0.74t 9.37 k 1.71.l 55.49 2 33.31 44.70 2 8.59 1.52 2 0.37t
Before
2.80 6.26 2.23 2.04 5.26 2.87 7.85 50.42 51.11 541.87 3.50 13.09 57.40 57.20 1.93
2 2 2 k k k 2 ” -t 2 2 2 ? 2 2
0.62 1.05 0.65 0.28 0.59 0.51 2.19 9.49 7.55 28.76 0.71 0.98 8.45 7.70 0.53
After
2.07 2 6.37 2 1.17 2 1.49 2 6.89 2 3.35 k 9.02 2 48.05 k 40.12 2 412.91 + 2.10 2 8.62 2 32.13 2 37.96 2 1.75 2
0.58.t 1.31 0.32’ 0.23t 0.78 0.75 2.34* 10.09 7.397 97.57 0.39t 1.75t 5.33* 3.92* 0.56
Values are given as mean ? SEM, n = 5 (saline), n = 5 (34 ngiminkg AVP, Lys: n = 3), n = 6 (68 nglminkg AVP). * p < 0.05, t p 2 0.01 (paired r-test) compared to before infusion.
surface of the BBB is not only related to the AA supply of the brain endothelium but is of some relevance to AA levels in the brain extracellular fluid and hence to AA availability and metabolism in the brain. Because there is a free diffusional exchange of small substances such as AA between cerebrospinal and interstitial fluid (5,12), possible changes in their interstitial levels should be seen in CSF. Indeed, infusion of AVP disturbed the reflection of changes in plasma AA levels within the CSF. Taking into account that only 10% of the total CSF volume is of extrachoroidal origin (36), the reduction in large neutral AA levels as measured in the present study (about 25%) is bigger than one would anticipate from a decline in the predicted unidi-
changes in hemodynamics. Reductions in both K,,, and V,,,,, for the transporter (the L-system) range from about 90% (VaI) to about 50% (Phe), revealing a negative correlation between respective substrate affinity and corresponding magnitude of transport depression (17). The decline in kinetic parameters of the
BBB transport of large neutral AA induced by AVP reduces the predicted unidirectional influx by approximately one-third for each of the five AA investigated so far (33). The findings indicate that elevated levels of circulating vasopressin, within the physiological range for periods of 60 min, reduce large neutral AA concentrations in the CSF, and provide evidence that the vasopressin-receptor interaction at the luminal
TABLE 3 CONCENTRATIONS
OF AA (nmol/g) IN RAT HLPPOCAMPUS BEFORE AND AFTER AVP INFUSION AVP Infusion
Amion Acids
Phe LeU
TYr Trp Met Ile Val Ser Ala Gln Asp Glu LYs Arg Tau
Control
7.17 29.31 6.74 13.86 77.77 10.36 208.01 214.01 3442.79 1543.55 460.38 2335.03 35.10 2127.14 2.66
5 5 + k 2 + 2 + + + 2 + 2 ? k
1.01 2.02 1.08 7.08 6.64 0.84 14.04 23.16 724.31 168.05 82.09 140.20 6.75 389.99 0.20
Values are given as mean + SEM, n = 4 (control), * p < 0.05 (ANOVA) compared to saline infusion.
Saline
8.04 35.77 6.52 12.80 82.67 12.04 209.02 312.72 4122.79 1780.37 514.73 2202.36 46.85 2011.17 3.47
2 2 t + + + 2 2 2 2 t 2 2 2 2
1.86 3.91 1.33 4.15 9.91 1.45 24.75 85.44 450.78 300.90 32.20 168.05 7.22 143.75 0.39
n = 6 (saline), n = 8 (34 ng/minkg
After 34 nglminikg
9.34 31.12 6.04 7.62 105.78 10.55 228.34 278.59 4141.79 1687.04 396.61 2140.72 37.53 1747.17 3.44
After 68 ng/min/kg
+ 1.84 + 3.55 ” 0.99 5 2.18 2 7.32* 2 1.26 5 7.41 2 52.50 t 512.45 2 240.61 + 46.74* 2 279.93 2 4.93 -)_ 190.19 k 0.54
AVP), n = 6 (68 nglminkg
16.02 37.69 17.69 11.18 97.01 12.72 232.38 310.64 4713.85 1898.19 322.51 1964.36 36.13 1922.44 4.47 AVP).
k 2 + + ? 2 + 2 2 + -t ? + + 2
3.41* 10.36 8.72* 4.54 12.14 4.04 30.36 88.45 114.45 191.47 121.87 518.54 12.52 521.89 0.95
EFFECTS
OF AVP IN PLASMA,
Ptle
Leu
969
CSF, AND BRAIN
Tyr
Trp
Met
ne
Val
SW
AIn
Gh
Amino
Asl,
Glu
Lys
Arg
TSJ
Acids
FIG. 1. Relative changes in AA levels in rat plasma given as percentage of the preinfusion level.
*p 5
0.05, **p s 0.01 (ANOVA of A-values) compared to saline infusion.
a logical site for the transport of solutes into and out of the CSF
rectional influx of LNAA from blood to brain by about 32% (33). Thus, we cannot escape the. conclusion that the depression by AVP of the unidirectional influx of LNAA across the BBB cannot solely account for the present findings. Therefore, additional events induced by AVP that lead, in combination, to the diminished AA levels in CSF should be considered. Apart from the effect on the BBB transport of large neutral AA, AVP has been described to affect the bulk flow of brain extracellular fluid (14) as well as the choroid plexus blood flow (18,19), both of which affect the rate of CSF production, enhancing its extrachoroidal portion. Both effects result in an enhanced extrachoroidal CSF production rate and thus intensify the effect of the AVP-induced reduction of the AA transport across the BBB on AA concentrations within the CSF. However, as such changes should be uniform for all AA, further events are likely to be involved. The choriod plexus has often been considered as
[(35) for review]. This might explain why AVP not only reduced CSF levels of AA, which are readily transported across the BBB, but also those that show no remarkable uptake at the BBB (i.e., small neutral, acidic AA, and Tau levels were also depressed after AVP infusion). Studies on the BBB transport reveal a considerable uptake only for large neutral and basic AA whereas small neutral and acidic AA are taken up to a much lesser degree, as only L and y+ but not A, ASC, or Gly transporters appear to be present at the luminal surface of endothelial cells [(28) for review]. At the blood side of the choroid plexus, however, a wide range of AA transporters could be identified, including uptake mechanisms for large neutral AA, cationic and acidic AA, and specific transporters for Gly, Pro, and Gln (31,35). Therefore, a movement of AA across the plexus epithelium has to be taken into account. It may well be that AVP affects these transfer pro-
x
250 l-
Ptw
Leu
Tyr
Trp
Met
Ie
VEIl
Ser
0
Salne
ia
34
I
68 ng/mln/kg
Ala
Amino
nghidkg
Gt,
AVP AVP
Asp
Glu
Lys
Arg
Ti3U
Acids
FIG. 2. Relative changes in AA levels in rat CSF given as percentage of the preinfusion level. *p 5 0.05, **p s 0.01 (ANOVA of A-values) compared to saline infusion.
970
REICHEL, BEGLEY AND ERMISCH
cesses as there are V, receptors located at the blood side of the choroid plexus with an affinity for AVP that is even higher (80 times) than at the luminal surface of the BBB (40). Moreover, no information is available as to whether AVP affects the transport of AA out of the brain, which is regarded as taking place for acidic and small neutral AA (4). The findings also indicate a dose dependency of the AVP effect on CSF levels of large neutral AA. At the lower infusion rate of AVP, levels of branched chain AA in CSF declined by about 25%, as levels of the other large neutral AA did; however, they were elevated (about 15%) when basal AVP levels rose by a factor of 4.4, whereas levels of the other large neutral (i.e., aromatic AA) declined as before. As proposed elsewhere (l), AA may share multiple transport systems at different anatomic locations in the brain. Circulating AVP could affect some of the transporters more than others, which may lead to gradually different changes in levels of certain AA. In contrast to CSF, vasopressin increased rather than decreased AA content of hippocampal tissue, as 10 out of 15 AA showed enhanced levels in response to AVP infusion. The concentration of Phe and Tyr more than doubled when plasma levels of AVP were raised by about 4.4 times. Although these are the first AA being released from AVP degradation (1 l), proteolytic conversion of AVP cannot explain this increase (nmol range) because the amounts of AVP infused are in the lower (pmol) range. The increases in these AA levels may influence neurotransmitter metabolism, as both AA serve as precursors for catecholamine synthesis (30). A significant effect was also found for levels of Met, which increased by 24% and 36%, respectively, dependent upon the amount of AVP infused. Because Met is a precursor for Tau synthesis (3). the accumulation of Met might be of significance for brain osmoregulation. Indeed, blood-borne vasopressin has repeatedly been suggested to be involved in brain osmoregulation. Elevated levels of the peptide were found in plasma and certain brain areas after osmotic stimulation (25). It has been hypothetized (9,10) that modulating the blood-to-brain transfer of AA into the brain extracellular fluid may be a mechanism for AVP involvement in brain osmoregulation. In fact, an adaptive decrease in brain free AA levels after chronic hyponatremia in mice to prevent brain edema has been reported (37). From that point of view, the enhanced AA levels in hippocampus homogenate, about 80% of which is intracellular fluid, could be a response to retain water in the cells and, thus, to maintain their normal cell volume (26). In terms of osmoprotective adaption, small neutral AA play a major role, whereby in certain aspects of cellular physiology, such as cellular excitability, Tau appears to be of special importance (27). Our findings do not contradict this hypothesis because even though CSF levels of Tau are depressed there is an increase in hippocampal Tau (by 30% or 70%, respectively, dependent upon the amount of AVP infused). Our results are also consistent with other findings (21) describing a stimulation of cellular uptake and utilization of AA by hypo-osmotic liver cell swelling. Similar mechanisms for cell volume regulation within the brain cannot be excluded. Our findings agree with this inasmuch as diminished levels of AA in cerebrospinal and thus interstitial
fluid may create an hypo-osmotic microenvironment that eventually might stimulate AA uptake into brain cells, as indicated by increased hippocampal AA levels. Although vasopressin is believed to be able to cross the BBB (38) the amounts of AVP reaching the abluminal side of the BBB under the conditions of the experiment could only be in the amol range according to previously reported kinetic data (39). Moreover, a rapid breakdown of AVP has been reported to occur at the brain side of the BBB (41). Therefore, a direct action of AVP molecules at neural elements as mechanisms to produce the effects seen is rather unlikely, suggesting that the acccumulation of AA in the hippocampal cells is indirectly stimulated. Our findings should be relevant for such situations in which vasopressin is physiologically released into the blood stream. Generally, but not exclusively, release of vasopressin occurs whenever the homeostasis of body fluids is disturbed. There is evidence that in such situations of stress plasma levels of AA are also changed (29). Because AA are potent osmolytes (27), an increased entry from blood into brain could disturb central volume regulation. Therefore, it is assumed that the effects of the peptide on the AA transport across the BBB become important, when plasma AA levels differ markedly from normal. The physiological function of the vasopressin-receptor interaction at the BBB may consist in maintaining the osmolytic homeostasis of the brain extracellular fluid by lowering the barrier transfer of AA, if the homeostasis is threatened by serious changes in the blood composition. The requirements for brain AA metabolism in such situations could be met by obtaining AA from the CSF, which then might act as an internal AA reservoir. In conclusion, peripheral application of AVP affects levels of AA in CSF and hippocampal tissue. The reason for the alterations we see as primarily AVP-induced changes in transport processes at the blood-brain and the blood-CSF barrier. The data give further evidence that peptides as ligands are involved in transport phenomena at barriers in organisms in general. In this special case, the interaction of the blood-borne nonapeptide with receptors at epithelial barriers of the brain produces changes in AA concentrations within the nervous tissue and thus affects their availability for brain metabolism. The detailed interpretation of the processes involved needs further investigation because, apart from effects of AVP on AA transporters, other mechanisms may come into play, such as altered production rates of interstitial and cerebrospinal fluid or changes in the metabolic utilization of AA by the nervous tissue. The changes in AA levels in CSF and brain elicited by AVP may be related to some physiological events, including osmoregulatory mechanisms. ACKNOWLEDGEMENTS We would like to thank Dr. Mary Forsling, St. Thomas’ Hospital, London, for performing the radioimmunoassay. This work was supported by a research grant from the Biomedical Sciences Division, King’s College London. Andreas Reichel was holding a grant by the German Academic Research Council (grant No. 312/332 4 00 075). David Begley was holding a British-German Academic Research collaboration grant for travel from the British Council. The authors would also like to thank Professor M. W. B. Bradbury for reading the draft manuscript and making many helpful suggestions.
REFERENCES 1. Begley, D. J.; Chain, D. G. Mechanisms regulating peptide levels in the cerebrospinal fluid. In: SegaJ, M. B., ed. Barriers and fluids of the eve and brain. Hampshire: Macmillan; 1992:82-105. 2. Begley, D. J.; Reichel,-A.; Ermisch, A. Simple HPLC analysis of free primary amino acid concentrations in rat plasma and cistemal cerebrospinal fluid. J. Chromatogr. [B] 657:185-191; 1994.
3. Bender, D. A. Amino acid metabolism. London: John Wiley; 1975. 4. Betz, A. L.; Goldstein, G. W. The basis for active transport at the blood-brain barrier. In: Eisenberg, H. M.; Suddith, R. Ll,, eds. Advances in experimental and medical biology. vol. 143. New York Plenum Press; 19815-16.
971
EFFECTS OF AVP IN PL.ASMA, CSF, AND BRAIN
5. Bradbury, M. W. B. The concept of a blood-brain barrier. ChiChester: Wiley; 1979. 6. Brust, P. Changes in regionad blood-brain barrier transfer of L-leutine elicited by arginine vasopressin. J. Neurochem. 46:534-547; 1986. 7. Brust, P.; Diemer, N. H. Blood-brain transfer of L-phenylalanine declines after peripheral bul: not central nervous administration of vasopressin. J. Neurochem. 55:2098-2104, 1990. 8. Brust, P.; Zicha, 1. Kinetics of regional blood-brain barrier transport of L-leucine in Brattleboro rats. Biomed. Biochim. Acta 47: 10131021; 1988. 9. Brust, P.; Shays, E. K.; Jeffties, K. J.; et al. Effects of vasopressin on blood-brain transfer 011methionine in dogs. J. Neurochem. 59:1421-1429; 1992. 10. Brust, P.; Christensen, T.; Diemer, N. H. Decrease of extracellular taurine in rat dorsal hippocampus after central nervous administration of vasopressin. J. Neurochem. 58: 1427- 143 1; 1992. 11. Burbach, J. P. H.; Lebouille, J. L. M.; Wang, X. C. Proteolytic conversion of neuropeptides into active fragments. In: Vizi, E. S.; Magyar, K., eds. Regulation of transmitter function: Basic and clinical aspects. Budapest: Akademia Kiado; 1984237-247. 12. Cserr, H. F.; Patlak, C. S. Regulation of brain volume under isosmotic and anisosmotic conditions. In: Gilles, R., ed. Advances in comparative and environmental physiology. vol. 9. Heidelberg: Springer; 1991:61-80. 13. Davson, H.; Welch, K.; Segal, M. B. The physiology and pathophysiology of the cerebrospinal fluid. New York: Churchill Livingstone; 1987. 14. DePasquale, M.; Patlak, C. S.; Cserr, H. F. Brain ion and volume regulation during acute hypematremia in Brattleboro rats. Am. J. Physiol. 256:Fl059-F1066: 1989. 15. Ennisch, A.; Landgraf, R.; 13rust, P.; Kretzschmar, R; Hess, J. Peptide receptors of the cerebral capillary endothelium and the transport of amino acids across the blood-brain barrier. In: Rakic, Lj.; Begley, D. J.; Davson, H.; Zlokovic, B. V., eds. Peptide and amino acid transport mechanisms in the central nervous system. New York: Stockton Press; 1988:41-53. 16. Ermisch, A.; Reichel, A.; Brust, P. Changes in the blood-brain barrier transfer of L-phenylalanine elicited by arginine vasopressin. Endoer. Regul. 26: II- 16; 1992. 17. Ermisch, A.; Brust, P.; Kreczschmar, R.; Riihle, H. J. Peptides and blood-brain barrier transport. Physiol. Rev. 73:489-527; 1993. 18. Faraci, F. M.; Mayhan, W. G.; Farrelli, W. J.; Heistad, D. D. Humoral regulation of blood flow to choroid plexus: Role of arginine vasopressin. Circ. Res. 63~373-379; 1988. 19. Faraci, F. M.; Mayhan, W. 13.; Heistad, D. D. Effect of vasopressin on production of cerebrospinal fluid: possible role of vasopressin (VI)-receptors. Am. J. Physiol. 258:R94-R98; 1990. 20. Hawkins, R. A.; Mans, A. IM.; Biebuyck, J. F. Amino acid supply to individual cerebral structures in awake and anesthetized rats. Am. J. Physiol. 242:El -El 1; 1982. 21. I%ussinger, D.; Lang, F. Cell volume in the regulation of hepatic function: A mechanism for metabolic control. B&him. Biophys. Acta 1071:331-350; 1991. 22. Hess, J.; Jensen, C. V.; Diemer, N. H. The vasopressin receptor of the blood-brain barrier in the rat hippocampus is linked to calcium signalling. Neurosci. Lett. 132:8-l& 1991. 23. Jones, B. N.; Gilligan, J. P. Amino acid analysis by o-phthaldialdehyde precolumn derivatization and reversed-phase HPLC. In: Mac-
24. 25. 26.
27. 28.
29. 30. 31. 32.
33.
34.
35. 36. 37.
38.
39. 40. 41.
Donald, J. C., ed. HPLC instrumentation and applications. Fairfield: International Scientific Communications, Inc; 1986:434-438. Kretzschmar, R.; Landgraf, R.; Gjedde, A.; Ermisch, A. Vasopressin binds to microvessels from rat hippocampus. Brain Res. 380:325330; 1986. Landgraf, R.; Neumann, I.; Schwarzberg, H. Central and peripheral release of vasopressin and oxytocin in the concious rat after osmotic stimulation. Brain Res. 457: 1291- 1300; 1988. Lang, F.; Vijlkl, H.; Hgussigner, D. General principles in cell volume regulation. In: Beyenbach, K. W., ed. Cell volume regulation. Basel: Karger; 1990: l-25. (Comp. Physiol., vol. 4.) Law, R. 0. Amino acids as volume regulatory osmolytes in mammalian cells. Comp. Biochem. Physiol. [A] 99:263-277; 1991. Lefauconnier, J.-M. Transport of amino acids. In: Bradbury, M. W. B., ed. Physiology and pharmacology of the blood-brain barrier. Heidelberg: Springer; 1992: 117- 150. Milakofsky, L.; Hare, T. A.; Miller, J. M.; Vogel, W. Rat plasma levels of amino acids and related compounds during stress. Life Sci. 36:753-761; 1985. Pardridge, W. H. Brain metabolism: A perspective from the bloodbrain barrier. Physiol. Rev. 63: 148I- 1535; 1983. Preston, J. E.; Segal, M. B. The steady-state amino acid fluxes across the perfused choroid plexus of the sheep. Brain Res. 525:275-279; 1990. Reichel, A.; Brust, P.; Ermisch, A. Effects of arginine vasopressin on the blood-brain transfer of certain large neutral amino acids (abstract). Third IBRO World Congress of Neuroscience Montreal, Canada, August 4-9; 1991:59.38. Reichel, A.; Begley, D. J.; Ermisch, A. Arginine vasopressin reduces the blood-brain transfer of L-tyrosine and L-valine: Further evidence of the effect of the peptide on the L system transporter at the blood-brain barrier (submitted). Sama, G. S.; Hutson, P. H.; Tricklebank, M. D.; Curzon, G. Determination of brain 5-hydroxytryptamine turnover in freely moving rats using repeated sampling of cerebrospinal fluid. J. N&rochei 40:383-388; 1983. Segal, M. B.; Zlokovic, B. V. The blood-brain barrier, amino acids and peptides. London: Kluwer Academic Publishers; 1990. Szentestvanyi, I.; Patlak, C. S.; Ellis, R. A.; Cserr, H. F. Drainage of interstitial fluid from different regions of rat brain. Am. J. Physiol. 246:F835-F844; 1984. Thurston, J. H.; Hauhart, R. E.; Nelson, J. S. Adaptive decrease in amino acids (taurine in particular), creatine, and electrolytes prevent cerebral edema in chronically hyponatremic mice: Rapid correction (experimental model of central pontine myelinolysis) causes dehydration and shrinkage of brain. Metab. Brain Dis. 2:223-241; 1987. Zlokovic, B. V.; Begley, D. J.; Segal, M. B.; et al. Neuropeptide transport mechanisms in the central nervous system. In: Rakic, Lj. J.; Begley, D. J.; Davson, H.; Zlokovic, B. V., eds. Peptide and amino acid transport mechanisms in the central nervous system. New York: Stockton Press; 1988:3- 19. Zlokovic, B. V.; Hyman, S.; McComb, J. G.; Lipovac, M. N.; Tang, G.; Davson, H. Kinetics of arginine-vasopressin uptake at the bloodbrain barrier. Biochim. Biophys. Acta 1025:191-198; 1990. Zlokovic, B. V.; Segal, M. B.; McComb, J. G.; Hyman, S.; Weiss, M. H.; Davson, H. Kinetics of circulating vasopressin uptake by choroid plexus. Am. J. Physiol. 260:F216-F224; 1991. Zlokovic, B. V.; Banks, W. A.; EIKadi, H.; et al. Transport, uptake, and metabolism of blood-borne vasopressin by the blood-brain barrier. Brain Res. 590:213-218; 1992.
Pergamon
Changes in Amino Acid Levels in Rat Plasma, Cisternal Cerebrospinal Fluid, and Brain Tissue Induced by Intravenously Infused Arginine-Vasopressin ANDREAS REICHEL,*-f DAVID J. BEGLEYt’
AND ARMIN ERMISCH*
*Section of Biosciences, University of Leipzig, Talstrasse 33, 04103 Leipzig, Germany and fBiomedica1 Sciences Division, King’s College London, Strand, London WC2R 2L.S, UK
Received 20 December 1994 REICHEL, A., D. J. BEGLEY AND A. ERMISCH. Changes in amino acid levels in rat plasma, cisternal cerebrospinaljuid, and brain tissue induced by intravenously infused arginine-vasopressin. PEPTIDES 16(5) 965-971, 1995.-Circulating argininevasopressin (AVP) is known to reduce the blood-to-brain transfer of large neutral amino acids (AA). As a first step to examine whether the reduced uptake by brain endothelial cells is reflected in changes in large neutral amino acid levels of the extracellular fluid environment of cells within the nervous tissue, we measured the concentrations of amino acids in plasma, cerebrospinal fluid (CSF), and hippocampal tissue of rats before and after infusion of AVP (34 and 68 nglminkg, respectively) over the time period of 60 min. AA levels changed in all compartments investigated during both saline and AVP infusions. Whereas in the salineinfused controls changes in CSF AA levels paralleled those in plasma, this correlation was abolished by raising AVP concentrations. The effect of AVP was found to be i) dependent on the AA, ii) different with respect to direction and iii) magnitude of changes in AA levels, and iv) in some cases dose dependent. In summary, AVP infusion increased plasma levels of 10 AA, but decreased all 15 AA measured by some 30% in CSF. In contrast to CSF, levels of AA were slightly enhanced in the hippocampal tissue. The results are not solely explicable by a reduced blood-to-brain transfer of AA. We conclude that further mechanisms by which AVP affects the availability of AA to the brain may exist. The physiological significance of the findings might be related to brain osmoregulation, espec:ially in situations of stress. Amino acids
Blood-brain
barrier
CSF
Hippocampus
THERE is considerable evidence that arginine-vasopressin (AVP) induces alterations of amino acid (AA) transport at the blood-brain barrier (BBB) [(17) for review], and AVP administration in physiological concentrations leads to a decrease in the kinetic parameters of the blood-to-brain transfer of large neutral AA [e.g., leucine (6,8), phenylalanine (7,16), valine (32), methionine (9), and tyrosine: (33)]. These effects are not dependent on hemodynamic changes (6) but have been ascribed as a consequence of the occupation of AVP receptors (24) at the lmninal endothelial cell membrane of brain capillaries (15), which are presumably of the V, type (22). The effect is most pronounced in the hippocampus, where the highest density of AVP rece:ptors has been found (17). It has not yet been investigated whether the peptide simply induces a decrease of the uptake of large neutral AA into the cerebral capillary endothelium or whether the reduced uptake results in a depressed transport across the endothelial cells to the brain interstitial fluid, which, supplies brain cells with substances
Plasma
Vasopressin
essential for neuronal performance. Changes in the availability of AA within this compartment may influence the brain metabolism, (e.g., neurotransmitter synthesis). As a first step to investigate this problem, we infused saline or AVP for 1 h into the jugular vein and measured the concentration of 15 AA in samples from blood plasma, cerebrospinal fluid (CSF), and hippocampal tissue with a recently adapted HPLC method (2). The peptide infusion induced specific changes in AA levels in all compartments examined; the effects were dose dependent, and are likeiy to differ in their underlying causes. Changes in the CSF levels of AA may be interpreted as an effect induced by the peptide on AA transporters on the barriers separating blood and brain compartments (i.e., BBB and blood-CSF barrier) after occupation of AVP receptors. The data provide evidence that AVPinduced AA transport alterations from blood to brain are followed by changed concentrations of AA in brain interstitial fluids. Interestingly, AVP also altered CSF and brain tissue levels
’ Requests for reprints should be addressed to David Begley.
965
REICHEL, BEGLEY AND ERMISCH
of AA in a way likely to be independent on the BBB transport of AA. METHOD
Animals and Surgery Adult Wistar rats of both sexes, weighing 250-300 g, were maintained on a 12-h light-dark cycle and allowed food (Rat & Mouse No. 3 Breeding Diet, SDS, Witham, Essex, UK) and water ad lib. Following pentobarbitone anesthesia (Sagatal, BDH, 60 mg/kg, IP), a flexible polyethylene catheter was inserted into the right femoral artery. A second catheter, which was connected with an infusion pump, was inserted into the right jugular vein of the animal. Subsequently, implantation of a polyethylene catheter in the cisterna magna followed, as previously described (34). Briefly, rats were positioned in a stereotaxic frame and a burrhole formed as a 2-3-mm slot at the end of the external occipital crest with a dental drill. For implantation, each catheter was made rigid by the insertion of an enamelled copper wire into the lumen. A button, 6.5 mm from the end of the catheter, formed by gently heating the plastic, indicated the correct position within the cistema magna when touching the brain surface after insertion. Dental acrylic and a second burr-hole 5 mm lateral to midline helped to stabilize the system. To minimize animal-to-animal variation, the experiments were commenced at the same time each morning. All animal procedures were in strict accordance with the Home Office guidelines, and specifically licensed under the Animals (Scientific Procedures) Act of 1986. Plasma, CSF, and Tissue Sampling Immediately after surgery the CSF catheter was connected for sampling and a gentle negative pressure helped the CSF to flow. A CSF sample of 20-40 ~1 was withdrawn for analysis. Additionally, a 250 ~1 blood sample was withdrawn from the femoral artery and immediately centrifuged (5 mitt, 12,000 X g) to prepare the plasma sample. Postinfusion, a second CSF sample was taken from the cistema magna, and a final blood sample was taken directly from the right carotid artery and prepared as above. The animals were subsequently killed, the brain removed, and each hippocampus was dissected from the brain and homogenized.
X g for hippocampal tissue and plasma) samples were stored at -20°C until detection. The analysis of AA was based upon a two buffer HPLC system with fluorometric detection of precolumn derivatized primary AA with o-phthaldialdehyde (2,23). Concentrations of 15 AA were calculated for the compartments tested by referring to the detection response to AA standard solutions essentially as described elsewhere (2). Statistical Analysis Data are presented as mean values 2 SEM. Statistical significance of changes between AA levels before and after infusion were determined using two-tailed t-test for paired comparisons. Intergroup comparisons were tested for significance by one-way analysis of variance (ANOVA) of the A-values (i.e., AA level after infusions, minus AA level before infusions, at different doses of AVP) and Tukey’s test, when a significant effect was obtained. Relative changes of AA levels in plasma and CSF were tested for correlations using linear regression analysis. RESULTS
Infusion of Saline In plasma (Table l), all branched chain and basic AA rose by about half and other AA did not show the same within-group tendency (e.g., Glu levels decreased, but Asp levels increased). Although the majority of AA levels increased (12 out of 15), statistical significance was only achieved for Phe, Tyr, Val, Ser, and Tau. In CSF (Table 2), 12 AA levels rose, including all branched chain AA (by about 50%), aromatic AA (20%), and basic AA (20%). In contrast, levels of acidic AA decreased. Changes in CSF, however, correspond to changes in plasma because there were no significant differences, but a significant correlation between changes in AA levels of both compartments (r = 0.63, p < 0.05). In the hippocampus (Table 3), though there were no significant differences, again 11 AA levels rose, including branched chain, small neutral AA, and Tau, whereas other AA showed different within-group tendencies compared with controls. These findings are interpreted to suggest that anesthesia, surgical, and sampling procedures per se cause changes in basal plasma AA levels that are paralleled in both brain tissue and CSF.
AVP Infusion The AVP infusion was started after the initial CSF and plasma samples were taken. The infusion was carried out with an Harvard syringe pump over the time period of 60 min at 5 and 10 $/mm of a 1.6 /IM AVP solution. This produced an infusion rate of 34 and 68 ng AVP/min/kg rat, and significantly raised basal plasma levels of AVP (1.7 + 0.2 pg/ml) by approximately 2.5 times (4.4 ? 0.7 pg/ml) and 4.4 times (7.3 2 1.3 pg/ml), respectively, measured by radioimmunoassay (n = 4). In a control group the infusate contained isotonic saline. Here, plasma AVP levels were not significantly changed (1.8 2 0.2 pg/ml). It was determined, in a separate group of animals, that the above experimental rates of infusion of AVP did not elevate blood pressure. Biochemical Analysis Blood plasma and homogenized hippocampal tissue were deproteinized by adding a fourfold excess of methanol (4:1), and CSF samples by adding an equivalent volume of methanol (1: 1). After centrifuging (10 min, 5000 X g for CSF, and 5 min 10,000
Infusion of AVP Infusion of AVP produced changes in plasma (Table l), CSF (Table 2), and hippocampal tissue AA levels (Table 3), which do not correspond to changes caused by the procedure. In plasma, either AVP infusion increased the average levels of 11 AA (e.g., branched chain and basic AA rose significantly by 50% and 25%, respectively). Ala, Gln, and Glu decreased but the change did not reach statistical significance. In comparison with saline infusion, Glu was more strongly depressed when AVP was infused, whereas Tyr, Ser, and Gln decreased significantly despite their increased plasma levels after saline infusion. In CSF, all 15 AA decreased after AVP infusion with statistical significance for 13 AA, despite their enhanced plasma levels. Aromatic, branched chain AA, and Tau declined by about 25%, small neutral and acidic AA by about one-third. A correspondence of AA levels in CSF and plasma as found in the control animals was not observed, because infusion of 34 ng/min/kg AVP abolished the correlation found in saline-infused animals. In hippocampus, levels of 10 AA (mainly large neutral and small neutral AA) increased after AVP infusion, with significant
967
EFFECTS OF AVP IN PL.4SMA, CSF, AND BRAIN
TABLE 1 CONCENTRATIONS OF AA (PM) IN RAT PLASMA BEFORE AND AFTER AVP INFUSION S,alineInfusion Amino Acids
Phe Leu TY~ Trp Met Ile Val Ser Ala Gin Asp Gill LYs kg Tau
Before
32.43 83.68 19.26 95.91 50.40 47.27 80.16 95.42 375.42 396.65 4.67 34.86 324.17 113.31 17.55
After
+ 7.98 k 15.80 ‘-t 4.97 2 26.49 2 13.14 2 12.43 * 22.04 -c 8.67 2 99.23 2 94.87 2 0.90 2 6.03 t 61.73 2 54.4.8 + 6.24
45.08 172.62 26.42 91.97 55.64 79.35 135.16 138.92 309.21 474.74 6.18 31.84 547.71 153.87 23.57
2 2 + % 5 2 ‘+ 2 2 2 2 2 ? 5
5.22* 32.02 5.15* 19.94 1.68 4.43 15.94* 8.067 82.27 36.88 1.13 7.59 66.91 60.00 5.21-i.
Before
34.86 103.50 23.04 98.32 38.67 62.45 105.42 132.47 344.10 443.58 6.70 52.82 477.33 229.53 14.36
2 2 ? 2 2 2 2 z 2 + k k 2 + +
Values are given as mean k SEM, n = 5 (saline), n = 8 (34 ng/min/kg * p 6 0.05,-t p 6 0.01 (paired t-test) compared to before infusion.
differences only for Phe, Tyr, and Met. In contrast, concentrations of acidic and basic amino acids were lowered in response to the infusion, with significance only for Asp. Dose-Dependent
AVP (68 ng/min/kg)
AVP (34 nghinikg)
AVP Effects
In contrast to the variety of changes induced by AVP in comparison with control, the am~ount of AVP to be infused had less effect on AA levels. In plasma (Fig. l), despite slightly enhanced Phe levels after infusion of 68 nglminfkg AVP compared with half the infusion rate, there were no differences in AA levels between both infusion rates. In CSF (Fig. 2), some large neutral AA showed striking changes dependent upon the amount of AVP infused. Whereas all large neutral AA were depressed with the lower infusion rate (34 ng/min/kg AVP), the group subdivided into aromatic AA had the same percentage of decrease (about 25%) and the branched chain AA showed increased levels (about 13%) when twice the amount of AVP was infused (68 ng/min/kg). With exception of Ser, which also showed this characteristic, no other AA showed dose effects. In the hippocampus (Table 3), 6 AA levels were dependent upon the amount of circulating AVP. Leu, Tyr, Ile, Set-, Gln, and Tau levels were higher when 68 nglminikg AVP was infused than in the control group, but lower when half the amount of AVP was given, although this effect was only significant for Tyr. DISCUSSION
The aim of the present study was to investigate whether AVP, which reduces the extraction of AA at the BBB (17), also produces a corresponding alteration in free AA concentrations in CSF and brain. Basal levels of AA measured in plasma and CSF are in good agreement with previously published values [see (13) for review], attesting to the suitability of the recently adapted HPLC analysis for AA in these compartments (2). However it has to be kept in mind that all AA levels derive from animals anesthetized
3.00 7.15 1.77 9.48 2.96 5.65 12.31 7.45 30.09 43.31 0.72 4.33 26.1 I 24.54 2.11
34.42 173.29 22.62 99.19 51.89 84.12 147.86 142.98 248.69 362.66 6.53 36.37 555.25 283.30 15.97
+ z 2 2 2 ? 2 ? 2 2 t + 2 2 ”
4.20 12.18t 3.27 11.48 7.14* 6.31t 18.07$ 15.95 40.68t 37.53 1.10 5.28t 32.38t 41.88 2.89
AVP), n = 6 (68 ng/min/kg
After
Before
After
29.95 78.23 27.96 84.61 38.82 89.07 85.14 101.30 298.38 492.78 6.64 46.52 271.13 194.25 29.30
2 2 + + + 2 + 2 + 2 k 2 ? 2 5
3.51 3.84 3.73 4.18 3.19 11.96 17.45 19.12 6.60 33.89 0.64 2.67 17.42 18.03 3.32
37.39 150.60 21.54 78.83 46.82 149.87 167.86 123.91 295.18 413.46 7.45 36.93 375.23 260.91 26.60
k 2 + 5 2 k 2 2 2 ? k ? t ? %
3.87* 23.42.t 4.62 6.73 1.42 15.60t 13.281_ 15.38 25.57 32.17* 0.67 2.81-i 72.32 82.13.t 3.25
AVP).
with pentobarbitone, which is known to change plasma AA levels (20). Discrepancies in AA levels before the different infusions due to animal-to-animal and day-to-day variations were unavoidable. Therefore, measurements of AA levels before and after infusion were always made on the same animal so that each animal was its own control and a paired t-test was applicable. The infusion of AVP changed AA levels in plasma, CSF, and hippocampal tissue, though there was no general effect of the peptide. The complexity of alterations induced excludes a uniform mechanism underlying the observed diversity. In plasma 7 out of 15 AA showed significant changes after AVP infusion. Interestingly, branched chain and basic AA levels increased, whereas levels of the acidic AA Glu and the small neutral AA Ala and Gln decreased. AVP exerted no effect on aromatic AA and Tau levels in plasma. In CSF the infusion of AVP significantly reduced levels of 12 AA, where the mean reduction was about 30%. In contrast to CSF, most of the AA increased after AVP infusion in rat hippocampus; however, statistical significance was found only for the increases of Phe, Tyr, and Met levels and the decline in Asp. Some changes were found when saline was infused. The reason remains unclear, but may be ascribed to an activation of the sympathetic nervous system and/or a possible release of AVP as a response to the anesthetic or to taking the blood samples. In fact, AVP levels were found to be slightly increased after 1 h of saline infusion (about 7%). However, despite these changes in basal AA levels caused by the procedure itself, the effect of AVP on CSF levels of AA remains highly significant (Fig. 2). In control animals, increases in plasma AA levels were paralleled in CSF as AA concentrations in both compartments correlated significantly for saline-infused animals. AVP infusion, however, abolished the reflection of changes in plasma AA in their corresponding CSF levels as the correlation of AA concentrations be-
tween both compartments ceased. The results are interpreted in relation to the known ability of vasopressin to dowmegulate the activity of the L-system AA transporter at the BBB (17). Blood-borne AVP reduces the kinetic constants of the blood-to-brain transfer of several large neutral AA, as reported for Phe, Leu, Tyr, Met, and Val, without
968
REICHEL,
TABLE CONCENTRATIONS
OF AA (@f)
Phe Leu 5r Trp Met Ile Val SW Ala Gln Asp Glu LYS Ar8 Tau
Before
4.25 6.24 3.35 3.44 5.50 3.23 5.95 65.18 56.88 510.91 3.42 10.01 100.94 39.53 5.27
? 2 5 z 2 + ” ” + -t k ? 2 2 2
Before
5.04 2 0.59 10.23 k 1.57t 3.36 z 0.28 4.77 -c 0.95 5.50 t 0.36 5.51 + 0.8l.t 7.12 ” 0.8Ot 64.84 5 4.28 46.21 + 1.427 501.95 2 20.06 2.40 I 0.33 7.65 + 1.52 99.97 k 5.24 36.14? 1.69 5.43 2 1.00
0.14
0.47 1.08 3.46 2.14 24.10 0.94 2.26 14.02 1.95 1.17
2
AVP (34 nglmitig) After
0.26 0.90 0.26 0.50
AND ERMISCH
IN RAT CSF BEFORE AND AFTER AVP INFUSION
Saline Infusion Amino Acids
BEGLEY
3.74 k 9.02 2 2.02 t 1.58 t 4.57 ? 3.98 2 6.44 2 74.22 ? 75.20 + 567.53 2 4.25 + 14.24 k 66.54 k 52.97 2 2.15 2
AVP (68 ng/min/kg) After
0.63 1.76 0.42 0.14 1.07 0.76 0.36 8.20 14.24 90.69 0.64 0.61 5.44 9.79 0.42
3.13 2 0.70t 8.08 2 1.56* 1.49 -t 0.52t 1.08 k 0.20t 3.67 k 0.81* 3.10 2 0.27 4.46 k 0.15.t 57.80 2 8.42* 63.09 2 14.82* 362.36 2 76.01t 2.74 -c 0.74t 9.37 k 1.71.l 55.49 2 33.31 44.70 2 8.59 1.52 2 0.37t
Before
2.80 6.26 2.23 2.04 5.26 2.87 7.85 50.42 51.11 541.87 3.50 13.09 57.40 57.20 1.93
2 2 2 k k k 2 ” -t 2 2 2 ? 2 2
0.62 1.05 0.65 0.28 0.59 0.51 2.19 9.49 7.55 28.76 0.71 0.98 8.45 7.70 0.53
After
2.07 2 6.37 2 1.17 2 1.49 2 6.89 2 3.35 k 9.02 2 48.05 k 40.12 2 412.91 + 2.10 2 8.62 2 32.13 2 37.96 2 1.75 2
0.58.t 1.31 0.32’ 0.23t 0.78 0.75 2.34* 10.09 7.397 97.57 0.39t 1.75t 5.33* 3.92* 0.56
Values are given as mean ? SEM, n = 5 (saline), n = 5 (34 ngiminkg AVP, Lys: n = 3), n = 6 (68 nglminkg AVP). * p < 0.05, t p 2 0.01 (paired r-test) compared to before infusion.
surface of the BBB is not only related to the AA supply of the brain endothelium but is of some relevance to AA levels in the brain extracellular fluid and hence to AA availability and metabolism in the brain. Because there is a free diffusional exchange of small substances such as AA between cerebrospinal and interstitial fluid (5,12), possible changes in their interstitial levels should be seen in CSF. Indeed, infusion of AVP disturbed the reflection of changes in plasma AA levels within the CSF. Taking into account that only 10% of the total CSF volume is of extrachoroidal origin (36), the reduction in large neutral AA levels as measured in the present study (about 25%) is bigger than one would anticipate from a decline in the predicted unidi-
changes in hemodynamics. Reductions in both K,,, and V,,,,, for the transporter (the L-system) range from about 90% (VaI) to about 50% (Phe), revealing a negative correlation between respective substrate affinity and corresponding magnitude of transport depression (17). The decline in kinetic parameters of the
BBB transport of large neutral AA induced by AVP reduces the predicted unidirectional influx by approximately one-third for each of the five AA investigated so far (33). The findings indicate that elevated levels of circulating vasopressin, within the physiological range for periods of 60 min, reduce large neutral AA concentrations in the CSF, and provide evidence that the vasopressin-receptor interaction at the luminal
TABLE 3 CONCENTRATIONS
OF AA (nmol/g) IN RAT HLPPOCAMPUS BEFORE AND AFTER AVP INFUSION AVP Infusion
Amion Acids
Phe LeU
TYr Trp Met Ile Val Ser Ala Gln Asp Glu LYs Arg Tau
Control
7.17 29.31 6.74 13.86 77.77 10.36 208.01 214.01 3442.79 1543.55 460.38 2335.03 35.10 2127.14 2.66
5 5 + k 2 + 2 + + + 2 + 2 ? k
1.01 2.02 1.08 7.08 6.64 0.84 14.04 23.16 724.31 168.05 82.09 140.20 6.75 389.99 0.20
Values are given as mean + SEM, n = 4 (control), * p < 0.05 (ANOVA) compared to saline infusion.
Saline
8.04 35.77 6.52 12.80 82.67 12.04 209.02 312.72 4122.79 1780.37 514.73 2202.36 46.85 2011.17 3.47
2 2 t + + + 2 2 2 2 t 2 2 2 2
1.86 3.91 1.33 4.15 9.91 1.45 24.75 85.44 450.78 300.90 32.20 168.05 7.22 143.75 0.39
n = 6 (saline), n = 8 (34 ng/minkg
After 34 nglminikg
9.34 31.12 6.04 7.62 105.78 10.55 228.34 278.59 4141.79 1687.04 396.61 2140.72 37.53 1747.17 3.44
After 68 ng/min/kg
+ 1.84 + 3.55 ” 0.99 5 2.18 2 7.32* 2 1.26 5 7.41 2 52.50 t 512.45 2 240.61 + 46.74* 2 279.93 2 4.93 -)_ 190.19 k 0.54
AVP), n = 6 (68 nglminkg
16.02 37.69 17.69 11.18 97.01 12.72 232.38 310.64 4713.85 1898.19 322.51 1964.36 36.13 1922.44 4.47 AVP).
k 2 + + ? 2 + 2 2 + -t ? + + 2
3.41* 10.36 8.72* 4.54 12.14 4.04 30.36 88.45 114.45 191.47 121.87 518.54 12.52 521.89 0.95
EFFECTS
OF AVP IN PLASMA,
Ptle
Leu
969
CSF, AND BRAIN
Tyr
Trp
Met
ne
Val
SW
AIn
Gh
Amino
Asl,
Glu
Lys
Arg
TSJ
Acids
FIG. 1. Relative changes in AA levels in rat plasma given as percentage of the preinfusion level.
*p 5
0.05, **p s 0.01 (ANOVA of A-values) compared to saline infusion.
a logical site for the transport of solutes into and out of the CSF
rectional influx of LNAA from blood to brain by about 32% (33). Thus, we cannot escape the. conclusion that the depression by AVP of the unidirectional influx of LNAA across the BBB cannot solely account for the present findings. Therefore, additional events induced by AVP that lead, in combination, to the diminished AA levels in CSF should be considered. Apart from the effect on the BBB transport of large neutral AA, AVP has been described to affect the bulk flow of brain extracellular fluid (14) as well as the choroid plexus blood flow (18,19), both of which affect the rate of CSF production, enhancing its extrachoroidal portion. Both effects result in an enhanced extrachoroidal CSF production rate and thus intensify the effect of the AVP-induced reduction of the AA transport across the BBB on AA concentrations within the CSF. However, as such changes should be uniform for all AA, further events are likely to be involved. The choriod plexus has often been considered as
[(35) for review]. This might explain why AVP not only reduced CSF levels of AA, which are readily transported across the BBB, but also those that show no remarkable uptake at the BBB (i.e., small neutral, acidic AA, and Tau levels were also depressed after AVP infusion). Studies on the BBB transport reveal a considerable uptake only for large neutral and basic AA whereas small neutral and acidic AA are taken up to a much lesser degree, as only L and y+ but not A, ASC, or Gly transporters appear to be present at the luminal surface of endothelial cells [(28) for review]. At the blood side of the choroid plexus, however, a wide range of AA transporters could be identified, including uptake mechanisms for large neutral AA, cationic and acidic AA, and specific transporters for Gly, Pro, and Gln (31,35). Therefore, a movement of AA across the plexus epithelium has to be taken into account. It may well be that AVP affects these transfer pro-
x
250 l-
Ptw
Leu
Tyr
Trp
Met
Ie
VEIl
Ser
0
Salne
ia
34
I
68 ng/mln/kg
Ala
Amino
nghidkg
Gt,
AVP AVP
Asp
Glu
Lys
Arg
Ti3U
Acids
FIG. 2. Relative changes in AA levels in rat CSF given as percentage of the preinfusion level. *p 5 0.05, **p s 0.01 (ANOVA of A-values) compared to saline infusion.
970
REICHEL, BEGLEY AND ERMISCH
cesses as there are V, receptors located at the blood side of the choroid plexus with an affinity for AVP that is even higher (80 times) than at the luminal surface of the BBB (40). Moreover, no information is available as to whether AVP affects the transport of AA out of the brain, which is regarded as taking place for acidic and small neutral AA (4). The findings also indicate a dose dependency of the AVP effect on CSF levels of large neutral AA. At the lower infusion rate of AVP, levels of branched chain AA in CSF declined by about 25%, as levels of the other large neutral AA did; however, they were elevated (about 15%) when basal AVP levels rose by a factor of 4.4, whereas levels of the other large neutral (i.e., aromatic AA) declined as before. As proposed elsewhere (l), AA may share multiple transport systems at different anatomic locations in the brain. Circulating AVP could affect some of the transporters more than others, which may lead to gradually different changes in levels of certain AA. In contrast to CSF, vasopressin increased rather than decreased AA content of hippocampal tissue, as 10 out of 15 AA showed enhanced levels in response to AVP infusion. The concentration of Phe and Tyr more than doubled when plasma levels of AVP were raised by about 4.4 times. Although these are the first AA being released from AVP degradation (1 l), proteolytic conversion of AVP cannot explain this increase (nmol range) because the amounts of AVP infused are in the lower (pmol) range. The increases in these AA levels may influence neurotransmitter metabolism, as both AA serve as precursors for catecholamine synthesis (30). A significant effect was also found for levels of Met, which increased by 24% and 36%, respectively, dependent upon the amount of AVP infused. Because Met is a precursor for Tau synthesis (3). the accumulation of Met might be of significance for brain osmoregulation. Indeed, blood-borne vasopressin has repeatedly been suggested to be involved in brain osmoregulation. Elevated levels of the peptide were found in plasma and certain brain areas after osmotic stimulation (25). It has been hypothetized (9,10) that modulating the blood-to-brain transfer of AA into the brain extracellular fluid may be a mechanism for AVP involvement in brain osmoregulation. In fact, an adaptive decrease in brain free AA levels after chronic hyponatremia in mice to prevent brain edema has been reported (37). From that point of view, the enhanced AA levels in hippocampus homogenate, about 80% of which is intracellular fluid, could be a response to retain water in the cells and, thus, to maintain their normal cell volume (26). In terms of osmoprotective adaption, small neutral AA play a major role, whereby in certain aspects of cellular physiology, such as cellular excitability, Tau appears to be of special importance (27). Our findings do not contradict this hypothesis because even though CSF levels of Tau are depressed there is an increase in hippocampal Tau (by 30% or 70%, respectively, dependent upon the amount of AVP infused). Our results are also consistent with other findings (21) describing a stimulation of cellular uptake and utilization of AA by hypo-osmotic liver cell swelling. Similar mechanisms for cell volume regulation within the brain cannot be excluded. Our findings agree with this inasmuch as diminished levels of AA in cerebrospinal and thus interstitial
fluid may create an hypo-osmotic microenvironment that eventually might stimulate AA uptake into brain cells, as indicated by increased hippocampal AA levels. Although vasopressin is believed to be able to cross the BBB (38) the amounts of AVP reaching the abluminal side of the BBB under the conditions of the experiment could only be in the amol range according to previously reported kinetic data (39). Moreover, a rapid breakdown of AVP has been reported to occur at the brain side of the BBB (41). Therefore, a direct action of AVP molecules at neural elements as mechanisms to produce the effects seen is rather unlikely, suggesting that the acccumulation of AA in the hippocampal cells is indirectly stimulated. Our findings should be relevant for such situations in which vasopressin is physiologically released into the blood stream. Generally, but not exclusively, release of vasopressin occurs whenever the homeostasis of body fluids is disturbed. There is evidence that in such situations of stress plasma levels of AA are also changed (29). Because AA are potent osmolytes (27), an increased entry from blood into brain could disturb central volume regulation. Therefore, it is assumed that the effects of the peptide on the AA transport across the BBB become important, when plasma AA levels differ markedly from normal. The physiological function of the vasopressin-receptor interaction at the BBB may consist in maintaining the osmolytic homeostasis of the brain extracellular fluid by lowering the barrier transfer of AA, if the homeostasis is threatened by serious changes in the blood composition. The requirements for brain AA metabolism in such situations could be met by obtaining AA from the CSF, which then might act as an internal AA reservoir. In conclusion, peripheral application of AVP affects levels of AA in CSF and hippocampal tissue. The reason for the alterations we see as primarily AVP-induced changes in transport processes at the blood-brain and the blood-CSF barrier. The data give further evidence that peptides as ligands are involved in transport phenomena at barriers in organisms in general. In this special case, the interaction of the blood-borne nonapeptide with receptors at epithelial barriers of the brain produces changes in AA concentrations within the nervous tissue and thus affects their availability for brain metabolism. The detailed interpretation of the processes involved needs further investigation because, apart from effects of AVP on AA transporters, other mechanisms may come into play, such as altered production rates of interstitial and cerebrospinal fluid or changes in the metabolic utilization of AA by the nervous tissue. The changes in AA levels in CSF and brain elicited by AVP may be related to some physiological events, including osmoregulatory mechanisms. ACKNOWLEDGEMENTS We would like to thank Dr. Mary Forsling, St. Thomas’ Hospital, London, for performing the radioimmunoassay. This work was supported by a research grant from the Biomedical Sciences Division, King’s College London. Andreas Reichel was holding a grant by the German Academic Research Council (grant No. 312/332 4 00 075). David Begley was holding a British-German Academic Research collaboration grant for travel from the British Council. The authors would also like to thank Professor M. W. B. Bradbury for reading the draft manuscript and making many helpful suggestions.
REFERENCES 1. Begley, D. J.; Chain, D. G. Mechanisms regulating peptide levels in the cerebrospinal fluid. In: SegaJ, M. B., ed. Barriers and fluids of the eve and brain. Hampshire: Macmillan; 1992:82-105. 2. Begley, D. J.; Reichel,-A.; Ermisch, A. Simple HPLC analysis of free primary amino acid concentrations in rat plasma and cistemal cerebrospinal fluid. J. Chromatogr. [B] 657:185-191; 1994.
3. Bender, D. A. Amino acid metabolism. London: John Wiley; 1975. 4. Betz, A. L.; Goldstein, G. W. The basis for active transport at the blood-brain barrier. In: Eisenberg, H. M.; Suddith, R. Ll,, eds. Advances in experimental and medical biology. vol. 143. New York Plenum Press; 19815-16.
971
EFFECTS OF AVP IN PL.ASMA, CSF, AND BRAIN
5. Bradbury, M. W. B. The concept of a blood-brain barrier. ChiChester: Wiley; 1979. 6. Brust, P. Changes in regionad blood-brain barrier transfer of L-leutine elicited by arginine vasopressin. J. Neurochem. 46:534-547; 1986. 7. Brust, P.; Diemer, N. H. Blood-brain transfer of L-phenylalanine declines after peripheral bul: not central nervous administration of vasopressin. J. Neurochem. 55:2098-2104, 1990. 8. Brust, P.; Zicha, 1. Kinetics of regional blood-brain barrier transport of L-leucine in Brattleboro rats. Biomed. Biochim. Acta 47: 10131021; 1988. 9. Brust, P.; Shays, E. K.; Jeffties, K. J.; et al. Effects of vasopressin on blood-brain transfer 011methionine in dogs. J. Neurochem. 59:1421-1429; 1992. 10. Brust, P.; Christensen, T.; Diemer, N. H. Decrease of extracellular taurine in rat dorsal hippocampus after central nervous administration of vasopressin. J. Neurochem. 58: 1427- 143 1; 1992. 11. Burbach, J. P. H.; Lebouille, J. L. M.; Wang, X. C. Proteolytic conversion of neuropeptides into active fragments. In: Vizi, E. S.; Magyar, K., eds. Regulation of transmitter function: Basic and clinical aspects. Budapest: Akademia Kiado; 1984237-247. 12. Cserr, H. F.; Patlak, C. S. Regulation of brain volume under isosmotic and anisosmotic conditions. In: Gilles, R., ed. Advances in comparative and environmental physiology. vol. 9. Heidelberg: Springer; 1991:61-80. 13. Davson, H.; Welch, K.; Segal, M. B. The physiology and pathophysiology of the cerebrospinal fluid. New York: Churchill Livingstone; 1987. 14. DePasquale, M.; Patlak, C. S.; Cserr, H. F. Brain ion and volume regulation during acute hypematremia in Brattleboro rats. Am. J. Physiol. 256:Fl059-F1066: 1989. 15. Ennisch, A.; Landgraf, R.; 13rust, P.; Kretzschmar, R; Hess, J. Peptide receptors of the cerebral capillary endothelium and the transport of amino acids across the blood-brain barrier. In: Rakic, Lj.; Begley, D. J.; Davson, H.; Zlokovic, B. V., eds. Peptide and amino acid transport mechanisms in the central nervous system. New York: Stockton Press; 1988:41-53. 16. Ermisch, A.; Reichel, A.; Brust, P. Changes in the blood-brain barrier transfer of L-phenylalanine elicited by arginine vasopressin. Endoer. Regul. 26: II- 16; 1992. 17. Ermisch, A.; Brust, P.; Kreczschmar, R.; Riihle, H. J. Peptides and blood-brain barrier transport. Physiol. Rev. 73:489-527; 1993. 18. Faraci, F. M.; Mayhan, W. G.; Farrelli, W. J.; Heistad, D. D. Humoral regulation of blood flow to choroid plexus: Role of arginine vasopressin. Circ. Res. 63~373-379; 1988. 19. Faraci, F. M.; Mayhan, W. 13.; Heistad, D. D. Effect of vasopressin on production of cerebrospinal fluid: possible role of vasopressin (VI)-receptors. Am. J. Physiol. 258:R94-R98; 1990. 20. Hawkins, R. A.; Mans, A. IM.; Biebuyck, J. F. Amino acid supply to individual cerebral structures in awake and anesthetized rats. Am. J. Physiol. 242:El -El 1; 1982. 21. I%ussinger, D.; Lang, F. Cell volume in the regulation of hepatic function: A mechanism for metabolic control. B&him. Biophys. Acta 1071:331-350; 1991. 22. Hess, J.; Jensen, C. V.; Diemer, N. H. The vasopressin receptor of the blood-brain barrier in the rat hippocampus is linked to calcium signalling. Neurosci. Lett. 132:8-l& 1991. 23. Jones, B. N.; Gilligan, J. P. Amino acid analysis by o-phthaldialdehyde precolumn derivatization and reversed-phase HPLC. In: Mac-
24. 25. 26.
27. 28.
29. 30. 31. 32.
33.
34.
35. 36. 37.
38.
39. 40. 41.
Donald, J. C., ed. HPLC instrumentation and applications. Fairfield: International Scientific Communications, Inc; 1986:434-438. Kretzschmar, R.; Landgraf, R.; Gjedde, A.; Ermisch, A. Vasopressin binds to microvessels from rat hippocampus. Brain Res. 380:325330; 1986. Landgraf, R.; Neumann, I.; Schwarzberg, H. Central and peripheral release of vasopressin and oxytocin in the concious rat after osmotic stimulation. Brain Res. 457: 1291- 1300; 1988. Lang, F.; Vijlkl, H.; Hgussigner, D. General principles in cell volume regulation. In: Beyenbach, K. W., ed. Cell volume regulation. Basel: Karger; 1990: l-25. (Comp. Physiol., vol. 4.) Law, R. 0. Amino acids as volume regulatory osmolytes in mammalian cells. Comp. Biochem. Physiol. [A] 99:263-277; 1991. Lefauconnier, J.-M. Transport of amino acids. In: Bradbury, M. W. B., ed. Physiology and pharmacology of the blood-brain barrier. Heidelberg: Springer; 1992: 117- 150. Milakofsky, L.; Hare, T. A.; Miller, J. M.; Vogel, W. Rat plasma levels of amino acids and related compounds during stress. Life Sci. 36:753-761; 1985. Pardridge, W. H. Brain metabolism: A perspective from the bloodbrain barrier. Physiol. Rev. 63: 148I- 1535; 1983. Preston, J. E.; Segal, M. B. The steady-state amino acid fluxes across the perfused choroid plexus of the sheep. Brain Res. 525:275-279; 1990. Reichel, A.; Brust, P.; Ermisch, A. Effects of arginine vasopressin on the blood-brain transfer of certain large neutral amino acids (abstract). Third IBRO World Congress of Neuroscience Montreal, Canada, August 4-9; 1991:59.38. Reichel, A.; Begley, D. J.; Ermisch, A. Arginine vasopressin reduces the blood-brain transfer of L-tyrosine and L-valine: Further evidence of the effect of the peptide on the L system transporter at the blood-brain barrier (submitted). Sama, G. S.; Hutson, P. H.; Tricklebank, M. D.; Curzon, G. Determination of brain 5-hydroxytryptamine turnover in freely moving rats using repeated sampling of cerebrospinal fluid. J. N&rochei 40:383-388; 1983. Segal, M. B.; Zlokovic, B. V. The blood-brain barrier, amino acids and peptides. London: Kluwer Academic Publishers; 1990. Szentestvanyi, I.; Patlak, C. S.; Ellis, R. A.; Cserr, H. F. Drainage of interstitial fluid from different regions of rat brain. Am. J. Physiol. 246:F835-F844; 1984. Thurston, J. H.; Hauhart, R. E.; Nelson, J. S. Adaptive decrease in amino acids (taurine in particular), creatine, and electrolytes prevent cerebral edema in chronically hyponatremic mice: Rapid correction (experimental model of central pontine myelinolysis) causes dehydration and shrinkage of brain. Metab. Brain Dis. 2:223-241; 1987. Zlokovic, B. V.; Begley, D. J.; Segal, M. B.; et al. Neuropeptide transport mechanisms in the central nervous system. In: Rakic, Lj. J.; Begley, D. J.; Davson, H.; Zlokovic, B. V., eds. Peptide and amino acid transport mechanisms in the central nervous system. New York: Stockton Press; 1988:3- 19. Zlokovic, B. V.; Hyman, S.; McComb, J. G.; Lipovac, M. N.; Tang, G.; Davson, H. Kinetics of arginine-vasopressin uptake at the bloodbrain barrier. Biochim. Biophys. Acta 1025:191-198; 1990. Zlokovic, B. V.; Segal, M. B.; McComb, J. G.; Hyman, S.; Weiss, M. H.; Davson, H. Kinetics of circulating vasopressin uptake by choroid plexus. Am. J. Physiol. 260:F216-F224; 1991. Zlokovic, B. V.; Banks, W. A.; EIKadi, H.; et al. Transport, uptake, and metabolism of blood-borne vasopressin by the blood-brain barrier. Brain Res. 590:213-218; 1992.