J. Inher. Metab. Dis. 13 (1990) 627-633 © SStEM and KluwerAcademicPublishers. Printed in the Netherlands
Maternal PKU Workshop
Transport of Amino Acids across the BloodBrain Barrier: Implications for Treatment of Maternal Phenylketonuria R. M. GARD1NER
Departments of Paediatrics and Human Metabolism, University College London, The Rayne Institute, University Street, London WCIE 6AU, UK; Present address: Department of Paediatrics, John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK Summary: Amino acid transport at the mammalian blood-brain barrier has been extensively characterized. Transport of L-phenylalanine and related neutral amino acids is known to be mediated by a stereospecific, sodium independent, saturable carrier. The affinity of this carrier is much higher than that of comparable systems in other tissues. This feature renders it susceptible to inhibition. It has been suggested that inhibition of neutral amino acid influx into the brain by hyperphenylalaninaemia contributes to the pathophysiology of brain damage in this condition. Methods for investigation of amino acid transport at the blood-brain barrier are discussed, a n d current knowledge of blood-brain barrier amino acid transport at the blood-brain barrier is reviewed. Developmental changes are delineated, with particular reference to recent work on the ovine blood-brain barrier. There is insufficient information concerning blood-brain barrier transport of amino acids in the fetal brain to allow firm conclusions to be drawn concerning implications for treatment of maternal PKU. Reasonable extrapolation from animal data suggests that transport inhibition may contribute to impaired fetal brain growth in maternal PKU, and can be minimized by attempts to maintain a normal milieu from the time of conception.
The precise mechanisms by which raised circulating phenylalanine levels exert a deleterious effect on the developing brain, either in utero or postnatally, remain uncertain. It has been suggested that the special properties of the transport mechanisms which mediate transfer of amino acids from the circulation to their site of metabolism within neurones may be relevant to the pathophysiology of cerebral damage in hyperphenylalaninaemia. In particular, it has been proposed that inhibition of blood627
628
Gardiner
brain barrier transport of related neutral amino acids by a high concentration of phenylalanine may be an important pathogenetic mechanism (Pratt, 1982). Information necessary to evaluate fully the r6te of amino acid transport at the blood-brain barrier in the pathophysiology of hyperphenylalaninaemia in utero remains incomplete. Amino acid transport at the blood-brain barrier has been extensively characterized in a variety of animat species, but there is limited information on changes occurring during development. However, a non-invasive approach to quantifying blood brain barrier transport in human patients, let alone the fetus, is not available, and interspecies extrapolation in this area is notoriously difficult. This paper reviews the methods available for the study of blood-brain barrier amino acid transport, and summarizes current knowledge of these transport systems in the mammalian brain. Developmental changes are considered, with particular reference to changes at the ovine blood-brain barrier. The possible r61e in pathophysiology of transport at this interface is discussed, with special emphasis on maternal PKU, and gaps in our present knowledge are identified. TRANSPORT AT THE BLOOD-BRAIN B A R R I E R It is now generally accepted that the anatomical basis of the blood-brain barrier lies in the characteristic structure of cerebral capillaries (Bradbury, t979). In particular, the tight-junctions between capillary endothelial cells create a vascular barrier impermeable to water soluble substrates (such as amino acids) in the absence of a specific transport mechanism. The plasma membrane of brain capillary endothelial cells is the site of several specific transport mechanisms including those for glucose, monocarboxylic acids, and amino acids. There is evidence for polarity: different systems may be present at the luminal and abluminal interfaces. Maturation of barrier function has been studied in several species (Bradbury, 1979). The old idea that the blood-brain barrier in the 'newborn' is 'open', is now known to be false and represents a gross oversimplification of the complex anatomical and functional changes which have been documented in a variety of species at various periods of gestation. M E T H O D S FOR INVESTIGATING AMINO ACID TRANSPORT AT THE BLOOD-BRAIN B A R R I E R During the last two decades both in vivo and in vitro techniques have been developed, but none are applicable in man. Quantitative information is obtained, including values for the kinetic parameters of transport, but the particular method employed has an important bearing on the interpretation of the results. Measurement of net uptake into the brain is extremely difficult as A-V concentration differences are small. In vivo methods:
Single-pass Single-pass Variable-infusion I n situ brain perfusion J. Inher, Metab. Dis, 13 (1990)
tissue sampling (Oldendorf, 1971) - indicator-dilution (Yudilevich et al., 1972) tissue sampling (Pratt, 1976) tissue sampling (Takasato et al., 1984)
629
Amino Acid Transport into Brain In vitro methods:
Observations on isolated brain capillaries (Joo, 1985). Sampling of brain tissue (following decapitation) after intracarotid injection of the test amino acid and a highly diffusible reference tracer was originally employed by Oldendorf (1971). Results are expressed as a brain uptake index. Although kinetic parameters may be estimated, quantitation of influx and permeability is not possible, and the degree of mixing of bolus and circulating blood is poorly defined. A recent modification of this approach, the in situ brain perfusion technique (Takasato et al., 1984) has several advantages. In particular, permeability can be estimated and the composition of the perfusion fluid is not changed appreciably by mixing with blood (Momma et al., 1987). The single-pass indicator dilution technique was first applied to the study of transcapillary exchange in the cerebrovascular bed by Crone (1965) and Yudilevich et al. (1972). If combined with measurements of flow and arterial concentration, quantitative values can be obtained (see below). It represents the obverse of the tissue sampling approach, the reference tracer being completely unextracted rather than freely diffusible. The variable intravenous infusion technique has been used extensively by Banos et al. (1975) and is particularly suited for detailed analysis of regional uptake. Studies on isolated brain capillaries (Joo, 1985) provide information in vitro, and allow the antiluminal side of the capillary to be examined. Interpretation difficulties arise concerning the structural and functional state of the capillaries following the isolation procedure, and the problem of distinguishing 'transport' from 'uptake'. The most important methodological considerations include the following: (i) A non-invasive method applicable to man is not available. Developments in NMR spectroscopy may provide a new approach. (ii) In general, influx rather than 'net uptake' is measured. (iii) Kinetic parameters obtained differ depending on whether transport of the amino acid is measured in isolation or in the presence of competing amino acids. A M I N O ACID TRANSPORT AT THE B L O O D - B R A I N BARRIER
Transport of amino acids in the mammalian cerebral nervous system has been extensively investigated (Yudilevich et al., 1988). The existence of separate transport systems at the blood-brain barrier and in brain cells is well established, and the anatomical site of blood-brain barrier transport is now recognized to be the plasma membrane of the brain capillary endothelial cells (Bradbury, 1979). Amino acid transport at the blood-brain barrier has been investigated using the methods described above in several animal species. Available evidence indicates the existence at the luminal interface of a stereospecific, saturable transport system mediating transport of large neutral amino acids. Fourteen neutral amino acids have been shown to demonstrate measurable affinity for this system (Smith et at., 1987). The order of affinity is similar to that of the L-system of Christensen (1973), but the measured Km values are both lower and extend over a wider range. Tryptophan is the only neutral amino acid showing significant binding to plasma J. lnher. Metab. Dis. t3 (1990)
630
Gardiner
proteins. The contribution of albumin-bound tryptophan to influx of this amino acid remains uncertain (Sarna et al., 1985). Transport of basic amino acids, arginine and lysine, has been documented in the rat and sheep but was not observed in the dog (Yudilevich et at., 1972). Transport of small neutral amino acids (glycine, serine, alanine) appears to be limited. Betz and Goldstein (1978) proposed from their observations on isolated brain capillaries that there may be a system-A carrier located at the abluminal side of the brain capillary. Transport of acidic amino acids, glutamate and aspartate, is not measurable by most methods although small, saturable extractions of these amino acids have been demonstrated by the Oldendorf technique. KINETIC PARAMETERS OF NEUTRAL AMINO ACID TRANSPORT I N C L U D I N G PHENYLALANINE Neutral amino acids are transported across the blood-brain barrier by a single facilitated system which is stereospecific and follows Michaelis-Menten saturation kinetics. Amino acid side chain hydrophobicity appears to be a critical determinant of transport affinity. The order of affinity is similar to that of the L-system of Christensen (1973), but the K m values are mostly between t-10% of values found in other tissues. At normal plasma concentrations the transporter is therefore nearly saturated and uniquely susceptible to competitive inhibition. Recent measurements using the in situ brain perfusion technique indicate that the true K m values are even lower than previous measurements suggested (Smith et aL, 1987). Measurements made in the presence of competing neutral amino acids (Km apparent) are elevated by competition effects. For example, the K m (app) for phenylalanine during plasma perfusion is approximately 20 times greater than the Km during saline perfusion. DEVELOPMENTAL CHANGES Available evidence suggests that blood-brain barrier transport systems are functional in the newborn of several species, including the rabbit (Braun et al., 1980), rat (Cremer et at., 1976; Sershen and Lajtha, 1976) and mouse (Seta et aL, t972). Fetal measurements are not available. Brain uptake indices have been measured for various amino acids during postnatal development in the rat: no difference was found for lysine, valine or glycine between 19-23-day-old and adult rats (Cremer et al., 1976). Kinetics of L-phenytalanine transport at the blood-brain barrier have been determined in the newborn rabbit (Pardridge and Mietus, 1982). Jmax and Km values were both higher than the corresponding values in the adult rat. TRANSPORT OF L-PHENYLALANINE AND RELATED AMINO ACIDS AT THE OVINE BLOOD-BRAIN BARRIER The aim of these experiments was to determine the kinetic parameters of bloodbrain transport of L-phenylalanine in the lamb and sheep in order to identify any postnatal developmental changes occurring in this species, and to quantify crossJ. lnher. Metab. Dis. 13 (1990)
631
Amino Acid Transport into Brain
inhibition between L-phenylalanine and several other amino acids (Brenton and Gardiner, 1988). Observations were made on 30 lambs and five adult sheep under sodium pentobarbitone anaesthesia. Extraction of amino acids was determined using a singlepass venous sampling technique. Venous sampling was from the sagittal sinus. Simultaneous measurements of cerebral blood flow were made using hydrogen clearance and influx calculated from the equation Jin = cerebral blood flow x fractional extraction x arterial blood concentration Kinetic parameters of transport were determined from measurements of influx made over a range of arterial blood concentrations. In the lamb, influx of both L-phenylalanine (14+--lnmolg-lmin -1) and L-alanine (12 + 2nmolg -1 rain-l) was greater than in the sheep (L-phenylalanine influx 9 + 1 nmol g- 1min- 1; L-alanine influx, 5 +_ 1 nmol g- 1min - 1 p .~ 0.01). This difference reflected higher blood concentrations of these amino acids in the younger animal. Concentration dependence of t-phenylalanine influx was best described by a model with a saturable and non-saturable component. Maximum influx (Jmax) was higher and the apparent affinity constant (Km app) lower in the lamb. Values obtained (mean +- SEM) were: lamb, Jmax 138_+6nmolg-lmin-1; Km app 0.85_+ 0.10nmolL-1; sheep, Jmax 107 ___7 n m o l g - l m i n - 1 ; K m app, 2.25 +- 0.25nmolL -1 L-Phenylalanine inhibited influx of L-leucine, L-tyrosine, L-valine and L-glutamine, but not L-arginine and L-lysine. In the lamb, L-phenylalanine inhibited L-histidine influx with an apparent inhibitor constant (K,) of 139 #mol L-1, and maximum inhibition of 92%. In the sheep, L-phenylalanine inhibited L-methionine influx with an apparent K n of 33 #mol L- 1 and a maximum inhibitor of 82°/;. Evidence was therefore obtained for developmental changes in the kinetic parameters of L-phenylalanine transport at the ovine blood-brain barrier, and significant inhibition of influx of related neutral amino acids by hyperphenylalaninaemia. DISCUSSION Implications for treatment of maternal PKU Although definitive evidence is lacking, current knowledge of transport of phenylalanine and related neutral amino acids at the blood-brain barrier suggests a plausible mechanism by which pathogenetic effects may arise from specific features of this transport system. In those species studied, it is known that several neutral amino acids share a single carrier and that their affinity constants for this system are within the range of plasma concentrations. The transporter is therefore nearly saturated with neutral amino acids at normal plasma concentrations, and an increase in concentration of one amino acid, such as phenylalanine, will inhibit influx of the others. If net uptake is thereby reduced, cerebral metabolic pathways may be deranged by impairment of substrate supply.
J. Inher. Metab. Dis.
13 (1990)
632
Gardiner
Does such a mechanism represent one way in which the developing CNS is damaged in the fetus of a mother with PKU? It is certainly possible, but the information necessary to examine the question rigorously is not available. The plasma neutral amino acid concentrations in the unstressed human fetus at 18 weeks gestation are very similar to those found in the adult rat. The changes which occur in response to maternal hyperphenylalaninaemia during the first trimester are of critical importance, but uncertain. It is possible that inhibition effects at the placental interface generate further distortions in the fetal plasma amino acid profile which could exacerbate inhibition by further reducing the concentration of vulnerable amino acids. Obviously there is no information concerning the kinetic parameters of neutral amino acid transport at the blood-brain barrier in the developing fetus, nor can there be certainty concerning the blood-brain barrier function in the human fetus during those early stages of gestation when maternal hyperphenylalaninaemia is known to be most deleterious. If influx of neutral amino acids into the brain is indeed inhibited, this may not be sufficient to reduce net uptake. A reduction in net uptake will not necessarily impair metabolic pathways such as those involved in protei n synthesis, although this is clearly more likely in the rapidly growing brain. If transport inhibition is an important component of the pathophysiology of the microcephaly found in infants of mothers with PKU, the described correlation with the degree of maternal hyperphenylalaninaemia could be anticipated (Levyjyet al., 1983; Drogari et al., 1987). Our present knowledge does not allow specific recommendations to be made concerning the management of this condition. Strenuous efforts to ensure a normal maternal amino acid profile from before conception would certainly minimize any deleterious effects mediated by inhibition of amino acid influx into the brain. In addition, if hyperphenylalaninaemia of some degree is present, attempts to ensure normal or raised concentrations of neutral amino acids known to share the same carrier (in other mammalian species) should tend to maintain a normal spectrum of amino acid influx into the brain.
REFERENCES
Banos, G., Daniel, P. M., Moorhouse, S. R. and Pratt, O. E. The requirements of the brain for some amino acids. J. Physiol. 246 (1975) 539-548 Betz, A. L. and Goldstein, G. W. Polarity of the blood-brain barrier: neutral amino acid transport into isolated brain capillaries. Science 202 (1978) 225-227 Bradbury, M. W. B. The Concept of a Blood-Brain Barrier, John Wiley, Chichester and New York, 1979 Braun, L. D., Cornford, E. M. and Oldendorf, W.H. Newborn rabbit blood-brain barrier is selectively permeable and differs substantially from the adult. J. Neurochem. 34 (1980) 147-152 Brenton, D. P. and Gardiner, R. M. Transport of L-phenylalanine and related amino acids at the ovine blood-brain barrier. J. Physiol. 402 (1988) 497-514 Christensen, H. N. On the development of amino acid transport systems. Fed. Proc. 32 (1973) 19-28 Cremer, J. E., Braun, L. D. and Oldendorf, W. H. Changes during development in transport processes of the blood-brain barrier. Biochim. Biophys. Acta 448 (1976) 633-637 J. Inher. Metab. Dis. 13 (1990)
Amino Acid Transport into Brain
633
Crone, C. Facilitated transfer of glucose from blood into brain tissue. J. Physiol. 181 (1965) 103-113 Drogani, E., Bearley, M., Smith, I. and Lloyd, J. K. Timing of strict diet in relation to fetal damage in maternal phenylketonuria. Lancet 2 (1987) 927-930 Joo, F. The blood-brain barrier in vitro: ten years of research on microvessels isolated from the brain. Neurochem. Int. 7(1) (1985) 1-25 Levy, H. L. and Waisbren, S. E. Effects of untreated maternal phenylketonuria and hyperphenylalaninaemia on the fetus. N. Engl. J. Med. 309 (1983) 1269-t274 Momma, S., Aoyagi, M., Rapoport, S. I. and Smith, Q. R. Phenylalanine transport across the blood-brain barrier as studied with the in situ brain perfusion technique, J. Neurochem. 48 (1987) 1291-1300 Oldendorf, W. H. Brain uptake of radiolabelled amino acid, amines and hexoses after arterial injection. Am. J. Physiol. 221 (1971) 1629-1639 Pardridge, W. M. and Mietus, L. J. Kinetics of neutral amino acid transport through the blood-brain barrier of the newborn rabbit. J. Neurochem. 38 (1982) 955-962 Pratt, O. E. The transport of metabolisable substances into the living brain. Adv. Exp. Med. Biol. 69 (1976) 55-75 Pratt, O. E. Transport inhibition in the pathology of phenylketonuria and other inherited metabolic diseases. J. Inher. Metab. Dis. 5 Suppl. 2 (1982) 75-81 Sarna, G. S., Kantamareni, B. D. and Curzon, G. Variables influencing the effect of a meal on brain tryptophan. J. Neuroehem. 44 (1985) 1575-1580 Sershen, H. and Lajtha, A. Capillary transport of amino acids in the developing brain. Exp. Neurol. 53 (1976) 465-474 Seta, K., Sershen, H. and Lajtha, A. Cerebral amino acid uptake in. vivo in newborn mice. Brain Res. 47 (1972) 4t5-425 Smith, Q. R., Momma, S., Aoyagi, M. and Rapoport, S. I. Kinetics of neutral amino acid transport across the blood-brain barrier. J. Neurochem. 49 (1987) 1651 1658 Takasato, Y., Rapoport, S. I. and Smith, Q. R. An in situ brain perfusion technique to study cerebrovascular transport in rat. Am. J. Physiol. 247 (1984) H484-H493 Yudilevich, D. L., De Rose, N. and Sepulveda, F. V. Facilitated transport of amino acids through the blood brain barrier of the dog. Studies in a single capillary circulation. Brain Res. 44 (1972) 569-578 Yudilevich, D. L., Wheeler, C. P. D. and Bustamante, J. C. A comparative view of amino acid transport across the blood-brain barrier (endothelium) and the placenta (trophoblast). In: Proceedings of Symposium on Peptide and Amino Acid Transport Mechanisms in the Central Nervous System, (In press)
J. Inher. Metab. Dis. 13 (1990)
Maternal PKU Workshop
Transport of Amino Acids across the BloodBrain Barrier: Implications for Treatment of Maternal Phenylketonuria R. M. GARD1NER
Departments of Paediatrics and Human Metabolism, University College London, The Rayne Institute, University Street, London WCIE 6AU, UK; Present address: Department of Paediatrics, John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK Summary: Amino acid transport at the mammalian blood-brain barrier has been extensively characterized. Transport of L-phenylalanine and related neutral amino acids is known to be mediated by a stereospecific, sodium independent, saturable carrier. The affinity of this carrier is much higher than that of comparable systems in other tissues. This feature renders it susceptible to inhibition. It has been suggested that inhibition of neutral amino acid influx into the brain by hyperphenylalaninaemia contributes to the pathophysiology of brain damage in this condition. Methods for investigation of amino acid transport at the blood-brain barrier are discussed, a n d current knowledge of blood-brain barrier amino acid transport at the blood-brain barrier is reviewed. Developmental changes are delineated, with particular reference to recent work on the ovine blood-brain barrier. There is insufficient information concerning blood-brain barrier transport of amino acids in the fetal brain to allow firm conclusions to be drawn concerning implications for treatment of maternal PKU. Reasonable extrapolation from animal data suggests that transport inhibition may contribute to impaired fetal brain growth in maternal PKU, and can be minimized by attempts to maintain a normal milieu from the time of conception.
The precise mechanisms by which raised circulating phenylalanine levels exert a deleterious effect on the developing brain, either in utero or postnatally, remain uncertain. It has been suggested that the special properties of the transport mechanisms which mediate transfer of amino acids from the circulation to their site of metabolism within neurones may be relevant to the pathophysiology of cerebral damage in hyperphenylalaninaemia. In particular, it has been proposed that inhibition of blood627
628
Gardiner
brain barrier transport of related neutral amino acids by a high concentration of phenylalanine may be an important pathogenetic mechanism (Pratt, 1982). Information necessary to evaluate fully the r6te of amino acid transport at the blood-brain barrier in the pathophysiology of hyperphenylalaninaemia in utero remains incomplete. Amino acid transport at the blood-brain barrier has been extensively characterized in a variety of animat species, but there is limited information on changes occurring during development. However, a non-invasive approach to quantifying blood brain barrier transport in human patients, let alone the fetus, is not available, and interspecies extrapolation in this area is notoriously difficult. This paper reviews the methods available for the study of blood-brain barrier amino acid transport, and summarizes current knowledge of these transport systems in the mammalian brain. Developmental changes are considered, with particular reference to changes at the ovine blood-brain barrier. The possible r61e in pathophysiology of transport at this interface is discussed, with special emphasis on maternal PKU, and gaps in our present knowledge are identified. TRANSPORT AT THE BLOOD-BRAIN B A R R I E R It is now generally accepted that the anatomical basis of the blood-brain barrier lies in the characteristic structure of cerebral capillaries (Bradbury, t979). In particular, the tight-junctions between capillary endothelial cells create a vascular barrier impermeable to water soluble substrates (such as amino acids) in the absence of a specific transport mechanism. The plasma membrane of brain capillary endothelial cells is the site of several specific transport mechanisms including those for glucose, monocarboxylic acids, and amino acids. There is evidence for polarity: different systems may be present at the luminal and abluminal interfaces. Maturation of barrier function has been studied in several species (Bradbury, 1979). The old idea that the blood-brain barrier in the 'newborn' is 'open', is now known to be false and represents a gross oversimplification of the complex anatomical and functional changes which have been documented in a variety of species at various periods of gestation. M E T H O D S FOR INVESTIGATING AMINO ACID TRANSPORT AT THE BLOOD-BRAIN B A R R I E R During the last two decades both in vivo and in vitro techniques have been developed, but none are applicable in man. Quantitative information is obtained, including values for the kinetic parameters of transport, but the particular method employed has an important bearing on the interpretation of the results. Measurement of net uptake into the brain is extremely difficult as A-V concentration differences are small. In vivo methods:
Single-pass Single-pass Variable-infusion I n situ brain perfusion J. Inher, Metab. Dis, 13 (1990)
tissue sampling (Oldendorf, 1971) - indicator-dilution (Yudilevich et al., 1972) tissue sampling (Pratt, 1976) tissue sampling (Takasato et al., 1984)
629
Amino Acid Transport into Brain In vitro methods:
Observations on isolated brain capillaries (Joo, 1985). Sampling of brain tissue (following decapitation) after intracarotid injection of the test amino acid and a highly diffusible reference tracer was originally employed by Oldendorf (1971). Results are expressed as a brain uptake index. Although kinetic parameters may be estimated, quantitation of influx and permeability is not possible, and the degree of mixing of bolus and circulating blood is poorly defined. A recent modification of this approach, the in situ brain perfusion technique (Takasato et al., 1984) has several advantages. In particular, permeability can be estimated and the composition of the perfusion fluid is not changed appreciably by mixing with blood (Momma et al., 1987). The single-pass indicator dilution technique was first applied to the study of transcapillary exchange in the cerebrovascular bed by Crone (1965) and Yudilevich et al. (1972). If combined with measurements of flow and arterial concentration, quantitative values can be obtained (see below). It represents the obverse of the tissue sampling approach, the reference tracer being completely unextracted rather than freely diffusible. The variable intravenous infusion technique has been used extensively by Banos et al. (1975) and is particularly suited for detailed analysis of regional uptake. Studies on isolated brain capillaries (Joo, 1985) provide information in vitro, and allow the antiluminal side of the capillary to be examined. Interpretation difficulties arise concerning the structural and functional state of the capillaries following the isolation procedure, and the problem of distinguishing 'transport' from 'uptake'. The most important methodological considerations include the following: (i) A non-invasive method applicable to man is not available. Developments in NMR spectroscopy may provide a new approach. (ii) In general, influx rather than 'net uptake' is measured. (iii) Kinetic parameters obtained differ depending on whether transport of the amino acid is measured in isolation or in the presence of competing amino acids. A M I N O ACID TRANSPORT AT THE B L O O D - B R A I N BARRIER
Transport of amino acids in the mammalian cerebral nervous system has been extensively investigated (Yudilevich et al., 1988). The existence of separate transport systems at the blood-brain barrier and in brain cells is well established, and the anatomical site of blood-brain barrier transport is now recognized to be the plasma membrane of the brain capillary endothelial cells (Bradbury, 1979). Amino acid transport at the blood-brain barrier has been investigated using the methods described above in several animal species. Available evidence indicates the existence at the luminal interface of a stereospecific, saturable transport system mediating transport of large neutral amino acids. Fourteen neutral amino acids have been shown to demonstrate measurable affinity for this system (Smith et at., 1987). The order of affinity is similar to that of the L-system of Christensen (1973), but the measured Km values are both lower and extend over a wider range. Tryptophan is the only neutral amino acid showing significant binding to plasma J. lnher. Metab. Dis. t3 (1990)
630
Gardiner
proteins. The contribution of albumin-bound tryptophan to influx of this amino acid remains uncertain (Sarna et al., 1985). Transport of basic amino acids, arginine and lysine, has been documented in the rat and sheep but was not observed in the dog (Yudilevich et at., 1972). Transport of small neutral amino acids (glycine, serine, alanine) appears to be limited. Betz and Goldstein (1978) proposed from their observations on isolated brain capillaries that there may be a system-A carrier located at the abluminal side of the brain capillary. Transport of acidic amino acids, glutamate and aspartate, is not measurable by most methods although small, saturable extractions of these amino acids have been demonstrated by the Oldendorf technique. KINETIC PARAMETERS OF NEUTRAL AMINO ACID TRANSPORT I N C L U D I N G PHENYLALANINE Neutral amino acids are transported across the blood-brain barrier by a single facilitated system which is stereospecific and follows Michaelis-Menten saturation kinetics. Amino acid side chain hydrophobicity appears to be a critical determinant of transport affinity. The order of affinity is similar to that of the L-system of Christensen (1973), but the K m values are mostly between t-10% of values found in other tissues. At normal plasma concentrations the transporter is therefore nearly saturated and uniquely susceptible to competitive inhibition. Recent measurements using the in situ brain perfusion technique indicate that the true K m values are even lower than previous measurements suggested (Smith et aL, 1987). Measurements made in the presence of competing neutral amino acids (Km apparent) are elevated by competition effects. For example, the K m (app) for phenylalanine during plasma perfusion is approximately 20 times greater than the Km during saline perfusion. DEVELOPMENTAL CHANGES Available evidence suggests that blood-brain barrier transport systems are functional in the newborn of several species, including the rabbit (Braun et al., 1980), rat (Cremer et at., 1976; Sershen and Lajtha, 1976) and mouse (Seta et aL, t972). Fetal measurements are not available. Brain uptake indices have been measured for various amino acids during postnatal development in the rat: no difference was found for lysine, valine or glycine between 19-23-day-old and adult rats (Cremer et al., 1976). Kinetics of L-phenytalanine transport at the blood-brain barrier have been determined in the newborn rabbit (Pardridge and Mietus, 1982). Jmax and Km values were both higher than the corresponding values in the adult rat. TRANSPORT OF L-PHENYLALANINE AND RELATED AMINO ACIDS AT THE OVINE BLOOD-BRAIN BARRIER The aim of these experiments was to determine the kinetic parameters of bloodbrain transport of L-phenylalanine in the lamb and sheep in order to identify any postnatal developmental changes occurring in this species, and to quantify crossJ. lnher. Metab. Dis. 13 (1990)
631
Amino Acid Transport into Brain
inhibition between L-phenylalanine and several other amino acids (Brenton and Gardiner, 1988). Observations were made on 30 lambs and five adult sheep under sodium pentobarbitone anaesthesia. Extraction of amino acids was determined using a singlepass venous sampling technique. Venous sampling was from the sagittal sinus. Simultaneous measurements of cerebral blood flow were made using hydrogen clearance and influx calculated from the equation Jin = cerebral blood flow x fractional extraction x arterial blood concentration Kinetic parameters of transport were determined from measurements of influx made over a range of arterial blood concentrations. In the lamb, influx of both L-phenylalanine (14+--lnmolg-lmin -1) and L-alanine (12 + 2nmolg -1 rain-l) was greater than in the sheep (L-phenylalanine influx 9 + 1 nmol g- 1min- 1; L-alanine influx, 5 +_ 1 nmol g- 1min - 1 p .~ 0.01). This difference reflected higher blood concentrations of these amino acids in the younger animal. Concentration dependence of t-phenylalanine influx was best described by a model with a saturable and non-saturable component. Maximum influx (Jmax) was higher and the apparent affinity constant (Km app) lower in the lamb. Values obtained (mean +- SEM) were: lamb, Jmax 138_+6nmolg-lmin-1; Km app 0.85_+ 0.10nmolL-1; sheep, Jmax 107 ___7 n m o l g - l m i n - 1 ; K m app, 2.25 +- 0.25nmolL -1 L-Phenylalanine inhibited influx of L-leucine, L-tyrosine, L-valine and L-glutamine, but not L-arginine and L-lysine. In the lamb, L-phenylalanine inhibited L-histidine influx with an apparent inhibitor constant (K,) of 139 #mol L-1, and maximum inhibition of 92%. In the sheep, L-phenylalanine inhibited L-methionine influx with an apparent K n of 33 #mol L- 1 and a maximum inhibitor of 82°/;. Evidence was therefore obtained for developmental changes in the kinetic parameters of L-phenylalanine transport at the ovine blood-brain barrier, and significant inhibition of influx of related neutral amino acids by hyperphenylalaninaemia. DISCUSSION Implications for treatment of maternal PKU Although definitive evidence is lacking, current knowledge of transport of phenylalanine and related neutral amino acids at the blood-brain barrier suggests a plausible mechanism by which pathogenetic effects may arise from specific features of this transport system. In those species studied, it is known that several neutral amino acids share a single carrier and that their affinity constants for this system are within the range of plasma concentrations. The transporter is therefore nearly saturated with neutral amino acids at normal plasma concentrations, and an increase in concentration of one amino acid, such as phenylalanine, will inhibit influx of the others. If net uptake is thereby reduced, cerebral metabolic pathways may be deranged by impairment of substrate supply.
J. Inher. Metab. Dis.
13 (1990)
632
Gardiner
Does such a mechanism represent one way in which the developing CNS is damaged in the fetus of a mother with PKU? It is certainly possible, but the information necessary to examine the question rigorously is not available. The plasma neutral amino acid concentrations in the unstressed human fetus at 18 weeks gestation are very similar to those found in the adult rat. The changes which occur in response to maternal hyperphenylalaninaemia during the first trimester are of critical importance, but uncertain. It is possible that inhibition effects at the placental interface generate further distortions in the fetal plasma amino acid profile which could exacerbate inhibition by further reducing the concentration of vulnerable amino acids. Obviously there is no information concerning the kinetic parameters of neutral amino acid transport at the blood-brain barrier in the developing fetus, nor can there be certainty concerning the blood-brain barrier function in the human fetus during those early stages of gestation when maternal hyperphenylalaninaemia is known to be most deleterious. If influx of neutral amino acids into the brain is indeed inhibited, this may not be sufficient to reduce net uptake. A reduction in net uptake will not necessarily impair metabolic pathways such as those involved in protei n synthesis, although this is clearly more likely in the rapidly growing brain. If transport inhibition is an important component of the pathophysiology of the microcephaly found in infants of mothers with PKU, the described correlation with the degree of maternal hyperphenylalaninaemia could be anticipated (Levyjyet al., 1983; Drogari et al., 1987). Our present knowledge does not allow specific recommendations to be made concerning the management of this condition. Strenuous efforts to ensure a normal maternal amino acid profile from before conception would certainly minimize any deleterious effects mediated by inhibition of amino acid influx into the brain. In addition, if hyperphenylalaninaemia of some degree is present, attempts to ensure normal or raised concentrations of neutral amino acids known to share the same carrier (in other mammalian species) should tend to maintain a normal spectrum of amino acid influx into the brain.
REFERENCES
Banos, G., Daniel, P. M., Moorhouse, S. R. and Pratt, O. E. The requirements of the brain for some amino acids. J. Physiol. 246 (1975) 539-548 Betz, A. L. and Goldstein, G. W. Polarity of the blood-brain barrier: neutral amino acid transport into isolated brain capillaries. Science 202 (1978) 225-227 Bradbury, M. W. B. The Concept of a Blood-Brain Barrier, John Wiley, Chichester and New York, 1979 Braun, L. D., Cornford, E. M. and Oldendorf, W.H. Newborn rabbit blood-brain barrier is selectively permeable and differs substantially from the adult. J. Neurochem. 34 (1980) 147-152 Brenton, D. P. and Gardiner, R. M. Transport of L-phenylalanine and related amino acids at the ovine blood-brain barrier. J. Physiol. 402 (1988) 497-514 Christensen, H. N. On the development of amino acid transport systems. Fed. Proc. 32 (1973) 19-28 Cremer, J. E., Braun, L. D. and Oldendorf, W. H. Changes during development in transport processes of the blood-brain barrier. Biochim. Biophys. Acta 448 (1976) 633-637 J. Inher. Metab. Dis. 13 (1990)
Amino Acid Transport into Brain
633
Crone, C. Facilitated transfer of glucose from blood into brain tissue. J. Physiol. 181 (1965) 103-113 Drogani, E., Bearley, M., Smith, I. and Lloyd, J. K. Timing of strict diet in relation to fetal damage in maternal phenylketonuria. Lancet 2 (1987) 927-930 Joo, F. The blood-brain barrier in vitro: ten years of research on microvessels isolated from the brain. Neurochem. Int. 7(1) (1985) 1-25 Levy, H. L. and Waisbren, S. E. Effects of untreated maternal phenylketonuria and hyperphenylalaninaemia on the fetus. N. Engl. J. Med. 309 (1983) 1269-t274 Momma, S., Aoyagi, M., Rapoport, S. I. and Smith, Q. R. Phenylalanine transport across the blood-brain barrier as studied with the in situ brain perfusion technique, J. Neurochem. 48 (1987) 1291-1300 Oldendorf, W. H. Brain uptake of radiolabelled amino acid, amines and hexoses after arterial injection. Am. J. Physiol. 221 (1971) 1629-1639 Pardridge, W. M. and Mietus, L. J. Kinetics of neutral amino acid transport through the blood-brain barrier of the newborn rabbit. J. Neurochem. 38 (1982) 955-962 Pratt, O. E. The transport of metabolisable substances into the living brain. Adv. Exp. Med. Biol. 69 (1976) 55-75 Pratt, O. E. Transport inhibition in the pathology of phenylketonuria and other inherited metabolic diseases. J. Inher. Metab. Dis. 5 Suppl. 2 (1982) 75-81 Sarna, G. S., Kantamareni, B. D. and Curzon, G. Variables influencing the effect of a meal on brain tryptophan. J. Neuroehem. 44 (1985) 1575-1580 Sershen, H. and Lajtha, A. Capillary transport of amino acids in the developing brain. Exp. Neurol. 53 (1976) 465-474 Seta, K., Sershen, H. and Lajtha, A. Cerebral amino acid uptake in. vivo in newborn mice. Brain Res. 47 (1972) 4t5-425 Smith, Q. R., Momma, S., Aoyagi, M. and Rapoport, S. I. Kinetics of neutral amino acid transport across the blood-brain barrier. J. Neurochem. 49 (1987) 1651 1658 Takasato, Y., Rapoport, S. I. and Smith, Q. R. An in situ brain perfusion technique to study cerebrovascular transport in rat. Am. J. Physiol. 247 (1984) H484-H493 Yudilevich, D. L., De Rose, N. and Sepulveda, F. V. Facilitated transport of amino acids through the blood brain barrier of the dog. Studies in a single capillary circulation. Brain Res. 44 (1972) 569-578 Yudilevich, D. L., Wheeler, C. P. D. and Bustamante, J. C. A comparative view of amino acid transport across the blood-brain barrier (endothelium) and the placenta (trophoblast). In: Proceedings of Symposium on Peptide and Amino Acid Transport Mechanisms in the Central Nervous System, (In press)
J. Inher. Metab. Dis. 13 (1990)
