brief review
Glutamate transport and renal function
TOMAS C. WELBOURNE1 AND JAMES C. MATTHEWS2
of Molecular and Cellular Physiology, Louisiana State University Medical Center,
Shreveport, Louisiana 71130; and 2Department of Animal Sciences, University of Kentucky,
Lexington, Kentucky 40546
1Department
␥-glutamyltransferase; EAAC1; GLT1; D-glutamate; phosphate-dependent glutaminase; paracellular permeability; urinary acidification
THE ECTOENZYME ␥-glutamyltransferase, phosphateindependent glutaminase (PIG), and the apical membrane glutamate transporter form a unit that functions
in regulating both glutamine metabolism and paracellular permeability (Fig. 1, A and B). PIG generates
glutamate from extracellular glutamine (and glutathione when available), which is transported across the
apical border by the Na⫹-dependent, high-affinity
EAAC1 transporter subtype (18). Mechanistically, this
‘‘functional unit’’ produces three intracellular signals
that modulate cellular energy production and paracellular permeability. One signal is the intracellular glutamate concentration (Fig. 1A), set by the rate of extracellular glutamate formation and transport into the cell.
Because cellular glutamate competes with glutamine
(17) for the cytosolic reactive site of the mitochondrial
glutaminase (phosphate-dependent glutaminase, PDG)
(9), increasing or decreasing the functional unit activity
results in suppression (Fig. 1A, left) or acceleration
(Fig. 1A, right) of glutamine utilization by this ratelimiting reaction and, hence, ATP and base formation
by this pathway. The glutamate transporter turnover
(movement of 3 Na⫹ in and 1 K⫹ out; see the accompanying article by Fairman and Amara, Ref. 5) generates a
second signal in the form of a localized increase in ADP,
which stimulates glycolysis and ATP production; this
ATP is distributed to the plasma membrane Na⫹-K⫹ATPase and the F-actin, maintaining the junctions
tight (Fig. 1B, left). Reducing the glutamate transport
This article is the fifth of five in this forum, which is based on a
series of reports on glutamate transport and glutamate metabolism
that was first presented at Experimental Biology ’98 in San Francisco, CA.
therefore slows glycolysis resulting in dispersion of the
F-actin and an opening of the tight junction. Finally, a
third signal is the rise in intracellular H⫹ concentration
resulting from the transport of glutamate and acid into
the cell (see accompanying article by Hediger, Ref. 8).
As a result of cellular acidification, oxidative deamination of glutamate is accelerated producing NH4⫹ and
␣-ketoglutarate (see accompanying article by Nissim,
Ref. 15). Therefore, the consequence of high functional
unit activity is extracellularly generated glutamate
(PIG pathway) being substituted for intracellularly
formed glutamate (PDG pathway) as the major fuel,
sparing cellular glutamine for biosynthetic pathways
and, at the same time, maintaining tight cell-to-cell
junctions. Conversely, diminishing the functional unit
activity releases the powerful glutaminase, with the
intracellularly produced glutamate diverted to the default transamination pathway (alanine formation) as
the tight junctions open (Fig. 1B, middle cell). High
functional unit activity is characteristic of differentiated cell monolayers in contrast to a low activity in
their rapidly proliferative phase.
Evidence in support of the functional unit concept
was obtained in vitro, from cells grown as monolayers
on plastic dishes (Fig. 1A) (12), or porous supports (Fig.
1B) (22), as well as in vivo (1, 2). The in vitro studies
were performed primarily, but not exclusively, using
LLC-PK1-F⫹ cells, which were selected (7) from the
parental LLC-PK1 cell line for growth in the absence of
glucose, relying upon glutamine as their metabolic fuel.
Because growing these proximal tubule-like LLCPK1-F⫹ monolayers on plastic dishes is the simplest
approach for demonstrating the role of the functional
unit in glutamine metabolism, evidence for the associ-
0363-6127/99 $5.00 Copyright r 1999 the American Physiological Society
F501
Downloaded from http://ajprenal.physiology.org/ by 10.220.33.6 on June 18, 2017
Welbourne, Tomas C., and James C. Matthews. Glutamate transport and renal function. Am. J. Physiol. 277 (Renal Physiol. 46): F501–
F505, 1999.—Brush border ␥-glutamyltransferase-glutaminase activity
and the high-affinity glutamate transporter EAAC1 function as a unit in
generating and transporting extracellular glutamate into proximal tubules as a signal that modulates intracellular glutamine/glutamate
metabolism, paracellular permeability, and urinary acidification. The
reported presence of a second glutamate transporter, GLT1, on the
antiluminal tubule surface points to specific functional roles for each
subtype in physiological and pathophysiological processes.
F502
GLUTAMATE TRANSPORT AND METABOLISM
ated function of PIG and apical glutamate transport
using this experimental model will be summarized first.
In Vitro Evidence: LLC-PK1-F⫹ Monolayers
Grown on Plastic
The ontologic expression of the functional unit in
cultured LLC-PK1-F⫹ cells parallels PIG activity and
tight junction formation in the developing kidney (3, 6).
The PIG activity is initially low after splitting monolayers and then increases. However, PIG still generates
less glutamate than the transporter’s capacity in this
phase of rapid proliferation; with differentiation (5–10
days), rates of extracellular glutamate production and
uptake converge, marked by dome formation and
transepithelial fluid movement; finally (after 10 days),
the two activities diverge as PIG continues to rise as
transporter activity declines with aging (13). Cellular
glutamate concentration reflects the developmental
changes in the functional unit; cellular glutamate
concentration rises in parallel with the PIG activity up
through differentiation and then levels off at ⬃175
nmol/mg (35 mM) as transporter activity declines,
leading to an accumulation of extracellular glutamate
in the media. Inhibiting the PIG activity eliminates
this extracellular glutamate accumlation (13). The
inverse relationship between the two functional unit
components is consistent with the hypothesis that the
glutamate transporter is regulated by the cellular
glutamate content (see accompanying article by McGivan and Nicholson, Ref. 11) resulting in the autoregulation of the functional unit’s activity.
From this perspective, monolayers must be studied
at a time when both PIG and transporter activity are
high, to best demonstrate the role of the functional unit
in metabolic regulation. With monolayers grown 8–10
days, postsplitting and acutely inhibiting PIG with
acivicin (AT-125) or blocking glutamate uptake with
D-aspartate (Fig. 1A, right cell) results in a 15–20%
reduction in cellular glutamate concentration within 1
h (12). Associated with this was a 30–50% increase in
glutaminase flux with formed intracellular glutamate
coupled to alanine formation (acivicin) or ammoniagenesis (D-aspartate). These results are consistent with the
functional unit modulating glutaminase flux coupled to
either alanine aminotransferase or glutamate dehydrogenase (GDH) dependent upon the transporter turnover (acid loading).
In Vitro Evidence: LLC-PK-F⫹ Monolayers
Grown on Porous supports
These cells show a markedly different phenotype
from those grown on plastic, being densely packed
columnar cells expressing well-developed brush borders and tight junctions and having structural characteristics of the in situ proximal tubule (21). Also, unlike
the cells grown on plastic, cells grown on porous
supports express a second glutamate transporter on
the basal cell surface, which functions in the regulation
of the cellular glutamate concentration (Fig. 1B). Net
glutamate efflux across this cell surface diminishes the
intracellular glutamate concentration to a level ⬃40%
below that for cells grown on plastic (20 mM) despite a
comparable apical glutamate uptake. In fact, inhibiting
the apical glutamate formation with acivicin and lowering the intracellular glutamate also eliminates basal
Downloaded from http://ajprenal.physiology.org/ by 10.220.33.6 on June 18, 2017
Fig. 1. ␥-Glutamyltransferase (GGt) and EAAC1 as a
‘‘functional unit’’ in LLC-PK1-F⫹ cell monolayers grown
on plastic dishes (A) or on porous supports (B). The
functional unit maintains high intracellular glutamate
concentration (GLU), which inhibits glutaminase (GA)
and glutamine (GLN) utilization. Reduced functional
unit activity as result of GGt inhibition (acivicin) or
blocking GLU uptake with D-aspartate (D-ASP) (A) or
limiting apical GLN (B) accelerates GA flux (A) and
increases paracellular permeability. ALA, alanine; PIG,
phosphate-independent glutaminase; PDG, phosphatedependent glutaminase.
GLUTAMATE TRANSPORT AND METABOLISM
F503
efflux associated with a 40% decrease in cellular glutamate. Furthermore, basal efflux can be regulated as
shown by the action of acidogenic hormones, i.e., growth
hormone, which primarily accelerate basal glutamate
efflux, lowering cellular glutamate (also contributed to
by an increased metabolic removal via GDH; see accompanying article by Nissim, Ref. 15) and accelerating
glutaminase flux (21). Although more than 80% of the
cell’s glutamate uptake capacity is present on the
apical surface, the reverse is true for the glutamine
uptake capacity. Consequently, the cellular glutamineto-glutamate concentration ratio is set higher in these
Fig. 2. Inhibition of GGt with acivicin (AT-125) increases urinary
glutamine (GLN), glutathione (GSH), aspartate (ASP), and alanine
(ALA) concentration. Rats were given intravenously either AT-125,
30 mg/100 g body wt, or vehicle (saline), and urine was collected
after 1 h. A is plasma profile, whereas B and C are control and
acivicin-injected urines, respectively.
monolayers (Fig. 1B) than those grown on plastic
(Fig. 1A), as is the dependent glutaminase flux. Nevertheless, when either PIG or glutamate uptake were
eliminated with acivicin or D-aspartate, cellular glutamate concentration fell 40–50% with a two- to threefold
acceleration of the glutaminase flux (22) coupled to
either deamination or transamination, depending upon
whether the transporter turnover was maintained (Daspartate) or not (acivicin). Thus the role of the functional unit in regulating cellular glutamine metabolism
can also be demonstrated in monolayers grown such
that both surfaces are exposed.
An additional advantage of studying monolayers
grown on porous supports is the opportunity to observe
changes in paracellular permeability after eliminating
the functional unit (Fig. 1, middle cell). With twoindependent measures of paracellular permeability, diffusion of
radiolabeled L-[14C]glucose and transepithelial electrical resistance (TER) and acivicin or DL-threo-3-hydroxyaspartate (THA) to reduce apical glutamate uptake,
monolayer ‘‘tightness’’ was shown to be modulated by
the functional unit activity (20). After 2 h, THA (0.5 mM)
resulted in a 2.6-fold increase in the L-glucose diffusion
rate from the outer to inner well and a 50% drop in
TER. Subsequent studies demonstrated that transporter turnover rather than the cellular glutamate concentration was the effective signal (19). Consequently, acute
reduction of the functional unit activity should open up
the paracellular pathway. At the same time, energy
formed from glutamine oxidation for transport would
be enhanced, enabling the cells to withstand a less
efficient transepithelial transport while generating base.
Downloaded from http://ajprenal.physiology.org/ by 10.220.33.6 on June 18, 2017
Fig. 3. Evidence for the 2 glutamate transporter subtypes EAAC1
and GLT1 in rat kidney cortex and the following 3 renal cell lines:
MDCK (canine kidney, duplicate lanes), LLC-PK1-F⫹ (porcine kidney,
GLN dependent), and LLC-PK1 (porcine kidney, glucose dependent).
Total RNA, ⬇20 µg/lane, was fractionated on a 1% agarose gel,
electroblot transferred to a Zeta-Probe GT membrane, covalently
cross-linked by ultraviolet light, and hybridized overnight with
[32P]dCTP-radiolabeled probes constructed using full-length EAAC1
and GLT1 rat cDNAs; dominant bands present at ⬇4 and 11 kb can be
observed for EAAC1 (left) and GLT1 (right).
F504
GLUTAMATE TRANSPORT AND METABOLISM
In Vivo: Role of the Functional Unit
in the Functioning Rat Kidney
Address for reprint requests and other correspondence: T. C.
Welbourne, Dept. of Molecular and Cellular Physiology LSUMC,
Shreveport, LA 71130 (E-mail: twelbo@lsunc. edu).
REFERENCES
1. Carter, P., and T. C. Welbourne. Glutamate transport regulation of renal glutaminase flux in vivo. Am. J. Physiol. 273
(Endocrinol. Metab. 36): E521–E527, 1997.
2. Carter, P., and T. C. Welbourne. Glutamate transport asymmetry in renal glutamine metabolism. Am. J. Physiol. 273 (Endocrinol. Metab. 36): E521–E527, 1997.
3. Curto, K. A., W. E. Sweeney, E. D. Avner, N. P. Piesco, and
N. P. Curthoys. Immunocytochemical localization of ␥-glutamyltranspeptidase during fetal development of mouse kidney. J.
Histochem. Cytochem. 36: 159–166, 1988.
4. Endou, H, H. Nonoguchi, J. Takehara, H. Yamada, and J.
Nakada. Intranephron heterogeneity of ammoniagenesis and
gluconeogenesis in rats. Contrib. Nephrol. 47: 98–104, 1985.
5. Fairman, W. A., and S. G. Amara. Functional diversity of
excitatory amino acid transporters: ion channel and transport
Downloaded from http://ajprenal.physiology.org/ by 10.220.33.6 on June 18, 2017
The functional unit as a regulatory mechanism elucidated in vitro would carry more relevance if it could be
demonstrated to function in vivo. In this regard, both
PIG and glutamate transporter EAAC1 are codistributed along the apical brush border, increasing in density from early to the late proximal tubule (4, 18) and
indicating that both components are present at the
appropriate sites. The intraluminal generation of glutamate from secreted glutamine along the proximal tubule would depend upon the prevailing plasma glutamine concentration and PIG activity. Importantly,
metabolic acidosis lowers the plasma glutamine by
20–40%. Hence, substrate is reduced along with the
acutely reduced PIG activity (14), curtailing the extracellular glutamate generation; consequently, cellular
glutamate would subsequently decrease accelerating
the intracellular glutaminase flux. Conversely, restoring the plasma glutamine concentration should return
the cellular glutamate and suppress the glutaminase
flux. To test this hypothesis, rats were maintained on
an elemental diet devoid of glutamine or fed an elemental diet supplemented with glutamine (23). After 3
days, their arterial plasma glutamine concentration,
renal cortical glutamate content, and glutamine utilization were measured. The glutamine-fed animals had an
18% rise in arterial plasma glutamine concentration, a
proportional rise in cellular glutamate, and a marked
reduction in glutamine utilization, consistent with
suppression of the intracellular glutaminase flux. To
confirm the role of the functional unit in this response,
acivicin was administered intravenously to ‘‘knockout’’
PIG activity, and the above parameters were reevaluated. Despite a slightly elevated arterial glutamine
concentration, renal cortical glutamate dropped 40%
with a twofold increase in renal glutamine utilization.
These in vivo results are consistent with the hypothesis
that the functional unit generates a glutamate flux that
acts as a governor on intracellular glutaminase flux.
Evidence that the functional unit also modulates paracellular permeability may be seen in the urinary amino
acid profile shown in Fig. 2. Control (Fig. 2B) and acivicintreated animals (Fig. 2C) had similar urine flow rates, yet
the concentrations of glutathione, aspartate, glutamine, and alanine, but not glutamate, were higher with
PIG inhibited. The rise in glutathione excretion (Fig.
2C) was not surprising, given that it is freely filtered
with subsequent absorption dependent on intraluminal
breakdown initiated by the PIG. However, the rise in
glutamine (and alanine) excretion is surprising, because proximal transport removes virtually all of the
filtered glutamine, consistent with movement (secretion) from plasma to lumen via the paracellular pathway. Of course, this leak would normally be selfcorrecting as intraluminal glutamine concentration
rises, driving the glutamate flux of the functional unit
(Fig. 1B, right). Evidence consistent with the generation of intraluminal glutamate from secreted glutamine may also be seen in EAAC1 knockout mice (16), in
which urinary glutamate excretion was increased 1,400fold above control compared with only 10-fold for aspartate;
this 140-fold greater increase in glutamate is consistent
with leaky tight junctions and backflux of plasma
glutamine resulting in increased PIG hydrolysis.
One can also disrupt the functional unit glutamate
signal by displacing L-glutamate from the apical surface transporter using D-aspartate while retaining the
proton signal, since D-aspartate is effectively reabsorbed (1). As a consequence of infusing D-aspartate,
renal cortical glutamate content decreased and was
associated with a two- to threefold increase in glutamine utilization and ammonium formation (1). Because
the rat proximal convoluted tubule lacks the transamination pathway, the ratio of NH4⫹ produced to glutamine utilized approximated a value of 2, which is
consistent with the activation of both the glutaminase
and GDH fluxes as the consequence of the reduced
cellular glutamate and pH, respectively, similar to the
in vitro responses (12, 22). On the other hand, the D-isomer
of glutamate does not inhibit apical L-glutamate transport, but, rather, reduces the blood side L-glutamate
uptake and activates glutamine utilization and ammonium formation with the ratio approaching a value of 2
(2). These findings point to asymmetrically distributed
glutamate transporter subtypes: EAAC1 on the apical
border and a second transporter subtype (GLT1?) on the
basal border. To test for the expression of another glutamate transporter in rat kidney, total RNA was isolated
from rat renal cortex and LLC-PK1-F⫹, MDCK, and
LLC-PK1 cells and probed for the presence of EAAC1
and GLT1 mRNA (Fig. 3). Dominant bands at around 4
kb for EAAC1 and 11 kb for GLT1 could be detected in
both the rat renal cortex and and the cultured kidney
cell lines. An example of a separate subtype of glutamate transporter that is localized to the basal membrane surface and that may express novel stereospecificity has been demonstrated in rat placenta (10). This
emphasizes the importance of studying glutamate transport regulation of cellular processes in a physiological
context. Further studies are obviously necessary to
characterize this curious transporter asymmetry and
the regulatory role played in glutamine and glutamate
metabolism as well as in transepithelial transport.
GLUTAMATE TRANSPORT AND METABOLISM
6.
7.
8.
9.
10.
12.
13.
14.
15. Nissim, I. Newer aspects of glutamine/glutamate metabolism:
the role of acute pH changes. Am. J. Physiol. 277 (Renal Physiol.
46): F493–F497, 1999.
16. Peghini, P., J. Jaryen, and W. Stoffle. Glutamate transporter
EAAC1 deficient mice develop dicarboxylic aminoaciduria and
behavioral abnormalities but no neurodegeneration. EMBO J.
16: 3822–3832, 1997
17. Shapiro, R. A., R. F. Morehore, and N. P. Curthoys. Inhibition by glutamate of phosphate dependent glutaminase of rat
kidney. Biochem. J. 207: 561–566, 1982.
18. Shayayakul, C., U. Kanai, W-S Lee, D. Brown, J. D. Rothstein and M. A. Hediger. Localization of the high-affinity
glutamate transporter EAAC1 in rat kidney. Am. J. Physiol. 273
(Renal Physiol. 42): F1023–F1029, 1997.
19. Welbourne, T., and D. Chevalier. Glutamate transport and
not cellular content modulates paracellular permeability in
LLC-PK1-F⫹ cells. Am. J. Physiol. 272 (Endocrinol. Metab. 35):
E367–E370, 1997.
20. Welbourne, T., D. Chevalier, and X. Mu. Glutamate transport
modulation of paracellular permeability across LLC-PK1-F⫹
monolayers. Am. J. Physiol. 271 (Endocrinol. Metab. 34): E889–
E895, 1996.
21. Welbourne, T., and J. Fuseler. Growth hormone-enhanced
acid production and glutamate and glutamine utilization in
LLC-PK1-F⫹ cell lines. Am. J. Physiol. 265 (Endocrinol. Metab.
28): E874–E879, 1993.
22. Welbourne, T., and X. Mu. Extracellular glutamate flux regulates intracellular glutaminase activity in LLC-PK1-F⫹ cells.
Am. J. Physiol. 268 (Cell Physiol. 37): C1418–C1424, 1995.
23. Welbourne, T., X. Mu, and S. Evans. Enteral glutamine
modulates renal glutamine utilization. J. Nutr. 126.: 1137S–
1141S, 1996.
Downloaded from http://ajprenal.physiology.org/ by 10.220.33.6 on June 18, 2017
11.
modes. Am. J. Physiol. 277 (Renal Physiol. 46): F481–F486,
1999.
Goldmann, D. R., and S. Segal. ␥-Glutamyltranspeptidase
activity in newborn rat kidney brush border. Enzymes 22:
301–311, 1977.
Gstraunthaler, G., and J. S. Handler. Isolation, growth and
characterization of a gluconeogenic strain of renal cells. Am J.
Physiol. 250 (Cell Physiol. 19): C232—C238, 1987.
Hediger, M. A. Glutamate transporters in kidney and brain.
Am. J. Physiol. 277 (Renal Physiol. 46): F487–F492, 1999.
Kvamme, E., I. A. Torgner, and B. Robert. Evidence indicating that pig kidney renal phosphate-activated glutaminase has a
functionally predominant external localization in the inner mitochondrial membrane. J. Biol. Chem. 226: 12185–12192, 1991.
Matthews, J. C., M. J. Beveridge, M. S. Malandro, J. D.
Rothstein, M. Campbell-Thompson, J. W. Verlander, M. S.
Kilberg, and D. A. Novak. Activity and protein localization of
multiple glutamate transporters in gestation day 14 vs. day 20
rat placenta. Am. J. Physiol. 274 (Cell Physiol. 43): C603–C614,
1998.
McGivan, J. D. and B. Nicholson. Regulation of high-affinity
glutamate transport by amino acid deprivation and hyperosmotic
stress. Am. J. Physiol. 277 (Renal Physiol. 46): F498–F500, 1999.
Meade, D., C. Chess, and T. C. Welbourne. Glutamate transport and cellular glutamine metabolism: regulation in LLC-PK1
vs. LLC-PK1-F⫹ cell lines. Am. J. Physiol. 274 (Cell Physiol. 43):
C1616–C1624, 1998.
Milford, L., and T. C. Welbourne. Cellular glutamate content
modulates extracellular uptake in aging LLC-PK1-F⫹ cells (Abstract). FASEB J. 12: 316, 1998.
Mu, X., and T. Welbourne. Response of LLC-PK1-F⫹ cells to
metabolic acidosis. Am. J. Physiol. 270 (Cell Physiol. 39): C920–
C925, 1996.
F505