| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
* Molecular Neurobiology Branch, NIDA-IRP, National Institutes of Health, and
Departments of Neurology and Neuroscience, Johns Hopkins University School of Medicine, Baltimore, Maryland 21224, USA
1Correspondence: Molecular Neurobiology Branch, NIDA-IRP, National Institutes of Health, 5500 Nathan Shock Dr., Baltmore, MD 21224, USA.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: affinity • translocation • ion gradients • turnover rate • plasma membrane targeting
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2
3
4
5
6)
. Many rewarding effects of cocaine have been
attributed to its ability to block DAT, enhancing synaptic dopamine
concentrations in brain reward pathways (7
, 8)
.
Elucidating DAT cDNAs and genes provides DAT’s primary structure and
relationship to other neurotransmitter transporters, including the
norepinephrine (NET) and serotonin (SERT) transporters that also serve
as cocaine recognition sites. However, little is known about the
molecular details of major DAT functions, including the ways in which
it interacts with substrates, ligands, and ions and the mechanisms by
which it translocates dopamine. Increased understanding of DAT function
could benefit from information about DAT tertiary structure. Only one
12-transmembrane (TM) domain transporter has been subject to successful
crystallographic structural determination. Current DAT topological
models are based largely on hydrophobicity analyses, mutagenesis
studies, sequence comparisons with other neurotransmitter transporters,
and analyses of posttranslational modifications. Figure 1A
thus represents some of the best current understanding of
possible DAT topologies, in light of the limited supporting evidence
available.
|
Models for interactions between dopamine and DAT have been influenced
by studies of chimeras and initial DAT mutants. Pharmacological
analyses of chimeras between DAT and NET suggested that DAT TMs 9–12
or TMs 1–3 be important for dopamine uptake affinity whereas TMs 5–8
could contribute to cocaine binding affinity (12
13
14)
.
Mutations of DAT TM polar and charged residues have suggested that
dopamine recognition might occur through mechanisms with analogies to
those used for catecholamine recognition by seven-transmembrane domain,
G-protein-linked catecholamine receptors. Dopamine’s catechol could
interact with paired serine residues disposed in putative DAT TM 7, for
example (15)
.
Proline residues represent an especially attractive target for
mutagenesis studies aimed to enhance understanding of static and
dynamic DAT structures important for its assembly and its many
functions. Important prolines dynamic roles in other membrane proteins,
through cis-trans isomerizations and other mechanisms,
render them strong candidates to contribute to intramolecular
mobilities likely to be required for DAT transport functions.
Spectroscopic studies reveal that bacteriorhodopsin transmembrane
proline structural rearrangements at Xaa-ProC-N peptide bonds accompany
proton pumping (16)
. Prolines, found more frequently in
transporter proteins than in globular proteins (17)
, could
well contribute to molecular mobilities important for
transporter-mediated substrate and ion translocation.
Prolines could also play important roles in forming and maintaining
more static DAT structural features. They can break or kink alpha
helical motifs commonly found in protein secondary structures
(18
, 19)
, since their amide nitrogens cannot form all of
the polypeptide backbone hydrogen bonds possible with other amino
acids. The bulk of proline side chains also discourages polypeptide
backbone hydrogen bond formation by amine nitrogens from neighboring
amino acids. Prolines bend helices in globular proteins by ~26°
(20)
, often producing bent or curved helices. Such helices
can pack into funnel- or cage-like structures with obtuse kink angles
facing inwardly or outwardly, respectively (21)
. These
proline configurations appear to contribute to numerous functions of
complex proteins; in many cases, eliminating them degrades function.
Proline deletions from the sarcoplasmic reticulum
Ca2+-ATPase reduce its Ca2+
uptake, perhaps by eliminating proline-induced kinks important for
proper positioning of Ca2+ binding sites
(22)
. Bacteriorhodopsin function is degraded by proline
mutants; modeling suggests that these proline residues normally provide
molecular rigidity, which stabilizes optimal relationships between the
positions of other functionally important residues (23)
.
Finally, proline residues can provide sites for interactions with
cationic ligands. The failure of prolines to form the intrahelical
hydrogen bonding characteristic of most other amino acids leaves
peptide backbone carbonyl oxygens from residues even several positions
amino- or carboxyl-terminal to the proline uninvolved in normal
intrahelical hydrogen bonding and exposed for interactions with ligands
(24
, 25)
. Proline nitrogens can also interact more
directly with cationic ligand features, providing ‘proton holes’
(26)
that can contribute significantly to the functional
activities of several molecular classes.
DAT sequences display 16 proline residues located within or adjacent to
8 of its 12 putative TM domains. Several of these proline residues are
conserved in DAT sequences from several species, in NET and in SERT
(Fig. 1B
). Mutating these prolines provides a powerful
approach to the problem of defining structural features responsible for
DAT-mediated molecular recognition and translocation events. Altered
patterns of dopamine uptake and cocaine binding observed in proline
mutants need to be interpreted in light of important caveats, however.
Although altered function could arise from direct mutation influences
on DAT/small molecule interactions, other effects can also be observed.
Mutations disrupting amino acid side chains important for stabilizing
interactions between adjacent DAT hydrophobic domains or membrane
lipids could influence function by gross effects on molecular
configurations, including those necessary for proper trafficking and
membrane insertion. Altered DAT functions observed in such mutants may
not necessarily indicate that the mutated residue participates directly
in ligand or substrate recognition. Dopamine transport could be
influenced indirectly, since it requires steps that include substrate
recognition, ion recognition, substrate and ion translocation, release
of substrate and ions into cytoplasm, and transporter reorientation.
DAT’s transport processes and turnover rate estimates thus could be
influenced by alterations at sites important for any of these transport
steps. Each of these interpretations needs to be carefully considered
in interpreting the effects of proline mutations on DAT’s affinities
for small molecules and its ability to transport them.
Despite such cautions in interpretation, identification of DAT amino
acids and domains important for specific features of dopamine and/or
cocaine recognition can add substantially to understanding dopamine
transport and cocaine action. For these reasons, we have produced DAT
mutants in each of these 16 hydrophobic amino acids distributed near or
through 8 of the 12 putative DAT TM domains (Fig. 1)
. We now report the
characterization of the properties of these mutants and discuss these
data in light of the mutants’ differential influences on specific DAT
properties.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
as described (Muta-Gene Phagemid
In Vitro Mutagenesis Version 2, Bio-Rad,
Hercules, Calif.). Mutagenesis was undertaken by annealing the
oligonucleotides to the single-stranded wild-type template, in
vitro synthesis and ligation of the mutant strand, nicking and
digestion of nonmutant strand, and repolymerization and ligation of the
gapped DNA as described by the manufacturer. Mutations are defined
using a single letter for the wild-type amino acid’s position number
and the substituted amino acid. Prefix numbers represent the putative
transmembrane domain in which the mutation is located. Mutations in TMs
1–2 were isolated in NotI-BglII fragments;
mutations in TMs 3–7 and P136A, P234A, P235A, P287A, P394A were
isolated in BglII-PvuI fragments and mutations in
TMs 8–12 and P515A, P544A, P545A, P553A were isolated in
PvuI-PstI fragments of pBluescript/rDAT. Each
mutation was confirmed by DNA sequencing.
Subcloning into a modified expression vector, pcDNA3.1/ZL-rDAT
Mutation-bearing restriction fragments were shuttled into the
rDAT-expressing mammalian plasmid pcDNA3.1/ZL-rDAT and correct
sequences were reconfirmed. pcDNA3.1/ZL-rDAT was derived from
pcDNA3.1+ (Invitrogen). The pcDNA3.1
BglII site was removed by digestion, fill-in
reactions, and religation. PvuI and PstI sites
were removed from sites outside the multiple cloning site using
site-directed mutagenesis as described above. Subcloning the 3.4 kb rat
DAT cDNA fragment from the pBluscript/rDAT cDNA (9)
into
the EcoRI-XbaI sites of the modified pcDNA3.1,
designated pcDNA3.1/ZL, produced pcDNA3.1/ZL-rDAT thus displayed single
sites for shuttling of the NotI-BglII,
BglII-PvuI, and
PvuI-PstI pBluscript/rDAT cDNA fragments that
carried the DAT mutations studied here.
Immunostaining transfected COS cells
Cells transfected with a truncated and promotorless version of
pcDNADAT1 (9)
, pcDEDAT, provided a negative control. Cells
were grown to 80% confluence in 6-well plates and cellular patterns of
DAT immunoreactivity were assessed by immunohistochemistry using
specific polyclonal rabbit anti-DAT sera, as described
(28)
. Stained cells were washed three times with TBS,
dehydrated, mounted on microscope slides, and examined for
semiquantitative assessments of the patterns of DAT immunoreactivity by
an observer unaware of the mutations.
Functional analyses
107 COS cells were transfected with 20
µg of pcDNA3.1/ZL-rDAT or mutant DNAs, grown in 6-well plates,
allowed to express the plasmids for three days, and then assayed for
their abilities to accumulate [3H] dopamine (49
Ci/mmol, NEN) or to bind the cocaine analog
[3H] CFT (83.5 Ci/mmol, NEN) by incubation in
Krebs-Ringer HEPES-buffered solution (KRH; 125 mM NaCl, 4.8 KCl, 1.3 mM
CaCl2, 1.2 mM MgSO4, 1.2 mM
KH2PO4, 5.6 mM glucose and 25.0 mM HEPES).
Kinetic and saturation analysis determined
Km,
Vmax, or
KD, and Bmax
values, respectively, as described (8)
. For uptake assays,
10 nM [3H] dopamine and 0.1, 1, 5, 10, 20, 30,
and 50 µM unlabeled dopamine concentrations were used. For initial
binding assays, 2 nM [3H] CFT was adjusted to
1.5, 3, 5, 15, 30, and 60 nM concentrations using unlabeled CFT. Cells
transfected with wild-type pcDNA3.1/ZL-rDAT served as controls.
Parallel incubations with 50 µM unlabeled (-) cocaine allowed
estimation of nonspecific uptake and binding. Uptake assays were
carried out for 5 min at 37°C, followed by two complete washes with 2
ml of KRH containing 50 µM ascorbic acid. Binding assays were carried
out for 2 h at 4°C, followed by three washes with 2 ml of 4°C
KRH buffer. Cells were solubilized in 0.5 ml of 1% sodium dodecyl
sulfate (SDS) solution and radioactivity was determined using a Beckman
LS 6000 liquid scintillation counter at ~50% efficiency. Studies of
dopamine inhibition of 2 nM [3H] CFT binding
used varying concentrations of dopamine in 50 µM ascorbic acid.
[3H] Alanine uptake assays were carried out by
incubating DAT-expressing COS cells at 37°C for 5 min with 300 µM
alanine that contained 10 nM radiolabeled alanine (92.6 Ci/mmol, NEN)
and 290 µM unlabeled alanine. For experiments demonstrating the
Na+ and Cl- dependence of
uptake, a dopamine concentration of the mutant’s
Km value was prepared to contain
99.5% unlabeled and 0.5% tritium-labeled dopamine for uptake assays
in different concentrations of Na+ or
Cl- in which lithium substituted for
Na+ and acetate for Cl- to
maintain osmolarity. Pargyline (50 µM) and 1 µM of the COMT
inhibitor RO 41–0960 (RBI, Natick, Mass.) were included in assay
buffers.
Analyses and definitions
Km and
Vmax values for
[3H] dopamine uptake,
KD and Bmax
values for [3H] CFT binding activities,
IC50 values, calculation of Hill coefficients,
curve fit to data using sigmoidal curve models for binding competition
data, calculation of data fitting, and t tests or analysis
of variance (ANOVA), followed by Tukey’s multiple comparison tests,
were all carried out using GraphPad PRISM Version 2 programs (GraphPad
Software, Inc., San Diego, Calif.). Ki
values were calculated as described (30)
.
We defined two criteria for significance of mutation effects. To meet
the first arbitrary criterion, the mutants needed to display
Km,
Vmax,
KD, or Bmax
values more than threefold different from wild-type values. The second
criterion required that the nominal t test statistical
comparisons were at P 
0.05. We list results from
mutants that did not meet the first criterion but achieve statistical
significance separately in Figs. 3
and 6
. Transporter turnover rate
reflected the number of dopamine molecules transported per second per
site, defined as [Vmax for dopamine
uptake in fmol/µg/min]/[Bmax for CFT binding
in fmol/µg x 60 (s/min)]. A ratio (termed
KD/Km
ratio) between mutation influences on cocaine and dopamine recognition
was calculated as [KD for CFT
bindingmutant/KD
WT]/[Km for dopamine
uptakemutant/Km
WT].
|
|
Molecular modeling
Transmembrane domain modeling used Sybyl 6.4 and STATFIT.SPL
programs (Tripos, Inc., St. Louis, Mo.). Transmembrane domains for
wild-type and mutant DATs were built as 
-helices and energy was
minimized. Weighted root mean square deviations (WRMSD) (Å) were
obtained for atomic relocation of backbones, side chains, or both in
comparing wild-type and mutant helices. When the proline to be mutated
divides the helix into two fragments, the larger fragment was used to
expedite comparisons.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Eight of the 16 Pro
Ala mutants displayed patterns of
DAT immunostaining similar to those of wild-type DAT (Fig. 2)
. These
included substitutions 2P101A, P234A, and P235A in ECL2, 5P272A,
5P287A, and P394A in ECL4, and P544A and P545A in ECL6.
|
Mutations of another eight proline residues altered expression
patterns. For mutants 8P401A, P515A in ICL5, and 11P528A, one-half to
one-third of the protein was distributed to the plasma membrane, while
the rest was found in perinuclear patterns (Fig. 2B
). Five
mutants (1P87A, 2P112A, P136A in ICL1, P553A in ECL6 and 12P572A)
displayed the most disruption of plasma membrane targeting.
Semiquantitative observations suggested that less than one-third of the
DAT immunoreactivity expressed by these mutants was inserted into the
cells’ plasma membranes (Fig. 2C
). Seven of these
eight expression-disrupting mutants are substitutions for prolines
conserved among many
Na+/Cl- dependent
transporters. Only two (2P101A and 5P272A) of the eight mutants that
left expression indistinguishable from wild-type are conserved across
most members of this family (Fig. 1B
).
[3H] Dopamine uptake
Wild-type DAT displayed a Km
value of 2.6 µM for dopamine uptake. Six of the eight mutants that
displayed normal patterns of cell surface immunostaining (Fig. 2)
—P234A and P235A in ECL2, 5P287A, and P394A in ECL4, and P544A and
P545A in ECL6—displayed affinities for dopamine uptake similar to
wild-type values (Fig. 3
, upper left). Each of these residues is conserved in the DAT sequences
from different species, but none is absolutely conserved among the 20
transporters arrayed in Fig. 1B
.
Several mutations influenced affinities for dopamine uptake. Mutation of the two highly conserved residues that did not affect the plasma membrane expression pattern did significantly change affinities for dopamine. Mutant 2P101A displayed a Km value of 0.5 µM, a fivefold increase in affinity compared to wild-type. Mutant 5P272A displayed a Km value of 74.9 µM, a 29-fold decrease in affinity.
Six of the eight mutants that displayed abnormal immunostaining
patterns displayed less than one-third of the affinity for dopamine
displayed by wild-type DAT and reached our threshold for
‘significant’ differences (see Materials and Methods). 1P87A
displayed a 21-fold decrease; 2P112A, 17-fold; P136A, 9-fold; P528A,
44-fold; P553A, 8-fold, and 12P572A, 4-fold (Fig. 3
, upper left). Only
8P401A and P515A yielded small alterations in dopamine affinities that
did not reach our criteria for significance.
Dopamine competition for CFT binding was examined in 10 mutants,
allowing an independent assessment of mutation-induced changes in
dopamine affinity and elucidation of the Hill slopes of displacement
curves. Competition binding curves (Fig. 4
) revealed that dopamine could almost totally eliminate specific
[3H] CFT binding from wild-type DAT and from
each of the three representative mutants. Most mutants produced
alterations in dopamine affinity in uptake experiments that paralleled
mutation-induced changes in dopamine affinity in competing for cocaine
analog binding, with
Km/Ki ratios ranging
between 2 and 3.6 (Table 1
). Mutants such as P234A that did not influence dopamine
Km values in uptake experiments
displayed near wild-type values for dopamine competition for specific
[3H] CFT binding. Mutants 1P87A, 2P112A, P136A
5P272A, 11P528A, and 12P572A conferred less potency on dopamine
competition for specific [3H] CFT binding,
consistent with their effects on dopamine
Km values obtained from uptake
experiments. Conversely, 2P101A gained potency in both assays. Two
mutants produced Km/Ki
ratios of around 5 (Table 1)
. Dopamine was slightly more potent in
inhibiting [3H] CFT binding to the 2P101A and
P136A mutants than expected based on mutation effects on dopamine
uptake. However, no mutant produced any dramatic dissociation between
shifts in dopamine potency in uptake assays and shifts in its potency
in competition experiments. No mutant examined significantly changed
the Hill coefficient for displacement of [3H]
CFT binding by dopamine (data not shown).
|
|
[3H] CFT binding
[3H] CFT bound to wild-type DAT with an
average dissociation constant (KD) of
22 nM. Thirteen mutants retained normal or near-normal CFT affinities.
Three mutants displayed significantly decreased CFT binding affinities.
Mutations of 5P272, 5P287, and 11P528 each decreased CFT affinity by
more than threefold. 2P101A, 8P401A, P515A, P553A, and 12P572A
displayed statistically significant decreases in CFT affinity that were
within threefold of wild-type values (Fig. 3
, upper right). No mutation
increased CFT affinity.
Selectivity of ProAla mutation influences
The effects of proline mutations at 16 positions in DAT fell into
five categories (Fig. 5
). Mutations in Group 1 changed affinity for neither dopamine nor CFT.
Seven mutants, six located in putative nontransmembrane loops and one
(8P401A) in TM 8, fell into this category. Only one of these seven
mutants, P515A, involved a highly conserved residue (Fig. 1B
). Mutations in Group 2 enhanced affinity for dopamine
without altering CFT affinity. The sole mutant with these properties,
2P101A, changed a residue that is highly conserved among transporters
(Fig. 1B
). Five mutations in Group 3 selectively decreased
affinity for dopamine. These mutations (1P87A, 2P112A, P136A, P553A,
and 12P572A) each disrupted plasma membrane expression of DAT (Fig. 2)
.
Two mutations in transmembrane domains, 5P272A and 11P528A, comprised
Group 4; they reduced DAT affinities for both dopamine and CFT. P287 is
the single mutant in Group 5 that decreased affinity for CFT without
reducing dopamine affinity. This residue is conserved among the
monoamine transporters, which serve as principal cocaine recognition
sites in the brain (Fig. 1B
). Each of the eight residues at
which mutations influenced dopamine uptake affinity is absolutely
conserved among the 20 neurotransmitter transporter sequences assessed.
However, one of the three residues at which mutations influenced CFT
affinity, 5P287, is not absolutely conserved.
|
ProGly substitutions at positions 101, 272, and 287
The methyl group alanine side chain provides a modest hydrophobic
contribution and tends to propagate helix structures into
neighboring amino acid residues. Glycine lacks this methyl, is less
hydrophobic, and can fail to propagate helix structures into
neighboring residues. We thus examined effects of glycine substitution
for prolines at which 1) alanine substitution substantially
reduced dopamine uptake or CFT binding affinity and 2)
expression of the alanine substitution mutant was similar to wild-type
patterns. Prolines 101, 272, and 287 fit these criteria. Data in Fig. 2
reveal that glycine substitution mutants P101G and P272G expressed in a
fashion indistinguishable form wild-type DAT. Some COS cells expressing
P287G appeared to express normally, whereas some subpopulations of the
cells display greater perinuclear staining than cells expressing
wild-type DAT (Fig. 2B
). Glycine substitutions for either
proline101 or proline 287 influenced dopamine uptake and CFT
recognition in fashions parallel to those of alanine substitutions at
these same sites (Fig. 6
). However, glycine substitution at position 272 yielded much more
dramatic reductions in CFT affinity than the corresponding alanine
substitution. Mutants with this glycine substitution (except 5P272G)
also revealed a trend toward smaller changes in dopamine affinity than
those induced by alanine substitution.
Seeking ion gradient alterations or alterations in ion dependence
of selected Ala substitution mutants
The relatively large influence of alanine substitutions for P101,
P272, P515A, and P528A on dopamine uptake, in the face of near-normal
expression patterns, led us to assess whether these mutations could
influence uptake through altering transmembrane gradients of ions or by
altering the Na+ or Cl-
dependence of uptake.
To assess possible mutation influences on COS cells’ abilities to maintain ion gradients necessary for ion-dependent transport processes, we examined the abilities of cells expressing these four mutants to manifest sodium-dependent [3H] alanine uptake. The cells’ initial rates of accumulating 300 µM [3H] alanine for 5 min were 100 ± 8% (2P101A), 103 ± 13% (5P272A), 87 ± 1% for P515A, and 90 ± 1% for 11P528A of wild-type values. These data provided no evidence that expression of any mutant substantially reduced the transmembrane sodium gradients sufficiently to greatly reduce alanine uptake, although the small changes noted with mutants P515A and 11P528A might conceivably reflect modest alterations.
Rates for uptake of dopamine at the Km
concentration by the wild-type DAT were markedly
Na+ or Cl- dependent.
Modest differences in the sodium dependence of uptake were observed in
each normally expressed mutant 2P101A and 5P272A, but not in P287A
(Fig. 7
). There were slight increases in sodium EC50
values noted for the 2P101 and 5P272 mutants and decreases for the P287
mutant (110, 103, and 68 mM, respectively, compared to 79 mM wild-type
values). Chloride dependence was altered most strikingly by mutant
5P572A; its 72 mM EC50 value doubled the
wild-type value of 37 mM. The maximal effect at 100 mM for
Cl-, different from the maximal effects at 150
mM for Na+, was noted for wild-type or the two
other mutants. Chloride EC50 values were also
altered slightly in 2P101A (43 mM) and in P287A (33 mM).
|
Pharmacological parameters
KD/Km ratios
KD/Km ratios
were calculated to quantitatively dissociate mutation influences on
dopamine and cocaine analog affinities (Fig. 8A
).
KD/Km values were
1 for wild-type, <1 for mutations with greater effects in reducing
dopamine affinity, and >1 for mutants with selectively decreased CFT
affinity. 2P101A displayed a
KD/Km value of
13, due to both increased dopamine affinity and decreased CFT affinity
(Fig. 3
, upper). 2P101G shows a
KD/Km value of
4.9; the glycine substitution has a smaller effect on CFT binding (Fig. 6B
). Mutation P287A displayed a
KD/Km value of
2.7; the corresponding glycine substitution P287G displayed a similar
value, 2.8. These data characterize the most selective mutation
influences on reducing CFT affinity identified here. 1P87A, 2P112A,
P136A, and P553A each displayed
KD/Km values of
<0.3. These mutants thus provided the most selective mutation
influence on dopamine affinity since they left CFT affinities
relatively intact. 5P272A and 11P528A display low
KD/Km ratio
values, since the mutations influenced dopamine uptake more than they
affected CFT binding. Each also had significant effects of CFT binding.
|
Uptake rates
One group of mutants displayed concordant expression, binding, and
uptake properties consistent with near-normal expression of almost
normally transporting DAT variants. P234A, P235A, 5P287A, P394A, P544A,
P545A, and P553A mutants displayed normal uptake rates. Each also had a
CFT Bmax value near wild-type values and
near-normal expression patterns determined by immunohistochemistry. In
contrast, in a second group of mutants including 1P87A, 2P112A, and
P136A, each decreased dopamine uptake Vmax, CFT
binding Bmax, and reduced densities of
immunostaining of plasma membrane DAT immunoreactivity compared to
wild-type values (Figs. 2
, 3)
. These mutants provide a picture
consistent with observed mutation-induced reductions in expression.
8P401A, P515A, 11P528A, and 12P572A each significantly decreased
dopamine uptake Vmax values and
reduced plasma membrane expression. The CFT binding
Bmax values for these mutants, 3.4–5.9
fmol/µg, were numerically lower that 6.8 fmol/µg wild-type values,
although variability rendered this difference statistically
insignificant (Fig. 3
, lower right). Mutation 2P101A reduced dopamine
uptake Vmax to 7.2 fmol/µg/min, only
2.7% of the wild-type value of 266 fmol/µg/min. 5P272A reduced
dopamine uptake Vmax and CFT binding
Bmax values in a more concordant manner.
Turnover numbers
Information about Bmax and
Vmax values allowed calculation of turnover
rates, assessing the number of dopamine molecules transported per DAT
per second. Wild-type DAT showed a turnover rate of 0.65
(s-1). 2P101A and 2P101G decreased this rate by
27- and 13.5-fold, respectively. 1P87A, 2P112A, 8P401A, 12P572A, and
5P272G decreased the rate to less than one-third of wild-type values.
Each of the other 12 mutants displayed turnover rates within threefold
those of wild-type DAT (Fig. 8B
).
Relationships between
KD/Km ratios
and turnover numbers
Mutation-induced changes in turnover rates generally correlated
with alterations in
KD/Km
ratios (Fig. 8C
). Mutations that decreased the dopamine
uptake turnover rates tended to increase
KD/Km
ratios. Results with mutant 2P101A were outstanding in this regard.
This mutant displayed the lowest turnover rate and an especially high
KD/Km
ratio.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Prolines important for DAT assembly and membrane insertion
Evidence for important roles for transmembrane domain proline
residues in the assembly of an appropriate overall DAT structure comes
from an evaluation of the cellular expression patterns of the mutants
reported here. Comparisons of the frequencies of disrupted expression
for proline mutants with the parallel values obtained from studies of
mutants in other DAT residues can help put these data in context. Five
of the seven transmembrane domain proline mutants (70%) substantially
disrupt expression (Fig. 1B
). By contrast, only 24% of
previously studied DAT TM domain phenylalanine mutants
(31)
and 33% (3 of 9) non-TM proline mutants studied in
the current report disrupt expression.
The eight proline mutations that disrupt plasma membrane expression of
DAT lie at transmembrane and nontransmembrane positions that are
remarkably conserved among members of this neurotransmitter transporter
family. Only two substitutions are tolerated at these positions in
other monoamine transporters (see Fig. 1B
). By contrast, the
8 mutants with normal expression patterns lie at positions at which 15
substitutions are found in other monoamine transporters. These lines of
evidence suggest that the eight expression-disrupting mutants—1P87A,
2P112A, P136A. 8P401A, P515A, 11P528A, P553A, and 12P572A—could
identify prolines important for the proper configurations of many
transporters.
Proline contributions to dopamine recognition: alanine
substitutions
Uptake studies reveal the most impressive reductions in dopamine
affinities for mutants 1P87A, 2P112A, 5P272A, and 11P528A (Fig. 3)
.
Since the short loop connecting TM1 and TM2 is likely to make these TM
domains near neighbors, each of these amino acids might be important
for configurations required for recognition of specific dopamine
features. If the TM1 D79 aspartic acid is involved in recognition of
dopamine’s amine moiety, a possibility suggested (but not proved) by
prior mutagenesis work, then these two prolines could even be involved
in maintaining local positioning appropriate for these DAT/dopamine
interactions. It is also conceivable that TM 11 could be near TMs 1 and
2 in the assembled DAT. Such proximity would provide the opportunity
for the 11P528 residue to contribute, even interactively, to a dopamine
recognition pocket that also involved TMs 1 and 2.
No current model places TM 5 next to the TM1/2 + 11/12 domains. However, 5P272 does lie near other TM 4 and 5 polar and aromatic residues at which alanine substitutions altered dopamine affinities. 5P272 might normally contribute to recognition of a different part of the dopamine molecule than that which might interact with TM 1/2 residues. It is interesting that 1P87, 2P112, and 11P528 are highly conserved in the transmitter transporter family. Although the influence of substitutions at most of these positions on dopamine affinities suggests that they could play roles in dopamine recognition, the conservation of these residues could also indicate that these prolines could contribute to the abilities of DAT to perform functions shared with other transporters. Obviously, such shared functions could include the abilities to attain conformations appropriate for proper membrane insertion or recognition and/or movement of ions.
Proline contributions to dopamine recognition: glycine
substitutions
Mutation 2P101A increased dopamine uptake affinity from 5- to
10-fold (Figs. 3
and 4
, Table 1
). When we substituted glycine 2P101G
mutant and observed strong parallels between results with the
helix-breaking 2P101G and helix-favoring 2P101A mutants, we found
little support for the idea that helical interruption provides a major
explanation for the wild-type 2P101’s role in maintaining a
physiological dopamine affinity. Lack of statistically significant
differences between glycine and alanine substitution effects on
dopamine affinities for 5P272 and P287 also suggests that
helix-breaking functions may not be major functions of the proline
residues found in the wild-type transporters (32)
.
Proline contributions to cocaine analog recognition
Several proline substitutions resulted in substantial losses of
affinity for the cocaine analog CFT. Substitutions at 2P101, 5P272,
P287, 8P401, P515, 11P528, and 12P572 yielded at least two- to
threefold losses of CFT affinity. These observations at least
tentatively implicate TM domains 2, 5, 8, 11, and 12 in CFT
recognition. Polar residue substitutions in TM2, TM5, TM8, TM11, and
TM12 also result in lost CFT affinity. Some of the largest losses in
other mutagenesis work from our laboratory have come from the TM5
mutant Y273A that lies adjacent to the P272A mutant (M. Itokawa, Z.
Lin, and G. R. Uhl, unpublished results). Conceivably, 5P272 could
be important for orienting the Y273 side chain in a fashion especially
important for proper cocaine analog recognition.
Selectivity of proline contributions to cocaine analog vs. dopamine
recognition
Five of the mutations, 1P87A, 2P112A, P136A, P553A, and 12P572A,
influenced dopamine uptake affinity with selectivity. A single
mutation, P287A, selectively influenced CFT binding affinity (Fig. 5)
.
The residue at which mutations selectively reduce CFT affinities is
located in an extracellular loop, whereas the prolines at which
substitutions selectively reduce dopamine affinities are found in TM
domains as well as loops that connect TM domains. It is possible that
either the proline hydrophobicity or its nitrogen electron pair
contribute to affinity for dopamine or cocaine analogs. Mutations
5P272A and 11P528A significantly reduced affinities for both dopamine
and cocaine analogs (Fig. 3
and Table 1
). These residues could thus
conceivably contribute to sites at which dopamine and cocaine
recognitions overlap. The substantial depth of these amino acids in
current models of DAT topology also suggests cocaine recognition pocket
elements lying deep within the DAT protein (Fig. 5)
.
Limitations of mutagenesis approaches
Overviews of the alterations in expression, dopamine uptake, and
cocaine binding observed in these proline mutants suggest the need to
invoke several of the cautions noted in the introduction. The fact that
half of the mutations change expression patterns suggests important
structural roles for many of these proline side chains and invokes
caution in interpretation of the affinity values obtained for these
mutants. However, removing these proline side chains is one of the
smaller changes in DAT that can be readily produced. It thus seems
unlikely that studies with other indirect means of examining DAT
structure–function relationships will reduce the need for such
interpretive caution. The molecular weights of the sulfhydryl reagents
commonly used to infer functional features of transmembrane proteins,
for example, are at least three times the size of the alterations
produced here (33)
.
Contributions of selected prolines to dopamine
Vmax: implications for proline contributions
to DAT mobilities important for dopamine translocation
Evidence for important roles for a number of transmembrane domain
proline residues in DAT transport functions comes from evaluation of
the mutants’ dopamine transport rates. DAT TM domain proline mutations
reduce Vmax values by 84%, on
average. These values contrast with average reductions of only 35% in
the non-TM proline mutants and of 60% in TM domain phenylalanine
mutants studied in this laboratory (31)
. Turnover numbers
parallel these observations. Reductions were 71%, 8%, and 49% for
the TM proline mutations, non-TM proline mutations, and TM domain
phenylalanine mutations, respectively. 2P101A is especially interesting
in this regard, as it is the only alanine substitution that eliminates
dopamine uptake activity but retains normal plasma membrane expression
and near-normal CFT binding affinity. Glycine substitution for 2P101
displayed effects similar to those of alanine substitution.
Conceivably, both 2P101 substitutions could load dopamine molecules but
translocate them poorly. If dopamine affinities for an initial
recognition site on DAT were so high that the molecule could not be
subsequently ‘handed off’ to other more cytoplasmically disposed
sites, mutants could display high
KD/Km
ratios but low turnover rates. These properties are consistent with the
results from substitutions at this residue (Fig. 8)
.
Prolines critical for DAT functions
Alanine substitution for either 5P272 or 11P528 significantly
decreases the affinity for both dopamine and CFT and the maximal
velocity of dopamine translocation (Fig. 3)
, indicating that both of
these highly conserved residues are central to the DAT functioning.
None of other 14 mutations produces such devastating influences. Five
mutations including 1P87A, 2P112A, P136A, and 12P572A significantly
reduce both dopamine uptake affinity and
Vmax, but not CFT affinity. The
reductions in Vmax, most likely, are
resulting from the disruptions in plasma membrane expression of these
mutant DATs (Fig. 2)
. Eight (50%) of the 16 mutations decrease
dopamine uptake activities compared to three (19%) mutations that
decrease CFT affinity. Those important for dopamine uptake are located
in DAT domains from outside the membrane, through and inside the
membrane, whereas those important for CFT affinity are located in
domains from outside to the middle of the membrane. These data suggest
that proline residues are much more important for dopamine uptake than
for cocaine analog binding, probably because they play important roles
in DAT mobility through isomerization.
Modeling specific features
Molecular modeling of helices with wild-type or 2P101A and 2P112A
mutations can provide insight into the kinds of alterations that are
possible in a putative TM 2 helix after such mutations. 2P101A gains
dopamine affinity, whereas 2P112A loses both dopamine uptake (17-fold)
and dopamine affinity (61-fold). These two mutants could provide
several classes of distinctive changes from the wild-type proline,
including local helix disruption. The wild-type TM 2, as currently
modeled, disrupts the helical configuration of its three
extracellularly located residues (G96, A97, F98) due to the presence of
2P101. A second disruption could occur at I108 and A109 due to 2P112.
By contrast, the 2P101A substitution could allow the TM 2 helical
configuration to extend to the extracellular membrane border. 2P112A,
by contrast, would allow a longer helical stretch within most of the
second TM domain, except its most extracellular portion (Fig. 9A
). Relocation of TM2 amino acid side chains could also
differ between the 2P101A and 2P112A. 2P101A changes the side chain
locations of G96, A97, F98, L99, and V100 in current models, whereas
2P112A changes F98, Y102, F105, I108, L113, F114, Y115, and M116 side
chains. The two mutations also provide differing effects on side chain
orientation (Fig. 9B
). 2P101A relocates only residues at the
amino-terminal side of the site of the altered proline, altering only
the F98 aromatic side chain orientation (Fig. 9B
, left
side). 2P112A relocates residues at both the amino- and
carboxyl-terminal sides and influences the disposition of the five
aromatic side chains in this helix, F98, Y102, F105, F114, and Y115
(Fig. 9B
, right side).
|
Modeling allows assessments of mutagenesis-induced side chain
displacements from their wild-type locations. 2P101A produces weighted
root mean square deviations (WRMSDs) of 0.65 Å. 2P112A produces
changes of 1.12 Å (Fig. 9C
). After optimal superimposition
on wild-type, 2P101A produces a WRMSD of 2.39 Å for F98, but less than
0.06 Å for other four aromatic residues in 2P101A. For 2P112A, these
displacements are 2.63 Å for Y115 and range from 1.28 A to 2.05 Å for
other four aromatic resides. The relatively large relocations of many
TM 1 side chains produced by 2P112A is compatible with its disruption
of plasma membrane expression.
These results can be compared to those from modeling the effects of
proline mutations in other TM domains, including 1P87A, 5P272A, 8P401A,
11P528A, and 12P572A. Those mutations cause less atomic relocation than
2P112A (see Fig. 9
legend). 5P272A causes 0.05 A changes, whereas
1P87A, 8P401A, 11P528A, and 12P572A lead to average changes of 0.1–0.2
Å.
This type of modeling may identify interesting DAT residues. Mutation
1P87A causes its largest relocations for isoleucine 74 (side chain 0.02
Å), phenylalanine 76 (0.07 Å), D79 (0.05 Å), and isoleucine 74 (0.04
Å) whereas other TM1 residues are affected less than one-tenth as
much. These data agree with previous observations that D79 and F76
appear to be functionally important (16
, 31)
and suggest
further study of I74.
Proline mutations thus provide novel information about the structure/function relationships of the dopamine transporter, add to information about the selective and nonselective features of cocaine and dopamine recognition and dopamine translocation, and point to DAT domains that could be targeted by selective novel pharmacological agents such as competitive cocaine antagonists that could spare dopamine.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  N. Hussainzada, A. Khandewal, and P. W. Swaan Conformational Flexibility of Helix VI Is Essential for Substrate Permeation of the Human Apical Sodium-Dependent Bile Acid Transporter Mol. Pharmacol., February 1, 2008; 73(2): 305 - 313. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Kinclova-Zimmermannova, M. Zavrel, and H. Sychrova Identification of Conserved Prolyl Residue Important for Transport Activity and the Substrate Specificity Range of Yeast Plasma Membrane Na+/H+ Antiporters J. Biol. Chem., August 26, 2005; 280(34): 30638 - 30647. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. E. Martin and K. Kirk The Malaria Parasite's Chloroquine Resistance Transporter is a Member of the Drug/Metabolite Transporter Superfamily Mol. Biol. Evol., October 1, 2004; 21(10): 1938 - 1949. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Koike, G. Conseil, E. M. Leslie, R. G. Deeley, and S. P. C. Cole Identification of Proline Residues in the Core Cytoplasmic and Transmembrane Regions of Multidrug Resistance Protein 1 (MRP1/ABCC1) Important for Transport Function, Substrate Specificity, and Nucleotide Interactions J. Biol. Chem., March 26, 2004; 279(13): 12325 - 12336. [Abstract] [Full Text] [PDF]
|
|
