A 3D structure model of the melibiose permease
of Escherichia coli represents a distinctive fold
Mohammad S. Yousefa,b,1and Lan Guana,b,2
aDepartment of Cell Physiology and Molecular Biophysics,bCenter for Membrane Protein Research, Texas Tech University Health Sciences Center,
Lubbock, TX 79430
Edited by H. Ronald Kaback, University of California, Los Angeles, CA, and approved July 15, 2009 (received for review May 19, 2009)
The melibiose permease of Escherichia coli (MelB) catalyzes the
coupled stoichiometric symport of a galactoside with a cation
(either Na?, Li?, or H?), using free energy from the downhill
translocation of one cosubstrate to catalyze the accumulation of
the other. Here, we present a 3D structure model of MelB threaded
through a crystal structure of the lactose permease of E. coli (LacY),
manually adjusted, and energetically minimized. The model con-
tains 442 consecutive residues (?94% of the polypeptide), includ-
ing all 12 transmembrane helices and connecting loops, with no
steric clashes and superimposes well with the template structure.
The electrostatic surface potential calculated from the model is
typical for a membrane protein and exhibits a characteristic ring of
positive charges around the periphery of the cytoplasmic side. The
3D model indicates that MelB consists of two pseudosymmetrical
6-helix bundles lining an internal hydrophilic cavity, which faces
the cytoplasmic side of the membrane. Both sugar and cation
binding sites are proposed to lie within the internal cavity. The
model is consistent with numerous previous mutational, biochem-
ical/biophysical characterizations as well as low-resolution struc-
tural data. Thus, an alternating access mechanism with sequential
binding is discussed. The proposed overall fold of MelB is different
from the available crystal structures of other Na?-coupled trans-
porters, suggesting a distinctive fold for Na?symporters.
bioenergetics ? ligand binding ? MelB ? protein threading ?
representative of the glycoside-pentoside-hexuronide/cation
symporter family of membrane transporters (2). MelB utilizes
free energy released from the energetically downhill movement
of a cation (Na?, Li?, or H?), in response to an electrochemical
cation gradient, to drive the uphill stoichiometric accumulation
of a galactopyranoside (3–5). The type of the cotransported
cation depends on the stereostructure of the transported sugar
(6). ?-Galactopyranosides (melibiose, raffinose, and p-nitrophe-
nyl-?-galactoside) are cotransported with Na?, H?, or Li?,
whereas ?-galactopyranosides (lactose, methyl-1-?-D-galactopy-
ranoside, and p-nitrophenyl-?-galactoside) are cotransported
with Na?or Li?but not H?(6). In the absence of an electro-
chemical cation gradient, MelB catalyzes the reverse reaction,
using free energy from the downhill translocation of the sugar to
drive the stoichiometric transport of the coupled cation in either
direction across the membrane (5, 7, 8). Similar features of
transport exist in the lactose permease of E. coli (LacY) (9, 10),
the best-studied representative of the major facilitator super-
family (MFS) of membrane transporters (2). Both MelB and
LacY transport D-galactopyranosides, with similar substrate
specificity; however, sugar transport in LacY is coupled solely
with H?(9, 10).
MelB consists of 473 aa with ?65% apolar/hydrophobic residues
(1, 11), and the Met-1 residue does not appear by N-terminal
sequencing (12). The single polypeptide is responsible for the Na?,
elibiose permease of Escherichia coli (MelB), encoded by
the melB gene in the mel operon (1), is a well-studied
H?, or Li?/sugar symport (11). The membrane topology of MelB
has been determined by hydropathy and phoA-fusion analyses (11,
exhibits 12 transmembrane helices connected by hydrophilic loops
with both N- and C-termini located at the cytoplasmic side of the
membrane. A signature of ‘‘6 helices-middle loop-6 helices’’ is
observed in both permeases. A projection map of MelB obtained
by cryo-EM of 2D crystals at a resolution of 8 Å (15) confirms the
total number of transmembrane helices as well as the feature of
2-helix bundles lining a central cavity. Furthermore, a 3D EM map
at a resolution of 10 Å reveals that MelB exhibits a heart shape and
Five X-ray 3D crystal structures of LacY have been solved for
the wild-type protein (17) and a conformationally restricted
mutant (18, 19). All crystal structures show that LacY is
organized into two 6-helix bundles related by 2-fold pseudosym-
metry, which are separated by a large hydrophilic cavity open to
the cytoplasmic side, representing an inward-facing conforma-
tion. The side chains important for binding both sugar and H?
lie at the apex of the central cavity in the middle of the protein.
Similar overall folds have been observed for two other members
of MFS, namely, the glycerol-3-phosphate transporter (GlpT)
(20) and the postulated multidrug efflux pump (EmrD) (21) of
E. coli. Here, we report a 3D structure model of MelB obtained
by threading analysis using LacY crystal structure as the tem-
plate. The model suggests that MelB shares a similar overall fold
with MFS members and exhibits a distinctive fold for Na?
Results and Discussion
Structure Prediction. The full-length sequence of MelB was sub-
jected to threading analysis using several server-based tools as
described in Materials and Methods. Hit lists of potential tem-
plates were obtained from the FUGUE (22), LOOPP (23), and
Phyre (24) programs [supporting information (SI) Table S1].
The crystal structure of LacY was consistently selected as a top
candidate by all three programs. Among other templates, two
other MFS permeases (GlpT and EmrD) and three non-MFS
proteins [LeuTAa (Na?/alanine transporter, PDB ID 2a65),
Amt-1 (an ammonium transporter, PDB ID 2b2f), and CLC (an
H?/Cl?antiporter, PDB ID 1ots)] were also recognized with
Author contributions: M.S.Y. and L.G. designed research; M.S.Y. and L.G. performed
research; M.S.Y. and L.G. analyzed data; and M.S.Y. and L.G. wrote the paper.
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
Freely available online through the PNAS open access option.
1On leave from: Biophysics Department, Faculty of Science, Cairo University, Egypt.
2To whom correspondence should be addressed. E-mail: firstname.lastname@example.org.
This article contains supporting information online at www.pnas.org/cgi/content/full/
www.pnas.org?cgi?doi?10.1073?pnas.0905516106 PNAS ?
September 8, 2009 ?
vol. 106 ?
no. 36 ?
All 3D threading models of MelB were converted to mem-
brane topological representations (Fig. S1). The topological
S1 D and E) clearly disagree with the well-established topology
of MelB (Fig. 1). The conversion of the CLC-based model was
rather difficult because of the large number of irregular and
broken helices. Therefore, these three models are excluded from
the analysis. On the other hand, all topological profiles based on
the MFS templates (Fig. S1 A–C) are similar to the experimen-
tally determined topology of MelB (11, 13, 14), where the
characteristic signature of 6 helices-middle loop-6 helices is
present. LacY shares similar substrate specificity and overall
transport mechanism with MelB; hence, the LacY-based model
was chosen for further systematic evaluation.
Careful examination of the threading model shows that the
register of the residues in helix X needed to be modified by half
a turn so that both hydrophilic and hydrophobic faces of the helix
orient properly. Therefore, manual adjustments, followed by
energy minimization, were performed as described in Materials
Alignment of 3D Structures. The primary sequence alignment
between MelB and LacY is relatively poor with ?37% sequence
MFS (2). However, it is remarkable that a secondary structure-
weighted alignment yields an excellent match between the
predicted secondary structure of MelB and the crystal structure
of LacY, as observed by the LOOPP program (Fig. S2). Subse-
quently, functionally important residues align well between the
two permeases as discussed below. The poor outcome of align-
presence of four longer extramembrane loops on both ends of
helices V and XI in MelB (Figs. S2 and S5).
The backbone model (FUGUE program) and the all-atom
model (LOOPP program) suggest that MelB adopts an overall
fold resembling that of LacY (Fig. 2A). The all-atom model of
MelB includes ?94% of the residues (positions 6–448) covering
all 12 transmembrane helices and connecting loops with accept-
able stereochemistry and no steric clashes. A Ramachandran
plot shows that 80% of the residues concentrate in the ?-helical
region and only 1% of the total residues are located in the
stereochemically disallowed area. These outliers are mainly
located in the flexible extramembrane loops. Furthermore,
superposition of the main chain atoms of the MelB model and
LacY structure, restricting the 3D alignment to the correspond-
ing transmembrane helices (Fig. 2A), yields a rmsd of only 0.5
Å. These results indicate that MelB can adopt an inward-facing
conformation similar to that of LacY. In addition, threading
models based on GlpT and EmrD exhibit arrangement of helices
similar to that of LacY but with different overall conformations.
threading models of four MelB orthologues from Salmonella
typhimurium, Citrobacter freundii, Klebsiella pneumoniae, and
Vibrio shilonii yield rmsd values of less than 1.5 Å. Moreover,
evolutionary conservation analyses were performed using the
patterns are mapped on the MelB model (Fig. S3A) and LacY
structure (Fig. S3B), the most conserved residues are located at
the interhelix interfaces within transmembrane regions, whereas
the most variable residues are located in the periphery (Fig. S3).
Overall Architecture. MelB, like LacY (Fig. 2 A and C), is heart
shaped when viewed parallel to the membrane and oval shaped
when viewed from the cytoplasmic side (Fig. 2 A and B). The
overall shape of the MelB model is consistent with previous
structural results based on 2D crystallographic projections and
3D EM maps (15, 16). The calculated electrostatic surface
potential of the MelB model reveals a hydrophobic transmem-
brane region and charged/polar extramembrane loops (Fig. 2B).
In both permeases, it is interesting that a positively charged belt,
like ‘‘blue lips,’’ is clearly observed around the periphery of the
cytoplasmic opening, with negative charges concentrated at both
the top and bottom surfaces of the proteins (Fig. 2 B and C). The
asymmetrical surface charge distribution in both permeases
supports the positive inside rule (25), which is important for
controlling membrane topology of the protein, and extends the
rule to the tertiary structure level. Moreover, the blue lips at the
cytoplasmic opening are expected to facilitate interactions with
the phosphate oxygens of phospholipids and might also imply a
preferential association with anionic phospholipids at the inner
leaflet of the membrane. Additionally, the negatively charged
presented as blue cylinders; the light-blue color indicates extramembrane segments. Helices are numbered with Roman numerals. Charged residues are shown
in red (negative) and blue (positive). Eight cavity-exposed charged residues from transmembrane regions are marked as filled square boxes. Residues known to
participate in the sugar binding are highlighted with a pink background. Residues expected to be involved in cation binding are highlighted with a yellow
background (see the text). Residues important for substrate(s) binding and/or coupling are highlighted with a blue background. The Met-1 residue is removed,
as suggested by N-terminal sequencing (12). The threaded region is between residues Thr-6 and Leu-448, as indicated by black lines.
Membrane topology of MelB. The topology model of MelB (11) is matched to the threading 3D structure (see the text). Transmembrane helices are
www.pnas.org?cgi?doi?10.1073?pnas.0905516106Yousef and Guan
surfaces at the top and bottom sides may contribute to the local
Domain Structure. MelB is organized in 2-helix bundles connected
with a central loop and separated by an internal cavity facing the
cytoplasmic side. Both N- and C-termini are located at the
cytoplasmic side of the protein (Fig. 2A). The N- and C-helix
bundles are related by a 2-fold pseudosymmetry. Within each
domain, there is another 2-fold inverted pseudosymmetry. The
primary sequence alignment between the N- and C-terminal
halves shows significant homology of 57.8% (17.5% identity),
with a nearly perfect match in the secondary structure (Fig. S4).
A pseudosymmetry between the two 6-helix bundles is evident
in the crystal structures of LacY. However, the homology
between the two halves is ?57.4%, with only ?14.2% identity
(Fig. S4). Thus, the threading structure is consistent with the
notion that MelB, like other MFS members, contains two
internal tandem repeats from the same genetic origin (2). The
pseudosymmetry between the two domains is not observed in the
low-resolution 2D projection map (15). Possible interpretations
would be the presence of several irregular or tilted helices and/or a
tilt in the pseudosymmetrical axis relative to the membrane plane.
Conserved Hydrophobic Interactions. Both the N- and C-helices
bundles contain conserved hydrophobic patches. Three Phe
residues in positions 17 and 21 (helix I), as well as position 146
(IV) in LacY, contribute to a hydrophobic patch, which seems
to be involved in stabilizing the N-terminal domain (Fig. 3B).
The corresponding positions in MelB are all Phe residues in
positions 16, 20 (helix I), and 151 (helix IV), respectively (Fig.
3A). In addition, this characteristic packing seems to play a role
in the optimal orientation of a conserved Arg residue, Arg-149
in MelB (26) and Arg-144 in LacY, which is important for sugar
binding (9) (Fig. 3). Furthermore, residues Phe-49 (helix II);
Leu-222, Leu-225, Phe-246, and Phe-247 (helix VII); Val-315
(helix X); and Leu-385 (helix XII) of LacY contribute to the
packing around helix VII in the C-terminal helices bundle (Fig.
3B). In MelB, residues Phe-48 (helix II); Leu-233, Leu-236,
Tyr255, and Phe-256 (helix VII); Val-328 (helix X); and Leu-417
(helix XII) occupy similar positions (Fig. 3A). The well-
conserved helical-packing environments support the predicted
overall arrangement of helices in the model.
Characteristic Cytoplasmic Loops of MelB. The presence of four
distinguishable loops is observed in the 3D model of MelB (Fig.
S5). These loops are significantly longer than the corresponding
regions in LacY. The length and location of these loops are
highly conserved across all MelB orthologues. Loops 4–5/5–6
and 10–11/11–12 are located on both sides of the corresponding
helices V and XI. It is remarkable that more than two-thirds of
the cytoplasmic loops 4–5 and 10–11 contain charged/polar
residues, (Fig. 1) most of which are highly conserved (Fig. S3).
Loop 4–5 contains two residues, Arg-141 and Glu-142, that are
essential for sugar/cation translocation (26, 27). In the model,
Arg-141 is in close proximity to the important residue Asp-120
of helix IV. In loop 10–11, residues Asp-351, Asp-354, and
6–448, blue), obtained from the LOOPP program, are superimposed on the LacY crystal structure (green, PDB ID 1pv6). The alignment was restricted to the
corresponding transmembrane helices. (B and C) Surface electrostatic potential maps of the MelB threading model and the crystal structure of LacY were
calculated using Adaptive Poisson-Boltzmann Solver (APBS) software (53). The scale indicates color-coded values of the electrostatic potentials (kT/e). A positive
electrostatic potential is noticeable around the cytoplasmic opening of the central cavity in both permeases. Front and back, viewing parallel to the membrane;
top, viewing from the cytoplasmic side; bottom, viewing from the periplasmic side.
Comparison between the threading model of MelB and the crystal structure of LacY. (A) Superposition. The main chain coordinates of MelB (residues
Yousef and Guan PNAS ?
September 8, 2009 ?
vol. 106 ?
no. 36 ?
and the model may only suggest one possible orientation.
Nevertheless, it is highly likely that rearrangements of loops 4–5
and 10–11 may play important role(s) in ligand recognition
Central Hydrophilic Cavity with Asymmetrical Charge Distribution.
Helices I, II, IV, and V (N-terminal domain) as well as helices
VII, VIII, X, and XI (C-terminal domain) line the central cavity
in the MelB model (Fig. 2A). The cavity is closed to the
periplasmic side by the helical contacts between helices V/VIII,
I/VII, and II/XI and is open to the cytoplasmic side. Based on the
model, it is possible that interactions between or involving loops
4–5 and 10–11 might cover the cytoplasmic opening of the
cavity. Remarkably, all helices in the inner layer (I, IV, VII, and
X) tilt and have kinks near the cavity as observed in LacY
(17–19), which may suggest unsynchronized movements of these
Unlike LacY, a clear asymmetrical distribution of charges is
observed within the cavity of MelB. Seven of eight charged
residues exposed to the cavity are from the N-terminal helices
(Figs. 1 and 4), and the only C-terminal–charged residue (Lys-
377, helix XI) is in close proximity to the N-terminal helices
bundle. In LacY, charged residues involved in sugar and H?
bindings are mainly located at the N- and C-terminal helices of
the cavity, respectively. Residues Asp-240 (helix VII); Glu-269
(helix VIII); Arg-302 (helix IX); and Lys-319, His-322, and
Glu-325 (helix X) in LacY form the charged/H-bond network
responsible for the H?binding and translocation (Fig. 4B) (29).
The equivalent positions in MelB are mainly occupied with polar
residues with no charged side chains.
In addition, apolar/hydrophobic residues dominate the
periplasmic half of the MelB molecule; even the periplasmic
external loops are not as charged as the cytoplasmic loops (Fig.
1). In the central cavity, the majority of the charged residues are
interactions at the periplasmic side and the charged environment
at the cytoplasmic side and inner surface of the cavity may
stabilize an inward-facing conformation, which is predicted to be
the most populated conformer of MelB, similar to what is
established for LacY (17, 30, 31).
helices I, V, and VI in the N-terminal domain and around helix VII in the
C-terminal domain of the MelB threading model (A, blue) and LacY crystal
structure (B, green).
Conserved helical packing. Hydrophobic patches (gray) between
cavity of MelB (A, blue) and LacY (B, green). The N- and C-terminal helices of MelB and LacY are shown in blue/light-blue and green/light-green colors,
respectively. Residues important for cosubstrate binding are shown as sticks. The large spheres reflect postulated positions of the ligands. Dotted lines show
possible interactions. (A) In the threading model of MelB, a melibiose molecule and a Na?ion are manually docked in putative binding sites. The modeling of
and Gly-117) are colored yellow. The functionally important loop 4–5 is shown in gray, with 2 important positions (Arg-141 and Glu-142) highlighted. Residues
colored in blue are important for substrate(s) binding and/or coupling (see the text). The Arg-149 is important for sugar binding (26) and is colored pink. In LacY
(B, PDB 1pv7), residues that are essential or important for the binding of sugar and H3O?are shown in pink and magenta, respectively. A ß-D-galactopyranosyl
1-thio-ß-D-galactopyranoside (TDG) molecule is shown in green, and manually docked H3O?is shown as an enlarged brown sphere.
The central cavity in the MelB model containing putative binding sites for Na?and sugar. Identical views from the cytoplasmic side down the central
www.pnas.org?cgi?doi?10.1073?pnas.0905516106Yousef and Guan
Putative Na?Binding Site. Among eight charged residues in the
cytoplasmic cavity, four Asp residues in positions 19 (helix I), 55
and 59 (helix II), and 124 (helix IV) have been identified to
contribute to the Na?-dependent sugar binding and transport in
MelB (32–37). Cys or Asn mutations at either Asp-55 or Asp-59
abolish completely and specifically the Na?-dependent increase
in both sugar affinity and stimulation of transport (35, 37).
Unlike the D59C mutant that uncouples the cations (Na?, Li?,
the H?-coupled melibiose transport (37). It is implied by atten-
uated total reflection Fourier transform infrared spectroscopy
studies that some carboxyl groups may directly interact with the
Na?ion in MelB (38). Moreover, it was concluded that Trp-54,
adjacent to Asp-55, is the major contributor to the Na?binding-
induced Trp fluorescence changes (39). Based on these obser-
vations, one Na?ion was manually docked with no steric clashes
between the carboxyl groups of the conserved Asp-55 and
Asp-59 (helix II) (Fig. 4A). The backbone carbonyl oxygen of the
conserved Gly-117 (helix IV) is also in close proximity to the
docked Na?ion. It is notable that the available crystal structures
of Na?-coupled symporters (40–43) reveal a common Na?
binding motif, wherein the Na?ions are buried between un-
wound regions of two helices, which is different from the Na?
binding site suggested for MelB. On the other hand, Asp-19
(helix I) and Asp-124 (helix IV) are not in close proximity to the
proposed Na?binding pocket in the model. Cys replacements at
these two positions affect both sugar- and Na?binding affinities.
It is likely that these residues contribute to the charged/polar
interactions around helix IV through the pairs Lys-18/Thr-119
(helices I/IV) Asp-124/Arg-141 (helix IV/loops 4–5), Arg-52/
Asn-248 (helices II/VII), Arg-52/Asp-55 (helix II) (44), and
Asp-59/Lys-377 (helices II/XI). Consistently, second-site muta-
tion analyses suggest close proximities between the pairs Lys-
18/Met-123 (1 helical turn below Thr-119) (44), Arg-52/Asn-248
(44), and Lys-377/Asp-59 (45). It is likely that the interhelix-
in optimizing the cation binding site and sugar recognition as
well as the coupling between sugar and H?translocation.
Residues Involved in Sugar Binding. Although both MelB and LacY
share similar substrate specificities, there is no detectable match
of residues important for sugar binding in the traditional align-
ment. In LacY, the salt bridge between Arg-144 (helix V) and
Glu-126 (helix IV) is essential for sugar binding and selectivity
(9, 10). In MelB, Arg-149 (helix V) is important for sugar
recognition (26), and Asp-124 and Tyr-120 (helix IV) are also
suggested to be involved in both sugar and/or cation coupling
(36). It is remarkable that the two pairs Arg-144/Glu-126 and
Arg-149/Asp-124, which are strictly conserved among all LacY
and MelB orthologues, respectively, align well at the tertiary
structure level (Fig. S2 and Fig. 4). Trp-151 of LacY stacks
hydrophobically against the galactopyranosyl ring of the sugar
and plays a role in sugar recognition (46). In MelB, there is no
Trp residue at the corresponding position, and Tyr-120 (36) is at
a position corresponding to Ala-122 of LacY, which abuts the
nongalactosyl moiety of D-galactosides (10). Phe-20 (helix I) of
LacY is in close proximity to the sugar binding pocket (17, 47),
and the side chain of the corresponding residue Asp-19 of MelB
could be within a salt-bridge distance with Arg-149. Arg-52, on
the same face of helix II as Asp-55 and Asp-59, is critical for
H?-coupled melibiose transport but is not essential for Na?-
coupled melibiose transport (44).
One molecule of melibiose was manually docked in the
internal cavity of the MelB model (Fig. 4A), guided by the
coordinates of ß-D-galactopyranosyl 1-thio-ß-D-galactopyrano-
side in the LacY crystal structure (Fig. 4B) (18). The docked
melibiose exhibits no steric clashes within the proposed binding
pocket and is surrounded to within 4 Å by potential hydrogen-
bonding partners: Arg-149 (helix V), Arg-52 (helix II), and
no sugar transport (45).
Although it is not clear whether Arg-149 in MelB confers
sugar specificity similar to the essential residue Arg-144 in LacY,
it is striking that the two residues occupy almost identical
positions (Fig. 3 A and B and Fig. 4). Based on the model, the
aromatic residues Tyr-26 (helix I) and Tyr-120 (helix IV) may
interact hydrophobically with the galactopyranosyl rings. More-
over, the sugar seems to be in close proximity to loop 4–5, which
is consistent with a previous photoaffinity labeling study (48).
The cooperative binding of the cosubstrates has been established
for MelB (26, 35, 39, 49). It is interesting that both binding sites
appear to be in close proximity in the model. It is also interesting
that precedence has been shown in LeuTAa(40), wherein the
substrate participates in the Na?coordination.
In the MelB model, helix IV is in the middle of the N-terminal
helices bundle with a kink at the center (Figs. 2A and 4A). Both
cation and sugar binding sites are near the hinge on the adjacent
faces of the helix, which implies a physical role in the coupling
between the two cosubstrates (39). The charged/H-bond net-
work observed between helices IV, I, and V, loops 4–5, as well
as in helices II and XI could account for the cooperative binding
of the cation and sugar in MelB (35, 36).
Postulated Mechanism. Combined with other studies (27, 50), the
alternating access mechanism can be postulated for the galac-
toside/Na?symport in MelB (i.e., both cation and sugar binding
sites are reciprocally exposed to either side of the membrane
during turnover). The symport mechanism for the efflux mode
can be explained by a simplified 6-step scheme similar to that
proposed for LacY (10) (Fig. 5). In the reversible and sequential
binding model, the Na?ion binds initially (step 1) and the sugar
then binds at the inner surface (step 2). Binding of the cosub-
strates in the inward-facing conformation causes a transition to
the outward-facing conformation (step 3). The sugar releases
initially (step 4), and the Na?ion then releases at the outer
surface (step 5). The empty carrier returns to the inward-facing
influx mode. Thus, it is likely that the turnover occurs when both
MelB. A cross section of the membrane is shown as a gray rectangle.
A kinetic scheme of the efflux mode of galactoside/Na?symport for
Yousef and GuanPNAS ?
September 8, 2009 ?
vol. 106 ?
no. 36 ?
cosubstrate binding sites are concurrently occupied (step 3) or
unoccupied (step 6).
Several lines of evidence, including bioinformatics, previous
functional studies, and biochemical/biophysical characteriza-
the threading model. Two pseudosymmetrical 6-helix bundles
the binding sites for both cosubstrates. As suggested from the
model, the overall fold of the Na?-coupled symporter MelB is
different from the available structures of other Na?-coupled
transporters, including LeuTAa, the sodium/galactose symporter
(41), the Na?/aspartate symporter (43), and the Na?/H?anti-
porter (51), but is similar to MFS members. Therefore, the MelB
model may represent a unique fold for Na?-coupled permeases.
Although exact side-chain orientations/interactions in the model
to be similar to that of LacY; however, the cation binding site is
different between the two permeases.
Materials and Methods
The detailed methodology is provided in SI Text. In summary, using the
full-length primary sequence of MelB as the sole input, three threading
programs [FUGUE, (22); LOOPP, (23); and Phyre, (24)] were used for the
and evaluation of the stereochemistry. The electrostatic surfaces were calcu-
lated using Adaptive Poisson-Boltzmann Solver (APBS) software (53). The
evolutionary conservation analysis was performed using the Web-based
server ConSurf (54).
ACKNOWLEDGMENTS. The authors thank Ge ´rard Leblanc and H. Ronald
Kaback for their critical comments and insightful suggestions and Arathi
Krishnakumar and Richa Chandra for their critical reading of the manu-
script. This work is supported by Award number R21HL087895 from the
National Heart, Lung, and Blood Institute (to L.G.) and Center for Mem-
brane Protein Research, the Texas Tech University Health Sciences Center.
The coordinates of the MelB model are available in SI Appendix.
1. Yazyu H, et al. (1984) Nucleotide sequence of the melB gene and characteristics of
deduced amino acid sequence of the melibiose carrier in Escherichia coli. J Biol Chem
3. Tokuda H, Kaback HR (1977) Sodium-dependent methyl 1-thio-?-D-galactopyranoside
transport in membrane vesicles isolated from Salmonella typhimurium. Biochemistry
4. Tsuchiya T, Raven J, Wilson TH (1977) Co-transport of Na?and methul-beta-D-
thiogalactopyranoside mediated by the melibiose transport system of Escherichia coli.
Biochem Biophys Res Commun 76:26–31.
5. Bassilana M, Pourcher T, Leblanc G (1987) Facilitated diffusion properties of melibiose
permease in Escherichia coli membrane vesicles. Release of co-substrates is rate
limiting for permease cycling. J Biol Chem 262:16865–16870.
6. Wilson DM, Wilson TH (1987) Cation specificity for sugar substrates of the melibiose
carrier in Escherichia coli. Biochim Biophys Acta 904:191–200.
7. Tsuchiya T, Wilson TH (1978) Cation-sugar cotransport in the melibiose transport
system of Escherichia coli. Membr Biochem 2:63–79.
8. Bassilana M, Pourcher T, Leblanc G (1988) Melibiose permease of Escherichia coli.
Characteristics of co-substrates release during facilitated diffusion reactions. J Biol
9. Kaback HR, Sahin-Toth M, Weinglass AB (2001) The kamikaze approach to membrane
transport. Nat Rev Mol Cell Biol 2:610–620.
10. Guan L, Kaback HR (2006) Lessons from lactose permease. Annu Rev Biophys Biomol
11. Pourcher T, Bibi E, Kaback HR, Leblanc G (1996) Membrane topology of the melibiose
permease of Escherichia coli studied by melB-phoA fusion analysis. Biochemistry
coli: Large scale purification and evidence that H?, Na?, and Li?sugar symport is
catalyzed by a single polypeptide. Biochemistry 34:4412–4420.
13. Botfield MC, Naguchi K, Tsuchiya T, Wilson TH (1992) Membrane topology of the
melibiose carrier of Escherichia coli. J Biol Chem 267:1818–1822.
14. Gwizdek C, Leblanc G, Bassilana M (1997) Proteolytic mapping and substrate protec-
tion of the Escherichia coli melibiose permease. Biochemistry 36:8522–8529.
15. HacksellI,etal.(2002)Projectionstructureat8A Åresolutionofthemelibiosepermease,
an Na-sugar co-transporter from Escherichia coli. EMBO J 21:3569–3574.
structure of the sugar symporter melibiose permease from cryo-electron microscopy.
J Struct Biol 152:76–83.
17. Guan L, Mirza O, Verner G, Iwata S, Kaback HR (2007) Structural determination of
wild-type lactose permease. Proc Natl Acad Sci USA 104:15294–15298.
18. Abramson J, et al. (2003) Structure and mechanism of the lactose permease of Esche-
richia coli. Science 301:610–615.
19. Mirza O, Guan L, Verner G, Iwata S, Kaback HR (2006) Structural evidence for induced
fit and a mechanism for sugar/H?symport in LacY. EMBO J 25:1177–1183.
20. Huang Y, Lemieux MJ, Song J, Auer M, Wang DN (2003) Structure and mechanism of
the glycerol-3-phosphate transporter from Escherichia coli. Science 301:616–620.
21. Yin Y, He X, Szewczyk P, Nguyen T, Chang G (2006) Structure of the multidrug
transporter EmrD from Escherichia coli. Science 312:741–744.
22. Shi J, Blundell TL, Mizuguchi K (2001) FUGUE: Sequence-structure homology recogni-
alties. J Mol Biol 310:243–257.
23. Teodorescu O, Galor T, Pillardy J, Elber R (2004) Enriching the sequence substitution
matrix by structural information. Proteins 54:41–48.
24. Kelley LA, Sternberg MJ (2009) Protein structure prediction on the Web: A case study
using the Phyre server. Nat Protoc 4:363–371.
25. von Heijne G (1989) Control of topology and mode of assembly of a polytopic
membrane protein by positively charged residue. Nature 341:456–458.
26. Abdel-Dayem M, Basquin C, Pourcher T, Cordat E, Leblanc G (2003) Cytoplasmic loop
in the process of Na?-coupled sugar translocation. J Biol Chem 278:1518–1524.
27. Meyer-Lipp K, et al. (2006) The inner interhelix loop 4–5 of the melibiose permease
from Escherichia coli takes part in conformational changes after sugar binding. J Biol
28. Ding PZ (2004) Loop X/XI, the largest cytoplasmic loop in the membrane-bound
melibiose carrier of Escherichia coli, is a functional re-entrant loop. Biochim Biophys
29. Kaback HR (2005) Structure and mechanism of the lactose permease. C R Biol 328:557–
30. Smirnova I, et al. (2007) Sugar binding induces an outward facing conformation of
LacY. Proc Natl Acad Sci USA 104:16504–16509.
31. Kaback HR, et al. (2007) Site-directed alkylation and the alternating access model for
LacY. Proc Natl Acad Sci USA 104:491–494.
32. Matsuzaki S, Weissborn AC, Tamai E, Tsuchiya T, Wilson TH (1999) Melibiose carrier of
Escherichia coli: Use of cysteine mutagenesis to identify the amino acids on the
hydrophilic face of transmembrane helix 2. Biochim Biophys Acta 1420:63–72.
33. Pourcher T, Deckert M, Bassilana M, Leblanc G (1991) Melibiose permease of Esche-
richia coli: Mutation of aspartic acid 55 in putative helix II abolishes activation of sugar
binding by Na? ions. Biochem Biophys Res Commun 178:1176–1181.
34. Hama H, Wilson TH (1994) Replacement of alanine 58 by asparagine enables the
melibiose carrier of Klebsiella pneumoniae to couple sugar transport to Na?. J Biol
35. Pourcher T, Zani ML, Leblanc G (1993) Mutagenesis of acidic residues in putative
membrane-spanning segments of the melibiose permease of Escherichia coli. I. Effect
on Na(?)-dependent transport and binding properties. J Biol Chem 268:3209–3215.
36. Zani ML, Pourcher T, Leblanc G (1994) Mutation of polar and charged residues in the
hydrophobic NH2-terminal domains of the melibiose permease of Escherichia coli.
J Biol Chem 269:24883–24889.
37. Zani ML, Pourcher T, Leblanc G (1993) Mutagenesis of acidic residues in putative
membrane-spanning segments of the melibiose permease of Escherichia coli. II. Effect
on cationic selectivity and coupling properties. J Biol Chem 268:3216–3221.
38. Leon X, Lemonnier R, Leblanc G, Padros E (2006) Changes in secondary structures and
39. Cordat E, Leblanc G, Mus-Veteau I (2000) Evidence for a role of helix IV in connecting
cation- and sugar-binding sites of Escherichia coli melibiose permease. Biochemistry
homologue of Na?/Cl?-dependent neurotransmitter transporters. Nature 437:215–
41. Faham S, et al. (2008) The crystal structure of a sodium galactose transporter reveals
mechanistic insights into Na?/sugar symport. Science 321:810–814.
42. Weyand S, et al. (2008) Structure and molecular mechanism of a nucleobase-cation-
symport-1 family transporter. Science 322:709–713.
43. Boudker O, Ryan RM, Yernool D, Shimamoto K, Gouaux E (2007) Coupling substrate
and ion binding to extracellular gate of a sodium-dependent aspartate transporter.
44. Franco PJ, Wilson TH (1999) Arg-52 in the melibiose carrier of Escherichia coli is
important for cation-coupled sugar transport and participates in an intrahelical salt
bridge. J Bacteriol 181:6377–6386.
helices II and XI in the melibiose carrier of Escherichia coli. Biochim Biophys Acta
46. Guan L, Hu Y, Kaback HR (2003) Aromatic stacking in the sugar binding site of the
lactose permease. Biochemistry 42:1377–1382.
47. Kasho VN, Smirnova IN, Kaback HR (2006) Sequence alignment and homology thread-
ing reveals prokaryotic and eukaryotic proteins similar to lactose permease. J Mol Biol
48. Ambroise Y, Leblanc G, Rousseau B (2000) Active-site-directed photolabeling of the
melibiose permease of Escherichia coli. Biochemistry 39:1338–1345.
49. Mus-Veteau I, Leblanc G (1996) Melibiose permease of Escherichia coli: Structural
organization of cosubstrate binding sites as deduced from tryptophan fluorescence
analyses. Biochemistry 35:12053–12060.
charge translocation in the melibiose permease from Escherichia coli. Biochemistry
51. Hunte C, et al. (2005) Structure of a Na?/H?antiporter and insights into mechanism of
action and regulation by pH. Nature 435:1197–1202.
52. Collaborative Computational Project N (1994) The CCP4 suite: Programs for protein
crystallography. Acta Crystallogr D 50:760–763.
53. Baker NA, Sept D, Joseph S, Holst MJ, McCammon JA (2001) Electrostatics of nanosys-
54. Landau M, et al. (2005) ConSurf 2005: The projection of evolutionary conservation
scores of residues on protein structures. Nucleic Acids Res 33:W299–W302.
www.pnas.org?cgi?doi?10.1073?pnas.0905516106Yousef and Guan