In vitro functional characterization of BtuCD-F, the Escherichia coli ABC transporter for vitamin B12 uptake.

Abstract

BtuCD is an ATP binding cassette (ABC) transporter that facilitates uptake of vitamin B(12) into the cytoplasm of Escherichia coli. The crystal structures of BtuCD and its cognate periplasmic binding protein BtuF have been recently determined. We have now explored BtuCD-F function in vitro, both in proteoliposomes and in various detergents. BtuCD reconstituted into proteoliposomes has a significant basal ATP hydrolysis rate that is stimulated by addition of BtuF and inhibited by sodium ortho-vanadate. When using different detergents to solubilize BtuCD, the basal ATP hydrolysis rate, the ability of BtuF to stimulate hydrolysis, and the extent to which sodium ortho-vanadate inhibits ATP hydrolysis all vary significantly. Reconstituted BtuCD can mediate transport of vitamin B(12) against a concentration gradient when coupled to ATP hydrolysis by BtuD in the liposome lumen and BtuF outside the liposomes. These in vitro studies establish the functional competence of the BtuCD and BtuF preparations used in the crystallographic analyses for both ATPase and transport activities. Furthermore, the tight binding of BtuF to BtuCD under the conditions studied suggests that the binding protein may not dissociate from the transporter during the catalytic cycle, which may be relevant to the mechanisms of other ABC transporter systems.

Full text

Page 1

In Vitro Functional Characterization of BtuCD-F, the Escherichia coli ABC
Transporter for Vitamin B12Uptake†
Elizabeth L. Borths,‡Bert Poolman,§Rikki N. Hvorup,|Kaspar P. Locher,*,|and Douglas C. Rees*,‡,⊥
DiVision of Chemistry and Chemical Engineering, Howard Hughes Medical Institute, MC 114-96, California Institute of
Technology, Pasadena, California 91125, Department of Biochemistry, Groningen Biomolecular Science and Biotechnology
Institute, UniVersity of Groningen, Nijenborgh 4, 9747 AG Groningen, The Netherlands, and Institute of Molecular Biology and
Biophysics, Swiss Federal Institute of Technology, ETH Zurich, CH-8093, Zurich, Switzerland
ReceiVed July 8, 2005; ReVised Manuscript ReceiVed October 4, 2005
ABSTRACT: BtuCD is an ATP binding cassette (ABC) transporter that facilitates uptake of vitamin B12
into the cytoplasm of Escherichia coli. The crystal structures of BtuCD and its cognate periplasmic binding
protein BtuF have been recently determined. We have now explored BtuCD-F function in vitro, both in
proteoliposomes and in various detergents. BtuCD reconstituted into proteoliposomes has a significant
basal ATP hydrolysis rate that is stimulated by addition of BtuF and inhibited by sodium ortho-vanadate.
When using different detergents to solubilize BtuCD, the basal ATP hydrolysis rate, the ability of BtuF
to stimulate hydrolysis, and the extent to which sodium ortho-vanadate inhibits ATP hydrolysis all vary
significantly. Reconstituted BtuCD can mediate transport of vitamin B12against a concentration gradient
when coupled to ATP hydrolysis by BtuD in the liposome lumen and BtuF outside the liposomes. These
in vitro studies establish the functional competence of the BtuCD and BtuF preparations used in the
crystallographic analyses for both ATPase and transport activities. Furthermore, the tight binding of BtuF
to BtuCD under the conditions studied suggests that the binding protein may not dissociate from the
transporter during the catalytic cycle, which may be relevant to the mechanisms of other ABC transporter
systems.
ABC1transporters constitute a large family of membrane
transport proteins that use energy from the binding and
hydrolysis of ATP to pump substrates into or out of the
cytoplasm, often against a concentration gradient. ABC
transporters are found in bacteria, archaea, plants, and
animals, where they facilitate nutrient uptake, osmotic
regulation, antigen processing, and toxin/drug export (1-
4). Clinically relevant examples are implicated in at least
thirteen different genetic diseases in humans (5) or contribute
to resistance of tumor cells to chemotherapeutic agents (3)
and multi-drug resistance in bacteria (6).
ABC transporters consist of two membrane spanning
domains (MSDs), that form a substrate translocation pathway,
and two attached, cytoplasmic ATP-binding cassette (ABC)
domains that bind and hydrolyze nucleotide (1, 7). Bacterial
ABC importers also require an extracellular substrate binding
protein (SBP) that specifically recognizes and delivers the
substrate to the translocation channel (8). Transported
substrates range in size from single ions to polypeptides and
include hydrophobic drugs as well as water-soluble nutrients,
a diversity reflected in the sequences and architectures of
MSDs (9). In contrast, the ABC domains feature several
highly conserved sequence motifs, including the P loop
(Walker-A), Walker-B, Q loop, and ABC signature motif,
that bind and hydrolyze ATP (10).
At present, the only crystal structure of a bacterial ABC
import system is for BtuCD, the vitamin B12transporter of
Escherichia coli (11). A role for the membrane-spanning
BtuC subunit in B12 transport across the cytoplasmic
membrane was elucidated (12) before both BtuC and the
ABC subunit BtuD were cloned (13). After the B12binding
protein BtuF from Salmonella typhimurium was identified
(14), the E. coli homologue was cloned, purified, and
determined to bind B12with an affinity of ∼15 nM (15).
We have previously determined the crystal structures of
E. coli BtuCD and BtuF (11, 16); structures of BtuF in the
presence and absence of vitamin B12 have also been
determined by Hunt and co-workers (17). In this study, we
set out to characterize the function of the BtuCD-F transport
system in vitro. ATP hydrolysis rates were analyzed both in
detergent solution and in reconstituted lipid vesicles, and
ATP-powered B12 transport was monitored in proteolipo-
somes. While our proteoliposome assays more closely reflect
the in vivo situation, the studies in detergent solution assess
†This research was supported in part by the Swiss National Science
Foundation, Grant No. 3100A0-103765 (K.P.L.), a Fulbright grant from
NACEE to support the stay of B.P. in the laboratory of D.C.R., and
the Howard Hughes Medical Institute (D.C.R.).
* To whom correspondence
phone,
+41-44-633-3991; fax,
locher@mol.biol.ethz.ch.D.C.R.:
1-626-744-9524; e-mail, dcrees@caltech.edu.
‡California Institute of Technology.
§University of Groningen.
|ETH Zurich.
⊥Howard Hughes Medical Institute.
1Abbreviations: ABC, ATP binding cassette; ARS, ATP regenerat-
ing system; DDM, dodecyl maltoside; FOS12, Fos-choline 12; LDAO,
lauryl dimethylamine-N-oxide; MSD, membrane spanning domain;
SBP, substrate binding protein; TNM, (Tris-NaCl-MgCl2buffer); TX-
100, Triton X-100.
should beaddressed. K.P.L.:
e-mail,
fax,
+41-44-633-1182;
phone,1-626-395-8393;
16301
Biochemistry 2005, 44, 16301-16309
10.1021/bi0513103 CCC: $30.25© 2005 American Chemical Society
Published on Web 11/15/2005

Page 2

protein activity under conditions similar to those used to
obtain three-dimensional crystals.
MATERIALS AND METHODS
Protein Expression and Purification. BtuCD was prepared
as described (11), and for reconstitution, the detergent was
exchanged on the nickel affinity column from 0.1% lauryl
dimethylamine-N-oxide (LDAO) to 0.14% Triton X-100
(TX-100). BtuF was expressed with an N-terminal decahis-
tidine tag, using the vector pET22b (Novagen) and E. coli
BL21(DE3) cells (Novagen). The protein was purified in the
absence of vitamin B12from periplasmic extract using Ni-
NTA affinity chromatography (Qiagen). After elution from
the nickel column, the buffer was exchanged to 20 mM Tris-
HCl pH 7.5, 250 mM NaCl, using a HiPrep desalting column
(Amersham Biosciences), and the protein was stored at 4
°C. For stability studies, the BtuCD-F complex was produced
as described earlier (16) and the detergent was exchanged
by gel filtration chromatography. All detergents were from
Anatrace (Maumee, Ohio) and were Anagrade, except TX-
100, which was purchased from Sigma.
Membrane Reconstitution of BtuCD. An overview of the
general methodologies employed in this study for the
reconstitution and functional analysis of ABC transporters
may be found in ref 18. Egg L-R-phosphatidylcholine and
E. coli polar lipid extract (Avanti Polar Lipids) at 20 mg/
mL in chloroform were combined in a 1:3 ratio (w/w).
Chloroform was removed by rotary evaporation, and residual
solvent was removed at reduced pressure (ca. 2 mmHg, 10
min). Dried lipids were hydrated in 50 mM Tris-HCl pH
7.5 at 20 mg/mL by incubation at room temperature with
periodic vortexing or stirring. The solution was sonicated
three times on ice (15 s on/45 s off) using a rod sonicator.
The hydrated lipid solution was frozen in liquid nitrogen and
stored at -80 °C. BtuCD was reconstituted essentially as
described for the OpuA transporter (19). The lipids were
thawed in a room temperature water bath, subjected to two
rounds of freezing and thawing, and extruded through a 400
nm polycarbonate membrane using a Mini-Extruder (Avanti
Polar Lipids). Protein, buffer, and detergent were added to
the mixture to final concentrations of 10 mg/mL lipids, 50
mM Tris-HCl pH 7.5, 0.14% TX-100, with BtuCD at a 1:50
ratio (w/w) of protein/lipids. The mixture was equilibrated
at room temperature for 30 min with gentle agitation.
BioBeads SM2 (BioRad) (40 mg/mL wet weight) were
added, and the solution was agitated at room temperature
for 15 min. Similar aliquots of BioBeads were added four
more times for the following incubation periods at 4 °C: 15
min, 30 min, overnight, and 60 min. BioBeads were then
removed by filtration through a disposable spin filter column.
The solution was diluted ∼5× with 50 mM Tris-HCl pH
7.5 and centrifuged at 150000g for 90 min in a TLA 100.3
rotor. The supernatant was removed and the lipids were
washed two more times, centrifuging as above for 15 min.
Proteoliposomes were finally resuspended in 50 mM Tris-
HCl pH 7.5 at 20 mg/mL lipids and were stored in liquid
nitrogen.
To incorporate components such as BtuF, vitamin B12, or
ATP-regenerating system (ARS) into the vesicle lumen,
proteoliposomes plus additives were mixed, frozen in liquid
nitrogen and thawed in a room temperature water bath. The
freeze-thaw cycle was repeated three times, after which the
proteoliposomes were made homogeneous by extrusion
through polycarbonate filters as specified below.
ATPase ActiVity Assays. ATPase activity in proteolipo-
somes was measured in 300 µL reactions containing 1.7 mg/
mL lipids (∼0.13 µM BtuCD), 50 mM Tris-HCl pH 7.5,
and 150 mM NaCl. 50 µM vitamin B12 and various
concentrations of BtuF were present during the freeze-thaw
cycles (see figures and figure captions). Reactions were
incubated in a 37 °C water bath for 3 min before adding 2
mM ATP and 10 mM MgCl2 (or MgSO4) to initiate the
reaction. 50 µL samples were removed at various time points
and were added to 50 µL of 12% SDS. Inorganic phosphate
was then assayed by the modified molybdate method (20).
ATPase activity in LDAO was measured in 300 µL
reactions containing 35 nM BtuCD, 50 mM Tris-HCl pH
7.5, 150 mM NaCl, and 0.1% LDAO. 50 µM vitamin B12, 2
mM sodium ortho-vanadate, and/or various concentrations
of BtuF were present as indicated in the figures and figure
captions. Reactions were incubated in a 37 °C water bath
for 3 min before 2 mM ATP and 10 mM MgCl2(or MgSO4)
were added to initiate the reaction. 50 µL samples were
removed at various times and were added to 50 µL of 12%
SDS. Inorganic phosphate was assayed as above. ATPase
activity in the detergent Fos-choline 12 (FOS12) was
measured as above, except that 0.1% FOS12 and 175 nM
BtuCD were used. ATPase activities in dodecyl maltoside
(DDM) and TX-100 were measured as above except that
0.1% DDM or 0.14% TX-100 and 70 nM BtuCD were used.
In all cases, the new detergent was added to BtuCD purified
in LDAO, with the result that LDAO was diluted to
subcritical micelle concentrations.
Vitamin B12 Transport Assays. To measure uptake of
vitamin B12, an ATP regenerating system (ARS) was added
to BtuCD-containing proteoliposomes in the presence or
absence of ATP (50 mM Tris-HCl pH 7.5, 150 mM NaCl,
5 mM MgCl2, 24 mM creatine phosphate, 2.4 mg/mL
creatine kinase, and 0 or 2 mM ATP). In one set of
experiments, the concentration of creatine phosphate in the
ARS was varied between 0 and 50 mM. This mixture was
frozen in liquid nitrogen and thawed in a room temperature
water bath three times to incorporate all components into
the vesicle lumen, and then extruded through a 400 nm
polycarbonate membrane using a Mini-Extruder (Avanti
Polar Lipids). The extruder was rinsed with ARS-containing
buffer, and the rinse solution was added to the proteolipo-
some mixture. The proteoliposome mixture was diluted with
TNM buffer (50 mM Tris-HCl pH 7.5, 150 mM NaCl, 5
mM MgCl2) and centrifuged in a TLA 100.3 rotor at 60 000
rpm for 15 min at 4 °C. Proteoliposomes were washed once,
resuspended in TNM buffer at 10 mg/mL lipid, and kept on
ice to limit ATP hydrolysis by BtuCD.57Co-labeled vitamin
B12(MPBiomed), unlabeled B12, and BtuF were premixed
at various concentrations in TNM buffer and incubated for
2 min at 37 °C before ARS-loaded proteoliposomes were
added to start the transport reaction. The final concentrations
of lipids and BtuCD in the transport reaction were 4 mg/mL
and ∼0.3 µM, respectively, in a total volume of 0.5 mL. 50
µL aliquots were removed at various times and diluted into
2 mL of cold stop buffer (50 mM Tris-HCl pH 7.5, 150 mM
NaCl, 8% PEG-6000, 100 µM unlabeled B12). Samples were
then filtered through 0.2 µm cellulose acetate filters using a
16302 Biochemistry, Vol. 44, No. 49, 2005
Borths et al.

Page 3

vacuum filter manifold. Filters were washed twice with 2
mL of cold stop buffer and placed in test tubes. The amount
of57Co-vitamin B12 retained on each filter was measured
using a γ counter. The concentration of vitamin B12inside
the vesicles was determined from this quantity, using a value
of 1 µL/mg phospholipid for the internal volume of vesicles
(21).
RESULTS
The experimental approaches used to characterize the
ATPase and transport activities of BtuCD are schematically
depicted in Figure 1. Transporters reconstituted into proteo-
liposomes may be inserted in either right-side-in or inside-
out orientations, depending on whether the ABC domains
are facing the inside or the outside of the vesicle, respec-
tively. The inside-out transporters may be used to monitor
ATPase activity with exogenously supplied nucleotide
(Figure 1A), while right-side-in transporters are correctly
oriented to assay the transport activity of vitamin B12into
the vesicle, when ATP is incorporated into the interior
(Figure 1B). Although detergent solubilized transporters
cannot be used to measure substrate transport, their ATPase
activity can be characterized (Figure 1C).
BtuCD Proteoliposomes. BtuCD containing proteolipo-
somes were formed by mixing together detergent-solubilized
BtuCD and detergent-destabilized liposomes, followed by
application of BioBeads to remove detergent. As judged by
the intensity of protein bands on SDS-PAGE, the recon-
stitution efficiency of the transporter was approximately 50%,
to give an overall incorporation level of 0.01 mg BtuCD/
mg phospholipid.
Stability of BtuCD-F Complex in Various Detergents. The
binding protein-transporter complex BtuCD-F was found to
be stable in the detergents FOS12, LDAO, DDM, and TX-
100 used for the ATPase assays. After mixing BtuCD and
BtuF, excess BtuF was removed by gel filtration chroma-
tography. The integrity of the complex was assayed by gel
filtration chromatography after storing the protein for several
days at either 4 °C or room temperature. A typical experiment
is shown in Figure 2. Similar results were obtained when
BtuCD and BtuF were incubated together with 2 mM
MgATP for 8 min at 37 °C prior to gel filtration (not shown),
indicating that the complex remains stable while hydrolyzing
ATP. Shorter chain detergents (alkyl chains of 10 carbons
FIGURE 1: Schematic representation of the in vitro assay setups used in the present study. (A) BtuCD reconstituted in liposomes for ATP
hydrolysis assays in the presence and absence of BtuF and vitamin B12. (B) Transport of radioactive vitamin B12into proteoliposomes. To
keep the concentration of ATP constant for prolonged periods of time, an ATP regeneration system (ARS) was entrapped in the vesicle
lumen. (C) ATP hydrolysis assays in detergent micelles.
FIGURE 2: Stability of BtuCD-F in various detergents. A repre-
sentative experiment is shown for dodecyl maltoside, but similar
results were obtained for all other detergents used in this study.
(A) BtuCD-F was generated in LDAO as described (16), and the
detergent was exchanged to DDM using two consecutive gel
filtration chromatography runs (Superdex 200, Amersham Bio-
sciences). The first run is shown by the dashed line and demon-
strates excess binding protein. The main peak was collected and
applied again to the same column (solid line). A single peak is
obtained that is stable for several days. (B) SDS-PAGE analysis
of the BtuCD-F peak obtained in (A).
Functional Characterization of BtuCD-F
Biochemistry, Vol. 44, No. 49, 2005 16303

Page 4

or less) were unable to maintain the structural integrity of
the transporter (not shown).
ATPase ActiVity in Proteoliposomes and Detergents. The
ATPase activity of BtuCD reconstituted in proteoliposomes
was tested under various conditions (Figures 3 and 4A). In
the absence of BtuF and vitamin B12, BtuCD had a basal
rate of hydrolysis of 180 nmol/min/mg (Figure 4A). The
ATPase activity was stimulated by BtuF, reaching 440 nmol/
min/mg at 20 µM BtuF (Figure 4A). The presence of vitamin
B12 had little effect on the ATPase activity of BtuCD in
proteoliposomes, regardless of how much BtuF was present
(Figure 3).
ATP hydrolysis rates were also examined in four different
detergents: LDAO, a zwitterionic detergent used for the
crystal structure determination of BtuCD (11); TX-100,
which was used for reconstitution (see Materials and
Methods); DDM, the nonionic detergent most commonly
used to solubilize ABC transporters (18, 22); and the
phosphocholine containing FOS12. LDAO was present at a
subcritical micelle concentration with TX-100, DDM, and
FOS12 under all conditions, since concentrated BtuCD
purified in LDAO was diluted into the other detergents. It
was demonstrated for TX-100 that ATPase rates were
unaffected by residual LDAO by repeating the experiments
after complete removal of LDAO (not shown).
ATPase rates varied widely in the different detergents
(Figure 3). The highest rates were observed in LDAO, with
a basal activity of 980 nmol/min/mg. While BtuF was found
to moderately stimulate hydrolysis in LDAO (16% over the
basal rate, Figure 4A), the presence of vitamin B12had no
effect. The lowest ATP hydrolysis rates were observed in
FOS12 where the basal rate was 300 nmol/min/mg (Figure
3). In contrast to its stimulatory effect in proteoliposomes
or LDAO, BtuF depressed the ATPase rate in FOS12 over
4-fold compared to the basal rate (Figure 4A). ATPase rates
in DDM and TX-100 were intermediate between those found
in LDAO and FOS12, and BtuF and vitamin B12had different
effects (Figure 3). In both DDM and TX-100, addition of
apo-BtuF depressed the hydrolysis rate below the basal level
(Figure 3). However, this decrease was not observed in the
presence of vitamin B12. Indeed, when vitamin B12 was
present µM with BtuF in TX-100, hydrolysis was stimulated
over 50% (Figures 3 and 4A).
The ATPase activity of BtuCD was measured at various
ATP concentrations, both in proteoliposomes and in LDAO,
in the presence of BtuF and vitamin B12 (Figure 4B).
Although the Kmfor ATP could not be accurately determined
due to limitations of the assay, it was estimated to be below
50 µM, confirming that ATP was initially present at a
saturating concentration (2 mM) in the ATPase reactions.
Inhibition of ATPase ActiVity by Sodium Ortho-Vanadate.
The ability of 2 mM sodium ortho-vanadate to inhibit BtuCD
was tested in proteoliposomes and in each of the four
detergents in the presence of BtuF and vitamin B12(Figure
3). ATPase activity in proteoliposomes, DDM, and TX-100
was reduced by >97%. The ATPase activities in LDAO and
FOS12 were less sensitive to the vanadate, being reduced
by only 88% and 69%, respectively, under this condition.
Vitamin B12Uptake by BtuCD Proteoliposomes. Transport
by reconstituted BtuCD was monitored using radioactive,
57Co-labeled vitamin B12. To prevent fast depletion of internal
FIGURE 3: BtuCD ATPase activity in proteoliposomes (PLS) and
various detergents. Rates were measured at 2 mM ATP, using
various concentrations of BtuF as indicated by the numbers (0, 1,
10 (in µM)) under the bars, in the presence or absence of 50 µM
vitamin B12(+ or - under each bar) and using 2 mM sodium ortho-
vanadate for inhibition (labeled “van”). The inset illustrates the
specific hydrolysis of ATP over time by BtuCD proteoliposomes
measured in the presence of 2 mM ATP at 0 (b, 9), 1 ([, 2), or
10 (1, right triangle) µM BtuF, with (b, [, 1) or without (9, 2,
right triangle) 50 µM B12 and is representative of all ATPase
reaction time courses used to prepare this figure.
FIGURE 4: ATP hydrolysis rates at various ATP and substrate
concentrations. (A) ATPase rates as a function of BtuF concentra-
tion in proteoliposomes (PLS) and various detergents at 2 mM ATP
and 50 µM vitamin B12. (B) ATPase rates as a function of ATP
concentration at 50 µM vitamin B12and 1 µM BtuF for proteoli-
posomes (b) and LDAO (9).
16304 Biochemistry, Vol. 44, No. 49, 2005
Borths et al.

Page 5

ATP through the uncoupled hydrolysis catalyzed by BtuCD,
an ATP regenerating system (ARS) was incorporated within
the lumen of the proteoliposomes (Figure 1B). The proteo-
liposomes were then washed to remove external ATP and
uptake was measured after addition of BtuF and radioactive
B12 (Figure 5, closed circles). The B12 uptake rate during
the first 5 min of transport under these conditions (1 µM
BtuF, 5 µM B12) was 0.3 nmol/min/mg. This value was
somewhat variable between proteoliposome preparations and
has been measured as high as 1.0 nmol/min/mg. This may
reflect difficulties in accurately determining the protein
concentration in liposomes. When excess unlabeled B12was
added to a parallel reaction after 10 min (Figure 5, closed
squares), the uptake of labeled B12stopped. The amount of
radioactive substrate trapped within the proteoliposomes
remained constant, demonstrating that B12had indeed been
transported into the liposome lumen and that there was no
leakage or export.
Uptake of vitamin B12was found to be strictly dependent
on the presence of BtuF on the outside and ATP within the
lumen of the vesicles. When BtuF was present in the absence
of ATP, a small amount of B12became associated with the
proteoliposomes (Figure 5, open circles). This radioactivity
was not chased away by 100 µM unlabeled B12in the reaction
stop buffer (kept at 4 °C, see Materials and Methods).
However, upon addition of 100 µM unlabeled B12 to the
reaction and incubation at 37 °C, the amount of labeled B12
associated with the proteoliposomes decreased over a time
scale of several minutes (Figure 5, open squares). The
implications of this observation are discussed below.
The maximum amount of B12 uptake was measured at
various initial concentrations of B12(Table 1). At low B12
concentrations (5-100 nM), BtuCD-F was capable of
establishing a >300-fold concentration gradient between the
inside and outside of the proteoliposomes. At higher initial
concentrations of B12 (0.5-10 µM), this value decreased,
presumably due to a depletion of ATP and ATP-equivalents
and/or the buildup of inhibitory concentrations of ADP. To
determine the relationship between vitamin B12uptake and
the amount of nucleotide equivalents (ATP plus creatine
phosphate) in the proteoliposome interior, the extent of B12
transport by BtuCD-F was measured in the presence of 2
mM ATP and variable concentrations of creatine phosphate
ranging from 0 to 50 mM (Figure 6). As the creatine
phosphate concentration increased, the total amount of B12
transported into the proteoliposomes (estimated from the
amount of vitamin B12 accumulated at 120 min) also
increased. For creatine phosphate concentrations less than
10 mM, the total uptake of B12varied nearly linearly with
the sum of the concentrations of adenine nucleotides (ADP
+ ATP) and creatine phosphate initially present.
DISCUSSION
The protocol used in this study for reconstituting purified,
detergent-solubilized BtuCD is based upon that developed
by Rigaud and colleagues (23, 24). It differs from the original
method by the application of freeze-thaw-extrusion steps
to incorporate components (e.g. BtuF, ATP) into the vesicle
lumen. The method has been successfully used for the
functional reconstitution of a variety of transporters (18, 19,
25, 26). The combination of E. coli phospholipids and
chicken egg phosphatidylcholine is obviously not identical
to the native lipid composition of E. coli, and our proteoli-
FIGURE 5: Time course of vitamin B12 transport into BtuCD
proteoliposomes. (A) Uptake into BtuCD proteoliposomes was
measured as described in Materials and Methods using 5 µM57Co-
labeled vitamin B12and (b) 1 µM BtuF and 2 mM ATP; (9) 1 µM
BtuF, 2 mM ATP, and 100 µM unlabeled vitamin B12chase; (O)
1 µM BtuF; (0) 1 µM BtuF and 100 µM unlabeled vitamin B12
chase; ([) 2 mM ATP; (]) no addition. Each point represents the
average of 3 experiments. For clarity, error bars for only the two
upper curves are indicated. (B) Magnified view of the lower curves
of (A), with error bars, are depicted using the same symbols. In
addition, liposomes devoid of BtuCD were used at 5 µM57Co-
labeled vitamin B12and 1 µM BtuF (cross); this background of
∼1.2 nmol of B12binding per mg of protein has not been subtracted
from the uptake data.
Table 1: BtuCD-F Mediated Uptake at Various Initial
Concentrations of Vitamin B12a
initial [B12]
(µM)
uptake (pmol)
0.0051.4
0.01 2.9
0.1 27
0.5 120
1 170
10170
vitamin B12
[B12]in
(µM)
0.71
1.5
14
58
84
84
[B12]out
(µM)
0.0022
0.0042
0.045
0.27
0.67
9.7
[B12]in/[B12]out
330
350
300
210
130
8.7
aTransport into proteoliposomes was measured in the presence of
0.5 µM BtuF and various concentrations of57Co-labeled vitamin B12,
with an ATP-regenerating system incorporated in the proteoliposome
lumen. The concentration of phospholipids in the proteoliposomes was
4 mg/mL, in a total reaction volume of 0.5 mL. To calculate the final
vitamin B12concentration gradient, the internal volume of the proteo-
liposomes was estimated as 1 µL/mg lipids (21).
Functional Characterization of BtuCD-F
Biochemistry, Vol. 44, No. 49, 2005 16305