Download Channel-mediated lactic acid transport: a novel function for

Biochem. J. (2013) 454, 559–570 (Printed in Great Britain)
559
doi:10.1042/BJ20130388
Channel-mediated lactic acid transport: a novel function
for aquaglyceroporins in bacteria
Gerd P. BIENERT*1,2 , Benoı̂t DESGUIN*1 , François CHAUMONT*3 and Pascal HOLS*3
*Institut des Sciences de la Vie, Université catholique de Louvain, Croix du Sud 4-5, 1348 Louvain-la-Neuve, Belgium
MIPs (major intrinsic proteins), also known as aquaporins,
are membrane proteins that channel water and/or uncharged
solutes across membranes in all kingdoms of life. Considering
the enormous number of different bacteria on earth, functional
information on bacterial MIPs is scarce. In the present
study, six MIPs [glpF1 (glycerol facilitator 1)–glpF6] were
identified in the genome of the Gram-positive lactic acid
bacterium Lactobacillus plantarum. Heterologous expression
in Xenopus laevis oocytes revealed that GlpF2, GlpF3 and
GlpF4 each facilitated the transmembrane diffusion of water,
dihydroxyacetone and glycerol. As several lactic acid bacteria
have GlpFs in their lactate racemization operon (GlpF1/F4
phylogenetic group), their ability to transport this organic acid
was tested. Both GlpF1 and GlpF4 facilitated the diffusion of
D/L-lactic
INTRODUCTION
have more than two, with Pediococcus pentosaceus and Lc. lactis
having four and Lactobacillus brevis three [9]. The existence of
GLPs in these organisms may reflect their ability to use glycerol
or other substrates in the presence of a second carbon source, since
they cannot utilize glycerol alone [9]. However, the large number
of MIPs in lactic acid bacteria suggests functions other than water
and glycerol transport. Interestingly, a few MIPs have been shown
to transport lactic acid in humans (AQP9), Arabidopsis [NIP2;1
(Nodulin26-like intrinsic protein)] and trematodes (SmAQP) [10–
12], suggesting that MIPs can facilitate the diffusion of lactic acid
across membranes.
Lactic acid, a monocarboxylic acid, and the lactate anion are in
chemical equilibrium at a pK a of 3.86. Although the dissociated
(ionized) lactate is the predominant species, the nondissociated
(uncharged) lactic acid is always present at physiological pH.
Unlike lactate, which requires protein transmembrane transport
systems, lactic acid is able to diffuse through lipid bilayers
[13]. However, the high lactate concentrations measured in lactic
acid bacteria, such as Lactobacillus plantarum, show that some
membranes are rather poorly permeable to lactic acid [14]. This
implies that lactic acid transport may be regulated and that
this regulation is a major factor in determining the cellular
lactate concentration. Until now, no lactic acid channel has been
functionally characterized in lactic acid bacteria. Lactic acid
transport is often linked to energy production. The different
energy-producing systems include lactate–proton symport [15],
malolactic fermentation [16,17] and citrolactic fermentation [18];
in the last two cases, non-facilitated passive diffusion [19] of lactic
acid is conceivable in addition to protein-mediated transport [20].
MIPs (major intrinsic proteins) facilitate the movement of water
and non-ionic solutes across membranes and are required for
osmoregulation, water conductance, gas and nutrient uptake and
translocation, metalloid homoeostasis, and signal transduction
in eubacteria, archaea, fungi, plants and animals. MIPs have
been classified into two phylogenetically functional subgroups,
AQPs (aquaporins), which originally consisted of water-specific
channels, but now include channels shown to be permeable to
other small uncharged solutes, and aquaglyceroporins, or GLPs
(glycerol facilitators), which are permeable to glycerol and urea,
with some also being permeable to water and metalloids [1].
Although the substrate specificity and roles of many plant and
mammalian MIPs have been thoroughly studied [2], less is known
about the physiological functions of MIPs in micro-organisms.
Most of the known bacterial AQP-type isoforms have been identified in Gram-negative bacteria, whereas most of the GLP-type
sequences have been found in Gram-positive ones. This asymmetry might be related to the different structures and diffusion
properties of the membranes and cell walls in the two bacterial
groups [3].
The best studied microbial MIPs are GlpF and AqpZ
from Escherichia coli (EcGlpF and EcAqpZ) [4,5], GlpF from
Lactococcus lactis (Gla) [6] and AqpX from Brucella abortus [7].
Phylogenetic analysis suggested that osmoregulation in eubacteria
requires either one AQP and one GLP or a single GLP transporting
both water and glycerol [3]. Some bacteria do not have any MIPencoding genes in their genome [8], but some lactic acid bacteria
acid. Deletion of glpF1 and/or glpF4 in Lb. plantarum
showed that both genes were involved in the racemization of lactic
acid and, in addition, the double glpF1 glpF4 mutant showed a
growth delay under conditions of mild lactic acid stress. This
provides further evidence that GlpFs contribute to lactic acid
metabolism in this species. This lactic acid transport capacity was
shown to be conserved in the GlpF1/F4 group of Lactobacillales.
In conclusion, we have functionally analysed the largest set of
bacterial MIPs and demonstrated that the lactic acid membrane
permeability of bacteria can be regulated by aquaglyceroporins.
Key words: aquaporin, bacterial aquaglyceroporin, GlpF, lactic
acid transport, Lactobacillus plantarum.
Abbreviations used: AQP, aquaporin; At, Arabidopsis thaliana ; dak, dihydroxyacetone kinase; Ec, Escherichia coli ; GLP/GlpF, glycerol facilitator; Lar,
lactate racemase; Lp, Lactobacillus plantarum; Ls, Lb. sakei ; MIP, major intrinsic protein; MRS, de Man, Rogosa and Sharpe; NIP, Nodulin26-like intrinsic
protein; NPA, asparagine-proline-alanine; P f , osmotic water permeability coefficient; Pp, Pediococcus pentosaceus ; rAQP9, rat AQP9; YNB, yeast nitrogen
base; ZmPIP2;5, Zea mays plasma membrane intrinsic protein 2;5.
1
These authors contributed equally to this work.
2
Present address: Leibniz Institute of Plant Genetics and Crop Plant Research, 06466 Gatersleben, Germany.
3
Correspondence may be addressed to either of these authors (email [email protected] or [email protected]).
c The Authors Journal compilation c 2013 Biochemical Society
560
G. P. Bienert and others
In the present study, we cloned six MIPs from Lb. plantarum,
analysed their substrate specificity and characterized their
function, with a particular focus on GlpF1 and GlpF4, which were
shown to be lactic acid transporters. These are the first bacterial
MIPs demonstrated to transport lactic acid.
MATERIALS AND METHODS
prepared as described previously [24]. PCR was performed using
Phusion high-fidelity DNA polymerase (Finnzymes) or PfuTurbo
Cx Hotstart DNA polymerase (Stratagene) in a 2720 Thermal
Cycler (Applied Biosystems). The primers used in the present
study were purchased from Eurogentec or Eurofins MWG Operon
and are listed in Supplementary Table S2 (at http://www.biochemj.
org/bj/454/bj4540559add.htm).
Strains, plasmids and growth conditions
The strains and plasmids used in the present study are listed
in Supplementary Table S1 (at http://www.biochemj.org/bj/454/
bj4540559add.htm). Plasmids were constructed in E. coli TOP10.
Lb. plantarum was grown in MRS (de Man, Rogosa and Sharpe)
broth (Difco Laboratories) at 28 ◦ C without shaking. When
required, erythromycin (250 μg/ml for E. coli, 10 μg/ml for Lb.
plantarum) or chloramphenicol (10 μg/ml for both E. coli and
Lb. plantarum) was added to the medium. Solid agar plates were
prepared by adding 2 % (w/v) agar to the medium. The urea
complementation assay and H2 O2 toxicity growth assay were
carried out as described previously [21,22]. For the urea complementation assay, the dur3 mutant YNVW1 Saccharomyces
cerevisiae strain was transformed with the vector described above
containing the test GLP gene and transformants were selected on
synthetic medium A [2 % agar, 2 % glucose, pH 5.5, and 0.17 %
YNB (yeast nitrogen base) without amino acids and ammonium
(Difco)], supplemented with 1 mM arginine, and spotted on to
synthetic medium A supplemented with 1 mM arginine or different concentrations of urea as the sole nitrogen source. For the H2 O2
toxicity growth and survival assays, the above-described yeast
strains were grown on synthetic medium B [2 % agar, 2 % glucose
and 0.17 % YNB without amino acids (Difco)] containing different concentrations of H2 O2 . For the lactic acid complementation
assay, the Δjen1 S. cerevisiae mutant strain (Euroscarf: BY4741;
Mata; his3D1; leu2D0; met15D0; ura3D0; YKL217w::kanMX4)
was transformed with pRS426-pTPIu containing the test GLP
gene and transformants, selected on synthetic medium B supplemented according to the auxotrophic requirements with histidine,
methionine and leucine, and spotted on to synthetic medium B
supplemented according to the auxotrophic requirements with
histidine, methionine and leucine and with glucose or different
concentrations of D/L-lactate as the sole carbon source. After 5–
11 days of incubation at 28 ◦ C, differences in growth and survival
in the different assays were recorded. Two to three independent
experiments were performed and gave consistent results.
Growth rate measurement
Lb. plantarum cells were grown in 96-well sterile microplates
with a transparent bottom (Greiner) and the attenuance (D600 ) was
measured every 10 min with a Varioskan Flash multimode reader
(Thermo Scientific).
Growth competition experiment
Lb. plantarum wild-type and the double glpF1 glpF4 mutant
were grown overnight in MRS medium, then MRS medium with
or without 200 mM L-lactate was inoculated with an equal volume
of the two strains, resulting in a final D600 of 0.002. Inoculation was
performed every 12 h in order to keep the cells in the exponential
phase (D600 <1.0). Cultures were plated every 24 h on MRS agar
and 96 PCRs were performed on isolated colonies in order to
determine the percentages of wild-type and double mutant.
DNA techniques and transformation
General molecular biology techniques were performed as in
Sambrook et al. [23]. Electrocompetent Lb. plantarum cells were
c The Authors Journal compilation c 2013 Biochemical Society
Construction of plasmids for transport assays
Primers (Supplementary Table S2) matching the different glpF
sequences were used to PCR amplify the glpF ORFs (open reading
frames) from genomic DNA isolated from the different bacterial
strains. The PCR products were directionally subcloned using a
uracil excision-based improved high-throughput USER cloning
technique [25] into the USER-compatible Xenopus expression
vectors pNB1u and pNB1YFPu (N-terminal fusion of YFP with
the protein of interest) [26] or the yeast expression vector pRS426pTPIu (Supplementary Table S1).
Construction of deletion mutants
Construction of strains LR0003 (glpF1) and LR0004 (glpF4)
expressing deletion mutants of the Lb. plantarum MIP-encoding
genes was performed as described previously [24]. In order to
obtain the double-mutant strain LR0005 (glpF1 glpF4), strain
LR0004 (ΔglpF4) was used and glpF1 was deleted following the
same procedures described above. The strains and plasmids are
listed in Supplementary Table S1. The primers used to construct
the deletion vectors and to validate the deletions are listed in
Supplementary Table S2.
In vitro RNA synthesis and oocyte transport assays
Ready-to-use capped complementary RNAs encoding N-terminal
YFP-tagged or non-tagged GlpFs were synthesized in vitro
as described previously [27]. Xenopus laevis oocytes were
isolated, defolliculated and injected, and the Pf (osmotic water
permeability coefficient) was determined as described previously
[27]. Localization of the YFP-tagged fusion proteins was verified
by confocal microscopy (LSM 710; Carl Zeiss) on fixed oocytes
as described previously [28]. YFP was excited at 514 nm and
the emitted fluorescence detected between 530 and 570 nm.
Glycerol, dihydroxyacetone or lactic acid permeability was
measured in an iso-osmotic swelling assay as described previously
[29] with the following adaptations. Prior to the experiment,
the oocytes were osmotically equilibrated in standard Barth’s
solution (200 mOsm/l) for 15–25 min. Solute uptake was tested by
measuring the swelling rate at room temperature (20 ◦ C) by video
microscopy on incubation in an iso-osmotic modified Barth’s
solution (200 mOsm/l) in which NaCl was replaced with 160 mM
glycerol, 80 mM 1,3-dihydroxyacetone dimer (corresponding to
approximately 160 mM dihydroxyacetone in solution) or 80 mM
sodium D-, L- or D/L-lactate (corresponding to approximately
58 μM non-dissociated lactic acid at pH 7.0). The swelling rate of
the oocytes was expressed as the rate of change in oocyte volume
[d(V/V 0 )/dt, where V is the volume at a specific time and V 0 the
initial oocyte volume at zero time], as described previously [29].
Direct lactic acid uptake experiments in oocytes were
performed as follows. Oocytes were incubated at room
temperature for 0, 6 or 12 min in a modified Barth’s solution
containing 1 mM unlabelled sodium lactate and 14 C-labelled
sodium lactate (12 μCi/ml) (PerkinElmer) with the osmolarity
adjusted to 200 mOsm/l with NaCl, then were washed four times
with 250 ml of ice-cold modified Barth’s solution containing
Aquaglyceroporins from Lb. plantarum
80 mM non-labelled sodium lactate, lysed overnight with 1 ml
of 10 % (w/v) SDS, and the radioactivity was counted in
a scintillation counter after addition of 4 ml of Lumasafe
scintillation liquid (PerkinElmer).
561
Sequencing results showed that the database sequences were
in agreement with the cloned sequences. All six Lb. plantarum
GlpF (LpGlpF) sequences contained the typical selectivity filter
residues and NPA (asparagine-proline-alanine) boxes of bacterial
MIPs (Supplementary Table S5 and Supplementary Figure S1 at
http://www.biochemj.org/bj/454/bj4540559add.htm).
Lactate racemization assays
Lb. plantarum cells were grown in MRS medium and Lar (lactate
racemase) activity was induced by addition of L-lactate sodium
salt (Sigma–Aldrich) at a final concentration of 200 mM during
the mid-exponential phase (D600 = 0.6–0.7), then the cells were
collected 4 h later, washed twice with 1 volume of 60 mM
Mes buffer, pH 6 (R buffer), and resuspended in 1/20 volume
of R buffer containing 0.17–0.18 mm glass beads (Sartorius
Mechatronics Biomedicals). The cells were homogenized for 2×
1 min at 6.5 m/s in a FastPrep-24 (MP), the homogenate was
centrifuged at 13 000 g for 15 min (4 ◦ C), and the supernatant
was collected. The cytoplasmic Lar-specific activity was assayed
by incubating the supernatant with 20 mM D- or L-lactate in R
buffer at 35 ◦ C for 10 min, stopping the reaction by incubation for
10 min at 90 ◦ C, and measuring the conversion to L- or D-lactate
using the D-lactic acid/L-lactic acid UV test (R-Biopharm). The
protocol was adapted to 100 μl reaction volumes in 96-well, halfarea microplates (Greiner). Total protein in the supernatant was
measured according to the method of Bradford [30], using the BioRad Laboratories protein assay and the cytoplasmic Lar-specific
activity expressed as μmol of D- or L-lactate produced per min per
mg of total protein. Whole-cell Lar activity was measured in cell
suspensions (D600 of 1.0) of Lar-induced Lb. plantarum, prepared
as described previously [31], but using 50 mM D- or L-lactate
instead of glucose; the cell suspension was incubated for 1 h at
30 ◦ C, then L- or D-lactate formation was measured as above and
whole-cell Lar activity was expressed as μmol of lactate produced
per h by 1 ml of cells.
Gene context analysis and phylogenetic alignment
Genes surrounding glpF homologues of Lb. plantarum WCFS1
were retrieved from the NCBI database (release 189) and are
listed in Supplementary Table S3 (at http://www.biochemj.org/
bj/454/bj4540559add.htm). The MIP protein sequences for all
other Lactobacillales were retrieved from the NCBI database
(release 189) and are listed in Supplementary Table S4 (at http://
www.biochemj.org/bj/454/bj4540559add.htm). They were aligned using ClustalX2 [32] and the phylogenetic tree was constructed
using the neighbour-joining method [33]. The tree was visualized using treeGraph2 [34].
RESULTS
In silico analysis and cloning of Lb. plantarum MIPs
The available genomes for Lb. plantarum WCFS1, Lb. plantarum
JDM1 and Lb. plantarum subsp. plantarum ST-III were screened
for MIP-encoding genes and six were identified in each of the
three genomes by gene annotation and homology search and
annotated as glpF1, glpF2, glpF3, glpF4, glpF5 and glpF6. The
deduced lengths of the different GlpF proteins ranged from 216
to 250 amino acids and the deduced protein sequences for a given
GlpF in the three different bacterial strains were 100 % identical
(Supplementary Figure S1 at http://www.biochemj.org/bj/454/
bj4540559add.htm). Using gene-specific primers based on the
genomic data, MIP-encoding ORFs were amplified by PCR
from the genomic DNA of Lb. plantarum WCFS1 and cloned.
Phylogenetic and genetic context analysis of Lactobacillales MIPs
To predict the role of glpF genes from Lb. plantarum, we performed a BLAST comparison of EcGlpF against Lactobacillales
genomes. The homologues from 17 species representative of
the diversity within this order of Gram-positive bacteria were
retained. A total of 49 MIP-encoding genes were found in
these species and their number in each species ranged from one
to six (denoted 1–6 in Supplementary Table S4). The encoded
proteins, together with EcAqpZ, EcGlpF, and human Aqp0 as an
outgroup, were aligned with ClustalX2 [32] and a phylogenetic
tree constructed using the neighbour-joining method [33]. Five
distinct phylogenetic groups were deduced from this alignment
(Figure 1A). In order to infer a potential role of each of these
groups and of the six glpF genes from Lb. plantarum, we looked at
their genetic context (Figure 1B). The first group (named AqpZ)
was composed of E. coli AqpZ and LpGlpF6 and their homologues. No function could be attributed to this group on the basis
of their genetic context (Supplementary Table S3), but, owing to
their homology with EcAqpZ, we hypothesize that it is composed
of water channels. LpGlpF1 and LpGlpF4 clustered together in the
second group (named GlpF1/F4) as expected, since their protein
sequences were 92 % identical. The glpF1 gene of Lb. plantarum
is part of the lar operon, consisting of larA, larB, larC, larD
(glpF1) and larE (Figure 1B). Since glpF1 is located in this operon, in which all the other genes have been shown to be involved
in the lactate racemization (Lar) activity of the bacterium [35], this
suggests it plays a direct role in lactic acid transport. Four of the
ten homologues in the GlpF1/F4 group were also found in lactate
racemization operons (lar operons), suggesting that this group
consists of lactic acid channels or proteins involved in a transport
process contributing to the Lar activity. The third group (named
GlpF2) was composed of LpGlpF2 homologues. The glpF2 gene
of Lb. plantarum and several LpGlpF2 homologues are part of
an operon composed of dak1B (dihydroxyacetone kinase 1B),
dak2 and dak3 that is similar to the operon coding for the dak
system in Lc. lactis [36] (Figure 1B). Since dihydroxyacetone is
chemically very similar to glycerol, and since other mammalian
and protozoan GLPs have been demonstrated to transport this
molecule [37], it is likely that LpGlpF2 is a dihydroxyacetone
transporter. The fourth group (named GlpF3) was composed of
LpGlpF3 homologues. Several of these (including Lb. plantarum
glpF3) have been assigned to operons composed of glpK and
glpD genes, which are similar to those responsible for glycerol
utilization in E. coli [38] (Figure 1B), so we hypothesized that
glpF3 encoded a glycerol transporter. Two exceptions were found,
these being Lactobacillus salivarius GlpF3 and S. agalactiae
GlpF3, which clustered, respectively, with dihydroxyacetone
(GlpF2) or glycerol (GlpF3) transporters, but whose genetic
context pointed respectively to glycerol or dihydroxyacetone
transport (Figure 1A). It seems that the dihydroxyacetone and
glycerol channels are functionally very closely related and that
interconversion of one into the other may occur during evolution.
The fifth group (named GlpF5) was composed of LpGlpF5
homologues and homologues of Gla, the aquaglyceroporin of Lc.
lactis, which has been shown to transport both water and glycerol
c The Authors Journal compilation c 2013 Biochemical Society
562
Figure 1
G. P. Bienert and others
Phylogenetic analysis and phylogenetic context of Lb. plantarum GlpFs
(A) Phylogenetic tree of 49 MIPs from 17 Lactobacillales species, GlpF and AqpZ from E. coli , and Aqp0 from Homo sapiens . The number after the species name corresponds to that used for this
isoform in this species in Supplementary Table S4 (at http://www.biochemj.org/bj/454/bj4540559add.htm). The relevant genetic context of the MIP-encoding genes is given in parentheses as Lar
(lactate racemization operon), Dha (dihydroxyacetone utilization operon), Gly (glycerol utilization operon) or Prop (propanediol utilization operon). The GlpFs that were functionally investigated in
the present study are shown in bold. En. faecalis, Enterococcus faecalis ; G. adiacens , Granulicatella adiacens ; Le ., Leuconostoc ; S., Streptococcus ; W. koreensis , Weissella koreensis . (B) Genetic
context of each of the six glpF genes from Lb. plantarum WCFS1. The glpF1–glpF6 genes are shown in white background and the other genes in black background. For details of gene annotation,
see Supplementary Table S3 (at http://www.biochemj.org/bj/454/bj4540559add.htm).
[6]. However, two members of this GlpF5 group were also found
to be located in the lar operon, indicating a potential role in
lactic acid transport (Figure 1A and Supplementary Table S4).
c The Authors Journal compilation c 2013 Biochemical Society
The glpF4, glpF5 and glpF6 genes of Lb. plantarum are not part
of any operon and their putative function cannot be deduced from
their genetic context (Figure 1B and Supplementary Table S3).
Aquaglyceroporins from Lb. plantarum
Figure 2
563
Water, glycerol and dihydroxyacetone membrane permeability of X. laevis oocytes expressing Lb. plantarum GlpFs
(A) P f values of oocytes injected with water (negative control) or cRNA encoding rAQP9 (positive control) or the indicated LpGlpF isoform. The swelling rates were determined on immersion for 70 s
in half-strength Barth’s solution. The data are expressed as the means of measurements for 11–14 oocytes in one experiment and are representative of those obtained in three separate experiments.
(B) Glycerol permeability of oocytes injected with water or cRNA encoding rAQP9 (positive control), ZmPIP2;5 (negative control), LpGlpF2, LpGlpF3 or LpGlpF4, or co-injected with cRNAs encoding
ZmPIP2;5 and LpGlpF1, LpGlpF5 or LpGlpF6. The initial swelling rates were determined on immersion for 170 s in iso-osmotic Barth’s solution containing 160 mM glycerol. The data are expressed
as the means of measurements for 8–11 oocytes in one experiment and are representative of those obtained in three separate experiments. (C) Dihydroxyacetone permeability of oocytes injected with
water or cRNA encoding rAQP9 (positive control), ZmPIP2;5 (negative control), LpGlpF2, LpGlpF3 or LpGlpF4 or co-injected with cRNAs encoding ZmPIP2;5 and LpGlpF1, LpGlpF5 or LpGlpF6.
The initial swelling rates were determined as described in (B) by immersion for 170 s in 160 mM dihydroxyacetone (35 s for rAQP9). The data are expressed as the means of measurements on 20–33
oocytes in two independent experiments. In all experiments, 25 ng of cRNA for each MIP-encoding gene was used, except in the case of ZmPIP2;5 (2 ng). V and V 0 indicate the volume at a given
time point and the initial volume respectively. The error bars are the 95 % confidence intervals.
Water, glycerol and dihydroxyacetone permeability of Lb.
plantarum GlpFs
The water channel activity of Lb. plantarum GlpFs was
determined by heterologous expression in X. laevis oocytes
and swelling assays in a hypo-osmotic medium. Water-injected
oocytes (negative control) or oocytes expressing the respective
GlpF isoform (LpGlpF1–LpGlpF6) were suspended in halfstrength Barth’s solution, which resulted in an outwardly directed
osmotic gradient of 100 mOsm. As shown in Figure 2(A),
the water-injected oocytes had a low Pf owing to limited
transmembrane water movement, whereas cells expressing
the positive control rAQP9 (rat AQP9), a well-characterized
mammalian GLP protein, had a high Pf [39]. No significant
increase in Pf was detected for cells injected with LpGlpF1,
LpGlpF5 or LpGlpF6 cRNA, whereas the Pf value of oocytes
injected with LpGlpF2, LpGlpF3 or LpGlpF4 cRNA was
approximately 4-, 6- or 7-fold higher respectively than that of
the water-injected oocytes.
To determine whether the lack of a Pf increase in
oocytes expressing LpGlpF1, LpGlpF5 or LpGlpF6 was due
to a problem of protein expression or trafficking, oocytes
were injected with cRNAs encoding the GlpFs fused to
YFP (YFP–LpGlpF1 to YFP–LpGlpF6) and the localization
of the YFP-tagged proteins was analysed using confocal
microscopy. As shown in Supplementary Figure S2 (at http://
www.biochemj.org/bj/454/bj4540559add.htm), no specific fluorescent signal was detected in water-injected oocytes, whereas
all six sets of YFP–LpGlpF-expressing cells showed a similar
fluorescence distribution and intensity at their periphery,
indicating a plasma membrane localization, but, in contrast with
the positive control YFP–ZmPIP2;5 (Zea mays plasma membrane
intrinsic protein 2;5), an additional strong internal YFP signal was
detected with the Lb. plantarum GlpFs. These results suggest that
all Lb. plantarum isoforms are expressed in oocytes and are, at
least partially, targeted to the plasma membrane.
The glycerol permeability of the LpGlpF isoforms was
measured in an iso-osmotic swelling assay with a 160 mM
glycerol inwardly-directed chemical gradient. Under these
conditions, glycerol enters the cells due to the chemical
concentration difference, resulting in a secondary osmotic
gradient and an influx of water. To observe cell swelling, the
membrane permeability for water must be sufficiently high; this
is the case if the channel of interest is also permeable to water
or if the water-impermeable MIP is co-expressed with a channel
that is water-permeable, but impermeable to the solute tested
[29,40]. Oocytes were injected with cRNA encoding the waterpermeable LpGlpF2, LpGlpF3 or LpGlpF4 isoform or co-injected
with cRNAs encoding LpGlpF1, LpGlpF5 or LpGlpF6 and
ZmPIP2;5, a water-permeable and glycerol-impermeable plant
AQP [41]. As shown in Figure 2(B), oocytes injected with water
or expressing ZmPIP2;5 alone did not swell during the monitored
time period, whereas oocytes expressing LpGlpF2 or LpGlpF4
showed significantly high swelling rates, similar to those of cells
expressing the positive control rAQP9, and LpGlpF3-expressing
oocytes displayed an even higher swelling rate. Co-expression of
LpGlpF1, LpGlpF5 or LpGlpF6 with ZmPIP2;5 did not result in
any volume increase. To verify that co-expression of ZmPIP2;5
with LpGlpFs did not modify its water permeability, the Pf of
oocytes expressing ZmPIP2;5 alone or together with LpGlpF1,
LpGlpF5 or LpGlpF6 was measured and oocytes co-expressing
the two channels were found to have a significantly higher Pf
than water-injected oocytes, similar to that of oocytes expressing
only ZmPIP2;5 (Supplementary Figure S3A at http://www.
biochemj.org/bj/454/bj4540559add.htm), demonstrating that the
water channel activity of ZmPIP2;5 was not impaired when coexpressed with GlpF proteins and that the lack of detected glycerol
permeability was not due to a restrictive water permeability of the
oocyte membrane.
The facilitated diffusion of dihydroxyacetone through LpGlpF
isoforms was analysed in an iso-osmotic swelling assay in the
presence of a 160 mM inwardly-directed chemical gradient.
c The Authors Journal compilation c 2013 Biochemical Society
564
G. P. Bienert and others
As shown in Figure 2(C) and as previously shown [37],
rAQP9 was a highly efficient dihydroxyacetone channel. In
addition, the swelling rate of oocytes expressing LpGlpF2,
LpGlpF3 or LpGlpF4 was significantly increased after transfer to
dihydroxyacetone-containing buffer. Compared with the waterinjected cells, the volume of oocytes expressing ZmPIP2;5 alone
or co-expressing ZmPIP2;5 with LpGlpF1, LpGlpF5 or LpGlpF6
increased only slightly. As this increase was small and was
also observed for ZmPIP2;5, it could result from influx of
water via ZmPIP2;5 following the non-facilitated transmembrane
diffusion of dihydroxyacetone. The water and dihydroxyacetone
permeability studies shown in Supplementary Figure S3(B) and
Figure 2(C) were performed on the same batches of oocytes
and the results therefore strongly suggest that the increased
swelling rate seen in oocytes expressing LpGlpF2, LpGlpF3
or LpGlpF4 was due to dihydroxyacetone permeability and not
to their intrinsic water permeability. Indeed, oocytes expressing
these three channels had much lower Pf values than ZmPIP2;5expressing oocytes (Supplementary Figure S3B), whereas a much
greater swelling rate was seen in dihydroxyacetone-containing
buffer (Figure 2C). Taken together, these swelling assay results
show that LpGlpF2, LpGlpF3 and LpGlpF4 are each permeable to
water, glycerol and dihydroxyacetone, whereas no transport was
seen with LpGlpF1, LpGlpF5 or LpGlpF6.
Urea and H2 O2 permeability of Lb. plantarum GlpFs
Lb. plantarum GlpFs were tested for urea permeability in a
complementation assay by heterologous expression in the yeast
mutant strain YNVW1 (dur3). This strain is deficient for the
yeast intrinsic urea transporter DUR3 and grows only on medium
supplemented with high concentrations of urea as the sole nitrogen
source or with an alternative nitrogen source (i.e. arginine) [42].
As shown in Figure 3(A), expression of LpGlpF1 or LpGlpF4
rescued yeast growth on medium containing low concentrations
of urea (1, 2 or 3 mM), as also seen with the positive control, the
urea/H + symporter AtDUR3 (where At is Arabidopsis thaliana).
In contrast, none of the other GlpFs from Lb. plantarum were
able to facilitate uptake of urea into yeast cells. The results of
this complementation assay show that LpGlpF1 and LpGlpF4
facilitate urea diffusion.
The H2 O2 permeability of LpGlpFs was investigated in the
same yeast mutant used for the urea complementation assay.
Yeast cells expressing LpGlpFs were exposed to different H2 O2
concentrations in the growth medium and survival was monitored
on plates. As shown in Figure 3(B), expression of LpGlpF1,
LpGlpF3 or LpGlpF4 resulted in complete growth inhibition
at 0.8 mM H2 O2 , whereas yeast cells carrying an empty vector
(control) or expressing the urea/H + symporter AtDUR3, LpGlpF5
or LpGlpF6 grew normally. LpGlpF2 showed an intermediate
sensitivity to H2 O2 and displayed complete growth inhibition
at 1.2 mM H2 O2 . The results of this toxicity assay show
that LpGlpF1, LpGlpF3, LpGlpF4 and, possibly, LpGlp2 are
permeable to H2 O2 .
Lactic acid permeability of Lb. plantarum GlpFs
As (i) Lb. plantarum belongs to the lactic acid-producing bacteria,
(ii) LpGlpF1 is located in a lactic acid-responsive operon
and (iii) a few other MIPs facilitate lactic acid diffusion, the
lactic acid channel activity of LpGlpFs was investigated in a
complementation assay by heterologous expression in the yeast
mutant strain jen1. Disruption of Jen1p abolishes uptake of
lactate into yeast cells [43] and prevents their growth on medium
c The Authors Journal compilation c 2013 Biochemical Society
Figure 3 Urea and H2 O2 transport assays in yeast expressing Lb. plantarum
GlpF isoforms
(A) Urea transport complementation assay for LpGlpF isoforms. The YNVW1 (dur3 ) yeast
mutant defective in urea uptake [21] transformed with the empty vector pRS426-pTPIu (negative
control) or pRS426-pTPIu containing the indicated LpGlpF gene or AtDUR3 (positive control)
at a D 600 of 1, 0.01 or 0.0001 was spotted on to medium containing various concentrations of
urea or arginine as the sole nitrogen source and growth was recorded after 10 days at 28 ◦ C.
The results shown are for yeast spotted at a D 600 of 1; a similar sensitivity was seen in the
other two sets of cultures. Control indicates the empty vector. (B) H2 O2 sensitivity of yeast cells
expressing Lb. plantarum glpF s. Cultures of dur3 yeast cells transformed with the empty
vector pRS426-pTPIu (control) or pRS426-pTPIu carrying the indicated LpGlpF isoform or
AtDUR3 at a D 600 of 1, 0.01 and 0.0001 were spotted on to medium containing the indicated
concentrations of H2 O2 and growth was recorded after 6 days at 28 ◦ C. The growth behaviour
and survival rates of the different transformants is shown for the yeasts spotted at a D 600 of 0.01;
a similar sensitivity was seen in the other two sets of cultures. All yeast growth assays were
performed at least twice, with consistent results.
containing low amounts of lactic acid/lactate as the sole carbon
source and only transformation with a plasmid encoding a protein
able to transport lactic acid/lactate allows this mutant strain to
grow in this medium. As shown in Figure 4(A), expression of
LpGlpF1 or LpGlpF4 rescued yeast growth on medium containing
low concentrations of D/L-lactate (0.1, 0.25 or 0.5 % w/v),
but none of the other LpGlpFs were able to complement the
growth of jen1 strain under these conditions. However, all
GlpF-expressing strains grew on an alternative carbon source,
glucose. It was interesting to observe that GlpF1- and GlpF4expressing strains grew less than the other strains on glucose
medium, but rescued the growth on medium containing D/Llactate. This complementation assay showed that LpGlpF1 and
LpGlpF4 facilitate lactic acid diffusion.
To further investigate the lactic acid channel activity of
LpGlpFs, an oocyte iso-osmotic swelling assay using a 80 mM
sodium D/L-lactate inwardly-directed chemical gradient was performed. rAQP9 was used as a positive control, as it facilitates lactic
acid diffusion [44]. Neither the co-expression of LpGlpF1, LpGlpF5 or LpGlpF6 with ZmPIP2;5, nor the expression of LpGlpF2
and LpGlpF3 alone, led to an significant increase in the oocyte
swelling rate compared with the water- and ZmPIP2;5 cRNAinjected cells (Figure 4B and Supplementary Figure S4 at
http://www.biochemj.org/bj/454/bj4540559add.htm). In contrast,
oocytes expressing LpGlpF4 or the positive control rAQP9
showed a 330- and 750-fold higher swelling rate respectively,
showing that LpGlpF4 is a functional lactic acid channel
(Figure 4B and Supplementary Figure S4). To confirm these
results in an independent experiment, we directly measured
[14 C]lactic acid/lactate accumulation in oocytes expressing
LpGlpF1, LpGlpF4 or rAQP9 as a positive control after different
incubation times in 12 μCi/ml sodium [14 C]lactate. As shown in
Figure 4(C), oocytes expressing LpGlpF4 or rAQP9 accumulated
Aquaglyceroporins from Lb. plantarum
Figure 4
565
Lactic acid membrane permeability of X. laevis oocytes and S. cerevisiae cells expressing Lb. plantarum GlpF isoforms
(A) Lactic acid transport complementation assay of LpGlpF isoforms. Cultures of Δjen1 yeast mutants defective in lactate uptake capacity [43] transformed with the empty vector pRS426-pTPIu
(negative control) or pRS426-pTPIu containing the indicated LpGlpF gene at a D 600 of 1, 0.1 and 0.01 were spotted on to medium containing the indicated concentration of D/L-lactate or 2 % glucose
as the sole carbon source and growth was recorded after 10 days at 28 ◦ C. (B) D/L-lactic acid membrane permeability of oocytes injected with water (negative control) or cRNA encoding rAQP9
(positive control), LpGlpF2, LpGlpF3 or LpGlpF4 or co-injected with cRNAs coding for ZmPIP2;5 and LpGlpF1, LpGlpF5 or LpGlpF6. The initial swelling rates were determined on immersion for
350 s in iso-osmotic Barth’s solution containing 80 mM D/L-lactate. The results are the means of measurements of 8–12 oocytes in a single experiment and are representative of those obtained in
three separate experiments. The error bars are the 95 % confidence intervals. (C) Lactic acid uptake by oocytes injected with water or cRNA encoding rAQP9, LpGlpF1 or LpGlpF4. Oocytes were
incubated for the indicated time in modified Barth’s solution supplemented with 12 μCi/ml sodium [14 C]lactate and the lactate/lactic acid content was determined as described in the Materials and
methods section. The results are expressed as the means of measurements of 16–20 oocytes in one experiment and are representative of those obtained in three separate experiments. The error
bars are the 95 % confidence intervals. (D) D- or L-lactic acid membrane permeability of oocytes injected with water (negative control) or cRNA coding for LpGlpF4. The initial swelling rates were
determined as in (B) by immersion for 350 s in 80 mM D/L-lactic acid (open bars), 80 mM D-lactic acid (light grey bars) or 80 mM L-lactic acid (dark grey bars). The results are expressed as the
means of measurements for four (D-/L-lactic acid) or 10–12 (D- or L-lactic acid) oocytes in a single experiment and are representative of those obtained in three experiments. The error bars represent
the 95 % confidence intervals. V and V 0 indicate the volume at a given time point and the initial volume respectively.
approximately twice as much lactic acid/lactate as water-injected
control oocytes when incubated for 6 or 12 min with 12 μCi/ml
sodium [14 C]lactate, confirming that LpGlpF4 is an efficient lactic
acid transporter in the oocyte system. In contrast, expression of
LpGlpF1 resulted in similar lactic acid/lactate uptake to that seen
with water-injected control oocytes.
Stereoselectivity of LpGlpF4 for lactic acid
E. coli GlpF shows stereospecificity for the transport of substrates
with different chain lengths and chiralities, which influence
the hydrophobic contacts between the transported solute and the
channel residues [45]. Lactic acid has two optical isomers, L and
D, which are both synthesized and interconverted in Lb. plantarum
[35]. To test whether LpGlpF4 showed stereoselectivity for one
of the two optical isomers, its channel activity for L-lactic acid, Dlactic acid and a D/L racemic mixture were compared in the oocyte
iso-osmotic swelling assay. As shown in Figure 4(D), LpGlpF4expressing oocytes showed a similar increase in swelling rate
under the three different conditions, showing that the D- and Llactic acid isomers were equally well transported by this protein.
Transport selectivity of GlpF1/F4 homologues from P. pentosaceus
and Lactobacillus sakei
To investigate the transport selectivity of GLPs from other lactic
acid bacteria belonging to the same phylogenetic GlpF1/F4
group as LpGlpF1 and LpGlpF4 and located in a lactate
racemization operon (Supplementary Table S4), the glpF1
orthologues of P. pentosaceus and Lb. sakei were cloned and
the corresponding proteins were tested for water, glycerol
and lactic acid permeability by heterologous expression in oocytes
as described above. As shown in Figure 5, expression of PpGlpF1
(where Pp is P. pentosaceus) and LsGlpF1 (where Ls is Lb.
sakei) induced a significant increase in the oocyte membrane
permeability for all three tested compounds, demonstrating that
GlpF1 orthologues facilitate the diffusion of commonly tested
solutes and that PpGlpF1 and LsGlpF1 are functional lactic acid
channels.
LpGlpF1 and LpGlpF4 transport lactic acid in Lb. plantarum WCFS1
To confirm the lactic acid channel activity of LpGlpF4 in bacterial
cells, three marker-free deletion mutant strains (glpF1, glpF4
c The Authors Journal compilation c 2013 Biochemical Society
566
Figure 5
G. P. Bienert and others
Transport characteristics of X. laevis oocytes expressing glpF1 orthologues from P. pentosaceus and Lb. sakei
(A) P f values. (B) Glycerol permeability. (C) D/L-lactic acid permeability. Oocytes were injected with water (negative control) or cRNA encoding rAQP9 (positive control), PpGlpF1 or LsGlpF1. The
amount of injected cRNA and the measurement of the initial swelling rate were identical to those in Figures 2 and 4. Mean values are shown for (A) a total of 14 water-injected oocytes, 16 AQP9-injected
oocytes, 18 PpGlpF-injected oocytes and 17 LsGlpF-injected oocytes in two independent experiments; (B) a total of 14 water-injected oocytes, 13 AQP9-injected oocytes, 11 PpGlpF-injected oocytes
and 12 LsGlpF-injected oocytes in two independent experiments; (C) a total of 14 water-injected oocytes, 13 AQP9 injected-oocytes, 16 PpGlpF-injected oocytes, and 14 LsGlpF-injected oocytes in
two independent experiments. The error bars represent the 95 % confidence intervals.
and glpF1 glpF4) were generated in Lb. plantarum WCFS1
and assayed for lactic acid transport by incubating non-growing
cells with 50 mM D- or L-lactate and monitoring L- or Dlactate production. This whole-cell lactate racemization (Lar)
activity was standardized to the cytoplasmic Lar-specific activity
measured in the supernatant of cell homogenates, giving the
whole-cell Lar activity/cytoplasmic Lar specific activity ratio,
which is used as a means of evaluating the transport of lactic
acid across cell membranes. As shown in Figure 6(A), at pH 6.5,
both of the single mutant strains showed decreased lactic acid
transport ability expressed as a percentage of that for the D- to
L-lactate conversion in the wild-type, whereas the double mutant
strain showed an even greater decrease, which corresponded to
the additive effect of the two simple inactivations. When the pH
of the medium was reduced to pH 5.5 or increased to pH 7.5, the
lactic acid transport ability of the wild-type and all three types of
mutant cells was enhanced or reduced to the same extent, showing
that the transported molecule was lactic acid and not the lactate
anion (Figure 6B shows the result for the wild-type). This is in
agreement with the assumption that, with very few exceptions,
MIPs only facilitate the transport of uncharged solutes. These
results clearly show that both GlpF1 and GlpF4 transport lactic
acid in Lb. plantarum WCFS1.
Physiological importance of lactic acid transport
As shown in Figure 7(A), when wild-type Lb. plantarum and
the double mutant strain (glpF1 glpF4) were grown in MRS
medium, the double mutant showed a growth delay during
the exponential growth phase that was even more marked when the
medium was supplemented with 200 mM L-lactate, suggesting
that the double mutant had an increased sensitivity to mild lactic
acid stress. To determine whether the growth delay affected the
fitness of the double mutant compared with the wild-type, coculture (1:1 ratio) competition experiments were performed in
which the two strains were grown in MRS medium with or
without 200 mM L-lactate and kept in exponential phase for three
days, then the wild-type/mutant ratio was assessed by PCR on
isolated colonies. As shown in Figure 7(B), the wild-type clearly
prevailed over the mutant after three days of exponential growth
(> 90 % of the global population) and, under mild lactic acid
c The Authors Journal compilation c 2013 Biochemical Society
stress conditions, this effect was even more striking, since none
of the 96 colonies tested after three days growth in MRS medium
plus 200 mM L-lactate contained the mutant strain. These data
show that the double mutant had a reduced fitness, probably due
to its reduced lactate transport capability, and suggests that, under
lactic acid stress, it was even less efficient in excreting the lactic
acid it produced, resulting in an even lower fitness under these
conditions.
Interestingly, the mutant was able to grow to a higher attenuance
than the wild-type in medium with or without added lactate
(Figure 7A). Lactic acid is known to have an inhibitory effect on
growing cells, especially at low pH [46], and this inhibitory effect
would be mitigated in cells showing limited lactic acid transport,
giving the mutant a growth advantage at the end of the exponential
phase, when medium acidification becomes inhibitory.
DISCUSSION
MIPs are a large family of channel proteins that are present in
all kingdoms of life. Although higher eukaryotes have a high
number of MIP isoforms, micro-organisms have either only a
few (e.g. two and four in E. coli and S. cerevisiae respectively)
or none at all, as is the case in many bacteria [8]. This can be
explained by the absence of membrane-surrounded subcellular
organelles and/or the high surface-to-volume ratio, which makes,
in theory, channel-mediated membrane diffusion optional. The
present study identified a number of MIP-encoding genes in
the order Lactobacillales, which includes all lactic acid bacteria
species. Interestingly, six of these were found to be present in
Lb. plantarum, a highly adaptable and versatile species, which is
found in a variety of environmental niches in which the different
growth conditions require different transporters. This number of
MIP-encoding genes is the highest so far reported for any bacterial
strain.
The phylogenetic analysis of 49 MIPs from diverse
Lactobacillales species, including Lb. plantarum, identified five
different MIP groups, each containing at least one Lb. plantarum
isoform, underlining the potential flexibility of this species
in its use of different solutes. The analysis of the genetic
context revealed that three MIP-encoding genes were located
in operons involved in the metabolism of glycerol (LpGlpF3),
Aquaglyceroporins from Lb. plantarum
Figure 6
567
Lactic acid transport assay of Lb. plantarum cells deficient in GlpF1 and/or GlpF4
(A) Bidirectional lactic acid racemization (Lar) assay of cell suspensions of Lb. plantarum WCFS1 (WT), mutant glpF1 (1), mutant glpF4 (4) or the double mutant glpF1 glpF4 (14)
at pH 6.5. The whole-cell Lar activity/cytoplasmic Lar-specific activity ratio was calculated and expressed as a percentage of that for the D- to L-lactate conversion by Lb. plantarum WCFS1 cells.
Mean values for four experiments are shown. (B) Bidirectional Lar assay of cell suspensions of Lb. plantarum WCFS1 (WT) at pH 7.5, pH 6.5 and pH 5.5. The whole-cell Lar activity/cytoplasmic
Lar-specific activity ratio was calculated for each pH value and expressed as a percentage of that for the D- to L-lactate conversion at pH 6.5. Mean values for three experiments are shown. The error
bars represent the 95 % confidence intervals.
Figure 7
Growth and fitness of Lb. plantarum cells deficient in GlpF1 and GlpF4
(A) Growth curves of wild-type (WT) Lb. plantarum WCFS1 (continuous lines) and its isogenic mutant ΔglpF1 ΔglpF4 (broken lines) in MRS medium (black lines) or MRS supplemented with
200 mM L-lactate (grey lines). The mean value for 16 individual cultures is shown. (B) Competition experiment between the WT and double mutant (initial ratio 1:1) in MRS medium (black bars) or
MRS supplemented with 200 mM L-lactate (grey bars). The percentage of the double mutant in the whole population was assessed by PCR on 96 individual colonies from three independent cultures
(32 each) after inoculation (day zero) and after 1, 2 or 3 days. The error bars represent the 95 % confidence intervals.
dihydroxyacetone (LpGlpF2) and lactic acid (LpGlpF1).
Transport assays performed in Xenopus oocytes, S. cerevisiae
cells and Lb. plantarum deletion mutants confirmed the substrate
predictions based on the genetic context analysis. Although
Lb. plantarum GlpFs were found to be mixed-solute channels,
LpGlpF2, LpGlpF3 and LpGlpF1 transported dihydroxyacetone,
glycerol and lactic acid respectively in addition to other solutes
(see Table 1 for a summary). The present study highlights the fact
that the determination of the genetic context of MIP-encoding
genes in bacterial genomes helps identify their substrates and
elucidates their physiological function. No substrate could be
identified for LpGlpF5 and LpGlpF6, even though the homology
of LpGlpF6 with EcAqpZ suggests a water channel activity.
The GlpF5 group appeared more diverse and no clear transport
specificity could be assigned to this group, since some members
transport glycerol and water [6] and others are found in lar
operons, suggesting an involvement in lactic acid transport.
The absence of transport activity when LpGlpF1, LpGlpF5 or
LpGlpF6 was expressed in oocytes is intriguing, since they seem
to be, at least partly, targeted to the plasma membrane, as shown
by the use of YFP-tagged variants (Supplementary Figure S2).
One possible explanation might be that they are non-functional
in this heterologous host membrane, as reported for other
microbial AQPs [47]. Whether this is due to a problem of protein
conformation, selectivity or regulation remains to be determined.
Even though LpGlpF1 appeared to be non-functional in oocytes,
it was shown to facilitate the passage of lactic acid, urea and H2 O2
across membranes when heterologously expressed in yeast and to
contribute to lactic acid metabolism in Lb. plantarum (Table 1).
To shed light on the mechanisms and structural requirements
determining substrate selectivity in bacterial MIPs, we tested their
permeability to urea and H2 O2 in the yeast system. Like water and
glycerol, H2 O2 and urea are typically transported by aquaporins or
aquaglyceroporins respectively. The best-characterized bacterial
c The Authors Journal compilation c 2013 Biochemical Society
568
Table 1
G. P. Bienert and others
Summary of the substrate permeability of GlpFs from Lb. plantarum , Lb. sakei and P. pentosaceus
+ , transport; + / − , transport; ND, no transport detected; NT, the substrate was not tested. The type of assay used is indicated in parentheses as O (uptake assay in oocytes), B (bacterial cell
racemization assay) or Y (yeast complementation or toxicity growth assay).
Tested substrate and functional assay system
GlpF isoform
Water
Glycerol
Dihydroxyacetone
Lactic acid
Urea
H2 O2
LpGlpF1
LsGlpF1
PpGlpF1
LpGlpF2
LpGlpF3
LpGlpF4
LpGlpF5
LpGlpF6
ND (O)
+ (O)
+ (O)
+ (O)
+ (O)
+ (O)
ND (O)
ND (O)
ND (O)
+ (O)
+ (O)
+ (O)
+ (O)
+ (O)
ND (O)
ND (O)
ND (O)
NT
NT
+ (O)
+ (O)
+ (O)
ND (O)
ND (O)
+ (Y, B)
+ (O)
+ (O)
ND (Y, O)
ND (Y, O)
+ (Y, O, B)
ND (Y, O)
ND (Y, O)
+ (Y)
NT
NT
ND (Y)
ND (Y)
+ (Y)
ND (Y)
ND (Y)
+ (Y)
NT
NT
+ / − (Y)
+ (Y)
+ (Y)
ND (Y)
ND (Y)
aquaglyceroporin, EcGlpF, is highly permeable to glycerol, but
only very weakly permeable to urea [48]. Of the lactic-acidpermeable aqua(glycero)porins, rAQP9 is permeable to urea,
whereas AtNIP2;1 is not [12]. In the case of the tested bacterial
aquaglyceroporins, urea permeability was seen together with
lactic acid permeability (LpGlpF1 and LpGlpF4), but not
with glycerol permeability (LpGlpF4 and LpGlpF3). These data
suggest that aquaglyceroporins that are permeable to urea are also
permeable to lactic acid, whereas this is not the case for the only
known lactic acid-transporting AQP AtNIP2;1 [12], supporting
the notion that, although the size and chemical environment of the
selectivity filter are important for the passage of urea, glycerol
and lactic acid, there must be additional restriction sites for the
passage of each substrate. A molecular dynamics simulation study
[49] suggested that the pore environment next to the NPA filter
region (towards the intracellular side) is a high-energy barrier for
solutes such as urea. Examination of the protein sequences of the
different isoforms (Supplementary Figure S1) did not indicate any
specific amino acids that might be responsible for distinguishing
between these three solutes. As expected, the H2 O2 sensitivity of
yeasts expressing LpGlpF isoforms correlated closely with the
measured water permeability.
We also tried to identify key residues responsible for lactic
acid permeability by comparing the sequences of previously
described MIPs and the bacterial GlpF aquaglyceroporins. Three
highly conserved residues, Gly64 , Ala65 and His66 , upstream of the
first NPA motif, are known to interact with substrate molecules
inside the channel of EcGlpF [49]. With the exception of Ala65 ,
which is replaced by a valine residue in all members of this
group, these residues are much less well conserved in GlpF1–
GlpF4 homologues, in which Gly64 is replaced by an asparagine,
aspartate or glutamate residue and His66 by a cysteine, serine,
phenylalanine or tyrosine residue, showing that these residues
are of minor importance in lactic acid selectivity. The proline
residue in the second NPA motif is also less well conserved
in the GlpF1–GlpF4 homologues, being replaced by leucine or
glutamine residue. However, replacement of the proline by other
hydrophobic residues (leucine, methionine or valine) or polar
uncharged residues (serine, threonine, asparagine or glutamine)
is observed in the NPA motifs of MIPs from yeast, filamentous
fungi, nematodes, frogs and sea urchins [50,51]. GlpF1–GlpF4
homologues and EcGlpF or GlpF2 homologues transport different
molecules, but no difference was noted in the residues forming
the ar/R selectivity filter (Supplementary Figure S1 and Table S5),
indicating that these residues are not responsible for the lactic
acid selectivity. It will be interesting to determine which residues
account for the lactic acid transport of GlpF1–GlpF4 homologues.
The Val71 residue located just downstream of the first NPA motif
c The Authors Journal compilation c 2013 Biochemical Society
is replaced by methionine in all GlpF1-4 homologues and in two
GlpF5 homologues found in lar operons. This residue may be a
hallmark of bacterial lactic acid transporters. However, in rAQP9,
AtNIP2;1 and SmAQP, this residue is a valine.
The existence of a glpF gene coding for an aquaglyceroporin
transporting lactic acid in the lar operon of lactic acid bacteria
is of particular interest. To the best of our knowledge, no lactic
acid channel has been functionally characterized in lactic acid
bacteria, so GlpF1 and GlpF4 represent the first proteins that
contribute, in addition to the intrinsic membrane permeability,
to lactic acid transport in these bacteria. In Xenopus oocytes,
both lactic acid isomers were transported equally well by
LpGlpF4, demonstrating that this channel does not favour the
transport of one isomer over the other. Deletion of both glpF1 and
glpF4 in Lb. plantarum WCFS1 demonstrated the involvement of
these channels in the cellular racemization of lactic acid, justifying
the presence of a lactic acid transporter in many lar operons
from various Lactobacillales genomes. However, the location
outside the lar operon of GlpF1 homologues, such as GlpF4
from Lb. plantarum, suggests a more general physiological role
for these channels than only lactic acid racemization. Indeed,
the role of GlpF-mediated lactic acid transport seems to be
important during exponential growth phase, as shown by the
competition experiment between the wild-type and the double
mutant. Growth experiments also suggested that the uptake of
lactic acid at the onset of stationary phase via LpGlpF1 and
LpGlpF4 was responsible for the increased sensitivity to the
extracellular non-dissociated form and provided further evidence
that GlpFs function as lactic acid channels. Furthermore, our
growth analyses of the different bacterial strains at different
pH values suggested that lactic acid is channelled, but the
lactate anion is not. The ability of MIPs to conduct lactic acid
seems to be a conserved and common feature of the GlpF1/F4
subgroup. This was supported by the observation that PpGlpF1
and LsGlpF1, which are located in the lar operon, have the
same substrate specificity as their Lb. plantarum homologues.
The transport ability of GlpF isoforms belonging to the GlpF1/F4
group might be the basis for the protein-mediated facilitated
diffusion of the non-dissociated lactic acid molecule, which
was first suggested a long time ago, based on translocation
specificity and kinetic analyses in several Lactobacillales species
[52,53].
Whether aquaglyceroporins from Lb. plantarum are the only
transporters responsible for the regulation of the transmembrane
transport of lactic acid remains to be elucidated. The Lb.
plantarum genome encodes a potential lactate permease protein
(lp_1814) belonging to the lactate permease protein family, known
as a lactate/H + symporter, and its E. coli homologue is involved
Aquaglyceroporins from Lb. plantarum
in uptake of lactate into bacteria [54]. Whether these two types of
transporter are responsible for lactic acid transport under different
physiological conditions remains to be elucidated.
AUTHOR CONTRIBUTION
Gerd Bienert and Benoı̂t Desguin performed the experimental work. Gerd Bienert, Benoit
Desguin, François Chaumont and Pascal Hols designed the experiments, analysed the
data and wrote the paper.
FUNDING
This work was supported by grants from the Belgian National Fund for Scientific Research
(FNRS), the Interuniversity Attraction Poles Programme-Belgian Science Policy and the
“Communauté Française de Belgique-Actions de Recherches Concertées”. G.P.B. was
an FNRS Postdoctoral Researcher. B.D. and P.H. are a Research Fellow and Research
Associate at the FNRS respectively.
REFERENCES
1 Engel, A. and Stahlberg, H. (2002) Aquaglyceroporins: channel proteins with a conserved
core, multiple functions, and variable surfaces. Int. Rev. Cytol. 215, 75–104
2 Ishibashi, K. (2009) New members of mammalian aquaporins: AQP10–AQP12. Handb.
Exp. Pharmacol. 190, 251–262
3 Zardoya, R. (2005) Phylogeny and evolution of the major intrinsic protein family. Biol.
Cell 97, 397–414
4 Maurel, C., Reizer, J., Schroeder, J. I., Chrispeels, M. J. and Saier, Jr., M. H. (1994)
Functional characterization of the Escherichia coli glycerol facilitator, GlpF, in Xenopus
oocytes. J. Biol. Chem. 269, 11869–11872
5 Calamita, G., Bishai, W. R., Preston, G. M., Guggino, W. B. and Agre, P. (1995) Molecular
cloning and characterization of AqpZ, a water channel from Escherichia coli . J. Biol.
Chem. 270, 29063–29066
6 Froger, A., Rolland, J. P., Bron, P., Lagree, V., Le Caherec, F., Deschamps, S., Hubert, J. F.,
Pellerin, I., Thomas, D. and Delamarche, C. (2001) Functional characterization of a
microbial aquaglyceroporin. Microbiology 147, 1129–1135
7 Rodriguez, M. C., Froger, A., Rolland, J. P., Thomas, D., Aguero, J., Delamarche, C. and
Garcia-Lobo, J. M. (2000) A functional water channel protein in the pathogenic bacterium
Brucella abortus . Microbiology 146, 3251–3257
8 Tanghe, A., Van Dijck, P. and Thevelein, J. M. (2006) Why do microorganisms have
aquaporins? Trends Microbiol. 14, 78–85
9 Lorca, G. L., Barabote, R. D., Zlotopolski, V., Tran, C., Winnen, B., Hvorup, R. N.,
Stonestrom, A. J., Nguyen, E., Huang, L. W., Kim, D. S. and Saier, Jr, M. H. (2007)
Transport capabilities of eleven Gram-positive bacteria: comparative genomic analyses.
Biochim. Biophys. Acta 1768, 1342–1366
10 Faghiri, Z., Camargo, S. M., Huggel, K., Forster, I. C., Ndegwa, D., Verrey, F. and Skelly, P.
J. (2010) The tegument of the human parasitic worm Schistosoma mansoni as an
excretory organ: the surface aquaporin SmAQP is a lactate transporter. PLoS ONE 5,
e10451
11 Tsukaguchi, H., Weremowicz, S., Morton, C. C. and Hediger, M. A. (1999) Functional and
molecular characterization of the human neutral solute channel aquaporin-9. Am.
J. Physiol. 277, F685–F696
12 Choi, W. G. and Roberts, D. M. (2007) Arabidopsis NIP2;1, a major intrinsic protein
transporter of lactic acid induced by anoxic stress. J. Biol. Chem. 282, 24209–24218
13 Salema, M., Poolman, B., Lolkema, J. S., Dias, M. C. and Konings, W. N. (1994) Uniport
of monoanionic L-malate in membrane vesicles from Leuconostoc oenos . Eur. J.
Biochem. 225, 289–295
14 Tseng, C. P., Tsau, J. L. and Montville, T. J. (1991) Bioenergetic consequences of
catabolic shifts by Lactobacillus plantarum in response to shifts in environmental oxygen
and pH in chemostat cultures. J. Bacteriol. 173, 4411–4416
15 Otto, R., Sonnenberg, A. S., Veldkamp, H. and Konings, W. N. (1980) Generation of an
electrochemical proton gradient in Streptococcus cremoris by lactate efflux. Proc. Natl.
Acad. Sci. U.S.A. 77, 5502–5506
16 Olsen, E. B., Russell, J. B. and Henick-Kling, T. (1991) Electrogenic L-malate transport by
Lactobacillus plantarum : a basis for energy derivation from malolactic fermentation.
J. Bacteriol. 173, 6199–6206
17 Poolman, B., Molenaar, D., Smid, E. J., Ubbink, T., Abee, T., Renault, P. P. and Konings,
W. N. (1991) Malolactic fermentation: electrogenic malate uptake and malate/lactate
antiport generate metabolic energy. J. Bacteriol. 173, 6030–6037
569
18 Marty-Teysset, C., Posthuma, C., Lolkema, J. S., Schmitt, P., Divies, C. and Konings, W.
N. (1996) Proton motive force generation by citrolactic fermentation in Leuconostoc
mesenteroides . J. Bacteriol. 178, 2178–2185
19 Konings, W. N. (2006) Microbial transport: adaptations to natural environments. Antonie
van Leeuwenhoek 90, 325–342
20 Pudlik, A. M. and Lolkema, J. S. (2011) Citrate uptake in exchange with intermediates in
the citrate metabolic pathway in Lactococcus lactis IL1403. J. Bacteriol. 193, 706–714
21 Liu, L. H., Ludewig, U., Frommer, W. B. and von Wiren, N. (2003) AtDUR3 encodes a new
type of high-affinity urea/H + symporter in Arabidopsis . Plant Cell 15, 790–800
22 Bienert, G. P., Moller, A. L., Kristiansen, K. A., Schulz, A., Moller, I. M., Schjoerring, J. K.
and Jahn, T. P. (2007) Specific aquaporins facilitate the diffusion of hydrogen peroxide
across membranes. J. Biol. Chem. 282, 1183–1192
23 Sambrook, J., Fritsch, E. F. and Maniatis, T. (1989) Molecular Cloning: A Laboratory
Manual, 2nd edn, Cold Spring Harbour Laboratory Press, Cold Spring Harbor
24 Lambert, J. M., Bongers, R. S. and Kleerebezem, M. (2007) Cre-lox-based system for
multiple gene deletions and selectable-marker removal in Lactobacillus plantarum . Appl.
Environ. Microbiol. 73, 1126–1135
25 Nour-Eldin, H. H., Hansen, B. G., Norholm, M. H., Jensen, J. K. and Halkier, B. A. (2006)
Advancing uracil-excision based cloning towards an ideal technique for cloning PCR
fragments. Nucleic Acids Res. 34, e122
26 Nour-Eldin, H. H., Norholm, M. H. and Halkier, B. A. (2006) Screening for plant
transporter function by expressing a normalized Arabidopsis full-length cDNA library in
Xenopus oocytes. Plant Methods 2, 17
27 Fetter, K., Van Wilder, V., Moshelion, M. and Chaumont, F. (2004) Interactions between
plasma membrane aquaporins modulate their water channel activity. Plant Cell 16,
215–228
28 Bienert, G. P., Cavez, D., Besserer, A., Berny, M. C., Gilis, D., Rooman, M. and Chaumont,
F. (2012) A conserved cysteine residue is involved in disulfide bond formation between
plant plasma membrane aquaporin monomers. Biochem. J. 445, 101–111
29 Carbrey, J. M., Gorelick-Feldman, D. A., Kozono, D., Praetorius, J., Nielsen, S. and Agre,
P. (2003) Aquaglyceroporin AQP9: solute permeation and metabolic control of expression
in liver. Proc. Natl. Acad. Sci. U.S.A. 100, 2945–2950
30 Bradford, M. M. (1976) A rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72,
248–254
31 Lorquet, F., Goffin, P., Muscariello, L., Baudry, J. B., Ladero, V., Sacco, M., Kleerebezem,
M. and Hols, P. (2004) Characterization and functional analysis of the poxB gene, which
encodes pyruvate oxidase in Lactobacillus plantarum . J. Bacteriol. 186, 3749–3759
32 Larkin, M. A., Blackshields, G., Brown, N. P., Chenna, R., McGettigan, P. A., McWilliam,
H., Valentin, F., Wallace, I. M., Wilm, A., Lopez, R. et al. (2007) Clustal W and Clustal X
version 2.0. Bioinformatics 23, 2947–2948
33 Saitou, N. and Nei, M. (1987) The neighbor-joining method: a new method for
reconstructing phylogenetic trees. Mol. Biol. Evol. 4, 406–425
34 Stover, B. C. and Muller, K. F. (2010) TreeGraph 2: combining and visualizing evidence
from different phylogenetic analyses. BMC Bioinf. 11, 7
35 Goffin, P., Deghorain, M., Mainardi, J. L., Tytgat, I., Champomier-Verges, M. C.,
Kleerebezem, M. and Hols, P. (2005) Lactate racemization as a rescue pathway for
supplying D-lactate to the cell wall biosynthesis machinery in Lactobacillus plantarum .
J. Bacteriol. 187, 6750–6761
36 Zurbriggen, A., Jeckelmann, J. M., Christen, S., Bieniossek, C., Baumann, U. and Erni, B.
(2008) X-ray structures of the three Lactococcus lactis dihydroxyacetone kinase subunits
and of a transient intersubunit complex. J. Biol. Chem. 283, 35789–35796
37 Uzcategui, N. L., Szallies, A., Pavlovic-Djuranovic, S., Palmada, M., Figarella, K.,
Boehmer, C., Lang, F., Beitz, E. and Duszenko, M. (2004) Cloning, heterologous
expression, and characterization of three aquaglyceroporins from Trypanosoma brucei .
J. Biol. Chem. 279, 42669–42676
38 Durnin, G., Clomburg, J., Yeates, Z., Alvarez, P. J., Zygourakis, K., Campbell, P. and
Gonzalez, R. (2009) Understanding and harnessing the microaerobic metabolism of
glycerol in Escherichia coli . Biotechnol. Bioeng. 103, 148–161
39 Ishibashi, K., Kuwahara, M., Gu, Y., Tanaka, Y., Marumo, F. and Sasaki, S. (1998) Cloning
and functional expression of a new aquaporin (AQP9) abundantly expressed in the
peripheral leukocytes permeable to water and urea, but not to glycerol. Biochem. Biophys.
Res. Commun. 244, 268–274
40 Hansen, M., Kun, J. F., Schultz, J. E. and Beitz, E. (2002) A single, bi-functional
aquaglyceroporin in blood-stage Plasmodium falciparum malaria parasites. J. Biol.
Chem. 277, 4874–4882
41 Chaumont, F., Barrieu, F., Jung, R. and Chrispeels, M. J. (2000) Plasma membrane
intrinsic proteins from maize cluster in two sequence subgroups with differential
aquaporin activity. Plant Physiol. 122, 1025–1034
42 Liu, L. H., Ludewig, U., Gassert, B., Frommer, W. B. and von Wiren, N. (2003) Urea
transport by nitrogen-regulated tonoplast intrinsic proteins in Arabidopsis . Plant Physiol.
133, 1220–1228
c The Authors Journal compilation c 2013 Biochemical Society
570
G. P. Bienert and others
43 Casal, M., Paiva, S., Andrade, R. P., Gancedo, C. and Leao, C. (1999) The lactate-proton
symport of Saccharomyces cerevisiae is encoded by JEN1. J. Bacteriol. 181,
2620–2623
44 Tsukaguchi, H., Shayakul, C., Berger, U. V., Mackenzie, B., Devidas, S., Guggino, W. B.,
van Hoek, A. N. and Hediger, M. A. (1998) Molecular characterization of a broad
selectivity neutral solute channel. J. Biol. Chem. 273, 24737–24743
45 Fu, D., Libson, A., Miercke, L. J., Weitzman, C., Nollert, P., Krucinski, J. and Stroud, R. M.
(2000) Structure of a glycerol-conducting channel and the basis for its selectivity. Science
290, 481–486
46 Pieterse, B., Leer, R. J., Schuren, F. H. and van der Werf, M. J. (2005) Unravelling the
multiple effects of lactic acid stress on Lactobacillus plantarum by transcription profiling.
Microbiology 151, 3881–3894
47 Kozono, D., Ding, X., Iwasaki, I., Meng, X., Kamagata, Y., Agre, P. and Kitagawa, Y. (2003)
Functional expression and characterization of an archaeal aquaporin. AqpM from
Methanothermobacter marburgensis . J. Biol. Chem. 278, 10649–10656
Received 14 March 2013/20 June 2013; accepted 25 June 2013
Published as BJ Immediate Publication 25 June 2013, doi:10.1042/BJ20130388
c The Authors Journal compilation c 2013 Biochemical Society
48 Heller, K. B., Lin, E. C. and Wilson, T. H. (1980) Substrate specificity and transport
properties of the glycerol facilitator of Escherichia coli . J. Bacteriol. 144, 274–278
49 Hub, J. S. and de Groot, B. L. (2008) Mechanism of selectivity in aquaporins and
aquaglyceroporins. Proc. Natl. Acad. Sci. U.S.A. 105, 1198–1203
50 Ishibashi, K. (2006) Aquaporin superfamily with unusual npa boxes: S-aquaporins
(superfamily, sip-like and subcellular-aquaporins). Cell. Mol. Biol. 52, 20–27
51 Pettersson, N., Filipsson, C., Becit, E., Brive, L. and Hohmann, S. (2005) Aquaporins in
yeasts and filamentous fungi. Biol. Cell. 97, 487–500
52 Harold, F. M. and Levin, E. (1974) Lactic acid translocation: terminal step in glycolysis by
Streptococcus faecalis . J. Bacteriol. 117, 1141–1148
53 Gätje, G., Müller, V. and Gottschalk, G. (1991) Lactic acid excretion via carrier-mediated
facilitated diffusion in Lactobacillus helveticus . Appl. Microbiol. Biotechnol. 34, 778–782
54 Nunez, M. F., Kwon, O., Wilson, T. H., Aguilar, J., Baldoma, L. and Lin, E. C. (2002)
Transport of L-lactate, D-lactate, and glycolate by the LldP and GlcA membrane carriers of
Escherichia coli . Biochem. Biophys. Res. Commun. 290, 824–829
Biochem. J. (2013) 454, 559–570 (Printed in Great Britain)
doi:10.1042/BJ20130388
SUPPLEMENTARY ONLINE DATA
Channel-mediated lactic acid transport: a novel function
for aquaglyceroporins in bacteria
Gerd P. BIENERT*1,2 , Benoı̂t DESGUIN*1 , François CHAUMONT*3 and Pascal HOLS*3
*Institut des Sciences de la Vie, Université catholique de Louvain, Croix du Sud 4-5, 1348 Louvain-la-Neuve, Belgium
Table S1
Strains and plasmids used in the present study
r
Cm , resistance to chloramphenicol; Emr , resistance to erythromycin; NCIMB, The National Collections of Industrial and Marine Bacteria, Aberdeen, Scotland; StrR , resistance to streptomycin.
Strain or plasmid
Strain
S. cerevisiae
YNVW1
Lb. plantarum
WCFS1
LR0003
LR0004
LR0005
E. coli
TOP10
Plasmid
pNB1u
pNB1YFPu
pRS426-pTPIu
pNZ5319
pNZ5348
pGIR801
pGIR802
1
2
3
Characteristic(s)
Source or reference
dur3 , ura3 ; mutant defective in urea uptake
[1]
Wild-type
WCFS1 glpF1
WCFS1 glpF4
WCFS1 glpF1 ΔglpF4
NCIMB
The present study
The present study
The present study
F − mcrA (mrr-hsdRMS-mcrBC ) ϕ80 lacZ M15 lacX 74 nupG recA1 araD 139 (ara-leu)7697 galE 15 galK 16 rpsL (StrR ) endA1 λ −
Invitrogen
Oocyte expression vector for MIP homologues
Oocyte expression vector for the N-terminal YFP-tagged protein of interest fusion
Yeast expression vector for MIP homologues
Cmr Emr ; pNZ5318 derivative for multiple gene replacements in Gram-positive bacteria
Emr ; pGID023 derivative containing cre under the control of the lp_1144 promoter
Cmr Emr ; pNZ5319 derivative containing homologous regions up- and down-stream of glpF1 (used for the construction of LR0003)
Cmr Emr ; pNZ5319 derivative containing homologous regions up- and down-stream of glpF4 (used for the construction of LR0004)
[2]
[2]
Laboratory collection
[3]
[3]
The present study
The present study
These authors contributed equally to this work.
Present address: Leibniz Institute of Plant Genetics and Crop Plant Research, 06466 Gatersleben, Germany.
Correspondence may be addressed to either of these authors (email [email protected] or [email protected]).
c The Authors Journal compilation c 2013 Biochemical Society
G. P. Bienert and others
Figure S1 ClustalX2 alignment of the six MIPs from Lb. plantarum (LpGlpF1–LpGlpF6), a Schistosoma mansoni AQP (SmAQP), Rattus norvegicus AQP9
(rAQP9), E. coli GlpF and AqpZ (EcGlpF and EcAqpZ), and Arabidopsis thaliana NIP2-1 (AtNIP2-1)
The selectivity ar/R filter residues (H2, H5, LE1 and LE2 ) and the NPA motifs (NPA1 in loop B and NPA2 in loop E) are boxed (see Table S5 for details).
c The Authors Journal compilation c 2013 Biochemical Society
Aquaglyceroporins from Lb. plantarum
Figure S2
Confocal microscopy analysis of YFP-tagged GlpF proteins expressed in X. laevis oocytes
Oocyte injected with 25 ng of cRNA coding for YFP–LpGlpF1 (A), YFP–LpGlpF2 (B), YFP–LpGlpF3 (C), YFP–LpGlpF4 (D), YFP–LpGlpF5 (E) or YFP–LpGlpF6 (F), or with 4 ng of cRNA coding for
YFP–ZmPIP2;5 (G) or 50 nl of water (H) were fixed and observed 3 days after injection as described previously [9]. The arrows indicate labelling of the oocyte membrane. Scale bars, 100 μm.
Figure S3
Water permeability of X. laevis oocytes expressing Lb. plantarum GlpFs
(A) Osmotic water permeability coefficients (P f ) of oocytes injected with water (negative control) or cRNA coding for ZmPIP2;5 or co-injected with cRNAs coding for ZmPIP2;5 and LpGlpF1, LpGlpF5
or LpGlpF6. The data are expressed as the means of measurements of 8–13 cells. (B) P f values for oocytes injected with water (negative control) or cRNA coding for rAQP9, ZmPIP2;5, LpGlpF2,
LpGlpF3 or LpGlpF4. The results are expressed as the means of measurements of 6–9 cells. In both experiments, 25 ng of cRNA for each MIP-encoding gene was used, except in the case of ZmPIP2;5
(2 ng). The error bars are the 95 % confidence intervals.
Figure S4 Lactic acid membrane permeability of X. laevis oocytes
expressing Lb. plantarum GlpFs
D/L-lactic acid membrane permeability of oocytes injected with water (negative control) or cRNA
coding for rAQP9 (positive control), LpGlpF2, LpGlpF3 or LpGlpF4, or co-injected with cRNAs
coding for ZmPIP2;5 and LpGlpF1, LpGlpF5 or LpGlpF6. The initial swelling rates were measured
by immersion for 350 s in an isotonic modified Barth solution containing 80 mM D/L-lactate
with the osmolarity adjusted to 200 mOsm/l with NaCl. V (t) indicates the volume at a given time
point and V (0) the initial volume. The data are expressed as the mean of measurements of 8–12
oocytes in one representative experiment out of three. The error bars are the 95 % confidence
intervals.
c The Authors Journal compilation c 2013 Biochemical Society
G. P. Bienert and others
Table S2
List of primers used in the present study
(a) Primers for USER-compatible Xenopus gene expression
Primer name
Sequence (5 →3 )
Mutant strain generation or plasmid construction
GlpF1_A
GlpF1_B
GlpF2_A
GlpF2_B
GlpF3_A
GlpF3_B
GlpF4_A
GlpF4_B
GlpF5_A
GlpF5_B
GlpF6_A
GlpF6_B
GlpSak_A
GlpSak_B
GlpPen_A
GlpPen_B
GGCTTAAUATGGTACATCAGTTGATTGCTGAATTTATGG
GGTTTAAUTTAATTAATTCCAAAAAAGCCATGCATAAAC
GGCTTAAUATGCACGGTTTTTTAGGTGAATTTTTAG
GGTTTAAUTTAACTAATAATTGTTTCCAGACCAG
GGCTTAAUATGATGAAAGATCCATTAGCACTAC
GGTTTAAUTTATGGCAATACATTAAACAACAG
GGCTTAAUATGATTCATCAGCTTTTGGCAGAG
GGTTTAAUCTACCCAATCCCAAAGAACC
GGCTTAAUATGACAGGTTCATGGGAAG
GGTTTAAUTTATAAACCAAAAAAGATCTTATAAATCC
GGCTTAAUATGCGCAAGTATCTCGCC
GGTTTAAUCTAGTCTTCTGATCCCATAACC
GGCTTAAUATGGTTCATCAGTTATTAGCAGAATTCATGG
GGTTTAAUTTAGATACCAAAGAAGCTCCGCATGAAG
GGCTTAAUATGGTAAATCATCTATTAGCCGAATTTATGG
GGTTTAAUTTAGTAAATGTGTAAGTAAAATCTAACAAAAG
LpGlpF1 expression
LpGlpF1 expression
LpGlpF2 expression
LpGlpF2 expression
LpGlpF3 expression
LpGlpF3 expression
LpGlpF4 expression
LpGlpF4 expression
LpGlpF5 expression
LpGlpF5 expression
LpGlpF6 expression
LpGlpF6 expression
LsGlpF1 expression
LsGlpF1 expression
PpGlpF1 expression
PpGlpF1 expression
(b) Primers used for the construction of deletion vectors
Primer name
Sequence (5 →3 )
Mutant strain generation or plasmid construction
-D_UP_A
-D_UP_B
-D_DW_A
-D_DW_B
-GlpF4_UP_A
-GlpF4_UP_B
-GlpF4_DW_A
-GlpF4_DW_B
TGATATGTTTCTGGGCGCGTTAC
TACCAAAAAATTACGCCTCCTCATC
TGCCAGGAATTGCGCCCTTTG
TTGCACCGTGACCAACTTGTAAC
GAATTTAAAAGCCCGTTACCC
AAGCTGATGAATCACTATCTTTCTC
ATGGGTTCTTTGGGATTGGGTAG
TCACGTTGCCCTCCTGCTTTG
pGIR801
pGIR801
pGIR801
pGIR801
pGIR802
pGIR802
pGIR802
pGIR802
Primer name
Sequence (5 →3 )
Mutant strain generation or plasmid construction
2Hy_LarC_A
-D_diag_A
-GlpF4_diag_A
-GlpF4_diag_B
GATGCAAACACTTTATTTAGACGCTTTTTC
GCCGCCTTATTAGCTTTGG
GTTTCCCGCGGCATTCCATTG
TTGGTGCATATCATAGTACCGATATTG
LR0003
LR0003
LR0004
LR0004
(c) Primers used for the validation of deletions
(d) Primers used for the sequencing of deletion vectors
Primer name
Sequence (5 →3 )
Mutant strain generation or plasmid construction
85
85_compl
87_bis
87_bis_compl
GTTTTTTTCTAGTCCAAGCTCACA
TTATTCGTTTGATTTCGCTTTCG
TTGATGATTGGTTCGGAAGGCACG
TATATAGTTTACCCCGTCAGC
pGIR801 and pGIR802
pGIR801 and pGIR802
pGIR801 and pGIR802
pGIR801 and pGIR802
c The Authors Journal compilation c 2013 Biochemical Society
Aquaglyceroporins from Lb. plantarum
Table S3
Genetic context of Lb. plantarum glpFs
(a) glpF1 gene context
Accession number
Protein
Annotation
YP_004888174.1
YP_004888175.1
YP_004888176.1
YP_004888177.1
YP_004888178.1
YP_004888179.1
larA gene product
larB gene product
larC1 gene product
larC2 gene product
glpF1 gene product
larE gene product
Lactate racemization operon protein LarA
Lactate racemization operon protein LarB
Lactate racemization operon protein LarC, N-terminal domain
Lactate racemization operon protein LarC, C-terminal domain
Glycerol uptake facilitator protein GlpF1
Lactate racemization operon protein LarE
(b) glpF2 gene context
Accession number
Protein
Annotation
YP_004888225.1
YP_004888226.1
YP_004888227.1
YP_004888228.1
YP_004888229.1
dak1B gene product
dak2 gene product
dak3 gene product
dhaP gene product
PurR
Phosphotransferase, dihydroxyacetone-binding subunit
Dihydroxyacetone phosphotransferase, ADP-binding subunit
Dihydroxyacetone phosphotransferase, phosphoryl donor protein
Dihydroxyacetone transport protein GlpF2
LacI family transcriptional regulator, maltose-related
(c) glpF3 gene context
Accession number
Protein
Annotation
YP_004888401.1
YP_004888402.1
YP_004888403.1
glpK gene product
glpD gene product
glpF3 gene product
Glycerol kinase
Glycerol-3-phosphate dehydrogenase, FAD-dependent
Glycerol uptake facilitator protein GlpF3
(d) glpF4 gene context
Accession number
Protein
Annotation
YP_004889075.1
YP_004889076.1
YP_004889077.1
Virul_fac_BrkB
glpF4 gene product
glf1 gene product
Ribonuclease BN family protein
Glycerol uptake facilitator protein GlpF4
UDP-galactopyranose mutase
(e) glpF5 gene context
Accession number
Protein
Annotation
YP_004890928.1
YP_004890929.1
YP_004890930.1
YP_004890931.1
cadA gene product
glpF5 gene product
trxA3 gene product
BsYrkD-like
Cadmium transporting P-type ATPase
Glycerol uptake facilitator protein GlpF5
Thioredoxin
Hypothetical protein
Accession number
Protein
Annotation
YP_004890947.1
YP_004890948.1
YP_004890949.1
MoxR
glpF6 gene product
ubiB gene product
MoxR family ATPase
Glycerol uptake facilitator protein GlpF6
Ubiquinone biosynthesis protein, ABC1 family
(e) glpF6 gene context
c The Authors Journal compilation c 2013 Biochemical Society
G. P. Bienert and others
Table S4
List of 49 MIPs from 17 Lactobacillales species
The number in front of the MIP name corresponds to that after the species name in Figure 1 of the main text and Figure S1. Dha, dihydroxyacetone utilization operon [4]; Gly, glycerol utilization
operon [5]; Lar, lactate racemization operon [6]; and Prop, propanediol utilization operon [7].
Species
S. agalactiae FSL S3-025
1
2
3
Weissella koreensis KACC 15509
1
2
Enterococcus faecalis OG1RF
1
2
3
Leuconostoc mesenteroides subsp. mesenteroides J17
1
2
Lactobacillus acidophilus 30SC
1
Lb. salivarius SMXD51
1
2
3
Lactobacillus suebicus KCTC 3549
1
2
3
4
Lactobacillus mali KCTC 3596 = DSM 20444
1
2
3
Lc. lactis subsp. cremoris NZ9000
1
2
3
Leuconostoc argentinum KCTC 3773
1
2
Lactobacillus delbrueckii subsp. lactis DSM 20072
1
2
Streptococcus australis A.T.C.C. 700641
1
2
3
4
Granulicatella adiacens A.T.C.C. 49175
1
Lb. plantarum WCFS1
1
2
3
4
5
6
Lb. brevis ATCC 367
1
2
3
P. pentosaceus ATCC 25745
1
2
3
Lb. sakei subsp. sakei 23K
1
2
3
4
c The Authors Journal compilation c 2013 Biochemical Society
Accession number
Annotation
Context
EGS28598.1
EGS26706.1
EGS26627.1
Glycerol uptake facilitator protein
Hypothetical protein FSLSAGS3026_10000
Glycerol uptake facilitator protein
Gly
Dha
YP_004726868.1
YP_004726362.1
MIP family facilitator protein
Aquaporin
-
YP_005708647.1
YP_005708605.1
YP_005708021.1
glpF2 gene product
glpF gene product
aqpZ gene product
Gly
Lar
-
YP_005174634.1
YP_005173456.1
Glycerol uptake facilitator-related permease (MIP family)
Glycerol uptake facilitator-related permease
-
YP_004292761.1
Glycerol uptake facilitator protein
-
EIA33094.1
EIA32770.1
EIA32549.1
Glycerol uptake facilitator protein
Glycerol uptake facilitator protein
Glycerol uptake facilitator protein
Dha
Gly
ZP_09451087.1
ZP_09450612.1
ZP_09449835.1
ZP_09449755.1
Glycerol uptake facilitator-related permease
Glycerol uptake facilitator protein
MIP family glycerol uptake facilitator protein GlpF
Glycerol uptake facilitator-related permease (MIP family)
-
ZP_09449322.1
ZP_09448664.1
ZP_09448262.1
Glycerol uptake facilitator protein
Glycerol uptake facilitator protein
Glycerol uptake facilitator-related permease (MIP family)
Dha
Lar
YP_006357717.1
YP_006356524.1
YP_006356304.1
Putative glycerol uptake facilitator protein
Glycerol uptake facilitator
Transporter
Gly
-
ZP_08230752.1
ZP_08230396.1
Glycerol uptake facilitator-related permease (MIP family)
Aquaporin Z
-
EGD27475.1
EGD27267.1
MIP family glycerol uptake facilitator protein GlpF
MIP family glycerol uptake facilitator protein GlpF
-
ZP_08021527.1
ZP_08021095.1
ZP_08021080.1
ZP_08020846.1
Glycerol facilitator-aquaporin
MIP family glycerol uptake facilitator protein GlpF
MIP family glycerol uptake facilitator protein GlpF
MIP family major intrinsic protein water channel AqpZ
Prop
Gly
-
ZP_05738516.1
Glycerol uptake facilitator
Lar
YP_004888178.1
YP_004888228.1
YP_004888403.1
YP_004889076.1
YP_004890929.1
YP_004890948.1
glpF1 gene product
dhaP gene product (GlpF2)
glpF3 gene product
glpF4 gene product
glpF5 gene product
glpF6 gene product
Lar
Dha
Gly
-
YP_795947.1
YP_795727.1
YP_796082.1
Glycerol uptake facilitator-related permease (MIP family)
Glycerol uptake facilitator-related permease (MIP family)
Glycerol uptake facilitator-related permease
Lar
-
YP_805154.1
YP_805088.1
YP_804732.1
Glycerol uptake facilitator-related permease (MIP family)
Glycerol uptake facilitator-related permease (MIP family)
Glycerol uptake facilitator-related permease
Lar
Gly
-
YP_395402.1
YP_395266.1
YP_395780.1
YP_396456.1
MIP family facilitator protein
glpF gene product
MIP family facilitator protein
aqpZ gene product
Lar
Gly
-
Aquaglyceroporins from Lb. plantarum
Table S5 Selectivity filter residues and NPA boxes in Lb. plantarum , Lb.
sakei and P. pentosaceus MIPs
The ar/R filter is defined by one amino acid residue in helix 2 (H2), one in helix 5 (H5) and two
in loop E (LE). The ar/R region is the narrowest part of the pore and the electro-physicochemical
properties of the four amino acids forming this selectivity filter are assumed to determine the
pore selectivity of MIPs in general. As in EcGlpF, this selectivity filter in GlpF1 and GlpF2
is composed of WGFR (see also Figure S1), GlpF3 has a WGYR ar/R selectivity filter, which is
physicochemically very similar to WGFR, GlpF5 a rather unusual YVPR ar/R filter and GlpF6
(see also Figure S1) F(H/I)XR, in which the X residue is a small uncharged residue (e.g. A, G,
L or T), which is typical of the water-specific AQP family in several organisms [8]. The amino
acids corresponding to the characteristic MIP NPA motifs in loop B and E are shown on the
right-hand side; changes from the classical NPA motifs are shaded in grey.
ar/R selectivity filter
NPA motifs
Protein
H2
H5
LE1
LE2
LoopB
LoopE
LpGlpF1
LpGlpF2
LpGlpF3
LpGlpF4
LpGlpF5
LpGlpF6
LsGlpF1
PpGlpF1
W
W
W
W
Y
F
W
W
G
G
G
G
V
I
G
G
F
F
Y
F
P
G
F
F
R
R
R
R
R
R
R
R
NPA
NPA
NPA
NPA
NPA
NPA
NPA
NPA
NLA
NPA
NPA
NLA
NPA
NPA
NLA
NQR
REFERENCES
1 Liu, L. H., Ludewig, U., Frommer, W. B. and von Wiren, N. (2003) AtDUR3 encodes a new
type of high-affinity urea/H + symporter in Arabidopsis . Plant Cell 15, 790–800
2 Nour-Eldin, H. H., Hansen, B. G., Norholm, M. H., Jensen, J. K. and Halkier, B. A. (2006)
Advancing uracil-excision based cloning towards an ideal technique for cloning PCR
fragments. Nucleic Acids Res. 34, e122
3 Lambert, J. M., Bongers, R. S. and Kleerebezem, M. (2007) Cre-lox-based system for
multiple gene deletions and selectable-marker removal in Lactobacillus plantarum . Appl.
Environ. Microbiol. 73, 1126–1135
4 Zurbriggen, A., Jeckelmann, J. M., Christen, S., Bieniossek, C., Baumann, U. and Erni, B.
(2008) X-ray structures of the three Lactococcus lactis dihydroxyacetone kinase subunits
and of a transient intersubunit complex. J. Biol. Chem. 283, 35789–35796
5 Durnin, G., Clomburg, J., Yeates, Z., Alvarez, P. J., Zygourakis, K., Campbell, P. and
Gonzalez, R. (2009) Understanding and harnessing the microaerobic metabolism of
glycerol in Escherichia coli . Biotechnol. Bioeng. 103, 148–161
6 Goffin, P., Deghorain, M., Mainardi, J. L., Tytgat, I., Champomier-Verges, M. C.,
Kleerebezem, M. and Hols, P. (2005) Lactate racemization as a rescue pathway for
supplying D-lactate to the cell wall biosynthesis machinery in Lactobacillus plantarum . J.
Bacteriol. 187, 6750–6761
7 Johnson, C. L., Pechonick, E., Park, S. D., Havemann, G. D., Leal, N. A. and Bobik, T. A.
(2001) Functional genomic, biochemical, and genetic characterization of the Salmonella
pduO gene, an ATP:cob(I)alamin adenosyltransferase gene. J. Bacteriol. 183, 1577–1584
8 Savage, D. F., O’Connell, 3rd, J. D., Miercke, L. J., Finer-Moore, J. and Stroud, R. M.
(2010) Structural context shapes the aquaporin selectivity filter. Proc. Natl. Acad. Sci.
U.S.A. 107, 17164–17169
9 Bienert, G. P., Cavez, D., Besserer, A., Berny, M. C., Gilis, D., Rooman, M. and Chaumont,
F. (2012) A conserved cysteine residue is involved in disulfide bond formation between
plant plasma membrane aquaporin monomers. Biochem. J. 445, 101–111
Received 14 March 2013/20 June 2013; accepted 25 June 2013
Published as BJ Immediate Publication 25 June 2013, doi:10.1042/BJ20130388
c The Authors Journal compilation c 2013 Biochemical Society