Functional reconstitution of a rice aquaporin water channel, PIP1;1, by a micro-batchwise methodology

Plant Physiology and Biochemistry 85 (2014) 78e84

Contents lists available at ScienceDirect

Plant Physiology and Biochemistry journal homepage: www.elsevier.com/locate/plaphy

Research article

Functional reconstitution of a rice aquaporin water channel, PIP1;1, by a micro-batchwise methodology Vito Scalera a, Patrizia Gena a, Maria Mastrodonato b, Yoshichika Kitagawa c, Salvatore Carulli a, Maria Svelto a, Giuseppe Calamita a, d, * a

Department of Biosciences, Biotechnologies and Biopharmaceutics, University of Bari “Aldo Moro”, Bari, Italy Department of Biology, University of Bari “Aldo Moro”, Bari, Italy Graduate School of Bioresources Sciences, Akita Prefectural University, Shimoshinjo, Akita, Japan d Network of Apulian Public Research Laboratories “WAFITECH”, Bari, Italy b c

a r t i c l e i n f o

a b s t r a c t

Article history: Received 6 September 2014 Accepted 29 October 2014 Available online 30 October 2014

Assessing the selectivity, regulation and physiological relevance of aquaporin membrane channels (AQPs) requires structural and functional studies of wild type and modified proteins. In particular, when characterizing their transport properties, reconstitution in isolation from native cellular or membrane processes is of pivotal importance. Here, we describe rapid and efficient incorporation of OsPIP1;1, a rice AQP, in liposomes at analytical scale. PIP1;1 was produced as a histidine-tagged form, 10His-OsPIP1;1, in an Escherichia coli-based expression system. The recombinant protein was purified by affinity chromatography and incorporated into liposomes by a micro-batchwise technology using egg-yolk phospholipids and the non-polar Amberlite resin. PIP1;1 proteoliposomes and control empty liposomes had good size homogeneity as seen by quasi-elastic light scattering and electron microscopy analyses. By stoppedflow light scattering, indicating correct protein folding of the incorporated protein, the osmotic water permeability exhibited by the PIP1;1 proteoliposomes was markedly higher than empty liposomes. Functional reconstitution of OsPIP1;1 was further confirmed by the low Arrhenius activation energy (3.37 kcal/mol) and sensitivity to HgCl2, a known AQP blocker, of the PIP1;1-mediated osmotic water conductance. These results provide a valuable contribution in fully elucidating the regulation and waterconducting property of PIP1;1, an AQP that needs to hetero-multimerize with AQPs of the PIP2 subgroup to reach the native plasma membrane and play its role. The micro-batchwise methodology is suitable for the functional reconstitution of whichever AQPs and other membrane transport proteins. © 2014 Elsevier Masson SAS. All rights reserved.

Keywords: Oryza sativa AQP channels PIP1 Osmotic water transport Liposomes Membrane protein reconstitution Micro-batchwise technology

1. Introduction Membrane proteins are arguably one of the most challenging areas of proteomics being central in cell functions including signaling, energy generation, transport and recognition. Membrane proteins are naturally embedded in a lipid bilayer that is, even in the simplest organism, a complex, heterogenous, and dynamic environment. Functional and structural studies of membrane proteins are often hampered by the complexity of the native membranes, a constraint that limits quite a lot the use of standard

Abbreviations: AQP, aquaporin; NMR, nuclear magnetic resonance; PIP, plasma membrane intrinsic protein; OG, N-octyl-b-D-glucopyranoside; RPS, rotating plate stirrer. * Corresponding author. Department of Biosciences, Biotechnologies and Biopharmaceutics, University of Bari “Aldo Moro”, Via Orabona 4, 70125 Bari, Italy. E-mail address: [email protected] (G. Calamita). http://dx.doi.org/10.1016/j.plaphy.2014.10.013 0981-9428/© 2014 Elsevier Masson SAS. All rights reserved.

biophysical techniques such as light scattering, circular dichroism, X-ray crystallography, NMR, etc. Indeed, many biophysical methodologies are not performable in the native environment as they require the protein to be extracted from its native membrane and reconstituted in vitro, into artificial lipid membranes (Girard et al., 2004; Seddon et al., 2004). Hence, a number of distinct approaches have been developed for the detergent-mediated solubilization and subsequent efficient incorporation of purified membrane proteins into liposomes. Aquaporins (AQPs) are a family of membrane channel proteins largely expressd in nature where they play a number of important biological functions (Calamita, 2005; Gena et al., 2011). In membranes, AQPs are organized as tetramers with each monomer forming a single pore. Depending on their transport selectivity, AQPs are grossly subdivided into orthodox aquaporins and aquaglyceroporins although some members of the family remain to be classified (Wu and Beitz, 2007; Ishibashi et al., 2011). Orthodox

V. Scalera et al. / Plant Physiology and Biochemistry 85 (2014) 78e84

AQPs permeate only water while aquaglyceroporins allow passage of glycerol, urea, and some other small neutral solutes, in addition to water. Some AQPs are poorly permeable to water and show ability to conduct anions (Yasui et al., 1999) while some others show specializations in transporting hydrogen peroxide (Almasalmeh et al., 2014; Bienert and Chaumont, 2014), ammonia and/or gases (Boron, 2010; Saparov et al., 2007; Soria et al., 2010, 2013), in addition to water. Tetramer assembly and cellular trafficking is critical for regulating the expression and function of some AQPs. Trafficking of plant plasma membrane intrinsic proteins belonging to the PIP1 subgroup (PIP1s) is facilitated by the heterotetramerization with members of the PIP2 subgroup (Fetter et al., 2004; Maurel et al., 2008; Yaneff et al., 2014; Zelazny et al., 2007). However, valuable information in assessing the transport functions and regulation of AQPs may come from studies involving their functional incorporation into phospholipid vesicles. To date, a number of recombinant AQPs have been functionally reconstituted using methodologies consisting in solubilizing the related proteins in suitable non-ionic detergents at a relatively high micellar concentration followed by their reconstitution into artificial liposomes by dilution approach (Borgnia et al., 1999; Ding et al., 2013; Müller-Lucks et al., 2013; Liu et al., 2013; Zeidel et al., 1994). According to these procedures, the commercially available Escherichia coli polar lipid extract is required as mixture of membrane lipids to prepare liposomes in which the AQP protein is reconstituted. A feasible and convenient method for the rapid functional reconstitution of transport systems in liposomes has been recently described (Spagnoletta et al., 2008) taking advantage of the property possessed by the Amberlite resin to absorb the exceeding nonionic detergent through the interaction of the hydrophobic detergent tail with the hydrophobic surface of the resin. When membrane proteins are present together with the resin, this allows the generation of proteoliposomes by a micro-batchwise technique. With this approach, the reconstitution efficiency is considerably improved when using a rotating plate stirrer instead of the classical column procedure. The micro-batchwise methodology has the important advantage of bringing to the almost complete removal of the detergent leading to tight proteoliposomes where the transport activity of the reconstituted protein is preserved. Unlike classical procedures, this technique results particularly convenient as it can be carried out using the common egg yolk phospholipids instead of polar lipids extracted from E. coli. In the present study, we describe a micro-batchwise-based methodology for the efficient reconstitution of PIP1;1, a plant AQP that has to heteromerize with PIP2s to reach the native plasma membrane playing a major role in mediating transcellular water exchange. Valuable information on the regulation and waterconductance property of PIP1 proteins is therefore provided.

79

recombinant 10His-OsPIP1; 1 protein was purified by Ni-NTA affinity chromatography according to a previous study (Müller-Lucks et al., 2013). 2.2. Functional reconstitution of 10His-OsPIP1; 1 protein According to the rotating plate stirrer (RPS) procedure described by Spagnoletta and coworkers (Spagnoletta et al., 2008), reconstitution of 10His-OsPIP1;1 into liposomes was performed by removing the excess of non-ionic detergent with Bio-Beads made of Amberlite (Bio-Beads SM-2, Bio-Rad, Hercules, CA), a hydrophobic ion-exchange resin, and using a rotating plate stirrer of 25-cm diameter (FALC-F250, Merton, London). The mixture used for the reconstitution of 10His-OsPIP1;1 was prepared by sequentially adding: 10 mM Pipes (pH 7.0), 1 mM NaN3, 1 mM DTT, 1.25% (w/v) N-octyl-b-D-glucopyranoside, 100 mg/ml of purified 10His-OsPIP1; 1 protein, and 6 mg of egg yolk phospholipids (phosphatidylcoline from fresh turkey egg yolk), in a final volume of 700 ml. The reconstitution mixture was vortex mixed and incubated at 4
C for 20 min. The reconstitution mixture was then incubated at room temperature with 0.4 g of Bio-Beads filled into a 2 ml Eppendorf tube under gentle stirring (32 rpm or 480 total revolutions) on a rotating plate stirrer. The 10His-OsPIP1;1 proteoliposomes were removed and collected from the detergent-coated resin by gently aspiration with a 27-gauge syringe. 2.3. Immunoblotting analysis Samples of crude E. coli cell lysate, purified 10His-OsPIP1;1 protein and liposome specimens prepared as above were separated by SDS-PAGE as previously described (Calamita et al., 2008). Acrylamide gels were stained with Coomassie Brilliant Blue (Sigma, St Louis, MO) or submitted to immunoblotting using an anti-6His tag monoclonal antibody (GenScript Corporation, Piscataway, NJ) at a final concentration of 0.2 mg/mL blocking buffer according to a previous study (Soria et al., 2010). 2.4. Electron microscopy

2. Materials and methods

Pellets of empty liposomes or 10His-OsPIP1; 1 proteoliposomes obtained by centrifugation at 152,000 g for 1 h at 4
C were fixed in a mixture of 3% paraformaldehyde and 1% glutaraldehyde in 0.1 mol/l PBS at pH 7.4 for 4 h at 4
C. Specimens were postfixed in 1% OsO4 in PBS for 1 h at 4
C. Fixed specimens were dehydrated in ethanol and then embedded in Epoxy Resin-Araldite (M) CY212 (TAAB, Aldermaston, UK) as previously reported (Mastrodonato et al., 2009). Ultrathin sections were mounted on formwar-coated nickel mesh grids and stained with uranyl acetate and lead citrate (Reynolds, 1963). The grids were observed using a Morgagni 268 electron microscope (FEI, Hillsboro, OR).

2.1. Production and purification of 10His-OsPIP1; 1

2.5. Osmotic water permeability assay

A recombinant form of Oryza sativa PIP1;1 with a 10-histidine fusion tag at its N-terminus (10His-OsPIP1; 1) was produced in an E. coli expression system as previously described (Liu et al., 2013). Briefly, the E. coli strain BL21-CodonPlus (DE3)-RIL (Stratagene, Santa Clara, CA) was transformed with the pTrc10His-10HisOsPIP1;1 plasmid. After inducing protein expression, the E. coli cells were broken and the membrane fraction recovered by centrifugation and dissolved overnight in solubilization buffer composed of 1.25% N-octyl-b-D-glucopyranoside (SigmaeAldrich, St. Louis, MO), 100 mM K2HPO4/KH2PO4, 10% (v/v) glycerol, 5 mM b-mercaptoethanol and 200 mM NaCl, pH 8.0, at 4
C. The

The size of the liposome specimens prepared as described above was determined by quasi-elastic light scattering using an N5 size distribution analyzer (Beckman Coulter Inc., Palo Alto, Ca) and by transmission electron microscopy. Light scattering experiments were carried out to assess the water permeability of the liposome specimens. The time course of liposome volume change was followed from changes in intensity of scattered light at the wavelength of 450 nm using a BioLogic SFM20 stopped-flow reaction analyzer (BioLogic, Claix, France) having a 1.6 ms dead time and 99% mixing efficiency in <1 ms. The sample temperature was kept at 20
C by a circulating water bath.

80

V. Scalera et al. / Plant Physiology and Biochemistry 85 (2014) 78e84

Water permeability was measured as previously reported (Calamita et al., 2006, 2008). Briefly, 100 ml of a concentrated sample suspension were diluted into 2.5 ml of reconstitution mixture having an osmolarity of 220 mOsm. One of the syringes of the stoppedflow apparatus was filled with the specimen suspension, whereas the other was filled with the same buffer to which sucrose was added to reach a final osmolarity of 500 mOsm in order to establish an outwardly directed gradient of sucrose (140 mM) upon mixing. Being sucrose an impermeant solute, after applying the hypertonic gradient, water outflow occurs and liposomes shrink causing an increase in scattered light intensity. The data were fitted to a single exponential function and the related rate constant (Ki, s1) of the water efflux out of the analyzed specimen was measured. The osmotic water permeability coefficient (Pf), an index reflecting the osmotic water permeability of the vesicular membrane, was calculated using the van Heeswijk and van Os equation:

Pf ¼ Ki $V0 =Av $Vw $DC; where Ki is the fitted exponential rate constant, V0 is the initial mean vesicle volume, Av is the mean vesicle surface, Vw is the molar volume of water, and DC is the osmotic gradient. The medium osmolarity was verified by a vapor-pressure osmometer (Wescor Inc., Logan, UT). In some experiments, the liposome specimens were pre-incubated for 5 min with 300 mM HgCl2, a sulfhydryl compound blocking most AQP channels. In other experiments, to verify the blocking action of the Hg2þ ion, the HgCl2 treatment was followed by a 15-min exposure to 10 mM of the reducing agent bmercaptoethanol.

2.6. Statistical analysis Experiments were performed at least in triplicate. All data resulted from at least three independent preparations and were

Fig. 1. Cell-based production of 10His-OsPIP1; 1. Recombinant OsPIP1;1 was produced with a 10-Histidines tag fused at its amino-terminus by employing an E. coli-based system. Total bacterial membrane proteins were resolved by SDS-PAGE and analyzed by Coomassie blue staining (CBS; left lane) or by immunoblotting (IB) using a polyclonal antibody directed against the poly-histidine tag (right lane). The 10His-OsPIP1;1 polypeptide is detected in its monomeric form, at a molecular mass of about 26 kDa (arrow). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

expressed as mean ± SEM. Differences between experimental groups were examined for statistical significance using the Student's t test. P < 0.01 denoted presence of a statistically significant difference. 3. Results 3.1. Heterologous expression and purification of an histidine-tagged form of OsPIP1; 1 As a first step in the study, we employed an E. coli-based expression system to produce a recombinant form of the O. sativa PIP1;1 bearing a 10His tag at its N-terminus (10His-OsPIP1;1). In order to optimize the heterologous expression of OsPIP1;1 in E. coli, the plant preferential codons of OsPIP1;1 were replaced with E. coli preferential codons as done in a previous work (Liu et al., 2013). The resulting recombinant protein was purified by Ni-NTA affinity chromatography. An aliquot of the purified protein was resolved in SDS-PAGE gel and its identity analyzed by immunoblotting by using a polyclonal antibody directed against the poly-His tag (Fig. 1). As expected and in line with a previous work (Liu et al., 2013), an immunoreactive band of about 26 kDa, likely corresponding to the monomeric form of the protein, was observed. 3.2. Incorporation of OsPIP1; 1 into liposomes by a micro-batchwise approach The purified recombinant 10His-OsPIP1;1 protein was reconstituted into liposomes prepared with egg yolk phospholipids following the valuable information acquired in a micro-batchwise methodology described to reconstitute a plant ADP/ATP carrier at analytical scale (Spagnoletta et al., 2008). Proteomicelles were prepared with the non-ionic detergent Noctyl-b-D-glucopyranoside (OG), the 10His-OsPIP1;1 protein and the egg yolk phospholipids as described in the Materials and Methods section. The resulting suspension was exposed to the nonpolar polystyrene Amberlite resin (Bio-Beads SM-2) to remove the excess of detergent bound to the 10His-OsPIP1;1 protein allowing the formation of proteoliposomes. We decided to use an RPS procedure to allow better adsorbing action by the Bio-Beads compared to the usual procedure using the column. In reconstituting 10His-OsPIP1; 1, the incorporation mixture was added of 0.4 g of Bio-Beads, an amount of resin shown to be a good compromise between the extent of protein incorporation and the functional activity of the protein. The recombinant protein was solubilized using OG at a final concentration of 1.25%, a mild nonionic detergent known to solubilize membrane proteins without affecting their native conformation. Purified 10His-OsPIP1;1 protein was added at the final concentration of 100 mg per milliliter of incorporation mixture. The same formulation was followed to make the control liposomes (empty liposomes) except of adding the protein. The morphometry of the prepared liposome specimens was analyzed by both transmission electron microscopy (Fig. 2A,B) and quasi-elastic light scattering (Fig. 2C) while the extent of 10HisOsPIP1;1 protein incorporation was checked by immunoblotting (Fig. 2D). By quasi-elastic light scattering, the mean diameter of the 10His-OsPIP1; 1 proteoliposomes and empty vesicles was of 409 ± 36 nm and 382 ± 14 nm, respectively. These sizes were very close to the diameters of 398 ± 49 nm (n ¼ 1530) and 369 ± 39 nm (n ¼ 1670) showed by the proteoliposomes and control empty liposomes, respectively, when analyzed by electron microscopy. These diameters were consistent with those reported in a previous work where the liposomes prepared by a micro-batchwise approach (Spagnoletta et al., 2008) where characterized

V. Scalera et al. / Plant Physiology and Biochemistry 85 (2014) 78e84

81

Fig. 2. Morphometric and immunoblotting analyses of liposome specimens. (A,B) Electron micrographs showing samples of the 10His-OsPIP1;1 proteoliposomes (A) and control empty liposomes (B) specimens prepared by the micro-batchwise methodology described in the Materials and Methods section. (C) Quasi-elastic light scattering analysis. Gaussian size distribution profiles of proteoliposomes (PL) and empty liposomes (EL) evaluated by volume. (D) Immunoblotting analysis of liposome specimens. Samples of purified 10HisOsPIP1;1 protein (PP; positive control), proteoliposomes (PL) and empty liposomes (EL) were precipitated with trichloroacetic acid and washed before being submitted to immunoblotting with an anti-polyHis antibody. An immunoreactive band of about 26 kDa corresponding to the 10His-OsPIP1;1 polypeptide is seen in the PP and PL lanes but not in the EL lane (arrow). Bar, 0.5 mm.

morphometrically by Non-Invasive Back Scattering (NIBS). By immunoblotting, indicating efficient incorporation of the PIP1 recombinant protein, an immunoreactive band of 26 kDa, likely corresponding to the 10His-OsPIP1;1 monomer, was seen with the proteoliposomes whereas no bands were observed with the control empty liposomes (Fig. 2C). 3.3. Functional analysis of reconstituted 10His-OsPIP1;1 The functionality of the reconstituted 10His-OsPIP1; 1 protein was checked by stopped-flow light scattering. This permitted to assess the rate constant (Ki, s1) of the osmotic shrinkage to which the liposome specimens underwent following an inwardly directed osmotic upshift of 140 mOsm. This series of experiments was also done to confirm whether PIP1;1 was able to exert its water channel activity independently of the physical association (hetero-oligomerization) with PIP2s needed to reach the plasma membrane in native cells. Aliquots of each liposome specimen were subjected rapidly to a hypertonic osmotic gradient of 140 mosM made by the impermeant

solute sucrose and the resulting time course of vesicle shrinkage was followed from the change in scattered light at 20
C, as previously described (Calamita et al., 2008; Liu et al., 2013). The rate constant (Ki) of liposome shrinkage as the result of water conductance activity of reconstituted 10His-OsPIP1;1 was measured as increase in scattered light. The Ki of the 10His-OsPIP1;1 proteoliposomes resulted of 11.2 ± 0.6 s1, a value markedly higher than that of the control empty liposomes (3.4 ± 0.4 s1) suggesting that the reconstituted 10His-OsPIP1;1 water channel was functional (Fig. 3A,B). Functional reconstitution of recombinant PIP1; 1 was confirmed with another series of light scattering experiments where the water permeability of the 10His-OsPIP1;1 proteoliposomes was strongly reduced (z54%) after 5-min treatment with 300 mM HgCl2, a known aquaporin blocker (Fig. 4). In line with this result, no considerable reduction was seen with the control empty liposomes challenged with HgCl2. Proving absence of artifacts, the inhibition exerted by the mercurial compound was reversed following 15 min incubation with 10 mM b-mercaptoethanol, a reducing agent that regenerates the essential thiol-residues of the protein (Fig. 4).

82

V. Scalera et al. / Plant Physiology and Biochemistry 85 (2014) 78e84

Fig. 4. Stopped-flow light scattering analysis of the sensitivity of 10His-OsPIP1;1 proteoliposomes to the AQP blocker HgCl2. The Ki of the proteoliposomes (PL) is strongly reduced after 5 min treatment with 300 mM HgCl2, a mercurial compound known for blocking the AQP water conductance, whereas no changes are seen with the empty liposomes (EL). Indicating absence of artifacts, the inhibition exerted by HgCl2 on the PIP1;1-mediated water conductance was reversed by 15-min treatment with 10 mM of the reducing agent b-mercaptoethanol (b-ME). The b-mercaptoethanol challenging does not change the water permeability of the empty liposome specimen. Data are mean ± S.E.M. from three independent preparations where each preparation yielded at least twelve light scattering traces. *, P < 0.01.

Fig. 3. Water permeability of 10His-OsPIP1;1 vesicles. (A) Representative stopped-flow light scattering traces of 10His-OsPIP1;1 proteoliposomes (PL) and control empty liposomes (EL) in response to a 140 mM inwardly directed osmotic gradient. (B) Indicating efficient protein incorporation, the PIP1;1 proteoliposomes exhibit a higher osmotic water permeability (Pf; mm/s) than empty liposomes. Data are mean ± S.E.M. from four independent preparations where each preparation yielded at least twelve light scattering traces. *, P < 0.01.

becomes particularly important when characterizing the function of a single transporter or transport complex in isolation from other cellular or membrane processes. The present study describes a rapid and efficient methodology to reconstitute functionally the O. sativa PIP1;1 at analytical scale. The methodology appears to be suitable for investigating other AQPs or, more in general, all membrane transport proteins, especially those with complex trafficking/regulation dynamics. A valuable aspect of this work comes from the easiness and convenience with which PIP1;1, an AQP that has been shown to hetero-oligomerize with PIP2s to reach the native membrane (Fetter et al., 2004; Yaneff et al., 2014; Zelazny et al., 2007), was incorporated into liposomes. PIP1;1, one of the most abundant AQPs expressed in rice leaves and roots, belongs to the subgroup of PIP1s, AQPs acting as major gateways of transcellular water

To examine further both the incorporation and the water permeation property of 10His-OsPIP1;1, another series of stoppedflow light scattering experiments was run to determine the Arrhenius activation energy (Ea) of the osmotic water transport of the liposome specimens from temperature dependence data at 10, 20 and 30
C. A slight increase in the Ki was observed when the 10His-OsPIP1;1 proteoliposomes were subjected to the same osmotic gradient (140 mosM) at increasing temperatures. A relatively high increase in water permeability was observed with the empty liposomes, instead (Fig. 5). Plotting of the Ki values obtained at the different temperatures resulted in Arrhenius activation energies (Ea) values of 3.37 and 10.43 kcal/mol for the proteoliposomes and empty liposomes, respectively. The low value of Ea seen with the proteoliposomes was biophysically consistent with the PIP1;1 incorporation and the channel-mediated permeation of water.

4. Discussion Functional reconstitution of transport systems in liposomes permits manipulating conditions on either membrane sides. This

Fig. 5. Stopped-flow light scattering analysis of the temperature dependence of proteoliposomes and empty liposomes water permeability. Rate constants of shrinkage of proteoliposomes and liposomes subjected to osmotic upshifts of 140 mOsm at 10, 20 and 30
C. A slight increase in Ki is seen with the 10His-OsPIP1;1 liposomes whereas the increase in Ki showed by the empty liposomes is relatively high. Arrhenius plotting of the Ki values results in activation energies (Ea) of 3.37 and 10.43 kcal/mol for the proteoliposomes and empty liposomes, respectively. *, P < 0.01.

V. Scalera et al. / Plant Physiology and Biochemistry 85 (2014) 78e84

exchange in plant plasma membranes (Maurel et al., 2008). PIP1;1 is suggested to be involved in important biological processes such as water homeostasis, stress tolerance and seed germination (Li et al., 2000; Lian et al., 2004; Liu et al., 2013; Maurel et al., 2008; Sade et al., 2014; Yu et al., 2006). A PIP1 AQP expressed in tobacco leaves, NtAQP1, has been reported to act as CO2 pore having a function in photosynthesis and stomatal opening (Uehlein et al., 2003). The PIP1 genes of the fast growing tree Populus tremula x alba have been recently suggested to be of particular critical relevance in the leave water and CO2 uptake underlying photosynthesis in conditions of water stress (Secchi and Zwieniecki, 2013). To our knowledge, this is the first time egg yolk phospholipids are employed to prepare functional AQP proteomicelles instead of using the costly E. coli polar lipids. So far, egg yolk phospholipids had been employed to reconstitute a variety of membrane proteins including carriers and membrane transporters such as the 2oxoglutarate mitochondrial carrier (Indiveri et al., 1987), the lysosomal sialic acid carrier (Mancini et al., 1992), and the aminoacid transport system L (Yao et al., 1993). The micro-batchwise method using egg yolk phospholipids for preparing the reconstitution mixture may be also suitable for the rapid reconstitution of other AQPs at analytical scale. As regards the technology, keeping the reconstitution mixture in a tube on a rotating plate stirrer (RPS) in the presence of detergent removing Bio-beads provided a number of advantages in comparison to the usual method of repetitively passing the reconstitution mixture through a column. First, the use of the RPS was time saving and did not require the presence of the operator during the reconstitution. Second, as regards the efficiency of reconstitution into liposomes, besides allowing high protein incorporation, the extent of functional PIP1;1 appeared to be elevated. Based on the values of rate constant (Ki), morphometric data and osmotic gradient and using the van Heeswijk and van Os equation, the coefficient of osmotic membrane permeability (Pf) of proteoliposomes and empty liposomes can be calculated resulting of 297.2 ± 17.1 and 83.6 ± 9.8 mm/s (P < 0.01), respectively. The Pf of the PIP1;1 proteoliposomes is considerably high, in line with the low Arrhenius activation energy (Ea; 3.37 kcal/mol) that was calculated based on temperature dependence data (Fig. 5). In fact, Ea values lower than 6 kcal/mol are synonymous of facilitated diffusion whereas values higher than 10 kcal/mol normally relate to simple diffusion across the phospholipid bilayer. The Pf of the empty liposomes analyzed in this study results considerably higher than the Pf of empty liposomes measured in a previous work (Liu et al., 2013) (83 and 10 mm/s, respectively). The different intrinsic membrane permeability/elasticity between the present and previous liposomes may come from differences in the lipid composing (the present work was using egg yolk phospholipids whereas the previous one was employing E. coli polar lipids) and/or differences in the methodologies employed to make the liposomes (microbatchwise vs. dilution methodology). The orientation of 10HisOsPIP1;1 within the liposomes did not seem to have effects on the water permeation measurements, an observation reflecting the bidirectionality of the AQP channel proteins (Meinild et al., 1998). A third advantage of having recourse to the micro-batchwise approach regarded the quality of the prepared liposomes. By both electron microscopy and quasi-elastic scattering, the diameter of the micro-batchwise-made liposomes was reasonably homogenous ranging between 370 and 440 nm. As shown in Fig. 2C, the diameter of both liposome specimens was distributed according to a Gaussian distribution. Diameter homogeneity is critical when performing kinetic studies such as the stopped-flow light scattering studies carried out in this work where the initial rate of water flux across the liposome membrane was particularly elevated. The micro-batchwise method used in this study was also highly efficient allowing incorporation of fully functional OsPIP1;1. The

83

proteoliposomes incorporating 10His-OsPIP1;1 were about 3.5 more permeable to water than control empty liposomes. This observation also suggests that the presence of the 10His tag does not affect the water permeation property of the pore formed by recombinant OsPIP1;1. On the contrary, as expected, the pore was blocked following treatment with the sulfhydryl compound HgCl2, an agent blocking most AQPs. The cysteine at position 244 in the OsPIP1;1 primary sequence may represent the residue responsible for the Hg2þ inhibition. Altogether, these results indicate fully functionality for the reconstituted protein channel. Importantly, this confirms the water-conducting property of PIP1;1 and shows that such AQP is able to permeate water independently of the oligomerization state and trafficking constraints that characterize the protein in native cells. PIP1s were described as being inactive AQP water channels (Sakurai et al., 2005). The inability in conducting water seen with some PIP1 isoforms expressed in Xenopus oocytes was explained as due to the their failure to traffic to the oocyte plasma membrane (Fetter et al., 2004). In vivo, PIP1;1 has to hetero-tetramerize with PIP2s to reach the plasma membrane (Bellati et al., 2010; Jozefkowicz et al., 2013; Yaneff et al., 2014; Zelazny et al., 2007). The key participation of PIP1;1 in plant plasma membrane water permeability is also indicated by the observation that PIP1-PIP2 hetero-multimerization enhances the water conductance activity of PIP2; 1 (Yaneff et al., 2014). In conclusion, our work describes a micro-batchwise methodology for the rapid and efficient reconstitution of rice PIP1;1 at analytical scale. Valuable information is provided in elucidating the regulatory mechanism and water-conducting property of major plant water channels such as PIP1s. The devised method appears to be also suitable for the efficient reconstitution of whichever AQP and other membrane transport proteins. Author contributions Conceived the experiments: GC, VS. Designed the experiments: GC, PG, VS. Performed the experiments: PG, GC, SC, MM, VS. Analyzed the data: GC, PG, VS, MM. Contributed reagents/materials/analysis tools: GC, YK, VS. Wrote the paper: GC. Reviewed/edited manuscript: GC, PG, VS, MM, MS. Researched data: GC, PG, VS, YK. Acknowledgments Financial support from Regione Puglia (PO Puglia FESR 20072013, Asse I, Linea 1.2, Rete di Laboratori Pubblici di Ricerca WAFITECH) to G.C. is gratefully acknowledged. References Almasalmeh, A., Krenc, D., Wu, B., Beitz, E., 2014. Structural determinants of the hydrogen peroxide permeability of aquaporins. FEBS J. 281 (3), 647e656. Bellati, J., Alleva, K., Soto, G., Vitali, V., Jozefkowicz, C., Amodeo, G., 2010. Intracellular pH sensing is altered by plasma membrane PIP aquaporin co-expression. Plant Mol. Biol. 74, 105e118. Bienert, G.P., Chaumont, F., 2014. Aquaporin-facilitated transmembrane diffusion of hydrogen peroxide. Biochim. Biophys. Acta 1840, 1596e1604. Borgnia, M.J., Kozono, D., Calamita, G., Maloney, P.C., Agre, P., 1999. Functional reconstitution and characterization of AqpZ, the E. coli water channel protein. J. Mol. Biol. 291, 1169e1179. Boron, W.F., 2010. Sharpey-Schafer lecture: gas channels. Exp. Physiol. 95, 1107e1130. Calamita, G., 2005. Aquaporins: highways for cells to recycle water with the outside world. Biol. Cell. 97, 351e353. Calamita, G., Gena, P., Meleleo, D., Ferri, D., Svelto, M., 2006. Water permeability of rat liver mitochondria: a biophysical study. Biochim. Biophys. Acta 1758, 1018e1024.

84

V. Scalera et al. / Plant Physiology and Biochemistry 85 (2014) 78e84

Calamita, G., Ferri, D., Gena, P., Carreras, F.I., Liquori, G.E., Portincasa, P., Marinelli, R.A., Svelto, M., 2008. Altered expression and distribution of aquaporin-9 in the liver of rat with obstructive extrahepatic cholestasis. Am. J. Physiol. Gastrointest. Liver Physiol. 295, G682eG690. Ding, X., Matsumoto, T., Gena, P., Liu, C., Pellegrini-Calace, M., Zhong, S., Sun, X., Zhu, Y., Katsuhara, M., Iwasaki, I., Kitagawa, Y., Calamita, G., 2013. Water and CO2 permeability of SsAqpZ, the cyanobacterium Synechococcus sp. PCC7942 aquaporin. Biol. Cell. 105, 118e128. Fetter, K., Van Wilder, V., Moshelion, M., Chaumont, F., 2004. Interactions between plasma membrane aquaporins modulate their water channel activity. Plant Cell. 16, 215e228. Gena, P., Pellegrini-Calace, M., Biasco, A., Svelto, M., Calamita, G., 2011. Aquaporin membrane channels: biophysics, classification, functions, and possible biotechnological applications. Food Biophys. 6, 241e249. Girard, P., Julicher, F., Prost, J., 2004. Fluid membranes exchanging material with external reservoirs. Eur. Phys. J. E. Soft Matter 14, 387e394. Indiveri, C., Palmieri, F., Bisaccia, F., Kramer, R., 1987. Kinetics of the reconstituted 2oxoglutarate carrier from bovine heart mitochondria. Biochim. Biophys. Acta 890, 310e318. Ishibashi, K., Kondo, S., Hara, S., Morishita, Y., 2011. The evolutionary aspects of aquaporin family. Am. J. Physiol. Regul. Integr. Comp. Physiol. 300, R566eR576. Jozefkowicz, C., Rosi, P., Sigaut, L., Soto, G., Pietrasanta, L.I., Amodeo, G., Alleva, K., 2013. Loop A is critical for the functional interaction of two Beta vulgaris PIP aquaporins. PloS One 8, e57993. Li, L., Li, S., Tao, Y., Kitagawa, Y., 2000. Molecular cloning of a novel water channel from rice: its products expression in Xenopus oocytes and involvement in chilling tolerance. Plant Sci. 154, 43e51. Lian, H.L., Yu, X., Ye, Q., Ding, X., Kitagawa, Y., Kwak, S.S., Su, W.A., Tang, Z.C., 2004. The role of aquaporin RWC3 in drought avoidance in rice. Plant Cell. Physiol. 45, 481e489. Liu, C., Fukumoto, T., Matsumoto, T., Gena, P., Frascaria, D., Kaneko, T., Katsuhara, M., Zhong, S., Sun, X., Zhu, Y., Iwasaki, I., Ding, X., Calamita, G., Kitagawa, Y., 2013. Aquaporin OsPIP1;1 promotes rice salt resistance and seed germination. Plant Physiol. Biochem. 63, 151e158. Mancini, G.M., Beerens, C.E., Galjaard, H., Verheijen, F.W., 1992. Functional reconstitution of the lysosomal sialic acid carrier into proteoliposomes. Proc. Natl. Acad. Sci. U S A. 89, 6609e6613. Mastrodonato, M., Calamita, G., Rossi, R., Scillitani, G., Liquori, G.E., Ferri, D., 2009. Expression of H(þ),K(þ)-ATPase and glycopattern analysis in the gastric glands of Rana esculenta. J. Histochem. Cytochem 57, 215e225. Maurel, C., Verdoucq, L., Luu, D.T., Santoni, V., 2008. Plant aquaporins: membrane channels with multiple integrated functions. Annu. Rev. Plant Biol. 59, 595e624. Meinild, A.K., Klaerke, D.A., Zeuthen, T., 1998. Bidirectional water fluxes and specificity for small hydrophilic molecules in aquaporins 0-5. J. Biol. Chem. 273, 32446e32451. Müller-Lucks, A., Gena, P., Frascaria, D., Altamura, N., Svelto, M., Beitz, E., Calamita, G., 2013. Preparative scale production and functional reconstitution of a human aquaglyceroporin (AQP3) using a cell free expression system. New. Biotechnol. 30, 545e551.

Reynolds, E.S., 1963. The use of lead citrate at high pH as an electron-opaque stain in electron microscopy. J. Cell. Biol. 17, 208e212. Sade, N., Shatil-Cohen, A., Attia, Z., Maurel, C., Boursiac, Y., Kelly, G., Granot, D., Yaaran, A., Lemer, S., Moshelion, M., 2014. The role of plant membrane aquaporins in regulating the bundle sheath-mesophyll continuum and leaf hydraulics. Plant Physiol. http://dx.doi.org/10.1104/pp.114.248633 [In press]. Sakurai, J., Ishikawa, F., Yamaguchi, T., Uemura, M., Maeshima, M., 2005. Identification of 33 rice aquaporin genes and analysis of their expression and function. Plant Cell. Physiol. 46, 1568e1577. Saparov, S.M., Liu, K., Agre, P., Pohl, P., 2007. Fast and selective ammonia transport by aquaporin-8. J. Biol. Chem. 282, 5296e5301. Secchi, F., Zwieniecki, M., 2013. The physiological response of Populus tremula x alba leaves to the down-regulation of PIP1 aquaporin gene expression under no water stress. Front. Plant Sci. 4, 507. Seddon, A.M., Curnow, P., Booth, P.J., 2004. Membrane proteins, lipids and detergents: not just a soap opera. Biochim. Biophys. Acta 1666, 105e117. Soria, L.R., Fanelli, E., Altamura, N., Svelto, M., Marinelli, R.A., Calamita, G., 2010. Aquaporin-8-facilitated mitochondrial ammonia transport. Biochem. Biophys. Res. Commun. 393, 217e221. Soria, L.R., Marrone, J., Calamita, G., Marinelli, R.A., 2013. Ammonia detoxification via ureagenesis in rat hepatocytes involves mitochondrial aquaporin-8 channels. Hepatology 57, 2061e2071. Spagnoletta, A., De Palma, A., Prezioso, G., Scalera, V., 2008. A micro-batchwise technique method for rapid reconstitution of functionally active mitochondrial ADP/ATP carrier from Jerusalem artichoke (Helianthus tuberosus L.) tubers. J. Biochem.. Biophys. Methods 70, 954e957. Uehlein, N., Lovisolo, C., Siefritz, F., Kaldenhoff, R., 2003. The tobacco aquaporin NtAQP1 is a membrane CO2 pore with physiological functions. Nature 425, 734e737. Wu, B., Beitz, E., 2007. Aquaporins with selectivity for unconventional permeants. Cell. Mol. Life Sci. 64, 2413e2421. Yaneff, A., Sigaut, L., Marquez, M., Alleva, K., Pietrasanta, L.I., Amodeo, G., 2014. Heteromerization of PIP aquaporins affects their intrinsic permeability. Proc. Natl. Acad. Sci. U S A 111, 231e236. Yao, S.Y., George, R., Young, J.D., 1993. Reconstitution studies of amino acid transport system L in rat erythrocytes. Biochem.. J. 292, 655e660. Yasui, M., Kwon, T.H., Knepper, M.A., Nielsen, S., Agre, P., 1999. Aquaporin-6: an intracellular vesicle water channel protein in renal epithelia. Proc. Natl. Acad. Sci. U S A 96, 5808e5813. Yu, X., Peng, Y.H., Zhang, M.H., Shao, Y.J., Su, W.A., Tang, Z.C., 2006. Water relations and an expression analysis of plasma membrane intrinsic proteins in sensitive and tolerant rice during chilling and recovery. Cell. Res. 16, 599e608. Zeidel, M.L., Nielsen, S., Smith, B.L., Ambudkar, S.V., Maunsbach, A.B., Agre, P., 1994. Ultrastructure, pharmacologic inhibition, and transport selectivity of aquaporin channel-forming integral protein in proteoliposomes. Biochemistry 33, 1606e1615. Zelazny, E., Borst, J.W., Muylaert, M., Batoko, H., Hemminga, M.A., Chaumont, F., 2007. FRET imaging in living maize cells reveals that plasma membrane aquaporins interact to regulate their subcellular localization. Proc. Natl. Acad. Sci. U S A 104, 12359e12364.