Isoprenoid modification of proteins distinct from membrane acyl proteins in the prokaryote Acholeplasma laidlawii.

Isoprenylation is an important posttranslational modification that affects the activity, subunit interactions and membrane anchoring of different eukaryotic proteins. The small, cell-wall-less prokaryote Acholeplasma laidlawii has more than 20 membrane acyl-proteins enriched in myristoyl and palmitoyl chains. Radioactive mevalonate, a precursor to isoprenoids, was incorporated into several specific membrane proteins of 20 to 45 kDa and two soluble proteins of 23-25 kDa, respectively. No acyl proteins and none of the polar acyl lipids became labelled but these are all labelled by radioactive fatty acids. Mevalonate was incorporated mainly into a minor neutral, non-saponifiable lipid which migrated just above a C30-isoprenoid (squalene) on TLC-plates. The isoprenoid chains could not be released by mild alkaline hydrolysis from most of the isoprenylated proteins, although this procedure releases acyl chains from lipids and all acylated proteins. Isoprenylated proteins were enriched in the detergent phase upon partition with the non-ionic detergent Triton X-114. This behaviour is similar to the acyl proteins of this organism and indicates that the isoprenoid chains give the proteins a hydrophobic character.     

Determining the stoichiometry and interactions of macromolecular assemblies from mass spectrometry

The growing number of applications to determine the stoichiometry, interactions and even subunit architecture of protein complexes from mass spectra suggests that some general guidelines can now be proposed. In this protocol, we describe the necessary steps required to maintain interactions between subunits in the gas phase. We begin with the preparation of suitable solutions for electrospray (ES) and then consider the transmission of complexes through the various stages of the mass spectrometer until their detection. Subsequent steps are also described, including the dissociation of these complexes into multiple subcomplexes for generation of interaction networks. Throughout we highlight the critical experimental factors that determine success. Overall, we develop a generic protocol that can be carried out using commercially available ES mass spectrometers without extensive modification.     

Finding the right balance: a personal journey from individual proteins to membrane-embedded motors: based on a lecture delivered at the 36th FEBS Congress in Torino, Italy, June 2011

It is now more than 20 years since the prophetic words of John Fenn, announcing his discovery, stated that 'Electrospray spectra have been obtained for biopolymers including oligonucleotides and proteins, the latter having molecular weights up to 130 000, with as yet no evidence of an upper limit' (Fenn JB, Mann M, Kai Meng C, Fu Wong S & Whitehouse CM (1989) Science246, 64-71). Today, with the mass spectra of intact ribosomes at 2.3 MDa becoming almost routine and the first electrospray spectra of membrane-embedded motors being recorded recently, new challenges are emerging. Knowledge of the intact mass of a protein or complex is only part of the MS information available. Data from the disruption of protein complexes in solution and gas phases are leading to subunit interaction maps and architectural models. Such models are enhanced by coupling with ion mobility in which the collision cross-section of a protein complex can be defined. Linking these attributes with knowledge of subunit dynamics and the role of post-translational modifications on the stability and interactions within complexes is increasing our understanding of the factors that stabilize and convert protein complexes between different quaternary states. From our earliest experiments, studying the folding of individual proteins, through to the characterization of membrane-embedded motors, it is clear that the full potential of electrospray in structural biology has yet to be realized. The present review offers a personal view of the transition from determining the mass of an individual protein to elucidating the structure and dynamics of heterogeneous assemblies in the megadalton mass range.     

Molecular architecture of the thylakoid membrane: lipid diffusion space for plastoquinone

We have determined the stoichiometric composition of membrane components (lipids and proteins) in spinach thylakoids and have derived the molecular area occupied by these components. From this analysis, the lipid phase diffusion space, the fraction of lipids located in the first protein solvation shell (boundary lipids), and the plastoquinone (PQ) concentration are derived. On the basis of these stoichiometric data, we have analyzed the motion of PQ between photosystem (PS) II and cytochrome (cyt.) bf complexes in this highly protein obstructed membrane (protein area about 70%) using percolation theory. This analysis reveals an inefficient diffusion process. We propose that distinct structural features of the thylakoid membrane (grana formation, microdomains) could help to minimize these inefficiencies and ensure a non-rate limiting PQ diffusion process. A large amount of published evidence supports the idea that higher protein associations exist, especially in grana thylakoids. From the quantification of the boundary lipid fraction (about 60%), we conclude that protein complexes involved in these associations should be spaced by lipids. Lipid-spaced protein aggregations in thylakoids are qualitatively different to previously characterized associations (multisubunit complexes, supercomplexes). We derive a hierarchy of protein and lipid interactions in the thylakoid membrane.     

Laser‐induced liquid bead ion desorption‐MS of protein complexes from blue‐native gels,a sensitive top‐down proteomic approach

We have developed an experimental approach that combines two powerful methods for proteomic analysis of large membrane protein complexes: blue native electrophoresis (BNE or BN‐PAGE) and laser‐induced liquid bead ion desorption (LILBID) MS. Protein complexes were separated by BNE and eluted from the gel. The masses of the constituents of the multiprotein complexes were obtained by LILBID MS, a detergent‐tolerant method that is especially suitable for the characterisation of membrane proteins. High sensitivity and small sample volumes required for LILBID MS resulted in low demands on sample quantity. Eluate from a single band allowed assessing the mass of an entire multiprotein complex and its subunits. The method was validated with mitochondrial NADH:ubiquinone reductase from Yarrowia lipolytica. For this complex of 947 kDa, typically 30 μg or 32 pmol were sufficient to obtain spectra from which the subunit composition could be analysed. The resolution of this electrophoretic small‐scale approach to the purification of native complexes was improved markedly by further separation on a second dimension of BNE. Starting from a subcellular fraction obtained by differential centrifugation, this allowed the purification and analysis of the constituents of a large multiprotein complex in a single LILBID spectrum.     

Characterization of the Tween 20-soluble membrane proteins of Acholephasma laidlawii

Methods for the purification of the four major proteins and two of the minor proteins in the Tween 20-soluble extract of A. laidlawii membranes have previously been described. The last step in the purification procedure involved an electrophoresis in a detergent-free buffer, where the concentration of Tween could be reduced by up to 2000 times. The purified proteins were found to remain soluble after removal of the bulk of the detergent. Solutions of the different protein samples contained 3 30 detergent molecules per protein molecule as determined by gas-liquid liquid chromatography. Some of the protein solutions also contained natural membrane lipids. It was probably only a fraction of detergent molecules and lipids, which was bound to the protein. Complete amino acid analysis showed that none of the proteins contained amino sugars and only one of them contained half-cystine. The specific absorbances and the molar absorption coefficients were calculated from the absorption spectra. The hydrophobicities and the partial specific volumes were calculated from the amino acid composition. The hydrophobicity values did not differ significantly from those of non-membrane proteins. Attempts to determine the sedimentation coefficients and the molecular weights were done ultracentrifugation after removal of the bulk of the detergent. The molecular weights, as determined by ultracentrifugation, were in general higher than the molecular weights determined by polyacrylamid-gel electrophoresis in sodium dodecyl sulphate (SDS), indicating that most of the proteins formed aggregates upon reducing the concentration of Tween 20. The size of these aggregates was not influenced by storage of the proteins at 0 C but it seemed to be highly affected by the speed and the time of centrifugation. The electrophoretic mobolities of the proteins were determined by free zone electrophoresis. Crossed immunoelectrophoresis was utilized to demonstrate that the Tween 20-soluble membrane proteins were not undergoing proteolysis during the preparation procedure.     

F<Subscript>1</Subscript>F<Subscript>0</Subscript>-ATP Synthase Complex Interactions <Emphasis Type="Italic">In Vivo</Emphasis> Can Occur in the Absence of the Dimer Specific Subunit e

Evidence suggests membrane bound F₁F₀-ATPase complexes form stable associations such that dimers can be retrieved from detergent lysates of mitochondria isolated from a range of sources including algae, higher plants, yeast and bovine heart, and plant chloroplasts. The physiological relevance of these interactions is not clear but may be connected with the formation and structure of mitochondrial cristae. We sought to demonstrate, in vivo, the association of F₁F₀-ATPases in yeast cells co-expressing two b subunits each fused at its C-terminus to a GFP variant appropriate for fluorescence resonance energy transfer (FRET; BFP as the donor and GFP as the acceptor fluorophore). Both subunit b-GFP and b-BFP fusions were assembled into functional complexes. FRET was observed from enzyme complexes in molecular proximity in respiring cells providing the first demonstration of the association, in vivo, of F₁F₀-ATPase complexes. Moreover, FRET was observed within cells lacking the dimer specific subunit e, indicating structured associations can occur within the inner membrane in the absence of subunit e.     

A role for lipids in the functional and structural properties of the rat liver aminoacyl-tRNA synthetase complex

The high molecular weight aminoacyl-tRNA synthetase complexes found in extracts of many eukaryotic cells often contain lipids and other non-protein components. Since hydrophobic interactions play an important role in maintaining synthetases in the complex, it has been suggested that the lipids present may also participate in its functional and structural integrity. In order to learn more about the role of lipids in the complex, we have compared the properties of the normal complex to one which has been delipidated by treatment with Triton X-114. Delipidation does not affect the size or activity of the aminoacyl-tRNA synthetase complex, but a variety of functional and structural properties of individual synthetases in the complex are altered dramatically. These include sensitivity to salts plus detergents, temperature inactivation, hydrophobicity, and sensitivity to protease digestion. In the latter case, removal of lipids also affects the low molecular weight products released by protease digestion. Purification of the synthetase complex by various chromatographic procedures can remove the lipids and lead to a structure that behaves like the delipidated complex prepared by detergent treatment. The significance of these findings for the intracellular location of aminoacyl-tRNA synthetases and for the study of purified complexes are discussed.     

Characterization of multiprotein complexes of the Borrelia burgdorferi outer membrane vesicles

Among bacterial cell envelopes, the Borrelia burgdorferi outer membrane (OM) is structurally unique in that the identities of many protein complexes remain unknown; however, their characterization is the first step toward our understanding of membrane protein interactions and potential functions. Here, we used two-dimensional blue native/SDS-PAGE/mass spectrometric analysis for a global characterization of protein-protein interactions as well as to identify protein complexes in OM vesicles isolated from multiple infectious sensu stricto isolates of B. burgdorferi. Although we uncovered the existence of at least 10 distinct OM complexes harboring several unique subunits, the complexome is dominated by the frequent occurrence of a limited diversity of membrane proteins, most notably P13, outer surface protein (Osp) A, -B, -C, and -D and Lp6.6. The occurrence of these complexes and specificity of subunit interaction were further supported by independent two-dimensional immunoblotting and coimmunoprecipitation assays as well as by mutagenesis studies, where targeted depletion of a subunit member (P66) selectively abolished a specific complex. Although a comparable profile of the OM complexome was detected in two major infectious isolates, such as B31 and 297, certain complexes are likely to occur in an isolate-specific manner. Further assessment of protein complexes in multiple Osp-deficient isolates showed loss of several protein complexes but revealed the existence of additional complex/subunits that are undetectable in wild-type cells. Together, these observations uncovered borrelial antigens involved in membrane protein interactions. The study also suggests that the assembly process of OM complexes is specific and that the core or stabilizing subunits vary between complexes. Further characterization of these protein complexes including elucidation of their biological significance may shed new light on the mechanism of pathogen persistence and the development of preventative measures against the infection.     

Morphological analysis on the distribution of membrane lipids and a membrane protein,NAP-22, during neuronal development in vitro

Specific localization of membrane proteins based on the interactions with membrane lipids at various microdomains (MDs) is under active investigation, since the elucidation of the molecular mechanism of the interactions could reveal a novel concept of cell organization. Due to the strong interactions not only between lipids but also between lipids and proteins, these MDs are considered to be recovered in a detergent-resistant low-density membrane fraction (DRM) after detergent extraction and density-gradient centrifugation. Neurons take well-developed membrane systems during maturation and specific localization of various membrane components, not only proteins but also lipids, is essential for the establishment of the nervous system. In previous studies, we showed that NAP-22 is a major protein of neuronal DRM and binds liposomes in a cholesterol-dependent manner. In this study, we analyzed the localization of membrane lipids during neuronal maturation in vitro and compared their distribution with that of NAP-22. In an attempt to detect DRM-associated lipids, we observed the staining patterns of neurons treated with Triton-X-100 at 4 degrees C before fixation. Our results showed the less staining patterns of cholesterol and sphingomyelin at the axonal tips and a different staining pattern of two gangliosides, GM(1) and GD(3). The enrichment of cholesterol at the NAP-22 localizing spots was observed after the treatment of the detergent. Since the application of maitotoxin, a calcium ion channel, caused the diminution of NAP-22 and cholesterol positive spots, the distribution of these molecules are considered under the calcium regulation.     

Subunit mass fingerprinting of mitochondrial complex I

We have employed laser induced liquid bead ion desorption (LILBID) mass spectrometry to determine the total mass and to study the subunit composition of respiratory chain complex I from Yarrowia lipolytica. Using 5-10 pmol of purified complex I, we could assign all 40 known subunits of this membrane bound multiprotein complex to peaks in LILBID subunit fingerprint spectra by comparing predicted protein masses to observed ion masses. Notably, even the highly hydrophobic subunits encoded by the mitochondrial genome were easily detectable. Moreover, the LILBID approach allowed us to spot and correct several errors in the genome-derived protein sequences of complex I subunits. Typically, the masses of the individual subunits as determined by LILBID mass spectrometry were within 100 Da of the predicted values. For the first time, we demonstrate that LILBID spectrometry can be successfully applied to a complex I band eluted from a blue-native polyacrylamide gel, making small amounts of large multiprotein complexes accessible for subunit mass fingerprint analysis even if they are membrane bound. Thus, the LILBID subunit mass fingerprint method will be of great value for efficient proteomic analysis of complex I and its assembly intermediates, as well as of other water soluble and membrane bound multiprotein complexes.     

Two-dimensional benzyldimethyl-n-hexadecylammonium chloride/SDS-PAGE for membrane proteomics

Despite the importance of membranes in any living system, the global analysis of membrane subproteomes is still a common obstacle. In particular, the widely used 2-DE technique consisting of IEF in the first dimension and SDS-PAGE in the second dimension has some major drawbacks regarding the separation of hydrophobic proteins. Therefore, we applied an alternative electrophoretic technique for separating membrane proteins: two-dimensional BAC/SDS electrophoresis (2-DB) using the cationic detergent benzyldimethyl-n-hexadecylammonium chloride in the first and the anionic detergent SDS in the second dimension. The use of 2-DB resulted in an improved separation of hydrophobic proteins. Thus, extremely hydrophobic proteins such as cytochrome-c oxidase subunit I with a grand average hydrophobicity (GRAVY) index of 0.74 and a total of 12 known transmembrane domains (TMD) or Sec61alpha with a GRAVY index of 0.56 and a total of ten known TMD could be identified by MS/MS analyses of protein spots derived from 2-DB gels.     

Interaction of the K+ channel KcsA with membrane phospholipids as studied by ESI mass spectrometry

In this study we have used electrospray ionization mass spectrometry (ESI-MS) to investigate interactions between the bacterial K(+) channel KcsA and membrane phospholipids. KcsA was reconstituted into lipid vesicles of variable lipid composition. These vesicles were directly analyzed by ESI-MS or mixed with trifluoroethanol (TFE) before analysis. In the resulting mass spectra, non-covalent complexes of KcsA and phospholipids were observed with an interesting lipid specificity. The anionic phosphatidylglycerol (PG), and, to a lesser extent, the zwitterionic phosphatidylethanolamine (PE), which both are abundant bacterial lipids, were found to preferentially associate with KcsA as compared to the zwitterionic phosphatidylcholine (PC). These preferred interactions may reflect the differences in affinity of these phospholipids for KcsA in the membrane.     

Free fatty acids associated with the high molecular weight aminoacyl-tRNA synthetase complex influence its structure and function

Aminoacyl-tRNA synthetases from higher eukaryotes often are isolated as high molecular weight complexes associated with other components such as lipids. Since hydrophobic interactions are involved in the organization of the complex, it has been suggested that interaction of synthetases with these lipids might be important for their structure and function. Delipidation is known to affect certain properties of synthetases within the complex including sensitivity to detergents plus salts, temperature inactivation, hydrophobicity, sensitivity to proteases, and, as shown here, sensitivity to p-mercuribenzoate and sites of papain cleavage. Of the lipids known to co-purify with the complex, cholesterol esters, phospholipids and free fatty acids, we show that the particular lipids responsible for many of these changes are the free fatty acids. Specific removal of fatty acids results in a complex with properties similar to one totally delipidated by detergent treatment, and readdition of the fatty acid fraction reverses the effects. The fatty acid fraction contains both saturated and unsaturated fatty acids, but unsaturated fatty acids are much more effective in reversing the properties of the delipidated complex that are saturated fatty acids. These results indicate that the free fatty acids co-purifying with the synthetase complex bind to the synthetases and affect their structure and function.     

Phospholipid complexation and association with apolipoprotein C-II: insights from mass spectrometry

The interactions between phospholipid molecules in suspensions have been studied by using mass spectrometry. Electrospray mass spectra of homogeneous preparations formed from three different phospholipid molecules demonstrate that under certain conditions interactions between 90 and 100 lipid molecules can be preserved. In the presence of apolipoprotein C-II, a phospholipid binding protein, a series of lipid molecules and the protein were observed in complexes. The specificity of binding was demonstrated by proteolysis; the resulting mass spectra reveal lipid-bound peptides that encompass the proposed lipid-binding domain. The mass spectra of heterogeneous suspensions and their complexes with apolipoprotein C-II demonstrate that the protein binds simultaneously to two different phospholipids. Moreover, when apolipoprotein C-II is added to lipid suspensions formed with local concentrations of the same lipid molecule, the protein is capable of remodeling the distribution to form one that is closer to a statistical arrangement. These observations demonstrate a capacity for apolipoprotein C-II to change the topology of the phospholipid surface. More generally, these results highlight the fact that mass spectrometry can be used to probe lipid interactions in both homogeneous and heterogeneous suspensions and demonstrate reorganization of the distribution of lipids upon surface binding of apolipoprotein C-II.     

NMR Characterization and Membrane Interactions of the Loop Region of Kindlin-3 F1 Subdomain

Kindlins-1,2 and 3 are FERM domain-containing cytosolic proteins involved in the activation and regulation of integrin-mediated cell adhesion. Apart from binding to integrin β cytosolic tails, kindlins and the well characterized integrin-activator talin bind membrane phospholipids. The ubiquitin-like F1 sub-domain of the FERM domain of talin contains a short loop that binds to the lipid membrane. By contrast, the F1 sub-domain of kindlins contains a long loop demonstrated binding to the membrane. Here, we report structural characterization and lipid interactions of the 83-residue F1 loop of kindlin-3 using NMR and optical spectroscopy methods. NMR studies demonstrated that the F1 loop of kindlin-3 is globally unfolded but stretches of residues assuming transient helical conformations could be detected in aqueous solution. We mapped membrane binding interactions of the F1 loop with small unilamellar vesicles (SUVs) containing either zwitterionic lipids or negatively charged lipids using ¹⁵N-¹H HSQC titrations. These experiments revealed that the F1 loop of kindlin-3 preferentially interacted with the negatively charged SUVs employing almost all of the residues. By contrast, only fewer residues appeared to be interacted with SUVs containing neutral lipids. Further, CD and NMR data suggested stabilization of helical conformations and predominant resonance perturbations of the F1 loop in detergent containing solutions. Conformations of an isolated N-terminal peptide fragment, or EK21, of the F1 loop, containing a poly-Lys sequence motif, important for membrane interactions, were also investigated in detergent solutions. EK21 adopted a rather extended or β-type conformations in complex with negatively charged SDS micelles. To our knowledge, this is the first report describing the conformations and residue-specific interactions of kindlin F1 loop with lipids. These data therefore provide important insights into the interactions of kindlin FERM domain with membrane lipids that contribute toward the integrin activating property.     

Extraction method for analysis of detergent-solubilized bacteriorhodopsin and hydrophobic peptides by electrospray ionization mass spectrometry

The analysis of integral membrane proteins or transmembrane peptides by electrospray ionization mass spectrometry (ESI-MS) is difficult since detergents, used to solubilize these hydrophobic proteins and peptides, severely suppress analyte ion formation. This problem has been addressed previously by precipitating the protein, removing the detergent, and resolubilizing the protein in a nonpolar solvent. Here, we demonstrate a method that avoids protein precipitation and resolubilization. Detergent-solubilized bacteriorhodopsin is extracted into a nonpolar solvent phase by adding a chloroform/methanol/water solvent mixture to the aqueous detergent solution. ESI mass spectra of the nonpolar, chloroform-rich phase were dominated by peaks due to bacterioopsin. Bacterioopsin precursors with partially cleaved leader sequences were seen in all mass spectra. Additional peaks were likely due to intact bacteriorhodopsin, i.e., bacterioopsin with the retinal prosthetic group attached, and to bacterioopsin associated with lipid molecules. A separation process that occurred in the fused-silica capillary leading to the electrospray tip was essential for obtaining ESI mass spectra of bacterioopsin. The extraction-into-chloroform procedure also worked well with hydrophobic, transmembrane-type peptides that were insoluble in other electrospray solvents, including 100% formic acid, and the method has application to transmembrane peptides formed from digests of integral membrane proteins.     

Dissection of multi-protein complexes using mass spectrometry: Subunit interactions in transthyretin and retinol-binding protein complexes

Complexes formed between transthyretin and retinol-binding protein prevent loss of retinol from the body through glomerular filtration. The interactions between these proteins have been examined by electrospray ionization combined with time-of-flight mass analysis. Conditions were found whereby complexes of these proteins, containing from four to six protein molecules with up to two ligands, are preserved in the gas phase. Analysis of the mass spectra of these multimeric species gives the overall stoichiometry of the protein subunits and provides estimates for solution dissociation constants of 1.9 ± 1.0 × 10⁻⁷ M for the first and 3.5 ± 1.0 × 10⁻⁵ M for the second retinol-binding protein molecule bound to a transthyretin tetramer. Dissociation of these protein assemblies within the gas phase of the mass spectrometer shows that each retinol-binding protein molecule interacts with three transthyretin molecules. Mass spectral analysis illustrates not only a correlation with solution behavior and crystallographic data of a closely related protein complex but also exemplifies a general method for analysis of multi-protein assemblies. Proteins Suppl. 2:3–11, 1998. © 1998 Wiley-Liss, Inc.     

Properties of protein-chlorophyll complexes from pea (Pisum sativum L.) leaves. The organization of chlorophyll.

Chlorophyll-protein-detergent complexes were prepared from pea chloroplasts by using sodium dodecylbenzenesulphonate and polyacrylamide-gel electrophoresis. Circular-dichroism spectra showed that complex CPI has a dimeric arrangement of chlorophyll a, with additional weaker interactions. Ellipticities were determined for both complexes and for purified chlorophylls in solution, and it is argued that the circular dichroism of complex CPII is derived from chlorophyll-protein interaction rather than from interaction between chlorophylls a and b. The detergent could be removed from the complexes by using urea and gel filtration, leaving the chlorophyll-protein in solution, although in each case a diminished ellipticity indicated some loss of organization. Three-peaked circular-dichroism spectra of chloroplast fragments before and after addition of detergent were compared with a curve obtained by summing graphically the spectra of complexes CPI, CPII and the free-pigment fraction. There was good correspondence at 650 nm, and the longer-wavelength peaks agreed in form and magnitude, but with discrepancies in position. It was concluded that complexes CPI and CPII pre-exist in the original material, but that there is an environmental effect which is destroyed when the complexes are extracted.     

Isolation of thylakoid membrane complexes from rice by a new double-strips BN/SDS-PAGE and bioinformatics prediction of stromal ridge subunits interaction

Thylakoid membrane complexes of rice (Oryza sativa L.) play crucial roles in growth and crop production. Understanding of protein interactions within the complex would provide new insights into photosynthesis. Here, a new "Double-Strips BN/SDS-PAGE" method was employed to separate thylakoid membrane complexes in order to increase the protein abundance on 2D-gels and to facilitate the identification of hydrophobic transmembrane proteins. A total of 58 protein spots could be observed and subunit constitution of these complexes exhibited on 2D-gels. The generality of this new approach was confirmed using thylakoid membrane from spinach (Spinacia oleracea) and pumpkin (Cucurita spp). Furthermore, the proteins separated from rice thylakoid membrane were identified by the mass spectrometry (MS). The stromal ridge proteins PsaD and PsaE were identified both in the holo- and core- PSI complexes of rice. Using molecular dynamics simulation to explore the recognition mechanism of these subunits, we showed that salt bridge interactions between residues R19 of PsaC and E168 of PasD as well as R75 of PsaC and E91 of PsaD played important roles in the stability of the complex. This stromal ridge subunits interaction was also supported by the subsequent analysis of the binding free energy, the intramolecular distances and the intramolecular energy.