Bacterial expression, purification and preliminary kinetic description of the kinase domain of v-fps

The gene coding for the tyrosine protein kinase domain of v-fpswas subcloned into a plasmid vector expressing glutathione-S-transferase(GST). This new vector expresses a fusion protein in Escherichiacoli composed of the kinase domain linked with GST at the N-terminus(GST-kin). A portion of the total expressed protein was solubleupon cell lysis and was purified by affinity chromatographyusing glutathione cross-linked agarose. GST-kin (M_r 57 000)is a phosphoprotein as judged by ³²P autoradiography, consistentwith the known autophosphorylation site within the kinase core[Weinmaster et aL (1984) Cell, 37, 559–568]. Cleavageof the fusion protein with thrombin and purification on phosphocelluloseresin yielded the pure kinase domain (M_r 33 000). The activityof the kinase domain is indistinguishable from that of GST-kinusing the peptide substrate EEEIYEEIE, indicating that Nterminalfusion has no effect on the kinase domain. GSTkin phosphorylatesa second peptide, EAEIYEAIE, with improved catalytic efficiency.Initial velocity data are consistent with a random bireactantmechanism with no substrate synergism observed in the ternarycomplex. Steady-state kinetic analyses reveal that this peptideis phosphorylated, with a k_cat of 3.6 s^–1, a K_peptideof 500 µM and a K_ATP of 250 µM. The expression,purification and preliminary kinetic analysis of the kinasedomain of v-fps provide the first step in the application ofstructurefunction studies for this oncoprotein     

Enzymatic amidation of recombinant (Leu27) growth hormone releasing hormone-Gly45

By chemoenzymatic synthesis the gene for a (Leu²⁷) analogueof human growth hormone releasing hormone-Gly⁴⁵ [(Leu²⁷GHRH-Gly⁴⁵]was constructed, cloned and expressed in Escherichia coli asa fusion protein with ß-galactosidase under the controlof the lac promoter and operator. Upon induction with isopropyl-D-thio-ß-galactopyranosidethe fusion protein accumulated to a yield of 15–20% ofthe total cellular protein. After cyanogen bromide deavage ofthe fusion protein the precursor peptide (Leu²⁷)hGHRH-Gly⁴⁵was separated by extraction and purified by ion exchange andh.p.l.c.-RP18 chromatography. The purified peptide was analysedby sequencing, isolectric focusing, amino acid analysis andamino acid analysis after V8 protease digestion. The carboxy-terminalglydne was subsequently amidated by PAM (peptidylglycine--amidating-monooxygenase),an enzyme which was isolated and characterized from fresh bovinepituitaries. Correct amidatlon of the penultimate amino acid,leucine, was verified by peptide sequencing with an authenticleucine amide reference.     

Lipid-tagged antibodies: bacterial expression and characterization of a lipoprotein--single-chain antibody fusion protein

In order to achieve a stable and functional immobilization ofantibodies, we investigated the possibility of adding hydrophobicmembrane anchors to antibody fragments expressed in Escherichiacoli. The DNA sequence encoding the signal peptide and the nineN-terminal amino add residues of the major lipoprotein of E.coliwas fused to the sequence of an anti-2-phenyloxazolone single-chainFv antibody fragment [Takkinen et al. (1991) Protein Engng,4, 837–841]. The expression of the fusion construct inE.coli resulted in specific accumulation of an immunoreactive28 kDa polypeptide. Unlike the unmodified single-chain Fv fragment,the fusion protein was cell-associated, labelled by [³H]palmitatewhich is indicative of the presence of N-terminal lipid modification,partitioned into the detergent phase upon Triton X-114 phaseseparation and was localized predominantly in the bacterialouter membrane. The fusion antibody displayed specific 2-phenyloxazolone-bindingactivity in the membranebound form and after solubilizationwith non-ionic detergents. Furthermore, upon removal of detergentthe fusion antibody was incorporated into proteoliposomes whichdisplayed specific hapten-binding activity. Our results showthat antibodies can be converted to membrane-bound proteinswith retention of antigen-binding properties by introductionof lipid anchors during biosynthesis. This approach may proveuseful in the design of immunoliposomes and immunosensors.     

An internal cellulose-binding domain meidates adsorption of an engineered bifunctional xylanase/cellulase

A chimeric xylanase/endoglucanase (XynCenA) with an internalcellulose-binding domain was constructed by fusing the Bacillussubtilis xyn gene fragment to the 5'-end of the Cellulomonasfimi cenA. A polyhistidine-encoding sequence was also fusedto the 5'-end of the xyn gene. The gene fusion was overexpressedin Escherichia coli and the fusion poly-peptide purified fromthe cell extracts using the polyhistldine tail. The hybrid proteinbehaved like the parental endo-glucanase or xylanase when assayedon a number of soluble and insoluble cellulosic substrates orxylans. The presence of two distinct active sites and the internalcellulose-binding domain did not significantly affect the hydrolysisof any of these substrates. However, the fusion protein exhibiteda strong affinity for both mkrocrystaUine cellulose (Avicel)and regenerated chitin. Like the parental endoglucanase, boundXynCenA could not be duted from these porysaccharides with eitherlow or high salt buffer or distilled water. More stringent conditions,such as 1% SDS or 8 M guanidinium hydro-chloride, fully desorbedthe protein. The fusion protein did not adsorb significantlyto insoluble xylan     

Genetically engineered brain drug delivery vectors: cloning, expression and in vivo application of an anti-transferrin receptor single chain antibodystreptavidin fusion gene and protein

A single chain Fv antibody–streptavidin fusion proteinwas expressed and purified from bacterial inclusion bodies followingcloning of the genes encoding the variable region of the heavychain and light chain of the murine OX26 monoclonal antibodyto the rat transferrin receptor. The latter undergoes receptormediated transcytosis through the brain capillary endothelialwall in vivo, which makes up the blood–brain barrier (BBB);therefore, the OX26 monoclonal antibody and its single chainFv analog may act as brain drug delivery vectors in vivo. Attachmentof biotinylated drugs to the antibody vector is facilitatedby production of the streptavidin fusion protein. The bi-functionalityof the OX26 single chain Fv antibody–streptavidin fusionprotein was retained, as the product both bound biotin and therat transferrin receptor in vitro and in vivo, based on pharmacokineticand brain uptake analyses in anesthetized rats. The attachmentof biotin–polyethyleneglycol–fluorescein to theOX26 single chain Fv antibody–streptavidin fusion proteinresulted in illumination of isolated rat brain capillaries inconfocal fluorescent microscopy. In conclusion, these studiesdemonstrate that genetically engineered single chain Fv antibody–streptavidinfusion proteins may be used for non-invasive neurotherapeuticdelivery to the brain using endogenous BBB transport systemssuch as the transferrin receptor.     

Direct expression in E.coli of a functionally active protein A-snake toxin fusion protein

We constructed a recombinant expression plasmid encoding a proteinA–neurotoxin fusion protein. The fused toxin is directlyexpressed in the periplasmic space of Escherichia coli and canbe purified in the milligram range by a single immuno-affinitystep. The LD₅₀ values of the fused toxin and native toxin are130 and 20 nmol/kg mouse respectively. The K_d values characterizingtheir binding to the nicotinic acetylcholine receptor (AcChoR)are respectively 4.8 ± 0.8 and 0.07 ± 0.03 nM.In contrast, the fused and native toxins are equally well recognizedby a toxin-specific monoclonal antibody which recognizes theAcChoR binding site. The lower toxicity of the fused toxin mightresult, therefore, from a steric hindrance, due to the presenceof the bulky protein A moiety (mol. wt = 31 kd) rather thanto a direct alteration of the ‘toxic’ site. Thefused toxin is more immunogenic than native toxin, since 1 nmolof hybrid toxin and 14 nmol of native toxin give rise to comparabletiters of antitoxin antibodies which, furthermore, are equallypotent at neutralizing neurotoxicity. The work described inthis paper shows that the use of fused toxins may be of paramountimportance for future development of serotherapy against envenomationby snake bites.     

Overproduction of bovine {beta}-casein in Escherichia coli and engineering of its main chymosin cleavage site

A cDNA clone containing the entire coding region for bovineß-casein A³ flanked by 53 base pairs of 5' non-codingand 358 base pairs of 3' non-coding sequences was isolated froma bovine mammary cDNA phagemid library. The coding segment formature ß-casein was subcloned into the T7 expressionsystem, in which the expression of recombinant ß-caseinwas controlled by the T7 gene 10 promoter and ribosome bindingsite. High level expression of Met-ß-casein to 20%of the total soluble proteins was obtained in Escherichia coliwithin 2 h after induction of T7 RNA-polymerase synthesis. Inan attempt to induce secretion the coding segment for matureß-casein was coupled to the ompA translations initiationsignal and signal peptide coding sequence but no secretion ofthe fusion protein and no processing of the signal peptide fromthe fusion protein was observed. Instead, the Met-ß-caseincould be isolated in asoluble form from E.coli cells after anosmotic shock, indicative of a periplasmic location. This proceduredid not lyse the cells. The protein was purified to homogeneityafter a pH 4.8 isoelectric precipitation followed by reversed-phasehigh-performance liquid chromatography. The ß-caseincDNA was altered to change the main chymosin cleavage siteinß-casein at position 192–193 in two ways, namelyfrom Leu–Tyr to Pro–Pro and to Leu–stop. Thesemutations were designed to prevent generation of the bitterpeptide ßcasein(193–209) by chymosin cleavage.The mutant Met-ß-caseins were expressed in E.colito the same level as wild-type Met-ß-casein. Purifiedmutant Met-ß-casein(Prol92– Prol93) was no longerhydrolysed by chymosin at the 192–193 bond.     

Cloning, expression and purification of a sarcoplasmic calcium-binding protein from the sandworm Nereis diversicolor via a fusion product with chloramphenicol acetyltransferase

A gene coding for the Nereis sarcoplasmic calcium-binding protein(NSCP) was synthesized and expressed in Escherichia coli. Thesequence of the gene was derived from the protein sequence byreverse translation. It possesses a number of unique, regularlyspaced, restriction endonuclease cleavage sites to facilitatefuture site-directed mutagenesis. For the cloning strategy thegene sequence was divided into four parts. Three parts werecloned by ligation of hybridized oligomers and one part by inversePCR. The protein was expressed as a fusion protein with thebacterial chloramphenicol acetyltransferase (CAT), which couldbe easily purified by affinity chromatography. At the junctionof the CAT and NSCP moieties a recognition site for the proteolyticenzyme factor Xa was built in. However, the distance betweenthe moieties appeared to be crucial to warrant cleavage. A kineticanalysis showed that NSCP prepared from the sandworm and theone expressed by E.coli behaved in the same way. This systemprovides a basis for site-specific mutagenesis studies, in orderto elucidate the molecular mechanism of cation binding and concomitantconformational changes     

Engineering a de novo-designed coiled-coil heterodimerization domain for the rapid detection, purification and characterization of recombinantly expressed peptides and proteins

Using the techniques of genetic engineering and the principlesof protein de novo design, we have developed a unique affinitymatrix protein tag system as a rapid, convenient and sensitivemethod to detect, purify and characterize newly expressed recombinantpeptides or proteins from cell extracts. The method utilizestwo de novo-designed linear peptide sequences that can selectivelydimerize to form the stable protein motif, the two-stranded-helical coiled-coil. In this method, a recombinant bacterialexpression vector pRLDE has been engineered so that one of thedimerization strands (E-coil) is expressed as a C-terminal fusiontag on newly expressed peptides or proteins, while the other(K-coil) is either biotin-labeled for detection in a Westernblot-type format or immobilized on an insoluble silica supportfor selective dimerization affinity chromatography. Recombinantlyexpressed peptides from Escherichia coli containing the dimerizationtag have been produced, detected and purified using this method.The recombinant peptides were easily and clearly identifiedusing the biotin-labeled coil, while the single-step affinitypurification results indicated the purity of the affinity purifiedexpressed peptides to be >95%, as assessed by reversed-phasechromatography. The stability of the dimerization domain alsoallows for the purified peptide to be left attached to the matrix,thus creating a new peptide-bound column that can be used tostudy peptide–protein or peptide–ligand interactions.Therefore this system offers a new alternative to existing peptideor protein fusion tags and demonstrates the utility of a denovo-designed system.     

Efficient production of the C-terminal domain of secretory leukoprotease inhibitor as a thrombin-cleavable fusion protein in Escherichia coli

We have developed a high-level production system for the C-terminaldomain of secretory leukoprotease inhibitor (SLPI) to investigateits pharmacological activities. A gene for the C-terminal domainof SLPI, (Asn55-AlalO7)SLPI, was constructed from chemicallysynthesized deoxyoligonucleotides. It was fused to a gene forthe N-terminal portion of human growth hormone via a DNA sequenceencoding Leu-Val-Pro-Arg, which can he cleaved by thrombin.The fused gene was expressed in Escherichia coli under the controlof a trp promoter, and the fusion protein was obtained as aninclusion body. After sulfonation of the cysteine residues,the sulfonated fusion protein was cleaved at the desired siteby thrombin. Sulfonated (Asn55-Ala107)SLPI was refolded in Trisbuffer containing reduced and oxidized glutathione. The resulting(Asn55-Ala107) SLPI was purified by cation-exchange chromatographyand reverse-phase high performance liquid chromatography. Thefinal yield was 50 mg/l culture. (Asn55-Ala107)SLPI was as activeagainst elastase as, but had less trypsin inhibitory activitythan, native SLPI. This system is suitable for the large-scaleproduction of the C-terminal domain of SLPI, which is an elastase-specificinhibitor.     

Destabilizing interactions between the partners of a bifunctional fusion protein

Hybrid MalE–GVP is a bifunctional protein in vitro sinceit binds maltose as protein MalE of Escherichia coli and sinceit is dimeric and specifically binds single-stranded DNA asprotein GVP of phage M13. The oxidation rate of a unique cysteineresidue was used to compare the stabilities of GVP in its freeand hybrid forms, under conditions where MalE was either foldedor unfolded by a denaturing agent. The results showed that boththe covalent link and tertiary non-covalent interactions betweenMalE and GVP destabilized GVP in MalE–GVP. To test whetherGVP had identical structures in its free and hybrid forms, mutationswere used as local conformational probes. The effects of thesemutations on the capabilities of MalE–GVP to dimerizeand to bind single-stranded DNA were assayed in vitro. Theywere compatible with the effects of the same mutations on theglobal activity of free GVP in vivo and with the effects thatcould be predicted from the known data on free GVP, in particularits crystal structure. Thus, one partner of a hybrid proteincan be destabilized by the other partner while maintaining itsstructural and functional characteristics.     

Synthesis of a squash-type protease inhibitor by gene engineering and effects of replacements of conserved hydrophobic amino acid residues on its inhibitory activity

Cucurbita maxima trypsin inhibitor I (CMTI-I), a member of thesquash-type protease inhibitor family, is composed of 29 aminoacids and shows strong inhibition of trypsin by its compactstructure. To study the structure–function relationshipof this inhibitor using protein engineering methods, we constructedan expression system for CMTI-I as a fused protein with porcineadenylate kinase (ADK). A Met residue was introduced into thejunction of ADK and CMTI-I to cleave the fusion protein withCNBr, whereas a Met at position 8 of authentic CMTI-I was replacedby Leu. Escherichia coli JM109 transformed with the constructedplasmid expressed the fused protein as an inclusion body. Aftercleavage of the expressed protein with CNBr, fully reduced speciesof CMTI-I were purified by reversed-phase HPLC and then oxidizedwith air by shaking. For efficient refolding of CMTI-I, we used50 mM NH₄HCO₃ (pH 7.8) containing 0.1% PEG 6000 at higher proteinconcentration. Strong inhibitory activity toward trypsin wasdetected only in the first of three HPLC peaks. The inhibitorconstant of CMTI-I thus obtained, in which Met8 was replacedby Leu, was 1.4x10-¹⁰ M. The effect of replacement of Met withLeu at position 8 was shown to be small by comparison of theinhibitor constant of authentic CMTI-III bearing Lys at position9 (8.9x10-¹¹ M) with that of its mutant bearing Leu at position8 and Lys at position 9 (1.8x10-¹⁰ M). To investigate the roleof the well conserved hydrophobic residues of CMTI-I in itsinteraction with trypsin, CMTI-I mutants in which one or allof the four hydrophobic residues were replaced by Ala were prepared.The inhibitor constants of these mutants indicated that thosewith single replacements were 5–40 times less effectiveas trypsin inhibitors and that the quadruple mutant was –450times less effective, suggesting that the hydrophobic residuesin CMTI-I contribute to its tight binding with trypsin. However,each mutant was not converted to a temporary inhibitor.     

The de novo protein with grafted biological function: transferring of interferon blast-transforming activity to albebetin

The de novo protein albebetin has been designed recently toform a predetermined tertiary fold that has not yet been observedin natural proteins. An eight amino acid fragment (131–138)of human interferon ₂ carrying the blasttransforming activityof the protein was attached to the N-terminus of albebetin nextto its initiatory methionine residue. The gene of chimeric proteinwas expressed in a wheat germ cell-free translation system andsynthesized protein was tested for its compactness and stability.Its ability for receptor binding was also studied. We have shownthat albebetin with attached octapeptide is practically as compactas natural proteins of corresponding molecular weight and possesseshigh stability toward the urea-induced unfolding. It binds murinethymocyte receptor at a high affinity and activates the thymocyteblast transformation efficiently at a concentration of 10^-11M.     

Expression and characterization of Lactobacillus 30a histidine decarboxylase in Escherichia coli

The genes coding for histidine decarboxylase from a wild-typestrain and an autoactivation mutant strain of Lactobacillus30a have been cloned and expressed in Escherichia coli. Themutant protein, G58D, has a single Asp for Gly substitutionat position 58. The cloned genes were placed under control ofthe ß-galactosidase promoter and the products arenatural length, not fusion proteins. The enzyme kinetics ofthe proteins isolated from E. coli are comparable to those isolatedfrom Lactobacillus 30a. At pH 4.8 the K_m of wild-type enzymeis 0.4 mM and the k_cat = 2800 min^–1; the correspondingvalues for G58D are 0.5 mM and 2750 min^–1. The wild-typeand G58D have autoactivation half-times of 21 and 9 h respectivelyunder pseudophysiological conditions of 150 mM K⁺ and pH 7.0.At pH 7.6 and 0.8 M K⁺ the half times are 4.9 and 2.9 h. Therelatively slow rate of autoactivation for purified proteinand the differences in cellular and non-cellular activationrates, coupled with the fact that wild-type protein is readilyactivated in wild-type Lactobacillus 30a but poorly activatedin E. coli, suggest that wild-type Lactobacillus 30a containsa factor, possibly an enzyme, that enhances the activation rate.     

The cystine knot of a squash-type protease inhibitor as a structural scaffold for Escherichia coli cell surface display of conformationally constrained peptides

The Ecballium elaterium trypsin inhibitor II (EETI-II), a memberof the squash family of protease inhibitors, is composed of28 amino acid residues and is a potent inhibitor of trypsin.Its compact structure is defined by a triple-stranded antiparallelß-sheet, which is held together by three intramoleculardisulfide bonds forming a cystine knot. In order to explorethe potential of the EETI-II peptide to serve as a structuralscaffold for the presentation of randomized oligopeptides, weconstructed two EETI-II derivatives, where the six-residue inhibitorloop was replaced by a 13-residue epitope of Sendai virus L-proteinand by a 17-residue epitope from human bone Gla-protein. EETI-IIand derived variants were produced via fusion to maltose bindingprotein MalE. By secretion of the fusion into the periplasmicspace, fully oxidized and correctly folded EETI-II was obtainedin high yield. EETI-II and derived variants could be presentedon the Escherichia coli outer membrane by fusion to truncatedLpp'–OmpA', which comprises the first nine residues ofmature lipoprotein plus the membrane spanning ß-strandfrom residues 46–66 of OmpA protein. Gene expression wasunder control of the strong and tightly regulated tetA promoter/operator.Cell viability was found to be drastically reduced by high levelexpression of Lpp'–OmpA'–EETI-II fusion protein.To restore cell viability, net accumulation of fusion proteinin the outer membrane was reduced to a tolerable level by introductionof an amber codon at position 9 of the lpp' sequence and utilizingan amber suppressor strain as expression host. Cells expressingEETI-II variants containing an epitope were shown to be surfacelabeled with the respective monoclonal antibody by indirectimmunofluorescence corroborating the cell surface exposure ofthe epitope sequences embedded in the EETI-II cystine knot scaffold.Cells displaying a particular epitope sequence could be enriched10⁷-fold by combining magnetic cell sorting with fluorescence-activatedcell sorting. These results demonstrate that E.coli cell surfacedisplay of conformationally constrained peptides tethered tothe EETI-II cystine knot scaffold has the potential to becomean effective technique for the rapid isolation of small peptidemolecules from combinatorial libraries that bind with high affinityto acceptor molecules.     

Synthesis, cloning and expression of the single-chain Fv gene of the HPr-specific monoclonal antibody, Jel42. Determination of binding constants with wild-type and mutant HPrs

The monoclonal antibody Jel42 is specific for the Escherichiacoli histidine-containing protein, HPr, which is an 85 aminoacid phosphocarrier protein of the phosphoenolpyruvate:sugarphosphotransferase system. The binding domain (Fv) has beenproduced as a single chain Fv (scFv). The scFv gene was synthesizedin vitro and coded for pelB leader peptide–heavy chain–linker–lightchain–(His)₅ tail. The linker is three repeats from theC-terminal repetitive sequence of eukaryotic RNA polymeraseII. This linker acts as a tag; it is the antigen for the monoclonalantibody Jel352. The codon usage was maximized for E.coli expression,and many unique restriction endonuclease sites were incorporated.The scFv gene incorporated into pT7-7 was highly expressed,yielding 10–30% of the cell protein as the scFv, whichwas found in inclusion bodies with the leader peptide cleaved.Jel42 scFv was purified by denaturation/renaturation yieldingpreparations with K_d values from 20 to 175 nM. However, basedupon an assessment of the amount of active refolded scFv, thebinding dissociation constant was estimated to be 2.7 ±2.0 nM compared with 2.8 ± 1.6 and 3.7 ± 0.3 nMpreviously determined for the Jel42 antibody and Fab fragmentrespectively. The effect of mutation of the antigen HPr on thebinding constant of the scFv was very similar to the propertiesdetermined for the antibody and the Fab fragment. It was concludedthat the small percentage (~6%) of refolded scFv is a true mimicof the Jel42 binding domain and that the incorrectly foldedscFv cannot be detected in the binding assay.     

Overproduction, preparation of monoclonal antibodies and purification of E.coli asparagine synthetase A

In order to explore the structure-function relationship of theEscherichia coli asparagine synthetase A it was necessary todevise a system for overexpression of the gene and purificationof the gene product. The E.coli asparagine synthetase A structuralgene was fused to the 3' end of the human carbonic anhydraseII structural gene and overexpressed in E.coli. The gene product,a 66 kDa fusion protein, which exhibited asparagine synthetaseactivity, was purified in a single step by affinity chromatographyand used as the antigen for the production of monoclonal antibodies.The monoclonal antibodies were screened by ELISA. Colonies werechosen which were positive for purified fusion protein and negativefor purified human carbonic anhydrase II. The E.coli asparaginesynthetase A gene was then overexpressed and the gene productwas used without purification for the final screen. The antibodiesselected were used for immunoaffinity chromatography to purifythe recombinant overexpressed E.coli asparagine synthetase A.Thus, a procedure is now available so that asparagine synthetaseA can be purified to homogeneity in a single step.     

Display of {beta}-lactamase on the Escherichia coli surface: outer membrane phenotypes conferred by Lpp'-OmpA'-{beta}-lactamase fusions

Bacterial cell-surface exposure of foreign peptides and solubleproteins has been achieved recently by employing a fusion proteinmethodology. An Lpp'–OmpA(46–159)–Bla fusionprotein has been shown previously to display the normally periplasmicenzyme ß-lactamase (Bla) on the cell surface of theGram-negative bacterium Escherichia coli. Here, we have investigatedthe role of the OmpA domain of the tripartite fusion proteinin the surface display of the passenger domain (Bla) and havecharacterized the effects of the fusion proteins on the integrityand permeability of the outer membrane. We show that in additionto OmpA(46–159), a second OmpA segment, consisting ofamino acids 46–66, can also mediate the display of Blaon the cell surface. Other OmpA domains of various lengths (aminoacids 46–84, 46–109, 46–128, 46–141and 46–145) either anchored the Bla domain on the periplasmicface of the outer membrane or caused a major disruption of theouter membrane, allowing the penetration of antibodies intothe cell. Detergent and antibiotic sensitivity and periplasmicleakage assays showed that changes in the permeability of theouter membrane are an unavoidable consequence of displayinga large periplasmic protein on the surface of E.coli. This isthe first systematic report on the effects that cell surfaceengineering may have on the integrity and permeability propertiesof bacterial outer membranes.