Dimethyl sulfoxide binding to globular proteins: a nuclear magnetic relaxation dispersion study

Cosolvent Dimethyl Sulfoxide Influences Protein-Ligand Binding Kinetics via Solvent Viscosity Effects: Revealing the Success Rate of Complex Formation Following Diffusive Protein-Ligand Encounter

Protein-ligand-exchange kinetics determines the duration of biochemical signals and consequently plays an important role in drug design. Binding studies commonly require solubilization of designed ligands in solvents such as dimethyl sulfoxide (DMSO), resulting in residual amounts of DMSO following titration of solubilized ligands into aqueous protein samples. Therefore, it is critical to establish whether DMSO influences protein-ligand binding. Here, we address the general and indirect effect of DMSO on protein-ligand binding caused by solvent viscosity, which is strongly dependent on the relative concentrations of DMSO and water. As a model system, we studied the binding of a drug-like ligand to the carbohydrate recognition domain of galectin-3 in the presence of variable amounts of DMSO. We used isothermal titration calorimetry to characterize binding thermodynamics and ¹⁵N NMR relaxation to monitor kinetics. The binding enthalpy is not affected, but we observe a subtle trend of increasingly unfavorable entropy of binding, and consequently decreased affinity, with increasing DMSO concentration. The increasing concentration of DMSO results in a reduced association rate of binding, while the dissociation rate is less affected. The observed association rate is inversely proportional to the viscosity of the DMSO-water mixture, as expected from theory, but significantly reduced from the diffusion-controlled limit. By comparing the viscosity dependence of the observed association rate with that of the theoretical diffusion-controlled association rate, we estimate the success rate of productive complex formation following an initial encounter of proteins and ligands, showing that only one out of several hundred binding "attempts" are successful.

Heterogeneous Off-Target Effects of Ultra-Low Dose Dimethyl Sulfoxide (DMSO) on Targetable Signaling Events in Lung Cancer In Vitro Models

Targetable alterations in cancer offer novel opportunities to the drug discovery process. However, pre-clinical testing often requires solubilization of these drugs in cosolvents like dimethyl sulfoxide (DMSO). Using a panel of cell lines commonly used for in vitro drug screening and pre-clinical testing, we explored the DMSO off-target effects on functional signaling networks, drug targets, and downstream substrates. Eight Non-Small Cell Lung Cancer (NSCLC) cell lines were incubated with three concentrations of DMSO (0.0008%, 0.002%, and 0.004% v/v) over time. Expression and activation levels of 187 proteins, of which 137 were kinases and downstream substrates, were captured using the Reverse Phase Protein Array (RPPA). The DMSO effect was heterogeneous across cell lines and varied based on concentration, exposure time, and cell line. Of the 187 proteins measured, all were statistically different in at least one comparison at the highest DMSO concentration, followed by 99.5% and 98.9% at lower concentrations. Only 46% of the proteins were found to be statistically different in more than 5 cell lines, indicating heterogeneous response across models. These cell line specific alterations modulate response to in vitro drug screening. Ultra-low DMSO concentrations have broad and heterogeneous effects on targetable signaling proteins. Off-target effects need to be carefully evaluated in pre-clinical drug screening and testing.

Me2SO perfusion time for whole-organ cryopreservation can be shortened: Results of micro-computed tomography monitoring during Me2SO perfusion of rat hearts

Cryopreservation of whole organs and specific tissues is an important and continually expanding field of medicine. The protocols currently used for organ preservation do not ensure survivability and functionality; the protocols for ovarian tissue lead to acceptable outcomes, but these are still capable of further improvement. In general, cryopreservation protocols need to be optimized. One important approach to improving cryopreservation protocols in general involves reducing exposure to cytotoxic cryoprotective agents prior to freezing. This study, therefore, evaluated the real-time tissue penetration of dimethyl sulfoxide, a cryoprotective agent that is widely used in cryopreservation. Dimethyl sulfoxide penetration in rat hearts perfused with a 15% (v/v) dimethyl sulfoxide solution was examined in real-time using dynamic contrast-enhanced micro-computed tomography imaging. Viability of cardiomyocytes was not significantly affected by the dimethyl sulfoxide perfusion procedure. Two different perfusion rates were evaluated and compared with perfusion using a common iodine-based contrast agent (iomeprol). The dynamic contrast-enhanced micro-computed tomography imaging data showed that dimethyl sulfoxide flushes both the extracellular and intracellular spaces in rat heart tissue to 95% equilibration after ≈ 35 s via perfusion. Subsequent wash-out via perfusion is completed to 95% within ≈ 49 s. The equilibration duration routinely used in dimethyl sulfoxide–based protocols for cryopreservation should therefore be questioned. Shorter incubation duration would perhaps be sufficient, as well as being beneficial in relation to cell survivability. It would be helpful to have techniques for non-invasive real-time monitoring of the penetration of cryoprotective agents and such techniques should be used to revise cryopreservation protocols. Switching to perfusion-based equilibration procedures might be beneficial, if feasible.

Minute Additions of DMSO Affect Protein Dynamics Measurements by NMR Relaxation Experiments through Significant Changes in Solvent Viscosity

Studies of protein-ligand binding often rely on dissolving the ligand in dimethyl sulfoxide (DMSO) to achieve sufficient solubility, and then titrating the ligand solution into the protein solution. As a result, the final protein-ligand solution contains small amounts of DMSO in the buffer. Here we report how the addition of DMSO impacts studies of protein conformational dynamics. We used 15 N NMR relaxation to compare the rotational diffusion correlation time (τC ) of proteins in aqueous buffer with and without DMSO. We found that τC scales with the viscosity of the water-DMSO mixture, which depends sensitively on the amount of DMSO and varies by a factor of 2 across the relevant concentration range. NMR relaxation studies of side chains dynamics are commonly interpreted using τC as a fixed parameter, obtained from backbone 15 N relaxation data acquired on a separate sample. Model-free calculations show that errors in τC , arising from mismatched DMSO concentration between samples, lead to significant errors in order parameters. Our results highlight the importance of determining τC for each sample or carefully matching the DMSO concentrations between samples.

The kinetic analysis of the N-methylation of 4-phenylpyridine by nicotinamide N-methyltransferase: Evidence for a novel mechanism of substrate inhibition

The N-methylation of 4-phenylpyridine produces the neurotoxin 1-methyl-4-phenylpyridinium ion (MPP+). We investigated the kinetics of 4-phenylpyridine N-methylation by nicotinamide N-methyltransferase (NNMT) and its effect upon 4-phenylpyridine toxicity in vitro. Human recombinant NNMT possessed 4-phenylpyridine N-methyltransferase activity, with a specific activity of 1.7 ± 0.03 nmol MPP+ produced/h/mg NNMT. Although the K_m for 4-phenylpyridine was similar to that reported for nicotinamide, its k_cat of 9.3 × 10^-5 ± 2 × 10^-5 s^-1 and specificity constant, k_cat/K_m, of 0.8 ± 0.8 s^-1 M^-1 were less than 0.15% of the respective values for nicotinamide, demonstrating that 4-phenylpyridine is a poor substrate for NNMT. At low (<2.5 mM) substrate concentration, 4-phenylpyridine N-methylation was competitively inhibited by dimethylsulphoxide, with a K_i of 34 ± 8 mM. At high (>2.5 mM) substrate concentration, enzyme activity followed substrate inhibition kinetics, with a K_i of 4 ± 1 mM. In silico molecular docking suggested that 4-phenylpyridine binds to the active site of NNMT in two non-redundant poses, one a substrate binding mode and the other an inhibitory mode. Finally, the expression of NNMT in the SH-SY5Y cell-line had no effect cell death, viability, ATP content or mitochondrial membrane potential. These data demonstrate that 4-phenylpyridine N-methylation by NNMT is unlikely to serve as a source of MPP+. The possibility for competitive inhibition by dimethylsulphoxide should be considered in NNMT-based drug discovery studies. The potential for 4-phenylpyridine to bind to the active site in two binding orientations using the same active site residues is a novel mechanism of substrate inhibition.

Characterization of Chemical Exchange Using Relaxation Dispersion of Hyperpolarized Nuclear Spins

Chemical exchange phenomena are ubiquitous in macromolecules, which undergo conformational change or ligand complexation. NMR relaxation dispersion (RD) spectroscopy based on a Carr-Purcell-Meiboom-Gill pulse sequence is widely applied to identify the exchange and measure the lifetime of intermediate states on the millisecond time scale. Advances in hyperpolarization methods improve the applicability of NMR spectroscopy when rapid acquisitions or low concentrations are required, through an increase in signal strength by several orders of magnitude. Here, we demonstrate the measurement of chemical exchange from a single aliquot of a ligand hyperpolarized by dissolution dynamic nuclear polarization (D-DNP). Transverse relaxation rates are measured simultaneously at different pulsing delays by dual-channel ¹⁹F NMR spectroscopy. This two-point measurement is shown to allow the determination of the exchange term in the relaxation rate expression. For the ligand 4-(trifluoromethyl)benzene-1-carboximidamide binding to the protein trypsin, the exchange term is found to be equal within error limits in neutral and acidic environments from D-DNP NMR spectroscopy, corresponding to a pre-equilibrium of trypsin deprotonation. This finding illustrates the capability for determination of binding mechanisms using D-DNP RD. Taking advantage of hyperpolarization, the ligand concentration in the exchange measurements can reach on the order of tens of μM and protein concentration can be below 1 μM, i.e., conditions typically accessible in drug discovery.

Refolding of horseradish peroxidase is enhanced in presence of metal cofactors and ionic liquids

The effects of various refolding additives, including metal cofactors, organic co-solvents, and ionic liquids, on the refolding of horseradish peroxidase (HRP), a well-known hemoprotein containing four disulfide bonds and two different types of metal centers, a ferrous ion-containing heme group and two calcium atoms, which provide a stabilizing effect on protein structure and function, were investigated. Both metal cofactors (Ca(2+) and hemin) and ionic liquids have positive impact on the refolding of HRP. For instance, the HRP refolding yield remarkably increased by over 3-fold upon addition of hemin and calcium chloride to the refolding buffer as compared to that in the conventional urea-containing refolding buffer. Moreover, the addition of ionic liquids [EMIM][Cl] to the hemin and calcium cofactor-containing refolding buffer further enhanced the HRP refolding yield up to 80% as compared to 12% in conventional refolding buffer at relatively high initial protein concentration (5 mg/ml). These results indicated that refolding method utilizing metal cofactors and ionic liquids could enhance the yield and efficiency for metalloprotein.

Protein-Inhibitor Interaction Studies Using NMR

Solution-state NMR has been widely applied to determine the three-dimensional structure, dynamics, and molecular interactions of proteins. The designs of experiments used in protein NMR differ from those used for small-molecule NMR, primarily because the information available prior to an experiment, such as molecular mass and knowledge of the primary structure, is unique for proteins compared to small molecules. In this review article, protein NMR for structural biology is introduced with comparisons to small-molecule NMR, such as descriptions of labeling strategies and the effects of molecular dynamics on relaxation. Next, applications for protein NMR are reviewed, especially practical aspects for protein-observed ligand-protein interaction studies. Overall, the following topics are described: (1) characteristics of protein NMR, (2) methods to detect protein-ligand interactions by NMR, and (3) practical aspects of carrying out protein-observed inhibitor-protein interaction studies.

Protective effects of dimethyl sulfoxide on labile protein interactions during electrospray ionization

Electrospray ionization mass spectrometry is a valuable tool to probe noncovalent interactions. However, the integrity of the interactions in the gas-phase is heavily influenced by the ionization process. Investigating oligomerization and ligand binding of transthyretin (TTR) and the chaperone domain from prosurfactant protein C, we found that dimethyl sulfoxide (DMSO) can improve the stability of the noncovalent interactions during the electrospray process, both regarding ligand binding and the protein quaternary structure. Low amounts of DMSO can reduce in-source dissociation of native protein oligomers and their interactions with hydrophobic ligands, even under destabilizing conditions. We interpret the effects of DMSO as being derived from its enrichment in the electrospray droplets during evaporation. Protection of labile interactions can arise from the decrease in ion charges to reduce the contributions from Coulomb repulsions, as well as from the cooling effect of adduct dissociation. The protective effects of DMSO on labile protein interactions are an important property given its widespread use in protein analysis by electrospray ionization mass spectrometry (ESI-MS).

Simplicity within the complexity: bilateral impact of DMSO on the functional and unfolding patterns of α-chymotrypsin

New understanding of the fundamental links between protein stability, conformational flexibility and function, can be gained through synergic studies on their catalytic and folding/unfolding properties under the influence of stabilizing/destabilizing additives. We explored an impact of dimethyl sulfoxide (DMSO), the moderate effector of multilateral action, on the kinetic (functional) and thermodynamic (thermal unfolding) patterns of a hydrolytic enzyme, α-chymotrypsin (α-CT), over a wide range of additive concentrations, 0-70% (v/v). Both the calorimetric and kinetic data exhibited rich behavior pointing to the complex interplay of global/local stability (and flexibility) patterns. The complex action of DMSO is explained through the negative and positive preferential solvation motifs that prevail for the extreme opposite, native-like and unfolded states, respectively, implying essential stabilization of compact domains by enhancement of interfacial water networks and destabilization of a flexible active site by direct binding of DMSO to the unoccupied specific positions intended for elongated polypeptide substrates.

Influence of dimehylsulfoxide on protein-ligand binding affinities

Because of its favorable physicochemical properties, DMSO is the standard solvent for sample storage and handling of compounds in drug discovery. To date, little attention was given to how DMSO influences protein-ligand binding strengths. In this study we investigated the effects of DMSO on different noncovalent protein-ligand complexes, in particular in terms of the binding affinities, which we determined using nanoESI-MS. For the investigation, three different protein-ligand complexes were chosen: trypsin-Pefabloc, lysozyme-tri-N-acetylchitotriose (NAG3), and carbonic anhydrase-chlorothiazide. The DMSO content in the nanoESI buffer was increased systematically from 0.5 to 8%. For all three model systems, it was shown that the binding affinity decreases upon addition of DMSO. Even 0.5-1% DMSO alters the KD values, in particular for the tight binding system carbonic anhydrase-chlorothiazide. The determined dissociation constant (KD) is up to 10 times higher than for a DMSO-free sample in the case of carbonic anhydrase-chlorothiazide binding. For the trypsin-Pefabloc and lysozyme-NAG3 complexes, the dissociation constants are 7 and 3 times larger, respectively, in the presence of DMSO. This work emphasizes the importance of effects of DMSO as a co-solvent for quantification of protein-ligand binding strengths in the early stages of drug discovery.

Design and characterization of protein-quercetin bioactive nanoparticles

Background

The synthesis of bioactive nanoparticles with precise molecular level control is a major challenge in bionanotechnology. Understanding the nature of the interactions between the active components and transport biomaterials is thus essential for the rational formulation of bio-nanocarriers. The current study presents a single molecule of bovine serum albumin (BSA), lysozyme (Lys), or myoglobin (Mb) used to load hydrophobic drugs such as quercetin (Q) and other flavonoids.

Results

Induced by dimethyl sulfoxide (DMSO), BSA, Lys, and Mb formed spherical nanocarriers with sizes less than 70 nm. After loading Q, the size was further reduced by 30%. The adsorption of Q on protein is mainly hydrophobic, and is related to the synergy of Trp residues with the molecular environment of the proteins. Seven Q molecules could be entrapped by one Lys molecule, 9 by one Mb, and 11 by one BSA. The controlled releasing measurements indicate that these bioactive nanoparticles have long-term antioxidant protection effects on the activity of Q in both acidic and neutral conditions. The antioxidant activity evaluation indicates that the activity of Q is not hindered by the formation of protein nanoparticles. Other flavonoids, such as kaempferol and rutin, were also investigated.

Conclusions

BSA exhibits the most remarkable abilities of loading, controlled release, and antioxidant protection of active drugs, indicating that such type of bionanoparticles is very promising in the field of bionanotechnology.

The protective effects of osmolytes on arginine kinase unfolding and aggregation

Osmolytes are a series of different kinds of small molecules that can maintain the correct conformation of protein by acting as molecular chaperons. In this study, the protective effects of four compatible osmolytes, i.e., proline, sucrose, DMSO and glycerol, were studied during arginine kinase (EC 2.7.3.3) unfolding and aggregation. The results showed that all the osmolytes applied in this study obviously prevented AK unfolding and inactivation that was due to a GdnHCl denaturant by reducing the inactivation rate constants (k(i)), increasing the transition free energy changes (DeltaDeltaG(i)) and increasing the value for the midpoint of denaturation (C(m)). Furthermore, the osmolytes remarkably prevented AK aggregation in a concentration-dependent manner during AK refolding. Our results strongly indicated that osmolytes were not only metabolism substrates, but they were also important compounds with significant physiological protective functions for proteins, especially in some extremely harsh environments.

Role of osmolytes as chemical chaperones during the refolding of aminoacylase

The refolding and reactivation of aminoacylase is particularly difficult because of serious off-pathway aggregation. The effects of 4 osmolytes--dimethylsulphoxide, glycerol, proline, and sucrose--on the refolding and reactivation of guanidine-denatured aminoacylase were studied by measuring aggregation, enzyme activity, intrinsic fluorescence spectra, 1-anilino-8-naphthalenesulfonate (ANS) fluorescence spectra, and circular dichroism (CD) spectra. The results show that all the osmolytes not only inhibit aggregation but also recover the activity of aminoacylase during refolding in a concentration-dependent manner. In particularly, a 40% glycerol concentration and a 1.5 mol/L sucrose concentration almost completely suppressed the aminoacylase aggregation. The enzyme activity measurements revealed that the influence of glycerol is more significant than that of any other osmolyte. The intrinsic fluorescence results showed that glycerol, proline, and sucrose stabilized the aminoacylase conformation effectively, with glycerol being the most effective. All 4 kinds of osmolytes reduced the exposure of the hydrophobic surface, indicating that osmolytes facilitate the formation of protein hydrophobic collapse. The CD results indicate that glycerol and sucrose facilitate the return of aminoacylase to its native secondary structure. The results of this study suggest that the ability of the various osmolytes to facilitate the refolding and renaturation of aminoacylase is not the same. A survey of the results in the literature, as well as those presented here, suggests that although the protective effect of osmolytes on protein activity and structure is equal for different osmolytes, the ability of osmolytes to facilitate the refolding of various proteins differs from case to case. In all cases, glycerol was found to be the best stabilizer and a folding aid.

High-resolution protein hydration NMR experiments: Probing how protein surfaces interact with water and other non-covalent ligands

High-resolution solution NMR experiments are extremely useful to characterize the location and the dynamics of hydrating water molecules at atomic resolution. However, these methods are severely limited by undesired incoherent transfer pathways such as those arising from exchange-relayed intra-molecular cross-relaxation. Here, we review several complementary exchange network editing methods that can be used in conjunction with other types of NMR hydration experiments such as magnetic relaxation dispersion and 1J(NC') measurements to circumvent these limitations. We also review several recent contributions illustrating how the original solution hydration NMR pulse sequence architecture has inspired new approaches to map other types of non-covalent interactions going well beyond the initial scope of hydration. Specifically, we will show how hydration NMR methods have evolved and have been adapted to binding site mapping, ligand screening, protein-peptide and peptide-lipid interaction profiling.

DMSO-Related Effects in Protein Characterization

DMSO is the standard solvent for preparing stock solutions of compounds for drug discovery. The assay concentration of DMSO is normally 0.1% to 5% (v/v) or 14 to 715 mM. Thus, DMSO is often one of the principal additives in assay buffers. This standardization of stock solutions does not eliminate possible pitfalls associated with the effects of the DMSO-containing solutions on individual proteins. In this article, the authors want to emphasize the importance of detailed studies of these effects in the early stages of drug discovery. Two protein systems, the extracellular soluble domain of the human growth hormone receptor (hGHbp) and the phosphatase domain of PFKFB1 (BPase), were used for the study on effects of DMSO on protein stability, protein aggregation, and binding of drug compounds. The study revealed significant differences in the proteins’ behavior in the presence and absence of low amounts of DMSO. The addition of DMSO resulted in destabilization of the proteins investigated and also changed the apparent binding property of 1 protein. The authors have also shown that low DMSO concentrations influence the ionization process in electrospray ionization mass spectrometry (ESI-MS).

Water and urea interactions with the native and unfolded forms of a beta-barrel protein

A fundamental understanding of protein stability and the mechanism of denaturant action must ultimately rest on detailed knowledge about the structure, solvation, and energetics of the denatured state. Here, we use (17)O and (2)H magnetic relaxation dispersion (MRD) to study urea-induced denaturation of intestinal fatty acid-binding protein (I-FABP). MRD is among the few methods that can provide molecular-level information about protein solvation in native as well as denatured states, and it is used here to simultaneously monitor the interactions of urea and water with the unfolding protein. Whereas CD shows an apparently two-state transition, MRD reveals a more complex process involving at least two intermediates. At least one water molecule binds persistently (with residence time >10 nsec) to the protein even in 7.5 M urea, where the large internal binding cavity is disrupted and CD indicates a fully denatured protein. This may be the water molecule buried near the small hydrophobic folding core at the D-E turn in the native protein. The MRD data also provide insights about transient (residence time <1 nsec) interactions of urea and water with the native and denatured protein. In the denatured state, both water and urea rotation is much more retarded than for a fully solvated polypeptide. The MRD results support a picture of the denatured state where solvent penetrates relatively compact clusters of polypeptide segments.

Magnetic relaxation dispersion studies of biomolecular solutions

Although the MRD method has a long record in biomolecular systems, it has undergone a renaissance in the past few years as methodological developments have provided access to new types of information. In particular, MRD studies of quadrupolar nuclei such as 17O and 23Na have yielded valuable insights about the interactions of proteins and oligonucleotides with their solvent environment. The biomolecular MRD literature is still dominated by hydration studies, but the method has also been used to study the interaction of organic cosolvents and inorganic counterions with biomolecules. The MRD method can potentially make important contributions to the understanding of the mechanisms whereby protein conformational stability is affected by nonaqueous solvent components, such as denaturants, stabilizers, and helix promoters. Residence times of water molecules and other low molecular weight species in association with biomolecules can be determined by MRD. Such residence times are of general interest for understanding the kinetics of biomolecule-ligand interactions and, when exchange is gated by the biomolecule, can be used to characterize large-scale conformational fluctuations on nanosecond-millisecond time scales. By monitoring the integrity and specific internal hydration sites as well as the global solvent exposure, the MRD method can also shed light on the structure and dynamics of biomolecules in fluctuating nonnative states. Because it does not rely on high resolution, the MRD method is also applicable to very large biomolecules and complexes and has even been used to investigate protein crystals, gels, and biological tissues. In fact, dynamic studies of solids and liquid crystals were among the earliest applications of the MRD method. In many of its diverse applications, the MRD method provides unique information, complementing that available from high-resolution NMR.

Conformational isomers of a class II MHC-peptide complex in solution

A number of kinetic measurements of peptide dissociation from class II MHC-peptide complexes provide compelling evidence for the existence of conformational isomers in solution. There is evidence that T-lymphocytes can distinguish such isomers. However, virtually nothing is known about the structure of these isomers. Accordingly, we have investigated a water-soluble version of the murine class II MHC molecule I-Ek complexed with an antigenic peptide derived from pigeon cytochrome c residues 89-104 (PCC) by 19F-NMR. Two fluorine labels were placed on the PCC peptide; one fluorine label was placed at a MHC contact site, the other at a position involved in T-cell receptor (TCR) recognition. Introduction of these labels did not alter the observed kinetics of the PCC/I-Ek complex. The NMR data show two conformational isomers of this immunogenic complex. The presence of conformational isomers at a TCR contact site suggests that these structures may be recognized differently by the TCR. The agreement between the dissociation kinetics and the 19F-NMR data demonstrate that kinetic heterogeneity is correlated with structural counterparts observed by NMR. Dissociations in the presence of dimethyl sulfoxide were used to show that the rate of interconversion of these conformational isomers at pH 7.0 is low, with a lifetime on the order of hours or more. Modification of a peptide residue of PCC occupying the minor MHC binding pocket P6 alters the 19F-NMR spectra of both labels. This demonstrates that distant changes of amino acid residues can influence the conformation of the whole antigenic peptide inside the MHC binding cleft.

The modulation of membrane structure and stability by dimethyl sulphoxide (review)

Dimethyl sulphoxide is a widely used agent in cell biology. It is well known as a cryoprotectant, cell fusogen and a permeability enhancing agent. These applications depend, to a greater or lesser extent, on the effect of dimethyl sulphoxide on the stability and dynamics of biomembranes. The aim of this review is to examine progress of the research which has been directed towards studies of the interactions between dimethyl sulphoxide and membranes, particularly that with the lipid components of cell membranes, as seen in its effects on model membrane systems. Models are proposed to explain the mechanism whereby dimethyl sulphoxide may mediate its effects on biological functions by its effects on the stability and properties of the membrane lipid matrix.