Kalavelil M. Koshy

Highly deiminated isoform of myelin basic protein from multiple sclerosis brain causes fragmentation of lipid vesicles

Myelin basic protein (MBP) occurs as a number of charge isomers due to phosphorylation, deamidation, and deimination of arginine to citrulline. All of these modifications decrease the net positive charge of the protein and its ability to cause aggregation of negatively charged lipid vesicles. This is used as a model system for the ability of MBP to cause adhesion of the cytosolic surfaces of myelin. Therefore, the effect of two deiminated forms of MBP on lipid vesicles was compared with that of the unmodified, most positively charged isomer, C1, to determine how loss of positively charged arginines would affect the function of MBP. The deiminated forms were the isomer isolated from normal human brains, in which only 6 Arg are deiminated to citrulline (MBP‐Cit6), and an isomer isolated from the brain of a patient who died with acute, fulminating multiple sclerosis (Marburg type), in which 18 of the 19 Arg were deiminated (MBP‐Cit18). Whereas C1 caused aggregation of lipid vesicles, resulting in an increase in absorbance due to light scattering, MBP‐Cit18 caused a decrease in absorbance of the lipid vesicles. Size exclusion chromatography and negative staining electron microscopy showed that this was due to fragmentation of the large multilayered vesicles into much smaller vesicles. MBP‐Cit6 caused less aggregation of lipid vesicles than did C1. However, no fragmentation of the vesicles into smaller ones in the presence of C1 and MBP‐Cit6 was detected by size exclusion chromatography or electron microscopy. The membrane fragmentation caused by MBP‐Cit18 is dramatically different from the effects of other forms of MBP from normal brain and may indicate a pathogenic effect of this charge isomer, which may have contributed to the severity of the Marburg type of multiple sclerosis. Alternatively, the deimination may have been a secondary effect resulting from the disease process. Regardless of the role of MBP‐Cit18 in multiple sclerosis, the effect of this modification indicates that, when most of the arginines of MBP are modified to an uncharged amino acid, the protein acquires properties similar to an apolipoprotein; thus, it may take up an amphipathic structure when bound to lipid. A partly amphipathic character may also be related to the role of MBP‐Cit6 in normal immature myelin, where it is the predominant charge isomer. J. Neurosci. Res. 57:529–535, 1999.

Influence of structural modifications on the phase behavior of semi-synthetic cerebroside sulfate.

Cerebroside sulfate (galactosylceramide I3-sulfate) containing alpha-hydroxy lignoceric acid (C24:0h-CBS), nervonic acid (C24:1-CBS), or hexacosanoic acid (C26:0-CBS) was prepared by a semi-synthetic procedure and studied by differential scanning calorimetry. The phase behavior of these species in 2 M KCl was compared to that of shorter chain length hydroxy and non-hydroxy fatty acid species reported earlier. All three of the new lipids undergo metastable phase behavior similar but not identical to the other species. In addition, the metastable phase behavior of all of the non-hydroxy fatty acid species was found to be more complex than previously thought, with several phases of high transition temperatures and enthalpies possible. Fatty acid hydroxylation inhibits the transition from the metastable to some of the more stable phases. It also significantly increases the phase transition temperatures of both the metastable and stable phases indicating that it contributes to the hydrogen bonding network formed between the lipid molecules and helps overcome the lateral repulsive effect of the negatively charged sulfate. The C-15 cis double bond significantly lowers the temperature and enthalpy of the phase transition indicating that it increases the lateral separation of the lipid molecules and decreases the intermolecular hydrogen bonding interactions. However, it does not prevent formation of a more stable phase. By comparing the effect of various structural modifications reported here and earlier it could be concluded that fatty acid chain length has little effect on the phase transition temperature and enthalpy. This suggests that the forces between the lipid molecules may be dominated by head group interactions rather than interactions between the lipid chains. However, fatty acid chain length has a significant effect on the tendency of the hydroxy fatty acid species to form the more stable phase. The ease of formation of the stable phase increases with increase in chain length. Thus an increase in chain length helps overcome the kinetic barrier to stable phase formation presented by hydroxylation of the fatty acid.

Effect of hydrogen-bonding and non-hydrogen-bonding long chain compounds on the phase transitin temperatures of phospholipids

Abstract The effect of a number of long chain compounds on the phase transition temperature, Tm of several phospholipids was measured at different pH values by differential scanning calorimetry. Only single chain compounds capable of intermolecular hydrogen bonding interactions, such as palmitic acid, hexadecanol, and hexadecylamine, were able to form 2:1 (m/m) complexes with dipalmitoylphosphatidylcholine (DPPC), dihexadecylphosphatidylcholine (DHPC), and dipalmitoylphosphatidylglycerol (DPPG), which melted at a temperature 20 – 27°C higher than the pure lipid. The Tm-values of these complexes was similar to those of pure dipalmitoylphosphatidylethanolamine (DPPE) and dipalmitoylphosphatidic acid (DPPA) at neutral pH. These complexes formed only at pH values where both hydrogen bond donating and accepting groups were present. The hydrogen bonding compounds could be incorporated into DPPE and DPPA also but caused only a small increase in their Tm-values. Non-hydrogen bonding single chain C-16 compounds, such as methyl palmitate, hexadecane, and hexadecyl glycerol, had a smaller effect on the Tm-values of DPPC, DHPC, and DPPG, increasing them by only 4–12C°C. Furthermore these compounds decreased the Tm-values of DPPE And DPPA. These results suggested that the large increase in Tm produced in DPPC, DHPC and DPPG by the hydrogen bonding compounds and the large Tm-values of pure DPPE and DPPA are a result of intermolecular hydrogen bonding interactions involving the lipid phosphate and not of differences in size or charge of the lipid head groups. However, the smaller increase produced by the non-hydrogen bonding compounds may be the result of a reduction of the surface charge density of the bilayer. Consideration of hydrogen bonding interactions as well as head group size and charge characteristics helps to understand the behavior of these lipids and their role in biological membranes.

Effect of fatty acid chain length, fatty acid hydroxylation, and various cations on phase behavior of synthetic cerebroside sulfate

Abstract Synthetic species of CBS containing palmitic, stearic, lignoceric, D-2-hydroxy palmitic, or D-2-hydroxy stearic acid were prepared and their phase behavior in the presence of a number of mono- and divalent cations was studied by differential scanning calorimetry and the use of fatty acid spin labels. The results showed that both the non-hydroxy fatty acid (NFA) and hydroxy fatty acid (HFA) forms of cerebroside sulfate (CBS) can occur in two different gel states, a metastable state and a lower entropy stable state. The phase behavior is more sensitive to the type and concentration of cation present than in the case with acidic phospholipids. The sensitivity of the transition temperature (Tm) to cation concentration reflects, in part, increased participation of the lipid in intermolecular hydrogen bonding interactions as the negative charge of the sulfate is shielded. The extra hydroxyl group on the HFA also contributes to the intermolecular hydrogen bonding network causing a significant increase in the Tm. The HFA has an even more significant effect in causing inhibition of formation of the stable state. Formation of the stable state is also inhibited by Li+ and divalent cations. A similar mechanism may be involved, i.e.; cross-linking of adjacent lipids or increased intermolecular interactions inhibit the molecular rearrangement necessary to form the stable state. This inhibition is counteracted by an increase in fatty acid chain length. The results suggest that the stable state may be interdigitated as a result of the unequal chain length between the sphingosine base and the fatty acid.

Divalent Cation-Mediated Interaction Between Cerebroside Sulfate and Cerebrosides: An Investigation of the Effect of Structural Variations of Lipids by Electrospray Ionization Mass Spectrometry

Divalent cations mediate a carbohydrate-carbohydrate association between the two major glycolipids, galactosylceramide (GalCer) and its sulfated form, cerebroside sulfate (CBS), of the myelin sheath. We have suggested that interaction between these glycolipids on apposed extracellular surfaces of myelin may be involved in the stability or function of this multilayered structure. A mutant mouse lacking galactolipids because of a disruption in the gene that encodes a galactosyltransferase forms myelin that initially appears relatively normal but is unstable. This myelin contains glucosylceramide (GlcCer) instead of GalCer. To better understand the role of GlcCer in myelin in this mutant, we have compared the ability of divalent cations to complex CBS (galactosyl form) with GlcCer or GalCer in methanol solution by using positive ion electrospray ionization mass spectrometry. Because both the alpha-hydroxylated fatty acid species (HFA) and the nonhydroxylated fatty acid species (NFA) of these lipids occur in myelin, we have also compared the HFA and NFA species. In addition to monomeric Ca2+ complexes of all three lipids and oligomeric Ca2+ complexes of both GalCer and GlcCer, Ca2+ also caused heterotypic complexation of CBS to both GalCer and GlcCer. The heterotypic complexes had the greatest stability of all oligomers formed and survived better at high declustering potentials. Complexes of CBS with GlcCer were less stable than those with GalCer. This was confirmed by using the free sugars and glycosides making up the carbohydrate headgroups of these lipids. HFA species of CBS and GalCer formed more stable complexes than NFA species, but hydroxylation of the fatty acid of GlcCer had no effect. The ability of GlcCer to also complex with CBS, albeit with lower stability, may allow GlcCer to partially compensate for the absence of GalCer in the mouse mutant.

Interdigitated lipid bilayers of long acyl chain species of cerebroside sulfate: a fatty acid spin label study

The metastable phase behavior of semi-synthetic species of cerebroside sulfate (CBS), with hydroxy and non-hydroxy fatty acids from 16 to 26 carbons in length, was compared in Li+ and K+ using differential scanning calorimetry. The structure of the metastable and various stable phases formed in the presence of these two cations was investigated using a fatty acid spin label, 16-doxylstearate. A number of stable phases with successively higher phase transition temperatures and enthalpies occur in the presence of K+ (see the preceding paper). Li+ prevents formation of the most stable phases with the highest transition temperatures and enthalpies for all species of CBS. However, it does not prevent a transition from the metastable phase to the first stable phase of the longer chain C24 and C26 species. Furthermore, it allows C24:0h-CBS to undergo a similar transition, in contrast to a high K+ concentration, which prevents it. The spin label has anisotropic motion in the metastable gel phase formed by all species of CBS on cooling from the liquid crystalline phase. The spectra resemble those in gel phase phospholipids. The spin label is partially insoluble in the most stable phases formed by all the lipids, including the unsaturated C24:1 species, preventing further elucidation of their structure using this technique. However, the spin label is soluble in the first stable phase formed on cooling by the longer chain C24:0 and C26:0-CBS in Li+ and K+ and by C24:0h-CBS in Li+, and is motionally restricted in this phase. The motional restriction is similar to that observed in the mixed interdigitated bilayers of asymmetric species of phosphatidylcholine and fully interdigitated bilayers formed by symmetric phospholipids. It strongly suggests that the highly asymmetric long chain species of CBS form a mixed interdigitated bilayer in their first stable gel phases while the metastable phase of these and the shorter chain lipids may be partially interdigitated. The metastable phase of C24:1-CBS is more disordered suggesting that it may not be interdigitated at all. Thus the results suggest that (i) the hydroxy fatty acid inhibits but does not prevent formation of a mixed interdigitated bilayer by long chain species of CBS, (ii) an increase in non-hydroxy fatty acid chain length from 24 to 26 carbons promotes it, and (iii) a cis double bond probably prevents any form of interdigitation. These results may be relevant to the physiological and pathological roles of these structural modifications of CBS.

Do the long fatty acid chains of sphingolipids interdigitate across the center of a bilayer of shorter chain symmetric phospholipids

Novel cerebroside sulfate (CBS) spin labels containing long chain C24 or C26 fatty acids with a nitroxide spin label on the 22nd carbon were synthesized and used to investigate the ability of the long fatty acid chains of glycosphingolipids to interdigitate across the center of a non-interdigitated bilayer of phospholipids formed of symmetric saturated or unsaturated shorter fatty acid chain species, in the presence or absence of cholesterol. The motion of these long chain spin labels incorporated at 1 mole% in dimyristoylphosphatidylcholine (diC14-PC), dipalmitoylphosphatidylcholine (diC16-PC), distearoylphosphatidylcholine (diC18-PC), dibehenoylphosphatidylcholine (diC22-PC), spingomyelin (SM), 1-stearoyl-2-oleoylphosphatidylcholine (18:0.18:1-PC), and dimyristoylphosphatidylethanolamine (diC14-PE) was compared to that of CBS spin labels containing stearic acid spin labeled at the 5th carbon and at the 16th carbon. The results indicated that the C26 chain is interdigitated in the gel phase of diC14-PC, diC16-PC, SM, and possibly diC18-PC, but not diC14-PE, and the C24 chain may interdigitate in diC14-PC but not in the other phospholipids. Thus in order to interdigitate across the center of gel phase bilayers, the long acyl chain of the sphingolipid probably must be long enough to nearly span the phospholipid bilayer. The inability to interdigitate in diC14-PE is likely due to the close packing of this lipid in the gel phase. The C26 chain may also be interdigitated in these lipids in the presence of cholesterol at low temperatures. However, at physiological temperatures in the presence of cholesterol and in the liquid-crystalline phase of all the lipids, the results indicate that the long acyl chain of the glycosphingolipid is not interdigitated, but rather must terminate at the bilayer center. This may force the carbohydrate headgroup of the glycosphingolipid farther above the bilayer surface, allowing it to be recognized better by various carbohydrate binding ligands and proteins.

Photolabeling of myelin basic protein in lipid vesicles with the hydrophobic reagent 3-(trifluoromethyl)-3-(m-[125I]iodophenyl)diazirine.

The hydrophobic photolabel 3-(trifluoromethyl)-3-(m-[125I]iodophenyl)diazirine([125I]TID) was used to label myelin basic protein or polylysine in aqueous solution and bound to lipid vesicles of different composition. Although myelin basic protein is a water soluble protein which binds electrostatically only to acidic lipids, unlike polylysine it has several short hydrophobic regions. Myelin basic protein was labeled to a significant extent by TID when in aqueous solution indicating that it has a hydrophobic site which can bind the reagent. However, myelin basic protein was labeled 2-4-times more when bound to the acidic lipids phosphatidylglycerol, phosphatidylserine, phosphatidic acid, and cerebroside sulfate than when bound to phosphatidylethanolamine, or when in solution in the presence of phosphatidylcholine vesicles. It was labeled 5-7-times more than polylysine bound to acidic lipids. These results suggest that when myelin basic protein is bound to acidic lipids, it is labeled from the lipid bilayer rather than from the aqueous phase. However, this conclusion is not unequivocal because of the possibility of changes in the protein conformation or degree of aggregation upon binding to lipid. Within this limitation the results are consistent with, but do not prove, the concept that some of its hydrophobic residues penetrate partway into the lipid bilayer. However, it is likely that most of the protein is on the surface of the bilayer with its basic residues bound electrostatically to the lipid head groups.

Interaction of myelin basic protein and polylysine with synthetic species of cerebroside sulfate.

The effect of myelin basic protein on the myelin lipid cerebroside sulfate was studied by differential scanning calorimetry and use of the fatty acid spin label, 16-S-SL, in order to determine (i) the effect of basic protein on the metastable phase behavior experienced by this lipid, and (ii) to determine if basic protein perturbs the lipid packing as it does with some acidic phospholipids. The effects of basic protein on the thermodynamic parameters of the lipid phase transition were compared with those of polylysine which has an ordering effect on acidic phospholipids as a result of its electrostatic interactions with the lipid head groups. Different synthetic species of cerebroside sulfate of varying fatty acid chain length and with and without a hydroxy fatty acid were used. The non-hydroxy fatty acid forms of cerebroside sulfate undergo a transition from a metastable to a more ordered stable state while the hydroxy fatty acid forms remain in the metastable state at the cation concentration used in this study (0.01 M Na+ or K+). The non-hydroxy fatty acid forms were still able to go into a stable state in the presence of both basic protein and polylysine. At low concentrations, basic protein increased the rate of the transition to the stable state, while polylysine decreased it for the longest chain length form studied. However, at high concentrations, basic protein probably prevented formation of the stable state. The hydroxy fatty acid forms did not go into the stable state in the presence of basic protein and polylysine. It is argued that the increased rate of formation of the stable state in the presence of basic protein and decreased rate in the presence of polylysine are consistent with interdigitation of the lipid acyl chains in the stable state. Basic protein also had a small perturbing effect on the lipid. It decreased the total enthalpy of the lipid phase transition. When added to the non-hydroxy fatty acid forms it increased the temperature of the liquid crystalline to metastable phase transition and decreased the temperature of the stable to liquid crystalline phase transition. It significantly decreased the transition temperature of the hydroxy fatty acid forms but only a portion of the lipid was affected. In contrast, polylysine increased the transition temperature of the metastable and stable states of all forms of cerebroside sulfate but had a greater effect on the non-hydroxy fatty acids forms than on the hydroxy fatty acid forms.(ABSTRACT TRUNCATED AT 400 WORDS)

Adhesion of acidic lipid vesicles by 21.5 kDa (recombinant) and 18.5 kDa isoforms of myelin basic protein.

Myelin basic protein (MBP) is thought to be responsible for adhesion of the intracellular surfaces of compact myelin to give the major dense line. The 17 and 21.5 kDa isoforms containing exon II have been reported by others to localize to the cytoplasm and nucleus of murine oligodendrocytes and HeLa cells while the 14 and 18.5 kDa isoforms lacking exon II are confined to the plasma membrane. However, we show that the exon II(-) 18.5 kDa form and a recombinant exon II(+) 21.5 kDa isoform both caused similar aggregation of acidic lipid vesicles, indicating that they should have similar abilities to bind to the intracellular lipid surface of the plasma membrane and to cause adhesion of those surfaces to each other. The circular dichroism spectra of the two isoforms indicated that both had a similar secondary structure. Thus, both isoforms should be able to bind to and cause adhesion of the cytosolic surfaces of compact myelin. The fact that they do not could be due to differences in post-translational modification in vivo, trafficking through the cell and/or subcellular location of synthesis, but it is not due to differences in their lipid binding.