Modifications of cysteine residues in the solution and membrane-associated conformations of phosphatidylinositol transfer protein have differential effects on lipid transfer activity.

The alpha isoforms of mammalian phosphatidylinositol transfer protein (PITP) contain four conserved Cys residues. In this investigation, a series of thiol-modifying reagents, both alkylating and mixed disulfide-forming, was employed to define the accessibility of these residues and to evaluate their role in protein-mediated intermembrane phospholipid transport. Isolation and analysis of chemically modified peptides and site-directed mutagenesis of each Cys residue to Ala were also performed. Soluble, membrane-associated, and denatured preparations of wild-type and mutant rat PITPs were studied. Under denaturing conditions, all four Cys residues could be detected spectrophotometrically by chemical reaction with 4,4'-dipyridyl disulfide or 5,5'-dithiobis(2-nitrobenzoate). In the native protein, two of the four Cys residues were sensitive to some but not all thiol-modifying reagents, with discrimination based on the charge and hydrophobicity of the reagent and the conformation of the protein. With the soluble conformation of PITP, achieved in the absence of phospholipid vesicles, the surface-exposed Cys(188) was chemically modified without consequence to lipid transfer activity. Cys(188) exhibited an apparent pK(a) of 7.6. The buried Cys(95), which constitutes part of the phospholipid substrate binding site, was covalently modified upon transient association of PITP with a membrane surface. The Cys-to-Ala mutations showed that neither Cys(95) nor Cys(188) was essential for lipid transfer activity. However, chemical modification of Cys(95) resulted in the loss of lipid transfer activity. These results demonstrate that the Cys residues of PITP can be assigned to several different classes of chemical reactivity. Of particular interest is Cys(95), whose sulfhydryl group becomes exposed to modification in the membrane-associated conformation of PITP. Furthermore, the inhibition of PITP activity by thiol-modifying reagents is a result of steric hindrance of phospholipid substrate binding.

Structure of a multifunctional protein. Mammalian phosphatidylinositol transfer protein complexed with phosphatidylcholine.

Eukaryotic phosphatidylinositol transfer protein is a ubiquitous multifunctional protein that transports phospholipids between membrane surfaces and participates in cellular phospholipid metabolism during signal transduction and vesicular trafficking. The three-dimensional structure of the alpha-isoform of rat phosphatidylinositol transfer protein complexed with one molecule of phosphatidylcholine, one of its physiological ligands, has been determined to 2.2 A resolution by x-ray diffraction techniques. A single beta-sheet and several long alpha-helices define an enclosed internal cavity in which a single molecule of the phospholipid is accommodated with its polar head group in the center of the protein and fatty acyl chains projected toward the surface. Other structural features suggest mechanisms by which cytosolic phosphatidylinositol transfer protein interacts with membranes for lipid exchange and associates with a variety of lipid and protein kinases.

X-ray analysis of crystals of rat phosphatidylinositol-transfer protein with bound phosphatidylcholine.

Phosphatidylinositol-transfer protein (PITP) is a soluble, ubiquitously expressed, highly conserved protein encoded by two genes in humans, rodents and other mammals. A cDNA encoding the alpha isoform of the rat gene was expressed to high levels in Escherichia coli, the protein purified and the homogeneous protein used for crystallization studies. Crystals of rat PITP-alpha were obtained by vapor-diffusion techniques using the sitting-drop method. Crystals grow within two weeks by vapor-diffusion techniques in the presence of polyethylene glycol 4000. Both crystal forms pack in the monoclinic space group P21. Crystal form I has unit-cell parameters a = 44.75, b = 74.25, c = 48.32 A and beta = 114.14 degrees. Unit-cell parameters for crystal form II are a = 47.86, b = 73.59, c = 80.49 A and beta = 98.54 degrees. Crystal form I has a Vm of 2.295 A3 Da-1 and an estimated solvent content of 46.4% with one molecule per asymmetric unit, while crystal form II has a Vm of 2.196 A3 Da-1 and an estimated solvent content of 44.0%, assuming two molecules per asymmetric unit.

The C-terminus of phosphatidylinositol transfer protein modulates membrane interactions and transfer activity but not phospholipid binding.

Rat phosphatidylinositol transfer protein (PITP) is a 32 kDa protein containing 271 amino acids. It is involved in a number of cell functions including secretion and cell signaling. To further characterize structure/activity relationships of PITP, two C-terminal truncated derivatives, PITP(1-259) and PITP(1-253), were produced in Escherichia coli and purified to homogeneity. PITP(1-259) had transfer activity equal to 30-40% to that of native PITP in transfer of either phosphatidylcholine (PC) or phosphatidylinositol (PI) when transfer was measured using 95/5 mol% PC/PI donor and acceptor vesicles; PITP(1-253) had only slight transfer activity, even under the most favorable assay conditions. Thus, amino acids 254-258 are critical for transfer activity. The transfer activity of PITP(1-259) was strongly dependent on the composition of the donor and acceptor vesicles. With 100 mol% PC donor and acceptor vesicles, PITP(1-259) transfer activity ranged from 70 to 100% to that of PITP. The presence of 2 mol% phosphatidic acid (PA) in either donor or acceptor vesicles reduced transfer activity to between 10 and 20% that of full-length PITP under the same conditions. If both donor and acceptor contained 2% PA, PITP(1-259) was essentially inactive, though the activity of PITP was not affected significantly under these conditions. PITP(1-253) and PITP(1-259) bind much more avidly to vesicles than does PITP, and this enhanced binding reflects increased electrostatic interactions. Thus, the C-terminal residues modulate the affinity of PITP for vesicles and the efficiency of phospholipid transfer.

Importance of phospholipid in the folding and conformation of phosphatidylinositol transfer protein: comparison of apo and holo species.

The significance of noncovalently bound phospholipid as a structural component of phosphatidylinositol transfer protein (PITP) and its role in acquisition and maintenance of the native conformation of the protein have been addressed by studying the refolding of PITP after exposure to 6 M guanidinium chloride (GdnCl). Protein conformations were characterized by (1) the intrinsic tryptophan fluorescence, circular dichroism, and absorbance spectroscopy, (2) the degree of binding of the fluorescent probe 1,8-ANS, and (3) limited proteolytic digestion. When the GdnCl concentration was reduced 100-fold by rapid dilution at 25 degrees C, practically all of the native transfer activity was regained within 20 min. Endogenous phospholipid demonstrated a strong interaction with the native PITP. Separation of the phospholipid from the protein by chromatography on a lipophilic matrix was achieved only under denaturing conditions and resulted in spontaneous oxidation of the apo-protein, accompanied by almost complete loss of recoverable transfer activity. Under reducing conditions, however, apo-PITP recovered more than 80% of the native transfer activity and was similar to holo-PITP in the kinetics of phospholipid transfer. Renatured apo-PITP demonstrated a significant relaxation of the tertiary structure, compared to native and renatured holo-PITP. Incubation of apo-PITP with phospholipid vesicles resulted in a more compact protein conformation. We conclude that the polypeptide can spontaneously fold to a native-like conformation, sufficient for interaction with a lipid membrane and acquisition of a phospholipid ligand. Binding of a phospholipid ligand brings about the final adjustments of protein conformation to the more compact native structure.

Truncations of the C-terminus have different effects on the conformation and activity of phosphatidylinositol transfer protein.

Contributions of the C-terminus toward the conformation and activity of phosphatidylinositol transfer protein (PITP) were studied by comparing properties of the 271 amino acid, full-length protein, PITP(1-271), and two truncated species, PITP(1-259) and PITP(1-253). Using recombinant proteins and an in vitro phospholipid transfer assay with phosphatidylcholine vesicles, the activities of PITP(1-271) and PITP(1-259) were identical, while the activity of PITP(1-253) was almost totally abolished. By most physical and chemical criteria, however, PITP(1-259) and PITP(1-253) were virtually indistinguishable and differed significantly from the full-length protein. Results of second derivative analysis of absorbance spectra were consistent with an additional two Tyr residues being exposed to the solvent in PITP(1-259) and PITP(1-253) in comparison to PITP(1-271). Only one out of four Cys residues in PITP(1-271) reacted with dithiobisnitrobenzoic acid, while two Cys residues were accessible in both truncated species. Quenching of intrinsic Trp fluorescence by acrylamide demonstrated an increase in exposure of Trp residues in both PITP(1-259) and PITP(1-253); binding of the fluorescence probe 1,8-ANS to these proteins was also significantly higher compared to PITP(1-271). These results describe a more relaxed overall tertiary structure brought about by the C-terminal truncations. This altered structure did not affect the stability of the truncated proteins, as indicated by equilibrium unfolding in guanidinium chloride. Refolding of the denatured PITP(1-259), however, was considerably slower than that of full-length PITP. Our study suggests a critical role of the C-terminal residues 254-259 in transfer activity of PITP. Residues 260-271, on the other hand, appear to be more important for the rapid folding and maintenance of a compact native conformation of the protein.

Limited proteolysis of rat phosphatidylinositol transfer protein by trypsin cleaves the C terminus, enhances binding to lipid vesicles, and reduces phospholipid transfer activity.

Rat phosphatidylinositol transfer protein (PITP) is a 32-kDa protein of 271 amino acids that transfers phosphatidylinositol and phosphatidylcholine between membranes. The alpha isoform of rat PITP was expressed in Escherichia coli and purified in high yields. The purified protein contained 1 mol of phosphatidylglycerol and had a transfer activity for phosphatidylinositol and phosphatidylcholine equal to or greater than that of PITP purified from mammalian brain. Limited protease digestion was used to further define structure, activity, and function relationships in PITP. PITP alone is relatively resistant to digestion by chymotrypsin, trypsin, and Staphylococcus V8 protease but is readily cleaved by subtilisin. Phospholipid vesicles containing phosphatidic acid enhance susceptibility to digestion by all four proteases. In the presence of vesicles, PITP, which migrates as a 36-kDa protein in SDS-polyacrylamide gel electrophoresis, is cleaved rapidly by trypsin to a form that appears to be 2-3 kDa smaller than the native form. The tryptic fragment retains partial phospholipid transfer activity and shows an enhanced affinity for phospholipid vesicles containing phosphatidic acid. Analysis of the tryptic digestion products by immunoblotting, N-terminal sequencing, and electrospray mass spectrometry showed that trypsin cleaves the C terminus of PITP at Arg253 and Arg259. Thus, removal of the C terminus enhances the affinity of PITP for vesicles and results in a dimunition of transfer activity. Overall, the data show that PITP undergoes conformation changes and that the C terminus becomes more accessible to trypsin when bound to vesicles. Hence, the C terminus is not an essential component of the membrane binding site and may be located distal to it.

Sequence of a human cDNA encoding phosphatidylinositol transfer protein and occurrence of a related sequence in widely divergent eukaryotes.

Phosphatidylinositol (PtdIns) transfer protein (PtdInsTP) is a phospholipid transfer protein that has been detected in all mammalian tissues examined. It catalyzes the transfer in vitro of PtdIns and phosphatidylcholine between membranes in a number of natural and artificial membrane systems and may be involved in secretion in vivo. In previous studies, we isolated and sequenced a cDNA encoding a rat PtdInsTP. A rat cDNA probe was used to isolate clones from a lambda gt11 human testis cDNA library which encoded full-length human PtdInsTP. The cDNA sequence defines a 270-amino-acid, 31.8-kDa protein whose sequence shares 98.9% identity to that of rat, making it one of the most conserved proteins known between the two species. DNA blot hybridization studies suggest that there may be more than one gene encoding this protein in humans. A comparison of rat and human PtdInsTP cDNAs revealed strong sequence similarity (88 and 84%) in portions of the corresponding 5'- and 3'-untranslated regions (UTR) of the rat and human mRNAs.

Localization of the gene encoding human phosphatidylinositol transfer protein (PITPN) to 17p13.3: a gene showing homology to the Drosophila retinal degeneration B gene (rdgB).

The human gene for phosphatidylinositol transfer protein (PITPN) has previously been shown to share sequence and functional homology to part of the Drosophila retinal degeneration B gene (rdgB). In view of the possible involvement of the PITPN locus in the etiology of retinal disease, the gene has been mapped to human chromosome 17p13.3 and mouse Chromosome 11.

Phospholipid transfer activity is relevant to but not sufficient for the essential function of the yeast SEC14 gene product.

To investigate several key aspects of phosphatidylinositol transfer protein (PI-TP) function in eukaryotic cells, rat PI-TP was expressed in yeast strains carrying lesions in SEC14, the structural gene for yeast PI-TP (SEC14p), whose activity is essential for Golgi secretory function in vivo. Rat PI-TP expression effected a specific complementation of sec14ts growth and secretory defects. Complementation of sec14 mutations was not absolute as rat PI-TP expression failed to rescue sec14 null mutations. This partial complementation of sec14 lesions by rat PI-TP correlated with inability of the mammalian protein to stably associate with yeast Golgi membranes and was not a result of rat PI-TP stabilizing the endogenous sec14ts gene product. These collective data demonstrate that while the in vitro PI-TP activity of SEC14p clearly reflects some functional in vivo property of SEC14p, the PI-TP activity is not the sole essential activity of SEC14p. Those data further identify an efficient Golgi targeting capability as a likely essential feature of SEC14p function in vivo. Finally, the data suggest that stable association of SEC14p with yeast Golgi membranes is not a simple function of its lipid-binding properties, indicate that the amino-terminal 129 SEC14p residues are sufficient to direct a catalytically inactive form of rat PI-TP to the Golgi and provide the first evidence to indicate that a mammalian PI-TP can stimulate Golgi secretory function in vivo.

Transport of phosphatidylinositol to rat hepatocyte plasma membrane catalyzed by phosphatidylinositol transfer protein.

Plasma membrane sheets were isolated from fresh rat liver and characterized by electron microscopy and marker enzyme activities. Plasma membrane sheets were used as the acceptor membrane in the measure of transport of phosphatidyl[3H]inositol from small unilamellar phospholipid vesicles or rough endoplasmic reticulum donor membranes. Catalysis of this transport was achieved with phosphatidylinositol transfer protein purified from rat or bovine brain. Assays were designed to separate donor and acceptor membranes by density gradient centrifugation. Rates of transfer were directly proportional to incubation time and the amounts of transfer protein and plasma membrane sheet added. These results are discussed in terms of cellular phosphatidylinositol metabolism, membrane phospholipid composition, and vesicle trafficking in rat hepatocytes.

Stimulation of cholinephosphotransferase activity by phosphatidylcholine transfer protein. Regulation of membrane phospholipid synthesis by a cytosolic protein.

The effect of rat liver phosphatidylcholine transfer protein on the incorporation of CDP-choline and dioleoylglycerol into phosphatidylcholine catalyzed by rat liver microsomal CDP-choline: 1,2-diacyl-sn-glycerol cholinephosphotransferase was studied. In the presence of phosphatidylcholine transfer protein, the incorporation of CDP-choline into phosphatidylcholine was markedly stimulated. Phosphatidylcholine transfer protein isolated from either rat or bovine liver was capable of this stimulatory effect; in contrast, phosphatidylinositol transfer protein from rat liver had no effect on phosphatidylcholine synthesis. Kinetic analysis showed that microsomal phosphatidylcholine synthesis increased 2.4-fold after 1 min and reached a maximum of approximately 10-fold within 10 min in the presence of phosphatidylcholine transfer protein; in the absence of this protein phosphatidylcholine synthesis stopped after 2-4 min. These results suggest that phosphatidylcholine transfer protein permits phosphatidylcholine synthesis to proceed further. With the addition of phospholipid vesicles, as an acceptor membrane in the reaction mixture, there was a significant amount of protein-mediated transfer of synthesized phosphatidylcholine to the vesicles. Measurable transfer of synthesized phosphatidylcholine to vesicles could only be detected after a lag of 2-4 min. The stimulation of cholinephosphotransferase could be nearly abolished by increasing the amount of added phospholipid vesicles; concurrently, a greater transfer to the vesicles was observed. These results describe a new property of phosphatidylcholine transfer protein which may be of physiological significance in the regulation of phosphatidylcholine synthesis in mammalian tissues.

Developmental patterns in rat brain of phosphatidylinositol synthetic enzymes and phosphatidylinositol transfer protein.

Phosphatidylinositol synthetic and intermembrane transfer activities were studied in rat in the developing whole brain and isolated cerebellum. Specific activities of CTP:phosphatidate cytidylyltransferase and CDPdiacylglycerol:inositol phosphatidyltransferase were found to have similar developmental patterns. Levels of phosphatidyltransferase seen in fetal animals (whole brain only) and neonatal (whole brain and cerebellum) were maintained through approximately postnatal day 15, peaked at day 28, and then declined to somewhat higher than fetal levels at day 60. Cytidylyltransferase activity varied from the phosphatidylinositol synthesizing enzyme in that specific activity continued to increase up to day 60. Whole brain phosphatidylinositol transfer specific activity showed a sharp peak at postnatal day 9 after which activity was maintained at or above the fetal levels to day 60. Cerebellum phosphatidylinositol transfer specific activity had a similar peak which was delayed 7-10 days compared to the whole brain. Phosphatidylinositol transfer protein was also determined immunologically: whole brain levels increased dramatically from fetal day 16 to 18 and then remained relatively constant, while cerebellum levels (measured from postnatal day 7) displayed a variable profile between days 7 and 28. The developmental pattern of CTP:phosphatidate cytidylyltransferase in rat brain is reported here for the first time.

Isolation and sequence of cDNA clones encoding rat phosphatidylinositol transfer protein.

Phosphatidylinositol (PtdIns) transfer protein is a cytosolic protein that catalyzes the transfer of PtdIns between membranes. It is expressed in organisms from yeast to man, and activity has been found in all animal tissues examined. Using antibodies prepared against bovine brain PtdIns transfer protein, lambda gt11 rat brain cDNA libraries were screened and several clones isolated. DNA sequence analysis showed that the cDNAs encoded a polypeptide of 271 amino acids with a mass of 31,911 Da. Comparison of the deduced amino acid sequence with N-terminal sequence data obtained for the intact purified bovine brain protein and rat lung phospholipid transfer protein verified that the cDNAs were PtdIns transfer protein clones. The predicted protein shows no significant sequence similarity to other known (phospholipid)-binding proteins. DNA blot hybridization suggests that the rat genome may contain more than one gene encoding PtdIns transfer protein. RNA blot hybridization reveals that the PtdIns transfer protein gene is expressed at low levels in a wide variety of rat tissues; all tissues examined showed a major mRNA component of 1.9 kilobases and a minor component of 3.4 kilobases. The isolation of clones encoding rat PtdIns transfer protein will greatly facilitate studies of the structure and function of PtdIns transfer proteins and their role in lipid metabolism.

Mature rat testis contains a high molecular weight species of phosphatidylinositol transfer protein.

Immunoblot analysis of a rat testis cytosol fraction revealed two proteins which reacted with a polyclonal rabbit antibody to bovine phosphatidylinositol transfer protein. These two proteins were separated by anion exchange and molecular sieve column chromatographic procedures and shown to catalyze the transfer of phosphatidylinositol and phosphatidylcholine between populations of small unilamellar vesicles. One protein was identified as the phosphatidylinositol transfer protein detectable in 16 other rat tissues and many eukaryotic species; the other phosphatidylinositol transfer protein was unique to testis. The molecular masses of the proteins, determined under denaturing electrophoretic conditions, were 35 and 41 kDa, respectively. When testis was examined in animals from birth to six weeks of age, the 35-kDa protein was present throughout, while the 41-kDa protein first appeared during week 4 and increased to adult levels by week 6; a small yet significant increase in tissue phosphatidylinositol transfer activity accompanied this expression of the testis-specific protein. Selective destruction of Leydig cells by ethylene dimethanesulfonate did not cause any detectable loss of the 41-kDa phosphatidylinositol transfer protein. The structural and catalytic relationships between the two testicular phosphatidylinositol transfer protein species remain to be elucidated.

Tissue distribution, purification and characterization of rat phosphatidylinositol transfer protein.

Phosphatidylinositol transfer activity is measured in cytosol fractions prepared from 13 rat tissues; specific activity is highest in brain and lowest in adipose and skeletal muscle. Based upon electrophoretic analysis phosphatidylinositol transfer protein is purified to homogeneity from whole rat brain. The protein has a molecular weight of 36,000 and exists as a mixture of species having isoelectric points of 4.9 and 5.3. In a vesicle-vesicle assay system, the intermembrane transfer rate is greatest for phosphatidylinositol and less by a factor of 2 for phosphatidylcholine; transfer of phosphatidylethanolamine, phosphatidylserine or sphingomyelin is not observed. Using a polyclonal rabbit antibody against bovine phosphatidylinositol transfer protein, immunologic cross-reactivity is noted between the rat protein and other mammalian phosphatidylinositol transfer proteins. A strong correlation is established between a tissue's capacity for phosphatidylinositol transfer and the amount of immunoreactive transfer protein seen in that tissue. Purified phosphatidylinositol transfer protein is capable of transporting newly synthesized phosphatidylinositol molecules from rat brain microsomes to small unilamellar phospholipid vesicles. The results are discussed within the context of cellular phosphoinositide metabolism and the maintenance of the metabolically responsive pool of phosphatidylinositol in the plasma membrane.

General kinetic model for protein-mediated phospholipid transfer between membranes.

Phospholipid transfer protein catalyzes the transfer of phospholipids between bilayer membranes. A general model is developed for describing the kinetics of this process. While previous models derive detailed expressions only for the initial rate of transfer from donor to acceptor membranes, this model takes into account donor-to-donor, acceptor-to-acceptor, and acceptor-to-donor transfers, in addition to the usual donor-to-acceptor transfer. The apparent rate of transfer along any of these specific routes is given as the product of the total rate of transfer (the sum of the rates of transfer along all four routes) and a probability function uniquely defined for each route. The model explains adequately the effects of membrane concentration on phospholipid transfer activity as well as the consequences of varying membrane surface charge and size. Using bovine liver phosphatidylcholine transfer protein, the model is applied to the kinetic analysis of phosphatidylcholine transfer between two populations of small unilamellar vesicles. Rates of protein-catalyzed phosphatidylcholine transfer between vesicles with identical phosphatidic acid content (2 or 6 mol%) are determined experimentally as a function of total vesicle concentration to calculate apparent dissociation constants and maximum rates of transfer; apparent rates of transfer between various combinations of vesicles containing 2 or 6 mol% phosphatidic acid are then deduced from the derived velocity expression. Reasonably good agreement is seen between theoretical apparent rate-vesicle concentration relationships and those measured experimentally. The results support the general treatment of the kinetics of protein-mediated phospholipid transfer and permit an estimation of useful kinetic parameters.

Effects of dietary fish oil on platelet function and plasma lipids in hyperlipoproteinemic and normal subjects.

We studied the effects of dietary supplementation with an encapsulated fish oil concentrate (Maxepa) on platelet function, fibrinolysis, and plasma lipids and lipoproteins in 9 normal subjects, 10 patients with type IV hyperlipoproteinemia, and 6 with type IIB hyperlipoproteinemia. After a baseline period, the subjects crossed over randomly between treatment periods with Maxepa (providing 3.24 g eicosapentaenoic acid and 2.16 g docosahexaenoic acid per day) and safflower oil (used as a control), given for 6 weeks each. Administration of Maxepa led to a slight prolongation of the bleeding time in all groups and to modest inhibition of platelet aggregation in the type IV hyperlipoproteinemics and normal subjects, with partial (41%) inhibition of thromboxane synthesis from baseline levels noted in the normal group. Plasma total fibrinolytic actively did not change significantly in any group. Maxepa treatment resulted in a marked decrease in triglyceride and VLDL-cholesterol and a slight increase in HDL-cholesterol was noted after Maxepa in the type IV hyperlipoproteinemics (4.11 +/- 0.13 mmol/l vs. 3.10 +/- 0.16 mmol/l, Maxepa vs. safflower oil). We conclude that dietary supplementation with fish oil results in a relatively minor degree of inhibition of platelet function in normal and hyperlipoproteinemic subjects, and a potentially adverse increase in LDL-cholesterol in type IV hyperlipoproteinemics.