Saturday, May 9, 2015

IMPLICATIONS OF THIOLESTER LINKED FATTY ACIDS IN APOLIPOPROTEIN B.

IMPLICATIONS OF THIOLESTER LINKED FATTY ACIDS IN APOLIPOPROTEIN B

INTRODUCTION

Apolipoprotein B (ApoB) is obligatory for triglyceride transport. Why ApoB is singled out among all apolipoproteins is not well understood. ApoB is the most hydrophobic apolipoprotein among all known apolipoproteins. It has the highest tendency to undergo aggregation. If one examines the amino acid composition of ApoB it resembles any other common water-soluble pro­ tein. Why, then, is ApoB so hydrophobic? We believe that the thiolester bound fatty acids recently found in ApoB 1•

may play an important role in contributing to the hydrophobic nature of ApoB.

HYDROPATHY OF ApoB

One can determine the hydrophobicity of ApoB by calculating the average hydropathy value for ApoB (i.e., determination of hydrophobicity or hydro­ philicity) using the hydropathy index of individual amino acids3•

The high negative value represents high hydrophilicity and the high positive value represents high hydrophobicity. These values were assigned based pri­ marily on the free energies of transfer for the side-chains of the amino acid from water phase into organic phase, and the numbers were tested with proteins with known characteristics 3 We tabulated the amino acid composition of ApoB from one of the sequence data5

(Table 1). The average hydropathy for ApoB was calculated to be -0.31 (Table 1). This negative value re­ flects the hydrophilic nature of ApoB based solely on the protein moiety. Comparing this value to that of the water-soluble serum albumin, which is calculated to be -0.39 (Table 1), the hydrophilicity of ApoB is not too far from that of albumin. For a protein with such a hydrophilic property, it is unreasonable to expect it to be insoluble in aqueous buffers.

image

The hydropathy values of ApoB and albumin may be compared to those of other known proteins as reported by Kyte and Doolittle3. As shown in Fig. 1, nearly all the points for hydrophobic membrane proteins are above the ero line, whereas most of the points for hydrophilic soluble proteins are below the zero line. ApoB and albumin would fall near the median of the hydrophilic points even though the chain length of ApoB would place it on the fifth frame toward the right. These data show that, based on the amino acid composition, ApoB more closely resembles a water-soluble globular pro­ tein than an insoluble hydrophobic membrane protein.

Alternatively, one can calculate the hydropathy profile along the se­ quence to locate the hydrophobic regions, as many of the investigators have done with ApoB5• •

Olofsson et a1.

found 39 sequences with high hydro­ phobicity, having an average hydropathy index of 2.6 ± 0.5, similar to membrane-spanning sequences. The average length of the sequences is only 7 ± 2 amino acid residues. Knott et al.8 found 11 sequences lying at the lipid/

aqueous interface with strong hydrophobicity. For an apolipoprotein with amino acid residues over 4500, necessary to encompass an enormous quantity of neutral lipid core, this small number of hydrophobic regions appears to be extremely meager, even with the inclusion of a number of regions containing amphipathic -sheet structure which has also been suggested to partici­ pate in lipid binding 7• •

Thus, it is not surprising that bound fatty acids are present in ApoB to enhance its hydrophobicity.

LINKAGES BETWEEN FATTY ACIDS AND ApoB--THIOLESTER BONDS

Methylamine (MA), a reagent specifically reactive with thiolester and which forms a covalently modified product 10, was used to study the presence of thiolester bonds in ApoB 1 •

Freshly isolated human plasma ApoB was first reduced and carboxymethylated to eliminate further participation of sulfhydryls and disulfide linkages in the carboxymethylation reaction of the newly generated sulfhydryls. The reduced and carboxymethylated (RCM-)ApoB

image

Fig. 1. Plot of average hydropathy (grease index) of various pro­ teins against their chain length: (X) 84 soluble enzymes with known sequences; (0)8 membrane-embedded proteins with known sequences; ( )8 putative proteins of human mitochondria. Figure reproduced from Kyte and Doolittle3 with permission.

image

in 6 M urea/Tris containing preservatives was incubated with [c]MA at pH 8.S and 30°C1. Covalent incorporation of [ C]MA was observed in RCM-ApoB with concomitant generation of new sulfhydryl groups, which could be blocked with [ H]- or [ c]iodoacetate. The molar ratio of S-[ H]CM/N-[c]MA in- corporation was near one at one-half hr incubation, suggesting the possible presence of intramolecular thiolester linkage(s) (see Table 2) •

incubation time increased, the incorporation of S-[increased in RCM­ ApoB without the concomitant increase in N-[ 14 incorporation, suggesting the possible presence also of intermolecular thiolester linkages, because the ligand containing N-[ 14 was separated from the protein. After 24 hr incubation, up to 10 mols of extra sulfhydryl groups were generated (Table ligands were soluble in organic solvent.

Following purification by thin-layer chromatography with two solvent systems (solvent system 1: hexane/diethyl ether/glacial acetic acid 113:3S:3 v/v; solvent system 2: hexane/ethyl acetate/methanol/glacial acetic acid 90:20:20:2 v/v) the [c]MA-derivatives were transesterified in methanolic HCl. Gas-liquid chromatography identified the methyl ester as palmitate and stearate at 1:1 molar ratio for human ApoB. Therefore, the palmitic and stearic acids are covalently linked to ApoB via intermolecular thiolester linkages to the side chain of cysteine residues. So what has happened is this (Fig. 2): thiolester linked fatty acids on ApoB are sensitive to alkaline attack. In the presence of [ C]MA, covalent products hexadecanoyl amide and [c]-methyl octadecanoyl amide are formed. After transesterification in methanolic HCl these fatty acyl amides are converted to methyl fatty acyl ester and identified by GLC as fatty acids. In ApoB, new -SH groups are generated.

The same method was applied to rat plasma RCM-ApoB, thiolester linked fatty acids were also observed in rat ApoB 2 However, in addition to the bound palmitate and stearate as found in human ApoB, a small amount of bound oleate was also observed in rat ApoB 2 difference is not yet kn?wn·

(Table 3). The significance of this Bound palmitic and stearic acids were also found by Fisher after hy­ drolysis of chymotryptic peptides of totally delipidized ApoB 12 13 He est­ imated at least 8 mols of fatty acids were present in ApoB 12 The presence of covalently bound palmitate was also observed recently in ApoB secreted by HepG2 cells14 Thus far, the only other apolipoprotein known to be acylated with long­ chain fatty acid is apolipoprotein A-I, by ester linkage15 However, the fatty acid was reportedly removed before apolipoprotein A-I was secreted into the circulation 15. Thus, ApoB is the only acylated apolipoprotein present in plasma and the only secreted apolipoprotein which is acylated via thiolester.

SIGNIFICANCE OF INTERMOLECULAR THIOLESTER BOUND FATTY ACIDS  clip_image006Thiolester is a high energy labile linkage. It is readily broken in vitro. Under alkaline conditions, free fatty acids and sulfhydryls are gen­ erated. The free fatty acids may cause ApoB to aggregate in aqueous solu-

image

image

tions. If the newly generated sulfhydryls are exposed to the surface of ApoB, they are susceptible to autoxidation. This autoxidation process is via the free radical pathway 16 Inter- or intramolecular -S-S- bridges may be produced by autoxidation of -SH groups and H o is generated as a by- product In the presence of low density lipoproteins (LDL), H o may oxidize lipids and causes lipid peroxidation of LDL. The presence of labile thiolester linked fatty acids in ApoB explains why ApoB has a strong tendency towards aggregat1.on, why ApoB sens1.t1.ve to autox1"dat1·on17, and why LDL is susceptible to lipid peroxidation.

PROTEINS ACYLATED WITH PALMITATE

Based on the literature information, a large number of proteins in eukaryotic cells are known to be acylated with palmitate (Table 4): the transforming protein of Harvey sarcoma virus, p21 ras 18-21 , the surface glycoproteins of several enveloped viruses22- , the histocompatibility complex antigens26, the mammalian transferrin receptor27, the erythrocyte membrane skeleton protein ankyrin28, the membrane glycoprotein rhodopsin 29, and the neuronal growth cone protein, GAP-4330 Palmitic acid is most often found to be linked through a thiolester bond to the cysteine residue. This has been established conclusively in the case of the G glycoprotein of vesicular stomatitis virus25, p21 ras19, the GAP-4330, the transferrin receptor31 , and the histocompatibility complex antigens26 Unlike myristic acid, which is only found attached through an amide bond to amino-terminal residues (most frequently, glycine), palmitic acid is found to modify cysteines present within the sequence of a protein. These residues are generally found near membrane-binding domains of a protein and usually on the cytoplasmic face of a membrane 34. Myristylation is a co-translational modification process while palmitylation is a post-translational process34. The presence of thiolester linked palmitate at Cys-186 in Harvey murine sarcoma (HMS) p21 ras protein was reported to be essential for the biological activities of ras proteins: for transformation of NIH/3T3 cells, for membrane localiza­ tion, and for lipid binding 21 The absence of Cys-186 or the replacement of Cys-186 with Ser-186 by point mutation causes the total loss of lipid bind ing capacity and the biological activities of HMS p21 ras proteins. In

image

addition, the presence of Cys at any other location in the sequence of ras proteins could not be palmitylated to acquire the biological activities. Interestingly, when the nonpalmitylated p21 ras derivatives were myristy­ lated at their amino termini, the p21 ras became activated and restored the membrane association and full transforming activity. Furthermore, myristy­ lated forms of normal cellular ras also became transforming 35 Thus, the normal function of cellular ras is diverted to transformation by myristate

and this suggests that the normal ras must be regulated ordinarily by some unique property of palmitate that myristate does not mimic 35 These data point out the importance of post-translational events. They suggest that fatty acid acylation is 1) for a functional purpose, 2) site specific and 3) fatty acid specific.

Acylation of p21 N-ras is reportedly a dynamic event with a rapid acyl­ ation-deacylation cycle, (half life about 20 min) even in the absence of prote1.n synthes1.s36 Acy1at1.on of p21 N 36 and of ankyr1.n24 is resistant to inhibitors of protein synthesis while most other cellular acyl proteins are sensitive to these agents. Most protein palmitylation occurs in the early Golgi complex 23 37, suggest1ng that the enzymatic machinery for acylation of p21 N-ras and ankyrin may differ from that of the bulk of cellular palmitylated proteins 36 Alternatively, there may be two pools of acylation substrates, one freely accessible to the acylation machinery (e.g., p21 N- ras) and the other inaccessible due to intracellular localization 36 .

The neuronal growth cone protein, GAP-43 is synthesized initially as a soluble protein that becomes attached to membranes posttranslationally after early fatty acylation at the only two cysteine residues, Cys-3 and Cys-4 and becomes an insoluble protein 30 Taking this information all together, we may predict that fatty acid acylation of ApoB may also be reversible, which offers the potential for dynamic regulation of ApoB and the assembly of its lipoproteins. It is pos­sible that ApoB may be synthesized initially as a soluble protein, like GAP-4330, which becomes insoluble after acyltation with long chain fatty acids.

EFFECT OF FATTY ACYL CHAIN LENGTH

Using acylated synthetic peptides with varying acyl chain length, Ponsin38 demonstrated that the hydrophobicity of the lipid-associating peptide increased with the acyl chain length and the binding constant to a phospholipid matrix or a high density lipoprotein model increased by 3 orders of magnitude as the length of acyl chain increased from 0 to 16 carbon units. It was concluded that for a given helical potential, the binding of a lipid-associating peptide to lipoprotein is governed by its hydrophobicity. Applying this knowledge to the acylation of ApoB, it sug­ gests that the hydrophobicity of the loci with sterylation is higher than those with palmitylation, and far higher than the loci without acylation.

CONCLUSIONS

The literature information implies that we can predict that fatty acid acylation of ApoE is a means of increasing the hydrophobicity of the ape­ lipoprotein for membrane attachment and lipid binding. Although the specif­ ic function of acylation of ApoB is still unknown, the selectivity suggests its possible role in assembly, secretion, lipid binding or transport. The acylation sites are most likely specific. The chain length of fatty acid and the possible reversibility of acylation may serve to modulate the func­tion of ApoB. Probably it is this regulated fatty acid acylation that singles out ApoB and makes it obligatory for triglyceride transport.

REFERENCES

1. G. Huang, D. M. Lee and S. Singh, Identification of the thiol ester linked lipids in apolipoprotein B, Biochemistry 27:1395 (1988).

2. V. S. Kamanna and D. M. Lee, Presence of covalently attached fatty acids in rat apolipoprotein B via thiolester linkages, Biochem. Biophys. Res. Commun. 162:1508 (1989).

3. J. Kyte and R. F. Doolittle, A simple method for displaying the hydro­ pathic character of a protein, J. Mol. Biol. 157:105 (1982).

4. T. P. Hopp and K. R. Woods, Predi ti of protein antigenic determi­ nants from amino acid sequences, Proc. Nat!. Acad. Sci. 78:3824 (1981). - -- -- -

5. S.-H. Chen, C.-Y. Yang, P. F. Chen, D. Setzer, M. Tanimura, W.-H. Li, A. M. Gotto and L. Chan, The complete eDNA and amino acid sequence of human apolipoprotein B-100, J. Biol. Chern. 261:12918 (1986).

6. B. Meloun, L. Moravek and V. K stka, Complete amino acid sequence of human serum albumin, FEBS Lett. 58:134 (1975).

7. S.-0. Olofsson, G. Bjursell, K. Bostrom, P. Carlsson, J. Elovson, A. A.

Protter, M. A. Reuben and G. Bondjers, Apolipoprotein B: structure,

biosynthesis and role in the lipoprotein assembly process, Athero­ sclerosis 68:1 (1987).

8. T. J. Knott, R. J. Pease, L. M. Powell, S. C. Wallis, S. C. Rall, Jr., T. L. Innerarity, B. Blackhart, W. H. Taylor, Y. Marcel, R. Milne, D.

Johnson, M. Fuller, A. J. Lusis, B. J. McCarthy, R. W. Mahley, B.

Levy-Wilson and J. Scott, Complete protein sequence and identifica­ tion of structural domains of human apolipoprotein B, Nature 323:734 (1986).

9. C.-Y. Yang, S.-H. Chen, S. H. Gianturco, W. A. Bradley, W.A., J. T.

Sparrow, M. Tanimura, W.-H. Li, D. W. Sparrow, H. DeLoof, M. Rosse­

neu, F.-S. Lee, Z.-W. Gu, A.M. Gotto, Jr. and L. Chan, Sequence, receptor-binding domains and internal repeats of human apolipoprotein B-100, Nature 323:738 (1986).

10. B.F. Tack, R.A. Harrison, J. Janatova, M.L. Thomas and J.W. Prahl, Evidence for presence of an internal thiolester bond in third compo­ nent of human complement, Proc. Nat!. Acad. Sci. USA 77:5764 (1980).

11.  D.M. Lee and S. Singh, Prese and localizati of o intramolecular thiolester linkages in apolipoprotein B, Circulation 76II:286 (1988).

12. W.R. Fisher, The structure of lipoproteins: Covalently bound fatty acids, Ph.D. Dissertation, University of Pennsylvania, University Microfilms, Inc., Ann Arbor, MI (1964).

13. W.R. Fisher and S. Gurin, Structure of lipoproteins. Covalently bound fatty acids. Science 143:362 (1964).

14. J.M. Hoeg, M.S. Meng, R. Ronan, S.J. Dernosky, Jr., T. Fairwell and H.B.

Brewer, Jr., Apolipoprotein B synthesized by Hep G2 cells undergoes fatty acid acylation, · Lipid Res. 29:1215 (1988).

15. J.M. Hoeg, M.S. Meng, R. Ronan, T. Fairwell and H.B. Brewer, Jr., Human apolipoprotein A-I. Post-translational modification by fatty acid acylation, J. Biol. Chern. 261:3911 (1986).

16. D.M. Lee, A.J Valente H. Kuo and H. Maeda, Properties of apolipo­ protein B in urea and in aqueous buffers. The use of glutathione and nitrogen in its solubilization, Biochim. Biophys. Acta 666:133 (1981).

17. Y.M. Torchinsky, The chemical properties of SH groups. Sulfhydryl reagents, in: "Sulfur in Proteins", Y.M. Torchinsky, ed., Pergamon Press, New-york, NY (1981).

18. J.E. Buss and B.M. Sefton, Direct identification of palmitic acid as the lipid attached to p21ras, Mol. Cell. Biol. 6:116 (1986).

19. z.-Q. Chen, L.S. Ulsh, G. DuBois d T.Y. Shih, Posttranslational pro­

cessing of p21 ras proteins involves palmitylation of the C-terminal

tetrapeptide containing Cysteine-186, · Virology 56:607 (1985).

20. B.M. Sefton, I.S. Trowbridge, J.A., Cooper and E.M. Scolnick, The

transforming proteins of Rous sarcoma virus, Harvey sarcoma virus and Abelson virus contain tightly bound lipid, Cell 31:465 (1982).

21. B.M. Willumsen, K. Norris, A.G. Papageorge, N.L. Hubbert and D.R. Lowy, Harvey murine sarcoma virus p21 ras protein: Biological significance

of the cysteine nearest the carboxy terminus, EMBO J. 3:2581 (1984).

M.F.G. Schmidt and M.J. Schlesinger, Fatty acid binding to vesicular stomatitis virus glycoprotein: a new type of posttrans1ational mod­ ification of the viral glycoprotein, Cell 17:813 (1979).

M.F.G. Schmidt and M.J. Schlesinger, Relation of fatty acid attachment to the translation and maturation of vesicular stomatitis and Sindbis virus membrane glycoproteins, · Biol. Chern. 255:3334 (1980).

M.F.G. Schmidt, M. Bracha and M.J. Schlesinger, Evidence for covalent attachment of fatty acids to Sindbis virus glycoproteins, Proc. Nat!. Acad. Sci. USA 76:1687 (1979). -- --

J.K. Rose, G.A. Adams and C.J. Gallione, The presence of cysteine in the cytoplasmic domain of the vesicular stomatitis virus glycoprotein is required for palmitate addition, Proc. Nat1. Acad. Sci. USA 81:2050 (1984). -- -- -- -- --

J.F. Kaufman, M.S. Krangel and J.L. Strominger, Cysteines in the trans­ membrane region of major histocompatibility complex antigens are fatty acylated via thioester bonds, · Biol. Chern. 259:7230 (1984).

M.B. Omary and I.S. Trowbridge, Covalent binding of fatty acid to the transferrin receptor in cultured human cells, J. Biol. Chern. 256:4713 (1981). - -- --

M. Staufenbiel and E. Lazaride, Ankyrin is fatty acid acylated in erythrocytes, Proc. Nat!. Acad. Sci. USA 83:318 (1986). 3 P.J. O'Brien and M. Zatz, Acylation of bovine rhodopsin by [ H]palmitic acid, J. Biol. Chern. 259:5054 (1984).

J.H.P. Sken d I. Virag, Posttranslational membrane attachment and dynamic fatty acylation of a neuronal growth cone protein, GAP-43, J. Cell Biol. 108:613 (1989).

S. Jing and I.S. Trowbridge, Identification of the intermolecular di­

sulfide bonds of the human transferrin receptor and its lipid-attach­

ment site, EMBO · 6:327 (1987).

M.F.G. Schmidt, Acylation of viral spike glycoproteins: A feature of

enveloped RNA viruses, Virology 116:327 (1982).

U. Klockmann and W. Deppert, Acylated simian virus 40 large T-antigen:

a new subclass associated with a detergent-resistant lamina of the plasma membrane, EMBO J. 2:1151 (1983).

B.M. Sefton and J.E. Bus;, The covalent modification of eukaryotic

proteins with lipid, J. Cell Biol. 104:1449 (1987).

J.E. Buss, P.A. Solski,-J.P. Schaeffer, M.J. MacDonald, C.J. Der, Acti­

vation of the cellular proto-oncogene produce p21 ras by addition of a myristylation signal, Science 243:1600 (1989).

A.I. Magee, L. Gutierrez, I.A. aXs C.J., Marshall and A. Hall, Dynamic fatty acylation of p21 r , EMBO J. 6:3353 (1987).

W.G. Dunphy, E. Fries, L.J. Urbani and J.E. Rothman, Early and late functions associated with the Golgi apparatus reside in distinct compartments, Proc. Nat!. Acad. Sci. USA 78:7453 (1981).

G. Ponsin, Relationship between structure and metabolism of HDL ape­ lipoproteins: Study with synthetic peptides, Adv. · Med. Biol. 243:139 (1988).

0 comments:

Post a Comment