An "i-4, i, i+4" "reductive and nucleophilic zipper" shared by both prion protein and beta-amyloid peptide sequences supports a common putative molecular mechanism

Yang Chiming (Chi Ming Yang)
(Neurochemistry Group, Department of Chemistry, Nankai University, Tian Jin 300071; Institute for Life Sci & Health, La Jolla (San Diego), California 92039-2035; College of Life Science & Biotechnology, Shanghai JiaoTong University, Shanghai 200030, China )

Received Apr. 21, 2000

Abstract To understand the protein molecular basis of genetic diseases, a hitherto undescribed "i-4, i, i+4" reductive, basic and nucleophilic zipper, XAXXXBXXXAX, evolved in natural protein sequences is investigated. A physical chemistry-based computer Algorithm allowed a protein sequence comparative study which argues that a His ^{(i - 4)} / Trp ⁽ⁱ⁾ / His ⁽ⁱ⁺⁴⁾ zipper in prion protein sequence and a His ^{(i - 4)} / Tyr ⁽ⁱ⁾ / His ⁽ⁱ⁺⁴⁾ zipper in beta-amyloid peptide sequence supports a common putative molecular mechanism in the sporadic forms of related neurodegeneative diseases. A similar "i-4, i, i+4" zipper is also identified from the active sites of a variety of copper(II)-dependent oxidases (as Trp^(i-4)/His⁽ⁱ⁾/Tyr^{( i+4)} zipper) used by aerobic bacteria for the respiration of dioxygen. It is suggested that an alignment of "i-4, i, i+4" triple or more oxidixable (reductive) and nucleophilic amino-acid residues may be crucial in the oxidative damage to both amyloid precursor protein (APP) and prion proteins in the initial stages of free-radical-based pathogenesis in Alzheimer's disease, and both sporadic and genetic forms of prion diseases. This type of zippers, H/Y/H, H/W/H, and W/H/Y, fully illustrate why octarepeats, i.e., 8 amino-acid residues in each peptide repeat, are evolved in prion proteins. The Gibbs free-energy change associated with the amyloid process may be expressed by the equation: delta G _{(amyloidosis)} = delta G₁⁰ _(mutation)+ [delta G_{2a ("i-4, i, i+4" basicity/nucleophilicity)} + delta G_{2b ("i-4, i, i+4" reductivity)} +......]
Keywords "i-4, i, i+4" "reductive, basic and nucleophilic zipper ", His (i) / Trp (i+4) or Tyr (i+4), beta-amyloid peptide, oxidative damage, genetic diseases, oxidase
Abbreviations  APP, amyloid precursor protein; AD, Alzheimer's disease; PrP, prion protein

Alzheimer's disease (AD)^[1,2] is a widespread progressive dementia happen in the elderly and its high death incidence is just next to cardiovascular disease and cancer in the developed world. The presence of non-crystalline, non-homogeneous and protease-resistant but ordered protein aggregate in the brains of patients is the predominant feature of AD^[2]. This protein aggregate is composed largely of a hydrophobic 39-43 amino acid peptide called beta-amyloid peptide^[3], the sequence of which is encoded in the amyloid precursor protein (APP) gene. Another neurodegenerative disorder, prion diseases, which occur as three types of incidences, i.e., sporadic, genetic and infectious, is characterized by the accumulation in the brain of an apparently similar type of protease-resistant and multimeric protein plaques^[4-7]. While prion disease can be transmissible by the infectious prion aggregates, AD is not transmissible^[4,5]. This is presumably because that the beta-amyloid peptide in AD is a small peptide composed of only about 42 amino-acid residues, but prion protein is a biomacromolecule with 254 amino-acid residues which may be able to confer sequence-specific biomolecular recognition ability^[4].
    Unfortunately the molecular mechanisms of these two neurodegenerative disorders have been in controversy for many decades^[4-11]. Although a free radical mechanism in AD was proposed years ago^[9,10], it was only until very recently that a sequence-specific protein radical mechanism for prion diseases had been eventually suggested from this lab^[4,12-15] and it has since received much attention duo to the disturbance of epidemic mad cow disease. Using bioinformative approaches to investigating sequence determinants of amyloidogenic proteins^[16], and the pathological and biological features of the neurodegenerative disorders, we were forced to have suggested that prion disease is initiated by both oxygen-containing free-radical damage and sequence-specific protein radicals, and disease propagation is facilitated by both sequence-specific prion radicals and reactive oxidative species in mammals^[4,12-15].
    To provide a protein molecular basis for the two neurodegenerative disorders with similar but not identical pathological features, in the present study a computational analysis of the sequences of benign primate prion protein, which are encoded in piron gene, and the sequence of amyloid peptide 1-42, which is encoded in the amyloid precursor protein (APP) gene, revealed a common "i-4, i, i+4" "reductive, basic and nucleophilic zipper", suggesting a common putative mechanism underlying the initiation stages of sporadic Alzheimer's disease and both sporadic and genetic forms of prion diseases.

1 METHODS AND RESULTS 
1.1 Thermodynamic consideration and theoretical basis of Algorithm for amyloid protein sequence analysis  
Since prion concept was arguably proposed by Prusiner^[7], protein misfolding (or amyloid formation) has been an unclear subject in protein biochemistry and biophysics. Here I use the term "amyloidosis" to describe the collective phenomenon: amyloidosis, misfolding or aggregation of a protein. Assuming that amyloid formation is resulted from both destabilized folding upon single-point mutation and microenvironment-induced protein aggregation, we have the following Gibbs energy equation, delta G _{(amyloidosis)} = delta G₁⁰ + delta G₂ = delta G₁⁰ + [ delta G_2a + delta G_2b +......] = delta G₁⁰ _(mutation)+ [delta G_{2a ("i-4, i, i+4" basicity/nucleophilicity)} + delta G_{2b ("i-4, i, i+4" reductivity)} + ......]                        (a)
where, deltaG₁⁰ is Gibbs free energy associated with destabilized folding upon single-point mutation; deltaG₂ is Gibbs free energy associated with microenviroment-associated chemistry-affected protein aggregation; deltaG_2a is Gibbs free energy associated with cooperative "i-4, i, i+4" basicity/nucleophilicity-affected protein aggregation; deltaG_2b is Gibbs free energy associated with "i-4, i, i+4" cooperative oxidative-reductive protein aggregation. Based on equation (a), a computer Algorithm is written to analyse the available amyloid protein sequences.
    Earlier reports demonstrated that the poly octarepeats in the prion protein molecules bind selectively with copper(II) ions, and it was postulated that Trp and His were the possible chelating residues and one Cu(II) ion presumably binds with two octarepeats (PHGGGWGQ) in prion proteins^[17]. Meanwhile, amyloid precursor protein (APP) also exhibits high affinity for Cu(II), which leads to reduction of Cu(II) to Cu(I), and APP undergoes a site-specific fragmentation in the reduction of hydrogen peroxide^[18,19]. However, the chemistry, especially the oxidative chemistry of this type of peptide sequences remains largely unexplored, although in a recent report it was postulated that beta-amyloid peptide directly produced hydrogen peroxide from its reaction with metal ions^[20].
    Whilst in the life-science research horizon, proteogenic amino acids histidine, tryptophan and tyrosine containing heterocyclic aromatic side-chains have been demonstrated to be highly reductive (oxidizable ) residues^[21-23]  towards reactive oxidative species, and they are basic and nucleophilic, being capable of reacting with a wide variety of electron-deficit species. The imidazol, indole, and phenol rings of histidine, tryptophan and tyrosine residues, respectively, are among the most susceptible side chains to free radical attack initiated by Fenton-type reactions. It can be suggested that the side chains of these residues may not only play a structural role in binding small ligands such as copper(II) ions, but also be prone to free radical damage by oxygen-containing radicals including the notorious hydroxyl radicals. In order to investigate the molecular basis of neurodegenerative disorders including prion diseases and AD, computer-assisted biochemistry-based comprehensive search was made for sequence similarity between APP, prion protein and other related protein sequences in nature. It can be shown both of these disease-associated proteins, APP and prion protein, maintain "i-4, i, i+4" triple or multiple reductive/nucleophilic amino-acid residues in the N-terminal domains, i.e., "Raa-X-X-X-Raa-X-X-X-Raa" motif. Here each "Raa" denotes a reductive/nucleophilic amino-acid residue such Trp, Tyr or His.
1.2 Zipper sequence of amyloid peptide A(beta)-42 and A(beta)-40   
Amyloid peptide sequence has been characterized for many years^[3,24], but its chemistry has been poorly understood. Its sequence is shown as follows:
DAEFR H^i-4DSGY ⁱ EVHH ⁱ⁺⁴Q KLVFF AEDVG SNKGA IIGLM VGGVV IA
    So far at least four genes (beta-APP mutations, apoE4 polymorphism, Presenilin 1 mutations and Presenilin 2 mutations) have been implicated in AD pathogenesis^[1,2], and accumulation of amyloid peptide plaques A(beta)-42 and A(beta)-40 has been observed in the brains of affected patients^[2,9,10]. Yet a detailed mechanism how and why the amyloid plaque is formed has never been clearly elucidated. This peptide was repeatedly demonstrated to be neuro-toxic in vitro and numerous reports provide evidence that nitrogen-containing heterocyclic aromatic compounds including nicotine are able to delay the amyloid aggregation process^[25,26]. Besides, tremendous amount of biological data from an oxidative stress theory-favoured camp supports a free radical molecular mechanism in AD pathogenesis^[27-33]. Nevertheless, since others often feel that only little in-depth insight into the free radical mechanism on a molecular basis has been available, a seemingly appealing view from protein misfolding-favoured club prefer claiming a "nucleation" or "seeding" model in interpreting the AD at a molecular level^[11,34].
1.3 Zipper sequences in prion proteins  
    The central event in prion diseases is the structural transformation of a single protein, i.e., prion protein, from its benign form to its diseased and infectious form^[4-8]. Prion gene encoding the benign prion protein (PrP) has been found from all the mammals so far examined^[4-8]. Its essentially requirement in the development of prion disease has been demonstrated by many transgenic mice experiments which showed that ablation of prion gene abolished the disease infection^[5-7]. It is revealed that the poly octarepeats in the N-terminal domain and the two hydrophobic segments in the C-terminal domain are found to be highly conserved among all the vertebrates^[5-7]. Characteristic parts of the sequence of the benign prion protein are as follows:
PQGGGGW⁵⁷GQ(PH ^i-4GGGW ⁱGQ) (PH ⁱ⁺⁴GGGWGQ)3..(A¹¹³GAAAAGAVVGGLGG Y)
......(L²³⁴FSSPPVILLISFLIFLIVG) [human numbering]
    Laboratory animal experiments indicated that polyoctapeptide repeats at the N-terminus of benign PrP play an important role in the aetiology of the diseases, since it is found that normal goats with four copies of octarepeats (PHGGGWGQ) are very susceptible to scrapie infection, but transgenic goats with 2 copies of the octarepeats are non-pathogenic. Meanwhile, expansion of this octarepeats was found to be associated with familial (genetic) human prion disease. For example, familial human prion disease patients have from 5 up to 8 extra octarepeats of (PHGGGWGQ) in their PrP sequence, compared with four octarepeats at the N-terminus encoded in prion genes for normal humans^[4].
    Other investigation provides firm evidence which indicates that His residues in scrapie PrP protein may plays a crucial role in the sequence specificity of prions infectivity, for the infectious scrapie agent is inactivated by treatment with diethylpyrocarbonate, which carboxyethylates the His residues of prion proteins. Interestingly, the infectivity of diethylpyrocarbonate-inactivated scrapie agent is restored by treatment with hydroxylamine^[35]. Besides, polyamines and nitrogen-containing compounds such as Congo red are also found to inhibit scrapie protein infection^[15,36], and copper ion dramatically affected the reversibility of scrapie prion inactivation^[37].
1.4 Zipper sequences in Cytochrome c oxidases
Cytochrome c oxidase [EC.1.9.3.1] is an important enzyme used by aerobic bacteria for the respiration of dioxygen ^[35,38,39]. This copper-dependent enzyme is located in the inner membrane of mitochondria in eucaryotes and in the plasma membrane of aerobic bacteria. It catalyzes the reduction of dioxygen. Active site of the copper-dependent enzyme cytochrome-c oxidases are shown in the following for comparison of membrane helix VI amino acid sequences withTrp 236, His240 and Tyr244 of subunits I from four cytochrome c oxidases. Bovine heart aa3-oxidase (Bh aa3); P. denitrificans aa3-oxidase (Pd aa3); T. thermophilus caa3-oxidase (Tt caa3); T. thermophilus ba3-oxidase (Tt ba3)^[40].
Bh aa3 D²²⁷PILYQHLFW^i-4 FFGHⁱ PEVYⁱ⁺⁴IL ILPGFGMISH IVTYYSGKKE PFGY²⁷⁰
Pd aa3 D²²⁷PVLYQHILW^i-4 FFGH ⁱ PEVY ⁱ⁺⁴IL ILPGFGIISH VISTFAKKPI FGY²⁷⁰
Tt caa3 D²²⁷PVLFQQFFW^i-4 FYSHⁱ PTVYⁱ⁺⁴VM LLPYLGILAE VASTFARKPL FGY²⁷⁰
Tt ba3 D²²⁷PLVARTLFW^i-4 WTGHⁱ PIVYⁱ⁺⁴FM LLPAYAIIYT ILPKQAGGKL VSDP²⁷⁰
   The –Trp-X-X-X-His-X-X-X-Tyr- motif in the sequence alignments is present in all cytochrome c and quinol oxidases of eucaryotes (mitochondria), eubacteria and archaea. It was previously postulated that this type of evolution may be to prevent the release of superoxide anion radical O^2-· , the peroxide anion O₂^2- and especially the dangerous hydroxyl radical · OH from the peroxide split^[38].

2 DISCUSSION AND OUTLOOKS
A reductive/nucleophilic zipper formed by "i-4, i, i+4" triple or more reductive amino-acid residues has been found in the essential peptide segments in both APP and prion protein sequences. Since the molecular mechanisms of both Alzheimer's disease and prion diseases have been proposed to be presumably associated with oxygen-containing free-radical damage to these proteins, the biochemical delineation of the reductive zipper further suggests and supports the conceptual idea that a common putative protein-radical chemistry-based mechanism likely underlie the initiation of Alzheimer's disease and both sporadic and genetic forms of prion diseases.
   It is known that the N-terminus of prion proteins containing poly octarepeat is highly flexible and disordered in vitro^[41], and secondary structure is formed upon its binding to copper(II) ion ^[17]. In a molten globular state of a protein, i.e., the widely believed folding intermediate towards the folded three-dimensional structure of the protein, the structural state being close to the folded state, the side chains at positions of i-4, i and i+4 shall reside in the closest spatial proximity with each other along the polypeptide chain. Therefore, the alignment of multiple i-4, i and i+4 reductive side-chains shall have been conceivably evolved by nature to be ready to bind with small ligands including the oxidative copper (II) ion (See Figure 1) and oxygen-containing free radicals. This uniformly existed "i-4, i, i+4" zipper is consistent with the well-known structural characteristic of an alpha-helical structure in both the folded and molten globular states of a polypeptide chain, whereby the number of amino-acid residues per normal alpha-helical turn is 3.6 (and 3.5 residues per turn for coiled-coil structures of coiled-coil polypeptides), and the distance between each pair of side-chains of "i, i+4" positions is about 4 x 1.5 angstroms = 6.0 angstroms. Hence, both structural and biochemical requirement for polypeptide-chain binding to small ligands are met by this unique alignment (Figure 1).

Figure 1. Molten globular state in the alignment of multiple i-4, i and i+4 reductive amino-acid residues in binding with small ligands including copper ion (II). Here Raa denotes reductive amino-acid residues.

    Of particular interesting is that an essentially similar zipper is also located at the active sites of many copper-dependent oxidases which catalyse electron-transfer-based oxygen respiration. Hence, the "i-4, i, i+4" multiple His and Trp/Tyr zipper may be a universal structural motif as a binding domain for copper ion (II). Meanwhile, it may imply that the amino-acid sequence of a protein not only encodes the three-dimensional structure of the protein, but also encodes sequence-dependent and microenvironment-associated oxidative chemistry property of the protein molecule^[42]. In addition, the "i-4, i, i+4" reductive zipper fully explain why octarepeats, i.e., 8 residues in each repeat, are evolved in mammalian prion protein sequence.
    Knowledge about oxidative protein free-radical biochemistry including tryptophan radicals in cytochrome c oxidases has been rapidly accumulating^{[11-15,20-23,27-30,43-47]}, and protein free-radical mechanism has been implicated in a wide variety of biological consequences including disease processes. Polar zipper sequences such as poly-glutamines sequence have been previously identified to be implicated in neurodegenerative Huntington's disease^[48], and poly-glutamine zipper in oligopeptide repeats - PQGGYQQYN was found to be crucial for spontaneous conformational conversion of yeast protein Sup35^[49], although the "i, i+4" Gln zipper (non-reductive zipper) may imply a different molecular mechanism in "yeast prion". In contrast, the reductive zipper "i-4, i, i+4" discussed here shall greatly facilitate better understanding into both the biochemical and biophysical aspects of protein participation in biological events, especially the molecular basis of genetic diseases in the brains of mammals which require oxygen. Recall that physiological function of a benign prion protein in mammals still remains unknown^[50,51], here the precise delineation of a distinct "i-4, i, i+4" reductive zipper shared by both prion protein and copper(II)-dependent oxidases may shed new light on future effort in elucidating what special task the prion protein molecule does. It is anticipated that a recognition of distinctive sequence of amyloid proteins shall provide guidelines for applying protein engineering and genetic engineering technique in the development of pharmaceutically valuable protein molecules in the treatment of related neurodegenerative disorders.

Acknowledgements   This work is supported by MOST of China and Nankai University. I thank my student Li Wei for assistance.

REFERENCES 
[1] Selkoe D J. Science, 1997, 275: 630-631.
[2] Hardy J, Gwinn-Hardy K. Science, 1998, 282: 1075-1079.
[3] Glenner G G, Wong C W. Biochem. Biophys. Res. Commun., 1984, 122: 1131-1135.
[4] Yang C M. Chemistry Online (Huaxue Tongbao), 1999, (13): 86. (http://www.chemistrymag.org/col/1999/c99086.htm)
[5] Chesebro B. Science, 1998, 279: 42-43.
[6] Weissmann C. FEBS Lett., 1996, 389: 3-11.
[7] Prusiner S B. Proc. Natl. Acad. Sci. U.S.A., 1998, 95: 13363-13383.
[8] Yang C M, Chen Y. Chinese Science Bulletin, 2000, 45: 285-289.
[9] Butterfield D A. Chemical Research in Toxicology, 1997, 10: 495-506.
[10] Naiki H, Hasegawa K, Yamaguchi I et al. Biochemistry, 1998, 37: 17882-17889.
[11] Jarrett J T, Lansbury P T. Cell, 1993, 73: 1053-1058.
[12] Yang C M. Chemical J. Internet, 1999, 1 (1), 6. (http://www.chemistrymag.org/cji/1999/011006le.htm)
[13] Ye X H. Science and Technology Daily (Keji Ribao), 1999, Oct. 21.
[14] Yang C M, Chen Y. Chemistry (Huaxue Tongbao), 2000, 63 (1), 60­63.
[15] Yang C M. Chinese Science Bulletin, 2000, 45: 1546-1554.
[16] Yang C M. Thirteenth Symposium of the Protein Society, Boston, July 24-28, 1999 and the Third European Symposium of the Protein Society, Sep.19-22, 1999, Garmisch-Partenkirchen, Germany.
[17] Stöckel J, Safar J, Wallace A C et al. Biochemistry, 1998, 37: 7185-7193.
[18] Multhaup G, Schlicksupp A, Hesse L et al. Science, 1996, 271: 1406-1409.
[19] Multhaup G, Ruppert T, Schlicksupp A et al. Biochemistry, 1998, 37: 7224-7230.
[20] Huang X, Atwood C, Hartshorn M A et al. Biochemistry, 1999, May 27 ASAP web release.
[21] Davies K J A, Delsignore M E, Lin S W. J. Biol. Chem., 1987, 262: 9902-9907.
[22] Stadtman E R. Science, 1992, 257: 1220-1224.
[23] Stubbe J. Proc. Natl. Acad. Sci. U.S.A., 1998, 95: 2723.
[24] Selkoe D J. J. Biol. Chem., 1996, 271: 18295-18298.
[25] Salomon A R, Marcinowski K J, Friedland R P, Zagorski M G. Biochemistry, 1996, 35: 13568-13578.
[26] Lorenzo A, Yankner B A. Proc. Natl. Acad. Sci. U.S.A., 1994, 91: 12243-12247.
[27] Butterfield D A, Howard B J, Yatin S et al. Proc. Natl. Acad. Sci. U.S.A., 1997, 94: 674-678.
[28] Manelli A M, Puttfarcken P S. Brain Res. Bull., 1995, 38: 569-576.
[29] Subramaniam R, Howard B J, Hensley K et al. Alzheimer's Res., 1995, 1: 141-144.
[30] Sayre L M, Zagorski M G, Surewicz W K et al. Chemical Research in Toxicology, 1997, 10: 518-526.
[31] Butterfield D A. Chemical Research in Toxicology, 1997, 10: 495-506.
[32] Pike C J, Burdick D, Walencewicz A J et al. J. Neurosci., 1993, 13: 1676-1687.
[33] Esler W P, Stimson E R, Ghilardi J R et al. Biochemistry, 1996, 35: 749-757.
[34] Costa P R, Kocisko D A, Sun B Q et al. J. Am. Chem. Soc., 1997, 119: 10487-10493.
[35] Voet D, Voet J G. Biochemistry. John Wiley & Son, Inc., 1995, 1026-1028.
[36] Supattapone S, Nguyen H O B, Cohen F E et al. Proc. Natl. Acad. Sci. USA, 1999, 96: 14529-14534.
[37] McKenzie D, Bartz J, Mirwald J et al. J. Biol. Chem., 1998, 273: 25545-25547.
[38] Ferguson-Miller S, Babcock G T. Chem. Rev., 1996, 96: 2889-2907.
[39] Musser S M, Chan S I. Mol. Evol., 1998, 46: 508-520.
[40] Buse G, Soulimane T, Dewor M et al. Protein Science, 1999, 8: 985-990.
[41] Donne D G, Viles J H, Groth D et al. Proc. Natl. Acad. Sci. USA., 1997, 94: 13452-13457.
[42] Yang C M, Protein chemical biology in prion diseases and BSE, invited colloquium on the opening ceremony of Bio-X center in the Shanghai Jiaotong University, 2000, China.
[43] Smith C D, Carney J M, Tatsumo T et al. Ann. N. Y. Acad. Sci., 1992, 663: 110-119.
[44] Starke-Reed P E, Oliver C N. Arch. Biochem. Biophys., 1989, 275: 559-567.
[45] Smith C D, Carney J M, Starke-Reed P E et al. Proc. Natl. Acad. Sci. U.S.A., 1991, 88: 10540-10543.
[46] Rigby S E J, Jünemann S, Rich P R, Heathcote P. Biochemistry, 2000, ASAP Article
[47] Goodin E B. J. Phy. Chem. B, 1998, 102: 8221-8228.
[48] Perutz M F, Johnson T, Suzuki M, Finch J T. Proc. Natl. Acad. Sci. U.S.A., 1994, 91: 5355-5358.
[49] Liu J J, Lindquist S. Nature, 1999, 400: 573-576.
[50] Brown D R, Besinger A. Biochem. J., 1998, 334: 423-429.
[51] Glockshuber R, Hornemann S, Billeter M et al. FEBS Letters, 1998, 426: 291-296.

[ Back ] [ Home ] [ Up ] [ Next ] Mirror Site in USA  Europe  China  CSTNet ChinaNet