An anti-prion protein?
Anticipating the anti-prion protein
Marsupial prion gene: no ORF for anti-prion
Bovine prion gene: amino acid and nucleotide sequence
Bovine prion gene: anti-sense amino acid and nucleotide sequence
Results of search for homology

An anti-prion protein?

Nature Letter 362: 213-2141993
Markus Moser, Bruno Oesch, Hansruedl Bueler Institut fur Hirnforschung Universitat Zurich, S029 Zurich, Switzerland

The physiological function of the prion protein (PrP) is unknown, but there is strong evidence that the infec ious agent of transmissible spongiform encephalopathies (the prion) consists at least in part of PrPsC, a mod)fied form of the normally occurring cellular PrP (PrPC)~. Both forms of PrP are encoded in a single exon of the PrP gene.

Goldgaber in his Scientific Correspondence noted a large open reading frame (ORF) on the antisense strand of the coding regions of mammalian PrP genes. Several genetic traits influencing spongiform encephalopathies in humans and animals have been linked to the PrP gene. If an 'anti-PrP' existed it could be responsible for the properties formerly ascribed to the prion protein. In support of this hypothesis, Hewinson and colleagues detected a 4.5-kilobase RNA in bovine brain which hybridized to PrP sense riboprobes, and which was thus expected to represent a PrP antisense transcript.

We have found a similar 4.5-kb mRNA in normal and scrapie infected hamsters and in mice; however, our results show that this presumed 'anti-prior protein' RNA is not derived from the antisense strand of the PrP gene. We detected a 4.5-kb 'antisense' RNA in brains of normal and scrapie infected hamsters at approximately the same level, as judged by northern analysis. The 4.5-kb RNA was polyadenylated and was not found in tissues other than brain. The fact that the 4.5-kb RNA was detected under stringent hybridization conditions initially suggested that the 'anti-PrP' RNA was transcribed from the opposite strand of the hamster PrP gene.

If indeed the 4.5-kb RNA were an antisense transcript of the PrP gene, then it should be absent or at least altered in mice homozygous for partially deleted PrP genes (Prn-p0~0)6. Using the above hamster probe A, which was shown to cross-hybridize to 'anti-PrP RNA from wild-type mouse brain, we detected the 4.5-kb RNA also in mice with Prp genes . Because probe A extended beyond the partial deletion in the PrP gene of PrP mice, we used an additional probe which lies entirely with in the deleted region. It de tected the same 4.5-kb RNA in both types of mice . These results unequivocally show that the 4.5-kb RNA is not derived from the opposite strand of the murine PrP gene; however, it appears to share considerable homology with the antisense strand of the PrP gene.

Our results confirm the presence of the 4.5-kb 'anti-PrP' RNA in two additional species, show that its concentration is not greatly altered after scrapie infection, and demonstrate that the RNA is not derived from the PrP gene. If the 4.5-kb mRNA, despite its being derived from a locus other than the PrP gene, encoded an 'anti-prior protein', this might have intriguing implications. It has recently been described for four independent proteins that peptides encoded by antisense sequences correspond structurally to binding proteins or receptors. Thus, a putative 'anti-prion protein' may be pertinent to the normal function of the prion protein.

1. Prusiner, s. s. Science ,
2. Oesch, s.(1985).
3 Basler, K etal. Cell46. 417
4: Goldgaber, D. Nature 351, 106 (1991)
5 Hewinson, R. G., et al Nature 352. 291 (1991).
6 Bueler, H. et al. Narure 356 (1992).
7. Ghiso, J., Saball, E,, Leoni, i., Rostagno, A &
Frangione, s, Proc.Nat'l Acad. Sci. U.S.A. 87,128
8.Srentani, R R er al Proc. natn. Acad. Sci. U.S.A 615,36~367 (1988)
9. Elton, T. S., Dion, L D30st, L, Oparil, s. &Proc. Nat'l. Acad. Sci. U. S. A 8s, 251~2522 (1988).
10. Bost, L, Smith, E. ; slal0ck. Proc. Nat'l.Acad. Sci. U S.A 82, 1372 1375 (1985)


Anticipating the anti-prion protein

Nature 351 106 1991 ...Goldgaber D

A host-encoded prion protein (PrP) is crucial in the pathogenesis of transmissible spongiform encephalopathies, such as scrapie, produce changes in complementary scrapie and scrapie-like diseases in man and PrP and anti-PrP mRNAs. The PrP gene is highly conserved in protein evolution and contains a single open reading. Several mutations in the human homologue prospective encoded protein, especially in PRNP gene were found in patients with view of its potential role in transmissible familial Gerstmann-Straussler-Scheinker's syndrome (GSS) and Creutzfeld-Jakob disease (CJD). While analysing PrP complementary DNA sequences with the DNA Strider computer program, I found a large overlapping ORF in the DNA strand opposite to the PrP transcriptional unit. The deduced amino acid sequence of the prospective encoded protein was unique. Its hydropathy plot is almost a mirror image of that of the PrP. For simplicity the name of anti-PrP is used in reference to this putative protein.

This ORF may be of biological importance first, because it is as large as the PrP ORF. Second, none of the differences in PrP sequences found in patients or animals resulted in additional stop codons in this ORF. It should be noted, however, that mutations found in GSS patients in codons 102 and 117 of the PRNP gene produced amino-acid changes in the anti-PrP, whereas mutations found in CJD patients in codon 178 and 200 did not. Third, there are ATG or CTG codons at the beginning of this ORF which could serve as translation initiation codons. Fourth, existing data on PrP gene expression do not exclude expression of the anti-PrP gene because only double-stranded DNA probes were used to detect PrP messenger RNA.

Obviously, only direct experimental analysis of infected and uninfected cells and tissues with single-stranded probes will show whether the anti-PrP gene is expressed.The expression of the anti-PrP gene would introduce a new player, present a different view of scrapie infection, and raise new and important questions. Do PrP and anti-PrP proteins interact in infected cells? Do RNA unwindases capable of modifying double-stranded RNA... Do GSS mutations, which, unlike CJD mutations, result in amino-acid changes in both proteins, explain the differences between the two? In conclusion, the existence of a long anti frame (ORF) located within a large exon. PrP ORF clearly warrants a search for the

1 Prusiner s. B. A Rev Microbiol 43 345-374 (1989).
2 . Marck c. Nucleic Acids Res. 18 1829 - 1836 (1988).
3 Kyte J. ; Doolittle R. F. J. molec. Bio 157 105-132(1982).
4 Hsiao c.  Science250, 1587-1590 (1990).
5 Doh-ura K. Biochem. biophys Res. Commun.163 974-979 (1989).
6.  Goldfarb L. G. Lancet t337 425 (1991).
7. Goldgaber D. Exp Neuro 106 204-206 ( 1990).
8. Prats Proc.natn. Acad sci. U.S.A. 86 1836-1840 (198g).
9. Srentani R. R. J. theor Bio 135 495 - 499 (1988).
10 Lamb. R. A, Creyfuss G. Nature 337 19 20 ( 1989).

Marsupial prion gene

Windl O; Dempster M; Estibeiro P; Lathe R
Centre for Genome Research, University of Edinburgh, UK.
Gene, 159: 2, 1995 Jul 4, 181-6

The normal function of the pathogenicity-related protein, PrP (or prion protein), is unknown. We have characterized a candidate marsupial PrP gene.A particular feature of the marsupial[and mink] gene is the lack of a continuous ORF on the antisense strand, as is found in most eutherian PrP. We propose that antisense ORFs found in other species are artefactual.

[The authors argue in the body of the paper that 5% of eukcaryotic genes have these open reading frames in the anti-gene, that this is a consequence of low useage of anti-codons to the stop codons to speed synthesis, that the hydrophobicity reversal is a simple property of the genetic code, and that all known long ORFs are in phase. -- webmaster]


Bovine prion gene: amino acid and nucleotide sequence

Here are the sequences for bovine prion (265 aa; 795 bps).
Note a minor polymorphism: nucleotide 576 exhibits C to T transition, creating a HindII polymorphism. Holstein-Friesian
Accession Numbers: SWISS-PROT: P10279" ... PID:g684
MVKSHIGSWILVLFVAMWSDVGLCKKRPKPGGGWNTGGSRYPGQ
GSPGGNRYPPQGGGGWGQPHGGGWGQPHGGGWGQPHGGGWGQPHGGGWGQPHGGGGWG
QGGTHGQWNKPSKPKTNMKHVAGAAAAGAVVGGLGGYMLGSAMSRPLIHFGSDYEDRY
YRENMHRYPNQVYYRPVDQYSNQNNFVHDCVNITVKEHTVTTTTKGENFTETDIKMME
RVVEQMCITQYQRESQAYYQRGASVILFSSPPVILLISFLIFLIVG

Base composition      197 a    193 c    244 g    161 t
ORIGIN      
        1 atggtgaaaa gccacatagg cagttggatc ctggttctct ttgtggccat gtggagtgac
       61 gtgggcctct gcaagaagcg accaaaacct ggaggaggat ggaacactgg ggggagccga
      121 tacccaggac agggcagtcc tggaggcaac cgttatccac ctcagggagg gggtggctgg
      181 ggtcagcccc atggaggtgg ctggggccag cctcatggag gtggctgggg ccagcctcat
      241 ggaggtggct ggggtcagcc ccatggtggt ggctggggac agccacatgg tggtggaggc
      301 tggggtcaag gtggtaccca cggtcaatgg aacaaaccca gtaagccaaa aaccaacatg
      361 aagcatgtgg caggagctgc tgcagctgga gcagtggtag ggggccttgg tggctacatg
      421 ctgggaagtg ccatgagcag gcctcttata cattttggca gtgactatga ggaccgttac
      481 tatcgtgaaa acatgcaccg ttaccccaac caagtgtact acaggccagt ggatcagtat
      541 agtaaccaga acaactttgt gcatgactgt gtcaacatca cagtcaagga acacacagtc
      601 accaccacca ccaaggggga gaacttcacc gaaactgaca tcaagatgat ggagcgagtg
      661 gtggagcaaa tgtgcattac ccagtaccag agagaatccc aggcttatta ccaacgaggg
      721 gcaagtgtga tcctcttctc ttcccctcct gtgatcctcc tcatctcttt cctcattttt
      781 ctcatagtag gatag

Reversed, in-phase bovine prion gene, sense strand 3' to 5'

gataggatgatactctttttactcctttctctactcctcctagtgtcctccccttctcttctcctagtgtgaacggggagcaaccattattcggaccctaa gagagaccatgacccattacgtgtaaacgaggtggtgagcgaggtagtagaactacagtcaaagccacttcaagagggggaaccaccaccaccactg acacacaaggaactgacactacaactgtgtcagtacgtgtttcaacaagaccaatgatatgactaggtgaccggacatcatgtgaaccaaccccattgc cacgtacaaaagtgctatcattgccaggagtatcagtgacggttttacatattctccggacgagtaccgtgaagggtcgtacatcggtggttccggggga tggtgacgaggtcgacgtcgtcgaggacggtgtacgaagtacaaccaaaaaccgaatgacccaaacaaggtaactggcacccatggtggaactggggt cggaggtggtggtacaccgacaggggtcggtggtggtaccccgactggggtcggtggaggtactccgaccggggtcggtggaggtactccgaccgggg tcggtggaggtaccccgactggggtcggtgggggagggactccacctattgccaacggaggtcctgacgggacaggacccatagccgaggggggtcac aaggtaggaggaggtccaaaaccagcgaagaacgtctccgggtgcagtgaggtgtaccggtgtttctcttggtcctaggttgacggatacaccgaaaagtggta


The anti-sense strand in 5' to 3' direction:

ctatcctactatgagaaaaatgaggaaagagatgaggaggatcacaggaggggaagagaagaggatcacacttgcccctcgttggtaataagcctgggattc tctctggtactgggtaatgcacatttgctccaccactcgctccatcatcttgatgtcagtttcggtgaagttctcccccttggtggtggtggtgactgtgtgttcc ttgactgtgatgttgacacagtcatgcacaaagttgttctggttactatactgatccactggcctgtagtacacttggttggggtaacggtgcatgttttcacgat agtaacggtcctcatagtcactgccaaaatgtataagaggcctgctcatggcacttcccagcatgtagccaccaaggccccctaccactgctccagctgcagc agctcctgccacatgcttcatgttggtttttggcttactgggtttgttccattgaccgtgggtaccaccttgaccccagcctccaccaccatgtggctgtccccag ccaccaccatggggctgaccccagccacctccatgaggctggccccagccacctccatgaggctggccccagccacctccatggggctgaccccagccaccc cctccctgaggtggataacggttgcctccaggactgccctgtcctgggtatcggctccccccagtgttccatcctcctccaggttttggtcgcttcttgcagaggc ccacgtcactccacatggccacaaagagaaccaggatccaactgcctatgtggcttttcaccat

Translation of anti-sense strand: note total absence of stop codons:

  LSYYEKNEERDEEDHRRGREEDHTCPSLVISLGFSLVLGNAHLLHHSLHHLDVSFGEVLPLGGGGDCVFL
  DCDVDTVMHKVVLVTILIHWPVVHLVGVTVHVFTIVTVLIVTAKMYKRPAHGTSQHVATKAPYHCSSCSS
  SCHMLHVGFWLTGFVPLTVGTTLTPASTTMWLSPATTMGLTPATSMRLAPATSMRLAPATSMGLTPATPS
  LRWITVASRTALSWVSAPPSVPSSSRFWSLLAEAHVTPHGHKENQDPTAYVAFHH

5'3' Frame 2: sense strand, frameshifted by one.

W-KAT-AVGSWFSLWPCGVTWASARSDQNLEEDGTLGGADTQDRAVLEATVIHLREGVAGVSPMEVAGAS LMEVAGASLMEVAGVSPMVVAGDSHMVVEAGVKVVPTVNGTNPVSQKPT-SMWQELLQLEQW-GALVATC WEVP-AGLLYILAVTMRTVTIVKTCTVTPTKCTTGQWISIVTRTTLCMTVSTSQSRNTQSPPPPRGRTSP KLTSR-WSEWWSKCALPSTRENPRLITNEGQV-SSSLPLL-SSSSLSSFFS--D

5'3' Frame 3: sense strand, frameshifted by two

GEKPHRQLDPGSLCGHVE-RGPLQEATKTWRRMEHWGEPIPRTGQSWRQPLSTSGRGWLGSAPWRWLGPA SWRWLGPASWRWLGSAPWWWLGTATWWWRLGSRWYPRSMEQTQ-AKNQHEACGRSCCSWSSGRGPWWLHA GKCHEQASYTFWQ-L-GPLLS-KHAPLPQPSVLQASGSV--PEQLCA-LCQHHSQGTHSHHHHQGGELHR N-HQDDGASGGANVHYPVPERIPGLLPTRGKCDPLLFPSCDPPHLFPHFSHSRI


3'5' Frame 2: anti-sense strand, frameshifted by one.

YPTMRKMRKEMRRITGGEEKRITLAPRW--AWDSLWYWVMHICSTTRSIILMSVSVKFSPLVVVVTVCSL TVMLTQSCTKLFWLLY-STGL-YTWLG-RCMFSR--RSS-SLPKCIRGLLMALPSM-PPRPPTTAPAAAA PATCFMLVFGLLGLFH-PWVPP-PQPPPPCGCPQPPPWG-PQPPP-GWPQPPP-GWPQPPPWG-PQPPPP -GG-RLPPGLPCPGYRLPPVFHPPPGFGRFLQRPTSLHMATKRTRIQLPMWLFT

3'5' Frame 3: anti-sense strand, frameshifted by two

ILL-EK-GKR-GGSQEGKRRGSHLPLVGNKPGILSGTG-CTFAPPLAPSS-CQFR-SSPPWWWW-LCVP- L-C-HSHAQSCSGYYTDPLACSTLGWGNGACFHDSNGPHSHCQNV-EACSWHFPACSHQGPLPLLQLQQL LPHASCWFLAYWVCSIDRGYHLDPSLHHHVAVPSHHHGADPSHLHEAGPSHLHEAGPSHLHGADPSHPLP EVDNGCLQDCPVLGIGSPQCSILLQVLVASCRGPRHSTWPQREPGSNCLCGFSP


Results of search for homology with Blastx

This amazing free and fast online search tool compares a submitted nucleotide sequence, here bovine prion, to a huge database of known protein sequences ( 184,520 sequences; 57,747,617 total letters, updated daily) in all six reading frames. This searches for holmology in the sense strant, the anti-sense strand, and various frame-shifts. No significant homology to any known sequence was found, leaving the origin of the prion gene family as mysterious as ever. Puzzling, because prions are quite conservative -- it is easly to align bovine with marsupial, lines that diverged 135,00,000 years ago. Chickens have a hexapeptide repeat instead of an octapeptide but align fairly well later. Going back farther, nothing homologous was found in nematode, fruit fly, or yeast, despite some false alarms. Better tools such as hidden markov matrices might find homology in the 'twilight zone' of sub-25% alignments.

Taken at face value, this leaves prions emerging more or less with the appearance of the vertebrates and not being part of any protein super-family. Normally, proteins appear from diverging gene duplications or domain assembly, but if so here, the orginal gene cannot yet be found. So we are left without the usual clues as to the biological function of normal prion protein or its domain.