Prion Deletion Mutants
Mad Cow Home ... Best Links ... Search this site

Prion repeat region deletions: introduction
Distinguishing deletions
Frequency of deletion classes: worldwide summary
Slippage mechanism of deletions
Phylgeny of deletion slippage
Acknowledgements and References
Supplemental material:
...deletions in other species
...phylogeny of hairpin C
...detection of repeats with streak tool
...wildtype human reference sequence

Prion repeat region deletions

Draft 1.17.97 ... Rev. 8.26.97 .... Last Rev 25 Aug 98 ... webmaster resource
Numerous deletions and insertions of octapeptide repeat units have been found in the human prion gene. These can occur as deleterious mutations causing CJD or as seemingly neutral polymorphisms.

A rather striking assemblege of human deletions and insertions -- always comprised of whole 24 base pair modules in the repeat region-- has accumulated from the screening of large numbers of CJD victims, their families, assorted neurological patients, and healthy controls for mutations in the prion gene. Similar events occur in other species, notably bovine.

DNA slippage during replication, not recombination, is proposed as the generating mechanism for these events, predicting for any species the allowed insertions and deletions while forbidding others. Any explanatory mechanism must address the issues of modularity, end points, mutation class frequency, microheterogenity, and non-observed but seemingly plausible events. DNA slippage is not a novel mechanism: in another gene, the mechanism was described as:

"... the mutation arose by slipped strand mispairing, creating a single-stranded loop, followed by DNA elongation, strand breathing and the formation of a mismatch bubble. An extensive literature search has revealed six additional deletion/insertion mutations in humans in which the inserted nucleotides come from the same DNA strand. Our model explains all six mutations, suggesting that rearrangement of a mismatch loop or bubble during DNA replication may be not uncommon."
Asymmetry between insertions and deletions arises from unequal roles of the template and nascent DNA strand in replication. Modularity arises from slipped hybridization of a DNA strand from a later repeat to that of an earlier one, enabled by tandem repeated sequences. The imperfect fidelity of some insertional repeats arises not from polymerase copy error or acceptance of a wobble pair, but from incomplete 3' exonuclease editing.

The beginning of the repeat region has a region of conserved secondary structure at the level of single-stranded nucleic acid, denoted as hairpin C. While the role is speculative (targeting or stability of mRNA), as a secondary structure helix C may delay the replication fork, providing the initial impetus to slippage.

 AC CGC TAC CCA CCT CAG GGC GGT GGT -- human helix C; first 5 codons of Ra underlined 
 N   R   Y   P   P   Q   G   G   G

Distinguishing among deletions

Last revised 18 Aug 98 ... webmaster
The wildtype human prion repeat region consists of a nonamer R1 followed by four similar octamers, R2, R2, R3, and R4. The basic repeat unit is PHGGGWGQ. The second and third repeats are indistinguisable at the DNA level. While the basic unit of genetic change is a repeat unit, not a single nucleotide, point mutations sometimes accompany deletion events: a modified repeat unit, denoted R3g [called R2c in ref 25], occurs rarely in deletions.

It is not possible to deduce the precise beginning and end points of a deletion. Any deletion of 24 consecutive bases within an ambiguity zone produces the same final sequence; deletions need not be in registration with repeat boundaries. Observed deletions can be clustered into classes based on microheterogeneity of repeat DNA; the end points cannot be further distinguised. The result at the protein level is always R1234. At the DNA level, the final product is a fusion or chimera of bases initially separated by 24 bp.

As an example, deletion of 24 bp anywhere within the residues underlined in the first line of the wildtype sequence below has exactly the same outcome. The deletion class is called R2 because it results in R1234, ie, has one less R2 than wildtype R12234.

R1-------------------------------R1 R2---------------------------R2 R2---------------------------R2 R3-
pro gln gly gly gly gly trp gly gln pro his gly gly gly trp gly gln pro his gly gly gly trp gly gln pro

51  52  53  54  55  56  57  58  59  60  61  62  63  64  65  66   67 68  69  70  71  72  73  74  75  76 
                                                                      C in R3g micro-variant

R3---------------------------R3 R4---------------------------R4 Rx----Rx  
pro his gly gly gly trp gly gln pro his gly gly gly trp gly gln gly gly gly

76  77  78  79  80  81  82  83  84  85  86  87  88  89  90  91  92  93  94
   ^ NcoI restriction site.  Variant residues in bold are critical in defining deletion classes.
The 5 known deletion classes are defined in the table below:

TypeOutcomeDeletion zoneWidthMicro#Pedigrees
R1 2 2  3 4
511-913 123 bp - -
del R2
R1 2    3 4
541-762 68 bp - 9
del R23
R1 23   3 4
691-822 41 bp - 1
del R3
R1 2 2    4
751-842 29 bp - 1
del R34
R1 2 3   34
771-902 41 bp - 41
del R3g34
R1 2 3g  34
771-902 41 bp R3g 2

Legend: The second line of the table is read, "the repeat deletion type 'del R2' has outcome R1234 with 24 bases deleted anywhere within the region beginning at the first base of codon 51 and ending at the second base of residue 76, a stretch of 68 bp. No irregular bases are associated with this deletion type. It has been found in 9 unrelated lineages; see citations in the pedigree table."

R23 is a fusion of the 5' end of R2 with the 3' end of R3; similarly for R34. Fusions are detected by small changes at third codon position [in bold type in the wildtype sequence above]. Ambiguity zones were first clearly articulated by Palmer (5) for R2, R23, and R34. Some authors only determined approximate positions of deletions relative to the Ncol restriction site at codon 76.]

Observed and predicted deletion classes (based on internal complementarity and slippage) are shown using nucleotide numbering:

Deletion classes and frequency: worldwide summary

Based on published data and correspondence to 18 August 98

Frequency reported for various deletions
CitationDeletionPedigreeswt allelesCommentary
Palmer (5)del R21737N Euro case
Perry (10) del R21115Alabama FAD, parent R2/R34, S80743
Pocchiari (6) del R21127opposite allele, V210I central Italian family
Cervenakova (25,26)del R25695control cases
Yamada (24)del R21100Japanese family, trans P105L
Vnecak-Jones (12)del R2?1115control, not R34
Masullo (13) del R2?129Italian, adopted, homozygous deletion
Palmer (5)del R231737only known occurence, N. Euro control
Cervenakova (25)del R32695familial AD, familial dementia-ataxia
Vnecak-Jones (12, 4)del R346115mid-Tennessee, one in E200K
Puckett (1)del R34230HeLa cell line and CDNA library, X83416
Laplanche (2, 5,22)del R3425Moroccan; Tunisian E200K family
Bosque (4, 11, 12)del R34 19Tennessee family with D178N
Palmer (5, 22 )del R345737N. Euro cases
Diedrich (3)del R3415 dementia cases; M81929
Perry (10)del R345115Alabama FAD + PD family, controls, S80732
Salvatore (7)del R342215downstream of NCol in controls
Brown (15)del R342135found among 26 iatrogenic cases, 110 controls
Brown (unpub)del R346[200]unpublished: kuru, 5 families with ataxic illness
Cervenakova (25,26)del R3413695Libyan, speech, Italian MJD, African, HGH, conjugal, PS1
Pocchiari (6) del R342127Italian control families
Cervenakova (25)del R3g342695African-Am (one R34/R3g34), ULL
Windl (16)none0120UK CJD referals, allele-specific hybridization
Owen (23)none0101methods could have detected deletions
Goldfarb (20)none0[535]methods focused on insertions, 10 deletions expected
Totalsall dels6325172.5% of the populations studied had some deletion*

* Care was taken not to double-count controls. This is an upper limit because studies screened mainly neurological patients whose deletion incidence may not be representataive. Note that the same distant mutational event may show up in apparently unrelated families due to adoption, extra-marital affairs, immigration, etc., causing the number of pedigrees to be over-estimated; for example there are 3,000 individuals in the Indiana kindred F198S. This would be somewhat offset by unsuspected kindreds within the controls. Studies used widely varying methods to detect and report deletions -- the three studies at the bottom of the table reporting no deletions in 756 cases are treated here as having ascertainment problems and are not included: they would drop deletion incidence to 1.9% , yet even this requires about 14 deletions in these 756 cases.

Frequency and proportions of deletions in Japanese populations may differ, as they do at codon 129 and 232. One deletion was found in 101 unrelated individuals, not enough data to reliably conclude that the frequency is lower.

Using the data at face value, 6.3 people per 10,000 would carry deletions on both chromosomes. Given the rarity of deletions and supposing neutrality, is it surprising to find 3 lineages in the collection with deletions on both alleles? (In 2 cases these are distinct types.) Not dramatically so: 1.6 are expected. This gives no particular support to an auxillary role for double deletions in mental disorders or CJD; single deletion data does not resolve this issue either.

While all deletions are equivalent in the sense of resulting in the same amino acid sequence, they might differ significantly in secondary structure of mRNA and thus in stability or efficiency of translation, ie, some could be over-producers. Perry (10, pg 16) notes a proximal region of the repeat region that 'sequences poorly' on the anti-sense strand, suggesting a hairpin. Thus it is premature to lump the five deletion classes in assessing their role in CJD and other dementias. (Within each class, despite ambiguity in end points, pooling is warranted. Favored end points may exist but they cannot be detected.)

Del R34 is the only deletion or insertion to attain the status of common polymorphism, with 47 of 63 (75%) observed deletions in this class for an incidence by itself of 1.9%. Curiously, 5 out of 7 deletions in other species are also del R34. The possible explanations are (1) that del R34 is a relative hot spot created over and over in independent events because of an inherent tendency of the replication fork to slip here, (2) that the allele has become common through neutral drift, founder effect, or some unknown selective bias. [There is not a single deletion pedigree in which the founder event can be dated.] The occurence of similar deletions in other species, the unusual long insertions, and similar patterns in unrelated genes with repeats argue for an intrinsic instability in such DNA.

Many authors model deletions by unequal recombination but in no case consider the reciprocal recombinant, an insertion. In the case of del R34, the reciprocal recombinant is R1R2R3R43R4 (using the scheme of Palmer (a NAME="5">5, pg.543). Only one insertion of a single repeat is known, R122234 even though ascertainment should be similar. Note 63 deletion pedigrees imply 63 single-insertion pedigrees whereas 1 is observed -- the imbalance is statistically most implausible.

Double deletions, eg R124, have not been seen; they might be exceedingly rare or fatal in early development (so observationally suppressed). They are predicted both by slippage and recombination models. Such mutations, unlike single deletions, could result in loss of structure/function with retention of regulation gene expression and be lethal dominants. Double deletions have regretably not been studied in transgenic animals.

The deletion data in summary:

R2  R23  R3  R34  R3g34   deletion class
11   1    2   47      2   number of distinct pedigrees = 63
A new German study found 0.9% of 578) patients (5 cases had deletion of one repeat; these were not further characterized.Windl O et al. "Molecular genetics of human prion diseases in Germany, Human Genetics, 105:244-252 1999.

Mechanism of deletion slippage

under development -- webmaster
Deletion end points in the slippage model vary according to the extent of 3' exonuclease excission during mismatch correction: the model suggests that 5' to 3' synthesis renews as soon as the last mismatch is corrected. This creates a hybrid repeat as end product: the 5' proximal part of the earlier repeat fused with the 3' end of the later repeat. The ambiguity zones for the three observed classes of human single repeat deletions, along with model-prefered end points are shown in Fig. 4 below.

Parsimonious application of the model to mammalian sequence data allows reconstruction of the repeat region sequence at earlier phylogenetic branch nodes. A terminal nonamer in the ferungulate lineage suggests that long-repeat bovine prion has arisen recently from an internal single insertion event. The repeat region as a whole may have been generated by ancient expansions of an upstream domain, illuminating the relationship of marsupial and chicken repeat to eutherians. These results sharpen the prion probe sequence for distant BLAST II homology searches for the currently orphaned prion gene.

Similar events occur in other species, notably bovine. The prion gene from the 80-odd mammals sequenced to date contains a repeat region highly similar in structure to that of humans. At the protein level, the consensus sequence is a direct tandem repeat consisting of a nonapeptide followed by three octapepetides, denoted here Ra, Rb, Rc, and Rd, with typical repeat element, PHGGGWGQ, that can be tracked back to the metatherian-eutherian divergence. Strong sequence conservation over a long period suggests a significant function; however, no role has yet been assigned to this domain nor to the prion protein as a whole. The prion gene is an orphan, having no known homologues earlier than teleosts.(salmon )

In fact, the role for this domain may reside mainly in nucleic acid rather than protein. The early part of the repeat domain is associated with an evolutionarily stable helix in mRNA called helix C. In other genes expressed in brain, mRNA structure has been suggested to target the site of translation or regulate its frequenccy. The repeat domain is easily cleaved by proteolysis and is seldom purified with intact protein; theoretical secondary structure prediction and NMR studies in mice indicate no fixed structure for the repeat domain. Silent mutations occur here with markedly reduced frequency compared with downstream domains that are equally well conserved at the amino acid sequence level, supporting selective bias at the nucleotide level (see Fig.9).

Recombination, in the form of unequal crossing-over, has been put forward by several authors(4, 5, 17, (20 as the mechanism by which insertions and deletions arise in the prion gene, though doubts have arisen as larger and more complex insertions were discovered. A recombination mechanism predicts that single repeat deletions are accompanied by an equal incidence ofreciprocal recombinants (correponding single repeat insertions); however these have never been observed. Recombination also does not account for wobble imperfections in repeats -- this would make recombination a potent point mutagen throughout the human genome since repeats are found in many genes, whereas recombination is usually viewed as a DNA repair and gene shuffling mechanism of high fidelity. Nonetheless, unequal crossing-over is capable of producing some observed products and may be responsible for a portion of them.
  • The animation shows how DNA slippage during replication of tandem repeats can lead to a repeat deletion. Here repeat Rd in the template slipped back to hybridize with Rc of the nascent chain, leading to a lack of Rd in the final new chain. Insertions differ only in that it is the nascent strand that slips, and the process can be iterated. Static view, all frames.
  • The animation shows how DNA slippage during replication of tandem repeats can lead to a repeat insertion. Here repeat Rd in the template slipped back to hybridize with Rc of the template chain, leading to an extra slightly modified Rc' in the final new chain. Insertions differ only in that it is the nascent strand that slips, and the process can be iterated. Static view, all frames.

    Phylogeny of deletion slippage

    under development -- webmaster
    Variation in repeat regions Variation in cross-species repeats

    Human prion repeat protein was used as a BLAST2 probe against a non-redundant set of public databases on August 11, 1997 using default settings. Homologous returned sequences (all prion orthologues) were parsed into individual repeat regions and sorted to eliminate redundancy.

    Terminology for repeat units from N to C is given in Column 1. Column 2 shows the number of species having a particular variant. Column 3 provides a proxy for species having the variant whose sequence is given in one-letter amino acid code in Column 4.

    The nonamer repeat Ra shows the most variants with eight. The sequence pqggggwgq is by far the most common and is suggested (with alternative pqgggt-wgq) to be ancestral by the marsupial sequence, pqgggtnwgq. Three point deletions occur in conjunction with various substitutions. Positions 1,2, 4, 7,8,9 are invariant. The kudu sequence tra.str is a weakly documented allele (ref ) with two surprising changes that may be attributable to cDNA error.

    The second repeat Rb is invariant as phgggwgq at the protein level in 75 species; marsupial has phpggsnwgq however. The third repeat Rc does not match Rb in three eutherian species even at the amino acid level, so this feature of the human sequence (which holds also at the DNA level) cannot be an required structural feature s. However, the slippage model, applied to an earlier era before the modern repeat structure was established, predicts ancient slippage upstream generated Ra, which in turn generated Rb-Re.

    The fourth and fifth repeats Rd, Re have minor point variation; the consensus sequence agrees with Rb and Rc. Mice and rats have an unusual serine preceding the tryptophan in both Rc and Rd.

    The fifth repeat Re has some interesting aspects. A fourth internal glycine occurs consistently in the ferrungulate clade (affecting artiodactyls, cetaceans, carnivores, and perissodactyls). This nonamer has no connection to Ra at the DNA level and probably represents a single micro-slippage event in a common ancestoral sequence.

    Certain placental mammals, most notably a bovine group, lama, giraffe, and squirrel monkey, have six repeats due to insertion events. These are similar to human insertion alleles and for alignment purposes must be considered on a case by case basis using microheterogenity in the DNA to pinpoint where the insertion occurred. Camel may have had an insertion followed by a deletion of the terminal nonamer. A single unit has been deleted in the black-handed spider monkey, Ateles geoffroyi, and five old world monkeys. Schaetztl (9 ) also reported an octamer deletion polymorphism in two of five orangutans.

    The most common allele in the bovine has six repeats; however, this is a fairly recent fixed insertion mutation. The five repeat polymorphism is the artiodactylan and mammalian norm. It may represent persistence of a more ancestral allele.

    RepeatFreq.Proxy spp.Sequence
    Ra 1 Tra.str pqeggdwgq
    Ra 2 Mus.mus pqgg-twgq
    Ra 6 Ory.cun pqggg-wgq
    Ra 35 Bos.tau pqggggwgq
    Ra 1 Cal.mol pqgggswgq
    Ra 3 Cri.gri pqgggtwgq
    Ra 1 Aot.tri pqsgg-wgq
    Ra 2 Rat.rat pqsggtwgq

    Rb 51 Tra.str phgggwgq

    Rc 48 Bos.tau phgggwgq
    Rc 2 Mus.mus phggswgq
    Rc 1 Tra.str phvggwgq

    Rd 1 Rat.rat phggg-gq
    Rd 47 Bos.tau phgggwgq
    Rd 3 Ceb.ape phggswgq

    Re 34 Pon.pyg phgggwgq
    Re 2 Rat.rat phgggwsq
    Re 7 Ate.geo -
    Re 10 Bos.tau phggggwgq

    Repeat regions in other species: amino acid level

    Special features of the artiodactyls

    [Under development]
    The artiodactyls present some unusual aspects in their repeat region, most notably a terminal nonamer repeat. Bovine and camel sequences have an insertion and deletion as well. Because so many artiodactyls have been sequence and because of their importance in the BSE epidemic, this whole area of the phylogenetic tree deserves special treatment. A concern of course is that certain polymorphisms might have heightened susceptibility to TSE as well as provide special dangers across species.

    X55882 is a GenBank cow sequence with 6 octapeptides; D10614 has 5 octapeptides:

    ...GG ...GG

    Ra 1 cct cag gga ggg ggt ggc tgg ggt cag 
    Rb 2 ccc cat gga     ggt ggc tgg ggc cag 
    Ri i cct cat gga     ggt ggc tgg ggc cag 
    Rc 3 cct cat gga     ggt ggc tgg ggt cag 
    Rd 4 ccc cat ggt     ggt ggc tgg gga cag 
    Rf 5 cca cat ggt ggt gga ggc tgg ggt caa 
                 ggt ggt
    Ra 1 cct cag gga ggg ggt ggc tgg ggt cag
    Rb 2 ccc cat gga     ggt ggc tgg ggc cag 
    Rc 3 cct cat gga     ggt ggc tgg ggt cag
    Rd 4 ccc cat ggt     ggt ggc tgg gga cag
    Re 5 cca cat ggt ggt gga ggc tgg ggt caa 
                 ggt ggt
    Comparing the two sequences, the simplest explanation is that (beginning with the short sequence) Rc in the nascent strand slipped back to Rb, and the insert Ri is a chimera of the proximal part of Rc with the distal part of Rb. In other words, the first codon of Ri, cct, derives from Rc whereas its last codon derives from Rb, in the slippage model.

    Y09760 camel Camelus dromedarius

    ccc cag gga ggg ggc ggc tgg ggt cag 
    ccc cac gga gga ggc     tgg ggt cag 
    ccc cac gga ggc ggc     tgg ggt caa
    ccc cac gga ggc ggc     tgg ggc cag 
    ccc cat ggt gga ggc     tgg ggt caa 
            ggt ggt ggc
    ccc cat ggt     ggt ggc tgg gga cag    cow  Rd 4 
    cca cat ggt ggt gga ggc tgg ggt caa    cow  Rf 5 
    The camel deletion is ambiguous but should be clarified by lama and any camel repeat polymorphism.  If camel originally had a terminal nonamer, then this could simply be deleted.

    U08309 black-handed spider monkey Ateles geoffroyi;
    U15164 black spider monkey x brown-headed Ateles paniscus x Ateles fusciceps
    GGG					AGG
    Ra ccc cag ggt ggt ggc tgg ggg caa 
    Rc ccc cat ggt ggc ggc tgg ggg cag 
    Rd ccc cat ggt ggc ggc tgg gga cag 
    Re cct cat ggt ggt ggc tgg ggt caa 
            gga ggt ggc
    Ra ccc cag ggt ggt ggc tgg ggg cag 
    Rb cct cat ggt ggt ggc tgg ggg caa
    Rc ccc cat ggt ggc ggc tgg ggg cag
    Rd ccc cat ggt ggc ggc tgg gga cag
    Re cct cat ggt ggt ggc tgg ggt caa 
            gca ggt ggc
    The Ateles geoffroyi results quite simply by deleting repeat Rb of the repeat region found in other species of this genus.

    Squirrel Monkey single repeat insertion

    Saimiri sciureu (squirrel monkey)
    Primates; Platyrrhini; Cebidae; Cebinae, Saimiri.
    The prion gene from two different animals were sequenced by different groups with accession numbers U15165 (Cervenakova) and U08310 (Schaetzl). At the protein level, the latter sequence has an extra copy of its third repeat [below] and changes of tyr to cys at codon 6 and lys to arg at codon 163. At the DNA level there are further silent changes of A to C preceding the trp codon in the penultimate repeat, and T to G at bp 606.

    Aligned squirrel monkey repeat DNA regions:

    the simplest explanation is that a tandem duplication of Rc occured; ie, Rc of the daughter strand looped out, Rd of the daughter hybridized with Rc of the template, 3' mismatches were removed, a new Rc was finished, and synthesis then continued on off Rd and Re, yielding 123345 for the newly synthesized strand.


    Pongo pygmaeus orangutan U08305

    Ambiguity zones:
    cct cag ggc ggt ggt ggc tgg ggg cag
    cct cat     ggt ggt ggc tgg ggg cag
    cct cat     ggt ggt ggc tgg ggg cag
    ccc cat     ggt ggt ggc tgg ggg cag
    cct cat     ggt ggt ggc tgg ggt caa
            gga ggt ggt 
    Signatures for distinguishable orangutan 24bp deletions:
    del A (pon.pyg) cct cat     ggt ggt ggc tgg ggg cag 
    del B (pon.pyg) ccc cat     ggt ggt ggc tgg ggt caa

    References for prion gene deletions

    The webmaster thanks Mark Palmer, Larisa Cervenakova, Katherine O'Rourke, Hermann Schaetzl, Stefan Kaluz, Otto Windl, Masanori Kubo, Lev Goldfarb,  and Roland Heynkes for valuable cooperation.
    1. Puckett C, Concannon P, Casey C, Hood L
    Genomic structure of the human prion protein gene 
    Am J Hum Genet 49(2), 320-329 (1991) 
    2. Laplanche JL, Chatelain J, Launay J-M, Gaxengel C, Vidaud M
    Deletion in prion protein gene in a Moroccan family. 
    Nucleic Acids Res 18(22), 6745 (1990) 
    3. Diedrich JF, Knopman DS, List JF, Olson K, Frey WH 2d, Emory CR, Sung JH, Haase AT
    Deletion in the prion protein gene in a demented patient. 
    Hum Mol Genet 1(6), 443-444 (1992) 
    4. Bosque PJ, Vnencak-Jones CL, Johnson MD, Whitlock JA, McLean MJ
    A PrP gene codon 178 base substitution and a 24-bp interstitial deletion in familial CJD. 
    Neurology 42(10), 1864-1870 (1992)
    5. Palmer MS, Mahal SP, Campbell TA, Hill AF, Sidle KC, Laplanche JL, Collinge J
    Deletions in the prion protein gene are not associated with CJD. 
    Hum Mol Genet 2(5), 541-544 (1993)
    6. Pocchiari M, Salvatore M, Cutruzzola F, Genuardi M, Allocatelli CT, Masullo C, Macchi G, Alema G, Galgani S, Xi YG,
    A new point mutation of the prion protein gene in CJD
    Ann Neurol 34(6), 802-807 (1993) 
    7  Salvatore M, Genuardi M, Petraroli R, Masullo C, D'Alessandro M, Pocchiari M et al.
    Polymorphisms of the prion protein gene in Italian patients with CJD 
    Hum Genet 94(4), 375-379 (1994); erratum in Hum Genet 1995 95(5):605 
    8. Laplanche JL, 1995
    Molecular genetics of familial and sporadic forms of human prion diseases 
    Ann Pharm Fr 53(5), 193-200 (1995)
    9. Schatzl HM, Da Costa M, Taylor L, Cohen FE, Prusiner SB
    Prion protein gene variation among primates. 
    J Mol Biol 245(4), 362-374 (1995); erratum in J Mol Biol 1997 Jan 17;265(2):257
    10. Perry RT, Go RC, Harrell LE, Acton RT
    SSCP analysis and sequencing of the human prion protein gene (PRNP) detects two
    different 24 bp deletions in an atypical Alzheimer's disease family. 
    Am J Med Genet 60(1), 12-18 (1995)
    11. Reder AT, Mednick AS, Cervenakova L, Goldfarb LG, Garay A, Ovsiew F
    Clinical and genetic studies of fatal familial insomnia. 
    Neurology 45(6), 1068-1075 (1995) 
    12. Vnecak-Jones CL and Phillips JA III
    Identification of heterogeneous PrP gene deletions in controls by detection of allele-specific heteroduplexes (DASH)
    AM J Hum Genet 50:871-872 (1992)
    13. Masullo, C, Salvatore, M, Macchi, G, Genuardi, M, and Pocchiari, M
    Progressive dementia in a Yyoung patient with a homoqygous deletion of the PrP gene
    Ann NY Acade Scie 724: 358-60 (1994)
    14. Goldfarb, LG, Brown, P, Cervenakova, L and Gajdusek, DC
    Genetic Analysis of CJD and Related Disorders 
    Phil.Trans.R.Soc.Lond. B 343: 379-384 (1994)
    15. Brown, P, Cervenakova, L, Goldfarb, L et al
    Iatrogenic CJD: Ancient genes and modern medicine
    Neurology  Feb;44(2):291-293 (1994)
    [Describes R34 in corneal graft and growth hormone case.]
    16. Windl, O, Demster, M, Estibeiro, JP, Lathe, R, et al
    Genetic basis of CJD in the UK: a systematic analysis of predisposing mutations and allelic variations in the PRNP gene
    Hum Genet 98: 259-264 (1996)
    17. Owen, F, Poulter M, Shah T
    An in-frame insertin in the prion protein gene in familial CJD
    Brain Res Mol Brain Res 7: 273-276 (1990)
    18. van Gool WA, Hensels GW, Hoogerwaard EM, Wiexer JHA, Wesseling P and Bolhuis PA
    Hypokinesia and presenile dementia in a Dkutch famiy with a novel insertion in the prion protein gene
    Brain 118: 1565-1571 (1995)
    19.  Owen, R, Poulter M, Collinge J, Leach M, Lofthouse R, Crow TJ, et al.
    A dementilng illness associatred with a novel insertion in the prion protein gene.
    Brain Res Mol Brain Res 13: 155-157 (1992)
    20.  Goldfarb L, Brown P, McCombie WR, Goldgaber D, Swergold GD, Wills PR, et al.
    Transmissible familial CJD associated with 5, 7, and 8 extra octapeptide coding repeats in the PRNP gene
    Proc Natl Acad Sci USA 88 10926-30 (1991)
    21. Cochran, EJ, Bennett DA, Cervenakova L, Kenney K, Bernard B, Foster NL, Benson DF, Goldfarb LG, and Brown, P.
    Familial CJD with a five-repeat octapeptide insert mutation
    Neurology 47: 727-733 (1996)
    22. Laplanche J-L,  Chatelain, J, Thomas S, Brown P, and Cathala F
    Analyse du gene PrP dans une famille d'origine Tunisienne atteinte de malade de Cretuzfeldt-Jakob.
    Rev Neurolog147: 825-827 (1991)
    23.Owen F, Poulter M, Collinge J, Leach M, Shah T, Lofthouse R, Chen YF, Crow TJ, Harding AE, Hardy J, et al
    Insertions in the prion protein gene in atypical dementias.
    Exp Neurol 1991 May;112(2):240-242
    24.Yamada M, Itoh Y, Fujigasaki H et al.
    A deletion in the prion protein gene in a Japanese family. [R2]
    Biomed Res 1994 15: 131-3
    25. Cervenakova L, Brown P, Piccardo P, Cummings JL, Nagle J, Vinters HV, Kaur P, Ghetti B, Chapman J, Gajdusek DC, Goldfarb LG
    24-nucleotide deletion in the PRNP gene: analysis of associated phenotypes
    in Transmissible Subacute Spongiform Encephalopathies, L. Court, B. Dodet Eds.
    Elsevier, Paris 1996  pp. 433-444
    26. Cervenakova L et al. [Pers. comm. 20 Aug 98]
    During 1997-98, found:
    1 additional case of R2  on 129 met
    6 additional cases of R34 on 129 met

    References for slippage in other genes

    1s. Oron-Karni V, Filon D, Rund D, Oppenheim A
    A novel mechanism generating short deletion/insertions following slippage is suggested by a mutation in the human alpha2-globin gene.
    Hum Mol Genet 1997 Jun;6(6):881-885
    2s.Hyland PL, McKinney MW, Keegan AL, McKenna PG, Curran MD, Middleton D, Barnett YA
    Sequence analysis of spontaneously-arising mutations at the aprt locus in wild-type and thymidine kinase-deficient Friend cells: evidence for strand slippage-misalignment mechanism in formation of deletions.
    Biochem Soc Trans 1997 Feb;25(1):127S 
    3s.Macey JR, Larson A, Ananjeva NB, Papenfuss TJ
    Replication slippage may cause parallel evolution in the secondary structures of mitochondrial transfer RNAs.
    Mol Biol Evol 1997 Jan;14(1):30-39 
    4s. Fitches AC, May SJ, Olds RJ
    A novel antithrombin gene mutation: slippage and mispairing as a mechanism of genetic disease.
    Pathology 1996 Nov;28(4):339-342 
    5s.Tran HT, Gordenin DA, Resnick MA
    The prevention of repeat-associated deletions in Saccharomyces cerevisiae by mismatch repair depends on size and origin of deletions.
    Genetics 1996 Aug;143(4):1579-1587
    6s.Luck R, Steger G, Riesner D
    Thermodynamic prediction of conserved secondary structure: application to the RRE element of HIV, the tRNA-like element of CMV and the mRNA of prion protein.
     J Mol Biol. 1996 May 24; 258(5): 813-826

    Supplementary Material

    18 Aug 98 webmaster
    Deletion polymorphisms in other species
    D34 Theropithecus gelada gelada baboon
    D34 Macaca sylvanus barbary macaque
    D34 Cercopithecus aethiops green monkey/grivet
    D34 Cercopithecus dianae dianae monkey
    D34 Cercocebus aterrimus mangabey
    D23 Pongo pygmaeus orangutan
    D12 Ateles geoffroyi black-handed spider monkey

    The silent versus sense anomaly: Hairpin C across species: The repeat region, despite the emphasis given here, has been very stable over evolutionary time at the amino acid level, even though its structure and function are uncertain. A sophisticated study of secondary structure in prion mRNA found (6s) a strongly conserved stem-loop region, called helix C, just upstream and partly including R1. The fulnction of hairpin C is not understood but might influence transcription or translation rates or ribosomal targeting within neurons. The structure might form at least transiently during replication in DNA as well and stabilize slipped strand intermediates, accounting in part for the marked instability of this region. Curiously, its length is exactly 24 bp if the bottom imperfect base pair is not included. It is not an easy matter to search databases for similar structures in other mRNAs.

    If this region, or the repeat region generally, functions mainly at the nucleotide level, then third codon position could be equally as important as first or second codon position,unlike the situation at the amino acid level. This predicts that evolutionarily fixed silent mutations would be less common here than in distal regions where selection acted mainly at the protein level. The prion protein affords an opportunity to test this idea by comparison to the invariant core domain 104-122: the repeat region should be changing relatively slower at third codon position than the invariant core, if indeed the repeat experiences strong selection at the nucleotide level.

    DNA repeats in octapeptides of normal human prion.

    The sequence is placed on horizontal and vertical axes and compared to itself with an online tool, receiving a white dot if there is a match. (This means that the diagonal is always a perfect match.) There is no new information below the diagonal because the matching process is symmetric. A moving convolution window of width 12 (four codons) removes clutter due to insignificant repeats.

    The three observed classes of deletions show up clearly on the intermittent line just above the diagonal. If slippage must be to the first adjacent region of significant complementarity, then del R2, del R23, and del R43 represent a complete set of allowable deletion classes. The extended streak for del A corresponds to the extended ambiguity zone first noted by Palmer et al. for deletion resulting in a 1234-pattern.

    Streak lines above the adjacency repeat line predict 6 additional deletion classes, corresponding to deletions to multiples of 24bp. These may be denoted as Ra/c, Rb/d, and Rc/e; Ra/d, Rb/e; and Ra/e which would result in final fusion repeats 1-3 45, 1 2-4 5, 12 3-5; 1-4 5, 1 2-5; 1-5, repectively. Slippage here must occur past the first region of possible complementarity to earlier regions; this has not been observed as yet.

    A similar graphic quickly predicts deletion patterns in any species.

    Wildtype reference human sequence

    GenBank accession P04156 D00015. (M13899 is original Kretschmar 1986 sequence, X83416 is Puckett sequence):
          1          11         21         31         41         51 
      241 ILLISFLIFL IVG      
     tggctgctgg atgctggttc tctttgtggc cacatggagt gacctgggcc tctgcaagaa
     gcgcccgaag cctggaggat ggaacactgg gggcagccga tacccggggc agggcagccc
     tggaggcaac cgctacccac ctcagggcgg tggtggctgg gggcagcctc atggtggtgg
     ctgggggcag cctcatggtg gtggctgggg gcagccccat ggtggtggct ggggacagcc
     tcatggtggt ggctggggtc aaggaggtgg cacccacagt cagtggaaca agccgagtaa
     gccaaaaacc aacatgaagc acatggctgg tgctgcagca gctggggcag tggtgggggg
     ccttggcggc tacatgctgg gaagtgccat gagcaggccc atcatacatt tcggcagtga
     ctatgaggac cgttactatc gtgaaaacat gcaccgttac cccaaccaag tgtactacag
     gcccatggat gagtacagca accagaacaa ctttgtgcac gactgcgtca atatcacaat
     caagcagcac acggtcacca caaccaccaa gggggagaac ttcaccgaga ccgacgttaa
     gatgatggag cgcgtggttg agcagatgtg tatcacccag tacgagaggg aatctcaggc
     ctattaccag agaggatcga gcatggtcct cttctcctct ccacctgtga tcctcctgat
     ctctttcctc atcttcctga tagtgggatg a
    DNA for normal 123 bp or 41 aa human repeat region: nucl: 151-274 or codons 51-91
    cct cag ggc ggt ggt ggc tgg ggg cag 
    cct cat     ggt ggt ggc tgg ggg cag 
    cct cat     ggt ggt ggc tgg ggg cag 
    ccc cat     ggt ggt ggc tgg gga cag 
    cct cat     ggt ggt ggc tgg ggt caa 
                gga ggt ggc

    Mad Cow Home ... Best Links ... Search this site