Familiar Familial Prionlike Disorders
Mad Cow Home ... Best Links ... Search this site

Familial Mediterranean Fever: indirect serum amyloid
Friedreich ataxia: non-amyloid intronic polyGAA repeat
Spinobulbar muscular atrophy -- X-linked androgen amyloid?
Oculopharyngeal dystrophy: autosomal dominant poly-alanine expansion
Wernicke encephalopathy (alcoholism) and CJD: chicken or egg?
Familial British dementia: read-thru of stop codon in cerebral amyloid angiopathy
Amyloid found in bacteriophage tailspike P22
Huntingtin amyloid or transglutaminase cross link?
Familiar familial prionlike disorders
Autosomal dominant 3' UTR CTG repeat expansion in SCA 8 and DM
Dentatorubralpallidoluysian atrophy: are fibrils congophilic?

Familial Mediterranean Fever: indirect amyloid

10 June 99 webmaster (based on Medline, OMIM, etc.)
Familial Mediterranean fever (FMF) is a autosomal recessive inflammatory disease frequently complicated by reactive systemic amyloidosis, recurring attacks of fever, synovitis, or serositis. The amyloid is formed of a common amyloid constituent, SAA (serum amyloid A) protein. However, after years of understandable confusion, the FMF susceptibility locus (MEFV) was mapped to chromosome 16p13 -- it encodes a 3.7-kb transcript expressed almost exclusively in mature granulocytes. The corresponding 781-amino acid protein, pyrin (or marenostrin), has no sequence similarity to the SAA amyloid .protein [see below].

Mutations clustered in exon 10 (M680V, M694I, M694V, I692del, K695R, A744S, and R761H and V726A), exon 2 (E148Q, E167D and T267I), exon 3(P369S), and exon 5 (F479L) account for most known familial cases.. Affected individuals are generally homozygotes or compound heterozygotes for these mutations. The MEFV carrier frequency was 21%, with E148Q the most common mutation. As with malarial resistance, a heterozygous selective advantage based on heightened inflammatory response to some class of pathogens endemic in the Mediterranean may account for the high frequency of heterozygotes.

AA amyloidosis can be studied in situ using serum amyloid P component scintigraphy; it is not restricted to any particular genotype. However M694V mutation is associated with a higher incidence of systemic amyloidosis which can be the sole manifestation of the disease, whereas V726A is the opposite. Abnormally high levels of serum amyloid A occur in attack-free intervals and very high levels at the onset of attacks. Ratios of amyloidosis type 1 and type II are reported as 123:157. Striking elevations of acute phase proteins, including serum amyloid A protein, manifest clinically as a nephropathy that passes through proteinuria, nephrotic and uremic stages to renal death. Elevation of SAA (serum amyloid A) is similar across ethnic groups and is also induced in many other unrelated diseases such as inflammatory bowel disease. Although FMF is not be caused by a mutation in the SAA or SAP, amyloidosis can be compounded with these.

Pyrin is a member of a family of nuclear factors homologous to the Ro52 antigen. It may shed light on the regulation of the acute inflammatory response. The periodic nature of inflammatory attacks in FMF is consistent with a protein that functions adequately at steady state but not under stress. Dramatic accumulation of neutrophils at the symptomatic serosal sites; is interpreted as the wildtype gene acting as an upregulator of an anti-inflammatory molecule or as a downregulator of a pro-inflammatory molecule; the role is to inhibit inflammation provoked by a minor insult, mutated pyrin is unable to inhibit unnecessary inflammatory events.

Transplanted normal kidneys are soon affected by amyloidosis. Conversely, transplanted FMF kidneys (not studied) would not affect normal kidneys because adequate levels of pyrin would be present -- FMF is a conventional loss-of-beneficial-function disease. This differs from fibrinogen alpha amyloidosis, also causing renal failure, but which is a gain-of-toxic-function disease. And unlike some secondary genes in familial Alzheimer, pyrin has no direct physical association with amyloid protein, and abnormal pyrin does not inactivate or recruit normal pyrin (the disease is recessive).

Familial Mediterranean fever is thus not an amyloid disease per se. It is of interest to prion research for two reasons:

(1) FMF leads indirectly to overproduction of another protein that is prone to amyloid formation. As FMF is (or was) extremely difficult to diagnose, its familial aspect might easily go undetected. Thus some cases of "sporadic" SAA amyloidosis (ie, where that gene -- and promoter -- have normal sequence) are actually sporadic in the sense that a 'rare' spontaneous conformational shift to a rogue conformer is not so rare when the protein is overproduced (which drives amyloid equilibrium by mass action). Any mechanism leading to overproduction could have this effect.

(2) The alleles frequencies are so high that a strong selective advantage must have accrued to heterozygotes in times past, along the lines of certain hemoglobinopathies. The homozygotes -- and their amyloidosis -- is then the price paid for a quick immune response to an unknown pathogen. There is no reason for such events be restricted to humans but not livestock. FMF is an unusual situation where a fatal amyloidosis is actually evolutionary favored.

Colchicine, an alkyloid from saffron inhibiting tubulin polymerization and thus mitotic chromatid movement, while effective in prevention and treatment of FMF-amyloidosis, will not be useful generally in amyloid diseases because it simply suppresses proliferation of cells involved in the inflammatory response. Colchicine, vinblastine, and taxol all bind to the beta subunit of tubilin (not to tau). While microtubules are 1D fibrils that can express strain types, unlike fibrils in amyloid diseases, their polymerization and depolymerization are under cellular control. Microtubules are not cross-beta structures and do not bind congo red in microtubular glomerulopathy fibrils. (Microtubule-associated protein tau is of course a very different story.)

Serum amyloid A (SAA) is an acute phase protein of unknown function. SAA circulates in plasma bound to high density lipoprotein-3 (HDL3). Serum amyloid A protein has length 122 amino acids in the unprocessed precursor:

  signal        1     18       
  chain        19    122       serum amyloid a protein.
  chain        19     94       amyloid protein AA.
  propep       95    122       often cleaved during amyloidogenesis.
  mod_res     101    101       methylation (di-).
  variant      15     15       g -> s.
  variant      70     70       v -> a (in 2-alpha, 2-beta, 1-beta and 1-gamma).
  variant      75     75       a -> v (in 2-alpha, 2-beta, and 1-beta).
  variant      78     78       d -> n (in 2-alpha, and 2-beta).
  variant      86     87       ff -> lt (in 2-alpha, and 2-beta).
  variant      89     89       h -> r (in 2-beta).
  variant      90     90       g -> d (in 1-beta).
  variant     102    102       e -> k (in 2-alpha, and 2-beta).
  variant     108    108       k -> r (in 2-alpha, and 2-beta).
          10         20         30         40         50         60 
           |          |          |          |          |          | 

          70         80         90        100        110        120 
           |          |          |          |          |          | 

Friedreich ataxia: non-amyloid replication slippage disease

12 June 99 webmaster (based on Medline, OMIM, etc.)
This ataxia is an interesting one. On the one hand, it is a replication slippage mutation like prion repeat disorders. It is often confused with familial alpha-tocopherol transfer deficiency and vitamin E may be therapeutic as in Alzheimer.

However, the repeat unit is a triplet GAA, not an octapeptide as in prion disease. However, it is not a triplet amyloid slippage disease like Huntington because the repeat occurs in an intron and does not appear in protein. However, the part of the intron where the slippage occurs is an AluSx retrotransposon, like that in prion gene 6977-7266 of U21985. However the prion AluSx does not bear strong homology in the GAA region. Even if it did, it would not cause CJD by knocking out transcription of one copy of the gene because CJD is over-production driven and autosomal dominant.

No one has ever looked for somatic mutation expansion in extra-repeat CJD. However, in Friedreich ataxia they report a case. FRDA researchers have also been much more aggressive in using extended haplotype markers to track down founder events, whereas in CJD they have relied on questionaires. Sporadic FDRA of course makes no sense.

Friedreich ataxia (FRDA), the most common autosomal recessive ataxia (1 in 50,000 live births) is caused in 96% of cases by homozygous expansions of an unstable GAA repeat localised in intron 1 (within an AluSx retrotransposon) of the FRDA gene on human chromosome 9q13-21.1. The expansion interferes with transcription. The 210 amino acid protein, frataxin, plays an important role in intramitochondrial iron homeostasis. Thoug a direct knockout mutation, in terms of oxidative damage it has common ground with various amyloid disorders.

Medline carries various confusing abstracts asserting frataxin to be a splice variant comprising exon 18 of the STM7 gene for phosphatidylinositol-4-phosphate 5-kinase activity, which has C-terminal similarity to the ataxia-telangiectasia gene. Less than 10 kb separates the CpG island identified in the X25/exon 1 from the 3-prime end of STM7/exon 16. Ataxia phenotype in mice lacking the type 1 inositol-1,4,5-triphosphate receptor also supported STM7 as the Friedreich ataxia gene. Finally, the structure of frataxin cDNAs, existense of mouse intronless pseudogenes, the nature of point mutations found in some patients, and the size of the endogenous frataxin protein ruled this out. In fact, a protein kinase intervenes between STM7 and FDRA:

Human chromosome 9q13 frataxin neighborhood

 CMD1B, Cardiomyopathy, dilated-1B, autosomal dominant
 FRDA, Friedreich ataxia (frataxin)
 PRKACG, Protein kinase, cAMP-dependent, catalytic, gamma 
 STM7, Phosphatidylinositol-4-phosphate 5-kinase, type I, beta

In yeast, lack of the yeast frataxin homolog (YFH1) results in mitochondrial iron accumulation, suggesting that FRDA results from oxidative damage secondary to mitochondrial iron overload: Yfh1p regulates mitochondrial iron efflux. The FRDA gene has weak homologues back to bacteria.

Friedreich ataxia is a classical autosomal recessive loss-of-function disorder with an uncommon mechanism for an inactivating mutations. Other cases are compound heterozygotes for a GAA expansion and one of at least 11 frataxin point mutation: M1I (cannot initiate translation), D122Y, G130V, various exon 1 truncating mutations (L106X), and missense mutations in the last three exons coding for the mature frataxin protein (eg I154F in exon 4), and an A-to-G transition disrupting the acceptor splice site at the end of the third intron.

Normal sequence already contains a moderate trinucleotide repeat with bimodal (polymorphic) size distribution. Small alleles have 6-10 repeats; large alleles (17% of normal alleles) range from 16 to 36 consecutive GAA. All longer normal alleles are interrupted by a hexanucleotide repeat (GAGGAA). These can expand into intermediate-size (42 - 60 GAA) that then undergo cataclysmic expansion to pathological range within a single generation (thus only subtle 'anticipation') , similar to myotonic dystrophy and fragile X. Pathological FRDA alleles range from 120 to 1,700 repeat units, or 360-5100 extra bp.

normal gene:     
      aggacgcggt ggctcatgcc cataatctca gcactttggg aggcctagga aggtggatca
      cctgaggtcc ggagttcaag actaacctgg ccaacatggt gaaacccagt atctactaaa
      aaatacaaaa aaaaaaaaaa aaaagaagaa gaagaagaag aagaagaaga aaataaagaa
      aagttagccg ggcgtggtgt cgcgcgcctg taatcccagc tactccagag gctgcggcag
      gagaatcgct tgagcccggg aggcagaggt tgcattaagc caagatcgcc caatgcactc
      cggcctgggc gacagagcaa gctccgtctc aaaaaataat aataataaat aaaaataaaa

A case of somatic expansion is on record in a case with mild ataxic phenotype and a family history of multiple sclerosis. While the patient was homozygous for the GAA triplet repeat expansion, the sural nerve biopsy revealed a significantly smaller expansion size, constituting direct demonstration of somatic mosaicism in FRDA [Muscle Nerve 1998 Mar;21(3):390-3].

Expansion from a 'premutation' is a concern inprion disease in cattle or humans with 6-7 octapeptide repeats, as is subclinical disease. Human prion (U29185) contains only a GAAGAA di-triplet and a tri-triplet, CAGCAGCAG so there is no risk of expansion of these triplets (and inactivated prion genes do not cause TSE, which is a gain-of-toxic-function disorder. Human prion does contain an Alu-Sx retrotransposon at 6977-7266.

FRDA founder effects are very pronounced in large normal alleles, one German mutation being dated to a minimum age of 107 generations. To look for a common founder of specific mutations, detailed linkage analysis employing six polymorphic chromosome 9 markers is used (extended haplotype), rather than compiling family histories (where parental parentage is often unknown. This method could also be applied to prion disease but it has not. It might be especially interesting in sheep.

Friedreich ataxia presents very similarly to ataxia with vitamin E deficiency (AVED) or familial isolated vitamin E deficiency, an autosomal recessive loss-of-function disease of alpha-tocopherol transfer protein (alpha-TTP). Conversely, it does not seem to be known if vitamin E is therapeutic in Friedreich ataxia, by protecting against oxidative damage from reactive oxygen species due to too much iron in the mitochrondria. Vitamin E is also used in Alzheimer to slow damage due to amyloid reactivity with iron and copper. However, in Friedreich ataxia there is no amyloid; oxidative damage is due directly to loss of mitochondrial function.

Various models of DNA in trinucleotide repeats have been put forward -- an unusual DNA structure during replication may trigger expansion (as with hairpin C in prion disease). One nmr-based model has FRDA repeats forming a triplex in which the TTC strand folds on either side of the same GAA strand. Another model, called sticky DNA, formed by the association of two purine.purine.pyrimidine (R.R.Y) triplexes in negative supercoils.

Genes with similar potential for disease could be sought by auto-searching the human genome project for tandem repeats near CpG islands. This has not been done (or at least not reported) -- Friedreich ataxia is not likely to be unique being within an AluSx retrotransposon.

Misdiagnosis of Friedreich ataxia and other ataxia is not unusual. A study of 361 families with adult-onset ataxia of unknown etiology (clinical diagnosis of FDRA specifically excluded) found among 178 autosomal dominant kindreds [Neurology. 1998 Dec;51(6):1666-71.]:

-- SCA1 expansion at a frequency of 5.6%,
-- SCA2 expansion at a frequency of 15.2%,
-- SCA3 expansion at a frequency of 20.8%,
-- SCA6 expansion at a frequency of 15.2%,
-- SCA7 expansion at a frequency of 4.5%.

Among patients with apparently recessive or negative family histories of ataxia, 6.8% and 4.4% tested positive for a CAG expansion at one of the autsomal dominant loci, and 11.4 and 5.2% of patients with apparently recessive or sporadic forms of ataxia had FDRA expansions. The repeat sizes for one or both FA alleles were relatively small, with sizes for the smaller allele ranging from 90 to 600 GAA repeats. Some 39% of the autosomal dominant kindreds could not be attributed to a known triplet repeat expansion.

Frataxin at SwissProt

  TRANSIT       1      ?       MITOCHONDRION.
  CHAIN         ?    210       FRATAXIN.
  VARIANT     122    122       D -> Y (IN FA).
  VARIANT     130    130       G -> V (IN FA).
  VARIANT     154    154       I -> F (IN FA).
  VARIANT     155    155       W -> R (IN FA).
  VARSPLIC    161    210       ISOFORM WITH 5B EXON).

     210 AA;  23218 MW 

Spinobulbar muscular atrophy (SBMA)

15 June 99 webmaster (based on Medline, OMIM, etc.)
This disease is of considerable interest in comparison to extra repeat familial prion disease. Somatic mosaicism is seen, chimpanzees carry a longer premutational expansion than humans, a polyglycine repeat does not result in fibrils, paternal anticipation is minimal, and misdiagnosis is rampant (lost in ALS). The androgen receptor research community has done an excellent job with specialized online mutational databases though there is a lot of work to do on characterizing protein aggregates.

Spinal and bulbar muscular atrophy (SBMA, formerly Kennedy disease) is an X-linked (Xq11-q12) motor neuron dystrophy caused by expansion of a polyglutamine CAG repeat within exon A of the androgen (dihydrotestosterone) receptor(AR), a 919-amino acid protein. (Bulbar refers to neurons of the medulla oblongata.) The androgen receptor has 3 major functional domains: a modulatory N-terminal encoded by exon 1 (1,586 bp) which already codes for a 19 residue glutamine repeat, a DNA-binding domain encoded by exons 2 and 3 (152 and 117 bp, resp.), a hinge region in the first half of exon 4, and an androgen ligand binding domain from 5 exons varying from 131 to 288 bp.

As in other diseases of this type, aggregation of the protein occurs in a polyglutamine length-dependent manner and the cellular toxicity is coupled to aggregation, suggesting that the molecular basis of neuronal degeneration in SBMA is toxic gain of function. Note that medulla oblongata and spinal chord are not exactly primary androgen targets and the several hundred mutations giving loss of function unsurprisingly are not primarily neurological in nature (none giving symptoms similar to SBMA), as befits a hypothetical prionlike disorder. Intranuclear inclusions are found in nonneural tissues as well as affected neural tissues and contain mutant peptide. Amazingly, no one has looked for congophilic birefringence.

Electron microscopy shows aggregates have a familiar ultrastructure. AR Q48 aggregates sequester mitochondria and steroid receptor coactivator 1 and stain positively for NEDD8, Hsp70, Hsp90 and HDJ-2/HSDJ. Co-expression of HDJ-2/HSDJ significantly represses aggregate formation in vitro. AR Q48 aggregates also stain with antibodies recognizing the PA700 proteasome caps but not 20S core particles, suggesting that ARQ48 accumulates due to protein misfolding and a breakdown in proteolytic processing.

While the size of CAG repeats in the AR gene is a determinant factors of the severity and progression rate of SBMA phenotypes, meiotic and somatic instability of CAG repeats are less than in other diseases caused by trinucleotide repeat expansions. However, detailed analysis of 20 SBMA cases including 4 autopsied patients established a tissue-specific pattern of mosaicism. The prominent somatic mosaicism was observed in the cardiac and skeletal muscles, which are predominantly composed of postmitotic cells, and in the skin, prostate, and testis. The central nervous system, liver, and spleen showed the smallest mosaicism.

The mutation frequency of the CAG repeat region at the androgen receptor locus has been measured using a rare SBMA semen sample. Among 258 X chromosome-containing sperm, 19% had a repeat number equal to the donor's somatic DNA (47 repeats), 66% were expansions and 15% were contractions. The average expansion was 2.7 repeats. More than half of the expansions involved one or two repeats; the largest was 11 repeats. 68% of the contractions were also one or two repeats but six (16%) were very large (12-25 repeats). One contraction generated an allele in an intermediate size range (33-39 repeats).

Disease-associated alleles (37-66 CAGs vs wild-type 19 CAG repeats)) change in length when transmitted from parents to offspring, with a significantly greater tendency to shift size when inherited paternally. Anticipation is relatively rare in SBMA in contrast to spinocerebellar ataxia type 1 and dentatorubralpallidoluysian atrophy which show obvious paternal anticipation. Spermatogenesis may cause paternal anticipation.

Point mutations in exons 2-8 encoding the DNA- and androgen-binding domain result androgen insensitivity, clustering clusers in exon 5 and in exon 7. The number of mutations in exon 1 is extremely low and no mutations have been reported in the hinge region in the first half of exon 4. The androgen receptor (AR) mutations database has several hundred mutations in it and an excellent map showing distribution. The mutation database is best searched at EMBL.

Inheritance: This disease is on the X chromosome (like fragile X repeat diseases FRAXA and FRAXE ) so is not directly comparable other polyglutamine diseases which are all autosomal dominant. The issue here is whether a toxic congophilic cross-beta fibril is responsible for the disease. Males have one X so it cannot be asked whether the disease is dominant nor whether a bad protein can recruit a good allele. Females have two X but only one active in a given cell. Females are affected by androgen insensitivity mutations [Exp Clin End Diab 1998;106(6):446-53 ] and heterozygous female carriers with repeat expansion are known [J Neurol 1996 May;243(5):388-92 etc.]. However, the SBMA literature treats this disease as a sex-linked recessive disease -- discoverer WR Kennedy reviews its history in Neurology 1998 Mar;50(3):583-94.

However, if the protein inclusions in this disease are cross-beta fibrils, then females should exhibit a mosaic pattern of inclusion formation and possibly have clinically detectable mild symptoms with very late onset. Amazingly, no one has looked for this. Because of polyglutamine repeat properties in other diseases, the protein inclusions are likely a small N-terminal fragment of androgen receptor itself, rather than some other protein over-produced in response to altered receptor (as in FMF).

Misdiagnosis: The clinical presentation of amyotrophic lateral sclerosis (ALS) is variable and overlaps with SBMA. Using the CAG repeat expansion for accurate molecular diagnosis in patients with motor neuron disease, 147 sporadic male ALS patients and 100 unrelated male patients from 100 familial ALS (FALS) kindreds were screened. ALS was clinically misdiagnosed in 2% of sporadic cases and in two of the 100 FALS kindreds. [Neurology 1997 Aug;49(2):568-72 ]

Androgen receptor also has polyglycine repeats encoded in the same exon 1, a tract of 24 in wildtype even longer than the 19 polyglutamine wildtype. Expansions and contractions of the glycine tract to 15-31 repeats are known but not associated with disease. Note only the last 18 GGC glycine codons are strictly repetitive in third codon position and presumbably are the ones subject to slippage. A prionlike disease is not expected here because glycine is unsuitable for cross-beta fibril; however, polyglycine makes for an interesting internal 'control' in this polyglutamine disease. The hexarepeat of glutamine is not known to experience expansion. A moderate expansion would probably still leave it short of the threshold needed for disease.

Evolution of polyglutamine repeats: Androgen receptor is a member of the nuclear receptor superfamily so a ligand-dependent transcription factor. Comparison of AR coding sequence from five primate species, Homo sapiens (human), Pan troglodytes (chimpanzee), Papio hamadryas (baboon), Macaca fascicularis (macaque), and Eulemur fulvus collaris (collared brown lemur) shows very strong conservation of the DNA and steroid binding domain sequences with an overall homology of 85%.

Note chimps have an even longer expansion than humans and SBMA could be expected to occur naturally in this species. Rats have a repeat length of 22 and mice of 20, so great apes appear to have an ancestral length while old world monkeys and lemurs have seen net contractions.

Human AR repeat regions in exon 1 (the second polyglutamine tract does not exhibits expansions):

 L  L  L  Q  Q  Q  Q  Q  Q  Q  Q  Q  Q  Q  Q  Q  Q  Q  Q  Q 
 Q  Q  Q  Q  E  T  S  P  R  Q  Q  Q  Q  Q  Q  G  E  D  G  S 
 C  G  G  G  G  G  G  G  G  G  G  G  G  G  G  G  G  G  G  G  G  G  G  G  G  E  A 
Comparison of polyglutamine repeat in primates:
chimp     1   ........................................................Q... 60
Papio     1   ......................................................------ 54
Macaca    1   ......................................................------ 54
Eulemur   1   ....................................................R.------ 54

chimp     1   ......................-......--............................. 117
Papio     55  .............-.......-............................. 103
Macaca    55  -............-.......Q............................. 103
Eulemur   55  -----........PQ......-........QS.......A.........Q. 99
Rat and mouse androgen receptors establish that the polyglutamine repeat is ancient but risk has disappeared in mouse.whereas the polyglycine is more primate-specific:




Familiar familial prionlike disorders

16 June 99 webmaster (based on Medline, OMIM, etc.)
By Jan 99, 14 disorders due to this type of mutation had been described. Fragile X and myotonic dystrophy are most similar to Friedreich's ataxia, involving an anomalous number of repetitions, reaching over a thousand copies of a given trinucleotide.

-- Friedreich ataxia: long GAA repeats
-- fragile X syndrome (FRAXA): long CGG repeats
-- fragile X syndrome FRAXE):long CGG repeats
-- myotonic dystrophy (DM): long CTG repeats, expanded CTG repeat in 3' UTR of dystrophia myotonica protein kinase (DMPK) on 19q13.
-- myotonic dystrophy 2 maps to 3q. chromosomal regions containing the genes for muscle sodium and chloride channels that are involved in other myotonic disorders are excluded -- spinocerebellar ataxia (SCA8): long CTG repeats (107-127) non-coding 13q21

Another group of 9 disease known as polyglutamine disorders have an increased number of CAG repetitions (producing an expanded polyglutamine sequence in the protein) but on the order of only a hundred repeats:

-- Huntington's disease: short CAG repeats
-- spinobulbar muscular atrophy: short CAG repeats: androgen receptor >
-- spinocerebellar ataxia (SCA1): short CAG repeats
-- spinocerebellar ataxia (SCA2): short CAG repeats
-- spinocerebellar ataxia (SCA3 or Machado-Joseph disease MJD: short CAG repeats
-- spinocerebellar ataxia (SCA6): short CAG repeats: voltage gated alpha 1A calcium channel
-- spinocerebellar ataxia (SCA7): short CAG repeats
-- dentatorubralpallidoluysian atrophy (DRPLA): CAG repeats
-- autosomal dominant pure spastic paraplegia (ADPSP): short CAG repeats

A third group has short extra polyalanine or polyaspartate repeats:

-- oculopharyngeal dystrophy: 2-7 extra GCG repeats: poly(A)-binding protein 2 PABP2
-- pseudoachondroplastic dysplasia (PSACH): polyaspartate: cartilage oligomeric matrix protein (COMP)

Prion disease involving extra tandem octapeptide repeats (but not point mutation disease) can also be classed with these replication slippage diseases.

Except for SBMA (which is X-linked), CAG repeat disorders are characterised by autosomal dominant heredity, somatic mosaicism, and anticipation (i.e., earlier onset age and increasing severity in successive generations). This is not observed in prion disease because the slippage events are orders of magnitude rarer. Somatic mutation has not been sought rigorously in sporadic CJD.

In normal individuals the loci contain a short length of triplet repeats (usually 5-40), which is polymorphic within the population. Increases in the lengths of the translated triplet repeats to 40-100 are associated with disease symptoms, whereas the untranslated triplet repeats to 200-3000 are associated with the disease.

(to be continued with individual updates).

Molecular Basis of the Neurodegenerative Disorders.

N Engl J Med 1999 Jun 24;340(25):1970-1980
Martin JB.
This review appeared on 24 June 1999. It covers many of the same diseases from a medical prespective reviewed here from a prion-oriented comparative molecular biology viewpoint.

There are many good aspects to the review, which covers the genetics and citations well and critically, and it brought up some new diseases on the threshold of classification. But why not simply provide instead a half-page set of links to OMIM, whick has fresh and far more detailed expert reviews of all of these diseases and more: where is the value-added?

Even though the disease screen was simply neurodegenerative disease, the adult onset autosomal dominant diseases in the article are prionlike or arguably so, though the author cannot bring himself to acknowledge this nor cite GG Glenner's very similar disease set and far-sighted organizing principle in his 1980 review. Very little has happened in the last 20 years other than Glenner's table now has more lines.

In fact, last week was remarkable in that 2 new gene products were added to Glenner's table: familial British dementia (FBD) and P22 tailspike.

Amyloid is not simply beta-pleated sheet. Silk is not an amyloidosis. This is found in numerous places in the article, eg, fig 1, pg 1972 ("beta-pleated sheets that stain with birefringent stains for amyloid"), and pg 1978 ("Homodimerization of protein fragments into beta pleated sheets occurs in Alzheimer and prion diseases"). Note that the first quote is a double error: only congo red exhibits birefringence (acquired optical activity from alignment along fibrils, the other dyes exhibit enhanced fluorescence) and the second quote is a triple error [AD and PrP form homo- and hetero-oligomers and dimers could not possibly exhibit detectable CR birefringence].

For the record, beta-pleated sheets, beta helix, beta-helical proteins, and cross-beta fibrils are not synonymous. The later terms are better thought of as super-secondary structure. CR birefringence is so far uniquely associated with cross-beta fibrils and is completely diagnostic for it. The question for this author is, 'is it coincidental that all these diseases are associated with an unique protein aggregate architecture that is never seen outside a disease setting?'

Oculopharyngeal dystrophy: expanded poly-alanine repeat

16 June 99 webmaster (based on Medline, OMIM, etc.)
Oculopharyngeal dystrophy (OPMD) is an autosomal dominant muscular dystrophy of world-wide distribution characterized by late onset eyelid drooping (ptosis), swallowing difficulties (dysphagia), and proximal muscle weakness mapping to chromosome14q11.2-q13. Nuclear filament inclusions (15- to 18-nm tubulofilaments) in skeletal muscle fibres are diagnostic.

The gene affected has recently been identified as poly(A) binding protein 2 gene (PABP2). Pathological expansions of the polyalanine tract may cause mutated PABP2 oligomers to accumulate as filament inclusions in nuclei. A GCG x 6 repeat encoding an N terminal polyalanine tract located at the of the protein is expanded to 8-13 repeats in 144 OPMD families screened.

More severe phenotypes were observed in compound heterozygotes for the 9x mutation and 7x allele (found in 2% of the population) than for 9x with 6x, whereas the GCG x 7 allele alone is autosomal recessive for OPMD. Thus the 7x repeat is a polymorphism which can act either as a modifier of a dominant phenotype or as a recessive mutation. This is reminiscent of codon 129 met/val in prion diseases.

The frequency of the OPMD mutation is 1 in 1,000 in the Province of Quebec, 1 in 200,000 in France, and 1 in 700 in Bukhara Jews living in Israel. The meiotic stability of a 9x repeat is interesting: a single ancestral OPMD carrier chromosome was introduced to the French-Canadian population by 3 sisters in 1648. Only 1 in 71 families show today a further expansion to12x, calculated to be 1 expansion in 598 meioses. This makes PABP2 repeat volatility considerably more extreme than that of prion octapeptide repeat.

OPMD has been misdiagnosed as familial cases of inclusion body myositis, which shares histologic characteristics with some advanced cases of oculopharyngeal dystrophy. However, no cases of sporadic IBM have been reassigned to OPMD.

Though no one has looked for congophilic birefringence nor characterized the fragment forming inclusions nor sought strain types in different repeat lengths, OPMD could possibly be a new class of prionlike diseases. Alanine is well suited for cross-beta structure and indeed the amyloidogenic prion peptide 106-126 has a similar alanine/glycine run. However, the short nature of the expansion would require that wild type and the 7x repeat be near the disease threshold. It may be that expansions per se are not the issue, but rather congruence with a peptide with just the wrong sequence for fibril formation.

For comparison purposes, human prion contains 112 MAGAAAAGAVVGGLGG 127 in its fibrilogenic region, which could be (non-homologously) aligned with PABP2 giving 10 /16 residues of agreement (62%) for the 10x alanine repeat given at GenBank, supportive of the interpretation of the prionlike interpretation of mutations in this protein:


The protein is extremely conserved: 304/306 are identical in human/cattle, 296 identical human/mouse, or 294/306 identical mouse/cattle. This alignment suggests that 10x is wildtype in mammals (to the extent a definite sequence length has meaning), in conflict with the OMIM entry and the OPMD literature: shorter repeats must be slippages of the human gene.

Of course, slippage and length variants might be found in mouse in cattle had more individuals been examined. These might have potential for 'infectious' dietary seeds.

Human PABP2
        1 maaaaaaaaa agaaggrgsg pgrrrhlvpg aggeagegap ggagdygngl eseelepeel
       61 llepepepep eeepprprap pgapgpgpgs gapgsqeeee epglvegdpg dgaiedpele
      121 aikarvreme eeaeklkelq nevekqmnms pppgnagpvi msieekmead arsiyvgnvd
      181 ygataeelea hfhgcgsvnr vtilcdkfsg hpkgfayief sdkesvrtsl aldeslfrgr
      241 qikvipkrtn rpgisttdrg fpraryrart tnynssrsrf ysgfnsrprg rvyrgrarat
      301 swyspy
bovine    1    ...........................................
mouse     1    ......................................D.... 

Polyaspartic acid has been found in pseudoachondroplasia (PSACH). Here the repeat is GAC x 5, coding for 5 consecutive aspartates within the calmodulin-like region of the COMP protein. Short expansions to 6x and 7x have been observed. Expansion as well as shortening of the repeat could cause the same disease. In cartilage, both patients had the rough endoplasmic reticulum inclusions in chondrocytes and tendon.

A review in 1987 of pseudoachondroplasia pedigrees previously reported as compatible with autosomal recessive inheritance concluded that most or all are instances of gonadal mosaicism for the autosomal dominant mutation. Inclusions are characteristic but not specific for pseudoachondroplasia since similar but smaller inclusions are found in multiple epiphyseal dysplasia. (Type I multiple epiphyseal dysplasia is autosomal dominant also with an expansion of a short GAC x 5 repeat to 6x.)

It is not clear at this point that either of these latter diseases is prionlike.

Wernicke encephalopathy and CJD: chicken or egg?

webmaster 17 June 99
Wernicke encephalopathy is commonly found in alcoholics, which is to say is commonly found. Thiamine deficiency is the immediate problem, with cofactor-depleted transketolase of the pentose shunt long suspected as the root of the problem. In Wernicke-Korsakoff syndrome, the Km of transketolase for thiamine pyrophosphate is off, an inborn error of metabolism clinically important only with inadequate dietary thiamine.

There are two mapped transketolases; regretably, neither shows sequence differencs in WK disease, leaving the whole involvement of TK up in the air. Recall there are some 14 other known thiamine deficiency diseases, many with neurological sign. Alpha-ketoglutarate dehydrogenase is also down in WE.

Wernicke encephalopathy might be a fairly common misdiagnosis of CJD, as well as a disease that apparently co-occurs with CJD. With a chronic alcoholic, is much effort expended on diagnosis when the symtoms fit WE? No, no more effort than would be expended in a case of Alzheimer in an elderly patient. One has to wonder if impaired oxidation brain damage in WE is a trauma predisposing that brain to CJD, especially in cases of early onset.

Wernicke encephalapathy was the initial diagnosis in the case of the 35 year old deer hunter in Missouri. However, he did not display appropriate symptoms of WE in detox and was eventually diagnosed with CJD (though the autopsy report has never been released).

Wernicke encephalopathy-like symptoms as an early manifestation of CJD in a chronic alcoholic.

J Neurol Sci 1999 Mar 1;163(2):192-8 
Nagashima T, Okawa M, Kitamoto T, Takahashi H, Ishihara Y, Ozaki Y, Nagashima K
A case of Creutzfeldt-Jakob disease (CJD) with presenting Wernicke encephalopathy (WE)-like symptoms and severe insomnia is presented. An 80-year-old alcoholic man with a 6 month history of tremors, ataxia, memory loss and confabulation, developed profound insomnia, confusion, and delirium with vivid hallucinations. Polysomnography revealed a marked reduction of sleep time, with central-type sleep apnea. Neither myoclonus nor periodic synchronous discharge (PSD) was observed. An autopsy revealed diffuse spongiform changes and astrocytosis throughout the cerebral gray matter, with severe involvement of the mammillary bodies and thalamus. Prion protein (PrP) immunostaining was positive in kuru plaques in the cerebellum, PrP polymorphism at codon 129 was heterozygous Met/Val, and proteinase K resistant PrP (PrP(res)) was demonstrated by Western blotting. The lack of necrotizing lesions in the mammillary bodies, thalamus, and periaqueductal gray matter could rule out WE. The data suggest that the present case of CJD is consistent with PrP(res) type 2 (CJD M/V 2), but was unique in the lack of some typical CJD signs and the presence of signs of WE and sleep abnormalities.

CJD associated with Wernicke encephalopathy.

Can J Neurol Sci 1988 May;15(2):156-60 
Gaytan-Garcia S, Gilbert JJ, Deck JH, Kaufmann JC
Wernicke disease (WD) is a complication of alcoholism and malnutrition and usually presents acutely and is characterized by disturbances of consciousness, paralysis of the external ocular muscles, and ataxia. The disease results from deficiency of vitamin B 1, or thiamine, an essential coenzyme in intermediate carbohydrate metabolism. On the other hand, CJD results from infection with an unconventional agent with a long incubation period and is characterized by a rapidly progressive dementia and histologically by a spongiform encephalopathy associated with neuronal destruction and pronounced astrogliosis. Combination of both diseases has not been reported in the literature previously and their relationship is uncertain. We present 3 cases with this interesting association and consider their relationship.

Creutzfeldt-Jakob disease presenting as Wernicke-Korsakoff syndrome.

J Neurol Sci 1992 Apr;108(2):149-53 
Pietrini V
A 47-year-old man began to suffer from progressive truncal ataxia and mental alterations typical of Wernicke-Korsakoff syndrome. He showed confusional state, hallucinations, delirium of jealousy and a serious impairment of recent memory. The symptomatology lasted 13 months, but only in the last weeks was it complicated by myoclonias. Triphasic pseudoperiodic sharp-waves characterized the EEG-recordings only in the final stage. Macroscopic examination of the brain showed marked atrophy of the mammillary bodies and superior vermis. However, the histological features were consistent with Creutzfeldt-Jakob disease (CJD) with focal accentuation of the changes in the latter structures. This case supports the hypothesis that CJD-changes begin focally in the CNS and, subsequently, spread along neuronal pathways, probably via central axons. Only in the final stage does the pathological process involve most parts of the gray matter. A focal accentuation of the CJD process in the cerebello-mammillo-thalamic system caused in this case a Wernicke-Korsakoff-like syndrome.

A stop-codon mutation in the BRI gene associated with familial British dementia

Nature 399, 776 - 781 (1999) 24 June 1999
RubenİVidal, BlasİFrangione, AguedaİRostagno, SimonİMead, ... JorgeİGhiso
Comment (webmaster): Here is a brilliant paper very relevent to prionlike diseases that appeared yesterday in Nature.

In a short note, they report the molecular basis for a prionlike disease known only in a single British kindred, sequence the gene and its splice variants in 6 species, map it to chromosome 13 in humans, study its tissue expression and covalent modifications, diagnose at-risk family members, offer pre-natal diagnosis, distinguish FBD from many similar diseases, and determine the gene's mode of evolution back to chickens. After direct mass spec of isolated amyloid, this miracle could be accomplished in hours by using conventional EST database searching, even tissue specific expression of splice variants.

This paper really shows the power of these database projects in accelerating disease research. The only bottlenecks are normal function and 3D structure -- they could not get at them in this instance.

Contrary to press releases, this disease has nothing whatsoever to do with Alzheimer. It is a subtype of cerebral amyloid angiopathy, a new gene, a new mechanism for generating amyloid unrelated to that of AD. The mechanism is simple: read-through of the stop codon into DNA never meant to code for protein. This creates a C-terminal fragment joined to something that cannot be folded properly; chaperones probably then default this to cross-beta fibril. This suggests that there will be a very large number of such diseases.

The really big news with this paper is that the results enable a powerful new search paradigm for unrecognized prionlike diseases: go through the 10,000+ diseases at OMIM for late onset, autosomal dominant, neurological diseases with known stop mutations alleles presenting with inclusions (hopefully congophilic but probably not looked for). In effect, the FBD people may have knocked off several diseases with just a single paper.

Now, the search tool at OMIM isn't the best, but they graciously allow downloading of the whole 23 meg database -- the webmaster's kind of people. This means anyone can auto-index it with their favorite compound boolean search engine with this particular search need in mind.

Familial British dementia (FBD), previously called familial cerebral amyloid angiopathy-British type, is an autosomal dominant disorder of undetermined origin characterized by progressive dementia, spasticity, and cerebellar ataxia, with onset at around the fifth decade of life. Cerebral amyloid angiopathy, non-neuritic and perivascular plaques and neurofibrillary tangles are the predominant pathological lesions. The disease was first described in 1933 as familial presenile dementia with spastic paralysis. See:

Plant, G. T.et al: Brain 113, 721-747 (1990): case histories of 26 affected individuals in 5 generations, noting congophilia and faint PAS-positivity noted in perivascular plaques.. The OMIM entry summarizes the literature under presenile dementia with spastic ataxia. The 1990 paper is apparently the most recent one.

A very curious aspect at OMIM is the nature of the Medline papers called up by an auto-search based on first paragraph textual clues:

Osborne JP, et al. 
       Huntington's chorea. Report of 3 cases and review of the literature. 
       Arch Dis Child. 1982 Feb;57(2):99-103. Review. 

     Myers RH, et al.  
       Late onset of Huntington's disease. 
       J Neurol Neurosurg Psychiatry. 1985 Jun;48(6):530-4. 

     Starkstein SE, et al.  
       Depression in patients with early versus late onset of Parkinson's disease. 
       Neurology. 1989 Nov;39(11):1441-5. 

The authors worked backwards with mass spectroscopy from tryptic digests of isolated amyloid fibrils. A novel 34 residue protein fragment called ABri was sequenced in pieces. By great good fortune, the partial sequence was represented in human, chicken, rat, mouse, rabbit, and pig EST collections, though the gene and protein were completely unknown otherwise. This allowed the full gene and promoter to be sequenced and mapped by FISH to human hromosome 13q14 and the evolutionary conserved regions to be delineated.

A single-exon gene coding for a protein of 266 amino acids was found. The composition is remarkable for 9 cysteine residues and is highly rich in leucine and isoleucine (each 9 %). The protein has a single identifiable type-II integral transmembrane-spanning domain between amino acids 52 and 74 with the C-terminal part being extracellular, a single N-glycosylation site at asparagine 170, but no signal sequence nor known protein motifs. Homology with mouse was 96%. There may be paralogues: J. Biol. Chem. 271, 19475-19482 (1996).

Seven heterozygously affected members of 200 member FBD kindred had a single nucleotide transition (T for A) at stop codon 267, resulting in read-through, producing 11 extra amino acids from previously untranslated 3' sequence. Most conveniently, this change creates a restriction site for rapid diagnosis. The amyloid thus consists of the carboxy-terminal 23 residues plus these 11 randomly generated amino acids: EASNC FAIRH FENKF AVETL ICSRT VKKNI IEEN. GenBank AAD40370AF152462], seems to give the correct sequence [below].

Thus we have a curious situation in which two groups came upon this protein by different approaches. Song's paper is entitled, 'Human putative transmembrane protein E3-16 mRNA, complete cds' and is isolated from normal pituitary.

 M  V  K  V  T  F  N  S  A  L  A  Q  K  E  A  K  K  D  E  P 
 K  S  G  E  E  A  L  I  I  P  P  D  A  V  A  V  D  C  K  D 
 P  D  D  V  V  P  V  G  Q  R  R  A  W  C  W  C  M  C  F  G 
 L  A  F  M  L  A  G  V  I  L  G  G  A  Y  L  Y  K  Y  F  A 
 L  Q  P  D  D  V  Y  Y  C  G  I  K  Y  I  K  D  D  V  I  L 
 N  E  P  S  A  D  A  P  A  A  L  Y  Q  T  I  E  E  N  I  K 
 I  F  E  E  E  E  V  E  F  I  S  V  P  V  P  E  F  A  D  S 
 D  P  A  N  I  V  H  D  F  N  K  K  L  T  A  Y  L  D  L  N 
 L  D  K  C  Y  V  I  P  L  N  T  S  I  V  M  P  P  R  N  L 
 L  E  L  L  I  N  I  K  A  G  T  Y  L  P  Q  S  Y  L  I  H 
 E  H  M  V  I  T  D  R  I  E  N  I  D  H  L  G  F  F  I  Y 
 R  L  C  H  D  K  E  T  Y  K  L  Q  R  R  E  T  I  K  G  I 
 Q  K  R  E  A  S  N  C  F  A  I  R  H  F  E  N  K  F  A  V 
 E  T  L  I  C  S  -  T  V  K  K  N  I  I  E  E  N  - 
Human-mouse proteins aligned:

mouse   1   ......................S.....V............G................ 58

mouse   59  ................................L...................R....... 118

mouse   119 ......DA...................................................K 178

mouse   179 .................................V.N.......................R 238

mouse   239 ...........T................ 266

The amyloid-forming proteolytic fragment bears no known resemblance to any other known congophilic peptide. (Immunoreactivity co-localized with yellow-green birefringent material after Congo red staining.) There is nothing in the sequence suggesting that amyloid could be formed. Obviously, there is no native fold for a protein that is one-third non-evolved sequence. Other stop-codon readthrough diseases are thus strong candidates to also form amyloid; certain thalassemias are given as read-through diseases (but these would not cause amyloid as hemoglobin is only expressed in short-lived cell types.

This amyloid had been purified from an individual with onset at age 56 years, 10 year duration, severe widespread cerebrovascular amyloidosis in the brain and spinal cord, non-neuritic amyloid plaques affecting the cerebellum, hippocampus, amygdala and occasionally the cerebral cortex, with changes in the periventricular white matter, perivascular amyloid plaques and neurofibrillar degeneration in hippocampal neurons.

Major messenger RNA transcript of 2.0 kb and1.6İkb were expressed in the brain, placenta, kidney and pancreas, whereas lower expression levels were observed in the heart, lung, liver and skeletal muscle. Northern analysis of different brain regions detected predominantly the 2.0-kb transcript in the cerebellum, spinal cord, subthalamic nucleus, substantia nigra and hippocampus, whereas lower expression levels were observed in the cerebral cortex, amygdala and thalamus. Transcripts using the first polyA signal, ATTAAA were found in ESTs from placenta, fetal liver/spleen , aorta, white blood cells, parathyroid tumours, pregnant uterus, and fetal heart. Transcripts that used the second polyadenylation signal, TATAAA, were found in breast, pregnant uterus, pancreas, cerebellum, parathyroid tumour, placenta , and heart.

Misdiagnosis: FBD had previously been called an atypical form of familial Alzheimer's disease, spongiform encephalopathy, and primary congophilic angiopathy, even though amyloid deposits did not stain correctly with appropriate antibodies (except amyloid-associated proteins such as apoliproproteins E and J, serum amyloid P-component, anad alpha and beta tubulin).

No immunoreactivity using two antibodies was observed in brain sections of sporadic cerebral amyloid angiopathy (CAA), sporadic Alzheimer's disease, Down's syndrome, hereditary cerebral haemorrhage with amyloidosis-Dutch type, hereditary cerebral haemorrhage with amyloidosis-Icelandic type, Hungarian transthyretin cerebral amyloidosis, nor systemic cases of light-chain amyloidosis (kidney), light-chain deposition disease (kidney) and amyloid A (heart).

FBD is completely unrelated to Alzheimer's disease, but presents similarly to it, especially to a presenilin 1 variant 9 described recently. Both disorders have in common the production of 4k peptides created by proteolytic processing of integral membrane proteins. The cytoskeletal pathology of FBD is undistinguishable from that observed in other neurodegenerative diseases, including sporadic and familial forms of Alzheimer's disease and the vascular form of prion protein (PrP)-CAA. [See Ghetti, B. et al. Vascular variant of prion protein cerebral amyloidosis: stop codon 145 mutation in PRNP. PNAS 93, 744-748 (1996) -- note this is not a read-through mutation].

Faulty gene find encourages Alzheimer researchers

Reuters Financial Report  Wed, Jun 23, 1999 
LONDON -- U.S. and British researchers have discovered a gene mutation that causes a type of hereditary senility similar to Alzheimer's disease, the most common form of dementia among the elderly. In a report in the science journal Nature on Wednesday, the researchers said the fault in a gene called BRI produces an unusual protein fragment that causes a build-up of amyloid plaque in the brain of sufferers.

Scientists at NYU School of Medicine in New York and the Institute of Neurology in London said their findings could lead to a better understanding of diseases such as Alzheimer's and ways of stalling their progression. "This is an important finding because it will give us another window onto the brain diseases associated with amyloid," Dr Jorge Ghiso, of NYU School of Medicine, said in a statement.

Amyloid is a term used to describe proteins that form unusual plaque deposits in the brain. Abnormal plaque and tangled bundles of fibres in the brain are two characteristics of Alzheimer's disease. Some scientists believe the build-up of amyloid deposits kills brain neurons and causes the memory loss and forgetfulness that are characteristic of diseases such as Alzheimer's.

"This exciting discovery provides new opportunities to uncover the cascade of events leading to the loss of synapses (relay stations between nerve cells) and neurons triggering the development of dementia," said Creighton Phelps, the director of the Alzheimer's Disease Centres Program at the National Institute on Ageing.

"If we can interfere with this cascade, common to many of the dementias, including Alzheimer's, we may be able to stall the disease process in its tracks," he added. The scientists found the faulty gene by comparing the DNA of a woman from a family with a long history of dementia, which the scientists named Familial British Dementia, with DNA in a public databank with thousands of DNA sequences.

In a separate report in Nature, scientists in the United States identified a protein, called Pin1, that could help to prevent or unwind the tangled fibres in the brains of Alzheimer's patients that are linked to dementia. The tangles are made of a protein called tau which becomes distorted in patients with the disease. Kun Ping Lu of Harvard Medical School in Boston showed that Pin1 can restore the function of tau in laboratory studies. "This is the first study to show a potential role for Pin1 in Alzheimer's disease," he said.

Formation of fibrous aggregates from a non-native intermediate: the isolated P22 tailspike beta-helix domain

J Biol Chem, Vol. 274, Issue 26, 18589-18596, June 25, 1999
Benjamin Schuler, Reinhard Rachel, and Robert Seckler 
Comment (webmaster 29 June 99):
Before reading this note, open a second window to get the protein under discussion in interative immersive 3D. This is at RCSB, the successor to Brookhaven PDB. Open a third window to compare this with generic authentic amyloid

About 5 years ago, a very bizarre protein structure called beta-helix turned up in an obscure family of proteins. In the webmaster's opinion, this structure amounts to intra-molecular amyloid. (Recall that prion and similar amyloid are inter-molecular cross-beta, ie, the cylinder is comprised of several strands.) The difference is that the intra-molecular case requires more non-beta connectors and loop-outs being a single connected peptide.

The question for amyloid has always been, if it is such a great default structure, why wasn't it ever used for something non-pathological? P22 tailspike and the pectin lyases answer that question. (Note: fungal researchers have put forward the idea of regulation by recruitmental sequestration in Podospora and Ure3 but have not established this in a natural setting.)

In essence, the beta-helix domain of native tail spike is close to a naturally occuring amyloid monomer but blocked from extension by other domains of the protein. This raises the spectre of a mutated virus inducing amyloid in its host. Congo red birefringence in fibrillar aggregates reported below show that beta helical proteins are essentially cross beta amyloid waiting to happen, ie that beta helix is structurally blocked cross beta.

In other words, it might be possible for a phage accumulating tailspike monomer to be partly proteolytically degraded, unblocking it to form amyloid within the host. (In amyloidoses, it is seldom de novo native folding that goes astray but rather chaparone-mediated refolding of fragments. Some of the Clp 'chaparones' actually are proteases.)

What exactly is the difference between ordinary pleated beta sheet, beta in cross-beta, and the beta in beta-helix (structurally and in terms of CR binding)? Working with 1TSP in Chime indicates that the beta strands of P22 are reasonably perpendicular to the cavity axis and might give the standard cross-beta spacing of 4.7 A in fiber diffraction. Individual P22 beta helices seem to have a typical Chothia right-handed twist.

One generic amyloid repeat takes 115.5 A with 24 strands, which proportionally for 13 stands of P22 would be 62.6 A, which also seems about right for 1TSP. The diameter of P22 cylinder but it seems within the range of amyloid fibril. Transthyretin cross-beta is usually modelled as parallel beta. Loops are allowed in cross beta though it seems to be more pure beta than P22 with all its connectors. The trimer of P22, while driven by intercalation of extraneous domains, may still illuminate how amyloid fibers are formed out of fibrils; the intra/inter distinction getting lost.

Native tailspike may bind congo red. This could not give rise to detectable birefringence unless many tail spikes were aligned , perhaps oriented parallel to an applied magnetic field as in Worcester 1978. Could CR not be diffused into crystals? The tryptophan -rich caudal fin possibly accounting for the quenching should be deleted out for purposes of studying the aggregate. Also, would not tandem repeats extend the beta helical region in a regular way?

Binding of congo red to parallel beta structure will all be the same, though only in select situations involving aligned fibril axes will birefringence be detectable (and non-specific binding to non-beta in protein could also occur). This is because CR in planar twofold form apparently has just the right length and chemical affinities relative to characteristic beta dimensional spacing.

Article highlights:

The Salmonella bacteriophage P22 recognizes its host cell receptor, lipopolysaccharide, by means of six tailspikes which are thermostable side-by-side homotrimers of parallel beta-helices, an architecture first discovered in 1994 and known only in 3 bacterial pectate lyases, a fungal pectin lyase, and rhamnogalacturonase A. P22 tailspike itself has endorhamnosidase activity. (Pectins are the polysaccharides of middle lamella and primary cell wall of dicotyledonous plants consisting of homogalacturonan 'smooth' regions and highly rhamnified 'hairy' regions of rhamnogalacturonan; pectate is a less methylated form.)

These proteins contain a large right-handed helix with 7-13 turns of about 30 amino acids each. The walls contain three parallel beta sheets formed by consecutive segments of the overall helix. These beta elements are typically bounded by an unusual left-handed alpha helix residue and looped out regions of variable length. A third of the residues have sidechains aligned at the surface or in the interior of the parallel beta helix. This new protein structure element (left-handed alpha bounded beta) gives rise to novel periodic features in the amino acid side chain interactions. These form linear stacks that include asparagine ladders, serine stacks, aliphatic stacks, and ringed-residue stacks.

The aligned residues at the surface are dominated by threonine, aspartic acid and asparagine, whereas valine, leucine and isoleucine are most frequently found in the interior, a very large hydrophobic cavity. This strong positional preference for specific amino acids interior to the beta helix led to a sequence profile that could be run against SwissProt to find proteins that fold into the parallel beta helix. Of the 19 protein families identified, 7 were known carbohydrate-binding proteins and 9 were a subset of the leucine-rich repeat (LRRs) domain-containing proteins. The CD spectra has distinct components attributable to parallel beta-helix.

For convenient interactive visualization of 1TSP, see the 3D image or view in immersive 3D.

The beta helix moiety, expressed without extraneous termini, folds reversibly at low ionic strength conditions. Increased ionic strength induces aggregation by a linear polymerization mechanism.While CD shows the secondary structure content of the aggregates is similar to that of the native state, at the same time tryptophan fluorescence was markedly quenched, implying a drastically changed environment of at least some of the tryptophans (but this could simply involve the tryptophan-rich caudal fin domain, a 64İresidue loop between the third and fourth coil of the beta-helix). Aggregates are reactive with anti-tailspike antibodies that do not recognize the native trimer and grow by incorporating tailspike chains in a specific, partly folded conformation rather than originating from unfolded chains.

Microscopic analysis of the aggregates revealed a variety of morphologies; among others, fibrils with fine structure were observed that exhibited bright green birefringence if viewed under cross-polarized light after staining with Congo red. These observations, together with the effects of folding mutations on the aggregation process, indicate the involvement of a partially structured intermediate distinct from both unfolded and native beta helix. Even in a light microscope, large, elongated, laterally associated fibrillary aggregates containing fine structure were clearly detectable, with diameters ranging from about 50 nm for single fibrils to about 50,000 nm for lateral assemblies.

Additionally, flat, sheet-like and partially curled structures were observed. Viewing samples stained with Congo red in a polarization microscope without analyzer showed all aggregates independent of morphology to bind Congo red, leading to an intense red color of the aggregates. Use of an analyzer with a polarization plane perpendicular to that of the polarizer revealed bright green birefringence for some of the aggregates described, especially those of large, fibrous morphology. For some of the thin fibrils without detectable fine structure, a faint birefringence was found, whereas neither the sheet-like structures nor the curled or amorphous aggregates exhibited detectable birefringence. Reasons for this could be either nonregular structure or insufficient thickness of the assemblies to polarize passing light to a detectable degree.

In the case of a beta-helix a drastic change of tertiary structure without simultaneous disruption of secondary structure is not plausible. Thus, possible explanations are (a) a rearrangement of the whole molecule to form a completely different beta-structure in the aggregated state with a similar beta-sheet content, (b) formation of an out-of-register beta-helix with unfavorable packing, or (c) a local denaturation process, involving either the caudal fin domain, a tryptophan-rich 64İamino acid insertion between the third and fourth coil of the beta-helix, or partial unfolding of terminal coils of the beta-helix.

In terms of aggregate structure, possibility (a) could be expected to lead to the formation of aggregates with the established cross-beta structure [Sunde, M., and Blake, C. (1997) Adv. Prot. Chem. 50, 123-155], but possibilities (b) and (c) might cause fibril formation according to a recently proposed model, assuming an assembly of beta-helical protofibrils for the formation of amyloid [Lazo, N. D., and Downing, D. T. (1998) Biochemistry 37, 1731-1735]. An amyloid-like structure might be the basis of aggregates, at least in view of some of the morphologies found, their binding to Congo red, and the bright green birefringence under cross-polarized light typically used as a diagnostic tool to identify amyloid . A significant shift of the point of maximum spectral difference in the Congo red binding assay in solution was not observed, probably because of the high proportion of amorphous aggregates present.

Earlier on-site articles on congo red:
fatal fibrils II 
fatal fibrils I 
conformational disease 
CR and huntingtin 
universal therapy
URE3 and Sup 35 
SH3 domains  

See also:
Science 1993 Jun 4;260(5113):1503-7
Structure 1993 Dec 15;1(4):241-51
Nat Struct Biol 1994 Oct;1(10):717-23
Science 1994 Jul 15;265(5170):383-6
J Struct Biol 1998;122(1-2):236-46
J Struct Biol 1998;122(1-2):216-22
Protein Eng 1997 Dec;10(12):1373-7
J Mol Biol 1997 Apr 11;267(4):865-80
see pdb entry header for length of beta sheets.

Autosomal dominant 3' UTR CTG repeat expansion in SCA 8 and DM

Nat Genet 1999 Apr;21(4):379-84 
Koob MD, Moseley ML, Schut LJ, Benzow KA, Bird TD, Day JW, Ranum LP 
Comment (webmaster):
This disease is so new that only a single article has been published on it, in April 1999, though OMIM already covers it. While the new gene represents an important refinement in the classification of spinocerebellar ataxias, being the 8th in the series, it also presented some paradoxical features.

First, though the screen was for a CAG polyglutamine repeat expansion (as found in SCA 1, 2, 3, 6,7) and this screen was successful for a new expansion location on chromosome 13q21 in a subset of the 361 kindred ataxia pool, the CAG expansion was not in a transcribed strand. Further it was flanked by numerous stop codons and a lack of splice sites. No polyglutamine tract can arise from this.

The other strand, the expansion counterpart a CTG expansion, was however expressed, predominantly in brain. It was found within a 4 exon transcript (numbered backwards D, C, B, A and found at AF126749; C lost in alternate splicing). For this strand, the expansion has possessed consensus splice sites and a polyA site distally. Thus SCA 8 would seem to be a CTG expansion in 3' UTR of messenger RNA of an unknown 3-4 exon gene. However, that putative gene had numerous stop codons.

Using the first exon D as primer, a novel polyadenylated transcript was found from the first strand. This transcript is said to contain a long ORF but -- outrageously -- its sequence is not shown or posted at GenBank. It overlapped with the CTG strand transcript solely in exon D. This strange turn of events led the authors to conclude that the repeat strand was an endogenous antisense regulatory RNA [see Gene 211:1-9 1998].

Antisense RNA to the prion gene needs to be revisited in light of these developments. That RNA was never sequenced and its signficance was pooh-poohed. However, it will be found by the human genome sequencing project within a few months (if it hasn't been already). [The last paper on this was Moser M et al Nature 1993 Mar 18;362(6417):213-4 ]

The only known precedent for SCA 8 is myotonic dystrophy (DM), caused by a CTG expansion in the 3' untranslated region of the dystrophia myotonica-protein kinase gene (DMPK) on chromosome 19. (Friedreich's ataxia is also a non-coding expansion of GAA but in an early intron and recessive, simply a loss of function.)

Why are these two expansions diseases autosomal dominant? There cannot be birefringent fibril based on polyglutamine etc. because the repeat is never manifested in protein. One explanation is simply dosage deficiency: the 3' UTR insert interferes with protein production from the affected allele, leaving an inadequate level from the normal allele that for some reason is not compensatable by feedback regulation. Yet this does not fit the data. Imprinting was specifically ruled out for SCA 8 -- both alleles are transcribed.

Current thinking for DM is that the expansion is a toxic gain-of-function mutation via creation of excessive binding sites for two hnRNPs (NAB50 and/or CUGBP), sequestering them to the extent that numerous muscle-specific mRNAs cannot be translated. It is too soon to say whether this model applies to SCA 8.

Note that DNA slippage is still the likely mechanism of repeat expansion. (Note SCA 8 is also polymorphic in normal controls for 11 CTA repeat expansions.) SCA 8 and DM have this much in common with certain forms of familial CJD and other forms of SCA (which are further similar in that the same triplit is amplified). Furthermore, analagously to amyloid, the repeat expansion may trap and sequester normal protein. However, the effect is not manifested by protein-protein aggregation because in SCA 8 and MD only normal protein is produced. Rather, the effect is through mRNA-protein sequestration at least in MD. Amyloid diseases so far seem to produce toxic product directely, rather than cause loss-of-normal function through sequestration. All of this makes for a rather complicated Venn diagram.

Medline and OMIM highlights:
SCA8 patients have expansions similar in size (107-127 CTG repeats) to those found among adult-onset DM patients. SCA8 is the first example of a dominant SCA not caused by a CAG expansion translated as a polyglutamine tract.

While searching for CAG repeat disorders in patients with undefined dominantly inherited ataxias, Koob et al. identified 80 CAG repeats followed by 11 TAG repeats in genomic DNA from a mother and daughter with adult-onset spinocerebellar ataxia (SCA). This expansion mapped to chromosome 13q21.

This repeat expansion was then identified in 7 other pedigrees with autosomal dominant SCA. The largest pedigree, of 7 generations, was clinically evaluated and tested. Onset of symptoms ranged from age 18 to 65, with a mean of 39 years. Dysarthria, mild aspiration, and gait instability were commonly the initial symptoms. Exam findings included spastic and ataxic dysarthria, nystagmus, limb and gait ataxia, limb spasticity, and diminished vibration perception. Progression was generally fairly slow, but severely affected family members were nonambulatory by the fourth to fifth decades.

Normal repeat length in 1,200 normal alleles was 16 to 37 repeats in 99% of alleles. Repeat lengths of up to 91 were seen in a small proportion of controls. Repeat length in the large SCA8 kindred ranged from 107 to 127 CTG repeats to isolated cases of up to 600.

As with the CTG expansion in myotonic dystrophy, repeat length contracted with paternal transmission (-86 to +7) and expanded with maternal transmission (-11 to +600). Maternal bias towards expansion has not been seen in the CAG repeat disorders causing other SCAs. There is no explanation available.

The two GenBank entries are an unsatisfactory accompaniment to a supposedly peer-reviewed paper. AF126749 has 1472 bp of cDNA for the CTG strand but is not annotated for exons A-D. AF126748 covers 1159 bp of the EcoR1 clone on the CAG strand with 310-822 annotated for a "3' terminal exon" only and the polyA signal in complement (shown below flanked in green as transferred to the CTG strand).

Blastn of the region preceding the repeat of the CTG strand produces one hit (112/113 identities) to a sequence called homo sapiens clone TGC13-7a trinucleotide repeat region, AF087653, which probably came up on a genome-wide screen and is basically identical to AF126748. This 1442 bp sequence was posted 20-Apr-1999 by a group from Toronto, ie, a day before the Nature Genetics paper appeared.

Blastn against full-length CTG sequence provides 397/402 hits. The plus to plus match begins extends from 1071/482 to 1472/843. The raises the question of what the first 481 bp of the Toronto sequence corresponds to: evidently spliced out regions of the SCA 8 gene missing from its processed cDNA.

The webmaster advised both groups on 1 July 99 of the overlap of the 13q21 sequence posted by JB Vincent tel: (416) 979-2221 2631 Neurogenetics, Clarke Institute of Psychiatry, Toronto group.

Their unpublished paper is provacatively entitled, "Unstable DNA in major psychoses: cloning of a new unstable trinucleotide repeat region on chromosome 13." Recent similar abstracts are below.

AF087653 agrees very well with a portion of the SCA 8 sequence: Blastn against full-length CTG sequence AF126749 provides 397/402 hits. The plus to plus match begins extends from 1071/482 to 1472/843. In other words, the first 481 bp of this sequence may represent genomic sequence spliced out of cDNA AF126749, and similarly for base pairs 884-1442. However, the SCA 8 EcoR1 sequence AF126748 has 310 bp of this latter material (720/727 identity) so only 249 is extensional, roughly 1193-1442 of AF087653. A Blastx search for known protein using a chimeric AF126749-AF087653 probe did not turn up a candidate protein.

spinocerebellar ataxia 8
repeat_region   1071..1348
all reading frames have numerous stop codons
        1 atccttcacc tgttgcctgg ctagagttgt ctggctccac tttgagctct tgcagaacca
       61 gccctttttc gtgtggtcca ggaaagtcca tgcctggcac cacctcctcc tctagtgact
      121 ccacgtagaa gagagtcctg gctggctgct gagtgccctg cccaggagcc ccttgctgca
      181 gcctcgtggc aactggaagc agggtgccat tcagcggatt gaaggaagag gaggaagagg
      241 acggggagga cgatgaagag gaagaggagg aaggcttctt ccagaaagtg ctcacaccgc
      301 ttctctcttg gcttttgagc aggcgactct ggctgggtcc ccagtgctca aagctgccac
      361 tgccgtcctg ttgcaggcag cctccccccg ccgggccgcc ggtggaagga gacgggtggc
      421 tgaagagttt ccagcggagt cgcagaatgt gcttcacatc gaagtctttt cgcccagagc
      481 ctgacatgct ttacgcacag aaggcaaaag gctggcagct cacgcagggt tctggaggct
      541 gggaagttca agaccaatgc acgagaattt ggtctaaaga gaatcttctt gctctgaaca
      601 cacatagtag aaggcagaag ggcaagagag agaacaaagt ctgtgtctcc acatggcaga
      661 agagcagagg agacagaacc tactcctcta tggcaaccac cccatcaatg acaaaaatcc
      721 tagaaggatg tatgtatagg aagttgaagt gttgagaaga gaatggctca gagtcaagcg
      781 ggaacaagat tcaaacttca gagagagagg gaagaaaaac atttaaatat atctggcata
      841 atccaagact atttacgaca agtgttctgt gtttctaata ataaaacaga cttcacctcg
      901 gagtacctgc agaactggga ccccaatgac cagggagaat gaagaacaac ttgtttgaag
      961 attgcctttt ctgactccca gcttccacgg agagattaac tctgttggct gaagccctat
     1021 cccaattcct tggctagacc ctgggtcctt catgttagaa aacctggctt tactactact
     1081 actactacta ctactactac tactactgct actgctgctg ctgctgctgc tgctgctgct
     1141 gctgctgctg ctgctgctgc tgctgctgct gctgctgctg ctgctgctgc tgctgctgct
     1201 gctgctgctg ctgctgctgc tgctgctgct gctgctgctg ctgctgctgc tgctgctgct
     1261 gctgctgctg ctgctgctgc tgctgctgct gctgctgctg ctgctgctgc tgctgctgct
     1321 gctgctgctg ctgctgctgc tgctgctgca ttttttaaaa atatattatc ttattttact
     1381 atttgatgtt ataattgtta tatatttttc catacttcct catactgctt atctcttact
     1441 taagaattta tgaataaaga attgattttt ca

Intergenerational CAG repeat expansion at ERDA1 on 17q21.3 in a family with childhood-onset depression, schizoaffective disorder, and recurrent major depression.

Vincent JB, Kovacs M, Krol R, Barr CL, Kennedy JL
Am J Med Genet 1999 Feb 5;88(1):79-82 
Reported evidence of anticipation for schizophrenia and bipolar disorder has recently precipitated a search for unstable trinucleotide repeats for these diseases. Several initial studies suggested an increase in the frequency of large CAG/CTG repeats in the genomes of schizophrenic and bipolar individuals. Published reports do not demonstrate expansion per se, and may be suggestive of allelic association with the disease rather than actual dynamic DNA mutations. This report documents evidence of a significant expansion of CAG/CTG repeats from one generation to the next in a family demonstrating evidence of anticipation for psychiatric disorders. Using the repeat expansion detection (RED) technique, we observed that a proband with multiple psychiatric diagnoses, including childhood-onset depression, inherited a larger CAG/CTG repeat than either parent. Analysis of the ERDA1 locus revealed that the proband inherited a very large allele from the father which increased in repeat number through transmission. The mother was diagnosed with schizoaffective disorder and the father with depression. While this DNA mutation may be a stochastic event unconnected with the disease, it could represent DNA instability as an etiologic factor in psychiatric diseases.

Transglutaminase aggregates huntingtin into nonamyloidogenic polymers, and its enzymatic activity increases in Huntington's disease brain nuclei

PNAS Vol. 96, Issue 13, 7388-7393, June 22, 1999
The protein huntingtin (htt), aggregated in neuronal nuclear inclusions, is pathognomonic of Huntington's disease (HD). Constructs, translated in vitro from the N terminus of htt, containing either polyQ23 from a normal individual, or polyQ41 or polyQ67 from an HD patient, were all soluble.

Transglutaminase (TGase) crosslinked these proteins, and the aggregations did not have the staining properties of amyloid. More TGase-catalyzed aggregates formed when the polyglutamine domain of htt exceeded the pathologic threshold of polyQ36.

Furthermore, shorter htt constructs, containing 135İaa or fewer, formed more aggregates than did larger htt constructs. TGase activity in the HD brain was increased compared with the control, with notable increases in cell nuclei. The increased TGase activity was brain specific. In lymphoblastoid cells from HD patients, TGase activity was decreased. TGase-mediated crosslinking of htt may be involved in the formation of the nonamyloidogenic nuclear inclusions found in the HD brain.

The staining properties of nuclear inclusions in the HD brain revealed that they were not amyloid [as claimed recently But note only 2 patients were examined, that there are many reasons why birefringence could be missed (as listed by Caughey), among them the authors' hostility to it. Absence means little in a small series of patients -- not all CJD can be demonstrated to have this effect.-- webmaster].

Dentatorubralpallidoluysian atrophy: filaments never tested for birefringence

30 June 99 Medline, OMIM, etc 
Comment (webmaster):
Dentatorubralpallidoluysian atrophy (DRPLA) is an autosomal dominant neurodegenerative disorder characterized by a loss of neurons in the dentate nucleus, rubrum, globus pallidus and Luys' body. Clinical features are myoclonus, epilepsy, dementia, and cerebellar ataxia. It is often found within Huntington misdiagnoses.

In reading Medline abstracts, we see words like filaments, bundles of filaments, filamentous structures, intranuclear inclusions, and ubiquitinated abnormal complex yet nowhere is there a test for congo red birefringence. The 1185 residue protein, atrophin-1, has been identified from a gene on chromosome 12p13.31 and contains a polyglutamine expansion similar to many of the spinocerebellar ataxias. Repeat size varies from 7 to 23 in normal individuals; in patients, as allele is expanded to 49 - 75 repeats or more. An oddity with this protein is the presence of other tandem repeated amino acids and also repeated glu-arg.

The function is unknown; rat and mouse are 93% identical to human but have a shorter repeat, averaging 15. Yeast 2-hybrid screens pull out interacting proteins like those of huntingtin. There is a second gene in both rat and human with about 50% homology suggesting a paralogue. The closest nematode protein, U41557, is not close.

Feature list
  DOMAIN       73     82       SER/GLU-RICH (MIXED CHARGE).
  DOMAIN      302    305       POLY-PRO.
  DOMAIN      376    382       POLY-SER.
  DOMAIN      386    397       POLY-SER.
  DOMAIN      442    447       POLY-PRO.
  DOMAIN      479    483       POLY-HIS.
  DOMAIN      484    497       POLY-GLN.
  DOMAIN      504    507       POLY-PRO.
  DOMAIN      564    574       POLY-SER.
  DOMAIN      704    707       POLY-PRO.
  DOMAIN      802    815       ARG/ALA-RICH (MIXED CHARGE).
  DOMAIN      816    827       ARG/GLU-RICH (MIXED CHARGE).
  DOMAIN      925    934       ARG/GLU-RICH (MIXED CHARGE).
SwissProt human and rat
       1 mktrqnkdsm smrsgrkkea pgpreelrsr graspggvst sssdgkaeks rqtakkarve
       61 eastpkvnkq grseeisese seetnapkkt kteqelprpq spsdldsldg rslnddgssd
      121 prdidqdnrs tspsiyspgs vendsdsssg lsqgparpyh ppplfppspq ppdstprqpe
      181 asfephpsvt ptgyhapmep ptsrmfqapp gappphpqly pggtggvlsg ppmgpkggga
      241 assvggpngg kqhpppttpi svsssgasga pptkppttpv gggnlpsapp panfphvtpn
      301 lppppalrpl nnasasppgl gaqplpghlp spyamgqgmg glppgpekgp tlapsphslp
      361 passsapapp mrfpysssss ssaaasssss sssssaspfp asqalpsyph sfppptslsv
      421 snqppkytqp slpsqavwsq gppppppygr llansnahpg pfppstgaqs tahppvsthh
      481 hhhqqqqqqq qqqqqqqhhg nsgppppgaf phpleggssh hahpyamsps lgslrpyppg
      541 pahlppphsq vsysqagpng ppvssssnss sstsqgsypc shpspsqgpq gapypfppvp
      601 tvttssatls tviatvassp agyktasppg pppygkraps pgayktatpp gykpgsppsf
      661 rtgtppgyrg tsppagpgtf kpgsptvgpg plppagpsgl pslppppaap asgpplsatq
      721 ikqepaeeye tpespvppar spspppkvvd vpshasqsar fnkhldrgfn scarsdlyfv
      781 plegsklakk radlvekvrr eaeqrareek ererererek ererekerel ersvklaqeg
      841 rapvecpslg pvphrppfep gsavatvppy lgpdtpalrt lseyarphvm spgnrnhpfy
      901 vplgavdpgl lgynvpalys sdpaarerer earerdlrdr lkpgfevkps eleplhgvpg
      961 pgldpfprhg glalqpgppg lhpfpfhpsl gplererlal aagpalrpdm syaerlaaer
     1021 qhaervaglg ndplarlqml nvtphhhqhs hihshlhlhq qdaihaasas vhplidplas
     1081 gshltripyp agtlpnpllp hplhenevlr hqlfaapyrd lpaslsapms aahqlqamha
     1141 qsaelqrlal eqqqwlhahh plhsvplpaq edyyshlkke sdkpl

Hereditary dentatorubral-pallidoluysian atrophy: detection of widespread ubiquitinated neuronal and glial intranuclear inclusions in the brain.

Acta Neuropathol (Berl) 1998 Dec;96(6):547-52 
Hayashi Y, Kakita A, Yamada M,  Tsuji S, Takahashi H
We examined the brains and spinal cords of seven patients with clinicopathologically and genetically confirmed hereditary dentatorubral-pallidoluysian atrophy (DRPLA) using an antibody against ubiquitin, and found small, round immunoreactive intranuclear inclusions in both neurons and glial cells in various brain regions.

Ubiquitinated neuronal intranuclear inclusions (uNIIs) were consistently found in the striatum, the pontine nuclei, the inferior olivary complex, the cerebellar cortex and the dentate nucleus. Ubiquitinated glial intranuclear inclusions (uGIIs) were found less frequently than uNIIs. Most of the inclusion-bearing nuclei were of an astrocytic nature.

Immunostaining with an antibody against DRPLA protein revealed similar immunoreactive neuronal and glial intranuclear inclusions, but in much smaller in numbers compared with uNIIs and uGIIs. Electron microscopy showed that such inclusions were composed of granular and filamentous structures. These findings strongly suggest that, in DRPLA, the occurrence of uNIIs and uGIIs is directly related to the causative gene abnormality (an expanded CAG repeat encoding polyglutamine), that neurons are affected much more widely than previously recognized and that glial cells are also involved in the disease process.

Hereditary dentatorubral-pallidoluysian atrophy: ubiquitinated filamentous inclusions in the cerebellar dentate nucleus neurons.

Acta Neuropathol (Berl) 1998 May;95(5):479-82 
Hayashi Y, Kakita A, Yamada M, Egawa S, Oyanagi S, Naito H, Tsuji S, Takahashi H
We examined the cerebellar dentate nucleus (CDN) in 16 patients with hereditary dentatorubral-pallidoluysian atrophy (DRPLA), one of the neurodegenerative diseases caused by expansion of a CAG repeat encoding a polyglutamine tract in the disease protein. In all patients, some CDN neurons were found to contain ubiquitinated filamentous inclusions in their cytoplasm.

On hematoxylin and eosin preparations, these filamentous inclusions were eosinophilic, basophilic or amphophilic, and were often found in areas of pale cytoplasm. Electron microscopy revealed that they consisted of bundles of filaments that were somewhat thicker than neurofilaments. These features of the present inclusions were indistinguishable from those of skein-like inclusions (SLI) previously described in the lower motor neurons in sporadic amyotrophic lateral sclerosis. We conclude that SLI can also occur in the CDN in DRPLA and believe that they reflect a characteristic pathological process in this disease.

Suppression of aggregate formation and apoptosis by transglutaminase inhibitors in cells expressing truncated DRPLA protein with an expanded polyglutamine stretch.

Nat Genet 1998 Feb;18(2):111-7 
Igarashi S, Koide R, ..Takahashi H, Tsuji S
To elucidate the molecular mechanisms whereby expanded polyglutamine stretches elicit a gain of toxic function, we expressed full-length and truncated DRPLA (dentatorubral-pallidoluysian atrophy) cDNAs with or without expanded CAG repeats in COS-7 cells. We found that truncated DRPLA proteins containing an expanded polyglutamine stretch form filamentous peri- and intranuclear aggregates and undergo apoptosis. The apoptotic cell death was partially suppressed by the transglutaminase inhibitors cystamine and monodansyl cadaverine (but not putrescine), suggesting involvement of a transglutaminase reaction and providing a potential basis for the development of therapeutic measures for CAG-repeat expansion diseases.

(To be continue: two key references are on order:
Kanazawa I Neurogenetics 1998 Dec;2(1):1-17 -- review of Japanese literature
Sisodia, S Cell 95: 1-4, 1998) -- Nuclear inclusions in glutamine repeat disorders: are they pernicious, coincidental, or beneficial?

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