Autosomal dominant ataxias are a genetically heterogeneous group of disorders for which spinocerebellar ataxia (SCA) loci on chromosomes 6p, 12q, 14q and 16q have been reported. We have examined 170 individuals (56 of whom were affected) from a previously unreported ten-generation kindred with a dominant ataxia that is clinically and genetically distinct from those previously mapped. The family has two major branches which both descend from the paternal grandparents of President Abraham Lincoln. Among those examined, 56 individuals have a generally non-life threatening cerebellar ataxia. Disease onset varies from 10-68 years and anticipation is evident. We have mapped this gene, spinocerebellar ataxia type 5 (SCA5), to the centromeric region of chromosome 11.

To understand the causes of CAG repeat tract changes that occur in the passage of human disease alleles, we are studying the effect of replication and repair mutations on CAG repeat tracts embedded in a yeast chromosome. In this report, we examine the effect of a mutation in the RTH1/RAD27 gene encoding a deoxyribonuclease needed for removal of excess nucleotides at the 5'-end of Okazaki fragments. Deletion of the RTH1/RAD27 gene has two effects on CAG tracts. First, the rth1/rad27 mutation destabilizes CAG tracts. Second, although most tract length changes in wild-type yeast cells are tract contractions, approximately half of the changes that occur as a result of the rth1/rad27 mutation are expansions of one or more repeat units. These results support the hypothesis that tract expansions that occur during passage of human disease alleles bearing expanded CAG tracts result from excess DNA synthesis on the lagging strand of replication.

To examine the chromosomal stability of repetitions of the trinucleotide CAG, we have cloned CAG repeat tracts onto the 3 end of the Saccharomyces cerevisiae ADE2 gene and placed the appended gene into the ARO2 locus of chromosome VII. Examination of chromosomal DNA from sibling colonies arising from clonal expansion of strains harboring repeat tracts showed that repeat tracts often change in length. Most changes in tract length are decreases, but rare increases also occur. Longer tracts are more unstable than smaller tracts. The most unstable tracts, of 80 to 90 repeats, undergo changes at rates as high as 3 ؋ 10 ؊2 changes per cell per generation. To examine whether repeat orientation or adjacent sequences alter repeat stability, we constructed strains with repeat tracts in both orientations, either with or without sequences 5 to ADE2 harboring an autonomously replicating sequence (ARS; replication origin). When CAG is in the ADE2 coding strand of strains harboring the ARS, the repeat tract is relatively stable regardless of the orientation of ADE2. When CTG is in the ADE2 coding strand of strains harboring the ARS, the repeat tract is relatively unstable regardless of the orientation of ADE2. Removal of the ARS as well as other sequences adjacent to the 5 end of ADE2 alters the orientation dependence such that stability now depends on the orientation of ADE2 in the chromosome. These results suggest that the proximity of an ARS or another sequence has a profound effect on repeat stability.Expansions of repetitions of the trinucleotide CAG are the cause of a number of human inherited, dominant neurological and neuromuscular diseases, including Huntington's disease (14), two forms of spinocerebellar ataxia (type 1 and MachadoJoseph disease) (17,20), and myotonic dystrophy (3,7,19). Although CAG trinucleotide repetitions are present in normal alleles of the genes giving rise to these disorders, mutant alleles have tracts which are longer than those within the normal range. The long tracts within disease alleles are unstable in that children often inherit a repeat length different from that found in their affected parent. The instability most likely reflects replicative errors which occur either during the meiotic division of gametogenesis or during the mitotic divisions preceding it.The underlying cause of the instability is thought to reflect the ability of CAG repeats to form palindrome-like structures (8,22). Such structures may present problems to the replication fork as it passes through them. One study using small CAG repeats embedded in palindromes carried on phage lambda showed that they were inhibitory to plaque formation (6). Studies with Escherichia coli have also shown that CAG repeats undergo both contractions and expansions when propagated in a bacterial plasmid (16).We decided to examine the stability of CAG repeats in Saccharomyces cerevisiae because the chromatin structure and chromosomal replication of this simple eukaryote have many similarities to the chromosomal mechanics of more complex euka...

The genes that encode the a and 1a subunits of protocatechuate 3,4-dioxygenase (3,4-PCD [EC 1.13.11.31) were cloned from a Pseudomonas putida (formerly P. aenuginosa) (ATCC 23975) genomic library prepared in A phage. Plaques were screened by hybridization with degenerate oligonucleotides designed using known amino acid sequences. A 1.5-kb SnaI fragment from a 15-kb primary clone was subcloned, sequenced, and shown to contain two successive open reading frames, designatedpcaH andpcaG, corresponding to the c and a subunits, respectively, of 3,4-PCD. The amino acid sequences deduced from pcaHG matched the chemically determined sequence of 3,4-PCD in all except three positions. Cloning ofpcaHG into broad-host-range expression vector pKMY319 allowed high levels of expression in P. putida strains, as well as in Proteus mirabilis after specific induction of the plasmid-encoded nahG promoter with salicylate. The recombinant enzyme was purified and crystallized from P. mirabilis, which lacks an endogenous 3,4-PCD. The physical, spectrmscopic, and kinetic properties of the recombinant enzyme were indistinguishable from those of the wild-type enzyme. Moreover, the same transient enzyme intermediates were formed during the catalytic cycle. These studies establish the methodology which will allow mechanistic investigations to be pursued through site-directed mutagenesis of P. putida 3,4-PCD, the only aromatic ring-cleaving dioxygenase for which the three-dimensional stmcture is known.The catabolic pathways for bacterial degradation of aromatic compounds converge, in most instances, on a small group of single-ring aromatic compounds. This group includes protocatechuate, catechol, gentisate, and a few other, similar compounds (9,11,12,22,34). The aromatic ring of these compounds is opened during reactions catalyzed by dioxygenase enzymes in which both atoms of oxygen from 02 are incorporated into the substrate. These enzymes usually contain nonheme iron stabilized in either the Fe(II) or Fe(III) oxidation state (34,37,43,44 (Fig. 1). This pathway catalyzes the conversion of protocatechuate (PCA) to P-ketoadipyl coenzyme A and subsequently to citric acid cycle intermediates.3,4-PCD has been isolated from many widely divergent bacteria (6, 13, 18, 34, 56), and its elaboration has been used as a taxonomic characteristic in classifying bacterial strains (53). All of the well-characterized 3,4-PCDs are composed of equimolar amounts of two nonidentical subunits, termed a and 3, organized as a1 protomers. The number of a3 protomers present in the enzymes from different bacteria is quite variable, with a known range of 3 to 12 (34). Soon after this general quatemary structure was recognized (33, 60), each of the subunits of P. putida 3,4-PCD (classified as P. * Corresponding author. aeruginosa at the time) was chemically sequenced (26,27,29,30). Subsequently, we reported the crystal structure of this enzyme (38). This is the only structure of a mononuclear nonheme iron-containing dioxygenase known. On the basis of this crystal stru...

To examine the genetic factors that affect the stability of disease-associated trinucleotide repeats, we have assessed the stability of CAG repeats in yeast strains with mutations in the mismatch repair system. We have found that both pms1 and msh2 mutations destabilize repeat tracts. Destabilization is evidenced both by the increased frequency of repeat length changes and in the pattern of changes that are observed. In wild-type cells repeats are relatively stable when CAG serves as the lagging strand template but relatively unstable when CTG serves as the lagging strand template. Large contractions in repeat length are the most common change. In pms1 and msh2 mutants the relatively stable tracts incur more tract length changes. In addition, many small deletions and some small additions, most often of one repeat unit, are frequent in repeats of the stable orientation. These small changes also are seen as a new class of events that occur in repeats in the unstable orientation. The results show that in yeast the mismatch repair system prevents small changes from occurring but cannot prevent larger changes from occurring.