Glutamine repeats and inherited neurodegenerative diseases: molecular aspects

Several dominantly inherited, late onset, neurodegenerative diseases are due to expansion of CAG repeats, leading to expansion of glutamine repeats in the affected proteins. These proteins are of very different sizes and, with one exception, show no sequence homology to known proteins or to each other; their functions are unknown. In some, the glutamine repeat starts near the N-terminus, in another near the middle and in another near the C-terminus, but regardless of these differences, no disease has been observed in individuals with fewer than 37 repeats, and absence of disease has never been found in those with more than 41 repeats. Protein constructs with more than 41 repeats are toxic to E. coli and to CHO cells in culture, and they elicit ataxia in transgenic mice. These observations argue in favour of a distinct change of structure associated with elongation beyond 37–41 glutamine repeats. The review describes experiments designed to find out what these structures might be and how they could influence the properties of the proteins of which they form part. Poly- -glutamines form pleated sheets of β-strands held together by hydrogen bonds between their amides. Incorporation of glutamine repeats into a small protein of known structure made it associate irreversibly into oligomers. That association took place during the folding of the protein molecules and led to their becoming firmly interlocked by either strand- or domain-swapping. Thermodynamic considerations suggest that elongation of glutamine repeats beyond a certain length may lead to a phase change from random coils to hydrogen-bonded hairpins. Possible mechanisms of expansion of CAG repeats are discussed in the light of looped DNA model structures.     

Insight into role of selection in the evolution of polyglutamine tracts in humans

Glutamine tandem repeats are common in eukaryotic proteins. Although some studies have proposed that replication slippage plays an important role in shaping these repeats, the role of natural selection in glutamine tandem repeat evolution is somewhat unclear. In this study, we identified all of the glutamine tandem repeats containing four or more glutamines in human proteins and then estimated the nonsynonymous (d(N)) and synonymous (d(S)) substitution rates for the regions flanking the glutamine tandem repeats and the proteins containing them. The results indicated that most of the proteins containing polyglutamine (polyQ) tracts of four or more glutamines have undergone purifying selection, and that the purifying selection for the regions flanking the repeats is weaker. Additionally, we observed that the conserved repeats were under stronger selection constraints than the nonconserved repeats. Interestingly, we found that there was a higher level of purifying selection for the regions flanking the polyQ tracts encoded by pure CAG codons compared with those encoded by mixed codons. Based on our findings, we propose that selection has played a more important role than was previously speculated in constraining the expansion of polyQ tracts encoded by pure codons.     

LIV-1 breast cancer protein belongs to new family of histidine-rich membrane proteins with potential to control intracellular Zn2+ homeostasis

Investigation of the protein product of the oestrogen-regulated gene LIV-1, implicated in metastatic breast cancer, has revealed 10 protein sequences of unknown function that belong to a new family with potential to control intracellular Zn2+ homeostasis. Sequence alignment highlights the similarity in transmembrane domains and extramembrane charged residues, indicating potential ion-transport ability. This family has a novel highly conserved motif of 66 residues, including a transmembrane domain and a catalytic zinc-binding sequence of zinc metalloproteases, containing conserved (indicated in bold type) proline and glutamine residues, HEXPHEXGD. These proteins contain more plentiful histidine-rich repeats than zinc transporters, suggesting an ability to bind or transport zinc across membranes. I propose that these 11 proteins form a new family with the potential to control intracellular Zn2+ homeostasis.     

A plant gene encoding a Myb-like protein that binds telomeric GGTTTAG repeats in vitro

A gene (AtTRP1) encoding a telomeric repeat-binding protein has been isolated from Arabidopsis thaliana. AtTRP1 is a single copy gene located on chromosome 5 of A. thaliana. The protein AtTRP1 encoded by this gene is not only homologous to the Myb DNA-binding motifs of other telomere-binding proteins but also is similar to several initiator-binding proteins in plants. Gel retardation assay revealed that the 115 residues on the C terminus of this protein, including the Myb motif, are sufficient for binding to the double-stranded plant telomeric sequence. The isolated DNA-binding domain of AtTRP1 recognizes each telomeric repeat centered on the sequence GGTTTAG. The almost full-length protein of AtTRP1 does not form any complex at all with the DNA fragments carrying four or fewer GGTTTAG repeats. However, it forms a complex with the sequence (GGTTTAG)(8) more efficiently than with the sequence (GGTTTAG)(5). These data suggest that the minimum length of a telomeric DNA for AtTRP1 binding consists of five GGTTTAG repeats and that the optimal AtTRP1 binding may require eight or more GGTTTAG repeats. It also implies that this protein AtTRP1 may bind in vivo primarily to the ends of plant chromosomes, which consist of long stretches of telomeric repeats.     

A census of protein repeats.

In this study, we analyzed all known protein sequences for repeating amino acid segments. Although duplicated sequence segments occur in 14 % of all proteins, eukaryotic proteins are three times more likely to have internal repeats than prokaryotic proteins. After clustering the repetitive sequence segments into families, we find repeats from eukaryotic proteins have little similarity with prokaryotic repeats, suggesting most repeats arose after the prokaryotic and eukaryotic lineages diverged. Consequently, protein classes with the highest incidence of repetitive sequences perform functions unique to eukaryotes. The frequency distribution of the repeating units shows only weak length dependence, implicating recombination rather than duplex melting or DNA hairpin formation as the limiting mechanism underlying repeat formation. The mechanism favors additional repeats once an initial duplication has been incorporated. Finally, we show that repetitive sequences are favored that contain small and relatively water-soluble residues. We propose that error-prone repeat expansion allows repetitive proteins to evolve more quickly than non-repeat-containing proteins.     

Evidence for a recruitment and sequestration mechanism in Huntington's disease.

Polyglutamine (polyQ) extension in the coding sequence of mutant huntingtin causes neuronal degeneration associated with the formation of insoluble polyQ aggregates in Huntington's disease. We constructed an array of CAG/CAA triplet repeats, coding for a range of 25-300 glutamine residues, which was used to generate expression constructs with minimal flanking sequence. Normal-length (25 glutamine residues) polyQ did not aggregate when transfected alone. Remarkably, when co-transfected with extended (100-300 glutamine residues) polyQ tracts, normal-length polyQ-containing peptides were trapped in insoluble detergent-resistant aggregates. Aggregates formed in the cytoplasm but were visible in the nucleus only when a strong nuclear localization signal was present. Intermolecular interactions between polyQ tracts mediated the localization of heterogeneous aggregates into the nucleolus by nucleolin protein. Our results suggest that extended polyQ can interact with cellular polyQ-containing proteins, transport them to ectopic cellular locations, and form heterogeneous polyQ aggregates. We provide evidence for a recruitment mechanism for pathogenesis in the polyQ neurodegenerative disorders. In susceptible cells, extended polyQ tracts in huntingtin might interact with and sequester or deplete certain endogenous polyQ-containing cellular proteins.     

Aggregation and toxicity of the proteins with polyQ repeats

Expansion of CAG triplet repeats is a cause of at least nine late-onset neurodegenerative disorders. The mutation manifests itself as a long stretch of glutamine repeats. The number of approximately 38 repeats is usually a threshold at which the disease develops and the longer the polyglutamine tract, the earlier the onset of disease. A common feature of these disorders is the presence of protein aggregates which are believed to be formed either by the formation of hydrogen bonds between amide residues or through the action of the enzyme transglutaminase. Mutated proteins may cause neurodegeneration by sequestering vital cellular proteins, inhibiting proteasomal system or by inducing apoptosis. It has been proved that molecular chaperones may block the negative effects of expression of mutated genes and for this reason they are a promising object for various therapeutic research.     

Negatively Charged Disordered Regions are Prevalent and Functionally Important Across Proteomes

Intrinsically disordered regions (IDRs) of proteins are often characterized by a high fraction of charged residues, but differ in their overall net charge and in the organization of the charged residues. The function-encoding information stored via IDR charge composition and organization remains elusive. Here, we aim to decipher the sequence–function relationship in IDRs by presenting a comprehensive bioinformatic analysis of the charge properties of IDRs in the human, mouse, and yeast proteomes. About 50% of the proteins comprise at least a single IDR, which is either positively or negatively charged. Highly negatively charged IDRs are longer and possess greater net charge per residue compared with highly positively charged IDRs. A striking difference between positively and negatively charged IDRs is the characteristics of the repeated units, specifically, of consecutive Lys or Arg residues (K/R repeats) and Asp or Glu (D/E repeats) residues. D/E repeats are found to be about five times longer than K/R repeats, with the longest found containing 49 residues. Long stretches of consecutive D and E are found to be more prevalent in nucleic acid-related proteins. They are less common in prokaryotes, and in eukaryotes their abundance increases with genome size. The functional role of D/E repeats and the profound differences between them and K/R repeats are discussed.     

The glutamine residues reactive in transglutaminase-catalyzed cross-linking of involucrin

The protein involucrin, synthesized by human keratinocytes, contains 585 amino acids, largely in the form of 10 amino acid repeats, each containing glutamines in 3 conserved positions. Involucrin is a substrate for the keratinocyte transglutaminase and is labeled by the cosubstrate amine, glycine ethyl ester. Study of tryptic peptides of involucrin shows that a single glutamine (residue 496), located 89 residues from the C-terminal end, is preferentially labeled by the enzyme. Additional glutamine residues become reactive when the molecule is fragmented. The C-terminal end, isolated as a cyanogen bromide fragment of 275 residues, is labeled equally at 2 glutamine residues. The polypeptide containing residues 148 to 280 accepts practically no amine while in intact involucrin but as a free fragment is labeled at multiple glutamine residues. It is concluded that the C-terminal and N-terminal ends of the protein are directive influences in that they suppress the reactivity of a number of glutamine residues in the intact molecule, leaving one glutamine highly preferred by the transglutaminase.     

Analysis of ankyrin repeats reveals how a single point mutation in RFXANK results in bare lymphocyte syndrome.

Ankyrin repeats are well-known structural modules that mediate interactions between a wide spectrum of proteins. The regulatory factor X with ankyrin repeats (RFXANK) is a subunit of a tripartite RFX complex that assembles on promoters of major histocompatibility complex class II (MHC II) genes. Although it is known that RFXANK plays a central role in the nucleation of RFX, it was not clear how its ankyrin repeats mediate the interactions within the complex and with other proteins. To answer this question, we modeled the RFXANK protein and determined the variable residues of the ankyrin repeats that should contact other proteins. Site-directed alanine mutagenesis of these residues together with in vitro and in vivo binding studies elucidated how RFXAP and CIITA, which simultaneously interact with RFXANK in vivo, bind to two opposite faces of its ankyrin repeats. Moreover, the binding of RFXAP requires two separate surfaces on RFXANK. One of them, which is located in the ankyrin groove, is severely affected in the FZA patient with the bare lymphocyte syndrome. This genetic disease blocks the expression of MHC II molecules on the surface of B cells. By pinpointing the interacting residues of the ankyrin repeats of RFXANK, the mechanism of this subtype of severe combined immunodeficiency was revealed.     

Evidence for polyproline II helical structure in short polyglutamine tracts

Nine neurodegenerative diseases, including Huntington's disease, are associated with the aggregation of proteins containing expanded polyglutamine sequences. The end result of polyglutamine aggregation is a beta-sheet-rich deposit. There exists evidence that an important intermediate in the aggregation process involves intramolecular beta-hairpin structures. However, little is known about the starting state, monomeric polyglutamine. Most experimental studies of monomeric polyglutamine have concluded that the backbone is completely disordered. However, such studies are hampered by the inherent tendency for polyglutamine to aggregate. A recent computational study suggested that the glutamine residues in polyglutamine tracts have a significant propensity to adopt the left-handed polyproline II (P(II)) helical conformation. In this work, we use NMR spectroscopy to demonstrate that glutamine residues possess a high propensity to adopt the P(II) conformation. We present circular dichroism spectra that indicate the presence of significant amounts of P(II) helical structure in short glutamine tracts. These data demonstrate that the propensity to adopt the P(II) structure is retained for glutamine repeats of up to at least 15 residues. Although other structures, such as alpha-helices and beta-sheets, become possible at greater lengths, our data indicate that glutamine residues in monomeric polyglutamine have a significant propensity to adopt the P(II) structure, although not necessarily in long contiguous helical stretches. We note that we have no evidence to suggest that the observed P(II) helical structure is a precursor to polyglutamine aggregation. Nonetheless, increased understanding of monomeric polyglutamine structures will aid our understanding of the aggregation process.     

Studies of the zein-like α-prolamins based on an analysis of amino acid sequences: Implications for their evolution and three-dimensional structure

α-Prolamins are the major seed storage proteins of species of the grass tribe Andropogonea. They are unusually rich in glutamine, proline, alanine, and leucine residues and their sequences show a series of tandem repeats presumed to be the result of multiple intragenic duplication. Two new sequences of α-prolamin clones from Coix (pBCX25.12 and pBCX25.10) are compared with similar clones from maize and Sorghum in order to investigate evolutionary relationships between the repeat motifs and to propose a schematic model for their three-dimensional structure based on hydrophobic membrane-helix propensities and helical “wheels.” A scheme is proposed for the most recent events in the evolution of the central part of the molecule (repeats 3 to 8) which involves two partial intragenic duplications and in which contemporary odd-numbered and even-numbered repeats arise from common ancestors, respectively. Each pair of repeats is proposed to form an antiparallel α-helical hairpin and that the helices of the molecule as a whole are arranged on a hexagonal net. The majority of helices show six faces of alternating hydrophobic and polar residues, which give rise to intersticial holes around each helix which alternate in chemical character. The model is consistent with proteins which contain different numbers of repeats, with oligomerization and with the dense packaging of α-prolamins within the protein body of the seed endosperm. © 1993 Wiley-Liss, Inc.     

Eukaryotic proteasomes cannot digest polyglutamine sequences and release them during degradation of polyglutamine-containing proteins

Long glutamine sequences (polyQ) occur in many cell proteins, and several neurodegenerative diseases result from expansion of these sequences. PolyQ-containing proteins are degraded by proteasomes, whose three active sites prefer to cleave after hydrophobic, basic, or acidic residues. We tested whether these particles can digest a polyQ chain. Eukaryotic 26S and 20S proteasomes failed to cut within stretches of 9-29Q residues in peptides. While digesting a myoglobin Q(35) fusion protein, the proteasomes spared the polyQ sequence. In contrast, archaeal proteasomes, whose 14 active sites are less specific, rapidly digested such polyQ repeats. Therefore, when degrading polyQ proteins, eukaryotic proteasomes must release aggregation-prone polyQ-containing fragments for further hydrolysis by unidentified peptidases. In polyQ diseases, such polyQ sequences (38-300Qs) exceed the lengths of normal proteasome products (2-25 residues). Occasional failure of these long undegradable sequences to exit may interfere with proteasome function and help explain why longer polyQ expansions promote early disease onset.     

Identification and analysis of novel amino acid sequence repeats and domains in Pyrobaculum aerophilum using computational tools

We have identified four repeats and five domains that are novel in proteins encoded by the Pyrobaculum aerophilum str. IM2 proteome using automated in silico methods. A "repeat" corresponds to a region comprising less than 55 amino acid residues that occurs more than once in the protein sequence and sometimes present in tandem. A "domain" corresponds to a conserved region comprising greater than 55 amino acid residues and may be present as single or multiple copies in the protein sequence. These correspond to (1) 85 amino acid residues AAG domain, (2) 72 amino acid residues GFGN domain, (3) 43 amino acid residues KGG repeat, (4) 25 amino acid residues RWE repeat, (5) 25 amino acid residues RID repeat, (6) 108 amino acid residues NDFA domain, (7) 140 amino acid residues VxY domain, (8) 35 amino acid residues LLPN repeat and (9) 98 amino acid residues GxY domain. A repeat or domain is characterized by specific conserved sequence motifs. We discuss the presence of these repeats and domains in proteins from other genomes and their probable secondary structure.     

Specific 33-residue repeat(s) of erythrocyte ankyrin associate with the anion exchanger.

Erythrocyte ankyrin contains an 89-kDa domain (residues 2-827) comprised almost entirely of 22 tandem repeats of 33 amino acids which are responsible for the high affinity interaction of ankyrin with the anion exchanger (Davis, L., and Bennett, V. (1990) J. Biol. Chem. 265, 10589-10596). The question of whether the repeats are equivalent with respect to binding to the anion exchanger was addressed using defined regions of erythrocyte and brain ankyrins expressed in bacteria. The conclusion is that the repeats are not interchangeable and that the 44 residues from 722 to 765 are essential for high affinity binding between erythrocyte ankyrin and the anion exchanger. Residues 348-765 were active whereas a polypeptide of the same size (residues 305-721) but missing the 44 residues was not active. The difference between the active and inactive polypeptides was not caused by the degree of folding based on circular dichroism spectra. The 44 residues from 722 to 765 were not sufficient for binding since deletions of residues from 348 to 568 resulted in a 10-fold loss of activity. However, the role of residues 348-568 may be at the level of folding rather than a direct contact since the deleted sequences were not active in the absence of 722-765 and since circular dichroism spectra revealed significant loss of structure in the smaller polypeptides. Further evidence that the 33-residue repeats are not equivalent in ability to bind to the anion exchanger is that a region of human brain ankyrin containing 18 33-residue repeats with 67% overall sequence identity to erythrocyte ankyrin was 8-fold less active than a region of erythrocyte ankyrin containing only 12 repeats. The fact that the anion exchanger binds to certain repeats suggests that the other 33-amino acid repeats could interact with proteins distinct from the anion exchanger and provide ankyrin with the potential for considerable diversity in association with membrane proteins as well as cytoplasmic proteins. Tubulin was identified as one example of a protein that can interact with ankyrin repeats that are not recognized by the anion exchanger.     

Molecular Evolution of Amelogenin in Mammals

An evolutionary analysis of mammalian amelogenin, the major protein of forming enamel, was conducted by comparison of 26 sequences (including 14 new ones) representative of the main mammalian lineages. Amelogenin shows highly conserved residues in the hydrophilic N- and C-terminal regions. The central hydrophobic region (most of exon 6) is more variable, but it has conserved a high amount of proline and glutamine located in triplets, PXQ, indicating that these residues play an important role. This region evolves more rapidly, and is less constrained, than the other well-conserved regions, which are subjected to strong constraints. The comparison of the substitution rates in relation to the CpG richness confirmed that the highly conserved regions are subjected to strong selective pressures. The amino acids located at important sites and the residues known to lead to amelogenesis imperfecta when substituted were present in all sequences examined. Evolutionary analysis of the variable region of exon 6 points to a particular zone, rich in either amino acid insertion or deletion. We consider this region a hot spot of mutation for the mammalian amelogenin. In this region, numerous triplet repeats (PXQ) have been inserted recently and independently in five lineages, while most of the hydrophobic exon 6 region probably had its origin in several rounds of triplet insertions, early in vertebrate evolution. The putative ancestral DNA sequence of the mammalian amelogenin was calculated using a maximum likelihood approach. The putative ancestral protein was composed of 177 residues. It already contained all important amino acid positions known to date, its hydrophobic variable region was rich in proline and glutamine, and it contained triplet repeats PXQ as in the modern sequences.Reviewing Editor: Dr. Cecilia Saccone     

Global analysis of tandem aromatic octapeptide repeats: the significance of the aromatic-glycine motif

MOTIVATION: Tandem peptide repeats play a key role in self-assembly and aggregation processes. A notable example is the occurrence of tandem peptide repeats in prionic proteins and their role in the aggregation process that leads to the formation of the prion. One of the structural characteristics that is evident from the comparison of mammalian and yeast prion proteins is the presence of aromatic residues in their tandem repeats. These residues are accompanied by glycine residues before and/or after the aromatic amino acid. Such aromatic-glycine conjugates are also present in the tandem repeats of the large family of the bacterial ice nucleation proteins. To study the significance of such aromatic-glycine occurrences, a global analysis of all the aromatic octapeptide repeats in the Swiss-Prot and TrEMBL databases was conducted. The search pattern was formulated to compare the number of conjugates of each of the 20 natural amino acids before or after the different aromatic residues. RESULTS: The presence of aromatic-glycine conjugates appears to be significantly higher than aromatic conjugates to any other amino acid. Furthermore, all the six various combination of glycine occurrences before or after the three aromatic residues are present. No such pattern was observed for any other amino acid. The significance of the findings is being discussed in the context of the physicochemical properties of aromatic-glycine conjugates and its possible role in the facilitation of aggregates formation.