We present the structure and nucleotide se- quence of a gene encoding the human epidermal 67-kDa keratin. Three genomic clones were isolated from a X Charon 4A human genomic library by hybridization to a specific cDNA probe. One clone of 12.3 kilobase pairs was shown by R-loop, DNA sequence, and primer-extension analyses to encode an entire gene of about 6.25 kilobase pairs. Of eight identified introns, seven are located' within the region that encodes the central coiled-coil a-helical domain of the protein. Except for one intron located at the end of the region encoding this domain, these do not delineate apparent structural subdomains. The positions of five of the introns exactly coincide with the positions of introns previously reported in the hamster gene for the intermediate filament protein vimentin (Quax, W., Egberts, W. V., Hendricks, W., Quax- Jeuken, Y. & Bloemandal, H. (1983) Cell 35, 215-233). These findings suggest that the human 67-kDa keratin and vimentin genes arose from a common ancestral gene.

To the Editor: Epidermolytic hyperkeratosis (EHK, MIM113800) is a rare autosomal dominant genodermatosis, although many cases occur sporadically as new mutations. It is characterized by blistering and erythema at birth, and development of hyperkeratosis with increasing age. Ultrastructural analyses reveal clumping of the keratin intermediate ®laments (KIF) within suprabasal keratinocytes of the epidermis. Previous studies have demonstrated that EHK is caused by mutations in the genes encoding the keratin 1 (K1) or the K10 proteins (Cheng et al, 1992; Chipev et al, 1992; Rothnagel et al, 1992; McLean and Lane, 1995; Korge and Krieg, 1996). Clinically, EHK has been divided into the palmoplantar type (palm sole, PS) and the nonpalmoplantar type (nonpalm sole), and each type has been subdivided into three subtypes according to the nature and severity of scaling (DiGiovanna and Bale, 1994; DiGiovanna et al, 1998). The mutation/clinical phenotype correlations available suggest that the PS and nonpalm sole types are due primarily to mutations in the KRT1 and KRT10 genes, respectively. The more severe PS-1 and PS-2 cases are caused by mutations that change sequences encoding the beginning of the 1A rod domain segment, or towards the end of the 2B rod domain segment (Yang et al, 1994, 1999; Arin et al, 1999, 2000; Cserhalmi-Friedman et al, 2000). In this study, we describe two separate cases of EHK, classi®ed as PS-2, caused by two different novel mutations: Asn187Lys located in exon 1, which is the 8th residue of the 1A rod domain segment; and Leu475Pro located in exon 7, which is the 10th residue from the end of 2B rod domain segment of K1. Two unrelated children, EHK-SJ and EHK-AS, have diffuse thick keratoderma on both palms and soles as well as moderate scaly patches on their entire bodies, but the morphology of their hair, nails, and teeth are normal. Light microscopic examination revealed vacuolar degeneration in the spinous and granular layers. As both sets of parents were normal, these represent sporadic cases. Both were classi®ed as having the PS-2 form of EHK. DNA was extracted (Chipev et al, 1994; Yang et al, 1994) from both the affected and unaffected family members, and from 50 unrelated normal individuals used as controls. Following sequencing of all exons of the KRT1 and KRT10 genes, patient EHK-SJ had a single nucleotide substitution on one allele of codon 187 in exon 1 of the KRT1 gene (AAC to AAA) so that the sequence gel of Fig 1(a) shows both the wild-type C and a mutant A at nucleotide position 564 (GenBank NM-006121; relative to the translation start point) in the proband. The mutant allele encodes a lysine residue instead of a wild-type asparagine residue in position 8 of the 1A rod domain segment. As this nucleotide substitution does not create or destroy a restriction enzyme site, we performed a polymerase chain reaction ampli®cation of a speci®c allele assay (Yang et al, 1994, 1999). As the parents and 50 other unrelated normal controls do not have this substitution, it represents a new mutation. The same mutation in the KRT2e gene was reported in ichthyosis bullosa of Siemens (Whittock et al, 2001). In patient EHK-AS, we found a single nucleotide substitution on one allele of codon 475 in exon 7 of the KRT1 gene (CTG to CCG) so that the sequence gel of Fig 1(b) shows both a wild-type T and a mutant C at nucleotide position 1457 (GenBank NM-006121; relative to the translation start point) in the proband. The mutant allele encodes a proline residue instead of a wild-type leucine residue in position 113 of 2B rod domain segment. This nucleotide substitution destroys an MspI restriction enzyme site (C/CGG), which was used to con®rm that the unaffected parents and 50 other unrelated unaffected persons do not posses this new mutation.

Epiplakin is a member of the plakin family with multiple copies of the plakin repeat domain (PRD). We studied the subcellular distribution and interactions of human epiplakin by immunostaining, overlay assays and RNAi knockdown. Epiplakin decorated the keratin intermediate filaments (IF) network and partially that of vimentin. In the binding assays, the repeat unit (PRD plus linker) showed strong binding and preferentially associated with assembled IF over keratin monomers. Epiplakin knockdown revealed disruption of IF networks in simple epithelial but not in epidermal cells. In rescue experiments, the repeat unit was necessary to prevent the collapse of IF networks in transient knockdown; however, it could only partially restore the keratin but not the vimentin IF network in stably knocked down HeLa cells. We suggest that epiplakin is a cytolinker involved in maintaining the integrity of IF networks in simple epithelial cells. Furthermore, we observed an increase of epiplakin expression in keratinocytes after the calcium switch, suggesting the involvement of epiplakin in the process of keratinocyte differentiation.

Transglutaminase activity was found to be present in highly purified non-synaptosomal rat brain mitochondria. A 78-kDa protein in these organelles was shown to be a transglutaminase 2 substrate, and incubation of a non-synaptosomal mitochondrial lysate with transglutaminase 2 yielded high-Mr proteins. The 78-kDa protein was identified as mitochondrial aconitase by MALDI-TOF analysis. Aconitase activity was decreased in a dose-dependent manner when non-synaptosomal rat brain mitochondria were incubated with transglutaminase 2. Transglutaminase activity is increased about 2-fold in the mitochondrial fraction of HD caudate. Moreover, Western blotting of the mitochondrial fraction revealed that most of the mitochondrial aconitase in HD caudate is present as high-Mr aggregates. Aconitase activity was previously shown to be decreased in Huntington disease (HD) caudate (a region severely damaged by the disease). The present findings suggest that an increase of transglutaminase activity in HD caudate may contribute to mitochondrial dysfunction by incorporating aconitase into inactive polymers.

Transglutaminase 3 (TGase 3) is a member of a family of Ca2+-dependent enzymes that catalyze covalent cross-linking reactions between proteins or peptides. TGase 3 isoform is widely expressed and is important for effective epithelial barrier formation in the assembly of the cell envelope. Among the nine TGase enzyme isoforms known in the human genome, only TGase 2 is known to bind and hydrolyze GTP to GDP; binding GTP inhibits its transamidation activity but allows it to function in signal transduction. Here we present biochemical and crystallographic evidence for the direct binding of GTP/GDP to the active TGase 3 enzyme, and we show that the TGase 3 enzyme undergoes a GTPase cycle. The crystal structures of active TGase 3 with guanosine 5% O-(thiotriphosphate) (GTPgammaS) and GDP were determined to 2.1 and 1.9 Angstrom resolution, respectively. These studies reveal for the first time the reciprocal actions of Ca2+ and GTP with respect to TGase 3 activity. GTPgammaS binding is coordinated with the replacement of a bound Ca2+ with Mg2+ and conformational rearrangements that together close a central channel to the active site. Hydrolysis of GTP to GDP results in two stable conformations, resembling both the GTP state and the nonnucleotide bound state, the latter of which allows substrate access to the active site.

The sequence of a protein chain determines both its conformation and its function in vivo. An attempt is made to gain an understanding of the classes of deformations that can arise in an important structural motif, the α‐helical coiled coil, as a consequence of mutations occurring in its underlying heptad substructure. In order to do so we consider the model structure of segment 1A in intermediate filaments and then investigate the structures arising from each of the 22 mutations observed in cytokeratin K5/K14 molecules that lead to variants of epidermolysis bullosa simplex. These are refined separately using a molecular dynamics protocol. The mutations often result in a significant distortion of the backbone over a turn or so of the α helix in either the chain itself or its constituent partner, leading to the likelihood of impaired chain aggregation and hence molecular assembly. One mutant (K14‐L143P; 1A‐28) gave rise to structural distortion along almost the entire length of segment 1A. The remaining structures showed less deformation, and normal‐looking intermediate filaments are likely in vivo. In addition, an identical mutation in the same position in each of the chains in the heterodimer did not necessarily give equivalent structural distortions. Although proline mutations frequently lead to the most severe structural deformations, a non‐proline substitution (K14‐R125S; 1A‐10) gave rise to the largest local structural disruption that was observed. Unexpectedly, mutations in positions a and d were not always of the greatest structural significance, although three in position a were shown by AGADIR to result in a significant increase in α‐helix stability. Proteins 2004. © 2004 Wiley‐Liss, Inc.

Epidermal-type Transglutaminase 3(TGase3) is a Ca2+-dependent enzyme involved in the cross-linking of structural proteins required in the assembly of the cell envelope. We have recently shown that calcium-activated TGase 3, like TGase 2, can bind, hydrolyze, and is inhibited by GTP despite lacking structural homology with other GTP-binding proteins. Here we report the crystal structure determined at 2.0 Angstrom resolution of TGase 3 in complex with GMP to elucidate the structural features required for nucleotide recognition. Binding affinities for various nucleotides were found by fluorescence displacement to be as follows: guanosine 5'-3-O-( thio) triphosphate (GTPgammaS) (0.4 muM), GTP (0.6 muM), GDP (1.0 muM), GMP (0.4 muM), and ATP (28.0 muM). Furthermore, we found that GMP binds as a reversible, noncompetitive inhibitor of TGase 3 transamidation activity, similar to GTPgammaS and GDP. A genetic algorithm similarity program ( GASP) approach ( virtual ligand screening) identified three compounds from the Lead Quest(TM) data base (Tripos Inc.) based on superimposition of GTPgammaS, GDP, and GMP guanine nucleotides from our crystal structures to generate the minimum align flexible fragment. These three were nucleotide analogs without a phosphate group containing the minimal binding motif for TGase 3 that includes a nucleoside recognition groove. Binding affinities were measured as follows: TP349915 (K-d = 4.1 muM), TP395289 (K-d = 38.5 muM), TP394305 (K-d = 1.0 mM). Remarkably, these compounds do not inhibit but instead activate TGase 3 transamidation by about 10-fold. These results suggest that the nucleotide binding pocket in TGase 3 may be exploited to either enhance or inhibit the enzymatic activity as required for different therapeutic approaches.

The accumulation of misfolded proteins in intracellular inclusions is a generic feature of neurodegenerative disorders. Although heavily ubiquitylated, the aggregated proteins are not degraded by the proteasomes. A possible reason for this phenomenon may be a modification of deposited proteins by transglutaminases forming gamma-glutamyl-epsilon-lysine (GGEL) cross-links between distinct proteins. Here, we show that the frequency of GGEL cross-links is an order of magnitude higher in Alzheimer's brain cortex than in age-matched or younger controls. This difference is due to the accumulation of GGEL cross-links in ubiquitin-immunopositive protein particles present in both Alzheimer's brains and those from aged individuals. The highly crosslinked protein aggregates show immunoreactivity to antibodies against tau and neurofilament proteins, and partially also to alpha-synuclein, indicating that these structures are inherent in Alzheimer's neurofibrillary tangles and Lewy bodies. Using mass sequence analysis, we identified the same six pairs of peptide sequences cross-linked in both senile and Alzheimer's specimens: Gln(31) and Gln(190) of HSP27 protein are cross-linked with Lys(29) and Lys(48) of ubiquitin and HSP27 therefore may cross-link two (poly) ubiquitin chains. One lysine residue of parkin and one of alpha-synuclein were also found to be cross-linked. The data suggest that cross-linking of (poly) ubiquitin moieties via HSP27 may have a role in the stabilization of the intraneuronal protein aggregates by interference with the proteasomal elimination of unfolded proteins.

Plakin family members envoplakin and periplakin have been shown to be part of the cornified cell envelope in terminally differentiating stratified squamous epithelia. In the present study, purified recombinant human envoplakin and periplakin were used to investigate their properties and interactions. We found that envoplakin was insoluble at physiological conditions in vitro, and co-assembly with periplakin was required for its solubility. Envoplakin and periplakin formed soluble complexes with equimolar stoichiometry. Chemical cross-linking revealed that the major soluble form of all periplakin constructs and of envoplakin/periplakin rod domains was a dimer, although co-assembly of the full-length proteins resulted in formation of higher order oligomers. Electron microscopy of rotary-shadowed periplakin demonstrated thin flexible molecules with an average contour length of 88 nm for the rod-plus-tail fragment, and immunolabeling EM confirmed the molecule as a parallel, in-register, dimer. Both periplakin and envoplakin/periplakin oligomers were able to bind synthetic lipid vesicles whose composition mimicked the cytoplasmic side of the plasma membrane of eukaryotic cells. This binding was dependent on anionic phospholipids and Ca(2+). These findings raise the possibility that envoplakin and periplakin bind to the plasma membrane upon elevation of intracellular [Ca(2+)] in differentiating keratinocytes, where they serve as a scaffold for cornified cell envelope assembly.

We have developed an assembly protocol for the intermediate filament (IF) protein vimentin based on a phosphate buffer system, which enables the dynamic formation of authentic IFs. The advantage of this physiological buffer is that analysis of the subunit interactions by chemical cross-linking of internal lysine residues becomes feasible. By this system, we have analyzed the potential interactions of the coiled-coil rod domains with one another, which are assumed to make a crucial contribution to IF formation and stability. We show that headless vimentin, which dimerizes under low salt conditions, associates into tetramers of the A22-type configuration under assembly conditions, indicating that one of the effects of increasing the ionic strength is to favor coil 2–coil 2 interactions. Furthermore, in order to obtain insight into the molecular interactions that occur during the first phase of assembly of full-length vimentin, we employed a temperature-sensitive variant of human vimentin, which is arrested at the “unit-length filament” (ULF) state at room temperature, but starts to elongate upon raising the temperature to 37 °C. Most importantly, we demonstrate by cross-linking analysis that ULF formation predominantly involves A11-type dimer–dimer interactions. The presence of A22 and A12 cross-linking products in mature IFs, however, indicates that major rearrangements do occur during the longitudinal annealing and radial compaction steps of IF assembly.

The accumulation of misfolded proteins in intracellular inclusions is a generic feature of neurodegenerative disorders. Although heavily ubiquitylated, the aggregated proteins are not degraded by the proteasomes. A possible reason for this phenomenon may be a modification of deposited proteins by transglutaminases forming gamma-glutamyl-epsilon-lysine (GGEL) cross-links between distinct proteins. Here, we show that the frequency of GGEL cross-links is an order of magnitude higher in Alzheimer's brain cortex than in age-matched or younger controls. This difference is due to the accumulation of GGEL cross-links in ubiquitin-immunopositive protein particles present in both Alzheimer's brains and those from aged individuals. The highly cross-linked protein aggregates show immunoreactivity to antibodies against tau and neurofilament proteins, and partially also to alpha-synuclein, indicating that these structures are inherent in Alzheimer's neurofibrillary tangles and Lewy bodies. Using mass sequence analysis, we identified the same six pairs of peptide sequences cross-linked in both senile and Alzheimer's specimens: Gln31 and Gln190 of HSP27 protein are cross-linked with Lys29 and Lys48 of ubiquitin and HSP27 therefore may cross-link two (poly)ubiquitin chains. One lysine residue of parkin and one of alpha-synuclein were also found to be cross-linked. The data suggest that cross-linking of (poly)ubiquitin moieties via HSP27 may have a role in the stabilization of the intraneuronal protein aggregates by interference with the proteasomal elimination of unfolded proteins.

High-sensitivity, rapid identification of proteins in proteomic studies normally uses a combination of one- or two-dimensional electrophoresis together with mass spectrometry. The simplicity and sensitivity of matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOFMS) have increased its application in recent years. The most common method of 'peptide fingerprinting' often may not provide robust identification. Normally additional sequence information by post-source decay (PSD) MALDI-TOFMS provides additional constraints for database searches to achieve highly confident results. Here we describe a derivatization procedure to facilitate the acquisition of such sequence information. Peptide digests from a skin-expressed protein were modified with 4-sulfophenyl isothiocyanate. The resulting peptides carry a fixed negative charge at the N-terminal end and the resulting PSD spectrum is dominated by C-terminal y-type ions. The sequence information in most cases can be obtained manually or with simple programming tools. Methods of optimizing the procedure and increasing the sensitivity are discussed.

The transglutaminase 3 enzyme is widely expressed in many tissues including epithelia. We have shown previously that it can bind three Ca2+ ions, which in site one is constitutively bound, while those in sites two and three are acquired during activation and are required for activity. In particular, binding at site three opens a channel through the enzyme and exposes two tryptophan residues near the active site that are thought to be important for enzyme reaction. In this study, we have solved the structures of three more forms of this enzyme by x-ray crystallography in the presence of Ca2+ and/or Mg2+, which provide new insights on the precise contribution of each Ca2+ ion to activation and activity. First, we found that Ca2+ ion in site one can be exchanged with difficulty, and it has a binding affinity of Kd = 0.3 μm (ΔH = –6.70 ± 0.52 kcal/mol), which suggests it is important for the stabilization of the enzyme. Site two can be occupied by some lanthanides but only Ca2+ of the Group 2 family of alkali earth metals, and its occupancy are required for activity. Site three can be occupied by some lanthanides, Ca2+,orMg2+; however, when Mg2+ is present, the enzyme is inactive, and the channel is closed. Thus Ca2+ binding in both sites two and three cooperate in opening the channel. We speculate that manipulation of the channel opening could be controlled by intracellular cation levels. Together, these data have important implications for reaction mechanism of the enzyme: the opening of a channel perhaps controls access to and manipulation of substrates at the active site.