The structure of a tandem pair of spectrin repeats of plectin reveals a modular organization of the plakin domain

Plectin is a large and versatile cytoskeletal linker and member of the plakin protein family. Plakins share a conserved region called the plakin domain located near their N terminus. We have determined the crystal structure of an N-terminal fragment of the plakin domain of plectin to 2.05 A resolution. This region is adjacent to the actin-binding domain and is required for efficient binding to the integrin alpha6beta4 in hemidesmosomes. The structure is formed by two spectrin repeats connected by an alpha-helix that spans these two repeats. While the first repeat is very similar to other known structures, the second repeat is structurally different with a hydrophobic core, narrower than that in canonical spectrin repeats. Sequence analysis of the plakin domain revealed the presence of up to nine consecutive spectrin repeats organized in an array of tandem modules, and a Src-homology 3 domain inserted in the central spectrin repeat. The structure of the plakin domain is reminiscent of the modular organization of members of the spectrin family. The architecture of the plakin domain suggests that it forms an elongated and flexible structure, and provides a novel molecular explanation for the contribution of plectin and other plakins to the elasticity and stability of tissues subjected to mechanical stress, such as the skin and striated muscle.     

Dissecting the sequence specific functions of alternative N-terminal isoforms of mouse bullous pemphigoid antigen 1

Bullous pemphigoid antigen 1 (BPAG1) is a member of the plakin family of proteins that is involved in cross-linking the cytoskeletal elements and attaching them to cell junctions. BPAG1 null mice develop severe degeneration of sensory neurons that was attributed in part due to the absence of a splice variant called BPAG1a that harbors an actin-binding domain at the N-terminus. Additional alternative splicing also results in BPAG1a isoforms with different first exons, leading to three additional types of BPAG1a called isoforms 1, 2 and 3 (or BPAG1a1, BPAG1a2, and BPAG1a3). These unique N-terminal extensions of the BPAG1a isoforms are of variable length. In this study, we characterized these N-terminal isoforms and evaluated the influence of these unique N-terminal sequences to the actin-binding properties. The unique N-terminal region of isoform 1 is very short and was not expected to affect the property of the ABD that followed it. In contrast, transfection studies and mutagenesis analyses signified that the N-terminal sequences of isoform 2 had the ability to bundle actin filaments and the N-terminal region that contained isoform 3 showed cortical localization. Isoforms 1, 2 and 3 also displayed differential tissue expression profiles. Taken together, these data suggested that the unique N-terminal regions of these isoforms have different roles that may be tailored to meet tissue specific functions.     

Novel alternative splicings of BPAG1 (bullous pemphigoid antigen 1) including the domain structure closely related to MACF (microtubule actin cross-linking factor).

BPAG1 (bullous pemphigoid antigen 1) was originally identified as a 230-kDa hemidesmosomal protein and belongs to the plakin family, because it consists of a plakin domain, a coiled-coil rod domain and a COOH-terminal intermediate filament binding domain. To date, alternatively spliced products of BPAG1, BPAG1e, and BPAG1n are known. BPAG1e is expressed in epithelial tissues and localized to hemidesmosomes, on the other hand, BPAG1n is expressed in neural tissues and muscles and has an actin binding domain at the NH(2)-terminal of BPAG1e. BPAG1 is also known as a gene responsible for Dystonia musculorum (dt) neurodegeneration syndrome of the mouse. Another plakin family protein MACF (microtubule actin cross-linking factor) has also an actin binding domain and the plakin domain at the NH(2)-terminal. However, in contrast to its high homology with BPAG1 at the NH(2)-terminal, the COOH-terminal structure of MACF, including a microtubule binding domain, resembles dystrophin rather than plakins. Here, we investigated RNAs and proteins expressed from the BPAG1 locus and suggest novel alternative splicing variants, which include one consisting of the COOH-terminal domain structure homologous to MACF. The results indicate that BPAG1 has three kinds of cytoskeletal binding domains and seems to play an important role in linking the different types of cytoskeletons.     

Crystal structure of the actin-binding domain of alpha-actinin-4 Lys255Glu mutant implicated in focal segmental glomerulosclerosis

Mutations in α-actinin-4 have been linked to familial focal segmental glomerulosclerosis (FSGS), a common renal disorder in humans, and produce an apparent increase in the actin-binding affinity of α-actinin-4 in vitro. One of the mutations, in particular, Lys255Glu, falls in the middle of the actin-binding interface of the actin-binding domain (ABD). The ABD consists of tandem calponin homology (CH) domains (CH1 and CH2). The crystal structures of most ABDs display a compact conformation, characterized by extensive inter-CH interactions. However, the conformation of F-actin-bound ABDs is unsettled. Some electron microscopy studies find that the compact conformation is preserved upon binding to F-actin, whereas other studies suggest that the CHs separate and the ABD becomes extended. The Lys255Glu mutation in CH2 is significant in this regard since it removes a crucial inter-CH interaction with Trp147 of CH1, thought to stabilize the compact conformation. Together, the increased actin-binding affinity and the removal of this important inter-CH contact suggested that the Lys255Glu mutation might facilitate the transition toward the open ABD conformation proposed by some of the electron microscopy studies. However, the crystal structure of the ABD of α-actinin-4 Lys255Glu mutant described here displays the canonical compact conformation. Furthermore, the sedimentation coefficients by analytical ultracentrifugation of wild-type and FSGS mutant ABDs (Lys255Glu, Ser262Pro, and Thr259Ile) are nearly identical (2.50 ± 0.03 S) and are in good agreement with the theoretical value calculated from the crystal structure (2.382 S), implying that the compact conformation is retained in solution. The absence of a structural change suggests that the compact ABD conformation observed in the majority of the structures is highly stable and is preserved in solution, even in FSGS mutant ABDs.     

Localization of the human gene for 230-kDal bullous pemphigoid autoantigen (BPAG1) to chromosome 6pter----q15.

Chromosome mapping of the human gene encoding the 230-kDal autoantigen of an autoimmune skin disease, bullous pemphigoid, was performed using flow-sorted human chromosomes of cells of normal karyotype and cells carrying a reciprocal translocation t(6;16)(q15;q24). The cDNA of the autoantigen hybridized with intact chromosome 6 and translocation chromosome 6p- (6pter----q15::16q24----qter). The gene (BPA230) was located to the chromosome region 6pter----q15.     

Comparative aspects of polyglutamine binding domain in PQBP-1 among Vertebrata

We investigated the evolutionary conservation of polyglutamine binding protein-1 (PQBP-1) among Vertebrata. PQBP-1s were highly conserved and shared the same domain features including a WW domain, a polar amino acid rich domain (PRD), a nuclear localization signal (NLS), and a C-terminal domain (CTD) among Eutheria, but not always among Vertebrata. PQBP-1s of Vertebrata contained a variable region in the middle portion corresponding to the position of PRD. The full form of PRD including both 7aa and DR/ER repeats was specific to Eutheria. PRD of non-eutherian Amniota was minimal. Amphibia had no PRD. The DR/ER repeat was solo in fishes. Agnatha PRD was also rich in polar amino acids, but contained no repetitive sequence. We investigated 3 polyQ-containing proteins known to interact with PQBP-1: BRN-2, Huntingtin, and ATAXIN-1, and showed a diverse nature of protein-protein interaction in Vertebrata. There appears to be no interaction between PQBP-1 and BRN-2, Huntingtin, or ATAXIN-1 in Amphibia, while the interaction between PQBP-1 and BRN-2 is expected to be conserved among Mammalia, and the interaction between PQBP-1 and Huntingtin or ATAXIN-1 depends on the lineage in Eutheria.     

Bullous pemphigoid antigen (BPAG1): cDNA cloning and mapping of the gene to the short arm of human chromosome 6

Bullous pemphigoid antigen (BPAG1), an integral component of the cutaneous basement membrane zone, serves as autoantigen in a blistering disease, bullous pemphigoid. In this study, we have generated cDNAs corresponding to human BPAG1 sequences. Two cDNAs, a 0.45-kb PCR product synthesized with human keratinocyte RNA as template and a 2.2-kb cDNA isolated from human keratinocyte lambda gt11 library, were utilized for chromosomal in situ hybridizations to establish the genomic location of the BPAG1 gene. Metaphase chromosomes of phytohemagglutinin-stimulated human peripheral blood leukocytes were examined by hybridizations with 3H-labeled cDNAs, and the chromosomes were identified by R-banding (fluorochrome-photolysis-Giemsa method). The results indicated that the human BPAG1 gene is at locus 6p11-6p12. This conclusion was supported by hybridizations with a panel of human X rodent hybrid cell DNA, which indicated concordance with human chromosome 6. Because different genes encoding human basement membrane components have been mapped previously to chromosomes other than 6, the results further indicate that human basement membrane zone genes are widely dispersed within the human genome.     

Structural analysis of the intracellular domain of (pro)renin receptor fused to maltose-binding protein

The (pro)renin receptor (PRR) is an important component of the renin-angiotensin system (RAS), which regulates blood pressure and cardiovascular function. The integral membrane protein PRR contains a large extracellular domain (∼310 amino acids), a single transmembrane domain (∼20 amino acids) and an intracellular domain (∼19 amino acids). Although short, the intracellular (IC) domain of the PRR has functionally important roles in a number of signal transduction pathways activated by (pro)renin binding. Meanwhile, together with the transmembrane domain and a small portion of the extracellular domain (∼30 amino acids), the IC domain is also involved in assembly of V₀ portion of the vacuolar proton-translocating ATPase (V-ATPase). To better understand structural and multifunctional roles of the PRR-IC, we report the crystal structure of the PRR-IC domain as maltose-binding protein (MBP) fusion proteins at 2.0 Å (maltose-free) and 2.15 Å (maltose-bound). In the two separate crystal forms having significantly different unit-cell dimensions and molecular packing, MBP–PRR-IC fusion protein was found to be a dimer, which is different with the natural monomer of native MBP. The PRR-IC domain appears as a relatively flexible loop and is responsible for the dimerization of MBP fusion protein. Residues in the PRR-IC domain, particularly two tyrosines, dominate the intermonomer interactions, suggesting a role for the PRR-IC domain in protein oligomerization.     

Diversity and evolutionary aspects of mucin binding (MucBP) domain repeats among Lactobacillus plantarum group strains through comparative genetic analysis

Mucin binding protein (Mub) of Lactobacillus plantarum- group (LPG) is considered as a potential marker gene due to its diversity and adherence property to host intestinal layer. We recently observed strain specific diversity in mub gene of LPG strains (n = 8) and grouped them into Type I, II and III based on the insertion and/or deletion of MucBP domain repeats (Devi et al., 2016). The mub gene expression pattern noted by real-time PCR revealed 67.64 increased fold change in MTCC 9483 with four MucBP repeats. Whereas, the isolates with 3 and 2 MucBP repeats showed ∼4 and 2 enhanced expression, respectively. Detection of integrase (int) and transposases (tra) in the flanking regions of mub, suggested variation in amplicon sizes and revealed strain diversity. The IS finder program identified tra nucleotide sequence similarity towards IS256 and IS5 family insertion elements that are present among other lactic acid bacteria. Phylogenetic dendrogram suggested the evolution of D₇–D₁₀ domains of L. paraplantarum MTCC 9483 from Lactobacillus namurensis, Lactobacillus hammesii and Lactobacillus spicheri. These results collectively gave information on the genetic diversity, functional MucBP domains and also facilitated in the selection of an efficient novel LPG strain with good adherence ability.     

Plakins in development and disease

Plakins are large multi-domain molecules that have various functions to link cytoskeletal elements together and to connect them to junctional complexes. Plakins were first identified in epithelial cells where they were found to connect the intermediate filaments to desmosomes and hemidesmosomes [Ruhrberg, C., and Watt, F.M. (1997). The plakin family: versatile organizers of cytoskeletal architecture. Curr Opin Genet Dev 7, 392-397.]. They were subsequently found to be important for the integrity of muscle cells. Most recently, they have been found in the nervous system, where their functions appear to be more complex, including cross-linking of microtubules (MTs) and actin filaments [Leung, C.L., Zheng, M., Prater, S.M., and Liem, R.K. (2001). The BPAG1 locus: Alternative splicing produces multiple isoforms with distinct cytoskeletal linker domains, including predominant isoforms in neurons and muscles. J Cell Biol 154, 691-697., Leung, C.L., Sun, D., Zheng, M., Knowles, D.R., and Liem, R.K. (1999). Microtubule actin cross-linking factor (MACF): a hybrid of dystonin and dystrophin that can interact with the actin and microtubule cytoskeletons. J Cell Biol 147, 1275-1286.]. These plakins have also indicated their relationship to the spectrin superfamily of proteins and the plakins appear to be evolutionarily related to the spectrins, but have diverged to perform different specialized functions. In invertebrates, a single plakin is present in both Drosophila melanogaster and Caenorhabditis elegans, which resemble the more complex plakins found in mammals [Roper, K., Gregory, S.L., and Brown, N.H. (2002). The 'spectraplakins': cytoskeletal giants with characteristics of both spectrin and plakin families. J Cell Sci 115, 4215-4225.]. In contrast, there are seven plakins found in mammals and most of them have alternatively spliced forms leading to a very complex group of proteins with potential tissue specific functions [Jefferson, J.J., Leung, C.L., and Liem, R.K. (2004). Plakins: goliaths that link cell junctions and the cytoskeleton. Nat Rev Mol Cell Biol 5, 542-553.]. In this review, we will first describe the plakins, desmoplakin, plectin, envoplakin and periplakin and then describe two other mammalian plakins, Bullous pemphigoid antigen 1 (BPAG1) and microtubule actin cross-linking factor 1 (MACF1), that are expressed in multiple isoforms in different tissues. We will also describe the relationship of these two proteins to the invertebrate plakins, shortstop (shot) in Drosophila and VAB-10 in C. elegans. Finally, we will describe an unusual mammalian plakin, called epiplakin.     

Structural evidence for the interaction of C-protein (MyBP-C) with actin and sequence identification of a possible actin-binding domain

C-protein (MyBP-C) is a myosin-binding protein that is usually seen in two sets of seven to nine positions in the C-zones in each half of the vertebrate striated muscle A-band. Skeletal muscle C-protein is a modular structure containing ten sub-domains (C1 to C10) of which seven are immunoglobulin-type domains and three (C6, C7 and C9) are fibronectin-like domains. Cardiac muscle C-protein has an extra N-terminal domain (C0) and also some sequence insertions, one of which provides phosphorylation sites. It is conceivable that C-protein has both a structural and regulatory role within the sarcomere. The precise mode of binding of C-protein to the myosin filament has not been determined. However, detailed ultrastructural studies have suggested that C-protein, which binds to myosin, can give rise to a longer periodicity (about 435A) than the intrinsic myosin filament repeat of 429A. The reason for this has remained a puzzle for over 25 years. Here we show by modelling and computation that the presence of this longer periodicity could be explained if the myosin-binding part of C-protein binds to myosin with the expected 429A repeat, but if there are systematic interactions of the N-terminal end of C-protein with the neighbouring actin filaments in the hexagonal lattice of filaments in the A-band. We also show that if they occur these interactions would probably only arise in defined muscle states. Further analysis of the MyBP-C sequence identifies a possible actin-binding domain in the Pro-Ala-rich sequence found at the N terminus of skeletal MyBP-C and between domains C0 and C1 in the cardiac sequence.     

Structural and mechanistic comparisons of the metal-binding members of the vicinal oxygen chelate (VOC) superfamily

The vicinal oxygen chelate family of enzymes catalyzes a highly diverse set of chemistries that derives from one common mechanistic trait: bidentate coordination to a divalent metal center by a substrate or intermediate or transition state through vicinal oxygen atoms. The array of reactions catalyzed by this family is mediated structurally by a common fold and protein-chelating residues that secure and localize a metal ion. The common fold has topological symmetry being comprised of two βαβββ units that form an incompletely closed barrel of β-sheet about the metal ion. Interestingly, despite the diversity of the reactions catalyzed and the large number of metals observed to bind and promote the chemistry, this semi-symmetrical open barrel extends metal liganding side chains inward from a highly positionally conserved constellation of amino acid residues within the structure. Moreover, the core barrel fold arises from an array of possible intra/inter domain and subunit arrangements of the individual βαβββ units that are universally observed to stack side-by-side contacting along the first β-strand of each. While there are examples of enzymes that use this fold and do not bind a metal ion, this review is concerned with summarizing the key structural and mechanistic correlations that can be made for the metal-dependent vicinal oxygen chelate enzyme family members.     

Structural analysis of the receptor binding domain of botulinum neurotoxin serotype D

Botulinum neurotoxins (BoNTs) are the most toxic proteins known. The mechanism for entry into neuronal cells for serotypes A, B, E, F, and G involves a well understood dual receptor (protein and ganglioside) process, however, the mechanism of entry for serotypes C and D remains unclear. To provide structural insights into how BoNT/D enters neuronal cells, the crystal structure of the receptor binding domain (S863-E1276) for this serotype (BoNT/D-HCR) was determined at 1.65 Å resolution. While BoNT/D-HCR adopts an overall fold similar to that observed in other known BoNT HCRs, several major structural differences are present. These structural differences are located at, or near, putative receptor binding sites and may be responsible for BoNT/D host preferences. Two loops, S1195-I1204 and K1236-N1244, located on both sides of the putative protein receptor binding pocket, are displaced >10 Å relative to the corresponding residues in the crystal structures of BoNT/B and G. Obvious clashes were observed in the putative protein receptor binding site when the BoNT/B protein receptor synaptotagmin II was modeled into the BoNT/D-HCR structure. Although a ganglioside binding site has never been unambiguously identified in BoNT/D-HCR, a shallow cavity in an analogous location to the other BoNT serotypes HCR domains is observed in BoNT/D-HCR that has features compatible with membrane binding. A portion of a loop near the putative receptor binding site, K1236-N1244, is hydrophobic and solvent-exposed and may directly bind membrane lipids. Liposome-binding experiments with BoNT/D-HCR demonstrate that this membrane lipid may be phosphatidylethanolamine.     

Functional and structural characterization of HIV-1 gp41 ectodomain regions in phospholipid membranes suggests that the fusion-active conformation is extended

HIV-1 entry into its host cell involves a sequential interaction whereby gp41 is in direct contact with the plasma membrane. Understanding the effect of membrane composition on the fusion mechanism can shed light on the unsolved phases of this complex mechanism. Here, we studied N36, a peptide derived from the N-heptad-repeat (NHR) of the gp41 ectodomain, its six helix bundle (SHB) forming counterpart C34, together with the N-terminal 70-mer wild-type peptide (N70), and additional gp41 ectodomain-derived peptides in the presence of two membranes, modeling inner and outer leaflets of the plasma membrane. Information on the structure of these peptides, their affinity towards phospholipids and their ability to induce vesicle fusion was gathered by a variety of fluorescence, spectroscopic and microscopy methods. We found that N36, having strong affinity towards phospholipids, prominently shifts conformation from alpha-helix in an outer leaflet-like zwitterionic membrane to beta-sheet in a membrane mimicking the negatively charged inner leaflet environment, leading to pronounced fusion-activity. Real-time atomic force microscopy (AFM) was used to study the peptides' effect on the membrane morphology, revealing severe bilayer perturbation and extensive pore formation.We also found, that the N36/C34 core is destabilized by electronegative, but not zwitterionic phospholipids. Taken together, our data suggest that the fusion-active pore forming conformation of gp41 is extended, upstream of the SHB. In this manner, folding of the ectodomain into a SHB might also serve as a negative regulator of fusion by impeding gp41 fusion-active surfaces, thus preventing irreversible damage to the cell membrane. This assumption is supported by the finding that pre-incubation of large unilamellar vesicles (LUV) with C-heptad repeat (CHR)-derived fusion inhibitors reduces the fusogenic activity of N-terminal peptides in a dose-dependant manner, and suggests that CHR-derived fusion inhibitors inhibit HIV entry in an analogous mechanism.     

Structural analysis of the intracellular domain of (pro)renin receptor fused to maltose-binding protein

In Synechocystis sp. PCC 6803, the loop domain (aa 1–70) of the phycobilisome core-membrane linker, L_CM, was found to interact with the glycosyl transferase homolog, Sll1466. Growth of a Sll1466 knock-out mutant was slightly faster in low light, but strongly inhibited in high light; the phenotype is discussed in relation to the regulation of light energy transfer to photosystem II. At the molecular level, the mutant shows the following changes compared to the wild type: (1) a smaller size and higher mobility of phycobilisomes on the thylakoid membrane, and (2) a changed lipid composition of the thylakoid membrane, especially decreased amounts of digalactosyl diacylglycerol. These results indicate a profound regulatory role for Sll1466 in regulating photosynthetic energy transfer.     

Genome-wide identification and analysis of the aldehyde dehydrogenase (ALDH) gene superfamily of Gossypium raimondii

Background

Aldehyde dehydrogenases (ALDHs) are members of the NAD(P)⁺-dependent protein superfamily which catalyzes aliphatic and aromatic aldehyde oxidation to non-toxic carboxylic acids. ALDH genes may offer promise for improving plant adaptation to environmental stress. Recently, elucidated genome sequences of Gossypium raimondii provide a foundation for systematic identification and analysis of ALDH genes. To date, this has been accomplished for many plant species except G. raimondii.

Results

In this study, thirty unique ALDH sequences that code for 10 ALDH families were identified in the G. raimondii genome. Phylogenetic analysis revealed that ALDHs were split into six clades in G. raimondii, and ALDH proteins from the same families were clustered together. Phylogenetic relationships of ALDHs from 11 plant species suggest that ALDHs in G. raimondii shared the highest protein homology with ALDHs from poplar. Members within ALDH families possessed homologous exon–intron structures. Chromosomal distribution of ALDH did not occur evenly in the G. raimondii genome and many ALDH genes were involved in the syntenic region as documented by identification of physical locations among single chromosomes. In addition, syntenic analysis revealed that homologues of many G. raimondii ALDHs appeared in corresponding Arabidopsis and poplar syntenic blocks, indicating that these genes arose prior to G. raimondii, Arabidopsis and poplar speciation. Finally, based on gene expression analysis of microarray and RNA-seq, we can speculate that some G. raimondii ALDH genes might respond to drought or waterlogging stresses.

Conclusion

Genome-wide identification and analysis of the evolution and expression of ALDH genes in G. raimondii laid a foundation for studying this gene superfamily and offers new insights into the evolution history and speculated roles in Gossypium. These data can be used to inform functional genomic studies and molecular breeding in cotton.     

Crystal structure at 2.8 A of Huntingtin-interacting protein 1 (HIP1) coiled-coil domain reveals a charged surface suitable for HIP1 protein interactor (HIPPI)

Huntington's disease is a genetic neurological disorder that is triggered by the dissociation of the huntingtin protein (htt) from its obligate interaction partner Huntingtin-interacting protein 1 (HIP1). The release of the huntingtin protein permits HIP1 protein interactor (HIPPI) to bind to its recognition site on HIP1 to form a HIPPI/HIP1 complex that recruits procaspase-8 to begin the process of apoptosis. The interaction module between HIPPI and HIP1 was predicted to resemble a death-effector domain. Our 2.8-Å crystal structure of the HIP1 371-481 subfragment that includes F432 and K474, which is important for HIPPI binding, is not a death-effector domain but is a partially opened coiled coil. The HIP1 371-481 model reveals a basic surface that we hypothesize to be suitable for binding HIPPI. There is an opened region next to the putative HIPPI site that is highly negatively charged. The acidic residues in this region are highly conserved in HIP1 and a related protein, HIP1R, from different organisms but are not conserved in the yeast homologue of HIP1, sla2p. We have modeled ∼ 85% of the coiled-coil domain by joining our new HIP1 371-481 structure to the HIP1 482-586 model (Protein Data Bank code: 2NO2). Finally, the middle of this coiled-coil domain may be intrinsically flexible and suggests a new interaction model where HIPPI binds to a U-shaped HIP1 molecule.     

Interaction of the bullous pemphigoid antigen 1 (BP230) and desmoplakin with intermediate filaments is mediated by distinct sequences within their COOH terminus

The bullous pemphigoid antigen 1 (BP230) and desmoplakin (DP) are members of the plakin protein family of cytolinkers. Despite their homology, their COOH termini selectively bind distinct intermediate filaments (IFs). We studied sequences within their COOH termini required for their interaction with the epidermal keratins K5/K14, the simple epithelial keratins K8/K18, and type III IF vimentin by yeast three-hybrid, cell transfection, and overlay assays. The results indicate that BP230 interacts with K5/K14 but not with K8/K18 or vimentin via a region encompassing both the B and C subdomains and the COOH extremity, including a COOH-terminal eight-amino-acid stretch. In contrast, the C subdomain with the COOH-terminal extremity of DP interacts with K5/K14 and K8/K18, and its linker region is able to associate with K8/K18 and vimentin. Furthermore, the potential of DP to interact with IF proteins in yeast seems to be regulated by phosphorylation of Ser 2849 within its COOH terminus. Strikingly, BP230 and DP interacted with cytokeratins only when both type I and type II keratins were present. The head and tail domains of K5/K14 keratins were dispensable for their interaction with BP230 or DP. On the basis of our findings, we postulate that (1) the binding specificity of plakins for various IF proteins depends on their linker region between the highly homologous B and C subdomains and their COOH extremity and (2) the association of DP and BP230 with both epidermal and simple keratins is critically affected by the tertiary structure induced by heterodimerization and involves recognition sites located primarily in the rod domain of these keratins.     

Structural analysis of the human galectin-9 N-terminal carbohydrate recognition domain reveals unexpected properties that differ from the mouse orthologue

Galectins are a family of β-galactoside-binding lectins that contain a conserved carbohydrate recognition domain (CRD). They exhibit high affinities for small β-galactosides as well as variable binding specificities for complex glycoconjugates. Structural and biochemical analyses of the mechanism governing specific carbohydrate recognition provide a useful template to elucidate the function of these proteins. Here we report the crystal structures of the human galectin-9 N-terminal CRD (NCRD) in the presence of lactose and Forssman pentasaccharide. Mouse galectin-9 NCRD, the structure of which was previously solved by our group, forms a non-canonical dimer in both the crystal state and in solution. Human galectin-9 NCRD, however, exists as a monomer in crystals, despite a high sequence identity to the mouse homologue. Comparative frontal affinity chromatography analysis of the mouse and human galectin-9 NCRDs revealed different carbohydrate binding specificities, with disparate affinities for complex glycoconjugates. Human galectin-9 NCRD exhibited a high affinity for Forssman pentasaccharide; the association constant for mouse galectin-9 NCRD was 100-fold less than that observed for the human protein. The combination of structural data with mutational studies demonstrated that non-conserved amino acid residues on the concave surface were important for determination of target specificities. The human galectin-9 NCRD exhibited greater inhibition of cell proliferation than the mouse NCRD. We discuss the biochemical and structural differences between highly homologous proteins from different species.     

Crystal structure of the human carboxypeptidase N (kininase I) catalytic domain

Human carboxypeptidase N (CPN), a member of the CPN/E subfamily of "regulatory" metallo-carboxypeptidases, is an extracellular glycoprotein synthesized in the liver and secreted into the blood, where it controls the activity of vasoactive peptide hormones, growth factors and cytokines by specifically removing C-terminal basic residues. Normally, CPN circulates in blood plasma as a hetero-tetramer consisting of two 83 kDa (CPN2) domains each flanked by a 48 to 55 kDa catalytic (CPN1) domain. We have prepared and crystallized the recombinant C-terminally truncated catalytic domain of human CPN1, and have determined and refined its 2.1 A crystal structure. The structural analysis reveals that CPN1 has a pear-like shape, consisting of a 319 residue N-terminal catalytic domain and an abutting, cylindrically shaped 79 residue C-terminal beta-sandwich transthyretin (TT) domain, more resembling CPD-2 than CPM. Like these other CPN/E members, two surface loops surrounding the active-site groove restrict access to the catalytic center, offering an explanation for why some larger protein carboxypeptidase inhibitors do not inhibit CPN. Modeling of the Pro-Phe-Arg C-terminal end of the natural substrate bradykinin into the active site shows that the S1' pocket of CPN1 might better accommodate P1'-Lys than Arg residues, in agreement with CPN's preference for cleaving off C-terminal Lys residues. Three Thr residues at the distal TT edge of CPN1 are O-linked to N-acetyl glucosamine sugars; equivalent sites in the membrane-anchored CPM are occupied by basic residues probably involved in membrane interaction. In tetrameric CPN, each CPN1 subunit might interact with the central leucine-rich repeat tandem of the cognate CPN2 subunit via a unique hydrophobic surface patch wrapping around the catalytic domain-TT interface, exposing the two active centers.