14-3-3 protein binding blocks the dimerization interface of caspase-2.

Among all species, caspase-2 (C2) is the most evolutionarily conserved caspase required for effective initiation of apoptosis following death stimuli. C2 is activated through dimerization and autoproteolytic cleavage and inhibited through phosphorylation at Ser139 and Ser164 , within the linker between the caspase recruitment and p19 domains of the zymogen, followed by association with the adaptor protein 14-3-3, which maintains C2 in its immature form procaspase (proC2). However, the mechanism of 14-3-3-dependent inhibition of C2 activation remains unclear. Here, we report the structural characterization of the complex between proC2 and 14-3-3 by hydrogen/deuterium mass spectrometry and protein crystallography to determine the molecular basis for 14-3-3-mediated inhibition of C2 activation. Our data reveal that the 14-3-3 dimer interacts with proC2 not only through ligand-binding grooves but also through other regions outside the central channel, thus explaining the isoform-dependent specificity of 14-3-3 protein binding to proC2 and the substantially higher binding affinity of 14-3-3 protein to proC2 than to the doubly phosphorylated peptide. The formation of the complex between 14-3-3 protein and proC2 does not induce any large conformational change in proC2. Furthermore, 14-3-3 protein interacts with and masks both the nuclear localization sequence and the C-terminal region of the p12 domain of proC2 through transient interactions in which both the p19 and p12 domains of proC2 are not firmly docked onto the surface of 14-3-3. This masked region of p12 domain is involved in C2 dimerization. Therefore, 14-3-3 protein likely inhibits proC2 activation by blocking its dimerization surface. DATABASES: Structural data are available in the Protein Data Bank under the accession numbers 6SAD and 6S9K.

14-3-3 protein masks the nuclear localization sequence of caspase-2.

Caspase-2 is an apical protease responsible for the proteolysis of cellular substrates directly involved in mediating apoptotic signaling cascades. Caspase-2 activation is inhibited by phosphorylation followed by binding to the scaffolding protein 14-3-3, which recognizes two phosphoserines located in the linker between the caspase recruitment domain and the p19 domains of the caspase-2 zymogen. However, the structural details of this interaction and the exact role of 14-3-3 in the regulation of caspase-2 activation remain unclear. Moreover, the caspase-2 region with both 14-3-3-binding motifs also contains the nuclear localization sequence (NLS), thus suggesting that 14-3-3 binding may regulate the subcellular localization of caspase-2. Here, we report a structural analysis of the 14-3-3ζ:caspase-2 complex using a combined approach based on small angle X-ray scattering, NMR, chemical cross-linking, and fluorescence spectroscopy. The structural model proposed in this study suggests that phosphorylated caspase-2 and 14-3-3ζ form a compact and rigid complex in which the p19 and the p12 domains of caspase-2 are positioned within the central channel of the 14-3-3 dimer and stabilized through interactions with the C-terminal helices of both 14-3-3ζ protomers. In this conformation, the surface of the p12 domain, which is involved in caspase-2 activation by dimerization, is sterically occluded by the 14-3-3 dimer, thereby likely preventing caspase-2 activation. In addition, 14-3-3 protein binding to caspase-2 masks its NLS. Therefore, our results suggest that 14-3-3 protein binding to caspase-2 may play a key role in regulating caspase-2 activation. DATABASE: The atomic coordinates and structure factors have been deposited in the Protein Data Bank, www.ww pdb.org (PDB ID codes 6GKF and 6GKG).

14-3-3 protein directly interacts with the kinase domain of calcium/calmodulin-dependent protein kinase kinase (CaMKK2).

BACKGROUND: Calcium/calmodulin-dependent protein kinase kinase 2 (CaMKK2) is a member of the Ca2+/calmodulin-dependent kinase (CaMK) family involved in adiposity regulation, glucose homeostasis and cancer. This upstream activator of CaMKI, CaMKIV and AMP-activated protein kinase is inhibited by phosphorylation, which also triggers an association with the scaffolding protein 14-3-3. However, the role of 14-3-3 in the regulation of CaMKK2 remains unknown. METHODS: The interaction between phosphorylated CaMKK2 and the 14-3-3γ protein, as well as the architecture of their complex, were studied using enzyme activity measurements, small-angle x-ray scattering (SAXS), time-resolved fluorescence spectroscopy and protein crystallography. RESULTS: Our data suggest that the 14-3-3 protein binding does not inhibit the catalytic activity of phosphorylated CaMKK2 but rather slows down its dephosphorylation. Structural analysis indicated that the complex is flexible and that CaMKK2 is located outside the phosphopeptide-binding central channel of the 14-3-3γ dimer. Furthermore, 14-3-3γ appears to interact with and affect the structure of several regions of CaMKK2 outside the 14-3-3 binding motifs. In addition, the structural basis of interactions between 14-3-3 and the 14-3-3 binding motifs of CaMKK2 were elucidated by determining the crystal structures of phosphopeptides containing these motifs bound to 14-3-3. CONCLUSIONS: 14-3-3γ protein directly interacts with the kinase domain of CaMKK2 and the region containing the inhibitory phosphorylation site Thr145 within the N-terminal extension. GENERAL SIGNIFICANCE: Our results suggested that CaMKK isoforms differ in their 14-3-3-mediated regulations and that the interaction between 14-3-3 protein and the N-terminal 14-3-3-binding motif of CaMKK2 might be stabilized by small-molecule compounds.

Human procaspase-2 phosphorylation at both S139 and S164 is required for 14-3-3 binding.

Procaspase-2 phosphorylation at several residues prevents its activation and blocks apoptosis. This process involves procaspase-2 phosphorylation at S164 and its binding to the scaffolding protein 14-3-3. However, bioinformatics analysis has suggested that a second phosphoserine-containing motif may also be required for 14-3-3 binding. In this study, we show that human procaspase-2 interaction with 14-3-3 is governed by phosphorylation at both S139 and S164. Using biochemical and biophysical approaches, we show that doubly phosphorylated procaspase-2 and 14-3-3 form an equimolar complex with a dissociation constant in the nanomolar range. Furthermore, our data indicate that other regions of procaspase-2, in addition to phosphorylation motifs, may be involved in the interaction with 14-3-3.

Substitution Glu480Lys in erythroid band 3 corresponds to the Fr(a) blood group antigen and supports existence of the second ectoplasmic loop of band 3.

BACKGROUND: Polymorphisms in extracellular loops of RBC band 3 correspond to antigens of the Diego blood group system. Of the seven putative extracellular loops, no mutations have until recently been found in the second, fifth, and sixth loops. We detected a substitution Glu480Lys that would be located in its second ectoplasmic loop. We hypothesized the substitution may underlie a novel antigen of the Diego system. STUDY DESIGN AND METHODS: IAT was performed using two different multispecific sera containing anti-Fr(a) and a series of multispecific sera containing antibodies against other blood group antigens of the Diego blood group system but not agglutinating Fr(a+) RBCs. Biosynthesis of band 3 was studied by RT-PCR of reticulocyte RNA and electrophoresis of solubilized RBC membranes. Anion exchange function of band 3 was studied by measuring the influx of radiolabeled sulfate. RESULTS: RBCs from the Glu480Lys carrier were agglutinated with sera containing anti-Fr(a) and not by sera with specificities for other antigens of the Diego system. We detected identical quantities of mRNA corresponding to the two band-3 alleles and normal content of band 3 in the RBC membranes, as well as normal sulfate influx into RBCs from the Fr(a) heterozygote. CONCLUSIONS: We confirmed the previously reported molecular basis of the Fr(a) antigen, thus providing supportive evidence for the existence of the second extracellular loop of band 3. We also demonstrated that this substitution does not affect mRNA stability, surface expression, and anion exchange function of band 3.

Characterization of a highly polymorphic marker adjacent to the SLC4A1 gene and of kidney immunostaining in a family with distal renal tubular acidosis.

BACKGROUND: Mutations in the human SLC4A1 (AE1/band 3) gene are associated with hereditary spherocytic anaemia and with distal renal tubular acidosis (dRTA). The molecular diagnosis of AE1 mutations has been complicated by the absence of highly polymorphic genetic markers, and the pathogenic mechanisms of some dRTA-associated AE1 mutations remain unclear. Here, we characterized a polymorphic dinucleotide repeat close to the human AE1 gene and performed an immunocytochemical study of kidney tissue from a patient with inherited dRTA with a defined AE1 mutation. METHODS: One CA repeat region was identified in a phage P1-derived artificial chromosome (PAC) clone containing most of the human AE1 gene and the upstream flanking region. We determined its heterozygosity value in multiple populations by PCR analysis. Genotyping of one family with dominant dRTA identified the AE1 R589H mutation, and family member genotypes were compared with the CA repeat length. AE1 and vH(+)-ATPase polypeptides in kidney tissue from an AE1 R589H patient were examined by immunocytochemistry for the first time. RESULTS: This CA repeat, previously reported as D17S1183, is approximately 90 kb upstream of the AE1 gene and displayed considerable length polymorphism, with small racial differences, and a heterozygosity value of 0.56. The allele-specific length of this repeat confirmed co-segregation of the AE1 R589H mutation with the disease phenotype in a family with dominant dRTA. Immunostaining of the kidney cortex from one affected member with superimposed chronic pyelonephritis revealed vH(+)-ATPase-positive intercalated cells in which AE1 was undetectable, and proximal tubular epithelial cells with apparently enhanced apical vH(+)-ATPase staining. CONCLUSIONS: The highly polymorphic dinucleotide repeat adjacent to the human AE1 gene may be useful for future studies of disease association and haplotype analysis. Intercalated cells persist in the end-stage kidney of a patient with familial autosomal dominant dRTA associated with the AE1 R589H mutation. The absence of detectable AE1 polypeptide in those intercalated cells supports the genetic prediction that the AE1 R589H mutation indeed causes dominant dRTA.