Abstract. We have detected and begun to characterize a 17-kD centromere-specific protein, CENP-A (Earnshaw, W. C., and N. Rothfield, 1985, Chromosoma., 91:313-321). Sera from several humans with CREST scleroderma autoimmune disease (CREST: calcinosis, Raynaud's phenomenon, esophageal dsymotility, sclerodactyly, and telangiectasia) bind this protein in immunoblot assays of HeLa whole cell or nuclear extracts. We have affinity purified the anti-17-kD centromere protein (anti-CENP-A) specific antibodies from immunoblots of HeLa nuclear protein. The antibodies react with epitopes present on CENP-A derived from humans but apparently do not recognize specific epitopes in either rat or chicken nuclei. Only human nuclear protein is CENP-A positive by immunoblot. Furthermore, human cells show localization of anti-CENP-A antibody to centromeres by immunofluorescence microscopy, whereas rat cells do not. On extraction from the nucleus, CENP-A copurifies with core histones and with nucleosome core particles. We conclude that this centromerespecific protein is a historic-like component of chromatin. The data suggest that CENP-A functions as a centromere-specific core histone.

CENP-A, a centromere-specific 17-kDa protein, has histone-like properties. However, in contrast to the common somatic histones, CENP-A is quantitatively retained in bull spermatozoa, and we have exploited this fact to purify CENP-A to apparent homogeneity. Partial sequence analysis of the purified protein indicates that CENP-A is a distinctive gene product. Some CENP-A sequences are highly similar to regions of histone H3. Other segments of CENP-A are not related to H3 or any other histone. These unrelated segments are presumably involved in loaliing CENP-A to centromeric DNA or in centromere-specific functions of CENP-A..The centromeric region of the chromosome is responsible for its integration into the mitotic spindle and for its proper segregation poleward during anaphase (1)(2)(3)(4). Several proteins have been identified that are associated with the centromere (5-10). For the most part, their functions are presently unknown. Recently, however, Bischoff et al. (11) reported that a 47-kDa centromere-specific autoantigen is highly homologous to the translation product ofRCCI, a gene involved in regulating mammalian chromosome condensation. This antigen may be identical to CENP-D, a 50-to 60-kDa centromeric protein reported to be present in a variety of species (7, 9, 10).CENP-A, a centromere-specific protein of 17 kDa, appears to be associated with kinetochore chromatin or with chromatin closely apposed to the outermost domain of the kinetochore, as judged by indirect immunofluorescence (9) and immunoelectron microscopy (12). This localization suggests it may play a direct role in kinetochore function during mitosis.We have extensively analyzed CENP-A and have shown that it appears to be a core histone by the following criteria: (i) it is extracted with histones from chromatin by dilute mineral acids, (ii) it elutes with histones H3 and H4 in apparent tetrameric complexes during ion exchange (13) or sizing-column chromatography (D.K.P., K.O., and R.L.M., unpublished observations), and, most importantly, (iii) it is a component of highly purified nucleosome core particles (13). CENP-A is exceptional among somatic histones not only because it is centromere specific but also because it is quantitatively retained in chromatin during spermatogenesis in mammals, even in species where other histones are quantitatively replaced by protamines (14).The selective retention of this protein in bull sperm, where other histones are absent, has now allowed us to extract and purify CENP-A to homogeneity. We have also subjected the purified CENP-A from bull sperm to partial sequence analysis and report here that it is a distinctive histone, with sequences similar to those of H3, as well as segments that are not related to histones or homologous to any other known mammalian centromeric protein sequences (11, 15). MATERIALS AND METHODSPurification and Acid Extraction of Nuclei ad ReversePhase Chromatography of Acid-Extracted Proteins. Calf thymus nuclei and bull sperm nuclei were purified and extracted with 1 M NaCl/0.25...

Oxidoreductases of the thioredoxin superfamily possess the C-X-X-C motif. The redox potentials vary over a wide range for these proteins. A crucial determinant of the redox potential has been attributed to the variation of the X-X dipeptide. Here, we substitute Lys for Gly at the first X of Escherichia coli thioredoxin to investigate how a positive charge would affect the redox potential. The substitution does not affect the protein's redox potential. The equilibrium constant obtained from pairwise reaction between the mutant and wild-type proteins equals 1.1, indicating that the replacement does not significantly affect the thiol-disulfide redox equilibrium. However, the catalytic efficiency of thioredoxin reductase on the G33K mutant decreases approximately 2.8 times compared to that of the wild type. The mutation mainly affects K(m), with little effect on k(cat). The mutation also inhibits thioredoxin's ability to reduce insulin disulfide by approximately one-half. Whether the mutant protein supports the growth of phages T3/7 and f1 was tested. The efficiency of plating (EOP) of T3/7 on the mutant strain decreases 5 times at 37 degrees C and 3 x 10(4) times at 42 degrees C relative to that of the wild-type strain, suggesting that interaction between phage gene 5 protein and thioredoxin is hindered. The mutation also reduces the EOP of phage f1 by 8-fold at 37 degrees C and 1.5-fold at 42 degrees C. The global structure of the mutant protein does not change when studied by CD and fluorescence spectra. Therefore, G33K does not significantly affect the overall structure or redox potential of thioredoxin, but primarily interferes with its interaction with other proteins. Together with the G33D mutation, the overall results show that a charged residue at the first X has a greater influence on the molecular interaction of the protein than the redox potential.

No abstract

The linker for activation of T-cells (LAT) is a palmitoylated integral membrane adaptor protein that resides in lipid membrane rafts and contains nine consensus putative tyrosine phosphorylation sites, several of which have been shown to serve as SH2 binding sites. Upon T-cell antigen receptor (TCR/CD3) engagement, LAT is phosphorylated by protein tyrosine kinases (PTK) and binds to the adaptors Gads and Grb2, as well as to phospholipase Cgamma1 (PLCgamma1), thereby facilitating the recruitment of key signal transduction components to drive T-cell activation. The LAT tyrosine residues Y(132), Y(171), Y(191), and Y(226) have been shown previously to be critical for binding to Gads, Grb2, and PLCgamma1. In this report, we show by generation of LAT truncation mutants that the Syk-family kinase ZAP-70 and the Tec-family kinase Itk favor phosphorylation of carboxy-terminal tyrosines in LAT. By direct binding studies using purified recombinant proteins or phosphopeptides and by mutagenesis of individual tyrosines in LAT to phenylalanine residues, we demonstrate that Y(171) and potentially Y(226) are docking sites for the Vav guanine nucleotide exchange factor. Further, overexpression of a kinase-deficient mutant of Itk in T-cells reduced both the tyrosine phosphorylation of endogenous LAT and the recruitment of Vav to LAT complexes. These data indicate that kinases from distinct PTK families are likely responsible for LAT phosphorylation following T-cell activation and that Itk kinase activity promotes recruitment of Vav to LAT.