Figure 1.
Uncharacterized human proteins are extracted from the UniProt Knowledgebase. Steps 1–4 are fully automated. The number of proteins that remain after each filter step is summarized on the right.
Table 1.
Summary of sequence and structural properties of the best four predictions and chemokine CXCL17 as reference.
Table 2.
Fold recognition results for the top five hits obtained for B42 with the Chemokine fold library.
Table 3.
Fold recognition results for the five top fold hits obtained for B42 with the pdb95 fold library.
Figure 2.
Modelling of the B42 protein as an IL8-like chemokine.
A) Sequence-to-structure alignment of B42 with its best scoring template vMIP-I (PDBId: 1ZXT). The secondary structure of vMIP-I is indicated below (yellow arrows for β-strands, red cylinders for α-helices, and dark grey line for coil regions; the light grey line indicates no atom coordinates for those residues in template). Cysteines are highlighted in boxes coloured (green, red, and blue) according to their pairing in each protein sequence. B) The 3D model of B42 (left) is compared with the X-ray structure of vMIP-I (right) used as template. Disulfide bonds are shown as orange sticks, and corresponding cysteines are labelled according to their pairing with the same colour code as in panel A. C) Sequence alignment of human B42 with its orthologous proteins (Ensembl release 54). Cysteines are highlighted in black boxes. The sequence identity between human and chimpanzee (Chimp_B42: ENSPTRG00000034188) is 98%, and 91% with oranguthan (OranUthan_B42: ENSPPYG00000010339).
Figure 3.
Modelling of the N73 protein as an IL8-like chemokine.
A) Sequence-to-structure alignment of the mature N73 (without signal peptide) with its best scoring template vMIP-II (PDBId: 2FJ2). The secondary structure of vMIP-II is indicated below (yellow arrows for β-strands, red cylinders for α-helices, and dark grey line for coil regions; the light grey line indicates no atom coordinates for those residues in template). Cysteines are highlighted in boxes coloured (green, red, and blue) according to their pairing in each protein sequence. B) The 3D model of N73 (left) is compared with the X-ray structure of vMIP-II (right) used as template. Disulfide bonds are shown as orange sticks, and corresponding cysteines are labelled according to their pairing with the same colour code as in panel A. C) Sequence alignment of the human N73 precursor with its orthologous proteins. Cysteines are highlighted in black boxes. Predicted N-glycosylation sites are shown in pink boxes, and the RGD motif in orange boxes. Under the alignment, ‘s’ indicates location of signal peptide (SP), ‘*’ indicates location of low complexity region (LCR), ‘c’ indicates location of predicted chemokine domain, ‘g’ indicates motif for glycosaminoglycan attachment at a serine residue, and ‘∼’ indicates location of disordered region predicted by GlobPlot. The sequence identity between human and gorilla (pre_N73_Gorilla: ENSGGOP00000010336) is 96%, and 45% with bushbaby (pre_N73_Bushb: ENSOGAP00000012479).
Figure 4.
Expression and gene structure of the B42 isoforms.
A) Expression of two B42 isoforms in human tissues (expected size for isoform A: 272 bp). B) Sequencing results of the two different B42 isoforms. Isoform A codes for the full length B42 protein, whereas Isoform B contains a stop codon after 9 residues.
Figure 5.
Protein localization results of B42.
A) Westernblot analysis of the C-terminal GFP-tagged B42 transgenic cell pool using an antibody to GFP. A weak band of the expected size 39.8 kDa (GFP-tag: 31 kDa+B42 8.8 kDa) was detected (marked with *). B) Immunoflourescent staining of the transgenic cell line. Two positive and two negative cells of the cell pool are shown. Localization of the C-terminal GFP-tagged B42 protein is mainly in the cytoplasm of HeLa cells: GFP - green, alpha-tubulin - red, DNA (DAPI) – blue. C) Only GFP-signal of B is shown in grey scale.
Figure 6.
Deconvoluted mass spectrum of B42.
Upper panel: Deconvolution of experimental spectrum with monoisotopic peak labelled. Inlay: SDS-page gel electrophoresis results to check B42 protein recovery. Lower panel: Theoretical isotopic distribution for [GA-B42] assuming 3 disulfide bonds, with monoisotopic peak labelled.
Table 4.
Predicted and experimentally determined secondary structure content of B42.
Figure 7.
Circular dichroism (CD) spectra of vMIP-II and B42.
The CD spectrum of chemokine vMIP-II taken from literature [68] is shown (left) in comparison to the spectrum measured for B42 (right).