TI - DISCUSSION . AB - Here we present the first mass spectrometry-driven proteome analysis of metaphase chromosomes prepared from HeLa S3 cells . The preparative method used has been previously used successfully in the identification of CENP-E , DNA topoisomerase II and condensin ( 28-31 ) , all of which are found in the preparation we describe here . Despite the levels of cytoskeletal and mitochondrial material , the useful percentage of chromosome -associated material found in this preparation makes it an attractive source for further analysis . It is important to note that the relative amount of any given protein in the preparation cannot be assessed by the techniques we used here , as the various peptides from tryptic digests may behave differently during mass spectrometry--indeed , the relatively low scores obtained for the known scaffold component , topoisomerase II , emphasise this point . We attribute the absence of many of the proteins known to be in the scaffold ( eg the CENP proteins ) from the MALDI-TOF profile to their relatively low abundance and to the possibility of their being poor subjects for mass spectrometric analysis . Our observations suggest that the further use of mass spectrometry against a more stringently-extracted scaffold fraction might reveal novel components of the chromosome scaffold that have not been identified by antibodies . The further characterisation of the unknown and/or poorly-characterised proteins identified in the current screen will require their localisation . This was done for the putative GTP -binding protein NGB/CRFG by fluorescently tagging it and overexpressing it in tissue culture cells . The perichromosomal localisation , seen clearly in metaphase , does not clearly define the likely role of this protein ( 58-62 ) . It may reflect a nucleolar function for the protein ( 47 ) , as might be expected from its apparent localisation in interphase ( 46 ) . However , it may also indicate some role in cell cycle control--a similar mitotic localisation has also been observed for the BCR oncogene product ( 47,48 ) , which interacts with and activates GTP -binding proteins . The chromosomal localisation of NGB/CRFG we describe here may , therefore , be in some way related to its putative role in renal disease ( 46 ) , but very little is known about this protein to date . In the search for new Aurora B SUBstrates , the most likely candidates are those known to have an association with chromosomes or the spindle , such as CAP-C/SMC-4 and topo IIalpha . Investigation of the Xenopus condensin complex showed no mitotic PHOSphorylation of CAP-C ( 63 ) . Therefore , topo IIalpha , which has been well described as a phosPHOprotein ( 49-54 ) , was tested as an Aurora B SUBstrate . Recombinant Aurora B indeed PHOSphorylated recombinant topo IIalpha in vitro . The functions of DNA topo IIalpha during mitosis have long been of interest and its PHOSphorylation , which varies throughout the cell cycle and is regulated by a number of kinases , appears to be of great importance in modulating its activities ( 64,65 ) . While the localisation of topo IIalpha during mitosis is somewhat controversial ( 53,56,66-68 ) , its locations are consistent with its being a potential Aurora B SUBstrate . The localisation of topo IIalpha at the centromere ( 69 ) further enhances the likelihood of the Aurora B interaction being significant , but our inability to specify the target residues by mass spectrometry has hampered our efforts to define this significance . To date , no consensus Aurora B PHOSphorylation signal has been described , so we cannot speculate on the potential target residues on topo IIalpha . While there remain further candidate subSTRates to be identified , the complexity of the metaphase chromosome fraction may necessitate the use of two-dimensional electrophoresis to resolve PHOSphorylated proteins sufficiently for their conclusive identification . In conclusion , we present the proteomic analysis of partially-extracted metaphase chromosomes as a means by which novel mitotic chromosome components may be described and by which chromosomal kinase SUBstrates may be identified . The ability to remove kinase activity endogenous to the preparation by heat treatment means that any kinase may be tested on chromosomes and the more abundant of its potential SUBstrates described .