The biological nature of hierarchically organized splitting of bands of human chromosomes remained an enigma since the first banding methods were described in 1970 and 1971. The technique introduced by Lore Zech in Caspersson's laboratory involved quinacrine mustard (Q-banding) and fluorescence microscopy 1, while other used Giemsa (G-banding) 23. Though several methods producing chromosome bands were developed later, G-banding became the one most widely used. A uniform system of human chromosomal nomenclature, which allowed to design not only individual chromosomes but also chromosome regions and bands, was proposed for the first time in 1971 at the Fourth International Congress of Human Genetics in Paris 4, later it developed into the document entitled "An International System for Human Cytogenetic Nomenclature", the last edition of which was published in 2005 5. Although recently evolved molecular cytogenetic techniques 678 and array-CGH 9 allow precise characterization of chromosomal abnormalities, analysis of cytogenetic bands is still of great importance. It is often the first step for a clinical diagnosis and in research to understand the biology of an inherited or acquired disease.

The present nomenclature of human chromosomes has been chosen in a more or less randomly manner only by morphological comparison of chromosomal G-bands at different resolution levels and without any systematic investigation about the origin of chromosomal bands. This might be the reason why mapping data from the human genome project do not always fit to the cytogenetic gene mapping data. Thus, an accurate banding nomenclature is required for precise characterization of chromosomal abnormalities.

Recent studies of metaphase chromosomes have revealed that they are remarkably elastic and can be stretched 1011. This extensibility of mitotic chromosomes has been used to increase the resolution of chromosome banding and to do for the first time the systematic analyses of chromosome band splitting 1213. It was found that new sub-bands appeared during the chromosome stretching process and that these sub-bands arose only from G-dark bands. Recently, to confirm these observations we applied another approach which analyzed behavior of multicolor banding (MCB) based pseudo-color bands in respect to G bands on human chromosome 5 using chromosome preparations of different length 14. Here, we extended and complemented these studies and present for the first time the biologically based hierarchically organized splitting of chromosomal bands for all human chromosomes.

Analysis of the splitting of chromosomal bands into sub-bands during the decondensation process from 300 up to 850 bands per haploid karyotype by using array-proved MCB (aMCB) enabled to demonstrate that only G-dark bands split into dark and light chromosomal sub-bands (Figs. 1 and 2). Thus, this fact suggested before, after application of chromosome stretching was confirmed for the entire human karyotype by a second and independent approach. This makes the concept of chromosomal region specific protein swelling even more evident 15. It has to be stressed, that the probes underlying the aMCB-approach are the only molecular cytogenetic probes which are molecular defined, i.e. they are the only within the human DNA sequence anchored FISH banding probes 16. Thus, the probes are well defined and highly suited for such an approach as here done.

To confirm band splitting results after aMCB we selected 2 sets of BAC probes for chromosome 3 and 7 located according to UCSC Genome Browser Database in the neighboring bands 3p21.1 (RP11-876B11: 54,006,044-54,210,950 MB, RP11-673C15: 54,195,430-54,358,466 MB) and 3p14.3 (RP11-904G16: 54,423,824-54,621,639 MB), or 7q11.21 (RP11-458F8: 65,934,704-66,092,418 MB) and 7q11.22 (RP11-584N20: 66,119,511-66,292,087 MB, RP11-243C20: 66,333,985-66,515,748 MB), respectively. BACs for one chromosome were labeled with different fluorochromes and hybridized simultaneously to chromosomes of different length. After that their cytogenetic location was determined using inverted DAPI on short and long chromosomes (Fig. 3). On long chromosomes BAC probes were cytogenetically located at the sub-bands indicated in the UCSC database. At the same time, on shorter chromosomes we could localize all BAC probes for chromosome 7 at 7q11.1. This supports the band splitting data obtained by the aMCB approach, as it indicates that both sub-bands 7q11.21 and 7q11.22 arise indeed from the band 7q11.1. Thus, in a biological nomenclature they should be attributed as 7q11.2 and 7q11.3. BAC FISH results for chromosome 3 could also prove band splitting scheme for 3q14.3-3q21.1 region of chromosome 3 derived on the basis of aMCB FISH (see for the Fig. 3).

Basic on our data, we suggest a biologically correct nomenclature of human chromosome band splitting. We have to state explicitly that we do not intend to replace the ISCN nomenclature by that one, as the latter has been used now for nearly 40 years. Changing the band names would lead to a complete mess in the cytogenetic literature. However, especially to better align new molecular with cytogenetic data the underlying biological nature of band splitting has to be known. For that we started – as far as possible with the band-nomenclature of ISCN 2005 at the 300 and/or 400 band level. However, we had to correct in some cases the ideograms – especially at the very ends of some chromosomes (e.g. 4p, 8p, and others), by including yet not shown dark bands. This is why, according to figures of all human chromosomes in different decondensation states (see 5 pp 32 to 33) there are additional GTG dark bands in the human chromosomes, not yet included in the commonly used ideograms; these yet not recognized bands are listed in table 1. The observed aMCB band-splitting could be explained only by these additional bands, which is a molecular prove for their existence besides the visual proves of GTG-banded chromosomes present in each cytogenetic laboratory (Fig. 4). Similar observations like the here reported ones were previously discussed in other DNA-based chromosome structure studies. The combinations of extreme Alu-richness and GC-richness could be aligned with GTG-/R-bands to so-called isochores 171819.

As noted by Kowalska and coworkers (2007) 20 sub-band information provided in different human genome sequence databases are not identical. This might be partially due to the different initial sources of G-banded ideogram used and leads to discrepancies in gene mapping in different databases. ISCN ideograms are used in the NCBI Human Genome browser http://www.ncbi.nlm.nih.gov/ and are based on morphological comparison of chromosomes at different G-band resolution level. At the same time UCSC http://genome.ucsc.edu/ and Ensembl genome browsers http://www.ensembl.org/index.html are based on the ideograms resulted from so called cytogenetic band prediction 21. The latter method employed results from 9500 FISH experiments to approximate the locations of the 850 high-resolution bands, and thus, could define chromosome band lengths more precisely. Surprisingly, by applying this cytogenetic band prediction algorithm, it was shown that the lengths of the darkest G bands were consistently underestimated, while the opposite was true for the light bands. This finding might be also an indirect proof for the observation that only G-dark bands split into new G-dark and G-light sub-bands, the first ones containing hence potentially higher condensed DNA.

Finally, the present study was only in part able to enlighten the chromosomal architecture of the pericentromeric heterochromatin. It could be shown, that centromere-near subbbands arise e.g. from 1q12, 19q12, 20p11 and 20q11 but not from 9q12 or 16q11.2. Also no new information could be obtained for the heterochromatic short arms of the acrocentric chromosomes. So, neither current genome databases nor cytogenetic nomenclature attempts do consider the DNA or chromatin base of these still somehow enigmatic chromosomal structures.

The biological way of band splitting in peripheral blood lymphocytes is shown for each chromosome in Fig. 4 and for each chromosome ' [see Additional files 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24]. Note: these ideograms are NOT intended to replace neither ISCN ideograms nor nomenclature!

In summary, we achieved now a better understanding of the chromosomal architecture (Fig. 4). The biologically correct nomenclature of sub-bands was found to be in more than 50% different from the ISCN 2005 nomenclature. This may help explaining why mapping data from the human genome project do not always fit the cytogenetic gene mapping data.

The authors declare that they have no competing interests.

NK and HN performed the necessary aMCB studies. NK and AW did the BAC-FISH selection and studies. UC, AW, TL have been involved in drafting the manuscript and revising it critically for important intellectual content.