Individual karyotypes at the origins of cervical carcinomas
© McCormack et al.; licensee BioMed Central Ltd. 2013
Received: 21 September 2013
Accepted: 1 October 2013
Published: 17 October 2013
Skip to main content
© McCormack et al.; licensee BioMed Central Ltd. 2013
Received: 21 September 2013
Accepted: 1 October 2013
Published: 17 October 2013
In 1952 Papanicolaou et al. first diagnosed and graded cervical carcinomas based on individual “abnormal DNA contents” and cellular phenotypes. Surprisingly current papilloma virus and mutation theories of carcinomas do not mention these individualities. The viral theory holds that randomly integrated, defective genomes of papilloma viruses, which are often untranscribed, cause cervical carcinomas with unknown cofactors 20–50 years after infection. Virus-free carcinomas are attributed to mutations of a few tumor-suppressor genes, especially the p53 gene. But the paradox of how a few mutations or latent defective viral DNAs would generate carcinomas with endless individual DNA contents, degrees of malignancies and cellular phenotypes is unsolved. Since speciation predicts individuality, we test here the theory that cancers are autonomous species with individual clonal karyotypes and phenotypes. This theory postulates that carcinogens induce aneuploidy. By unbalancing mitosis genes aneuploidy catalyzes chain reactions of karyotypic evolutions. Most such evolutions end with non-viable karyotypes but a few become new cancer karyotypes. Despite congenitally unbalanced mitosis genes cancer karyotypes are stabilized by clonal selections for cancer-specific autonomy.
To test the prediction of the speciation theory that individual carcinomas have individual clonal karyotypes and phenotypes, we have analyzed here the phenotypes and karyotypes of nine cervical carcinomas. Seven of these contained papilloma virus sequences and two did not. We determined phenotypic individuality and clonality based on the morphology and sociology of carcinoma cells in vitro. Karyotypic individuality and clonality were determined by comparing all chromosomes of 20 karyotypes of carcinomas in three-dimensional arrays. Such arrays list chromosome numbers on the x-axis, chromosome copy numbers on the y-axis and the number of karyotypes arrayed on the z-axis. We found (1) individual clonal karyotypes and phenotypes in all nine carcinomas, but no virus-specific markers, (2) 1-to-1 variations between carcinoma-specific karyotypes and phenotypes, e.g. drug-resistance and cell morphology, (3) proportionality between the copy numbers of chromosomes and the copy numbers of hundreds of over- and under-expressed mRNAs, (4) evidence that tobacco-carcinogens induce cervical carcinomas via aneuploidy, consistent with the speciation theory.
Since the individual clonal karyotypes of nine carcinomas correlated and co-varied 1-to-1 with complex individual transcriptomes and phenotypes, we have classical genetic and functional transcriptomic evidence to conclude that these karyotypes encode carcinomas - much like the clonal karyotypes that encode conventional species. These individual karyotypes explain the individual “DNA contents”, the endless grades of malignancies and the complex individual transcriptomes and phenotypes of carcinomas.
In 1952 Papanicolaou et al. first diagnosed and graded cervical carcinomas microscopically based on individual “abnormal DNA contents” and cellular phenotypes . These individualities of cervical carcinomas have since been confirmed and extended by numerous cytometric and cytogenetic studies [2–21]. Based on these classic characteristics ‘Pap-tests’ are used to this day to diagnose and grade carcinomas.
Surprisingly, the current papilloma virus and mutation theories of carcinomas do not mention the individual DNA contents and phenotypes of carcinomas. The viral theory holds that randomly integrated defective viral genomes, which are non-immunogenic and often untranscribed, cause cervical carcinomas with unknown cofactors 20–50 years after infection [13, 17, 18, 21–36]. Since defective and inactive papilloma viral sequences are present in at least 50 million (30%) of cancer-free American females  as compared to 12,000 annual cervical carcinomas , these cofactors must be very rare events. On the other hand the virus-free carcinomas are attributed to mutations of certain tumor-suppressor genes, especially the p53 gene [12, 13, 16, 39–45]. But the paradox of how either a few mutations or latent defective viral genomes generate carcinomas with endless “abnormal DNA contents”, grades of malignancies and phenotypic individualities is unsolved .
The relevance of this theory to cervical carcinomas is supported by: (1) A plethora of classic evidence for abnormal DNA contents or aneuploidies in cervical carcinomas [1–21, 64–67], (2) Recent evidence that the copy numbers of the mRNAs of individual carcinomas [33, 45, 68–72], like those of other cancers [73–78], are proportional to the copy numbers of the corresponding carcinoma-specific chromosomes (See Results regarding transcriptomes of cervical carcinomas), and (3) Induction of cervical carcinomas by aneuploidogenic cigarette smoke-carcinogens as predicted by the speciation theory [27, 79–83].
In this study we have investigated the predictions of the speciation theory that, (1) Cervical carcinomas have individual clonal karyotypes, transcriptomes and phenotypes irrespective of integrated viral sequences, (2) There are 1-to-1 correlations between phenotypic and karyotypic variations, e.g. intrinsic and acquired drug-resistance and cell morphology, (3) The copy numbers of hundreds of over- and under-expressed mRNAs of carcinomas are proportional to the copy numbers of the corresponding chromosomes of individual carcinomas, (4) Cigarette smoke carcinogens can initiate carcinomagenesis by inducing aneuploidy. As we show below our investigations have indeed confirmed all of these predictions of the speciation theory.
Karyotypic and phenotypic individualities of the nine cervical carcinomas of this study
Cell morphology and sociology
Ploidy (Avg. Chromosome no)
Cobblestone shapes; tight 3D colonies
85% Near-diploid (46±1)
15% Near-tetraploid (87±2)
80% Near-triploid (68±3)
20% Near-hexaploid (133±1)
Polygonal; growing as dense monolayers
Round; tight 3D colony
Fusiform; tight 3D colony
Triangular; high density
Polygonal; refractile; tight 3D colonies
Round or oval; dense gradually growing 3D colonies
It is shown below that the morphological variants of SW-756 and of HeLa are karyotypically closely related (see, Do karyotypic variations or mutations alter the morphology and drug-sensitivity of carcinomas?). This suggests that cancer-specific karyotypic variations, rather than mutations would have generated these morphological variants.
The cervical carcinomas studied here also differed from each other in intrinsic resistances to cytotoxic drugs. According to the speciation theory this characteristic of cancers is a direct consequence of the karyotypic origins of cancers as follows: Since specific combinations of multi-genic chromosomes generate new autonomous cancer species, numerous new gene combinations with new phenotypes are inevitably generated non-selectively, which are not necessary for carcinogenesis – such as intrinsic drug-resistance [86–88]. Intrinsic drug-resistance is thus a drug treatment-independent, individual phenotype of cancers.
Intrinsic drug‒resistances of cervical carcinoma lines
Section-specific conclusions. We conclude that all nine distinct cervical carcinomas studied here have individual cell morphologies and individual intrinsic resistances against cytotoxic drugs. These results confirm the prediction of the speciation theory that individual carcinomas have individual phenotypes, because they have individual karyotypes.
A common viral etiology would instead have predicted non-individual, common carcinoma-specific phenotypes shared by the seven virus-positive carcinomas listed above. Such virus-specific phenotypes would be lacking in the two virus-free carcinomas C-33A and HT-3. However, no such virus-specific phenotypes were observed.
To test whether the individual phenotypes of cervical carcinomas have karyotypic origins, we have next analyzed the karyotypes of the nine cervical carcinomas.
Figures 6 shows the karyotype arrays of a normal human female and of the near-diploid carcinoma C-4II together with a table listing the clonality of all C-4II chromosomes in %. As can be seen in the table, the average chromosome number of C-4II was 43±1 and thus near clonal (shown on the top of the table). The array and the table further show that the copy numbers of the C-4II chromosomes were 80 to 100% clonal and thus formed mostly parallel lines. The non-clonal chromosomes consisted (1) of intact chromosomes with non-clonal copy numbers mostly oscillating in narrow margins of ±1 of clonal values, and (2) of two unique marker chromosomes, which were present in 2 of 20 cells. By contrast, the 20 karyotypes of the normal female were 100% clonal.
A comparison of the C-4II and C-33A arrays to each other (Figures 6 and 7) shows at a glance that the karyotypes of both carcinomas are highly clonal, and that they are individually very different, much like two different signatures. Moreover, the arrays of both carcinomas differ substantially from that of the normal female.
The presence of non-clonal chromosomes in both carcinomas indicates ongoing chromosomal variation despite overall clonality. In addition the C-33A displayed flexibility of the karyotype as a whole, namely with regard to its ploidy. An approximate two-fold polyploidization of the basic near-diploid karyotype was observed in 3 per 20 cells. Other examples of polyploidizations in cancers are listed as cancer-specific forms of karyotypic flexibility in Table 1, in figures below, in the Background and in the literature [21, 46, 50, 55, 58, 89].
Thus the karyotype arrays of C-4II and C-33A confirm the individualities and clonalities of the karyotypes of carcinomas predicted by the speciation theory. Moreover, these karyotypes indicate a 1-to-1 ‘genotype-phenotype’ correspondence between individual clonal karyotypes and cellular phenotypes (Figure 2A,B) of the two carcinomas: ‘one carcinoma – one karyotype’.
In the following we have analyzed the karyotype arrays of the seven near-triploid cervical carcinomas introduced above (Table 1). According to the speciation theory the near-triploid carcinoma karyotypes should be more flexible and thus more heterogeneous than the near-diploid carcinomas, because more mitotic genes will be unbalanced in near-triploid than in near-diploid cancer karyotypes [60, 62, 90] (Background, Figure 1).
Thus far the karyotype arrays of nine cervical carcinomas illustrate at a glance the individuality and clonality of each carcinoma. Comparisons of the karyotypes of all carcinomas with their cellular phenotypes (Figures 2, 3 and 4) further indicate classical, individual ‘genotype-phenotype’ correlations . Thus all cervical carcinomas studied here confirm the ‘one carcinoma-one karyotype’ rule shown to apply to the two near-diploid carcinomas C-4II and C-33A above .
As expected from the high degrees of aneuploidy (relative to normal cells) of the seven near-triploid carcinomas, the karyotypes of the near-triploid carcinomas (1) were more flexible and thus more heterogeneous, (2) included much higher numbers of clonal marker chromosomes, and (3) included higher percentages of cells with unique, non-clonal marker chromosomes than their near-diploid counterparts (see Background and next section).
Carcinoma-specific chromosome recombination indices
% cells with non-clonal markers
SW-756 C1 and C2
15 and 25
HeLa R and S
10 and 20
The relatively low chromosome recombination index of the near-triploid HeLa cell line shown in Table 3 may reflect evolutionary time-dependent selective balancing of mitosis genes in cancer species. The HeLa cell line was the first to be established in vitro, and has been stably maintained in many laboratories since 1952 .
Section-specific conclusions. 1) All (100%) of the carcinomas tested here (Table 1) contained individual, clonal karyotypes as predicted by the speciation theory (Figures 6, 7, 8, 9, 10, 11, 12, 13 and 14). The clonality of these carcinomas indicates that they derived from a single ancestral cell, in which this clonogenic karyotype was formed. Since these individual karyotypes correlated 1-to-1 with individual cell morphologies (Figures 2, 3 and 4), individual intrinsic drug-resistances (Table 2) and individual chromosome recombination indices (Table 3), and are not found in normal cells (see Figure 6), we conclude that these karyotypes generated and maintain the carcinomas studied here.
The nine carcinoma karyotypes analyzed here also confirmed the individual karyotypic flexibilities predicted by the speciation theory (Background). Moreover the arrays shown in Figures 6, 7, 8, 9, 10, 11, 12, 13 and 14 and the carcinoma-specific chromosome recombination indices listed in Table 3 show that the karyotypic flexibilities and corresponding karyotypic heterogeneities are roughly proportional to the karyotypic complexity of carcinomas, as predicted by the speciation theory.
With regard to the relevance of carcinoma karyotypes to Pap-tests we deduce that karyotype analyses could provide new, DNA content-independent distinctions between carcinomas with near-normal DNA contents and low Pap-grades such as C-4II and C-33A and normal diploid hyperplastic tissues (Figures 6 and 7). Among the Pap-tests with high, near-triploid DNA contents, karyotype analyses could provide new information like the chromosome recombination index as a measure to grade malignancy.
Since all nine individual carcinomas we analyzed contained individual clonal karyotypes and phenotypes, we deduced that individual karyotypes encode these carcinomas - just like the individual karyotypes that encode conventional species. This theory predicts that carcinomas are subject to phenotypic alteration by karyotypic variations, which would thus be sub-speciations. Previously, we have found that alterations in the sensitivity of cancers to toxic drugs [55, 88], and the origins of metastases [50, 57] correlated with karyotypic alterations.
By contrast, the viral and mutation theories of carcinomas postulate that poorly-defined viral sequences and gene mutations determine the phenotypes of carcinomas (Background).
To distinguish between the mutation and speciation theories as mechanisms of phenotypic variation, we compared here the karyotypes of two matched pairs of carcinomas differing in cell morphology, and of two matched pairs of carcinomas differing in acquired drug-resistance.
As can be seen in Figure 12 the karyotypes of the SW-756 clone with round cells (C1) differed from the karyotype of the clone with fusiform cells (C2) (Figure 4A,B) in seven clonal copy numbers of intact chromosomes and in six round clone C1-specific and four fusiform clone C2-specific marker chromosomes.
Likewise we show in Figure 13 that the karyotype of the HeLa R clone with round cells differed from the karyotype of the HeLa clone S with fusiform cells in the clonal copy numbers of five intact chromosomes and in six R-specific and eight S-specific marker chromosomes.
Similarly we show in Figure 13 that the karyotype of the methotrexate-resistant HeLa S clone (0.5 μg per 10^6 cells) differed from the karyotype of the drug-sensitive HeLa S clone in the copy numbers of 10 intact chromosomes and the loss of seven parental markers.
Likewise we show in Figure 14 that the karyotype of the methotrexate-resistant MS-751 clone differed from the karyotype of the drug-sensitive clone in the copy numbers of eight intact chromosomes, and the loss of one and the gain of 4 new clonal marker chromosomes.
By chance, we found that the four new methotrexate-specific MS-751 marker chromosomes were derived from one non-clonal cell of the native MS-751 carcinoma (Figure 14). To emphasize this accidental discovery, we flagged these markers in Figure 14 by an asterisk in both cases. Thus drug-resistance in this case was probably generated in part by the clonal selection of a non-clonal variant of the drug-sensitive native carcinoma MS-751. This confirms a previous example for phenotypic variation by selection of a non-clonal chromosome by Heng et al. [51, 56].
It could be argued, however, that the morphological variants of SW-756 and of HeLa, and the drug-resistant variants of HeLa and of MS-751 studied here were each caused by undetected mutations of cellular or viral genes, but were inadvertently accompanied by new clonal karyotypic variations. But it is unlikely that in all four of these variations undetected causal mutations were inadvertently accompanied by karyotypic alterations that were clonally stable in subsequent passages. It would follow that specific karyotypic variations, rather than mutations generated the morphological and drug-resistant carcinoma variants.
Several researchers have recently investigated the transcriptomes of cervical carcinomas. As is the rule with other cancers (Background), the transcriptomes of cervical carcinomas consist of hundreds of individually over- and under-expressed normal mRNAs [33, 45, 68–72]. In agreement with the individuality of the transcriptomes of carcinomas, no consistent or “recurrent” carcinoma-specific “drivers” emerged in searches for common carcinoma-specific transcripts [33, 45, 69, 70].
Moreover, it has not been possible to identify consistent transcriptomes of cervical carcinomas with HPV-functions [33, 45], “likely representing common signaling pathways triggered by HPV”  and thought to have “potential value of targeting E6 and E7 function … in the treatment of HPV(+) cancers” . This is again consistent with the absence of a common, virus-specific karyotypic pattern in the different virus-positive carcinomas described in section two above.
In view of this we propose here that the individual karyotypes of cervical carcinomas encode the complex, individual transcriptomes of carcinomas (Figure 1). This proposal is consistent with several independent studies, which have shown that the copy numbers of the mRNAs of cancers are directly proportional to the copy numbers of the corresponding chromosomes [71, 73–75, 77, 78]. Accordingly Landry et al. have even shown by quantitative sequence analyses that the copy numbers of all HeLa mRNAs are exactly proportional to the copy numbers of the corresponding chromosomes. In view of this they concluded that there is “no dosage compensation” for the chromosome copy number-dependent dosage of HeLa mRNAs . Adey et al. have just confirmed this conclusion .
Section-specific conclusions. We conclude that individual karyotypes of cervical carcinomas encode their highly complex, individual transcriptomes. In view of this we see the exact correlations between the copy numbers of the mRNAs and the copy numbers of the corresponding chromosomes of HeLa cells described by Landry et al.  and Adey et al.  as Rosetta stones of functional evidence for our theory.
All cancers have common characteristics in addition to those studied here, as for example (a) autonomy, (b) immortality and (c) the ability to metastasize [49, 59]. Indeed all nine carcinomas studied here are autonomous and immortal in cell culture  and three of these, HT-3, MS-751 and CaSki have been described to form metastases in vivo .
The following reasons indicate that these cancer-specific commonalities are properties of common cancer-specific karyotypes rather than of common mutations or virus-derived sequences: (a) Autonomy is the primary characteristic of a species conferring reproductive independence of the parental species. As we pointed out above, the origin of carcinoma-specific autonomy coincides with that of the clonal origins of carcinoma-specific karyotypes (Background, Figure 1, Results, sections two and three). (b) Immortality of carcinomas has been reduced to karyotypic flexibility, which generates new variants constitutively to replace the inevitable chromosomal and genetic mutations that accumulate over time  (Figure 1). As described in the Background karyotypic flexibility and the corresponding cancer-specific clonal heterogeneity are direct consequences of the congenital aneuploidy of all cancers. (c) Metastasis has also been reduced to karyotypic flexibility generating metastatic sub-species of carcinomas with variant tissue-tropisms [50, 57, 59] – much as karyotypic flexibility generated the drug-resistant and morphological variants described above in Figures 12, 13 and 14.
Section-specific conclusions. Common carcinoma-specific characteristics like autonomy and karyotypic flexibility, which generate immortality and metastases, lend additional support to the theory that carcinoma-specific karyotypes are sufficient to create and maintain carcinomas, independent of mutations and sequences of latent viruses.
Since the individual clonal karyotypes of nine carcinomas correlated 1-to-1 and co-varied 1-to-1 with the complex individual transcriptomes and phenotypes of these carcinomas, we conclude that individual clonal karyotypes encode these carcinomas, much like the individual clonal karyotypes that encode conventional species. In sum this evidence provides genetic and functional proof for the speciation theory of carcinomas.
Indeed 1-to-1 correlations between phenotypes and karyotypes are the classical genetic basis of defining phylogenetic identities and relationships between conventional species [92, 93]. Thus the endless individual “DNA contents” and corresponding individual karyotypes, grades of malignancy and histological phenotypes of the carcinomas studied here and of those first described by Papanicolaou et al. , each support the conclusion that carcinomas are species of their own.
Our conclusion that transcriptionally very active and complex individual karyotypes, instead of mutations or latent viral sequences, encode carcinomas thus explains our failures and those of numerous others to find common, virus-specific karyotypic or phenotypic characteristics in papilloma virus-positive carcinomas [5, 12, 16, 18–21, 29, 33, 67, 70]. The carcinoma-species-theory also explains the recent finding that “HPV status … showed no correlation to outcome” .
Above all there is no direct functional evidence that cervical carcinomas depend on (1) the plethora of “passenger mutations” and “single nucleotide polymorphisms”, which are found in all cancers and normal cells by genome sequencing [94–96], or on (2) the defective and latent papilloma viral sequences (Background), which are found in 50 million or about 30% of healthy American females . If carcinomas would depend on papilloma virus functions they would be immunogenic and subject to anti-viral immunity, which is not observed [31, 34]. Indeed, the occurrence of papilloma virus-free cervical carcinomas is by itself sufficient evidence that cervical carcinomas are virus-independent (Background).
In view of this the clonal, but nonidentical viral sequences and cellular mutations found recently in individual cancers by whole genome sequencing appear to be non-carcinogenic, preneoplastic mutations, which became clonal accidentally by subsequent evolutions of clonogenic cancer-specific karyotypes.
Because of their lack of carcinogenic function such preneoplastic mutations have been termed “passenger mutations” [95–97] or “implausible” mutations . Cancer cells are thought to acquire such individual passenger or implausible mutations during their long preneoplastic histories either from conception to cancer [95, 96, 99, 100] or from virus infections 20–50 years prior to carcinogenesis . Because of these long preneoplastic histories such mutations are also termed “archaeological” mutations [95, 96].
In sum, our data indicate that newly formed carcinoma-specific karyotypes generate and maintain carcinomas, independent of latent viral sequences or mutations of tumor suppressor genes. Based on our findings it is expected that a vaccine against human papilloma viruses will have no effect on the occurrence of cervical carcinomas.
In searching for alternatives of the putative viral and mutational cervical carcinogens we found in the literature a long-known cervical carcinoma-specific carcinogen, namely cigarette smoke. Surprisingly, in view of the current viral and mutational theories, smoking was already known in the 1970s to induce cervical carcinomas [27, 80–83]. In 1983 a study of the Centers for Disease Control even proved that “12 or more pack-years smoked” increase cervical carcinoma rates 12.4-fold compared to non-smokers . The steady decline of cervical cancer in parallel with declining smoking rates in the US supports this theory . Other studies have independently shown that smoking induces preneoplastic aneuploidy in proportion to the dose of cigarettes smoked [101–105]. In addition ionizing radiation has been found to induce cervical carcinomas too . Thus, the induction of aneuploidy by cigarette smoke or other carcinogens supports the karyotypic mechanism of carcinomagenesis proposed here, and explains its characteristic time-dependence owing to the inevitably slow karyotype mutation-selection scheme outlined by the speciation theory (Background, Figure 1).
The “early” preneoplastic or even neoplastic karyotypic alterations of cervical biopsies described by several investigators lend unexpected support to the karyotypic mechanism of carcinomagenesis proposed by our theory [5, 17, 21, 33, 66, 70, 107, 108] (see also previous section). These investigators may have used the proposition “early” for karyotypic alterations, because they expected early carcinoma causing mutations instead.
The speciation theory thus presents a coherent mechanism of cervical carcinomagenesis in which carcinogens initiate carcinogenesis by inducing aneuploidy, and aneuploidy then initiates automatic chain reactions of karyotypic evolutions that eventually generate new carcinoma-specific karyotypes.
Since conventional DNA cytometry has a blind spot for near-diploid carcinomas and cancers, we propose that the karyotype analyses performed here could provide new, absolute distinctions between normal diploid but hyperplastic cervical tissues, hyperplasias with random aneuploidies and near-diploid carcinomas with clonal aneuploid DNA contents and low Pap-grades, such as C-4II and C-33A (Figures 6 and 7).
Moreover, our results promise new distinctions in grading malignancy of biopsies with high near-triploid DNA contents - based on the number of clonal versus non-clonal chromosomes, and on the corresponding chromosome recombination indices described here (Table 3). This speculation seems justified, because our results confirm the rule (Background): The more “abnormal” the “DNA content” – the higher is the flexibility and thus the malignancy of the carcinoma [1, 3–6, 10, 17, 59, 65, 67].
Thus the karyotypic mechanism of carcinomagenesis described here offers new, testable karyotypic markers to distinguish between benign diploid biopsies, as yet benign biopsies with non-clonal preneoplastic aneuploidies , and malignant carcinomas with clonal near-diploid and near-triploid karyotypes like those shown in Figures 6, 7, 8, 9, 10, 11, 12, 13 and 14.
In view of this we think our results provide the basis for a new and improved approach to the diagnosis, treatment and prevention of cervical carcinomas along the lines of the original discoveries of Papanicolaou et al. .
The papilloma virus-free human cervical carcinoma cell lines C-33A and HT-3, and the viral DNA-positive lines C-4II and CaSki were obtained from American Type Culture Collection (Manassas, VA). The cervical carcinoma lines SiHa, ME-180, MS-751, SW-756 and a second strain of CaSki were kind gifts of Pulivarthi Rao (Laboratory of Molecular Cytogenetics, Texas Children's Cancer Center, Baylor College of Medicine, Houston, Texas). The HeLa line was a kind gift of Ann Fischer (Cell culture facility at UC Berkeley, Barker Hall). Cells were propagated in cell culture medium RPMI 1640 (Sigma Co) supplemented with 5% or 10% fetal calf serum, 1% Anti-Anti (GibCo company) and 1% Nystatin (Sigma company).
One to two days before karyotyping, cells were seeded at about 50% confluence in 3 ml medium containing 5% fetal calf serum in a 5-cm culture dish. After reaching ~75% confluence, 200 ng colcemid in 20 μl solution (KaryoMax, Gibco Invitrogen company) was added to 3 ml medium. The culture was then incubated at 37°C for 4–8 hrs. Subsequently cells were dissociated with trypsin, washed once in 3 ml of physiological saline, and then incubated in 0.075 molar KCl at 37°C for 15 min. After cooling the solution in ice-water, 0.1 volume of freshly mixed glacial acetic acid-methanol (1:3, vol. per vol.) was added and the mixture was centrifuged at 800 g for 6 minutes at room temperature. The cell pellet was then suspended drop-wise in ice-cold acetic acid-methanol (1:3) solution and incubated at room temperature for 15 min. This cell suspension was then pelleted by centrifugation as described above, resuspended in an appropriate small volume and either transferred onto a microscope slide with a micropipette tip to give a suitable density of metaphases for microscopic analysis or stored at -20C. Slides with suitable metaphase chromosomes were hybridized with chromosome-specific color-coded DNA probes (MetaSystems, Newton, MA 02458), as described by the manufacturer and by us previously [45, 83], to generate the chromosome-specific colors shown above in Figure 5.
Stepanenko and Kavsan have recently published an independent speciation theory of cancer similar to the one described here .
We thank Cormac Jennings (Cancer Center of the University of Hawaii at Honolulu, HI) for a critical review and helpful comments on the manuscript, and Lars Steinmetz (European Molecular Biology Laboratory, Heidelberg, Germany) for unpublished information on their recent analysis of the HeLa transcriptome . In addition we gratefully acknowledge Alfred Boecking (Universitaetsklinikum Duesseldorf, Germany), Michele Carbone (Cancer Center of the University of Hawaii at Honolulu, HI), Alice Fabarius and Ruediger Hehlmann (University of Heidelberg at Mannheim, Germany), Claus Koehnlein (Medical practice, Kiel, Germany), Josh Nicholson (Virginia Polytech, Blacksburg, VA), Iver Petersen and Thomas Liehr (University of Jena, Germany), Jonathan Pollack (Stanford University, CA), David Rasnick (former visiting scholar at UC Berkeley), Reinhard Stindl (Alpharm GesmbH, Perchtoldsdorf, Austria), Peter Walian (Lawrence Berkeley National Laboratory, Donner Lab) and Sandra and Eric Wolman (Potomac, MD) for valuable information and critical discussions. Finally we thank Aastha Grade, Julia McKenchie and Yash Rusconi, three undergraduate interns of 2013 for catalyzing this project by their help with experiments, lively discussions and critical ideas. Two of us (PD and MD) specifically thank Michele Carbone for sponsoring a month of working on and discussing this study at the new Cancer Center of Honolulu. Bong-Gyoon Han is specifically thanked for drafting and designing Figure 1 (Lawrence Berkeley National Laboratory, Donner Lab). Our study would not have been possible without the generous support of the philanthropists Robert Leppo (San Francisco), Rajeev and Christine Joshi (London, UK), Peter Rozsa of the Taubert Memorial Foundation (Los Angeles), Howard Urnovitz (Chronix Biomedical, San Jose, CA), a Foundation that prefers to remain anonymous and other private sources.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.