References
1. Campo E, Swerdlow SH, Harris NL et al (2011) The 2008 WHO classifi cation of lym- phoid neoplasms and beyond: evolving con- cepts and practical applications. Blood
117:5019–5032. doi: 10.1182/
blood-2011-01-293050
2. Morgan R, Chen Z, Richkind K et al (1999) PHA/IL2: an effi cient mitogen cocktail for cytogenetic studies of non-Hodgkin lymphoma and chronic lymphocytic leukemia. Cancer Genet Cytogenet 109:134–137
3. Gahrton G, Robert KH, Friberg K et al (1980) Nonrandom chromosomal aberrations in chronic lymphocytic leukemia revealed by poly- clonal B-cell-mitogen stimulation. Blood 56:640–647
4. Decker T, Schneller F, Hipp S et al (2002) Cell cycle progression of chronic lymphocytic leuke- mia cells is controlled by cyclin D2, cyclinD3, cyclin-dependent kinase (cdk) 4 and the cdk inhibitor p27. Leukemia 16:327–334
5. Decker T, Schneller F, Kronschnabl M et al (2000) Immunostimulatory CpG-
oligonucleotides induce functional high affi nity IL-2 receptors on B-CLL cells: costimulation with IL-2 results in a highly immunogenic phe- notype. Exp Hematol 28:558–568
6. Dicker F, Schnittger S, Haferlach T et al (2006) Immunostimulatory oligonucleotide-induced metaphase cytogenetics detect chromosomal aberrations in 80% of CLL patients; a study of 132 CLL cases with correlation to FISH, IgV H status, and CD38 expression. Blood 108:3152–3160 7. Shi M, Cipollini MJ, Crowley-Bish PA et al
(2013) Improved detection rate of cytogenetic abnormalities in chronic lymphocytic leukemia and other mature B-cell neoplasms with use CpG-oligonucleotide DSP30 and interleukin 2 stimulation. Am J Clin Pathol 139:662–669 8. Koczkodaj D, Popek S, Zmorzyński SZ et al
(2016) Detection of chromosomal changes in chronic lymphocytic leukemia using classical cytogenetic methods and FISH: application of rich mitogen mixture for lymphocyte cultures.
J Investig Med 64:894–898. doi: 10.1136/
jim-2015-000050
43
Thomas S.K. Wan (ed.), Cancer Cytogenetics: Methods and Protocols, Methods in Molecular Biology, vol. 1541, DOI 10.1007/978-1-4939-6703-2_5, © Springer Science+Business Media LLC 2017
Chapter 5
Cytogenetic Harvesting of Cancer Cells and Cell Lines
Roderick A. F. MacLeod , Maren E. Kaufmann , and Hans G. Drexler
Abstract
We describe an evidence-based approach toward optimizing chromosome preparation from cancer cells and cell lines. The procedures described here emphasize the utility of both cell culture—to maximize the yields of the dividing cells needed to harvest mitotic metaphase chromosome preparations and how an empirical evaluation of hypotonic treatments enables optimal conditions to be effi ciently determined.
Key words Cancer chromosomes , Cell culture , Hypotonic treatment , Fixation , Slide making
1 Introduction
Despite their uncontrolled proliferation and numerous dividing cells, cancer cells nevertheless often yield chromosome prepara- tions inferior to those from benign tissue, a discrepancy seldom encountered by molecular biological approaches for which cell cycle stage is largely immaterial. As well as cell cycle dependence, the legion variety of cell types encountered in cancer—each clone being effectively unique—defeats a “one size fi ts all” approach.
Thus, cytogenetic harvesting of cancer cells should be tailored to the individual needs of each intractable cell or cell type.
Thanks to the advent of fl uorescence in situ hybridization (FISH), cytogenetics now bridges the gap between classical and pure molecular biological methodologies. It is now apparent that in addition to point mutations, a major force in carcinogenesis is the production of molecular gene alterations by chromosome rearrangements, which if recurrent, may be used diagnostically.
Accordingly, a standard work on the cytogenetics of human cancer [ 1 ] has been published to cover the fi eld, in which the fourth edi- tion to date requires 27 authors when compared to merely 2 in the second edition. As well as in hematopoietic cancers, recurrent chromosome changes have now been described in solid tumors enabling target gene identifi cation. Hence, rarer and poorly char- acterized tumors lacking known cytogenetic involvement have 1.1 Background
acquired heightened research interest to those seeking novel can- cer mechanisms. Intractable tumors may be particularly challeng- ing to harvest, or bear the most subtle cytogenetic changes where chromosome quality is at a premium. Cancer cytogenetics helps interpretation of genomic array data by providing a window onto individual cells and intercellular heterogeneity [ 2 ].
A basic cytogenetic requirement is the provision of dividing cells which must be arrested at mitosis when individual chromo- somes and their structure become microscopically visible enabling cancer chromosome rearrangements to be discerned. We are now able to tackle these complexities thanks to FISH-based methodol- ogies which, in turn, demand chromosome preparations of the best quality possible. Clinical cytogenetic protocols rarely allow the empirical optimization of cell harvest conditions since repeat diag- nostic samples are seldom available. These exigencies may in some cases be obviated by recourse to cell culture which enables cell numbers to be expanded. Cancer cell lines offer infi nite replication and repetition by providing unlimited material, allowing analyses of a scope and depth that usually denied to those working with primary cancer cells.
Our experience has taught us that cell harvest conditions, espe- cially choice of hypotonic treatments, are the most critical to success with intractable cell types. Hence, a little effort invested in optimiz- ing these conditions to individual cultures is well worth the effort.
The complex chromosomal rearrangements of cancer cells often present a challenge to cytogenetic analyses and require multiple rounds of FISH to achieve meaningful results. Such approaches are greatly facilitated by cryopreservation, of metaphase cell sus- pensions at −20 °C, microscope slides bearing such metaphases at
− 80 °C, and living cells themselves in liquid nitrogen at
− 196 °C. Microscope slides bearing metaphase-enriched cell sus- pensions are the bread and butter of cytogenetics and their assess- ment underlies any scheme for evidence-based hypotonic optimization. Hence, mitotic metaphase chromosomes must be analyzed rationally. Consequently, three key criteria that are likely to impact on subsequent analyses are metaphase quantity, chromo- some spreading, and chromosome morphology.
1. Metaphase quantity ( see Fig. 1a, b ): To enrich metaphase num- bers growing cultures are treated with colcemid : its key activ- ity—microtubule depolymerisation—inhibits formation of mitotic spindles leading to metaphase arrest. Colcemid is toxic limiting the degree of effective exposure. Other agents have been proposed, e.g. vinblastine or colchicine, but they have proved to be more expensive or even more toxic.
To quantify mitotic suffi ciency (on a scale of A to C), as a rule of thumb, low power (100×) fi elds should carry on average 1.2 Harvesting Aims
Fig. 1 Impact of harvesting conditions on harvesting mitotic metaphase chromosomes for cytogenetic analysis. ( a, b ) Images show how severe hypotonic treatments can deplete harvests of mitotic cells which easily might be misinterpreted as “low mitotic index .” While cells shown in ( a ) were subjected to “standard” hypotonic treatment (0.075 M KCl for 7 min) which effectively burst mitoses, those shown in ( b ) received a milder treatment (1:1 = 0.075 M KCl + 0.9 % sodium citrate, 1 min). (c – e) The metaphase images show different levels of spreading without great impact on chromosome quality. Image ( c ) is clearly broken due to overlong hypotonic exposure with concomitant chromosome loss. That in ( e ) shows excessive crossovers hampering analysis, while that in ( d ) allows chromo- somes to be easily seen without undue risk of chromosome loss. (f–h) Images illustrate chromosome quality. While in ( f ) chromosomes are adequately spread and chromatids lie in parallel. The solid in appearance of chromosomes in ( f ) is likely to yield good G- banding and FISH, while those in ( g ) and ( h ) are substandard. The arms of those in image ( g ) lie akimbo, a confi guration inimical to subsequent G-banding or FISH. In ( h ) spreading and morphology are clearly substandard, the chromosomes too small and overlaid for analysis. Counterintuitively perhaps, the poor chromosomes in ( h ) were produced after an overly harsh hypotonic treatment
at least one usable metaphase to be deemed “A”; one per hori- zontal row scores a “B”; while fewer than that gets a “C.”
“AB” or better is necessary for G- banding , while “B” is neces- sary for chromosome painting and “A” for the most challeng- ing FISH- based methods, if only to justify the additional outlay involved.
In our experience, counterintuitively perhaps, the most common ground for mitotic insuffi ciency is not lackadaisi- cal proliferation, rather an overly harsh hypotonic treatment which increases the cell breakage and yields a “chromosome soup.” Hence, the appropriate responses to lack of metaphases should include the following measures: (a) increasing the amount of “soft” buffer (e.g., sodium citrate) at the expense of
“harsh” KCl, avoidance of warm incubation at 37 °C, or even performing this step on ice; (b) reducing incubation times down to even a few seconds; and (c) reducing the amount of hypotonic used in relation to residual medium remaining after centrifugation.
2. Chromosome spreading ( see Fig. 1c, e ): To facilitate analysis chromosomes must ideally lie grouped together and evidence mitotic integrity: broken cells which have lost chromosomes serve to confuse analysis and may yield false ascertainment of aneuploidies . On the other hand, chromosomal superimposi- tion—“crossovers”—should be avoided lest detail be sacrifi ced thereby. To improve spreading cells are exposed to what is termed “hypotonic shock”, though the underlying biophysical mechanisms linking it to chromosome morphology remain elusive [ 3 ]. After hypotonic treatment, nuclei release their chromosomal contents. Hypotonic treatments must be fi nely tuned to individual cell types, however, since excessive nuclear breakage may be accompanied by chromosome losses.
Importantly, we have found that matching hypotonic treatments to individual cell requirements is the single control- lable factor which contributes most to chromosome quality.
This requires testing replicate cell cultures using a variety of hypotonic treatments to determine optimal conditions, repeat- ing if necessary until successful [ 4 ].
In addition to hypotonic tuning, chromosome spreading may be controlled by local humidity and viscosity encoun- tered by fi xed cells when they hit the target microscope slide.
Hence, slides should be cleaned by washing in diluted HCl, followed by ethanol and polishing using a lint-free cloth.
Precleaned slides are then kept at −20 °C until immediately before use, their thin frost layers turning rapidly to moisture which encourages spreading. Paradoxically, overly spread metaphases and those remaining tight may both, again, be caused by overly harsh hypotonic treatments. In extreme cases, so much inherently sticky DNA is released that meta-
phases are effectively cocooned by it. The same considerations and remedial measures are applied again as described in the preceding section.
3. Chromosome quality ( see Fig. 1f–h ): The third basic criterion described here is best defi ned pictorially. Although the term
“quality” is subjective, “good chromosome quality” is seldom in doubt. Ideally, chromosomes should appear dense under phase contrast or after solid staining, but never retractile with troublesome halos. Arms should lie straight in parallel, never akimbo, as this impairs band formation and is associated with
“fl uffi ness” inimical to banding resolution.
There is little doubt that cancer chromosome quality fails to match normal cells, whether lymphocytes, keratinocytes, or amniocytes. This shortcoming is often blamed on the altered chromatin states involved in gene dysregulation present in can- cer. Alternatively, the immature cell types giving rise to various cancers may be unsuited to standard hypotonic treatments.
Poor chromosome quality may be due to the use of suboptimal media or fetal bovine serum (FBS) supplements in certain cases ( see Note 1 ).
Cells may be harvested directly from tumor biopsies or after cul- ture. In the case of continuous cell lines, cultures are long-term and the number of population cell doublings is seldom—if indeed ever—known with any precision. Clonal heterogeneity which is a key characteristic of cancer cell populations may be lost during pas- sage, but a recent study showing long-term subclonal stability sug- gests that this fear may be overstated [ 5 ]. In the case of primary cancer cells, short-term cultures (usually <72 h) may be used to expand the mitotic cell fraction. If cancer samples also contain nor- mal stroma tissue, the risk of its inadvertent analysis at the expense of the neoplastic moiety demands special consideration, e.g., via cytogenetic data, though this is obviously uninformative when poor quality hampers detection.