Page designed and maintained by Octavian Henegariu (Email:or).
WARNING: The information provided in these pages is copyrighted and is intended for individual use only. No parts of this work (text, tables or pictures) may be commercialized, published or otherwise reproduced without the written consent of the author.
Ref: Laboratory Investigation. 2001 Apr;81(4):475-81.Get article in PDF format here
Small marker chromosome identification in metaphase and interphase using centromeric multiplex FISH (CM-FISH)
Octavian Henegariu1, Patricia Bray-Ward1, Sevilhan Artan1, Gail H.Vance2, Mazin Qumsyieh1 and David C. Ward1
1Department of Genetics, Yale University School of Medicine, 333 Cedar Street, New Haven, CT 06510
2Department of Medical and Molecular Genetics, Indiana University School of Medicine, 975 W Walnut Street, Indianapolis, IN 46202
Multicolor karyotyping procedures, such as M-FISH, SKY or CCK, can be used to detect chromosomal rearrangements and marker chromosomes in prenatal diagnosis, peripheral blood cultures, leukemia and solid tumors, especially in cases where G-banding is not sufficient. A regular M-FISH analysis requires relatively large amounts of labeled DNA (micrograms), is not informative in interphase nuclei, hybridization can take up to 2-3 days and unlabeled human chromosome painting probes are not available commercially. Unique probes (plasmids, PACs), specific for centromeric or subtelomeric chromosomal regions, can replace the painting probes in M-FISH to answer specific questions, such as the identification of marker chromosomes and aneuploidies. A set of plasmid probes carrying repetitive sequences specific for the alpha-satellite region of all human chromosomes were combined in a metaphase assay and an interphase assay, allowing identification of aneuploidies in one hybridization step, on a single cytogenetic slide. The fluor-dUTPs and the labeled antibodies required to label and detect the DNA probes can be prepared in any laboratory. All DNA probes can be easily isolated and labeled using common molecular cytogenetic procedures. Because of the repetitive nature of the probes, hybridization time is short, usually less than one hour, and the analysis can be performed with non-specialized image processing software.
Several multicolor karyotyping procedures, such as multiplex FISH (M-FISH), spectral karyotyping (SKY) or color-changing karyotyping (CCK) (Henegariu et al, 1999; Schrock et al, 1996; Speicher et al, 1996) were introduced in the past several years, and were successfully used to identify complex structural and numerical chromosome aberrations. These techniques were applied in prenatal diagnosis (Uhrig et al, 1999), peripheral blood cultures and leukemia and solid tumors (Haddad et al, 1998; Huang et al, 1998), especially in cases where G-banding was not sufficient to identify the chromosome of origin. Any multicolor karyotyping procedure requires relatively large amounts of labeled chromosome painting probe DNA (micrograms), works only on metaphase chromosomes, takes up to 2-3 days to hybridize and unlabeled chromosome painting probes are not available commercially. As an alternative approach, the use of unique probes (plasmids, PACs), specific for centromeric or subtelomeric chromosomal regions, can replace the painting probes in multiplex FISH assays to answer specific questions such as aneuploidy or marker chromosome detection. Probes specific for the alpha satellite region of human chromosomes have been successfully used for many years to identify major prenatal aneuploidies, as well as numerical aberrations in human tumors (Huegel et al, 1995; Pergament, 2000; Ried et al, 1992; Zhao et al, 1998). In this article, we are describing a multiplex FISH procedure combining the centromeric probes of all human chromosomes in one single assay. This approach allows the identification of marker chromosomes and other aneuploides on one single cytogenetic slide, in less than two hours. A similar procedure was proposed in abstract format by others (Heller et al, 2000; Nietzel et al, 1999) but was not detailed. Probes can be prepared, labeled and combined in any cytogenetic laboratory, using standard molecular cytogenetic techniques, and the analysis can be performed on any computer platform using generic imaging software, such as Adobe Photoshop.
Results and discussion
We used a set of µ-satellite probes combined in an algorithm detailed in the tables of Figures 1 and 2, to achieve hybridization and separate detection of centromeres of all human chromosomes (except 13/21) on the same slide. This novel M-FISH diagnostic procedure is called centromeric M-FISH (or CM-FISH) and is used for one-step diagnosis of chromosomal aneuploidies (Fig. 1). Two separate strategies were developed: in (1), the "metaphase assay", or CM-FISH, human centromeric probes were labeled using combinatorial labeling, then mixed together and hybridized on the same cytogenetic preparation. This approach is used for small marker chromosome identification, and requires the presence of metaphases. In the second strategy, (2) the "interphase assay" or iCM-FISH, the centromeric probes are divided into three groups, which are hybridized on three separate areas of the same cytogenetic preparation. This approach allows identification of aneuploidies in interphase nuclei, and does not require cell culture prior to FISH.
CM-FISH provides several advantages over other multicolor diagnostic procedures. The DNA probes do not require competitor DNA during hybridization, FISH signals from just minute amounts of labeled DNA are very strong, and all probes are available as plasmids (Choo et al, 1991; Durm et al, 1998), thus offering a virtually unlimited source of fresh DNA. Hybridization time of 0.5-2 hours is sufficient, allowing same-day diagnosis of a clinical sample. The main disadvantage is the lack of separation between the centromeres of chromosomes 13 and 21 (Maratou et al, 1999) and the impossibility to detect marker chromosomes lacking µ-satellite sequences (Magnani et al, 1998; Wandall et al, 1998).
1. Metaphase assay (regular CM-FISH)
This approach is especially useful in those cases in which a small marker chromosome is diagnosed in a clinical sample by routine cytogenetic techniques (GTG banding). In most such cases, CM-FISH can identify the origin of the marker on the same day.
Combining centromeric probes in a CM-FISH assay faced specific problems not encountered with regular M-FISH. Because of common alphoid subfamilies (Grady et al, 1992; Jorgensen, 1997; Lee et al, 1997; Mitchell, 1996; Sullivan et al, 1996; Wevrick and Willard, 1989), many centromeric DNA probes hybridize to several chromosomes simultaneously, and yield hybridization signals quite different in size and strength. To overcome such problems, the probes were labeled based on a six fluor-algorithm , including both combinatorial labeling (same probe labeled with combinations of two or three fluors) (Ried et al, 1992) and ratio labeling (which uses different amounts of dye to differentiate probes from one another) (Tanke et al, 1999). A combination of these labeling strategies allowed labeling of each DNA probe by no more than two fluors at the same time, thus facilitating separation of signals at the microscope (Fig 1 table). The nucleotides used were FITC-dUTP, Cy3-dUTP, Cy5-dUTP, digoxigenin "DIG"-dUTP (detected with antiDIG-diethyl aminomethyl coumarin "DEAC"), dinitrophenyl "DNP"-dUTP (detected with antiDNP-Cy5.5) and biotin "BIO"-dUTP (detected with avidin Cy7 or aminomethyl coumarin "AMCA"). Cy3.5 and Texas Red (fluors with similar absorption/emission characteristics) could not be used in the assay, as the strength of the Cy3 signal made it detectable through the Cy3.5 filter. Simultaneous use of all centromeric probes, required a low post-hybridization washing stringency (0.2xSSC, 15 minutes at 42?C). Although complex, this multiplex approach allowed simultaneous detection of all centromeres. To decrease cross-hybridization, the procedure could benefit from replacing some of the plasmid probes with labeled oligonucleotide probes (Warburton et al, 1991), with a higher specificity for the respective chromosomes.
Based on their hybridization characteristics, the DNA probes were divided into two arbitrary groups, group A including probes hybridizing to chromosomes with shared alphoid subfamilies (chromosomes 1-5-16-19; 2-18-20; 4-9; 13-21; 14-22) and group B including probes hybridizing primarily to only one chromosome. Group A probes required combinatorial and ratio labeling, whereas group B probes required combinatorial labeling only (Fig 1 table and Fig 2). To illustrate this, the DNA probes for chromosomes 4 and 9, which partially cross-hybridize to one another, were both labeled with Cy5 and DIG (combinatorial labeling). The Cy5 signal was stronger on chromosome 4 and the DIG/DEAC signal was stronger on chromosome 9, this difference allowing chromosome identification (ratio labeling)."A particular case is that of chromosomes 13 and 21, which cannot be separately identified by FISH with µ-satellite probes. If a 13/21 marker is found, it can be subsequently identified using painting probes or pericentromeric unique probes, such as PACs (B39I12 and 126N20), on chromosome 21 and YACs (748f2 and 967b1) on 13. In over fifteen cases tested by CM-FISH (both normal controls and clinical samples) the hybridization patterns on every chromosome pair were constant. The only chromosome showing FISH signal of variable intensities even by visual inspection, were chromosomes 13 and 21.
CM-FISH does not require karyotyping software, as it is performed on cases already known cytogenetically by G-banding. In CM-FISH, once the marker is located in the metaphase either by DAPI counterstaining or by overexposing the image in the DEAC or FITC channel (where tissue autofluorescence is highest), only the flourescence signals on the marker are visualized or captured in all six channels. The marker is identified by interpreting the hybridization pattern. Image analysis and pseudocoloring can be easily performed with generic image processing software, such as Adobe Photoshop.
2.Interphase assay (iCM-FISH)
This CM-FISH approach allows detection of all centromeres in interphase nuclei on the same slide, an important advantage for early diagnosis of aneuploidies in leukemias, solid tumors and in prenatal screening. Three separate probe combinations (mixtures C1, C2, C3 - Fig 2 table) were necessary to reliably detect all centromeres, and six fluors were required to label all DNA probes. The three probe mixtures are simultaneously hybridized on different areas of the same slide. Although efforts were made to keep all probes used for a particular type of analysis (such as prenatal diagnosis) in the same mixture, this was possible only to some extent, as attention was also paid to preventing each probe mixture from becoming too complex. In every nucleus, any one of the six fluors used shows four to six hybridization signals. Of the three mixtures created (Fig 2), C1 is aimed more at diagnosing aneuploidies in leukemias, whereas C2 is aimed more at prenatal diagnosis. Mix C1 is the most complex, as it includes probes yielding 6 fluorescent signals in three channels (Fig. 2). Interphase CM-FISH requires a good slide preparation technique (Henegariu* et al, 2001), to make the nuclei as "flat" (bi-dimensional) as possible. To test the procedure we used bone marrow and peripheral blood samples in which virtually all nuclei showed the specific aneuploidy. In our hands, about 3% (bone marrow) and 8% (peripheral blood) of all nuclei showed analyzable signals in all six fluorescent channels. The reason for this seemingly low number is the complexity of the probe mixture: in each channel, four or six centromeres have to be simultaneously visible and the signals have to be in the same plane of the image. Additionally, some centromeres work better than other, and not all labeling reactions are identical. Small variations in labeling and sample preparation have a much higher impact in signal quality when using these probes than with the use of chromosome painting libraries. In fact, the numbers found are in accordance with theoretical calculations: when counting centromere signals in each channel (one fluor), we found that approximately 60 % of nuclei had all expected 4-6 centromeric signals visible in the same plane. If the hybridization efficiency is identical for each of the six fluorophores used in an assay, 60% useful nuclei/channel should result in 4.7% of nuclei with separable signals in all channels, a number close to the experimental values. In general the number of informative nuclei is higher for peripheral blood cultures because of the better nuclear morphology after spreading. Once an aneuploidy is identified by iCM-FISH screening, in order to count the number of interphase signals of that probe in a few hundred nuclei, the more convenient approach is to hybridize a separate slide with the respective centromeric probe labeled with a visible fluor (such as FITC, Cy3, rhodamine). This is useful especially when the respective aneuploidy is identified by a iCM-FISH probe, which happened to be labeled by a infrared dye (Cy5, Cy5.5 or Cy7). All infrared dyes require a CCD camera for visualization. Separate hybridization with a centromeric probe is easy, short (takes only 30 minutes) and the signals easy to score in numerous nuclei. It is conceivable that some researchers would wish to hybridize all centromere probes on one slide AND also have the flexibility of scoring hybridization signals in hundreds of nuclei on the same slide. This would require the use of only visible fluors for labeling. Consequently, the number of probes in a set needs to be reduced. A simple solution is to divide each of the three centromere sets described (C1, C2, C3) into two subsets and then hybridized each subset on a different area of the same cytogenetic slide. In our laboratory, up to eight different hybridizations could be simultaneously performed on the same slide, by cutting 22x22mm coverslips into quarters, using a regular diamond pen. Eight such coverslip pieces can easily fit on one slide. For laboratories equipped with more sophisticated equipment and commercial software packages (such as the PowerGene M-FISH package), scoring infrared signals in many nuclei is possible by using the live camera view provided by the software. This enables visualization and counting of the infrared dye signals on the computer screen.
The main advantage of iCM-FISH is the use of interphase nuclei, with no need for time-consuming cell culture. All µ-satellite sequences can be identified within 1-2 hours from the start of the analysis. As with regular CM-FISH, the main disadvantage of the procedure is the lack of separation between chromosomes 13 and 21, a problem which can be addressed by using pericentromeric BAC, PAC or YAC clones on either one of these two chromosomes.
Materials and Methods
The following centromeric probes ("a")were used: a 1= pZ16A* (HF Willard, similar to pE25b at ATCC); a 2= pBS4D* (A Baldini); a 3= p3-9 (HF Willard); a 4= pG-XbaII/340* (T Hulsebos); a 5= pC1.8 (T Hulsebos) ; a 6= a RI12 (ATCC); a 7= pZ7B (HF Willard); a 8= pJM128 (ATCC); a 9= pMR9A* (A Baldini); a 10= pa 10RP8 (HF Willard); a 11= pLC11A (HF Willard); a 12= pa 12H8 (ATCC); a 13/21= L1.26* (P Devilee); a 14/22= a XT(680) (AL Jorgensen)*; a 15= pTRA-25 (A Choo); a 16= see a 1 (if wanted, pSE16 can be added to the CM-FISH); a 17= p17H8 (HF Willard) or pYAM 7-29 (YB Yurov); a 18= L1.84 (P Devilee)*; a 19= pG-A16 (T Hulsebos) *; a 20= pZ20 (A Baldini)*; a 21= see 13; a 22= p22/1:2.1;2.8;0.73 (HE McDermid); a X= pXBR-1 (C Disteche); a Y= pDP97 (D Page, ATCC). * indicates probes constantly hybridizing on more than one chromosome pair. The name of the individuals who provided the initial DNA clone to the laboratory, either at Yale or at Indiana University, are also shown.
Probe labeling and hybridization
All probes (plasmids) were prepared by alkaline lysis and were labeled by regular nick translation, at 10-15 ng/µl DNA concentration. When the combination algorithm required the same probe to be labeled with two dyes, separate nick translation reactions were performed, and the labeled DNA combined in the proportions depicted in Figure 1,Table. Separate labeling reactions of 3-4 µg of every plasmid DNA were performed and stored. 2-3ul of every labeling reaction were used in separate hybridization experiments to assess hybridization quality, chromosomes yielding signals and relative signal strength (visual approximation). In the next step, the labeled probes were pooled as depicted in Figure 1 Table, using 1µl each labeling reaction for the "strong" probes (for example chromosomes 1;5;16;19; 3; 14; 15; 17; 18) and 2 µl each for the weaker probes. After hybridization, the strength of various hybridization signals on chromosomes was again visually assessed, and the amount of each labeled probe adjusted for identification. In our hands ten consecutive experiments were required until the working combination was found. Larger amounts of labeled probe were then combined into a working pool, to be used for 50-100 slides. This labeled DNA pool was precipitated, resuspended in hybridization buffer and stored at -20 C. One aliquot (10 µl) was used to hybridize one slide.
All labeled nucleotides used in nick translation were custom synthesized in our laboratory as previously described (Henegariu et al, 2000). The fluor- or hapten-dUTPs used were: FITC-dUTP, Cy3-dUTP, Cy5-dUTP, DEAC-dUTP, DIG-dUTP, DNP-dUTP and BIO-dUTP. Labeled antibodies were either purchased or prepared in our laboratory by standard protein-dye conjugation protocols (Molecular Probes, Eugene, OR and Amersham Pharmacia, Arlington Heights, IL). DIG was detected with sheep antiDIG (Accurate Chemical, Westbury, NY) conjugated with DEAC or Cy5.5; DNP was detected with rat antiDNP (Accurate Chemical) conjugated with Cy5.5; BIO was detected with avidin-AMCA (Accurate Chemical) or avidin (Vector Laboratories, Burlingame, CA) conjugated with Cy7. After nick translation labeling, DNA from every probe was pipetted into the same vial (common pool), according to the volumes depicted in Figure 1,Table. The DNA was ethanol precipitated, resuspended in hybridization buffer (containing 50% formamide) denatured and hybridized. Similar results were obtained using hybridization times from 1 hour to three days. After hybridization, slides were rinsed in 50% formamide/2xSSC for 10 minutes and 0.2xSSC for 15 minutes at 42? C, followed by antibody incubation. One microliter of each antibody stock solution (1mg/ml) was pipetted in the same vial containing 100 ml of 4xSSC, vortexed, placed on the slide and covered with a coverslip. After 5-10 minutes incubation at 37? C, each slide was rinsed for 10 minutes in wash solution (4xSSC, 0.1% Tween), stained with DAPI (if necessary), dried, mounted with antifade and analyzed. No particular modifications in the common FISH protocol are necessary, as centromeric probes yield very strong signals.
Image capturing and analysis
Images were captured using a Sensys (Photmetrics, Tucson, AZ) cooled CCD camera and were pseudocolored using either the PowerGene M-FISH package (PSI, Inc) or generic software (Adobe Photoshop). In CM-FISH, the role of DAPI counterstaining is only to provide the location of the marker chromosome in the metaphase. If DAPI staining is not performed (DAPI and AMCA have similar absorption spectra), it can be simply replaced by exposing the metaphase longer (usually 1.5-3 seconds) in the DEAC or DAPI/AMCA channel ("pseudo-DAPI" image). Tissue auto-fluorescence is more than sufficient to show the chromosomes and help identify the marker (Fig 1d). This allows the use of avidin AMCA, without the requirement for subsequent coverslip removal, slide rinsing and DAPI staining. Because CM-FISH is aimed at identifying a marker chromosome (usually with a morphology quite different than the normal chromosomes) in the metaphase, a specialized software package is not necessary for the analysis itself. Images in each channel can be captured using any available system (digital camera, video CCD camera, cooled CCD camera) and transferred to the hard drive as grayscale images. The six channels plus the DAPI or pseudoDAPI image (if available) can be merged as a "multichannel" image with 7 channels, using generic imaging software (Adobe Photoshop). Pseudo-coloring of all images is also not necessary. As the position of the marker in the metaphase is known from the pseudoDAPI image, marker analysis is easily performed by simple visualization of its hybridization signals in all six channels. The signal pattern found is compared with a known chart or table (such as Figure 1, table) to identify the origin of the marker. In interphase CM-FISH (iCM-FISH), aneuploidies are identified by counting the number of FISH signals in every fluorescence channel. Overlapping the gray-scale images is useful for precise identification of the origin of every signal. Any FISH software package or generic image analysis software are equally useful for this purpose.
Choo KH, Vissel B, Nagy A, Earle E, Kalitsis P (1991). A survey of the genomic distribution of alpha satellite DNA on all the human chromosomes, and derivation of a new consensus sequence. Nucleic Acids Res 19:1179-82
Durm M, Sorokine-Durm I, Haar FM, Hausmann M, Ludwig H, Voisin P, Cremer C (1998). Fast-FISH technique for rapid, simultaneous labeling of all human centromeres. Cytometry 31:153-62
Grady DL, Ratliff RL, Robinson DL, McCanlies EC, Meyne J, Moyzis RK (1992). Highly conserved repetitive DNA sequences are present at human centromeres. Proc Natl Acad Sci U S A 89:1695-9
Haddad BR, Schrock E, Meck J, Cowan J, Young H, Ferguson-Smith MA, du Manoir S, et al (1998). Identification of de novo chromosomal markers and derivatives by spectral karyotyping. Hum Genet 103:619-25
Heller A, Nietzel A, Rocchi M, Starke H, Fiedler W, Claussen U, Liehr T (2000) Centromere specific mult-color FISH (cenM-FISH) - a new and rapid method for the identification of marker chromosomes. Eur J Hum Genet 8 (Suppl 1):87
Henegariu O, Bray-Ward P, Ward DC (2000). Custom fluorescent-nucleotide synthesis as an alternative method for nucleic acid labeling. Nat Biotechnol 18:345-348
Henegariu O, Heerema NA, Bray-Ward P, Ward DC (1999). Colour-changing karyotyping: an alternative to M-FISH/SKY [letter]. Nat Genet 23:263-4
Henegariu* O, Heerema NA, Wright LL, Bray-Ward P, Ward DC, Vance GH (2001). Improvements in cytogenetic slide preparation: controlled chromosome spreading, chemical aging and gradual denaturing. Cytometry In press [Jan 2001]:In press
Huang B, Ning Y, Lamb AN, Sandlin CJ, Jamehdor M, Ried T, Bartley J (1998). Identification of an unusual marker chromosome by spectral karyotyping. Am J Med Genet 80:368-72
Huegel A, Coyle L, McNeil R, Smith A (1995). Evaluation of interphase fluorescence in situ hybridization on direct hematological bone marrow smears. Pathology 27:86-90
Jorgensen AL (1997). Alphoid repetitive DNA in human chromosomes. Dan Med Bull 44:522-34
Lee C, Wevrick R, Fisher RB, Ferguson-Smith MA, Lin CC (1997). Human centromeric DNAs. Hum Genet 100:291-304
Magnani I, Meneveri R, Marozzi A, Ginelli E, Fuhrman-Conti AM, Darfler M, Monard SP, et al (1998). Functional marker chromosomes lacking detectable alpha satellite DNA: a tool for centromere study. Chromosome Res 6:69-71
Maratou K, Siddique Y, Kessling AM, Davies GE (1999). Novel methodology for the detection of chromosome 21-specific alpha- satellite DNA sequences. Genomics 57:429-32
Mitchell AR (1996). The mammalian centromere: its molecular architecture. Mutat Res 372:153-62
Nietzel A, Starke H, Heller A, Fiedler W, Senger G, Gebhart E, Rocchi M, Claussen U, Liehr T (1999) Characterization of small marker chromosomes by centromere specific 24-color FISH. Cytogenet Cell Genet 85:40
Pergament E (2000). The application of fluorescence in-situ hybridization to prenatal diagnosis. Curr Opin Obstet Gynecol 12:73-6
Ried T, Baldini A, Rand TC, Ward DC (1992). Simultaneous visualization of seven different DNA probes by in situ hybridization using combinatorial fluorescence and digital imaging microscopy. Proc Natl Acad Sci U S A 89:1388-92
Schrock E, du Manoir S, Veldman T, Schoell B, Wienberg J, Ferguson-Smith MA, Ning Y, et al (1996). Multicolor spectral karyotyping of human chromosomes [see comments]. Science 273:494-7
Speicher MR, Gwyn Ballard S, Ward DC (1996). Karyotyping human chromosomes by combinatorial multi-fluor FISH. Nat Genet 12:368-75
Sullivan BA, Schwartz S, Willard HF (1996). Centromeres of human chromosomes. Environ Mol Mutagen 28:182-91
Tanke HJ, Wiegant J, van Gijlswijk RP, Bezrookove V, Pattenier H, Heetebrij RJ, Talman EG, et al (1999). New strategy for multi-colour fluorescence in situ hybridisation: COBRA: COmbined Binary RAtio labelling. Eur J Hum Genet 7:2-11
Uhrig S, Schuffenhauer S, Fauth C, Wirtz A, Daumer-Haas C, Apacik C, Cohen M, et al (1999). Multiplex-FISH for pre- and postnatal diagnostic applications. Am J Hum Genet 65:448-62
Wandall A, Tranebjaerg L, Tommerup N (1998). A neocentromere on human chromosome 3 without detectable alpha- satellite DNA forms morphologically normal kinetochores. Chromosoma 107:359-65
Warburton PE, Greig GM, Haaf T, Willard HF (1991) PCR amplification of chromosome-specific alpha satellite DNA: definition of centromeric STS markers and polymorphic analysis. Genomics 11(2):324-33
Wevrick R, Willard HF (1989). Long-range organization of tandem arrays of alpha satellite DNA at the centromeres of human chromosomes: high-frequency array-length polymorphism and meiotic stability. Proc Natl Acad Sci U S A 86:9394-8
Zhao L, Khan Z, Hayes KJ, Glassman AB (1998). Interphase fluorescence in situ hybridization analysis: a study using centromeric probes 7, 8, and 12. Ann Clin Lab Sci 28:51-6