Multicolor spectral karyotyping of human chromosomes

Science

Volumn 273, Issue 5274, 1996, Pages 494-497

  Schrock E ^a   Du Manoir, S ^a   Veldman, T ^a   Schoell, B ^a   Wienberg, J ^b   Ferguson Smith, M A ^b   Ning, Y ^a   Ledbetter, D H ^a   Bar Am, I ^c   Soenksen, D ^d   Garini, Y ^e   Ried, T ^a  

^a NATIONAL HUMAN GENOME RESEARCH INSTITUTE   (United States)

^b UNIVERSITY OF CAMBRIDGE   (United Kingdom)

^c TEL AVIV UNIVERSITY   (Israel)

^d Applied Spectral Imaging Inc   (United States)

^e Applied Spectral Imaging   (Israel)

Author keywords

[No Author keywords available]

Indexed keywords

ARTICLE; CHROMOSOME ANALYSIS; COLOR; COMPUTER ANALYSIS; DNA PROBE; FLUORESCENCE IN SITU HYBRIDIZATION; FOURIER ANALYSIS; GENOME; HUMAN; IMAGING; KARYOTYPE; MICROSCOPY; PRIORITY JOURNAL; SPECTROSCOPY;

EID: 0038214755     PISSN: 00368075     EISSN: None     Source Type: Journal    
DOI: 10.1126/science.273.5274.494     Document Type: Article

References (15)

1. M. M. Le Beau, in Important Advances in Oncology, V. T. DeVita, S. Hellman, S. Rosenberg, Eds. (Lippincott. Philadelphia, 1993), pp. 29-45.
2. Z, Malik et al., J. Microscopy, in press; Y. Garini et al., in Fluorescence Imaging Spectroscopy and Microscopy, X. F. Wang and B. Herman, Eds. (Wiley, New York, 1996), vol. 137, pp. 87-124.
3. Z, Malik et al., J. Microscopy, in press; Y. Garini et al., in Fluorescence Imaging Spectroscopy and Microscopy, X. F. Wang and B. Herman, Eds. (Wiley, New York, 1996), vol. 137, pp. 87-124.
4. M. R. Speicher, S. G. Ballard, D. C. Ward, Nature Genet. 12, 368 (1996).
5. Metaphase chromosomes from normal individuals, from patients with chromosomal abnormalities, and from tumor cell lines were prepared by standard methods. G-banding procedures followed standard protocols. We generated the chromosome painting probes by amplification of flow-sorted chromosomes with direct incorporation of fluorochrome-conjugated deoxyuridine 5′-triphosphate (dUTP) using degenerate oligonucleotideprimed polymerase chain reaction (DOP-PCR) [H. Telenius et al., Genes Chromosomes Cancer 4, 267 (1992)]. Five different fluorochrome-conjugated nucleotides (Cy2-dUTP, Spectrum Green-dUTP, Cy3-dUTP, Texas Red-dUTP, Cy5-dUTP) were used to label all 24 chromosomes (Amersham Life Science, Vysis, Molecular Probes). Combinatorial fluorescence was produced by combining differentially labeled chromosome painting probes. Hybridization and detection was essentially done as described [T. Ried, A. Baldini, T. C. Rand, D. C. Ward, Proc. Natl. Acad. Sci. U.S.A. 89, 1388 (1992)]: 100 ng of each chromosome-specific library was ethanol precipitated in the presence of 10 μg of human Cot-1 DNA and resuspended in 10 μl of hybridization solution (50% formamide, 2× standard saline citrate, 10% dextran sulfate). After hybridization, the chromosome specimens were counterstained with 4′,6′-diamidino-2-phenylindole (DAPI), embedded in 1,4-diazabicycls[2.2.2] octane (DABCO)-glycerol and visualized on a Leica DMIRBE microscope equipped tor epifluorescence. The DAPI banding was imaged with a cooled CCD camera (Photometrics, Tucson, AZ) or by spectral imaging through a DAPI-specific optical filter.
6. Metaphase chromosomes from normal individuals, from patients with chromosomal abnormalities, and from tumor cell lines were prepared by standard methods. G-banding procedures followed standard protocols. We generated the chromosome painting probes by amplification of flow-sorted chromosomes with direct incorporation of fluorochrome-conjugated deoxyuridine 5′-triphosphate (dUTP) using degenerate oligonucleotideprimed polymerase chain reaction (DOP-PCR) [H. Telenius et al., Genes Chromosomes Cancer 4, 267 (1992)]. Five different fluorochrome-conjugated nucleotides (Cy2-dUTP, Spectrum Green-dUTP, Cy3-dUTP, Texas Red-dUTP, Cy5-dUTP) were used to label all 24 chromosomes (Amersham Life Science, Vysis, Molecular Probes). Combinatorial fluorescence was produced by combining differentially labeled chromosome painting probes. Hybridization and detection was essentially done as described [T. Ried, A. Baldini, T. C. Rand, D. C. Ward, Proc. Natl. Acad. Sci. U.S.A. 89, 1388 (1992)]: 100 ng of each chromosome-specific library was ethanol precipitated in the presence of 10 μg of human Cot-1 DNA and resuspended in 10 μl of hybridization solution (50% formamide, 2× standard saline citrate, 10% dextran sulfate). After hybridization, the chromosome specimens were counterstained with 4′,6′-diamidino-2-phenylindole (DAPI), embedded in 1,4-diazabicycls[2.2.2] octane (DABCO)-glycerol and visualized on a Leica DMIRBE microscope equipped tor epifluorescence. The DAPI banding was imaged with a cooled CCD camera (Photometrics, Tucson, AZ) or by spectral imaging through a DAPI-specific optical filter.
7. Spectral images were acquired and analyzed with the SD200 spectral bio-imaging system (Applied Spectral Imaging, Ltd., Migdal Haemek, Israeli. The optical arrangement is schematically presented in Fig. 1. The SD200 imaging system attached to an inverted microscope (Leica DMIRBE) by means of a C-mount consists of an optical head with a special Fourier transform spectrometer (Sagnac common path interferometer) to measure the spectrum, and a cooled CCD camera (Princeton Instruments, Trenton, NJ) for imaging. The samples were illuminated with a Xenon lamp (OptiQuip 770/1600) and imaged with a 63× oil immersion objective through a custom-designed filter set (Chroma Technology, Brattleboro, VT) with broad emission bands (excitation filter: 486/28 nm, 565/16 nm, 642/22 nm; emission filter: 524/44 nm, 600/38 nm, 720/113 nm; beamsplitter: reflection 421 to 480 nm, 561 to 572 nm, 631 to 651 nm: transmission 495 to 564 nm, 580 to 620 nm, 660 to 740 nm). Excitation through this filter set allows all dyes to be excited and measured simultaneously without an image shift. The generation of a spectral image is achieved by acquiring ∼ 100 frames of the same image. Each two frames differ only in the optical path differences (OPDs) created by a scanner controller in the inter ferometer. In this way the interferogram as the modulated function of intensity (that is, the intensity as a function of OPD) is measured simultaneously for each pixel in the image. However, each pixel functions like a stand-alone Fourier transform spectrometer. Measurement times vary depending on the brightness and the size of the image, the desired spectral resolution, and the signal-to-noise ratio. A typical measurement for chromosome painting probes takes about 50 s with a 15-nm (at 600 nm) spectral resolution. The spatial resolution of the measurement is ∼0.24 μm and is limited by the CCD pixel size (15 μm) and the objective magnification (63×). After the measurement, ∼2 min are required to build the spectral image with a software-based fast Fourier transform (FFT] algorithm [E. O. Brigham, The Fast Fourier Transform and its Application (Prentice-Hall, Englewood Cliffs, NJ, 1988)]. The conversion of emission spectra to visualize the spectral image in display colors is achieved as follows: The measured spectrum at each pixel is divided into three spectral ranges (475 to 550 nm, 550 to 650 nm, and 650 to 750 nm). Each of the spectral ranges is visualized in a different color (blue, green, and red, respectively). The intensity for each color is proportional to the integrated intensity in the corresponding spectral range (Figs. 1 and 2).
8. x,y,n for all reference spectra, the smallest value is chosen and a classification color is assigned to that pixel in accordance with the classification color assigned to the most similar reference spectrum.
9. Y. Ning, Nature Genet., in press.
10. E. Schröck et al., unpublished data.
11. A. Kallioniemi et al., Proc. Natl. Acad. Sci. U.S.A. 91, 2156 (1994).
12. U. Koehler, F. Bigoni, J. Wienberg, R. Stanyon, Genomics 30, 287 (1995): P. van Tuinen and D. H. Ledbetter, Am. J. Phys. Anthropol. 61, 453 (1983).
13. U. Koehler, F. Bigoni, J. Wienberg, R. Stanyon, Genomics 30, 287 (1995): P. van Tuinen and D. H. Ledbetter, Am. J. Phys. Anthropol. 61, 453 (1983).
14. P. Meltzer, X.-Y. Guan, A. Burgess, J. Trent, Nature Genet. 1, 2 (1992).
15. E.S. received a stipend from the Deutsche Forschungsgemeinschaft. I.B. is a postdoctorate student in S. Lavi's lab. We would like to thank P. Millman (Chroma Technology) and A. Waggoner (Amersham Life Sciences) for valuable discussions, and D. Leja for help in preparing Fig. I. T. Knutsen, K. Precht, M. Macha, the cytogenetics laboratory of American Medical Laboratories, and Children's Hospital, Chantilly, VA, kindly provided metaphase chromosome preparations. The continued support of R. Buckwald, D. Cabib, N. Katzir, D. Wine, and M. Lavi (Applied Spectral Imaging, Ltd.) is gratefully acknowledged. We are indebted to J. Trent and M. Bittner for critically reading the manuscript.