Slide preparation and hybridization

One of the most important steps in successful color karyotyping is slide preparation. Metaphases need to be properly spread, with little or no cytoplasmic residua, and the slides subjected to very little aging prior to hybridization. Cell suspensions were stored at Ý20 C in standard 3:1 methanol:acetic acid fixative. Slides were prepared one hour before hybridization, using common spreading techniques. After preparation, slides were aged 10-30 seconds at 94 C in ethanol (chemical aging) and immediately subjected to 30-45 second treatment with 0.005% (w/v) pepsin, followed by several seconds rinse in PBS, pH 7.5, and an ethanol series (3 minutes each in 70%, 90%, 100%). Slides were denatured in 70% formamide/2xSSC at 75 C for 2 minutes, passed through another ethanol series and air-dried. The denatured DNA probe was placed on the slide, covered with a coverslip and hybridized. Alternatively, the probe cocktail was placed on the slide before denaturing, coverslipped, sealed with rubber cement and denaturing of the chromosomal DNA and the probe DNA was done "simultaneously", by placing the slide on a heated metal block, at 75 C for 2 minutes. Both denaturing techniques produce similar results and the entire slide preparation process requires 30-40 minutes. Hybridization was done overnight (14-20 hours). The freshness of the slides made hybridization very efficient, and longer hybridization times (2-4 days) were not necessary.

Probes, dyes and antibodies

Microdissected chromosome painting probes were kindly provided as PCR products by J. Trent, P. Meltzer and M. Bittner (National Center for Human Genome Research, Bethesda, MD). PCR re-amplification of each library was performed using a previously described degenerate primer (Telenius et al) and the following cycling conditions: 45 sec at 94 C, 45 sec at 54 C and 4 minutes at 68 C, for 30 cycles. To propagate these probe libraries, for each chromosome, 3-4ul original product was re-amplified in 100ul PCR volume. Labeling was done using the same PCR conditions and any fluorescent- or haptene-labeled dUTP. The proportion of labeled dUTP to dTTP was 1:8 for Texas Red, 1:5 for TAMRA, 1:3 for Cy3, Cy5, BIO, DIG and 1:2 for AMCA, FITC and DNP. Labeled dUTPs were either synthesized in this laboratory using amine-succinimide ester coupling reactions or were purchased from the following vendors: Boehringer Mannheim (DIG, BIO, AMCA, FITC); Amersham Life Sciences (Cy3, Cy5); Molecular Probes (TAMRA, Texas Red), NEN Life Sciences (DNP); Enzo Diagnostics (BIO). Prior to labeling, probes were mixed together in cocktails to be labeled with the same dye/nucleotide, either as indicated in Table 1 (for "3+3") or as proposed by Speicher et al. for Table 2 combinations ("3+2 and 2+3"). In every PCR cocktail, each probe template was added in an amount proportional to chromosomal size, then further adjusted by testing. After PCR labeling, DNA size was brought to below 500 bp, using a controlled DNase I digestion, 15 minutes at room temperature, followed by 2-3 minutes incubation at 95 C (stop reaction). The 10x DNase solution contains 10-20ng/ul DNaseI and 20mM MgCl2. In general, for one hybridization, the following amounts of labeled PCR products were used: 150ul for AMCA, Cy3, Cy5; 75ul for TAMRA, Texas Red, FITC, DNP; 50ul for BIO, DIG. Probes were ethanol-precipitated and resuspended in 12ul hybridization buffer. Antibodies were purchased from the following vendors: Accurate Chemical (Avidin Cy3, avidin Cy5); Boehringer Mannheim (Avidin FITC, sheep antimouse Cy3), Sigma (Mouse anti-digoxin, goat anti-rabbit FITC); Vector Laboratories (Avidin D, Avidin AMCA, horse anti-mouse, horse anti-mouse Texas Red). Cy3.5 and Cy5.5 labeled antibodies were prepared using standard dye-protein coupling protocols (Amersham, Molecular Probes)

Probe detection and image capture

After hybridization, 3x 5minute posthybridization washes were done at 42-45 C in 2xSSC/50% formamide, then in 0.2x SSC, followed by incubation with the primary antibody layer, which included mouse antidigoxin and rabbit antiDNP. For the 2+3 algorithm #4, avidin-AMCA was also added at this step. All antibodies were stored as 1mg/ml stock solution and were diluted 1:100 or 1:200 in 4xSSC for use. After 5-10 minutes incubation at 37 C, slides were rinsed 3x3 minutes in wash solution (4xSSC/0.1% Tween) and mounted with antifade medium without DAPI. Images of the fluorescent labeled probes were captured using either a digital photographic camera (Olympus DP-10) attached to an Olympus AX70 fluorescent microscope or using specialized software (Vysis or PSI, Inc) controlling cooled CCD cameras (Photometrics) on Leica Aristoplan or Olympus Provia microscopes, respectively. Images of 10-15 metaphases were stored and their position coordinates recorded. The coverslips were then removed using tweezers and the slides were rinsed in wash solution 10-15 minutes at 42 C. Secondary antibodies were added according to the detection algorithm (usually goat antirabbit FITC, sheep antimouse Cy3 or horse antimouse Texas Red, and avidin Cy3.5 or avidin Cy5). After 5-10 minutes incubation at 37 C and rinsing in wash solution, slides were stained with DAPI and mounted with antifade media. Images of the same metaphases were captured. The chip of the digital photographic camera tested (DP-10) allows imaging of fluorophores from DAPI to Texas Red, only. Unlike video CCD and cooled CCD cameras, the digital camera could not detect infrared emitting fluorophores (Cy5, Cy5.5). When using a digital camera, it is important to choose a brand allowing manual exposures, so that exposing times can be changed at will. With any digital or CCD camera, imaging sequence should proceed from the fluorophore with the longest wavelength towards the fluorophore with the shortest wavelength. Exposure to the higher energy blue light can decrease the signal from other fluorophores on the metaphase. For CCK, the three most common channels to use would be blue (DAPI and AMCA), green (FITC) and red(rhodamine, Cy3, TAMRA, Texas Red, Cy5.5). If more channels are desired, fluorescence filters can be purchased (Chroma) to visualize the fluorophore AQUA (Vysis, Inc) with emission between DAPI and FITC. Also, filters can be purchased to allow separate detection of orange-red dyes (Cy3) and red dyes (Cy3.5, Texas Red).

Image processing

Individual channel/fluorophore images captured with the digital photographic camera (DP-10) were stored in JPG format and transferred to a computer running Adobe Photoshop (versions 3.0. 4.0. or 5.0). As this camera records color images, the first step was to open every image in Photoshop and convert them to grayscale images, by choosing "mode" and "grayscale". When CCD or cooled CCD cameras were used for image capturing, the Vysis or PSI software used to control these cameras allowed raw images of every channel/fluorophore to be exported as individual grayscale image files in PICT or TIFF format. Three grayscale JPG or PICT files at one time were simultaneously opened with Photoshop, then merged into a tri-color image by clicking on the arrow located in the upper right corner of the "channels" window, and selecting "merge channels", and then "RGB image". When prompted, the colors red, green and blue can be selected for any of the three raw images opened. The software then provides a multicolor image of all three channels. Each channel can be selected individually by clicking on the channel name in the channel window, and can be processed for background subtraction, lightness and contrast, sharpening or smoothing using the commands under the top menus. Usually, two RGB files were created, one with the channels captured under the "first image" and another one with the channels captured under the "second" image (Fig. 3a-b, 3d-e, 4i-j and 4k-l). Chromosome analysis can be done at this time using the charts provided (Fig 3 and 4). To merge these two RGB files in Photoshop and obtain a true multicolor image, in which each chromosome has a different shade, the "second" RGB was further processed. Its colors were altered so as to be different from the "first" image by going to the top menu and selecting "image", "adjust" and "hue/saturation" and then moving the "hue" arrow left or right until a desired color combination was obtained. These changes were stored. The modified "second" RGB image was then copied and pasted on top of the "first" RGB image. In the layer/channel pallete, by clicking on "layers" the transparency of the "second" image was then adjusted and set to about 50%. The shift between the first and second images was apparent immediately and was corrected using the arrows on the keyboard, which move the pasted "second" image in the desired direction by one pixel for each click. As the two merged images have different color shades, each chromosome pair in the final image has a different color shade. When CCD or cooled CCD cameras and FISH/M-FISH software are available, the channels captured for the "first" and "second" CCK images can be stored all in a single file. The M-FISH software (PSI, Inc) will allow correction for the position shift and will automatically process, colorize, merge the images and create a karyotype, similar to a standard M-FISH. The version of the Vysis FISH software tested allowed for only three channels to be captured, pseudo-colored and merged at one time, yielding RGB images. The "first" and "second" colored CCK images were then merged using Photoshop.

References

1. Muller, S., O'Brien, P.C., Ferguson-Smith, M.A. & Wienberg, J. Cross-species colour segmenting: a novel tool in human karyotype analysis. Cytometry 33, 445-452 (1998). 
2. Schrock, E. et al. Multicolor spectral karyotyping of human chromosomes. Science 273, 494-497 (1996). 
3. Speicher, M.R., Gwyn Ballard, S. & Ward, D.C. Karyotyping human chromosomes by combinatorial multi-fluor FISH. Nature Genet 12, 368-375 (1996).
4. Tanke, H.J. et al. New strategy for multi-colour fluorescence in situ hybridisation: COBRA: COmbined Binary RAtio labelling. Eur J Hum Genet 7, 2-11 (1999).
5. Telenius, H. et al. Degenerate oligonucleotide-primed PCR: general amplification of target DNA by a single degenerate primer. Genomics 13, 718-725 (1992).