Interphase Fluorescence in Situ Hybridization Analysis: a Study Using Centromeric Probes 7, 8, and 12*F

Interphase Fluorescence In situ Hybridization Analysis: A Study Using Centromeric Probes 7, 8, and 12*f

LIAN ZHAO, M.D., CLSp(CG), ZEBUNNISA KHAN, M.S., CLSp(CG), KIMBERLY J. HAYES, B.S., CLSp(CG), and ARMAND B. GLASSMAN, M.D.

Division and Department of Laboratory Medicine, Section of Cytogenetics, The University of Texas M.D. Anderson Cancer Center, Houston, TX 77030

ABSTRACT

Interphase fluorescence in situ hybridization (I-FISH) is a useful technique for detecting chromosomal numerical abnormalities in tumors and is gaining acceptance as a tool in cytogenetics and clinical diagnoses. Performance and quality control information about commercial products are necessary in order to implement an individual FISH probe as a routine clinical laboratory test. Interphase FISH analysis was performed with three commercially available alpha-satellite chromosome- specific DNA centromeric probes (D7Z1/D7Z2; D8Z2; and D12Z3) on bone marrow material prepared for conventional cytogenetic analysis. The results were interpreted following enumeration of the signals in 500 interphase nuclei each by two different observers. A mean of 93.92 percent (±1.3 percent, 1 SD) was found for chromosome 7; a mean of 93.91 percent (±1.5 percent, 1 SD) was found for chromosome 8, and a mean of 92.85 percent (±1.4 percent, 1 SD) was found for chromosome 12. The results of the study demonstrated that I-FISH using chromosome centromeric probe(s) is a reliable, reproducible, and accurate technique. This technique can be integrated into routine clinical practice with proper quality control protocols.

Introduction have enabled the cytogeneticist to detect specific deoxyribonucleic acid (DNA) Fluorescence in situ hybridization (FISH) is sequences and to visualize their distribution in used increasingly in clinical practice.1 interphase nuclei. The technique has been Improved sensitivity, speed of detection, and successful in our research laboratory.2,3 To other inherent advantages of this technique apply this technique in the clinical laboratory, information about product performance needed to be generated. * This work is supported in part by the Olla S. Stribling Chair for Cancer Research. The purpose of this study was to establish t Address reprint requests to: Armand B. Glassman, reference ranges for FISH. Three chromo M.D., Division and Department of Laboratory Medicine, some-specific centromeric probes (alpha- Section of Cytogenetics, Box 73, The University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Blvd., satellite DNA 7, 8, 12) were chosen. These Houston, TX 77030. three chromosomes are commonly involved in

51 0091-7370/98/0100-0051 $01.50 © Institute for Clinical Science, Inc. 52 ZHAO, KHAN, HAYES, AND GLASSMAN human leukemias and lymphomas. Our data min, immediately cooled on ice, and then demonstrated that: (1) interphase FISH, using placed onto slides. The slides were then cov chromosome centromeric probe(s), is a reli ered with a 22 x 22 mm coverslip, sealed with able, reproducible, and accurate technique, rubber cement, and incubated overnight at and (2) this technique can be integrated into 37°C in a humidified chamber. clinical cytogenetics practice.

P r o b e D e t e c t io n a n d M ic r o s c o p y Material and Methods After incubation, the slides were treated in C o n t r o l S a m p l e P r e p a r t io n s 0.25x SSC for 5 min at 70°C. The signal of the cells was detected by anti-digoxigenin conju Bone marrow cells from 10 untreated leu gated with fluorescein isothiocyanate (FITC). kemia subjects with conventional cytogeneti Slides were then washed in phosphate buffer cally normal chromosomes 7, 8, 12 were used. with detergent [PBD] three times, for 2 min Chromosome preparations were made each. Finally, the slides were counterstained according to standard cytogenetic proce with propidium iodide/antifade solution. The dures.4 Briefly, cells were cultured overnight slides were examined under a Zeiss fluorescent in Roswell Park Memorial Institute (RPMI) microscope. A dual wave length filter (Zeiss) 1640 medium supplemented with 20 percent was used to view both FITC and propidium fetal bovine serum, and then harvested using iodide simultaneously. colcemid by routine cytogenetic methods. TABLE I DNA P r o b e s Criteria For Analyzing by Fluorescence Three centromeric alpha satellite probes In Situ Hydridization Results (D7Z1/D7Z2, D8Z2, and D12Z3)* were used 0 Signal No signal can be seen in a cell. to determine the number of chromosome cen Do Not include cells in S-phase, tromeres in both interphase nuclei and meta which has a diffuse FITC phase chromosomes. The probes were digoxi- background. genin labeled and commercially available. 1 Signals After focusing, just one clear signal can be seen in a cell. I n s it u H ybridization 2 Signals Two dear signals can be seen in a The hybridization protocol, as recom cell. mended by Oncor, Inc. with modification, was 3 Signals Three clear separated signals are applied. Slides were treated with 2x standard seen in a cell. Do Not confuse saline citrate [SSC] for 30 min at 37°C, dehy with sister chromatid split. drated in a 70 percent/85 percent/100 percent ethanol series for 2 min each, denatured (70 4 Signals Four or more separate, clear percent formamide, 2x SSC fpH 7.0] at 70°C or >4 signals in a cell. for 2 min, dehydrated again at room tempera Not There are signals in a cell, but the ture, and air-dried. The probes were mixed analyzable signal is hardly distinguishable. with Hybrisol VI and denatured at 72°C for 5 (N/A)

* Oncor, Gaithersburg, MD. FITC = fluorescein isothiocyanate. INTERPHASE FLUORESCENCE IN SITU HYBRIDIZATION ANALYSIS 5 3

E v a l u a t io n o f FISH R e s u l t s cells with signals are shown in figure 1. All slides were coded so that the observers had no The FISH results were evaluated by two knowledge of the probes that had been used trained observers (L.Z. & Z.K.) using the cri- on each slide. In each sample, a total of 500 teria in table I. Five categories of interphase nuclei were evaluated by each observer in dif-

—

_> 5 4 ZHAO, KHAN, HAYES, AND GLASSMAN 100% 90% 80% c=CO

20%

10%

0 % -K E m SbZÜ Z I___ iÜâSÉUi. 0 signai 1 signal 2 signals 3 signals ¿4 signals N/A FISH signals Chromosome 7 Chromosome 8 Chromosome 12

F igure 2. Fluorescence in sta data from 10 control samples (mean and standard deviation). ferent areas of the slides. In addition, certain Discussion numbers of metaphase were identified with FISH signals on each slide. In this study, commercially available digoxigenin-labeled centromeric DNA probes spe cific for chromosome 7, 8, and 12 were used Results for routine cytogenetic evaluation. The probes* provided a rapid and reliable method Ten non-leukemic diploid specimens based for detecting and quantifying chromosomes 7, on conventional cytogenetics were randomly 8, and 12 in interphase nuclei. The results selected for each group. Three probes had were interpreted following enumeration of the been used on each sample. The same hybrid signals in 500 interphase nuclei in different ization condition was used on a total of 30 areas of the slides by two trained observers. Six slides. The results were interpreted following categories for the control groups were estab enumeration of the signals in 500 interphase lished: no signal (0); 1 signal; 2 signals; 3 sig nuclei. From each group of 10 samples, a nals; 4 or more signals, and N/A cells (not ana- mean of 93.92 percent (±1.3 percent, 1 SD) lyzable). A cut-off value for each category was found for chromosome 7; a mean of 93.91 along with the two expected signals was deter percent (±1.5 percent, 1 SD) was determined mined. Similar, or even considerably higher, for chromosome 8, and a mean of 92.85 per percentages of false monosomic or trisomic cent (±1.4 percent, 1 SD) was observed for nuclei, have been seen by other investigators chromosome 12 (figure 2). On each slide, the for a variety of alpha-satellite probes.5,6> signal on metaphases was counted to confirm Loss of target DNA, poor penetration of the probe locus. Two signals on all intact meta probe, incomplete or inefficient hybridization, phases were found (table II). The probability distribution from each observer is summarized in table III. * Oncor, Gaithersburg, MD. TABLE II Fluorescence In Situ Hybridization Signal Distribution in Observed Metaphases Slide Number Chromosome 7 Comment Chromosome 8 Comment Chromosome 12 Comment NEPAE FLUORESCENCE INTERPHASE Control 1 19M:2S The metaphase with one 18M:2S The one and three signal, 20M:2S 1M:1S signal does not contain 1M:1S metaphases are poor 46 chromosomes 3M:1S quality Control 2 20M:2S 20M:2S 20M:2S Control 3 10M:2S 12M:2S 20M:2S Control 4 20M.2S 10M:2S 8M:2S N SITUIN Control 5 18M:2S 19M:2S 1M :3S The morphology of the 19M:2S metaphase with the ANALYSIS HYBRIDIZATION three signals is poor Control 6 1M:4S The metaphase with four 19M:2S The metaphase with four 18M:2S 19M:2S signals is tetraploid 1M:4S signals is tetraploid Control 7 20M:2S 17M:2S The metaphase with four No metaphases 1M:4S signals is tetraploid found Control 8 5M:2S 3M:2S 20M:2S Control 9 20M:2S 20M.2S 20M:2S Control 10 20M:2S 4M:2S 8M:2S M - metaphase. S « signal. 5 6 ZHAO, KHAN, HAYES, AND GLASSMAN TABLE III Mean and One Standard Deviation (SD) of Control Groups for Alpha-satellite Probes (7, 8,12) from 10 Samples FITC-Signal in a-satellite 7 a-satellite 8 a-satellite 12 10 Slides M eant 1SD Mean ± SD Mean ± SD (500 cells each) in % in % in % 0 signal 1.86 ±1.3 1.55 ±0.9 1.59± 1.0 1 signal 3.35 ± 1.1 2.89 ±1.1 4.30 ± 0.8 2 signals 93.92 ±1.3 93.91 ±1.5 92.85 ± 1.4 3 signals 0.74 ±0.3 0.98 ± 0.7 0.86 ± 0.4 4 or more signals 0.16 ±0.2 0.64 ± 0.4 0.34 ±0.2 N/A (not analyzable) 1.78 ±1.2 2.09 ±1.5 1.09 ±0.5 FITC = fluorescein isothiocyanate. or signal overlap are all possible causes leading References to a weak or absent signal. Some cells were difficult to score because of signal problems. 1. Le Beau M. Detecting genetic changes in human tumor cells: have scientists “gone fishing?” Blood These cells were classified as not analyzable 1993;8:1979-83. (N/A) cells. The percentage of these N/A cells 2. Zhao L, Van Oort J, Cork A, Liang JC. Comparison may have been a reflection of the counting skill between interphase and metaphase cytogenetics in of the laboratory person, the binding efficiency detecting chromosome 7 defects in hematological of the probe, or otherwise technically related. neoplasias. Am J Hematol 1993;43;205-11. The efficiency of each probe is indicated in 3. Zhao L, Chang KS, Estey EH, Hayes K, Deisseroth AB, Liang JC. Detection o f residual leukemic cells in part by the number of cells having no signal. patients with acute promyelocytic leukemia by the The N/A cells are not included in the total fluorescence in situ hybridization method: potential percentage for each category. Once the labo for predicting relapse. Blood 1995;85:495-9. ratory normal control range is established, the 4. Trujillo JM, Cork A, Ahearn MJ, Youness EL, McCredie KB. Hematologic and cytologic character percent N/A cells can be used to compare with ization of 8/21 translocation acute granulocytic leuke each specimen. Only the cells that contain mia. Blood 1979;56:695-706. clear signals should be counted. Fluorescence 5. Anastasi J, Le Beau M, Vardiman JW, Weatbrook CA. in situ hybridization, using alpha-satellite Detection of numerical chromosomal abnormalities in DNA centromeric probes 7, 8, and 12, is neoplastic hematopoietic cells by in situ hybridization with a chromosome-specific probe. Am J Pathol 1990; highly reproducible and capable of providing 136:131-9. reliable quantitative information in the clinical 6. Perez Losada A, Wessman M, Tiainen M, Hopman A, cytogenetic laboratory. Willard HF, Sole F, Caballin M, Woessner S, Knuu- tila S. Trisomy 12 in chronic lymphocytic leukemia: an inteiphase cytogenetic study. Blood 1991;78:775-9. Acknowledgment 7. Poddighe DJ, Mosesker O, Smoots D, Awwad BH, This work was supported in part by the Olla S. Stribling Ramaekers F, Hopman A. Interphase cytogenetics of chair for Cancer Research. The authors thank Leslie hematological cancer: comparison o f classical karyo Wildrick for her editorial assistance. The secretarial assis typing and in situ hybridization using a panel of eleven tance of Isabel Hemandez, Pinkie D. Oji, and Bemadine chromosome specific DNA probes. Cancer Res 1991; Williams is greatly appreciated. 51:1959-67.