ELSEVIER

The Characterization of Chromosomal Rearrangements by a Combined Micro-FISH Approach in a Patient with Myelodysplastic Syndrome

Jia Xu, Edward Cedrone, Michele Roberts, Guan Wu, Sten Gershagen and Nancy Wang

ABSTRACT: We report the application of a combined strategy: chromosome microdissection, degenerate oligonucleotide primed-PCR, and reverse chromosome painting (micro-FISH), as well as forward chro- mosome painting, for the characterization of chromosomal rearrangements in a MDS patient with the karyotype 46,XX,-11, +r analyzed by G-banding. The karyotype was refined as 46,XX, der(2)t(2;11) (q35;?p13),der(11)dic r(11)(:p13-~q13::p13~q13:). Our study demonstrated that the chromosome composi- tion of a neoplasia can be identified more systematically and accurately using these combined molecular cytogenetic approaches. The DOP-PCR methodology modified is suitable for the practical application of micro-FISH on specimens prepared for routine banding analysis.

INTRODUCTION

The analysis of chromosome abnormalities in malignant cells has become an integral part of the diagnostic and prognos- tic work-up of many human cancers [1, 2]. Their molecular analysis has facilitated the identification of genes related to the pathogenesis of malignancy [3, 4]. However, complex marker chromosomes frequently remain undiagnosed after conventional cytogenetic analysis. A recently developed tech- nique has proven extremely useful in solving this problem. This technique involves chromosome microdissection and subsequent PCR amplification of the dissected chromosomal DNA, followed by labelling for fluorescent in situ hybrid- ization (FISH) to normal metaphase chromosomes (micro- FISH) [5, 6]. The micro-FISH approach has also been termed reverse chromosome painting, as opposed to forward chro- mosome painting, in which one uses normal chromosomes as probes onto metaphases containing the chromosome of interest [7]. Even without any clue, reverse chromosome painting can be used to actively identify not only the origin of the marker chromosomes, but also the regions and break- points involved. However, when there is a complex karyo-

From the Departments of Pediatrics (J. X., E. C., N. W.), Pathol- ogy (M. R.), and Urology (G. W., S. G.), University of Rochester School of Medicine, Rochester, New York.

This study was supported in part by NIH grant CA52761. Address reprint requests to: Dr. Nancy Wang, Dept. of Pediatrics,

University of Rochester School of Medicine and Dentistry, Box 77Z 601 Elmwood Ave., Rochester, NY 14642.

Received September 16, 1994; accepted February 3, 1995.

Cancer Genet Cytogenet 83 :105-110 (1995) © Elsevier Science Inc., 1995 655 Avenue of the Americas , New York, NY 10010

type, the combination of forward (conventional) and reverse chromosome painting is needed for providing a more effi- cient and definitive analysis of the chromosomal rearrange- ments.

In the present study, we report the application of a com- bined strategy involving chromosome microdissection, DOP- PCR [8], and reverse and forward chromosome painting for precise characterization of chromosomal rearrangements in a myelodysplastic syndrome case.

MATERIALS AND METHODS

Clinical History

A bone marrow aspirate was obtained from a 65-year-old white woman with a 27-month history of clinically progress- ing myelodysplastic syndrome (MDS), with a FAB classifi- cation of refractory anemia (RA). The patient had two previ- ous bone marrow aspirates with a normal karyotype, 24 and 5 months prior to the current specimen. Recent worsening pancytopenia and increasing transfusion requirements prompted chromosome analysis to identify possible karyo- typic aberrations associated with clinical progression of MDS. After the detection by G-banding of a ring chromosome of unknown origin, a detailed molecular-cytogenetic analy- sis was performed.

Chromosome Microdissection and Amplification of Dissected DNA The procedure for chromosome microdissection and PCR was performed by following the method of Guan et al. [9], com- bined with a heat soak [10] at the denaturization step. Briefly,

0165-4608/95/$9.50 SSDI 0165-4608(95)00046-R

106 J. Xu et al.

cells were harvested from bone marrow using conventional cytogenetic techniques. Prior to microdissection, the cells had been fixed in 3:1 (methanol:acetic acid) fixative for 2 days. Metaphase spreads were prepared on 24 × 60 mm no. 1 1/2 coverslips and were stained by trypsin Giemsa band- ing. Microdissection was performed with glass micronee- dies controlled by Narashige micromanipulators (models MM-88 and MO-302) attached to an inverted microscope. The dissected ring chromosome adherent to the microneedle was transferred to a 0.2-mL microcentrifuge tube containing 5 ~L of collection buffer (40 mM Tris HC1, pH 7.5, 20 mM MgCI 2, 50 mM NaC1, 200 ~tM of each dNTP, 1 uni t Topo 1 [Promega], and 5 pmol of a universal primer CCGACTC- GAGNNNNNNATGTGG). A gene Amp PCR System 9600 (Perkin-Elmer Cetus) was used for DNA amplification.

After 5 copies of dissected ring chromosomes were col- lected, the collection buffer was covered with 45 ~tL of mineral oil and incubated at 37°C for 30 minutes, followed by heat soak at 94°C for 30 minutes. An initial 8 cycles of PCR (denaturization at 94°C for 1 minute, anneal ing at 30°C for 2 minutes, and extension at 37°C for 2 minutes) were per- formed by adding approximately 0.3 units of T7 DNA poly- merase (Sequenase version 2.0, USB) at each cycle.

Subsequently, 50 ~tL of PCR reaction mixture (10 mM Tris, pH 8.4, 2 mM MgC12, 50 mM KC1, 0.1 mg/mL gelatin, 200 ~M each of dNTP, 50 pmole universal primer, and 2 units

Taq DNA polymerase) were added to the same tube. The reac- tion was heated to 95°C for 3 minutes, followed by 35 cycles at 94°C for I minute, 56°C for I minute, 72°C for 2 minutes, with a final extension at 72°C for 5 minutes.

Reverse Chromosome Painting A 2-~tl aliquot of the PCR product was labeled for 16 cycles in a 50-~tL secondary PCR exactly as described above, ex- cept for the addit ion of 20 ~M biotin-11-dUTP (BMB). La- beled PCR products were purified using a Centricon 100 (Amicon) filter and 100 ~tg of this probe was added to 10 ~L hybridization mixture containing 55% formamide, 10% dex- tran, 1 x SSC, and 5-10 ~tg COT-1 DNA (BRL). The probe was painted back to metaphase spreads from the same pa- tient, and to spreads from normal individuals. FISH was per- formed using our procedure described previously [11]. G-banding was performed prior to FISH.

Forward Chromosome Painting Digoxigenin-labeled cL satellite probe (D11Z1, p5030-DG.5) for the centromere of chromosome 11 and digoxigenin-labeled total chromosome probe (p5202-DIG) for chromosome 11 were purchased from Oncor (Gaithersburg, MD). The FISH pro- cedure employed was essentially the same as that described by the manufacturer.

Metaphase cells were viewed with a Zeiss Axioplan

P

1 3 2 der(2)t(2;l l )

6 7 8 9 10

4 $

i?

11 12

i i ~= ! ~C~i~ i ~̧ :I~̧¸̧

1 3 14 15

19 20 21 22

16

X

17 1 8

P

der( l l )d ic r ( l l )

Figure 1 A bone marrow karyotype from a patient showing 46,XX,der(2)t(2;11)(q35;?p13),der(11)dic r(11)(:p13 ~ q13: :p13~q13:). Rearranged chromosomes are indicated by arrowheads. Note that der(2)t(2;11) and der(llldic r(11) could not be identified by G-banding alone.

Micro-FISH in MDS 107

microscope. Photomicrographs were taken on Kodak Ekta- chrome 400 color sl ide film.

RESULTS

Sixteen metaphase chromosome spreads were analyzed by G-banding. The absence of a normal chromosome 11 and the presence of a ring chromosome of unknown origin were noted in every metaphase analyzed. On the basis of G-banding, the karyotype was designated 46,XX,- 11, + r (Fig. 1).

The micro-FISH probe, generated by microdissect ion of the ring chromosome, was app l ied to previously G-banded metaphase spreads from a cytogenetically normal individual (reverse painting). The centromeric region of both copies of the normal chromosome 11, from p13 to q13, was painted with this probe. Subsequently, the micro-FISH probe was appl ied to metaphase spreads from the patient. The entire ring chro- mosome was painted, as was the centromeric region of the normal chromosome 11, again from p13 to q13 (Fig. 2). Thus, the karyotype designation was refined to 46,XX,r(11)(p13q13).

However, the ring is larger than a single copy of the painted segment of chromosome 11, so FISH was performed with a centromeric probe for chromosome 11 onto metaphase spreads from the patient. This probe revealed that the ring

chromosome contains two centromeres of chromosome 11 (Fig. 3). Therefore, the karyotype was refined further, to 46,XX,der(11)dic r(11)(:p13-*q13:: p13-*q13 :).

To verify the possible involvement of chromosome 11 in other structural aberrations, FISH with a commercial ly avail- able whole paint ing probe of chromosome 11 was app l ied to metaphase chromosome spreads from the patient. It painted the entire r ing chromosome, one normal chromo- some 11, and a short segment t ranslocated to the distal end of the long arm of a chromosome 2 (Fig. 4). Combined with G-banding results, the karyotype des ignat ion was refined even further, as 46,XX,der(2)t(2;11)(q35;?p13),der(11)dic r(11) (:p13--*q13::p13-*q13:). The patient 's spec imen is, therefore, t r isomic for the centromeric region of chromosome 11 (p13- q13), and monosomic for 11q13 to 11qter and 2q35 to 2qter.

DISCUSSION

In neoplasia , there frequently are complex chromosomal structural aberrations which are unident i f iable by conven- t ional chromosomal banding analysis. This hinders the pre- cise ident i f icat ion of specific chromosomes, regions, and genes that are crucia l ly related to the pathogenesis of a neo- plasia. Recently, the availabil i ty of FISH techniques has sig-

J

I .......... ~ . . . . . . . . . . . . . . . . B

O

Figure 2 Reverse chromosome painting analysis. A) Micro-FISH probe from the patient hybridized onto a normal G-banded metaphase spread, as shown in (B). The chromosome 11 material present in the ring chromosome is visual- ized on the two normal chromosome 11, from p13 to q13. C) Micro-FISH probe hybridized back onto a metaphase spread of the patient, as shown in (D). The entire ring chromosome, as well as the normal chromosome 11 (p13 to q13), is painted, which confirms that the origin of the micro-FISH probe is from the ring. Arrow and arrowhead point to normal chromosome 11 and ring chromosome, respectively.

108 J. Xu et al.

4

J B Figure 3 A) FISH with the probe specific for the centromeric re- gion of chromosome 11 onto a G-banded metaphase spread of the patient, as shown in (B). Note two signals on the ring chromosome, indicating the dicentric nature of the chromosome. Arrow and ar- rowhead indicate normal chromosome 11 and ring chromosome, respectively.

nificantly improved the accuracy in cancer cytogenetic anal- ysis, especially when it is applied repetitively and combined with G-banding analysis [11]. However, the efficacy of FISH analysis depends on the probes chosen, which, in turn, de- pends on the information obtained from the G-banding anal- ysis. When there is no clue available from the banding anal- ysis, it is very difficult to choose the adequate probe(s) for FISH analysis. Therefore, the nature of FISH application is more a verification and confirmation than an active iden- tification.

In contrast, the newly developed approach of micro-FISH provides a straightforward method to identify any chro- mosomal segment of unknown origin by dissecting the seg- ment of interest, having the DNA isolated, amplified, fluores- cence labelled, and in situ hybridized to normal metaphase spreads [6-12]. In the present study we applied this molec- ular approach, combined with G-banding, in the identifica- tion of a ring chromosome which turns out to be a der(11) dic r(11)(:p13-*q13::p13-*q13:). A ring chromosome is formed by the reunion of the two broken ends of a chromosome. To determine whether the broken-off segment of chromosome 11 is deleted or involved in translocation, FISH with whole chromosome 11 probe was applied and demonstrated that the segment of 11p13 to 11pter was actually translocated to

chromosome 2. The karyotype of this MDS case can, there- fore, be defined as 46,XX,der(2)t(2;11)(q35:?p13),der(11)dic r(11) (: p13-*q13: :p13 -*q13 :).

Structural rearrangements of chromosome 11, especially deletions of 11q, are seen frequently in MDS [19, 20]. In the present study, using G-banding combined with chromosome microdissection, and forward and reverse chromosome paint- ing, we identified the involvement of chromosome 11 in both a translocation and ring formation. A ring chromosome is a rare cytogenetic abnormality found in patients with MDS. A literature search revealed 27 cases of MDS with a ring chro- mosome. In 24 of the 27 reported cases (89%), ~additional complex cytogenetic abnormalities were observed. It has been reported that ring chromosomes, as well as complex cyto- genetic abnormalities, are associated with a poor prognosis in MDS [15-18]. In the present case, the emergence of a ring chromosome 11 and a t(2;11) translocation 27 months after initial MDS diagnosis may, therefore, portend a poor prog- nosis. The worsening pancytopenia and increasing transfu- sion requirements correlated well with the prognosis of the cytogenetic analysis.

The identification of the ring chromosome and the t(2 ;11) can not be accomplished without the application of multi- ple molecular approaches. Our study demonstrated that micro-FISH is the most straightforward and powerful mo- lecular cytogenetic approach in determining the origin of a marker chromosome, especially when there is no clue provided from the banding analysis. However, to identify the possible structural aberrations that show that a specific chro- mosome may be involved, a forward chromosome painting combined with banding analysis is the most efficient ap- proach. By using both micro-FISH and forward chromosome painting, the chromosome composit ion of a neoplasia can be systematically and accurately identified.

It has been reported by several investigators that chromo- somes should be exposed to fixative for only a couple of minutes before microdissection, to avoid damage to DNA by acetic acid [13, 14]. This significantly hinders the practical application of micro-FISH in clinical cases, as acetic acid is the main ingredient in the fixative used for routine band- ing analysis. By the time microdissection is applied after rou- tine G-banding analysis, the chromosomes have already been in the fixture containing acetic acid for days.

In our present study, we were able to generate a micro- FISH probe successfully from the bone marrow cells that had been kept in fixative for 2 days. This may be due to the in- creased efficiency of PCR painting resulting from pretreat- ment of the dissected chromosome with Tope I [9], followed by a heat soak [10]. This technical modification should ex- tend the application of micro-FISH in clinical oncology and, in general, facilitate the accuracy in chromosome analysis of neoplasia.

REFERENCES

1. Helm S, Mitelman, F (1987): Cancer Cytogenetics. New York, Alan R. Liss, Inc.

2. Sandberg AA (1990): The Chromosome in Human Cancer and Leukemia, 2nd Ed. Amsterdam, Elsevier Science Publishing Co., New York, pp. 50-61.

Micro-FISH in MDS 109

B

D F i g u r e 4 Forward chromosome painting analysis with painting probe for chromosome 11 onto metaphase spreads derived from the patient's bone marrow (A and C) showing hybridization results. Note three distinct fluorescence signals on the ring chromosome, chromosome 11, and the q terminal of der(2)t(2;11) (B and D) showing G-banded metaphases identical to (A) and (C), respectively. Note that one normal chromosome 2 is missing from the metaphase showing in (A) and (B), and the ring chromosome showing in (C) and (D) is tetracentric. Arrow and small arrow indi- cate normal chromosome 11 and chromosome 2; arrowhead and small arrowhead point to r (11) and der (2), respectively.

3. Nowell PC (1992): Cancer, Chromosomes, and Genes. Lab In- vest 66:407-417.

4. Rowley JD (1990): Molecular cytogenetics: Rosetta stone for un-. derstanding cancer. Twenty-ninth G.H.A. Clowes Memorial Award Lecture. Cancer Res 50:3816-3825.

5. Lfidecke HJ, Senger G, Claussen U, Horsthemke B (1989): Clon- ing of defined regions of the human genome by microdissecion of banded chromosomes and enzymatic amplification. Nature (Lond) 338:348-350.

6. Meltzer PS, Guan X-Y, Burgess A, Trent JM (1992): Rapid gener- ation of region specific probes by chromosome microdissection and their application. Nature Genet 1:24-28.

7. Carter NP, Ferguson-Smith MA, Perryman MT, Telenius H, Pel- meat AH, Leversha MA, Glancy MT, Wood SL, Cook K, Dyson HM,.Ferguson-Smith ME, Willat LR (1992): Reverse chromosome painting: a method for the rapid analysis of aberrant chromo- somes in clinical cytogenetics. J Med Genet 29:299-307.

8. Telenius H, Carter NP, Bebb CE, Nordenskjold M, Ponder BAJ, Tunnacliffe A (1992): Degenerate oligonucleotide-primed PCR: general amplification of target DNA by a single degenerate primer. Genomics 13:718-725.

9. Guan X-Y, Trent JM, Meltzer PS (1993): Generation of band- specific painting probes from a single micmdissected chromo- some. Hum Mol Genet 2:1117-1121.

I0. Ruano G, Pagliaro EM, Schwartz TP, Lamy K, Messina D, Gaenss- len RE, Lee HC (1992): Heat soaked PCR: an efficient method for DNA amplification with applications to forensic analysis. Bio Tech 13:266-274.

II. Xu J, Wang N (1994): Identification of chromosomal structural alterations in human ovarian carcinoma cells using combined GTG-banding and repetitive fluorescence in situ hybridization (FISH). Cancer Genet Cytogenet 74:1-7.

12. Zhang M, Meltzer P, Jenkins R, Guan X-Y, Trent JM (1993): Ap- plication of chromosome micmdissection probes for elucida- tion of BCR-ABL fusion and variant Philadelphia chromosome translocations in chronic myelogenous leukemia. Blood 81: 3365-3371.

13. Kao F-T (1993): Microdissection and microcloning of human chromosome regions in genome and genetic disease analysis. Bio Essays 15:141-146.

14. Hagag NG, Viola MV (1993): Preparation of chromosomes for microdissection. In: Chromosome Microdissection and Clon-

1 1 0 J. Xu et al.

ing. A Practical Guide. NG Hagag, MV Viola (eds), Academic Press, New York, pp. 25-39.

15. Sadamori N, Han T, Kakati S, Sandberg AA (1983): Chromo- somes and causation of human cancer and leukemia. LI. A hairy cell leukemia case with 14q. and ring chromosomes: sig- nificance of ring chromosomes in blood disorders. Cancer Ge- net Cytogenet 10:67-77.

16. Yoshida A, Ueda T, Imamura S, Uchida M, Kamada N, Nakamura T (1993): Two types of ring chromosomes found in a patient with myelodysplastic syndrome. Leukemia 7:1921-1923.

17. Yunis JJ, Rydell RE, Oken MM, Arnesen MA, Mayer MG, Labell M (1986): Refined chromosome analysis as an independent prog-

nostlc indicator in de novo myelodysplastic syndromes. Blood 67:1721-1730.

18. Jacobs RH, Cornbleet MA, Yardiman JW, Larson RA, LeBeau MM, Rowley JD (1986): Prognostic implications of morphology and karyotype in primary myelodysplastic syndromes. Blood 67:1765 -1772.

19. Sandberg AA (1993): Chromosome changes in leukemia and can- cer and their molecular limning. In: The Causes and Conse- quences of Chromosomal aberrations. IR Kirsch (ed), CRC Press, Inc., Boca Raton, pp. 141-163.

20. Potter AM, Watmore A (1992): Cytogenetics in myeloid leuke- mia. In: Human Cytogenetics. A Practical Approach. Vol. II, 2nd Ed. Malignancy and Abnormalities. DE Rooney, BH Czepulkowski (eds). IRL Press, Oxford, pp. 27-66.