Patients
Ninety-two paraffin embedded tissue samples from both tumor and normal tissues of randomly selected patients with colorectal cancer were obtained for evaluation of the loss of 18q chromosome. The patients were diagnosed and operated in the past five years at the Surgery Clinic, Faculty of Medicine, Skopje and histopathological diagnosis of resected tumors was done at the Institute of Oncology and Radiotherapy, Faculty of Medicine, Skopje. The primary criteria for inclusion was the availability of sufficient amount of parafin embeded tissue for analysis and the absence of distant metastasis at diagnosis, so that the distribution of Dukes stages in this group of patients is not representative for all colorectal cancer patients from the R.Macedonia. Clinical characteristics of patients and histopathological features of their tumors such as age, gender, localization, size, histopathological grade and clinical stage of the tumors were obtained from the patients' records.

DNA extraction
Six 5 mm thick sections both from paraffin embedded tumor and normal tissues of each patient were used for DNA extraction. The sections were examined under microscope and, if neccessary, the malignant tissue was clearly marked from the adjusant normal tissue before scraping them off by scalpel in separate tubes. Following deparaffinization with xylol, tissue samples were incubated in Digestion buffer containing 50 mM TRIS (pH 8.5), 1mM EDTA (pH 8.5), 1% TWEEN 20, 1% NONIDET and 0.8 mg/ml Proteinase K, on 56˚C over night. The following day, DNA was extracted with phenol/chloroform (1 vol.) and precipitated with 2 vol.absolute ethanol. The DNA pellet was eluted in 50-100 ml sterilized water [17].

Microsatellite markers and PCR amplification
Loss of heterozygosity of 18q chromosome was assessed using five microsatellite markers [D18S58, D18S61, DCC, D18S46 (DPC4) and D18S535], given in Table 1. Two methods, the conventional non-radioactive polymearase chain reaction/polyacrylamide gel electrophoresis (PCR-PAGE) analysis and fluorescent multiplex PCR/capillary electrophoresis assay, were used for evaluation of 18q loss.
Non-radioactive PCR/PAGE method was performed with procedures routinely in use in our laboratory, beginning with PCR amplification of the five microsatellite markers. For each analysis, 300 ng of purified DNA, from both tumor and normal tissues, were amplified in separate tubes in a final volume of 25 µl in a mixture containing 10 mM TrisHCl (pH8.3), 50 mM KCl, 2 mM MgCl2, 200 mM each of the four dNTPs, 25 pM of each primer and 1.5 U of Taq polymerase. The PCR amplification consisted of 30 cycles of denaturation at 95˚C for 1 min, annealing at 60˚C for 1 min and extension at 72˚C for 1 min. Two µl of each reaction were heat denatured in 2 vol. of loading dye (95% formamide, 1 mM EDTA, 0.05% bromophenol blue and 0.05% xylene cyanol), quickly chilled on ice and loaded on 6% denaturing polyacrylamide gels (40 cm x 20 cm x 0.4 mm) containing 6 M urea and 32% formamide. The electrophoresis was run at 70 W constant power for 2 hours and the fragments were transferred to positively charged nylon membranes (Hybond+, Amersham Pharmacia Biotech LTD, Amersham Place, Buckinghamshire, UK). The membranes were hybridized with nonradioactively labeled internal oligonucleotide probes (CA)9 or (A)18 and detected by chemiluminiscence. Finally, the allelic content was evaluated by visual inspection or by densitometry, when the allelic loss was not obvious.
The results obtained with the nonradioactive PCR-PAGE method were compared with the results obtained with fluorescent multiplex PCR/capillary electrophoresis assay. In this assay, the 5 loci were amplified in a multiplex PCR reaction, one primer of each primer pair being labeled at the 5' end with either 6-FAM (D18S58 and D18S535), HEX (D18S46 and DCC) or JOE (D18S61) fluorescent dyes (MWG-Biotech AG, Ebersberg, Germany). The PCR conditions were similar as above. Fluorescently labeled products were detected by capillary electrophoresis using the ABI PRISM 310 genetic analyzer and GeneScan Software (Applied Biosystems, Foster City, CA, USA). PCR samples (1 µl) were mixed with 24 µl formamide and 0.5 µl size standard TAMRA, denatured on 95˚C for 5 min, quickly chilled on ice and analyzed with GeneScan virtual filter set A and C. Loss of heterozygosity was scored when the ratio of the tumor signal to that of the normal signal was less than 0.65 or more than 1.35.

Table 1.Sequences of the primers and the sizes of the resulting PCR fragments used for the analysis of the allelic imbalance of the long arm of chromosome 18

* used only in multiplex PCR/capillary electrophoresis assay