Section 5 -

Molecular Analysis with Tissue Microarrays and Loss of Heterozygosity (LOH) Xavier Matias-Guiu and Judit Pallares Hospital Universitari Arnau de Vilanova de Lleida University of Lleida

Uses of Tissue Microarrays in Molecular Pathology Tissue Microarrays (TMAs) are powerful tools to study and validate large amounts of data generated by genetic approaches. TMAs allow the analysis of sets of proteins in hundreds of tumor samples, simultaneously. Multi-tissue blocks were first introduced by Batiffora et al. [1] in 1990. Then, in 1998 Kononen et al. [2], introduced a new method of combining multiple tissues in a single paraffin block. They used a novel sampling approach "the tissue arrayer". Tissue Microarrays can be used to assess the relevance of a given gene in a disease, and to analyse potential targets for prognosis and treatment in different types of cancer. TMAs can be used for the analysis of DNA by fluorescence in situ hybridization (FISH), RNA by mRNA in situ hybridisation and protein by immunohistochemistry [3]. Tissue Microarrays are composed of many cylindrical tissue cores embedded together in a paraffin block. The tissue cores are obtained from "donor blocks", from surgical pathology or autopsy files, or research material. After selection of the appropriate region in the donor block, from the corresponding hematoxylin-eosin stained slide, a tissue core is obtained by a thin-walled needle, and then inserted into the "recipient block" with defined array coordinates. Tissues are inserted at high density, with up to 1000 tissue cores in a single paraffin block. The selection of cylinders with 0.6 mm diameter allows preservation of histological specimens with minimal damage to the original block. At least 200 consecutive sections of 4 mm thickness can be cut from each tumor array block. TMAs offer tremendous advantages in the analysis of molecular markers [4]. The evaluation of a high number of tumors in a single experiment improves the accuracy and ability to standardize the technical procedure. It provides uniform reaction conditions and helps to economize reagents. TMAs may contain positive and negative controls and are ideal tools to test and optimise antibodies. There are numerous possible applications of TMAs in cancer research. These include validation of DNA microarray data, molecular profiling of large series of tumors with hundreds of biomarkers, rapid translation of data from cell lines and animal models to human cancer, evaluation of clinico pathological data and of newly discovered genes and molecules that can have an impact on prognosis and treatment of cancer. Usefulness of TMA in the study of signalling pathways in endometrial carcinoma will be shown in this lecture to illustrate potential indications of this technique. The main disadvantage of TMAs is the heterogeneity of tissues. To address this issue a large number of studies have been published [5, 6]. Most of them indicate that TMAs are usually valid for clinico-pathological studies, and that there is an excellent agreement between the use of TMAs and a standard tissue section, because they work with a high density of samples at the same time. The question of how many tissue cores are required to represent the expression of a given antigen in a specific tumor type, depends on the variability of the parameter being analysed. In the case of heterogeneously distributed markers, a validation study should be done to assess the agreement between the TMA and standard tissue sections. Molecular Analysis with Loss of Heterozygosity (LOH) Inactivation of a tumor suppressor gene typically occurs in two steps, thus fulfilling Knudson's hypothesis. The first hit is frequently a point mutation or a small deletion. The second alteration is usually a large genomic loss of part of a gene, or even part of a chromosome, or the whole chromosome. LOH analysis allows the identification of the second hit. LOH analysis requires identification of loss of polymorphic markers that flank a tumor suppressor gene. LOH is frequently identified by microsatellite PCR. Microsatellites are short tandem repeats made of di-or trinucleotides that are distributed throughout the genome. Tri- or tetranucleotide tandem repeats are often highly polymorphic. Since they are usually small in size, and they are flanked by DNA with specific sequences, they can be amplified by PCR. It is important to select markers with a high rate of heterozygosity in the population to minimize the number of uninformative cases. The informative status of a sample with respect to individual microsatellite loci should be screened. If a patient is homozygous for a given microsatellite locus, then the microsatellite alleles are the same size following PCR, generating a single peak,and a loss of one of the two alleles cannot be detected, and the patient is designated as non-informative for that marker. Because of the risk of uninformative markers, a group of microsatellites [5, 6] should be selected. Interpretation of LOH should be perfomed by following objective criteria. A double peak observed for the microsatellite marker amplified from normal DNA indicates a heterozygote. A single peak in tumor tissue DNA compared with normal tissue DNA indicates loss of one allele. The ratio of DNA peaks obtained for pairs of alleles should be were compared between normal and tumor tissue. A ratio greater than 1.5 is considered LOH of the shorter allele, and a ratio less than 0.5 considered LOH of the longer allele. Ratios from 0.5 - 1.5 should be interpreted as no LOH .It is important to know that a correct interpretation of LOH in a sample usually requires absence of contamination by normal tissue, since the presence of cells without LOH (normal cells) may alter the allelic peak heights ratios. For that reason, tissue microdissection is recommended. Although microsatellite PCR is the technique that is most frequently used to assess LOH, other different approaches can also used. For example, RFLP analysis can be used to identify DNA polymorphisms than create or abolish restriction sites for endonucleases. Oligonucleotide microarrays are also capable to simultaneously determine the genotype of thousands of Single-nucleotide polymorphism (SNP arrays), with specific software that analyzes the scanned image data, and generate SNP calls of paired normal and tumor samples. Comparative genomic hybridization (CGH) is also a powerful molecular cytogenetic method for the screening of a tumor genome for chromosomal imbalances, but the identification of small interstitial deletions can be difficult. Moreover, detection of mid-sized deletions can be achieved by simultaneous screening of several target sequences by multiplex amplification and probe hybridization (MAPH) and multiple ligation-dependent probe amplification (MLPA). Both techniques rely on comparative quantitation of specifically bound probes that are amplified by PCR with universal primers. The applications of LOH analysis in diagnostic molecular pathology are limited. Several of them will be illustrated in this lecture. In the experience of this author LOH can be used in patients with multiple tumors. A concordant pattern of LOH between two tumors found in the same patient should be taken as an indicator that one tumor is metastatic from the other. In contrast, discrepant LOH pattern may suggest that both tumors are independent [7, 8]. However, it is worth mentioning that occasionally, a metastasis can have a different LOH pattern in comparison to the primary tumor. This is because inactivation of tumor suppressor genes involved in tumor progression may occur during the process of invasion and metastasis; and these genes may be abnormal in the metastatic tumor, but normal in the primary tumor. LOH can also been used in familial cancer syndromes secondary to tumor suppression genes. An example of this situation is the breast and ovarian cancer syndrome. The disease can be caused by two genes BRCA-1 and BRCA-2 in chromosomes 17 and 13. The identification of a similar pattern of LOH in different tumors from several patients of a particular family may help in the identification of the gene responsible for the disease in a specific family [9]. LOH on different chromosomes has been proposed for diagnosis and prognosis in different tumors. In this lecture, the example of LOH in endometrial carcinoma will be illustrated. However, the usefulness of LOH analysis in different settings, such as chromosome 1 in neuroblastoma, or chromosome 1p and 19q in gliomas will be also discussed. References Battifora H. The multitumor (sausage ) tissue block: novel method for immunohistochemical antibody testing. Lab Invest 1986, 55:244-48. Kononen J, Bunbendorf L, Kallioniemi A, Barlund M, Schraml P, Leighton S, Torhost J, Mihatsch MJ, Sauter G, Kallioniemi OP: Tissue microarrays for high-troughput molecular profiling of tumor specimens. Nat Med 1998, 4:844-47. Moch, Kononen T, Kallioniemi OP, Sauter G. Tissue microarrays: what will they bring to molecular and anatomic pathology? Adv Anat Pathol 2001, 8:14-20. Bubendorf L, Nocito A, Moch H, Sauter G. Tissue microarray (TMA) technology: miniaturized pathology archives for high-throughput in situ studies. J Pathol 2001, 195:72-79. Camp RL, Charette LA, Rimm DL. Validation of tissue microarray technology in breast carcinoma. Lab Invest 2000, 80:1943-49. Rubin MA, Dunn R, Strawderman M, Pienta KJ. Tissue microarray sampling strategy for prostate cancer biomarker analysis. Am J Surg Pathol 2002, 26:2249-2256. Matias-Guiu X, Garcia A, Curell R, Prat J: Renal cell carcinoma metastatic to the thyroid gland. A comparative molecular study between the primary and the metastatic tumor. Endocrine Pathol 1998 9:255-260 Matias-Guiu X, Lagarda H, Catasus Ll, Bussaglia E, Gallardo A, Gras E, Prat J: Clonality analysis in synchronous or metachronous tumors of the female genital tract. Int J Gynecol Pathol 2002 21: 205-211 Gras E, Cortes J, Diez O, Alonso C, Matias-Guiu X, Baiget M, Prat J:Loss of heterozygosity on chromosome 13q12-q14, BRCA-2 mutations and lack of BRCA-2 promoter hypermethylation in sporadic epithelial ovarian tumors.Cancer 2001 92:787-95