CONSTITUTIONAL MUTATIONS IN GENES RELATED TO PREDISPOSITION FOR TUMORS

On the basis of literature data and own experience the authors review the current knowledge about the molecular basis of inherited predispositions for tumors. They hypothesize that in the near perspective 5-10 years studies using existing registry data/material and the latest novel technology will allow the identification of the molecular background for the majority of hereditary cancers which will have enormous practical consequences especially for the prevention of malignancies.

It is now well recognized that 5-10% of all tumors including so-called common malignancies such as cancers of the breast, colon and ovaries occur as a result of high monogenic predispositions.It is estimated that an addi-tional 30-50% of malignancies are also "inherited tumors", however they occur due to polygenic predispositions.The molecular basis of polygenic tumors is not known, although its existence is strongly suggested by results of studies on monozygotic twins.In contrast, monogenic background has already been proven for a significant proportion of hereditary cancers (Lichtenstein et al., 2000).
Genes/syndromes most frequently examined for the occurrence of constitutional mutations are summarized in Table 1.

STK11
Peutz-Jegers syndrome breast, pancreas Westerman et al., 1999;Yoon et al., 2000 reached about 50, there is still a large proportion of strong tumor aggregations without known molecular defects.For example: a) Since 2001 we performed the first worldwide population screening for familial cancers in West-Pomerania.During the first year of this programme we performed a complete diagnostic process including pedigree analysis, DNA testing and clinical examination of about 400 000 individuals.About 2600 families with at least three first degree relatives affected by tumors, thus showing the most characteristic feature of monogenic dominant genetic disorders, have been identified, but constitutional mutations were detected in only about 5% of them (Lubiñski et al., 2002).
b) We performed sequencing in 200 families with at least three cases of breast/ovarian cancers.Constitutional mutations of the BRCA1 gene were found in 63% (126/200) and of the BRCA2 gene in 3% (6/200) of families.Most probably about 30% of cancer aggregations are caused by BRCAX gene/genes mutations.Large BRCA1/BRCA2 deletions were found only in five cases (Górski et al., 2002).(Suchy et al., 2002).Using of lower case "c" in front of the nucleotide number is recommended for cDNA by the Nomenclature Working Group (Antonarakis SE and the Nomenclature Working Group (1998) Recommendations for a Nomenclature System for Human Gene Mutations.Hum Mutat.; 11: 1-3).
endometrium have been recognised and at least one of these tumors has been diagnosed under age of 50 (Kurzawski et al., 2002a).Thus, again there is very high probability that in a significant proportion of families with strong aggregation of colorectal/ endometrial cancers the genes responsible for a major molecular defect are not known yet.We were able to show that one of such genes may be MSH6.The MSH6 gene is one of the mismatch repair genes.Mutations in this gene, in opposite to MSH2/MLH1 genes, are much rarer and exhibit distinct phenotype.In our MSH6 family tumor spectrum include endometrioid ovarian cancer, colon cancer and two endometrial cancers.In patient, with previously excluded germline mutation in MSH2/MLH1 genes, we identified in exon 5 of MSH6 gene the frameshift mutation (Figs. 1 and 2) (Suchy et al., 2002).d) One of the genes recognized as a cause of stomach cancer familial aggregations is E-cadherin.We sequenced 53 patients with stomach cancers from families with clinical features of hereditary gastric tumors (at least three relatives affected by stomach cancer at any age or two relatives affected by stomach cancer with at least one diagnosed under age of 50) and in no case the E-cadherin constitutional mutations have been found (A.Jakubowska, unpublished data).

IDENTIFICATION OF MOLECULAR BASIS OF STRONG CANCER FAMILY AGGREGATIONS
Major work has still to be performed in order to identify molecular basis of strong cancer family aggregations.A promising methodology is linkage analysis with intragenic markers for candidate genes.It is reasonable to expect that for example for strong cancer aggregations with the involvement of breast cancer genes known to be connected with BRCA1/BRCA2, and for cancer aggregations with the involvement of colon cancer genes known for their aggregation with MSH2/MLH1 should be considered (Tables 2-5, Fig. 3).Interactions between BRCA1, BRCA2, MSH2 and MLH1 and associated gene proteins have been recognised using different systems, mainly by immunoprecipitation and mass spectrometric analysis of associated proteins (Deng & Brodie, 2000;Wang et al., 2000).

MODIFIERS
The cancer risk in BRCA1, MSH2 and MLH1 mutation carriers is at the level of about 50-80%.It has been shown recently that the risk heterogeneity may be related to modifying environmental and genetic factors (Table 6).
The results of studies on modifying features are not consistent.In our opinion this is caused by methodology problems: short series of cases studied, differences in ethnic origin and mutation types, the lack of matching for environmental features.Recently, we performed studies on the modifying effect of the polymorphic 135C®G Rad51 allele on breast cancer risk in carriers of BRCA1 5382insC.Analyses on 83 pairs matched very carefully for reproductive factors allowed us to show more than 2x risk reduction in women with the 135C®G form of Rad51.This form has been found in 37% of the unaffected BRCA1

CONCLUSIONS
The molecular basis of the genetic background for hereditary cancers has been discovered for a significant proportion of tumors.However, for the majority of hereditary cancers there is still lack of fundamental knowledge.Summarised results indicate directions of studies on identification of novel genes involved in pathogenesis of cancer family syndromes, which can be planned in the near future.It can be expected that in the near perspective of 5-10 years studies using existing registry data/material and the latest novel technology will allow the identification of the molecular background for the majority of hereditary cancers which will have enormous practical consequences especially for the prevention of malignancies.

Figure 1 .
Figure 1.The first Polish MSH6 mutation c.3311-3312 delTT.Detection of a germline MSH6 mutation by sequence analysis; arrow indicates the site of deletion; a) mutant; b) wilde type(Suchy et al., 2002).Using of lower case "c" in front of the nucleotide number is recommended for cDNA by the Nomenclature Working Group (Antonarakis SE and the Nomenclature Working Group (1998) Recommendations for a Nomenclature System for Human Gene Mutations.Hum Mutat.; 11: 1-3).