All biopsy and resection specimens were reviewed by the Pathology Department of the University Hospital of Coimbra, to verify non-small cell histology of the lung cancer samples and to determine the histologic subtype. Histological subtypes consisted of 80 cases of surgically staged pulmonary carcinomas as well as corresponding non-neoplastic and pre-neoplastic tissue of adenocarcinomas and squamous cell carcinomas of the lung.

Twenty-four females and 16 males with adenocarcinoma (ADC), and 36 males and 4 females with squamous cell carcinoma (SCC) were confirmed according to the latest World Health Organization Classification of Lung Cancer (2004).7 The average age was 64.9±9.88 years at the time of diagnosis.

The TNM staging was applied according to the 2010 TNM Classification of Malignant Tumours, 7th Edition8: thirty-nine patients had Stage I disease, 19 had Stage II and 9 had stage III.

Representative areas of carcinoma, preneoplastic lesion and normal bronchial epithelium and parenchyma, were carefully selected from the haematoxylin-eosin slide, and marked respectively in each slide and the each formalin-fixed, paraffin-embedded block of tissue. Tissue microarrays of 3mm were performed in triplicate in tumour and normal respiratory epithelium and in a single core for preneoplastic lesion, included on the same slide.

IHC satins for hMLH1 and hMLH2 were performed on 3μm thick sections cut from paraffin-embedded TMAs blocks. Commercially available antibodies against these markers were used as per manufacturer's protocols. Monoclonal antibodies used were (clone ZM001, 1:200, Zymed) for MLH1and (clone FE11, 1:40, Zymed) for MSH2, using the Streptavidin–Biotin Horseradish Peroxidase method. Epitope retrieval using microwave heat or steam pre-treatment was performed as required. Normal endometrium was used as positive control. The normal staining pattern for both MLH1 and MSH2 is nuclear. Lymphocytes and normal bronchial epithelium were used as internal positive control as follow: Normal expression (>75%/+++); Reduced expression: Negative (≤10%/−) and Positive (>10–50%/+ or 50–75%/++).

Following the explained score it was possible to define two definite groups, the positive when present +++/75% nuclear expression and the group with reduced expression. Staining results were examined without knowledge of the status of the molecular analyses.

For DNA extraction, areas from all the slides corresponding to the principal histological patterns were selected. The tumour-rich areas were manually micro-dissected. Five 10μM sections were cut from the paraffin blocks corresponding to the already selected areas and deparaffined by xylene extraction. The resulting tissue pellets were digested with Proteinase K at 56°C overnight. The genomic DNA was isolated using QIAamp DNA MiniKit according to the manufacturer’ protocol (Qiagen, Valencia, CA). By spectrophotometry (Nanodrop ND1000, Thermo Scientific, USA) the purity and concentration of the samples were verified.

The promoter methylation status of the MLH1 gene of all tumour samples and of the MSH2 gene of tumour samples was determined by chemical treatment with sodium bisulfite and subsequent methylation-specific PCR (MSP) analysis as described.9

Two sets of primers were specifically designed upstream of the promoter region to discriminate methylated from unmethylated alleles, using the Methprimer Software10 (Table 1).

The bisulfite modification was performed using the Epitect Bisulfite Kit (Qiagen Valencia, CA) to convert all unmethylated cytosines to uracils, while leaving methylated cytosines unaffected (Table 2).

Modified DNA was amplified in a total volume of 110μL containing 1× NH4 buffer, 2mM of MgCl2, 200μM of dNTPs, 1U Taq DNA Polymerase (Bioline), different primer pair concentrations were used after optimization (Table 1), considering that we are working with FFPE tissue, an optimization procedure confers better results. PCR was performed for 40 cycles with annealing temperatures of 61°C for 30s and primer extension at 72°C for 60s using 50ng bisulfite-modified DNA. All PCRs were performed with positive controls for both unmethylated and methylated alleles and no DNA control. DNA of healthy human lymphocytes served as positive control for methylated and unmethylated MLH1 and MSH2 state. The first was treated with Methyltransferase (CpG Methyltransferase M.SssI) (New England Biolabs), and subsequently submitted to sodium bisulfite transformation (Epitect Bisulfite, Qiagen), the second was only modified with sodium bisulphite. The CpG Methyltransferase, M.SssI, methylates all cytosine residues (C5) within the double-stranded dinucleotide recognition sequence 5′…CG…3′.

The PCR products were separated on a 4% agarose gel containing ethidium bromide and visualized under UV light, and its size estimated by comparison with a DNA Molecular Weight Marker XIII (50bp ladder) (Roche).

The association between clinicopathological variables (pTNM, gender, age) and frequency of methylation and protein expression silencing among the tumour subtypes was assessed using the Pearson χ2 and Fischer's exact test. All tests were two-sided and the level of significance was set at p<0.05.

We investigated MLH1 and MSH2 expression using immunohistochemistry analysis in tissue microarrays comprising 80 primary NSCLC patients, that were respiratory epithelium, preneoplastic lesion (basal cell hyperplasia, metaplasia, dysplasia) and tumoural patterns. Following the explained score it was possible to define two definite groups, normal expression with >75%/+++ nuclear expression and the group with reduced expression <75%/++. After applying these scores, the internal control of lymphocytes had normal expression (data not shown) but tumour adjacent respiratory epithelium and parenchyma were considered reduced for both MLH1 and MSH2. Twenty-three cases (57%) of ADC for MLH1 expression and 27 cases (67%) for MSH2, had reduced expression; in SCC 29 cases (72%) for MLH1 expression and 25 cases (62%) for MSH2 expression had reduced expression.

Co-reduction of both MLH1 and MSH2 was found in 38 cases (19 ADC and 19 SCCs), according with Table 5, probably due to different altered epigenetic or/and genetic mechanisms involved in down-regulation of gene expression. Tumour cells that exhibited an absence of nuclear staining in the presence of non-neoplastic cells and infiltrating lymphocytes with nuclear staining were considered to have an abnormal pattern.

In normal respiratory epithelium reduced expression was found in 70% MLH1 and 75% MSH2 in ADC, and 100% MLH1 and 97% in MSH2 in SCC, to study this we would need cancer-free controls to explain the possible time required to acquire the additional genetic and epigenetic changes that promote tumour progression.

Basal cell hyperplasia in cases of adenocarcinoma presented normal expression of MLH1 and MSH2. Squamous cell carcinomas revealed reduced expression of MLH1 and MSH2 in 72% and 62.5% respectively. Fig. 1 represents the IHC expression heterogeneity.

Figure 1.

MLH1 IHC expression in one mixed type adenocarcinoma, (A) basel cell hyperplasia, (B) acinar pattern with reduced expression (−), (C) bronchioloalveolar pattern with normal expression (+++) and (D) solid pattern with reduced expression (++); MSH2 IHC expression in epidermoid carcinoma in situ and (E) corresponding epidermoid carcinoma (F) 200×.

(0.76MB).

We examined the promoter hypermethylation status of MLH1 and MSH2 genes in 40 squaumous cell carcinoma and 33 adenocarcinoma using MSP assay in manually microdissected tissue. 40 DNA samples of squamous cell carcinoma and 33 samples of adenocarcinoma were analyzed (Table 3).

We observed two patterns of hypermethylation; the partial pattern was the most prevalent (58.33%) for the MLH1 gene (M-U) while for the MSH2 gene the hypermethylation fully pattern (M-M) was the one most verified (76.9%).

MLH1 methylation pattern seems to vary substantially by histological type (Table 3). The prevalence of MLH1 promoter hypermethylation was higher in squamous cell carcinoma (47.5%), with a correlation (p=0.003) between the histological subtype and the methylation status of the promoter of MLH1 gene. The difference is not so obvious in MSH2 promoter hypermethylation; however adenocarcinoma has the highest number of cases (42.4%). No correlation was found between histological subtype and methylation status of MSH2 (p=0.2699).

According to the relationship between the frequency of gender and histological subtype, women with adenocarcinoma as well as men with squamous cell carcinoma have a higher percentage of hypermethylation of MLH1 and MSH2 promoters (Table 4).

According to the distribution of gene promoter methylation and loss of expression at tumour type in 80 patients, no statistical correlation was observed between the promoter methylation and expression levels in tumour samples, p=0.09 for MLH1 and p=0.894 for MSH2 (Tables 5–7).

Moreover, no correlation was found between MLH1 and MSH2 methylation levels and any of the clinicopathological characteristics considered (age, gender, TNM stage and tumour histological type) (Table 7).

We observed two patterns of hypermethylation; the partial pattern was the most prevalent (58.33%) for the MLH1 gene (M-U) while for the MSH2 gene the hypermethylation fully pattern (M-M) was the one most verified (76.9%).

The authors declare that no experiments were performed on humans or animals for this study.

The authors declare that they have followed the protocols of their work centre on the publication of patient data and that all the patients included in the study received sufficient information and gave their written informed consent to participate in the study.

The authors declare that no patient data appear in this article.