Article Text


In situ amplification of oestrogen receptor α mRNA in breast cancer cell lines and tumours
  1. L Lespagnard1,
  2. D Gancberg1,
  3. Y Maaroufi2,
  4. G Leclercq2,
  5. D Larsimont1,
  6. A Verhest1
  1. 1Department of Pathology, Institut Jules Bordet, 1 rue Héger-Bordet, 1000 Brussels, Belgium
  2. 2Laboratoire Jean-Claude Heuson de Cancérologie Mammaire, Institut Jules Bordet
  1. Dr Lespagnard llespa{at}


The aim of this work was to develop a direct in situ reverse transcription polymerase chain reaction (in situ RT-PCR) assay for the detection of oestrogen receptor α (ERα) mRNA on in vitro cell lines and breast tumour cell smears. ERα mRNA amplification was performed on MCF-7 (ERα positive) and MDA-MB-231 (ERα negative) cell lines as well as on 51 cytological smears of breast tumour samples from patients. The in situ amplification of mRNA in cell lines and ex vivo breast tumours was successful. However, finding an equilibrium between optimal cell morphology and PCR performance varied with each tumour, leading to difficulty in standardisation for daily practice. Nonetheless, in situ RT-PCR is a useful tool for the detection of ERα mRNA in selected cases, both in vitro and ex vivo. J Clin Pathol: Mol Pathol

Statistics from

The detection of oestrogen receptors α (ERα) is essential for establishing the optimal treatment for breast cancer. Because of the early detection of tumours these days, small tumours are frequently encountered. This has led to the development of techniques that require only a few cells. We describe a method for assessing the ERα status of breast tumours performed on cell smears by in situ reverse transcriptase polymerase chain reaction (in situ RT-PCR). As recently reviewed by Tartour et al,1 in situ RT-PCR is a variant of PCR that is very attractive to pathologists because it combines the high sensitivity of PCR with morphological identification.

Essentially, this approach enables the ERα status of a tumour to be determined using a very low number of cells and theoretically might be applicable to all tumours, even the smallest ones, for which sampling for multiple assays is not possible.

We developed our approach on MCF-7 (ERα positive) and MDA-MB-231 (ERα negative) breast cancer cell lines. In a second step, we assayed the practical feasibility of this method in clinical practice by performing cytological smears from breast tumour samples.

Materials and methods

Cytospots were performed for MCF-7 and MDA-MB-231 breast cancer cell lines purchased from American Type Collection Cells (Rockville, Maryland, USA).

Breast tumour cytological smears were performed either by scraping or by fine needle aspiration of the tumour and spreading of cells with the edge of a coverslip on silane treated slides. Based on the literature, we compared three fixations procedures, namely: 10% neutral buffered formalin (NBF) at room temperature, for either (1) a 10 minute or (2) an overnight incubation period,2 and (3) 10% NBF for 20 minutes at room temperature followed by a 20 minute incubation in a mix of ethanol/acetone (45/5) at 4°C.3 Before in situ RT-PCR, the slides that had undergone the first two fixation processes were pretreated with proteinase K (5 μg/ml in phosphate buffered saline (PBS) pH 7.4) for 10 minutes at 37°C, washed in PBS, and then incubated for five minutes at 95°C. Slides that had undergone the third fixation process remained untreated.

We performed a direct in situ RT-PCR procedure adapted from the method of Gey et al,2 with 5′ end biotinylated primers and an RT Th DNA polymerase, allowing the combination of reverse transcription and PCR in one step. We used the 5′ end biotinylated ERα specific sense primer (U4.1) (5′-ATGGTCAG TGCCTTGTTGGATGC-3′) located in exon 4 (1175–1197) and the 5′ end biotinylated antisense primer (L1) (5′-GCCCTCTACACAT TTTCCCTGGT-3′) located in exon 6 (1470–1492). These sequences delineate a region of 318 bp in wild-type mRNA and 180 bp in variants with specific exon 5 deletion (Delta 5 mRNA).4 PC03 and GH21 primers were used to amplify the control β globin gene and gave amplified products of 254 bp.5

The in situ RT-PCR mixture contained 10 μl of 5× EZ buffer (PE Biosystems, Norwalk, Connecticut, USA), 2.5 mM Mn(OAc)2, 300 μM dNTPs (Roche Molecular Biochemicals, Brussels, Belgium), 10 pmoles of each primer, and 10 U RNAse inhibitor (PE Biosystems) in a final volume of 50 μl. Slides were covered with amplicover discs assembled with amplicover clips using an assembly tool provided by the manufacturer and transferred to the heating block of the GeneAmp in situ PCR 1000 system (PE Biosystems). PCR conditions were as follows: 30 minutes at 60°C, followed by three minutes at 94°C and by 30 cycles of 30 seconds at 94°C and one minute at 60°C, with a final five minute extension at 60°C.

After PCR cycling, slides were rinsed twice in PBS for five minutes, postfixed for 10 minutes in 10% NBF, and rinsed in PBS. PCR products were detected by incubating for 20 minutes with streptavidin peroxidase. Finally, 3,3′ diaminobenzidine hydrochloride (DAB) was used as chromogen and haematoxylin as counterstain. In some cases, the PCR supernatant was set apart and run on a 2% agarose gel to check the amplicon size under UV transillumination after ethidium bromide staining. Controls without Taq polymerase or without biotinylated primers were included in each run. RNAse free DNAse (10 U/μl) treatment was performed in some cases to ascertain the specificity of the reaction.

For conventional RT-PCR experiments, RNA was extracted using Trizol (Roche Molecular Biochemicals), according to the protocol recommended by the manufacturer. First strand complementary DNA (cDNA) was reverse transcribed using random oligonucleotide primers. Aliquots of this reverse transcription reaction equivalent to cDNA from 500 ng total RNA were used in the PCRs. The PCR mixture contained 20 pmoles of each primer (see above), 200 μM dNTP, 10 mM Tris/HCl (pH 9), 50 mM KCl, 2.5 mM MgCl2, and 1 U Taq DNA polymerase. Cycling parameters were as follows: three minutes at 95°C; 30 cycles at 95°C for one minute, one minute at 52°C, and one minute at 72°C; and a final step of seven minutes at 72°C. The amplified products were separated by electrophoresis on 2% agarose gels and stained with ethidium bromide.

Results and discussion

First, we performed in situ and conventional RT-PCR in parallel using the same primers to check whether similar amplifications could be achieved. We detected PCR products of the expected size in 2% agarose gel both in the supernatant of in situ RT-PCR and by conventional RT-PCR from MCF-7 cells, whereas no signal was detected with MDA-MB-231 cells (fig 1). We then tested different fixation procedures to determine the optimal fixation conditions that would produce the best quality cell morphology and RT-PCR results. We found that a 10 minute incubation in 10% NBF at room temperature was adequate for both cell lines: MCF-7 and MDA-MB-231. Positive signals were found in MCF-7 cells (fig 2A) whereas MDA-MB-231 cells remained negative (fig 2B); the data correlated with those obtained in agarose gel described above. Of note, we found that an overnight fixation in 10% NBF decreased staining, leading to frequent false negative PCR results. This suggests that a prolonged proteinase K treatment, which is not ideal for cell preservation, would be required. A 20 minute fixation in 10% NBF at room temperature followed by a mix of ethanol/acetone (45/5) at 4°C for 20 minutes also produced good morphological and PCR results in the MCF-7 cell line.

Figure 1

Reverse transcription polymerase chain reaction (RT-PCR) for oestrogen receptor α (ERα) mRNA detection: (A) after in situ RT-PCR cycling, slides were disassembled (amplicover discs and clips removed: see material and methods), 18 μl of supernatant was loaded on to a 2% agarose gel, and the gels were visualised under UV after ethidium bromide staining; (B) after conventional RT-PCR, 18 μl of PCR product was loaded on to a 2% agarose gel, which was visualised under UV. Lanes 1, MDA-MB-231 negative cells; lanes 2, MCF-7 positive cells. PCR products of expected sizes (318 and 180 bp) were detected.

Figure 2

In situ detection of oestrogen receptor α (ERα) mRNA after in situ reverse transcription polymerase chain rection (RT-PCR) with biotinylated ERα primers: (A) MCF-7 positive cells, (B) MDA-MB-231 negative cells, (C) a positive breast tumour cell smear. Staining (brown) is seen in positive cells after incubation with streptavidin-peroxidase, followed by visualisation with DAB as chromogen and counterstaining with haematoxylin (blue).

We next analysed breast tumour smear samples using these methods and found that 10% NBF for 10 minutes was optimal (fig 2C). For the cell lines, an overnight fixation in 10% NBF led to decreased PCR intensity. In contrast, frequent false negative PCR results were seen with ex vivo breast tumour cells when a 20 minute fixation in 10% NBF followed by a mix of ethanol/acetone (45/5) for 20 minutes at 4°C was used, despite the fact that positive results were also obtained with this method. This may be related to the heterogeneity of tumours. Therefore, we decided that 10% NBF for 10 minutes would produce a good compromise between morphology and PCR requirements. Nevertheless, in these conditions only 23 of 51 cases still had well preserved morphology after PCR cycling. As already mentioned above, this point was difficult to standardise, depending essentially on tumour type (fibrous, necrotic, haemorrhagic, etc). Another observation related to the intertumour variability of the technique was the detection of in situ RT-PCR products either in the cytoplasm or in both the cytoplasm and the nucleus. Similar results were obtained after DNase treatment.

In our series of 51 breast tumour cell smears fixed in 10% NBF for 10 minutes, 47 cases were positive for ERα mRNA by in situ RT-PCR. Of note, the ERα status of 44 tumours was routinely determined by immunohistochemistry: 40 cases were positive using this conventional technique.

In summary, we successfully amplified ERα mRNA in situ in both in vitro cell lines and ex vivo cell smears from breast tumour samples. Nevertheless, our observations illustrate the difficulty of standardising ERα mRNA detection by in situ RT-PCR for intensive use in daily practice, because finding an equilibrium between PCR performance and cell morphology preservation seems to be tumour dependent. However, our results suggest that in situ RT-PCR is a potentially useful tool for ERα mRNA detection in selected cases both in vitro and ex vivo.


This work was supported by Oeuvre Belge du Cancer and by Lambeau-Marteaux Foundation.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.