PCR Amplifiable DNA from Breast Disease FFPE Section for Mutational Analysis

Abstract

Formalin-fixed paraffin-embedded (FFPE) tissue specimens have been a staple of research, providing precious resources for molecular and genomic studies. However, the biggest challenge is the extraction of high-quality DNA from FFPE tissues, given that the integrity of DNA is critically affected by formalin fixation. Formaldehyde induces crosslinks in DNA that renders single or double-stranded DNA breaks. Such breaks cause extensive fragmentation that directly influences the quality of DNA purified and the number of templates available for PCR amplification. Thus, protocol for DNA purification from FFPE tissues must effectively extract highly fragmented DNA and reverse cross-linking caused by formalin fixation. DNA extraction methods available in the literature were selected and modified at different stages to optimize a protocol that extracts DNA of sufficient quality and fragment size to be detectable by PCR. Archived FFPE tissues belonged to patients with triple negative breast cancer (TNBC) and benign breast disease were used for the protocol optimization. The best optimized protocol was then used to amplify Exon 4 region of Proviral integration site for Moloney murine leukemia virus1 (Pim1) kinase gene to analyze any probable somatic mutations both in TNBCs and benign breast diseases. Of the 12 different protocols developed, best quality DNA in terms of fragment size and purity was obtained when Tween20 lysis buffer was used for both deparaffinization and overnight digestion along with high salt precipitation. Optimized protocol was then validated by extracting DNAs from 10 TNBCs and 5 benign breast disease specimens with consistent purity and fragment size. PCR amplification and subsequent Sanger’s sequencing revealed the presence of mutations in the Exon 4 region of Pim1 kinase. Deparaffinization and overnight digestion in Tween20 lysis buffer along with high salt precipitation yielded the best quality PCR amplifiable DNA for mutational analysis.

INTRODUCTION

Formalin-fixed paraffin-embedded (FFPE) technique is widely used for the long-term preservation of clinical tissues. With wide research focusing on the area of biomarker discovery, FFPE tissue sections have become a valuable tool in retrospective genetic research. However, the biggest challenge is the extraction of high-quality DNA from FFPE tissues, given that the integrity of DNA is critically affected by the formalin fixation.^{1, 2}

Formaldehyde, being a reactive electrophile, creates various crosslinks such as protein-protein, protein-DNA, DNA-formaldehyde as well as interstrand DNA adducts.^{3, 4} It essentially crosslinks DNA by reacting with imino groups of DNA bases, thereby reducing the number of hydrogen bonds in the DNA double helix.⁵ Furthermore, it initiates changes at Adenine:Thymine (AT)-rich regions of double-stranded DNA and induces generation of N-hydroxymethyl monoadducts on guanine, adenine, and cytosine. It also creates N-methylene crosslinks between adjacent purines, generates apurinic and apyrimidinic sites, and causes hydrolysis of phosphodiester bonds leading to short chains of polydeoxyribose with intact pyrimidines.⁶ Likewise, formaldehyde causes cross-linking of DNA bases with nearby histones, triggering changes in its conformation.⁴ Hence, formaldehyde-induced crosslinks of DNA reduce the stability of double-stranded DNA, resulting in a single- or double-stranded DNA breaks. Such DNA breaks cause extensive fragmentation that directly influences the quality of DNA purified and the number of templates available for PCR amplification.⁷ Thus, the protocol for DNA purification from FFPE tissues must effectively extract highly fragmented DNA and reverse cross-linking caused by formalin fixation.

Isolation procedure can be amended by tweaking deparaffinization, optimizing digestion condition of proteinase, and improving the procedure for extraction and purification. Our main objective was to improvise and fine-tune the existing methods so as to optimize a protocol that can extract DNA of sufficient quality and fragment size to be detectable by PCR.

Henceforth, we selected methods available in the literature and modified at various steps to isolate good quality DNA from archived FFPE tissues belonged to patients with triple negative breast cancer (TNBC) and benign breast disease. The best optimized protocol was then used to amplify Exon 4 region of Pim1 kinase gene to analyze any probable somatic mutation.

MATERIALS AND METHODS

Ten TNBC and 5 benign breast disease FFPE blocks archived at the Department of Pathology, Krishna Institute of Medical Science, Hyderabad, India, were retrieved for this study. All the FFPE blocks had been sampled between 2004 and 2009 and fixed in 10% neutral buffered formalin for up to maximum of 24 h at room temperature as per the hospital records.

Protocol optimization was done with 2 samples, 1 each from TNBC and benign breast disease. Tissue blocks were cut on a standard microtome to generate sequential sections ranging between 10 and 20 µm. Twenty to thirty sections from each sample amounting to ∼10–15 mg of deparaffinized tissue were collected into microcentrifuge tube and processed individually.

Control protocol using QIAamp isolation kit

Deparaffinization of tissue sections were done through immersion in 100% xylene for 5 min followed by 2 times centrifugation to remove xylene, washed twice serially in ethanol (1 × 100%, 1 × 70%, and 1 × 50%), and allowed to air dry at 75°C. The dried pellet was then further processed using QIAamp DNA FFPE tissue kit following manufacturers’ instructions.

Optimization of protocol

Different protocols were selected from published literature and modified at various stages of DNA extraction process (as depicted in Fig. 1).

FIGURE 1.

Schematic representation of protocols modified at various stages of DNA extraction process.

Deparaffinization

Deparaffinization of tissue section was achieved by following 4 different methods.

• A₁: Boiled in Tween 20 lysis buffer (0.5% Tween 20, 1 mM EDTA, 50 mM Tris-HCl pH8.5) at 95°C for 10 min followed by centrifugation at 140,000 rpm for 5 min. A second round of incubation with fresh lysis buffer at 65°C for 10 min followed by centrifugation at 140,000 rpm for 5 min.

• A₂: Immersion in 100% xylene for 5 min followed by centrifugation at 140,000 rpm for 5 min and serial ethanolic hydration (1 × 100%, 1 × 70%, and 1 × 50%).

• A₃: Boiled in hot water at 95°C for 20 min followed by centrifugation at 140,000 rpm for 5 min. A second round of incubation with fresh hot water at 95°C for 20 min followed by centrifugation at 140,000 rpm for 5 min.

• A₄: Boiled in alkali digestion buffer (0.1 M NaOH, 1% SDS pH 12.0) at 100°C for 40 min followed by centrifugation at 140,000 rpm for 5 min.

Digestion

Digestion of deparaffinized tissue pellet was performed with 2 types of lysis buffers (both containing proteinase K at the concentration of 20 μg/ml), at 2 different digestion condition.

• B₁: Tween20 lysis buffer at 65°C for O/N.

• B₂: Alkali digestion buffer at 100°C for 40 min.

A₁ and A₄ were digested with only B₁ and B₂, respectively. Whereas, A₂ and A₃ were split into 2 samples and digested with B₁ and B₂, respectively.

Extraction

Phenol:phenol:chloroform:isoamyl alcohol extraction was performed on the all the digested samples. Briefly, an equal volume of Tris-saturated phenol (pH 8) was added to lysates in 6 tubes (labeled as A₁B₁, A₂B₁, A₂B₂, A₃B₁, A₃B₂, A₄B₂) and vortexed gently. Tubes were then centrifuged for 5 min at 14,000 rpm and aqueous phase was then transferred to fresh microcentrifuge tubes. An equal volume of phenol:chloroform:isoamyl alcohol (25:24:1) was added to the aqueous phase, vortexed again, and then centrifuged for 5 min at 14,000 rpm. The upper aqueous phase was transferred to a new tube and treated with RNase A (100 µg/ml) for 1 h at 37°C.

Precipitation

DNA precipitation was performed with and without salt treatment as follows.

• D₁: Exposure to 5 M NaCl at the ratio of 1:25 at 65°C for 4 h followed by precipitation with 0.1 the volume of 3 M sodium acetate (pH 5.2) and 1 volume of 100% isopropanol at –20°C O/N.

• D₂: Precipitation with 0.1 the volume of 3 M sodium acetate (pH 5.2) and 1 volume of 100% isopropanol at –20°C O/N.

Aqueous phase collected in 6 microcentrifuge tubes were split into 12 tubes (labeled as A₁B₁CD₁, A₁B₁CD₂, A₂B₁CD₁, A₂B₁CD₂, A₂B₂CD₁, A₂B₂CD₂, A₃B₁CD₁, A₃B₁CD₂, A₃B₂CD₁, A₃B₂CD₂, A₄B₁CD₁, A₄B₁CD₂) and DNA was precipitated with or without addition of NaCl. All the samples were then centrifuged at 14,000 rpm for 5 min. The pelleted DNA was washed gently once with 70% ice cold ethanol, dried and reconstituted with nuclease-free water.

DNA quality and yield

The quality of DNA extracted from each protocol was evaluated using agarose gel electrophoresis and the absolute yield of extracted DNA was quantified using a NanoDrop ND-1000 (Thermo Fisher Scientific, Waltham, MA, USA).

DNA fragment analysis by PAGE

Small (<1000‐bp) DNA fragments were separated by conventional PAGE using the method described by Chory and Pollard.⁸

Downstream processing of DNA extracted from FFPE tissue using the optimized protocol

The protocol that extracted the best quality DNA in terms of fragment size, purity, and concentration was then employed to extract DNA from 10 TNBCs and 5 benign breast disease FFPE blocks. A multiplex PCR was done to evaluate the ability of extracted DNA to amplify housekeeping gene glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (226 bp) and Exon 4 of Pim1 kinase (414 bp) using the set of primers as shown in Table 1. The amplified products were subsequently subjected to Sangers’ sequencing for mutational analysis.

TABLE 1.

Primer sequences of GAPDH and Exon 4 of Pim1 kinase

Target gene	Forward primer, 5′–3′	Reverse primer, 5′–3′
GAPDH (226 bp)	GAAGGTGAAGGTCGGAGTC	GAAGATGGTGATGGGATTTC
Pim1 kinase Exon 4 (414 bp)	GCCCCTGCAGCCTAATGG	GAGTCACCTGGGCAGCTC

Statistical analysis

Every experiment was repeated at least 3 times and all the values were averaged. Data were expressed as means ± sd.

RESULTS

A comparison was made between commercial kit-based protocol and protocols modified at various stages of DNA extraction in terms of concentration, purity, and fragment size. The results presented in Table 2 showed that the quality and quantity of DNA from FFPE tissues (TNBCs and benign breast disease) depend on the methods of extraction at different stages. The best quality DNA in terms of fragment size and purity was obtained when deparaffinization done in Tween20 lysis buffer followed by overnight digestion in the same buffer and subsequent precipitation in high salt concentration.

TABLE 2.

Concentration, purity, and fragment size of DNA extracted from FFPE tissues (benign and TNBC) using protocol modified at various stages

Extraction method	Tissue type	Concentration (ng/μl)	A260/280	Fragment size (base pairs)
QIAamp	Benign	54.92 ± 6.54	1.743 ± 0.012	300–500
	TNBC	62.70 ± 3.11	1.756 ± 0.006
A1B1C1	Benign	689.73 ± 38.07	1.947 ± 0.031	700–800
	TNBC	618.39 ± 46.68	1.943 ± 0.021
A1B1C2	Benign	382.22 ± 33.14	1.803 ± 0.015	300–500
	TNBC	312.94 ± 29.88	1.800 ± 0.010
A2B1C1	Benign	207.32 ± 18.25	1.396 ± 0.025	300–400
	TNBC	213.47 ± 10.33	1.390 ± 0.044
A2B1C2	Benign	104.70 ± 9.33	1.303 ± 0.015	200–300
	TNBC	115.26 ± 13.79	1.293 ± 0.006
A2B2C1	Benign	79.27 ± 9.10	1.606 ± 0.041	200–300
	TNBC	67.43 ± 13.63	1.597 ± 0.0153
A2B2C2	Benign	57.42 ± 4.70	1.533 ± 0.012	200–300
	TNBC	54.04 ± 9.83	1.523 ± 0.006
A3B1C1	Benign	1496 ± 102.44	1.800 ± 0.020	400–500
	TNBC	1612.86 ± 97.63	1.787 ± 0.006
A3B1C2	Benign	917.74 ± 76.97	1.617 ± 0.029	300–400
	TNBC	840.52 ± 50.11	1.627 ± 0.040
A3B2C1	Benign	341.43 ± 25.60	1.893 ± 0.012	200–300
	TNBC	314.17 ± 30.33	1.910 ± 0.017
A3B2C2	Benign	369.15 ± 33.08	1.753 ± 0.012	200–300
	TNBC	360.18 ± 30.55	1.747 ± 0.012
A4B2C1	Benign	111.51 ± 9.84	1.703 ± 0.035	200–300
	TNBC	94.66 ± 14.87	1.710 ± 0.020
A4B2C2	Benign	51.67 ± 8.47	1.447 ± 0.006	200–300
	TNBC	61.13 ± 5.08	1.470 ± 0.017

Data shown are from 3 independent experiments and represented as mean ± sd. A₁B₁CD₁: deparaffinized in tween20 buffer + digested in tween20 buffer + PCI extraction + high salt precipitation. A₁B₁CD₂: deparaffinized in tween20 buffer + digested in tween20 buffer + PCI extraction + No salt treatment. A₂B₁CD₁: deparaffinized in xylene + digested in tween20 buffer + PCI extraction + high salt precipitation. A₂B₁CD₂: deparaffinized in xylene + digested in tween20 buffer + PCI extraction + No salt treatment. A₂B₂CD₁: deparaffinized in xylene + digested in alkali buffer + PCI extraction + high salt precipitation. A₂B₂CD₂: deparaffinized in xylene + digested in alkali buffer + PCI extraction + No salt treatment. A₃B₁CD₁: deparaffinized in hot water + digested in tween20 buffer + PCI extraction + high salt precipitation. A₃B₁CD₂: deparaffinized in hot water + digested in tween20 buffer + PCI extraction + No salt treatment. A₃B₂CD₁: deparaffinized in hot water + digested in alkali buffer + PCI extraction + high salt precipitation. A₃B₂CD₂: deparaffinized in hot water + digested in alkali buffer + PCI extraction + No salt treatment. A₄B₂CD₁: deparaffinized in tween20 buffer + digested in tween20 buffer + PCI extraction + high salt precipitation. A₄B₂CD₂: deparaffinized in tween20 buffer + digested in tween20 buffer + PCI extraction + No salt treatment. Bold indicates the best optimized protocol.

PCR amplification of 226 bp GAPDH gene fragment was evident in all DNA samples extracted through 12 different protocols (unpublished results). However, PCR amplification of 414 bp Exon 4 of Pim1 kinase gene was accomplished only in A₁B₁C₁ and to a certain extent in A₃B₁C₁ extraction method. From this, it can be deduced that appropriate deparaffinization, overnight digestion, and high salt treatment enabled more intact and amplifiable DNA for mutational analysis.

Efficacy of A₁B₁C₁ extraction process was then validated through DNA extraction and consequent GAPDH and Pim1 kinase amplification from 10 TNBCs and 5 benign breast disease specimens. All FFPE specimens yielded significantly good quality DNA in terms of concentration and purity (Fig. 2A) and higher fragment size (Fig. 2B). Successful PCR amplification of GAPDH and Pim1 kinase was apparent in all the studied specimens as intense bands of PCR products (Fig. 2C). Besides, we successfully detected several mutations in Exon 4 region of Pim1 kinase, notably deletion of 2 nt at 6020–6021 in TNBCs (Fig. 2D) through Sanger’s sequencing.

FIGURE 2.

Validation of optimized protocol (A₁B₁C₁) in breast FFPE specimens. A) Quality and quantity. B) Fragment analysis. C) PCR amplification of Pim1 kinase and GAPDH of DNA isolated from breast FFPE specimens. D) Sanger’s sequencing (chromatogram) showing deletion of adenine (A) and guanine (G) at position 6020–6021 of Exon 4 region of Pim1 kinase. Data shown are from 3 independent experiments. Quality and quantity analysis of extracted DNA are represented as means ± sd.

DISCUSSION

Of all the deparaffinization techniques, melting of wax in hot water⁹ yielded the highest concentration of DNA as compared with other methods, albeit with slightly lesser purity and predominately smaller fragments. In our study, xylene, the most common deparaffinized agent, proved to be a contributor to the DNA fragmentation with poor quality and quantity.¹⁰ Deparaffinization and extraction with alkali buffer did not show an efficient DNA extraction, although Gilbert et al.¹¹ have demonstrated that the quality of DNA obtained through alkali buffer digestion was optimum for PCR reactions.

Furthermore, we observed that overnight digestion at 65°C improved the efficiency of DNA extraction from FFPE tissues.¹² Phenol chloroform extraction is the conventional method for organic extraction of nucleic acids. However, we found better extraction efficacy with phenol:chloroform:isoamyl alcohol rather than with phenol chloroform extraction (unpublished results).

Exposure of too high salt concentration before precipitation with isopropanol enabled the neutralization of negatively charged DNA that subsequently allowed the isopropanol to remove the hydration shell of water molecules around the phosphate leading to effective precipitation of DNA from the solution.

CONCLUSIONS

Deparaffinization and overnight digestion in Tween20 lysis buffer followed by phenol:chloroform:isoamyl extraction and high salt precipitation yielded the best quality amplifiable DNA for mutational analysis.