A novel approach for extracting DNA from formalin-fixed paraffin-embedded tissue using microwave

Abstract

Background

Formalin-fixed paraffin-embedded (FFPE) tissue is the gold standard procedure for tissue preservation. However, the extraction of DNA is a cumbersome procedure as the extracted DNA is fragmented because of the cross-linking effect of formalin. Hence, the aim of the present study is to extract DNA from FFPE tissues using different techniques with a specific objective of comparing the extracted DNA, both quantitatively and qualitatively.

Method

Ten samples of FFPE tissues were retrieved from the archives of the Department of Oral Pathology. Total genomic DNA was extracted by different methods which included QIAamp DNA FFPE Tissue Kit, Norgen DNA FFPE Tissue Kit, phenol–chloroform method, mineral oil–based extraction, M/10 NaOH solution method, and microwave method. A 280–base pair sequence was selected for evaluation of downstream amplification.

Results

The statistical analysis was performed using unpaired student's t-test to compare the DNA yields and quality obtained by microwave methods with other methods using SPSS software. Total genomic DNA retrieved by the microwave method was superior to other methods both qualitatively and quantitatively.

Conclusion

DNA extraction from FFPE tissues is an onerous task as irreversible bonds form between the nucleic acid during fixation which are difficult to break during DNA retrieval. Hence, the microwave method provides good total genomic DNA which gives better downstream results when compared with other methods.

Introduction

With the advancement in molecular biology, interest in the field of genetic sciences has increased over the past decade. Genomic DNA extraction from fresh tissue is the easiest method. However, this is not the case as the DNA retrieval with preserved tissues is a tedious process. Hence, formalin-fixed paraffin-embedded (FFPE) tissues from hospitals and tissue banks are the main source of total genomic DNA extraction for retrospective studies planned over a long period of time.

Formalin fixation of tissues followed by paraffin embedding is the oldest and gold standard procedure for tissue preservation. It is the cheapest and most feasible way of preservation. However, strong cross-linking property of formaldehyde is the main disadvantage, which affects the effective retrieval of genetic material due to chemical modifications and degradation of DNA and RNA, resulting in their limited bioavailability. Fragmentation of the nucleic acids occurs during the extraction procedure as strong covalent bonds are formed by formalin during fixation procedures.^{1, 2}

Antigen retrieval method is a routine procedure in immunohistochemistry (IHC), which is supposed to break the cross-linking caused by formaldehyde. Shi et al.² was the first person to incorporate microwave method of antigen retrieval in IHC. Using the same phenomenon, the present study aims at extraction of total DNA from FFPE tissues by different methods and to compare it with our microwave method to assess the extracted DNA qualitatively and quantitatively.

Materials and methods

Ten samples of FFPE were retrieved from the archives of the Department of Oral Pathology and Microbiology, Maulana Azad Institute of Dental Sciences, New Delhi. The 10 samples comprised 4 benign odontogenic cysts, 2 ameloblastoma, odontogenic tumors, 2 keratocystic odontogenic tumors, and 2 oral submucous fibrosis, potentially malignant disorder. Genomic DNA from the same tissue samples was extracted by using six different methods. The total DNA was extracted using the commercially available kits, phenol–chloroform (PC) method, mineral oil method, M/10 NaOH solution method, and microwave extraction method. Five serial sections of 5-μm thickness with area (16–20 mm²) per sample were taken in microcentrifuge tubes, and deparaffinization was performed.

DNA extraction method

Commercially available kits

• a)QIAamp DNA FFPE Tissue Kit and
• b)Norgen DNA FFPE Tissue Kit
• c)PC method

The total genomic DNA was extracted from FFPE tissues by following a standardized protocol using xylene to deparaffinized as described by Shi et al.²

• d)Mineral oil method

At 90 °C for 20 min, FFPE tissues were deparaffinized in mineral oil followed by conventional PC method.³

• e)M/10 NaOH method as described by Shi et al.²

Five hundred microliters of M/10 NaOH was added to each microtube containing two 10-mm thick tissue sections. The mixture was heated at 120 °C using an autoclave for 20 min and then cooled down for 15 min; 0.2 ml of the retrieval solution was aspirated for Polymerase Chain Reaction (PCR) test, while the remaining sample was stored at 4 °C for further use.

• f)Microwave method

The FFPE tissue samples were deparaffinized and kept in 0.1M phosphate-buffered saline (PBS) solution preceding washing. The tissues in PBS solution were heated in microwave for 2 min at 400 MW and subsequently for 2 min at 800 MW. The tissue samples were homogenized and mixed with 500 micron-liter lysis buffer (10 mM Tris–HCl, pH 8.0; 100 mM EDTA, pH 8.0; 50 mM NaCl; 0.5% SDS; 200 μg/ml proteinase K [add just before use]). Subsequent steps were similar to those of the PC method.

Assessment of DNA quantity and quality and downstream process

The quantity of extracted DNA was assessed using the Nanodrop, BioPhotometer plus, and quality was assessed by visualizing the band after loading the DNA sample in 0.8% agarose gel electrophoresis at 80 W. The A260:A280 ratio was taken as ∼1.80 for assessing the purity of DNA.

For downstream application, the extracted DNA was run in PCR with a known primer of β actin, that is, forward sequence: GTCCCCGTCTGGCCTGGCT and reverse sequence: ACTGACTACCTCATGAAGAT. 150 ng of template DNA in a 2-μl volume was mixed with a 0.8-μl volume of each primer pair (β actin) at concentrations of 10 pmol and added to 22 μl of PCR Master Mix (New England Biolabs). The thermal cycling was then carried out with an initial 5-min denaturation step at 95 °C. This step was coupled to a repeating cycle of 1 min at 95 °C (denaturation), 1 min at 59 °C (annealing), and 1 min at 72 °C (extension) for 36 cycles, followed by a 7-min completion step at 72 °C, and final cooling at 20 °C. After obtaining the PCR product, band formation in agarose gel was assessed.

Statistical analysis

The statistical analysis was performed using unpaired student's t-test to compare the DNA yields and quality obtained by microwave methods with other methods using SPSS software. The differences were considered significant at p ≤ 0.05.

Results

Assessment of quantity and quality and downstream processing

Our study indicated that the total DNA extracted by QIAamp DNA FFPE commercial tissue kit was in good quantity in comparison with Norgen DNA FFPE Tissue Kit and other extraction methods, except microwave method that yielded maximum DNA concentration (Table 1).

Table 1.

Concentration and quality of DNA by different methods.

S. No	Protocol used	No. of samples	Total time duration	Concentration of nDNA (ng/μL)	A260/280 (range)
1.	QIAamp DNA FFPE Tissue Kit	10	3 h	95–135	1.75–2.10
2.	Norgen DNA FFPE Tissue Kit	10	3 h	28–50	1.55–2.05
3.	PC method	10	15 h	50–98	1.65–2.23
4.	Mineral oil	10	15 h	21–63	1.50–2.30
5.	M/10 NaOH	10	45 min	12–25	2.08–2.40
6.	Microwave method	10	15 h	100–150	1.70–2.00

PC, phenol–chloroform; FFPE, formalin-fixed paraffin-embedded.

Bold values indicate highest values in respective headings.

The DNA yields obtained by the microwave method were significantly higher when compared with other methods, whereas quality was significantly higher when compared with mineral oil (Table 2).

Table 2.

Comparison of DNA concentration and quality of DNA by microwave method with other methods using unpaired student t test.

Comparison of DNA concentration of different methods with microwave method				Comparison of OD value of different methods with microwave method
Methods	Mean difference	t value	p value	Mean difference	t value	p value
QIAamp DNA FFPE Tissue Kit	16.3	12.041	0.032*	0.008	0.183	0.857
Norgen DNA FFPE Tissue Kit	86.7	2.323	<0.001*	0.059	1.107	0.283
PC method	53.5	13.947	<0.001*	0.004	0.054	0.958
Mineral oil	83.0	12.041	<0.001*	0.168	1.77	<0.001*
M/10 NaOH	106.7	19.041	<0.001*	0.378	7.794	0.104

PC, phenol–chloroform; FFPE, formalin-fixed paraffin-embedded.

*p value ≤ 0.05 is considered as statistically significant.

Prominent and sharp bands were seen in DNA extracted by the microwave method and in few cases with QIAamp DNA FFPE Tissue kit, whereas in rest of the methods shearing of DNA was noticed (Fig. 1).

Fig. 1.

Gel image of extracted total genomic DNA by different methods, lane 1–2: microwave method; lane 3–4: QIAamp DNA FFPE Tissue Kit; lane 5–6: Norgen DNA FFPE Tissue Kit; lane 7–8: PC method; lane 9–10: mineral oil; lane 11–12: M/10NaOH; lane 15: DNA ladder 100–3000 bp. PC, phenol–chloroform; FFPE, formalin-fixed paraffin-embedded.

A desired amplicon of 280 base pairs was visualized on 1.5% agarose gel. Based on PCR amplification performance, the samples that were extracted with the microwave method and two samples with the commercial kit showed positive results (Fig. 2).

Fig. 2.

Gel image of PCR product of β actin, Product size: 280bp, lane 1: 100-1500bp ladder, lane 2–3: microwave method; lane 4: QIAamp DNA FFPE Tissue Kit, lane 5: Norgen DNA FFPE Tissue Kit; lane 6: PC method; lane 7: mineral oil; lane 8: M/10 NaOH. PC, phenol–chloroform; FFPE, formalin-fixed paraffin-embedded.

Discussion

Formaldehyde is a reactive electrophilic species that fixes the tissue through cross-linkage. Depending upon the affinity of proteins for formaldehyde, three types of reactive groups can be formed, which are as follows (1) methylol groups, (2) Schiff bases, and (3) methylene bridges. The initial cross-linking that occurs in the first 24–48 h is a reversible process and includes methylol groups and Schiff bases. In a span of 30 days, irreversible methylene bonds are formed. These covalent bonds are difficult to break, resulting in fragmentation of DNA during the extraction process. Breaking these cross-linkages before DNA extraction is the important aspect for achieving good quality DNA.⁴

Hence, since ages, we are trying to extract DNA from FFPE tissue by various means. However, regardless of our effort, we still are struggling to find the solution. In the present study, the main focus was given on breaking the cross-linkage of protein caused by formalin fixation. As per our results, the extracted DNA by QIAamp FFPE tissue commercial kit was of good quality in two samples which is in accordance with other studies.⁵ While in remaining samples, shearing occurred, which can be attributed to many factors including very harsh treatment during fixation by formalin which results in inevitable DNA fragmentation. The commercially available kits are advantageous in some ways that they reduce time and effort. The disadvantage is that they are expensive and protocols cannot be altered easily.

The PC extraction method is an organic method that is being used in the isolation of DNA routinely. The use of PC method for FFPE tissues with minor changes has been suggested by a few authors.^{6, 7} In the present study, we found reduced concentration of DNA in all the samples processed by PC method. The disadvantage of this method is that it is very labor intensive and samples get easily contaminated. Similar results have been elaborated by Sengüven et al.⁵

Using mineral oil in FFPE tissue, 21–63 ng/μL of DNA was extracted during deparaffinization, which is quite low. However, quality of the extracted DNA was satisfactory. Extraction of DNA using mineral oil instead of xylene reduces the time and hands on labor⁸ and also provides good quality DNA. Unfortunately, mineral oil is suited only for DNA fragments <200 base pairs even in archived specimens as old as 27 years in genetic studies.⁹

Shi et al.² advocated the use of an alkali (pH 11.5–13) during DNA extraction. In the present study, fragmented and very low amount of DNA in the range of 12–25 ng/μL was obtained using this method. As per observation, using temperature of 100 °C leads to fragmentation or denaturation of the extracted nucleic acid.²

The extracted DNA by our microwave method yielded good quantity (100–150 ng/μL) and quality of DNA (A260:A280 ratio = 1.70–2.00) in comparison with other methods. It is a known fact that the cross-linking of proteins to nucleic acid is thought to be heat reversible.^{10, 11, 12} To retrieve the epitopes for immunohistochemistry, microwave has been used to break the cross-linkage to achieve best results. The same concept has been used successfully in our study as well to achieve good quality and quantity of DNA.

Also, heating the FFPE tissues at high temperature preceded by treatment with various alcohols, detergents at high pressure, and buffer pH combined with heating results in breaking of cross-linkages over epitopes.^{13, 14, 15, 16, 17}

To break the cross-linkages of proteins, various time-dependent and power-related protocols have been used. Boiling the buffer at their respective boiling point at the maximum power settings were used to achieve better results in various studies.^{17, 18, 19, 20} In the present study, we have heated the samples in phosphate buffer solution at 400 MW initially for 2 min to create instability in the cross linkages followed by 850 MW (maximum) for 2 min to breaks these bonds without deteriorating the nucleic acid. Govender et al.²⁰ used different retrieval solutions, that is, 0.01 M citric acid, pH 6; 1 mM EDTA, pH 8; 10 mM Tris Base; 1 mM EDTA, pH 9; 10 mM Tris Base; 1 mM EDTA with 0.5% between pH 9 to get the better yield of DNA from FFPE successfully.

Based on literature and experience in molecular pathology, a few pertinent points should be considered to achieve DNA of good quality²¹: (1) low voltage recommended at 4–10 V/cm if size of DNA is between 1 and 12 kb; (2) sharp bands are seen in the agarose gel kept in buffer for longer period (e.g. overnight) or after prestorage in the refrigerator (3 °C); (3) incorporate DNAase inhibitor; (4) do not vortex or shake DNA solution containing the genomic DNA; (5) susceptibility to shearing is increased in long genomic DNA.

To conclude, extraction of nucleic acid from FFPE tissues is very tedious as formalin fixation results in strong cross-linkages of proteins with DNA. Our microwave method has shown extremely promising results for extraction of good quality DNA from FFPE tissues without degradation which is of superior quality and gives better downstream results when compared with other extraction methods. Further studies on a large sample size could probably authenticate the results obtained in the present study.

Conflicts of interest

All authors have none to declare.