Evaluation of cost-effective total nucleic acids extraction protocols for cultured Mycobacterium tuberculosis; a comparison by PCR amplification of genes associated with drug resistance

Justification

The application of most molecular biology techniques is largely dependent on the quantity and quality of extracted nucleic acids [1, 2] and which are, in turn, greatly influenced by the efficiency of cell lysis, DNA/RNA recovery and residual amounts of some extraction reagents [3–5].

Mycobacterium cell wall is made of an impermeable complex structure that makes the lyses of the cell difficult [3, 6, 7]. As a result, most of the simple and commonly used nucleic acid isolation procedures result in poor quality and low yield of nucleic acids, and are also affected by the type of specimen used [7–9]. Several complicated protocols and commercial DNA and RNA extraction kits have been developed that are mostly efficient on cultured Mycobacterium, but these are expensive, and have elaborate procedures [10, 11].

PCR amplifications, in our laboratory, of some of the resistance-associated genes of the Mycobacterium tuberculosis- complex genome have not been appropriate for further analysis. A number of PCR protocol modifications resulted in marginal improvement of the amplification of these genes. Therefore, we concede that the evaluation and modifications of nucleic acid extraction protocols that are less expensive, less sophisticated, and less prone to contamination would be useful in resource-constrained setting as in Ghana. We therefore present a comparison of the results of seven nucleic acids extraction protocols, and aim at determining a cost-effective and less time consuming extraction protocol for cultured Mycobacterium tuberculosis by PCR of drug resistance genes.

Culture and Cell Suspension Preparation

Colonies, from five sputum samples cultured on (Lowenstein Jensen) medium slants for seven weeks, were harvested to prepare a two loopfuls of colony per mL suspension in BBL™ Middlebrook ADC Enrichment broth.

For each DNA extraction protocol, an aliquot (200 μL) of each cell suspension was transferred into a sterile 1.5 mL screw-capped tube and centrifuged for 3 min at 10,000 × g. The supernatant was carefully removed and the pellet washed twice with sterile distilled water, centrifuging each time as above. The five resultant pellets were then subjected to total nucleic acid extraction.

Nucleic Acid Extraction

PCR amplification of Mycobacterium tuberculosis DNA

Five microlitres (5.0 μL) of each extract was used with Qiagen PCR kit for the gene amplification. Commercially purified DNA of Mycobacterium tuberculosis H₃₇RV standard strain and PCR water were used as positive and negative controls respectively. The following primers were used;

Electrophoresis

Amplicons were resolved on an ethidium bromide stained 2% agarose gel.

In avoiding contaminations, culture processing was performed in a separate biosafety cabinet located in separate room; nucleic acids extracts and PCR mixes were prepared in PCR work stations in separate rooms; and disposable plugged micropipette tips were used.

Relative Quantification of Nucleic Acids

Extracts obtained by each protocol was incubated with 2 volumes of absolute ethanol on ice for at least 1 hour and centrifuged at 12,000 × g for 10 min. Supernatant were decanted and pellets washed with 70% ethanol. The centrifugation was repeated; the pellets were air-dried and resuspended in 50 μL of warm 1× TE buffer.

The yield of total nucleic acid for each protocol was determined by measuring and calculating the average absorbance at 260 nm of the five extracts. Also, absorbance at 280 nm was measured and the ratio of UV absorbance 260/280 nm calculated.

Growth from Culture

Rough, eugonic buff colonies were first observed for all the five cultured sputum samples, three weeks after incubation, and stained acid-fast. At harvest, growth covered at least 50% of the media.

Nucleic Acids Quantity

Protocol 4 recorded a four fold increase in the yield of nucleic acids relative to that obtained with protocol 1 (table 1). This can be attributed to the fact that cells of Mycobacterium were better dispersed in the 1× TE buffer as used in protocol 4 [3]. Therefore, a comparison of the yield obtained for protocol 4 with those for protocols 2 and 3 (table 1) reveals that the inclusion of Tris-EDTA led to a more efficient lysis compared to the use of SDS alone at 1% (protocol 2), as well as freezing prior to heating (protocol 3). As presented in table 1, the yield recorded with protocol 5 (which involved the use of Tris-EDTA, SDS and heat) was not significantly different from that recorded by protocol 4 (which involved the use 1× Tris-EDTA and heat). This indicates that the inclusion of SDS in protocol 5 did little to improve the efficiency of lysis when it was used with Tris-EDTA. Coupled with the earlier finding that the efficiency of SDS (in protocol 2) could be achieved by freezing and heating (protocol 3), and also by heating in TE (protocol 4), the use of SDS at 1% could be conveniently avoided on the bases of extra cost [5].

Using TritonX-100 with Tris-EDTA (Protocol 6) clearly improved the efficiency of lysis; since the yield of nucleic acids was eight times that obtained with protocol 1 and four times that obtained with protocol 4. TritonX-100 is known to prevent the aggregation of Mycobacteria in suspension, thereby, enhancing the effect of heating on the cell wall structure [3]. The use of Tween-20 in addition to TE with heating (protocol 7) recorded a yield about half that obtained with protocol 4, (table 1).

Amplification by PCR

The difficulties we have had with the amplification of the KatG and rpoB genes using extracts obtained by protocol 1 were again observed in this study. This inconsistent amplification is likely to be a result of the low yield and/or poor quality of DNA, although the rpsL gene amplified for this extract (figure 1). Poor quality DNA is most likely the cause of this observation since the amount of nucleic acid did not correlate with the PCR amplification and also that the ratio of UV absorbance at260/280 nm was 0.2 (table 1), indicating the extract contained a relatively high amount of protein, which could inhibit PCR [4]. Extracts obtained with protocol 4 amplified for all three genes (table 2, rpoB, KatG and rrs), showing as bright bands in figure 2. Clearly, the higher nucleic acids yield and ratio of UV absorbance obtained by protocol 4 relative to protocol 1 influenced the successful amplifications since PCR is known to be influenced by the amount of nucleic acids and proteins amongst other factors [4, 5]. Interestingly, extracts obtained by protocol 3 amplified for two of the three genes (rpoB and rrs; table 2 and figure 3), and recorded yields and ratio of UV absorbance higher than those obtained with protocol 1, but lower than those obtained with protocol 4 (table 1). The non-amplification observed for extracts obtained with protocols 2 and 5 (figure 1b and 1c) is likely to be due to residual SDS (a known inhibitor of PCR) in the extracts; since their nucleic acids yields and UV absorbance ratios were higher than for other extracts that amplified by PCR (table 1). The use of SDS for the extraction of Mycobacterium DNA would require further purification steps in other to fully take advantage of the lysis efficiency of SDS. PCR amplification of DNA obtained with protocols 6 and 7 were not consistent and reproducible. This observation remains unclear, since the yields were quite high.

Genes amplified	Protocol 1	Protocol 2	Protocol 3	Protocol 4	Protocol 5	Protocol 6	Protocol 7
rpo B	X	X	√	√	X	-	-
Kat G	X	X	X	√	X	-	-
rrs	√	X	√	√	X	-	-
Treatments
Temperature	heating	heating	freezing and heating	heating	heating	heating	heating
Chemical		SDS		Tris/EDTA	Tris/EDTA/SDS	Tris/EDTA/TritonX-100	Tris/EDTA/Tween20

√ ⇒ expected amplification; X ⇒ no amplification; - ⇒ inconsistent amplification

We strongly suggest the use of 1× TE buffer in addition to temperature treatments for improved lysis of M. tuberculosis form culture. A combination of freezing and heating in Tris-EDTA (protocols 3 and 4) is strongly recommended for the preparation of M. tuberculosis nucleic acids useful for PCR; since in addition to its effectiveness, the reagents used in the two protocols are easily available.