Comparison of Sampling Procedures for the Molecular Diagnosis of Leishmaniases

ABSTRACT.

The present work evaluates sampling protocols, storage procedures, and DNA purification methods for Leishmania spp. detection and quantification in different biological samples. The efficiency of three preservation solutions, a phosphate buffer solution, an ethylenediaminetetraacetic acid (EDTA) buffer solution, and 70% ethanol, was compared in combination with three DNA extraction protocols: a commercial silica column kit, salting-out protein precipitation, and organic extraction with phenol-chloroform. Tissue samples from BALB/c mice experimentally infected with Leishmania (Leishmania) amazonensis, Leishmania (Viannia) braziliensis, or Leishmania (Leishmania) infantum were stored in the three preservation solutions and subsequently subjected to the three different DNA extraction methods. The extracted DNA was then used in real-time polymerase chain reaction (PCR) assays for the detection and quantification of parasite ribosomal small subunit DNA targets as well as mammalian glyceraldehyde-3-phosphate dehydrogenase (gapdh) targets. The results of the optimized protocols showed that the DNA extraction method did not influence test quality, but DNA from samples preserved with the EDTA buffer solution produced higher amounts of target amplicons. Based on these results, we concluded that samples from suspected cases of leishmaniasis for submission to molecular diagnostic procedures should be preferentially preserved in EDTA, followed by any one of the DNA purification methods evaluated.

INTRODUCTION

The leishmaniases are a public health problem in several developing countries. Approximately 0.7 to 1 million new cases are reported annually from nearly 100 endemic countries.¹ There are at least 54 Leishmania species described, of which more than 20 are involved in human infections with a wide range of clinical spectra.² Clearly, accurate diagnostic procedures are relevant in understanding epidemiological profiles and in the administration of optimized therapeutic protocols. Additionally, diagnostic tests are critical in conducting active surveillance and identifying risk factors under One Health approaches.³ Moreover, correct identification of parasite species in samples collected systematically from previous studies is useful for understanding the epidemiology of the leishmaniases.⁴ Among molecular techniques currently used for the diagnosis of leishmaniases, polymerase chain reaction (PCR)–based assays, sometimes combined with sequencing technologies, are the main approach in diagnosis research.⁵ The molecular techniques have increased the sensitivity and specificity of leishmaniasis diagnosis compared with those of conventional approaches to parasite culture and microscopy.⁶

The way in which clinical samples are stored before nucleic acid isolation influences the quality of the purified DNA. It has already been shown that deoxyribonucleases play an important role in the degradation of DNA extracted from tissues and that the presence of ethylenediaminetetraacetic acid (EDTA) in solutions to which these tissues are exposed inhibits the action of these nucleases.⁷ Another simple strategy used in tissue preservation for further DNA purification is ethanol fixation, which has been described as being capable of preserving mammalian tissues for more than 6 years at room temperature under satisfactory conditions for subsequent DNA extraction and DNA amplification through PCR assays.⁸

Purification of nucleic acids from cells or tissues is an essential step in diagnostic protocols through molecular biology techniques. The purity and integrity of the extracted DNA are factors that strongly influence the results obtained in subsequent molecular techniques, such as polymerase chain reaction (PCR), as target integrity is crucial for the pair of primers to complete extension and chain replication. To guarantee the quality of these parameters, some steps are desirable in the DNA extraction process: 1) obtaining sufficient mass to perform the tests; 2) ensuring the integrity of the molecules—mainly when the DNA will be used for the amplification of large targets; and 3) maintaining the purity of the preparation. The latter not only keeps the sample free of contaminants that can hinder the PCR, such as protein residues, free nucleotides, and RNA, but also avoids the presence of tissue contaminants, such as hemoglobin, and residual contaminants, such as organic solvents or other Taq DNA polymerase inhibitors.⁹

Formaldehyde is a popular preservative that is traditionally used in pathology laboratories. However, its use for molecular studies has met with variable degrees of failure, and overall, it is not recommended. A detailed study of the damage to DNA due to different fixatives showed that both the length of time that tissues were stored in formalin solutions before DNA extraction and the buffering agent used affected DNA quality.¹⁰ The major problem is that DNA obtained from samples preserved in formaldehyde becomes cross-linked to surrounding tissue components and cannot be extracted by standard extraction methods. It has also been shown that sample exposure to formaldehyde impairs the quality of DNA in a time-dependent manner and that recovery of DNA from samples fixed in ethanol-based fixatives is higher than that from formaldehyde-fixed samples; therefore, ethanol can be a useful alternative for molecular profiling studies.^11,12

The first description of a DNA extraction protocol dates from 1869, when the Swiss physician Johann Friedrich Miescher isolated nucleic acids from leukocytes using concepts still used in current protocols.¹³ DNA extraction protocols are based on strategies that use physicochemical properties to purify DNA from complex mixtures that contain RNA, proteins, lipids, and several other cellular components.

When DNA is fragmented, the efficiency of the PCR decreases, especially if the targeted PCR product is large, as the probability of a break inside the template is directly proportional to its length. Therefore, breaks make the template fall apart, preventing the formation of an amplicon and reducing the amount of PCR product. It was already demonstrated that the accuracy of DNA quantification by quantitative real-time PCR (qPCR) can be affected by DNA fragmentation in a study where experimentally fragmented DNA was systematically used as a template, even when ideal spectrophotometer parameters of purity were maintained.¹⁴

To accommodate the needs of template production and compensate for the difficulties mentioned above with different fixatives, we compared the number of copies of the Leishmania ribosomal small subunit (SSU rDNA) sequence in biological samples from in vivo infections using mice as a host model. After collection, tissue samples were stored in a phosphate buffer solution, an EDTA buffer solution, or 70% aqueous ethanol. The latter is used routinely in the preservation of animal tissues. Then, the samples were subjected to three different DNA purification protocols, a commercial silica column kit, salting-out protein precipitation, and organic extraction with phenol-chloroform, for further amplification of specific DNA targets through qPCR assays.

The present work shows a systematic evaluation of processing protocols for the PCR-based diagnosis of leishmaniasis in biological samples. Based on the results obtained, we suggest that samples for molecular diagnosis based on PCR methods should be preferentially preserved in EDTA buffer followed by any DNA purification.

MATERIALS AND METHODS

Cells and animals.

Promastigote forms of Leishmania (Leishmania) amazonensis (MHOM/BR/1973/M2269), Leishmania (Viannia) braziliensis (MHOM/BR/1975/M2903), and Leishmania (Leishmania) infantum chagasi (MCER/BR/1981/M6445) were grown in medium M199 (Gibco, Paisley, United Kingdom) with Earle’s salts supplemented with 40 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) (pH, 7.4); 0.1 mM adenine; hemin (5 mg/L); penicillin (4550 U/mL); streptomycin (0.05 mg/mL); and 10% fetal bovine serum (Gibco).

Three-month-old male BALB/c mice were used in the experimental infections, and all procedures involving their use had the approval of the Ethical Committee for the Use of Animals of the Biomedical Sciences Institute of the University of São Paulo (CEUA-ICB-USP, protocol #145, October 20, 2011).

Experimental infections.

Test tissue samples were obtained from in vivo infections with L. (L.) amazonensis, L. (V.) braziliensis, or L. (L.) infantum in 18 mice per parasite species. The animals were inoculated subcutaneously in the posterior left footpad with 1.0 × 10⁶ stationary promastigotes of each Leishmania species. After 6 weeks, the animals were killed in a CO₂ chamber; approximately 2 mm³ of different tissues—skin from the inoculation site of all animals or blood, liver, spleen, and bone marrow from the mice inoculated with L. (L.) infantum—was collected and kept in different storage solutions (six mice per group) and processed by the different DNA extraction methods (six mice per group).

Sampling and treatment of infected tissues.

Tissue fragments of approximately 2 mm³ of each studied sample were submerged in 1 mL of one of the three different storage solutions: phosphate-buffered saline (PBS): 150 mM NaCl and 50 mM Na₂HPO₄, pH 7.4; EDTA buffered solution (NET): 150 mM NaCl, 50 mM EDTA, and 100 mM Tris, pH 7.4; or aqueous solution of 70% ethanol (EtOH70). The samples were then stored at 4°C for 1 week. Each sample was then centrifuged at 3,000 × g, and the liquid was removed for later DNA extraction. Additionally, 100 µL of blood from 10 mice inoculated with L. (L.) infantum was collected and immediately processed.

DNA extraction from infected tissues.

The treated tissues were washed with PBS, macerated with sterile pestles, lysed in 630 µL of ATL buffer (Qiagen, Hilden, Germany) with 1.4 mg of proteinase K at 56°C for 3 hours, and agitated in a vortex mixer once every hour. After digestion, the samples were aliquoted in three equal amounts that were submitted to three different DNA extraction processes.

DNA extraction with a commercial kit (DNeasy).

The proteinase-digested samples (200 µL) were processed with a DNeasy Blood & Tissue Kit (Qiagen) according to the manufacturer’s instructions. The process briefly consisted of adding ethanol to the lysed samples and binding of precipitated DNA to silica columns, which were washed with a nonalcoholic solution for DNA elution in 200 µL of TE (1 mM EDTA and 10 mM Tris, pH 7.4). Blood samples were processed according to the manufacturer’s manual.

Salting-out deproteination extraction (SODE) (adapted from Miller and others¹⁵).

Eighty microliters of saturated NaCl solution was added under vortex agitation to the proteinase-digested samples (200 μL). The samples were then centrifuged at 3,000 × g for 15 minutes at room temperature, and 560 µL of cold absolute ethanol was added to the supernatant, followed by centrifugation at 20,000 × g for 10 minutes at 4°C. The precipitated DNA was washed with 1 mL of 70% ethanol, dried, and resuspended in 200 µL of TE.

Organic DNA extraction (ϕ:CHCl₃) (adapted from Uliana and others¹⁶).

The proteinase-digested samples (200 μL) were submitted to a three-step organic extraction: The first was the addition of 1 volume of buffered phenol, the second was the addition of 1 volume of phenol:chloroform:isoamyl alcohol (25:24:1), and the third was the addition of 1 volume of chloroform:isoamyl alcohol (24:1), with each step being followed by centrifugation at 20,000 × g for 5 minutes at 4°C. The nucleic acid present in the aqueous phase was precipitated by adding 0.1 volume of 3 M sodium acetate, pH 7.0, and 2 volumes of cold absolute ethanol, followed by centrifugation at 20,000 × g for 10 minutes at 4°C. The precipitated DNA was washed with 1 mL of 70% ethanol, dried, and resuspended in 200 µL of TE.

DNA concentrations and purity were determined by spectrophotometry on GeneQuant Pro equipment (Amersham Biosciences, Cambridge, United Kingdom). To assess the influence of sample storage solutions (PBS, NET, or EtOH70) as well as commercial (DNeasy) or in-house (SODE or ϕ:CHCl₃) DNA extraction protocols on the total amount of DNA obtained, DNA concentrations and purity of the samples were determined by spectrophotometric measurements and compared. For this analysis, only DNA from samples from skin was considered. All the tissue samples had approximately the same volume (around 2 mm³).

Detection and quantification of DNA.

Quantitative real-time PCR.

Quantitative reactions were performed on a PikoReal 96 Real-Time PCR System (Thermo Fisher Scientific, Waltham, MA) with a Maxima SYBR Green/ROX PCR Master Mix Kit (Thermo Fisher Scientific) in a final volume of 20 µL, containing 0.3 µM of each oligonucleotide and 50 ng of total DNA as the template. The thermocycling protocol consisted of an initial denaturation step at 94°C for 5 minutes and 40 cycles including steps of DNA denaturation at 94°C for 30 seconds and annealing/extension at 60°C for 30 seconds. Fluorescence signals were acquired at the end of the extension steps. A melting curve determination was programmed at the end of each qPCR assay. The copy number of the DNA targets in the experimental samples was calculated by standard curve interpolation constructed with plasmids containing the specific targets. Oligonucleotides based on the coding sequence of the Leishmania small subunit ribosomal RNA gene (SSU rDNA) (S12: 5′-GGTTGATTCCGTCAACGGAC-3′ and S19: 5′-GAATTCGACCGAATGCGGCC-3′) or mammalian glyceraldehyde-3-phosphate dehydrogenase (gapdh) (gapdF2: 5′-GCCCAGAACATCATCCCTG-3′ and gapdR2: 5′-GGAACACGGAAGGCCATG-3′) gene were used as primers.

Statistics.

Samples were tested in duplicate in all quantification assays. Statistical analysis was achieved using a Kruskal-Wallis nonparametric test followed by Dunn’s post hoc test for multiple comparisons among the groups. The results are presented as median differences, and statistical significance was set to P < 0.05. All statistics were calculated, and the graphs were made using GraphPad Prism 8 (San Diego, CA).

RESULTS

DNA concentration.

To evaluate the yield under the different conditions to which the biological samples were subjected, DNA concentrations and purity were estimated through spectrophotometry. Higher concentrations of DNA were obtained from tissues infected with L. (L.) amazonensis than from those infected with L. (V.) braziliensis or L. (L.) infantum, regardless of the storage solution or DNA extraction method (Figure 1). For these samples, irrespective of the DNA extraction chosen (DNeasy, SODE, or ϕ:CHCl₃), the results that showed the highest concentrations of total DNA were those from samples that were treated with EtOH70, followed by those from samples that were treated with PBS or NET (Figure 1A). The DNA concentrations of samples obtained from L. (V.) braziliensis and L. (L.) infantum infections did not differ significantly, regardless of the solution used to maintain the tissue prior to DNA extraction (Figure 1A). Among the most concentrated samples (i.e., those derived from the experimental L. (L.) amazonensis infections), the methodology used in the extraction of DNA did not influence the DNA yield (Figure 1B), irrespective of the storage solution chosen (PBS, NET, or EtOH70). On the other hand, in the less concentrated samples (i.e., those derived from L. (V.) braziliensis or L. (L.) infantum infections), the commercial kit based on silica columns (DNeasy) yielded significantly less total DNA than the “in-house” methods (SODE or ϕ:CHCl₃) (Figure 1B).

Figure 1.

Quantification of total DNA in samples from experimental infections. DNA quantification was performed through spectrophotometry using total DNA from skin samples of mice experimentally infected with Leishmania (Leishmania) amazonensis, Leishmania (Viannia) braziliensis, or Leishmania (Leishmania) infantum, maintained in different storage solutions—PBS = phosphate-buffered saline; NET = ethylenediaminetetraacetic acid (EDTA) buffered solution; or EtOH70 = aqueous solution of 70% ethanol—and processed according to different DNA extraction methods—DNeasy = silica columns–based commercial kit; SODE = salting-out deproteination extraction; or ϕ:CHCl₃ = organic extraction with phenol and chloroform. (A) DNA samples from tissues subjected to the different DNA extraction methods, grouped according to the storage solution used. (B) DNA samples from tissues preserved in the different storage solutions, grouped according to the DNA extraction method used. Six samples per experimental group were tested in duplicate. Horizontal bars indicate medians. Statistical analysis was achieved using a Kruskal-Wallis nonparametric test followed by Dunn’s post hoc test for multiple comparisons among groups.

The purity of the DNA samples was determined on the basis of spectrophotometric measures, considering A260/A280 and A260/A230 ratios as parameters. According to the parameters indicated by the spectrophotometer manufacturer, A260/A280 and A260/A230 ratios around 1.8 and 2.0–2.2, respectively, are generally accepted as “pure” for DNA preparations. Despite being the method with the lowest DNA yield (Figure 1B), the commercial DNeasy extraction kit produced the highest purity DNA samples, as seen from the spectrophotometric measures A260/A280 and A260/A230 Supplemental Table 1). The in-house protocols for DNA purification produced samples with spectrophotometric parameters different from those expected for pure DNA preparations Supplemental Table 1).

The results obtained from the analyses of DNA yielding—as well as those obtained from qPCR assays and described in the subsections below—are also demonstrated in heat maps and bar diagrams that show a combined panel where all variables are compared: storage condition versus DNA extraction method, with medians from technical duplicates used for plotting the graphs (Supplemental Figure 1). Despite the impossibility of a proper statistical analysis, these panels reveal patterns that corroborate all the results already presented in which statistical tools were applied (Figures 1 and 2).

Figure 2.

Specific quantification of host or parasite DNA and parasite load in samples from experimental infections. Specific DNA quantification was performed through qPCR assays using primers based on the mammalian glyceraldehyde-3-phosphate dehydrogenase (gapdh) gene or parasite ribosomal small subunit DNA (SSU rDNA) gene. (A–D) Total DNA from skin samples of experimentally infected mice was used as a template to determine the absolute number of the specific targets. (E–F) Parasite load was calculated by the ratio between the absolute numbers of parasite- and host-specific targets. PBS = phosphate-buffered saline; NET = ethylenediaminetetraacetic acid (EDTA) buffered solution; EtOH70 = aqueous solution of 70% ethanol; DNeasy = silica columns–based commercial kit; SODE = salting-out deproteination extraction; ϕ:CHCl₃ = organic extraction with phenol and chloroform. (A, C, and E) DNA samples from tissues subjected to the different DNA extraction methods, grouped according to the storage solution used. (B, D, and F) DNA samples from tissues preserved in the different storage solutions, grouped according to the DNA extraction method used. Six samples per experimental group were tested in duplicate. Horizontal bars indicate medians. Statistical analysis was achieved using a Kruskal-Wallis nonparametric test followed by Dunn’s post hoc test for multiple comparisons among the groups. qPCR = quantitative real-time PCR.

Host DNA quantification.

Taking into account that in a lesion most of the cells belong to the host, we inferred that the quantification of the host DNA may reflect the quality of the total DNA obtained from the skin samples. To examine this possibility, quantification of specific mammalian targets was performed through qPCR with primers based on the sequence of the constitutive gene gapdh using equal amounts of total DNA (50 ng) as the template in all reactions.

In general, DNA samples obtained from L. (L.) amazonensis and L. (V.) braziliensis experimental infections had higher amounts of host DNA detected than those from L. (L.) infantum infections (Figure 2A and B).

Considering all skin samples, regardless of the parasite species involved in the infection or the DNA extraction protocol used, the samples with the highest amount of host DNA were those that had been previously treated with NET compared with those treated with PBS or EtOH70 (Figure 2A). This trend, although it was seen in all samples regardless of the parasite species involved in the infection, exhibited statistically significant differences among the samples from mice infected with L. (L.) amazonensis treated with NET or EtOH70; significant differences were also found among samples from mice infected with L. (V.) braziliensis, where EtOH70 was the statistically less effective treatment (Figure 2A).

No differences were observed in the samples from tissues infected with L. (L.) amazonensis or L. (L.) braziliensis when the three methodologies used for DNA extraction were evaluated. On the other hand, samples obtained from infections with L. (L.) infantum processed by the commercial DNeasy kit showed a greater amplification of the specific target than the samples processed by the SODE method (Figure 2B).

Parasite DNA quantification.

To evaluate the effect of the different storage conditions and different DNA extraction protocols on the detection of parasite DNA, we performed qPCR assays with primers based on SSU rDNA coding sequence using equal amounts of total DNA (50 ng) as the template in all reactions.

All samples from mice infected with L. (L.) amazonensis had larger amounts of parasite DNA—approximately 10,000 and 100,000 times more copies—than samples from mice infected with L. (V.) braziliensis and L. (L.) infantum, respectively (Figure 2C and D). Nevertheless, regarding these samples, the NET solution was the most effective condition for parasite DNA detection compared with PBS, followed by EtOH70. This trend was also observed in samples obtained from infections with L. (V.) braziliensis, although no significant differences were observed. Moreover, no significant differences were found in the samples from infections with L. (L.) infantum considering the different storage solutions (Figure 2C).

The DNA extraction methodology did not interfere with the effectiveness of the detection and specific quantification of parasite DNA, regardless of the species involved in the infection or the storage solution used (Figure 2D).

Normalized parasite load data.

To determine the parasite load, correcting the errors inherent in the parasite DNA quantification process, the raw qPCR data obtained for that target were normalized by the data from the host DNA quantification.

As expected, and similar to the parasite’s absolute DNA quantification, the normalized data indicated a higher parasite load in all samples from mice infected with L. (L.) amazonensis—approximately 10,000 and 100,000 times—than that in samples from mice infected with L. (V.) braziliensis or L. (L.) infantum, respectively (Figure 2E and F). For these samples, NET and EtOH70 were the most effective storage solutions for parasite load evaluation, whereas for tissue samples infected with L. (V.) braziliensis or L. (L.) infantum, we did not observe significant differences (Figure 2E).

The different extraction protocols did not result in statistically significant differences in the efficiency of parasite load determination (Figure 2F).

Yield, absolute quantification of DNA, and parasite load in different tissues from mice infected with L. (L.) infantum.

Because preservation solutions did not seem to influence the quantification of parasite targets in L. (L.) infantum infections and the DNA extraction protocol with silica columns proved to be efficient for these samples (Figure 2), we analyzed DNA quantification results for samples from the liver, spleen, bone marrow, skin, and blood of mice infected with L. (L.) infantum previously preserved in PBS, NET, or EtOH70 and subsequently processed with the DNeasy kit (Figure 3).

Figure 3.

DNA quantification in samples of different tissues from mice infected with Leishmania (Leishmania) infantum. Plotted dots represent the yield of total genomic DNA (A), absolute quantification of host and parasite DNA (B and C, respectively), or normalized parasite load (D) in six samples of liver, spleen, bone marrow, skin, and 10 blood samples from infected mice subjected to a silica columns–based commercial kit (DNeasy), regardless of the storage treatment adopted (PBS, NET, or etOH70). All samples were tested in duplicate. Horizontal bars indicate medians. Statistical analysis was achieved using a Kruskal-Wallis nonparametric test followed by Dunn’s post hoc test for multiple comparisons among groups. EtOH70 = aqueous solution of 70% ethanol; NET = ethylenediaminetetraacetic acid (EDTA) buffered solution; PBS = phosphate-buffered saline; SSU rDNA = small subunit ribosomal RNA gene.

Spectrophotometric measures showed higher yield in liver and spleen samples than in blood samples (Figure 3A). The comparison among any other tissues did not result in significant differences. On the other hand, the specific quantification of host DNA proved to be more efficient in spleen samples than in liver and blood (Figure 3B). There were no significant differences in the specific quantification of the parasite DNA and, consequently, in the parasite load when all tissues were compared (Figure 3C and D). However, blood samples were the ones where parasite detection was less frequent (only two out of 10 samples).

DISCUSSION

In the present study, the quantity of mammalian DNA in samples from mice experimentally infected with L. (L.) amazonensis, L. (V.) braziliensis, or L. (L.) infantum was assessed by real-time PCR amplification using mammalian-specific primers for the gapdh gene. The results showed that tissue samples stored in NET buffer, in which the EDTA concentration was 50 mM, produced DNA with significantly higher quality for amplification than samples stored in other tissue storage solutions. All three storage solutions tested are readily available, and the advantage of NET is that because it is an aqueous buffer, there are no regulatory restrictions for any sample transportation methods.

The determination of the parasite load by the detection of Leishmania SSU rDNA in these samples indicated that the tissues maintained in NET or in EtOH70 were the ones that produced the greatest number of detectable amplicons. Although ethanol fixation also resulted in satisfactory DNA samples, in terms of total DNA, this fixation showed differences in parasite detection when the infectious agent was L. (L.) amazonensis or L. (V.) braziliensis (Figures 1 and 2). The parasitic burden in lesions caused by these species reflects the parasite’s biology. Many species of the subgenus L. (Leishmania) develop rapidly, producing extensive lesions, whereas species of the subgenus L. (Viannia) produce lesions that develop relatively slowly.¹⁷ This difference is reflected in the samples from L. (L.) amazonensis and L. (V.) braziliensis experimental infections, which contained different quantities of target DNA.

Despite producing samples with low amounts of total DNA in terms of concentration (Figure 1A), preservation with NET buffer was the method that produced samples with the highest amount of intact parasite target DNA (Figure 2C). These results, along with those observed in the mammalian host DNA detection tests (Figure 2A), support the choice of NET buffer as a simple and efficient option for the preservation of biological samples destined for Leishmania DNA detection tests through PCR.

A previous study evaluated the best storage method for preserving the DNA of insect intestinal bacteria, where five different methods were tested, ranging from freezing to dimethyl sulfoxide, including 95% ethanol, using a PCR targeting an ∼300-bp portion (V4 region) of the bacterial 16S ribosomal RNA gene without quantification of the target. The conclusion was that there was a significant variation between the fixation methods within a given insect species.¹⁸ Another study evaluated metatranscriptomic and metagenomic profiles of human gut flora in preserved samples by immediate freezing or fixation in ethanol and another commercial storage reagent. The findings showed that the method of preservation did not significantly affect species identification, but it significantly affected transcript detection.¹⁹ These studies show that the fixation method can affect the results of subsequent molecular tests.

In the present work, the samples obtained from three distinct DNA extraction methods were evaluated by amplifying the gapdh gene of mice and the SSU rDNA gene of Leishmania. Despite showing different yields in DNA purification (Figure 1B), the samples generated DNA templates that were similarly amplified, without showing significant differences in the effectiveness of detection and quantification of parasite DNA or, consequently, in the determination of parasite burden (Figure 2D and F). Thus, for samples to be used for diagnosis, the choice of the DNA extraction method can be made according to the advantages that each protocol offers. The commercial kit used in this work was the option that yielded the lowest amount of purified DNA but sufficient quantity and quality for use as template DNA to perform the tests. The main advantage of this choice is the practicality and the optimization of the processing time. On the other hand, the use of manufactured kits is financially more costly than the other methodologies that we used. Organic extraction with phenol-chloroform or protein precipitation by salting-out methods produced larger amounts of purified DNA than the commercial kit, with similar efficiency regarding the production of amplifiable DNA in PCR tests despite producing DNA with lower purity (Figure 2, Supplemental Table 1). DNA isolation after protein precipitation by salting-out is described as advantageous over the other two methods because it is inexpensive and nontoxic, as it does not use organic solvents and can be carried out in a short time²⁰; in a routine situation, it may be a good choice for DNA extraction because of these characteristics.

Although the liver samples yielded more DNA than those obtained from other tissues, the spleen was the tissue that generated the most amplifiable DNA (Figure 3A and B). When parasite DNA was the target of amplification, a very low number of molecules could be detected, and there seemed to be no difference among the tissues used for examination, except for blood, which was the tissue with the lowest rates of detection of the parasite (Figure 3C). These results, obtained from an intended model of visceral leishmaniasis, may reflect a low parasitic burden in these mice, which were inoculated intradermally and which may not have had time to develop a more chronic form of infection.

Studies that investigate and standardize the most efficient DNA preservation methods are essential for the successful detection of small numbers of parasites. In conclusion, our results indicate that NET buffer is an effective option for the preservation and maintenance of samples destined for DNA extraction and purification processes for use in PCR-based diagnosis of leishmaniasis.

The three tested DNA extraction and purification protocols, a commercial silica column kit, protein precipitation by salting-out, and organic extraction with phenol-chloroform, produced similar qualities of specific target amplification through PCR. They can therefore be chosen according to the advantages they offer.

Note: Supplemental figure and table appear at www.ajtmh.org.