Optimization of the Cetyltrimethylammonium bromide (CTAB) DNA extraction protocol using forest elephant dung samples

Among non-invasive biological samples, feces offer an important source of DNA and can easily be collected. However, working with fecal DNA from highly vegetarians species such as elephant is more challenging because plant secondary compounds have an inhibitory effect on PCR reactions. Working with forest elephant dung samples, we tested and adapted a protocol of DNA extraction developed on plants based on the Cetyltrimethylammonium bromide (CTAB) protocol. The protocol is relatively simple and yields a high DNA concentration. It is five-time less expensive compared to the methods of Benbouza et al. The extracted DNA is of good quality and easily amplified by PCR. The high-amplification percentage of mitochondrial genes in fecal DNA and subsequent sequencing of PCR products indicate that the proposed optimized method is reliable for molecular analysis of forest elephant dung samples.• Our optimized CTAB protocol has been adjusted by the addition of Sodium Dodecyl Sulfate (SDS) and proteinase K during the lysis phase. The combined effect of these reagents was capable of lysing cell walls and removing proteins efficiently.• Moreover, the prolonged time of incubation (overnight incubation at room temperature followed by 3 hours of incubation in a water bath) enhanced the increase of DNA yield but make the optimized protocol more time-consuming.


a b s t r a c t
Among non-invasive biological samples, feces offer an important source of DNA and can easily be collected. However, working with fecal DNA from highly vegetarians species such as elephant is more challenging because plant secondary compounds have an inhibitory effect on PCR reactions.
Working with forest elephant dung samples, we tested and adapted a protocol of DNA extraction developed on plants based on the Cetyltrimethylammonium bromide (CTAB) protocol.
The protocol is relatively simple and yields a high DNA concentration. It is five-time less expensive compared to the methods of Benbouza et al. The extracted DNA is of good quality and easily amplified by PCR. The high-amplification percentage of mitochondrial genes in fecal DNA and subsequent sequencing of PCR products indicate that the proposed optimized method is reliable for molecular analysis of forest elephant dung samples.
• Our optimized CTAB protocol has been adjusted by the addition of Sodium Dodecyl Sulfate (SDS) and proteinase K during the lysis phase. The combined effect of these reagents was capable of lysing cell walls and removing proteins efficiently. • Moreover, the prolonged time of incubation (overnight incubation at room temperature followed by 3 hours of incubation in a water bath) enhanced the increase of DNA yield but make the optimized protocol more time-consuming.

Background
Noninvasive techniques have been largely used in ecology for molecular studies because they present the advantage to not disturb or arm the studied animals [1] . Among non-invasive biological samples, feces offer an important source of DNA and can easily be collected.
Several methods have been proposed to extract DNA from feces, such as the guanidinium thiocyanate silica method [2][3][4] , the aqueous two-phase system method [5] , the phenol-chloroform method [6] , the Chelex 100 method [ 7 , 8 ], and the washing technique [9] . The application of these methods is however limited, especially for herbivorous species. Several limitations are presented by different methods actually used for DNA isolation. These include the presence of inhibitors, the quality of feces, the amount of feces used, and the preservative fluids [9] . Commercial kits offer efficient and convenient alternatives but are comparatively expensive, and sometimes fail in samples from herbivorous animals. Because, any single DNA extraction technique is efficient for all species, various techniques targeting different species have been proposed [10] .
Working with fecal DNA from highly vegetarians species such as elephant is more challenging. Indeed, plant secondary compounds have an inhibitory effect on PCR reactions. PCR success then depends on the maximization of DNA concentration, while reducing secondary compounds derived from plants in the diet [ 11 , 12 ]. Such a low amplification success reduces considerably the suitability of fecal samples for genetic studies in herbivores. Therefore, it is important to optimize the genetic techniques to obtain an adequate amplification success rate. CTAB method has been widely applied in molecular genetics of plants [13] . That detergent-based method has already been tested in some studies based on non-invasive samples [ 14 , 15 ]. Several other variations have been developed with the aim of adapting the method to a large number of organisms [ 16 , 17 ].
An alternative combining at the same time a low cost, a saving of time, a guarantee of good quality DNA extraction, and the use of non-harmful reagents not requiring special handling devices must therefore be sought.
Working on forest elephant, we tested and adapted a protocol developed by Vroh Bi et al. [18] on plants based on the CTAB (cetyltrimethylammonium bromide) protocol of Murray & Thompson [19] described by Chandellier [20] . CTAB is a frequently used surfactant in DNA extraction and several modifications of that protocol originally published by Doyle and Doyle [21] have been used.
To optimize the quantity and quality of DNA extracted from forest elephant dung samples, known for its high essential oil and polyphenolic content [22][23][24] , different parameters including optimizing tissue lysis phase, DNA precipitation, duration of incubation time, duration of DNA drying and temperature of DNA storage have been tested to maximize the amount and quality of DNA retrieved.
The standard extraction protocol is that of Doyle et al. [21] based on the cationic detergent Cetyltrimethylammonium bromide (CTAB) previously used by Murray and Thompson [16] and Benbouza et al. [25] . These authors used that protocol to respectively extract DNA from leaves of Table 1 Comparison of the main steps of the DNA extraction methods using the CTAB method proposed by Benbouza et al. [25] and our optimized protocol.
Store DNA extracts at 4 °C.

Biological sample: elephant dung samples
From 07 to 30 October 2019, dung samples of forest elephant were collected at stage 1 of decomposition (dung pile intact, very fresh, and moist, with odour) in the Bossématié, Dassioko and Port-Gauthier Forest Reserves respectively located in the rainforest area in Côte d'Ivoire. Each fecal sample was preserved in 50 ml cryotube containing 70 % ethanol. All samples were transported to the laboratory within 15-20 days and stored at -80 °C till extraction.

Determination of the quantity and purity of the extracted DNA
A volume of 1 μl of each DNA extract was used for DNA quantification using a Thermo Scientific Nanodrop spectrophotometer ( NanoDrop 20 0 0 C, USA ) by calculating the 260/280 and 260/230 absorbance ratio of each sample [26] . The integrity of the DNA was checked by 1 % agarose gel electrophoresis.

Validation of DNA quality and integrity using PCR and sequencing
Primer This mixture was processed in the thermal cycler Techne TC-512 for PCR reactions at different successive temperature cycles. A total of 40 cycles were performed. Each cycle consists respectively of an initial denaturation at 95 °C for 5 min, followed by denaturation at 95 °C for 30 s, a hybridization phase at 38.4 °C for 45 s, and an elongation at 72 °C for 30 s, then a final termination cycle at 72 °C for 5 min and finally a cooling phase at 4 °C. DNA sequencing was performed by BGI BIO-SOLUTIONS HONGKONG CO., LIMITED company following the techniques of Sanger et al. [27] .
The delimitation of the two methods is also presented by the percentile profile ( Fig. 2 ). The all total DNA samples extracted using the optimized method yields over 5 ng/μL whereas the majority of the samples extracted using the method of Benbouza et al. [25] yields less than 0 ng/μL, exception for six samples yielding from 0.3 to 26.9 ng/μL.
The 260/280 and 260/230 ratios of the method of Benbouza et al. [25] were not significantly different ( U -test of Mann-Whitney, p = 0.0 0 01 < 0.05), indicating low yield and the presence of high levels of contaminants ( Fig. 3 ). On the other side, the 260/280 and 260/230 ratios of the optimized method were close to 2.25 and 1.0, respectively ( Fig. 4 ). The 260/280 ratio of the optimized method showed a larger variance.
The large majority (65%) of the sample extracted using the method of Benbouza et al. [25] were out of the range of good quality DNA (1.8-2.0), indicating a poor quality DNA ( Fig. 5 ). The optimized method gave a relatively higher (14) number of sample in the range of good quality DNA or greater than 2 ( Fig. 5 ).
Of the 97 dung samples extracted with the new optimized method, the amplification success rate was 82.50 % (80 out of 97 samples).
The results showed that DNA samples extracted by the novel improved methods presented good amplification as proved by clear and thick bands of expected amplicons in the gel, indicating that the improved method resulted in good quality DNA of sufficient quantity ( Fig. 6a ). The PCRs with DNA extracted using the method of Benbouza et al. [25] , fail to amplify, indicating poor quality DNA. The absence of amplicons with the negative control indicates that there was no cross contamination ( Fig. 6b ).  Of the 80 samples that had a positive PCR, 67.5% (54 out of 80) were successfully sequenced. Approximately 600 base pairs (bp) located in the control region of mtDNA were amplified with primer pairs LafCR1_F and LafCR2_R. The resulting chromatogram depicted evenly-spaced peaks, each with only one color ( Fig. 7 ).
The resulting fragments sequenced were blasted against the NCBI database and the sequences had a 99 % similarity with previously reported sequences of Loxodonta cyclotis .

Estimation of the cost and time of DNA extraction
The time and the cost required to extract DNA from a sample using each of the two methods was determined by summing the times consumed by all the steps of each extraction protocol ( Table 2 ). The New optimized method is approximately eightfold more time-consuming (29 h 45 min) compare to the methods of Benbouza et al. [25] . The average cost of DNA extraction from a sample based on

Conclusions
Molecular genetic analyzes in elephant populations are sometimes limited by the availability of DNA extraction protocol, fresh animal material, and the time required for extraction as well as by the quality of DNA extracted from dung samples.
Here we proposed an optimized CTAB DNA extraction protocol using forest elephant dung samples. The protocol is relatively simple and it yields a high DNA concentration. The extracted DNA is of good quality and easily amplified by PCR. The high-amplification percentage of mitochondrial genes in fecal DNA and subsequent sequencing analysis for PCR products indicate that the proposed optimized method is reliable for molecular analysis of forest elephant dung samples. It also has the advantage to be cost-effective and is five times less expensive compared to the method of Benbouza et al. [25] .
It also offers significantly easier to prepare and transport samples, and thus is suitable for studies in countries where elephants occur. However, it has the disadvantage of being time-consuming. The results of the present fecal DNA studies could help to assess and evaluate conservation or management policies for forest elephant, a critically endangered species.

Sources of funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Additional information
The optimized CTAB DNA extraction method using forest elephant dung samples yields DNA concentrations ranging from 6.2 to 1716.9 ng/ μL, far greater than the one derived from the method of Benbouza et al. [25] that ranged from -49.4 to 16.3 ng/ μL. In addition to yield a high DNA concentration, our optimized method presents a relatively good DNA quality and is more appropriate. Our method amplified the mtDNA of 80 % of elephant fresh dung samples (n = 94). Compare to the study of Renan et al. [28] who were able to amplify mtDNA from 61 % of swab samples, our optimized DNA extraction method is more efficient. Renan et al. [28] had 100 % amplification success for mtDNA using the frozen swab combined with a QIAamp DNA. Using the CTAB extraction method, the mtDNA amplification rate obtained by these authors dropped to 50 % when extracting DNA from swabs taken from fresh samples frozen in the field (n = 8). Hence, when comparing our optimized CTAB method to the CTAB procedure used by these authors, our method has a greater amplification success.
The high-amplification and sequencing percentage of mitochondrial genes in elephant dung sample confirmed that the new optimized method is an efficient and reliable protocol for fecal DNA isolation.
Hence, our method is a good option to conduct molecular research on forest elephant dung samples.
The difference in performance observed between the proposed optimized methods and the one proposed by Benbouza et al. [25] could be explained by the efficiency of the optimized CTAB/SDS/Proteinase K procedure to lyse cell walls more easily [29][30][31][32] than the methods of Benbouza et al. [25] .
SDS is a key component very well for cell lysis [ 33 , 34 ], and is the basis of a very well-established method for protein purification and molecular weight estimation [ 35 , 36 ].
The prolonged incubation time (overnight incubation at room temperature followed by 3 h of incubation in a water bath) has certainly increased the DNA yield but prolonged eightfold DNA extraction time. The effect of incubation time and other parameters such as the temperature needs further evaluation since several studies [37][38][39][40] indicated that the incubation time differentially impacts on DNA yield depending on the type of sample or the preservation condition used.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.