A Mg²⁺-dependent high-yield method for extracting high-molecular-weight genomic DNA from a single planarian specimen

Abstract

Background

The isolation of intact, high-molecular-weight genomic DNA (HMW gDNA) is essential for achieving complete genome assemblies. However, extracting HMW gDNA from a single individual of Dugesia japonica remains a technical challenge using the standard protocol, probably due to the presence of abundant polysaccharides and nucleases.

Results

In this study, we have developed a more robust protocol for preparing HMM gDNA, with high yields and quality, from a single D. japonica. The key step in our protocol involves the use of a Mg²⁺-dependent lysis buffer, rather than using metal cation chelation to block the activities of DNase I as in the standard protocol. Using this approach were able to achieve a yield of about 10–15 µg of HWM gDNA per worm. Our method showed species- and region-specific effectiveness, with optimal results observed at 20 mM Mg²⁺ for our local D. japonica specimens. The extracted HMW gDNA is fully compatible with advanced long-read sequencing platforms such as PacBio HiFi and Oxford Nanopore. However, when applied to Schmidtea mediterranea and D. japonica specimens from Beijing, the method was ineffective and led to progressive gDNA degradation.

Conclusions

This protocol offers a simple and high-yield solution for isolating HMW gDNA from D. japonica. It also provides an alternative for organisms whose gDNA consistently exhibits unexplained degradation using established protocols.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12864-025-11731-6.

Introduction

The rapid advancement of long-read sequencing technologies, particularly PacBio High Fidelity (HiFi) reads, Oxford Nanopore Technologies (ONT) ultra-long sequencing, together with high-throughput chromosome conformation capture (Hi-C), has revolutionized genome assembly, enabling nearly contiguous telomere-to-telomere genomes for an expanding range of model organisms [1–5]. Complete genome assemblies offer invaluable insights into species development and evolution, as well as into the expression and regulation of key traits in key model organisms [6]. And the success of these long-read sequencing approaches critically depends on obtaining enough high-molecular weight genomic DNA (HMW gDNA), free from enzymatic inhibitors and has minimal metal ion contamination [7, 8]. Specifically, for HiFi sequencing, the standard protocol requires ~ 10 µg of gDNA per SMRT cell, while the low-input protocol typically requires ~ 3–5 µg. For ONT ultra-long-read sequencing, ~ 10 µg of gDNA per sample is generally used. Additionally, Hi-C sequencing typically requires a starting cell count of 10⁶ cells.

While various HMW gDNA extraction protocols have been developed, obtaining high-quality HMW gDNA remains challenging for some hydrobionts [9], especially the planarians. As a widely used model organism in regeneration research [10], planarians are small (~ 1 cm, 10–20 mg per worm) and rich in mucus and polysaccharides [11], making it difficult to isolate high-quality gDNA suitable for advanced sequencing applications [12]. Standard methods, including commercial DNA purification kits and hexadecyltrimethylammonium bromide (CTAB) DNA extraction protocol often result in unexplained degradation, rendering the gDNA unsuitable for long-read sequencing [12, 13].

To date, for Schmidtea mediterranea, the most well-studied and genomically resource-rich model planarian species, two methods have been used for HMW gDNA extraction. The first, the guanidine thiocyanate (GTC) lysis method [14], uses a high concentration of guanidine thiocyanate to rapidly denature proteins and inactivate nucleases during tissue dissociation. While this method was successfully in yield HMW gDNA suitable for PacBio HiFi sequencing, it suffered for low DNA yields (only 0.5–1 µg per worm), in addition to significant RNA contamination [14, 15], and it required at least 10–20 mixed individuals to get gDNA sufficient for one HiFi sequencing run (10 µg gDNA/cell for normal, 5 µg gDNA/cell for low starting quantity). Nevertheless, this approach has enabled S. mediterranea to achieve one of the best-assembled planarian genomes to date, with haplotype-phased, chromosome-scale assemblies and a contig N50 of 270 Mb [15]. The second method is based on the standard Tris-SDS proteinase K protocol, modified from the book, Molecular Cloning: A Laboratory Manual, 4th edition [16], which has been reported to be able to obtain gDNA of 30–150 kb from fifty 8-mm CIW4 S. mediterranea [13].

However, for Dugesia japonica, another widely studied East Asian model planarian species [10], genome assembly remains highly fragmented (contig N50 ~ 248 Kb) [17]. Using the GTC lysis protocol, we were able to obtain only ~ 2 µg HMW gDNA from a single D. japonica specimen (~ 15 mg), far below the expected yields (in theory, one should be able to obtain 1 µg of gDNA from 1 mg of fresh tissues). Moreover, using the Tris-SDS-proteinase K protocol, a mass of degraded gDNA was observed using Zibo (Shandong Province, China) D. japonica specimens, but the yield was up to ~ 10 µg/worm. Although the GTC lysis protocol can meet sequencing requirements by increasing the number of worms, sequencing gDNA from mixed samples of multiple worms would inevitably complicate the genome assembly process considering the significant chromosomal polymorphism and polyploidy in D. japonica [18]. Unlike the S. mediterranea, which has a stable diploid (2n = 8) [14], D. japonica exhibits considerable variation in chromosome number between different individuals and populations [18]. Even within the same population, individuals show karyotype variations, with diploid (2n = 16), triploid (3n = 24), and mixed-ploidy (2n & 3n) individuals coexisting [19]. Moreover, the extremely high AT content, and extensive repetitive sequences in the planarian genome significantly accelerate chromosomal evolution [15, 17], making single-individual sequencing crucial for accurate genome assembly. Therefore, how to maintain the intactness of gDNA without sacrificing yield from a single worm has become the key challenge for genome sequencing and assembly right now, at least for D. japonica.

Deoxyribonuclease II (DNase II), widely distributed in various organisms, can hydrolyze phosphodiester bonds to degrade DNA and plays an important role in evading host immune defense and in the process of pathogen invasion [20, 21]. A recent study has found that D. japonica from Zibo is rich in four of isoforms DNase II, which can efficiently degrade DNA in vitro over a broad temperature range (0–55 °C) within just 5–15 min, and such degradation can be inhibited using divalet ion, e.g. Mg²⁺, Ca²⁺ etc [22]. Based on this observation, we hypothesize that the activation of DNase II during gDNA preparation is the primary cause of genome degradation in D. japonica. Thus, we modified classic Tris-SDS-proteinase K protocol by adding Mg²⁺ to the lysis buffer to inhibit DNase II activity, a strategy that contrasts with standard DNA extraction methods that use EDTA to chelate divalent metal ions. Using this modified method, we successfully extracted ~ 12 µg of HMW gDNA from a single D. japonica worm (~ 15 mg), which was suitable for downstream sequencing applications such as PacBio HiFi and ONT ultra-long read sequencing. This method represents a practical and high-yield solution tailored for D. japonica, and it may serve as a valuable alternative for other organisms prone to DNA degradation during gDNA preparation.

Materials and methods

Planarian culturing and specimen collection

All planarians used in this paper were cultured and propagated asexually in the laboratory at 18–20 °C. S. mediterranea were maintained in Lushan spring^® water supplemented with 0.21 g/L Instant Ocean, 0.1 mM KCl, 0.1 mM MgSO₄, 0.12 mM NaHCO₃ [23]. D. japonica specimens were collected from multiple locations in China, including Beijing and Zibo City (Shandong Province), and maintained in Lushan spring^® water. The collected samples were classified into three groups based on their origins: Dj-BJ (Beijing), Dj-WT (Yiyuan), and Dj-BS (Boshan). Notably, Yiyuan and Boshan belong to distinct hydrological systems, potentially contributing to genetic or ecological variations among the groups. Planarians were fed shrimp approximately twice weekly and were starved for at least two weeks before gDNA extraction and sequencing.

Mucus removal from planarian surface

Mucus removal from planarian surfaces was performed using a 0.5% N-acetyl-L-cysteine (NAC)-HEPES-NaOH (pH 7.2–7.4) buffer, prepared as described in Additional file 1 (Supplementary Materials). This protocol was adapted from a previously published method [14]. Briefly, an individual worm measuring 1–1.5 cm (< 20 mg) was transferred into the NAC buffer and subjected to vigorous agitation (~ 200 rpm) at room temperature for at least 15 min using a rotator. After treatment, the planarians were washed three times in clean water to remove the residual buffer. The specimens were then dissected into smaller pieces, rinsed an additional three times to eliminate the remaining mucus in gut, and flash-frozen in liquid nitrogen for subsequent analyses.

Development and optimization of an Mg²⁺-dependent tissue lysis buffer for high-quality gDNA extraction

To extract high-quality gDNA of D. japonica, a modified Mg²⁺-dependent tissue lysis buffer (referred to as Mg²⁺-lysis buffer) was developed by adapting the standard Tris-SDS-proteinase K protocol [13]. This modification introduced Mg²⁺ ions to inhibit DNase II activity but omitted EDTA, a chelating agent that binds to Mg²⁺. To optimize DNA extraction conditions, Mg²⁺ concentration gradients (0, 5, 10, 15, 20, 30, 40, and 50 mM) were tested, and the quality of the extracted gDNA was evaluated through agarose gel electrophoresis and pulsed-field gel electrophoresis (PFGE). MgCl₂ was used to prepare solutions with varying Mg²⁺ concentrations. The composition of the Mg²⁺-lysis buffer was as follows: 20 mM Tris (pH 8.0), 100 mM NaCl, 0.4 mg/mL proteinase K, 1% SDS, 0.2% β-mercaptoethanol, 4 µg/µL RNase A (TaKaRa), and Mg²⁺ at specified concentrations. RNase A was included to degrade RNA, resulting in DNA suitable for downstream applications without further purification. For comparison, a control lysis buffer containing 50 mM EDTA but without Mg²⁺ was also prepared. All lysis buffers were freshly prepared before each experiment.

Genomic DNA extraction

Each single D. japonica worm was weighed individually. Individuals exceeding 20 mg in weight were divided into two segments for extraction. Frozen D. japonica tissues (< 20 mg) were thoroughly ground in 900 µL of cold lysis buffer (excluding SDS) using a pestle until no visible tissue fragments remained. After homogenization on ice, SDS was added to the suspension, and the final volume was adjusted to 1 mL with lysis buffer. Tissue digestion was performed in a 55 °C water bath for 4–12 h, with gentle inversion of the tubes 4–6 times every 30 min to ensure uniform mixing. After digestion, the samples were cooled to room temperature (RT). An equal volume of phenol–chloroform–isoamyl alcohol was added, gently inverted to mix, and incubated on a rotator at ~ 80 rpm for 10 min at RT. The samples were centrifuged at 10,000 × g for 5 min at 4 °C, and the aqueous phase was carefully transferred to a new tube. This step was repeated once to minimize protein contamination. To remove residual phenol, an equal volume of chloroform was added to the aqueous phase, gently mixed by inversion, and incubated on a rotator for 10 min at RT. The phases were separated by centrifugation at 10,000 × g for 10 min at 4 °C. This extraction was repeated 1–2 additional times if the organic solution phase was still visible. The final aqueous phase was transferred to a clean tube, and 0.3 volumes of 5 M ammonium acetate were added and mixed thoroughly by inversion. To precipitate gDNA, 0.7 volumes of isopropanol were carefully added, and the mixture was gently inverted until visible DNA strands appeared. The DNA was pelleted by centrifugation at 5,500 × g for 15 min at RT. The resulting DNA pellet was washed three times with 70% ethanol, centrifuged at 5,500 × g for 5 min at 4 °C, briefly air-dried, and resuspended in 200–450 µL of TE buffer. The DNA was allowed to dissolve overnight at 4 °C to maximize recovery. The entire extraction process was performed using a wide-bore pipette tip. Reagents, recipes and a stepwise protocol are provided in Additional file 1 (Supplementary Materials).

Quantification of extracted DNA

The concentration of extracted gDNA was measured using a Qubit™ fluorometer (dsDNA BR assay). DNA purity was assessed using a NanoDrop spectrophotometer (Thermo Scientific), and the measurements were performed in triplicate for accuracy. The intactness of gDNA was evaluated via agarose gel electrophoresis to confirm minimal shearing and optimal extraction quality.

Electrophoresis

The intactness and size distribution of the extracted gDNA were analyzed by loading 200 ng of each sample onto separate wells of a 0.7% agarose gel. Standard agarose gel electrophoresis was performed in 1× TAE buffer at 100 V for 60 min. To further assess the size and relative shearing of the gDNA, pulse-field gel electrophoresis (PFGE) was conducted as described in a previous protocol [24]. The procedure utilized 1% agarose gels prepared in 0.5× TBE buffer, with Lambda Ladder PFG markers serving as size standards. The PFGE conditions were set as follows: 6.0 V/cm, a switch time of 1 s, an angle of ± 60°, and a total runtime of 13 h at 14 °C.

Results

Mg²⁺ inhibits gDNA degradation during extraction from D. japonica

To prepare for gDNA extraction, similar-sized Dj-WT collected from Yiyuan were selected (Fig. 1). Using the standard Tris–SDS–proteinase K DNA extraction method, gDNA from Dj-WT showed extensive degradation (Fig. 2), most likely caused by DNase II activity. To mitigate this issue, Mg²⁺ was introduced into the lysis buffer to inhibit DNase II-mediated DNA degradation. To identify the optimal Mg²⁺ concentration for gDNA extraction, we tested a range of Mg²⁺ concentrations (0–50 mM) in the lysis buffer. DNA yield and quality were assessed using a Qubit™ fluorometer and a NanoDrop spectrophotometer (Thermo Scientific), respectively. As shown in Table 1, DNA yields were largely consistent across all Mg²⁺ concentrations, averaging ~ 12 µg per worm. The A260/A280 ratios confirmed that the extracted DNA was of high purity with minimal contaminations.

Fig. 1.

Top-view photographs of various planarian individuals. The D. japonica planarians Dj-WT, Dj-BS, and Dj-BJ were collected from Yiyuan, Boshan, and Beijing, respectively. All D. japonica and S. mediterranea were kept reproducing asexually. Scale bars, 0.2 mm

Fig. 2.

Agarose gel and pulse field gel electrophoresis of isolated D. japonica-WT gDNA. (A) Agarose gels showing the gDNA quality and intactness of a single Dj-WT worm extracted using different Mg2+-lysis buffers. Every lane represents an independent individual. Each gDNA sample was concurrently analyzed using both agarose gel electrophoresis (Top) and pulsed-field gel electrophoresis (Bottom). “E” indicates lysis buffer containing 50 mM EDTA without Mg²⁺. Numbers 0 to 50 represent lysis buffers with different Mg²⁺ concentrations. “M” represents either a 1 kb DNA marker, λ-Hind III digest DNA marker, or Lambda PFG Ladder. (B) gDNA extracted using 20 or 50 mM Mg²⁺-lysis buffer. (C) gDNA extracted using 20 mM Mg²⁺-lysis buffer. Every lane represents an independent individual. (D) gDNA extracted by GTC buffer

Table 1.

gDNA yields and quality for Dj-WT based on NanoDrop and Qubit dsDNA quantification

Sample	Weight (mg)	EDTA mM	Mg2+ mM	NanoDrop		Qubit		Main Band Size	Figure
				C (ng/µ)	260/280	C (ng/µl)	Yield (µg)
Tris-SDS buffer, 55 °C
WT-E	17.8	50	0	144.8	1.93	56.74	11.348	< 10 K	Figure 2A
WT-1	17.0	0	0	58.9	1.90	39.64	7.928	35 K	Figure 2A
WT-2	17.1	0	5	155.6	1.89	80.26	16.052	55 K	Figure 2A
WT-3	17.0	0	10	144.2	1.89	65.95	13.19	60 K	Figure 2A
WT-4	18.3	0	15	131.7	1.91	64.70	12.94	> 60 K	Figure 2A
WT-5	18.3	0	20	168.3	1.97	93.60	18.72	> 60 K	Figure 2A
WT-6	17.9	0	25	142.7	1.86	61.44	12.288	> 60 K	Figure 2A
WT-7	18.0	0	30	133.2	1.98	61.15	12.23	> 60 K	Figure 2A
WT-8	17.2	0	40	178.4	1.85	70.46	14.092	60 K	Figure 2A
WT-9	17.3	0	50	76.9	1.93	41.86	8.372	35 K	Figure 2A
WT-A	13.8	0	20	61.3	1.91	27.90	12.55	> 60 K	Figure 1.B
WT-B	12.1	0	50	67.8	1.89	30.60	15.30	45 K	Figure 1.B
WT-01	9.9	0	20	74.6	1.82	56.55	9.85	> 60 K	Figure 1.C
WT-02	14.9	0	20	124.7	1.80	87.87	15.82	> 60 K	Figure 1.C
WT-03	13.9	0	20	163.3	1.84	63.34	11.40	> 60 K	Figure 1.C
WT-04	10.7	0	20	106.2	1.82	56.03	10.09	> 60 K	Figure 1.C
WT-05	16.5	0	20	127.0	1.81	78.48	14.13	> 60 K	Figure 1.C
WT-06	13.8	0	20	146.2	1.86	48.20	8.68	> 60 K	Figure 1.C
WT-07	8.7	0	20	101.9	1.86	35.15	6.33	> 60 K	Figure 1.C
WT-08	9.9	0	20	85.4	1.85	52.72	9.49	> 60 K	Figure 1.C
GTC buffer
g-WT-01	15.8	0	0	97.0	2.11	8.93	3.58	> 60 K	Figure 2D
g-WT-02	17.3	0	0	70.8	2.14	6.20	2.48	> 60 K	Figure 2D

The size distribution of the extracted gDNA was evaluated using agarose gel electrophoresis and PFGE. As shown in Fig. 2A, gDNA extracted with lysis buffers lacking Mg²⁺ or containing 50 mM EDTA exhibited significant degradation. In contrast, the gDNA intactness was improved progressively with Mg²⁺ concentrations ranging from 5 mM to 20 mM. At 15 mM Mg²⁺, some degradation was observed (~ 33% incidence, Supplementary Fig. 1), indicating instability. Moreover, once the Mg²⁺ concentration reached 50 mM, gDNA underwent significant degradation and the sizes of DNA fragments were down to about 40–45 Kbp, smaller than that of the low concentration of Mg²⁺ (Fig. 2B). Notably, gDNA extracted with 20 mM Mg²⁺ consistently showed high molecular weight fragments exceeding 60 kbp. Under this extraction condition (i.e. 20 mM Mg²⁺), seven independent parallel experiments were conducted, each yielding stable and consistent HMW DNA electrophoretic profiles (Fig. 2C). These results were comparable to those obtained using the GTC method (Fig. 2D).

To assess compatibility with downstream applications, gDNA extracted with the 20 mM Mg²⁺ buffer was tested on long-read sequencing platforms, including PacBio HiFi (standard and low-input Revio) and ONT ultra-long sequencing. The gDNA extracted at this concentration consistently demonstrated high quality and intactness, making it well-suited for advanced sequencing technologies and we achieved chromosome-level genome assemblies of D. japonica with contig N50 values of 2.87 Mb and 6.38 Mb for diploid and triploid karyotypes, respectively (Manuscript in preparation). Even at 50 mM Mg²⁺, the gDNA remained usable for HiFi sequencing, albeit with reduced fragment sizes. These findings indicate that the Mg²⁺-dependent extraction method is effective for Dj-WT planarians, with 20 mM Mg²⁺ being the optimal concentration to obtain high-quality gDNA fragments up to 100 kbp.

Species-specific and regional differences in Mg²⁺ sensitivity for planarian gDNA extraction

To investigate whether the Mg²⁺-dependent gDNA extraction method is applicable to other planarian species (Fig. 1), we conducted gDNA extraction experiments with different concentrations of Mg²⁺ (0 to 50 mM) at 55 °C using planarians from different regions, including Dj-BS (collected from Boshan, representing a different water system than Dj-WT), Dj-BJ (collected from Beijing), and S. mediterranea. As shown in Table 2; Fig. 3A and C, the results revealed significant species- and region-specific differences in Mg²⁺ sensitivity. Specifically, gDNA extracted from Dj-BS exhibited similar intactness to that of Dj-WT, though it showed slightly reduced sensitivity at higher Mg²⁺ concentrations (Fig. 3A). In contrast, gDNA extracted from Dj-BJ and S. mediterranea displayed progressive degradation as the Mg²⁺ concentration increased (Fig. 3B-C). Notably, even within the same species, D. japonica, different regional populations (Dj-WT, Dj-BS, and Dj-BJ) exhibited distinct responses to Mg²⁺, suggesting the potential influence of environmental or genetic factors on DNA extraction efficacy. These findings suggest that the optimal Mg²⁺ concentration for gDNA extraction is not universally consistent across different planarian species or even within populations of the same species. Tailored optimization of Mg²⁺ concentration is crucial to ensure high-quality gDNA extraction, particularly when applying this method across diverse planarian species and populations.

Table 2.

gDNA yields and quality across different species and regional populations of planarians, assessed using Nanodrop and Qubit dsDNA quantification

Sample	Weight (mg)	EDTA mM	Mg2+ mM	NanoDrop		Qubit		Main Band Size	Figure
				C (ng/µ)	260/280	C (ng/µl)	Yield (µg)
D. japonica specimens collected from Boshan (Zibo City, Shandong Province)
BS-E	12.2	50	0	128.7	1.82	60.20	12.04	15 K	Figure 3A
BS-1	13.9	0	0	126.9	1.84	67.00	13.4	30 K	Figure 3A
BS-2	10.3	0	5	91.3	1.85	49.40	9.88	40 K	Figure 3A
BS-3	10.0	0	10	68.6	1.87	41.60	8.32	> 60 K	Figure 3A
BS-4	10.3	0	15	84.1	1.89	39.00	7.8	60 K	Figure 3A
BS-5	10.8	0	20	41.8	1.94	16.80	6.72	> 60 K	Figure 3A
BS-6	12.2	0	25	44.6	1.95	19.30	7.72	> 60 K	Figure 3A
BS-7	12.0	0	30	67.4	1.90	29.60	11.84	55 K	Figure 3A
BS-8	13.4	0	40	70.9	1.94	32.20	9.66	55 K	Figure 3A
BS-9	1.40	0	50	99.3	1.99	32.20	9.66	55 K	Figure 3A
D. japonica specimens collected from Beijing
BJ-E	14.4	50	0	88.4	1.82	29.41	8.24	> 60k	Figure 3B
BJ-1	16.5	0	0	78.3	1.89	42.11	9.36	> 60k	Figure 3B
BJ-2	14.0	10	5	120.5	1.83	53.24	11.82	< 10 K	Figure 3B
BJ-3	18.9	15	10	112.5	1.86	55.51	12.48	25 K	Figure 3B
BJ-4	17.3	20	15	121.2	1.85	57.25	10.65	40 K	Figure 3B
BJ-5	12.3	30	20	99.0	1.88	24.19	7.61	< 5 K	Figure 3B
D. japonica specimens collected from Yiyuan (Zibo City) and S. mediterranea specimens
WT-01	13.9	50	0	106.0	1.81	58.50	11.70	25 K	Figure 3C
WT-02	17.2	0	0	89.8	1.90	54.40	10.88	20 K	Figure 3C
WT-03	17.9	0	20	138.8	1.98	72.90	14.58	50 K	Figure 3C
WT-04	17.8	0	50	147.6	2.04	64.30	12.86	25 K	Figure 3C
Sme-05	4.83*	50	0	43.2	1.92	14.90	2.98	60 K	Figure 3C
Sme-06	6.27*	0	0	72.3	1.91	25.40	5.08	50 K	Figure 3C
Sme-07	5.27*	0	20	92.9	2.01	24.50	4.90	60 K	Figure 3C
Sme-08	5.34*	0	50	103.0	2.03	27.80	5.56	30 K	Figure 3C

* One adult S. mediterranea worm was cut into four segments, with each segment processed separately for gDNA extraction

Fig. 3.

Agarose gel and pulsed-field gel electrophoresis of isolated gDNA from different species and regional populations of planarians. (A) Agarose gels showing the quality and intactness of gDNA extracted from a single Dj-BS worm using different Mg²⁺- lysis buffers. Each lane represents an independent individual. Each gDNA sample was concurrently analyzed using both agarose gel electrophoresis (Top) and pulsed-field gel electrophoresis (Bottom). “E” indicates lysis buffer containing 50 mM EDTA without Mg²⁺. Numbers 0 to 50 represent lysis buffers with varying Mg²⁺ concentrations. “M” represents either a 1 kb DNA marker, λ-Hind III digest DNA marker, or Lambda PFG Ladder. (B) gDNA extracted from Dj-BJ using Mg²⁺-lysis buffers with different Mg²⁺ concentrations. (C) gDNA extracted from Dj-WT and S. mediterranea using Mg²⁺-lysis buffer with different Mg²⁺ concentrations. Each lane of Dj-WT represents an independent individual. Additionally, one S. mediterranea worm was cut into four segments, with each segment placed in a different lysis buffer for DNA extraction. Each lane represents an independent fragment

Discussion

D. japonica exhibits notable chromosomal polymorphism and polyploidy, extreme AT bias, and a high proportion of repetitive sequences in its genome [17, 18]. These features necessitate the extraction of HMW gDNA from individual specimens to achieve high-quality genome assembly. In this study, we addressed this challenge by optimizing the classic Tris-proteinase K-SDS method and achieved HMW gDNA yields exceeding 10 µg per individual. The extracted DNA is suitable for long-read sequencing platforms such as PacBio HiFi and ONT ultra-long sequencing, making this method particularly advantageous for genome research in small organisms like D. japonica.

The key innovation of our method lies in the use of a Mg²⁺-dependent lysis buffer, replacing the established lysis buffer that uses EDTA to chelate metal ions [25]. This adjustment aligns with the high activity of DNase II observed in D. japonica and some divalent cations, such as Mg²⁺ and Zn²⁺, can effectively inhibit DNase II activity [22, 26]. Compared to the commonly used GTC lysis method, our approach exhibits superior DNA extraction efficiency, achieving a higher yield per milligram of fresh tissue (~ 1 µg DNA per 1 mg tissue). This ensures the isolation of HMW gDNA from a single planarian at quantities sufficient for high-quality sequencing, significantly simplifying the challenges associated with genome assembly.

Interestingly, we have shown a dual role of Mg²⁺ in maintaining DNA intactness during the gDNA extraction process. At an optimal concentration of 20 mM, Mg²⁺ effectively inhibits DNase II, preserving the gDNA intactness. However, concentrations exceeding 30 mM led to significant DNA degradation, likely due to the activation of other Mg²⁺-dependent nucleases, such as DNase I [27, 28], TatD [29, 30], EndoIV [31], and TREX1 [28, 32]. These findings highlight the importance of carefully controlling Mg²⁺ concentrations to balance DNase II inhibition while avoiding the unintended activation of other nucleases. This precise control is critical for obtaining high-quality gDNA suitable for downstream applications.

Additionally, our study showed that the standard Tris-SDS lysis method was also effective in extracting gDNA from S. mediterranea and Dj-BJ. This effectiveness is likely due to the lowered DNase II activity or fewer isoforms of DNase II in these species. For Dj-WT, genomic analysis identified four DNase II isoforms, each containing two conserved “HKD” motifs essential for enzyme activity [22]. In contrast, no DNase II has been reported in S. mediterranea at present. We predicted the homologs of Dj-DNase II (1–4) in the published PlanMine3.0 database [33] as well as Planarian Anatomy Ontology [34], and named them Smed-DNase II (1–4), of which IDs were SMED30015950, SMED30007970, SMED30001221, and SMED30013360, respectively. However, the four predicted Smed-DNase II homologs lacked the complete open reading frames or the critical “HKD” motifs, likely resulting in lower nuclease activity. These differences may explain why the standard Tris-SDS methods remain effective for S. mediterranea. However, due to the absence of genomic data for Dj-BJ, further analysis of this population was not possible.

Although our method performed well in D. japonica specimens from local wild populations and laboratory strains, its applicability may be constrained by inter-population differences in genome characteristics, DNase II activity, and enzymatic regulation and individual difference. Given the limited number of individuals analyzed, future studies are expected to incorporate an expanded local sample pool and refined quantitative analyses, further validating and reinforcing the robustness of the approach. Additionally, precise control of Mg²⁺ concentrations presents practical challenges, as deviations can lead to DNA degradation by activating other nucleases. Current findings are based on a limited sample size, which restricts the generalizability of the method to other populations or species. Future work should focus on: (1) expanding the sample range to include geographically diverse populations and refining regional-specific extraction protocols; (2) exploring the use of alternative nuclease-specific inhibitors (such as ATA) or incorporating them into the Mg²⁺ based lysis buffer—coupled with optimizing detergent concentrations and employing advanced detection equipment— to further improve the quality of extracted gDNA. Moreover, multi-omics approaches could provide deeper insights into the regulation mechanism of DNase II, guiding further optimizations. Testing the method’s applicability to other small organisms with similar genomic features or biological characteristics will further extend its utility and relevance.

In conclusion, we have developed an efficient Mg²⁺-dependent extraction method that allows for the isolation of high-quality HMW gDNA from a single D. japonica worm. Compared to standard methods, this approach offers significant improvements in yield and is well-suited for PacBio HiFi and ONT ultra-long sequencing, making it a powerful tool for advancing genomic research in D. japonica and potentially, other planarian species.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

CTAB

Hexadecyltrimethylammonium bromide

DNase

Deoxyribonuclease

GTC

Guanidine thiocyanate

Hi-C

High-throughput chromosome conformation capture

HiFi

High Fidelity

HMW gDNA

High-molecular-weight genomic DNA

NAC

N-acetyl-L-cysteine

ONT

Oxford Nanopore Technologies

PFGE

Pulsed-field gel electrophoresis

RT

Room temperature

Author contributions

F.T.Y, Q.X.P, and A.L. conceived and designed this project. F.T.Y, A.L., and H.K.D drafted the manuscript. A.L, B.R.S, P.Y, L.L.G, and C.J.X performed the experiments. Y.Z and J.C.Q participated in data analysis and visualization. All authors contributed to manuscript revision, read and approved the submitted version.

Funding

This study was supported by the Natural Science Foundation of China (32370689, 32470518, and 31701845), the Natural Science Foundation of Shandong Province, China (ZR2022MC022 and ZR2024MC154).

Data availability

All data generated or analyzed during this study are included in this article and its supplementary information files. The datasets generated and analyzed during the current study are available in the GenBank repository, with the following accession numbers: Dj-DNase II-1: OL989992; Dj-DNase II-2: OL989993; Dj-DNase II-3: OL989994; Dj-DNase II-4: OL989995. Additionally, the Smed ID numbers for homologs of Dj-DNase II of S.mediterranea used in this study are available in the PLANOSPHERE: SÁNCHEZ LAB PLANARIAN RESOURCES: SMED30015950; SMED30007970; SMED30001221; SMED30013360. Sequencing data generated from the extracted gDNA are not included in this article, as they will be analyzed and published as part of a separate study focused on downstream applications and biological insights.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent to publish

Not applicable.

Competing interests

The authors declare no competing interests.