Agricultural soils support a wide diversity of organisms, with nematodes ranking among the most abundant and functionally important metazoans [1]. These microscopic organisms inhabit nearly all terrestrial and aquatic environments, including extreme habitats such as Antarctica, the Great Salt Lake and deep subsurface soils [2–4]. Within the food web, nematodes play key ecological roles by contributing to organic matter decomposition, nutrient mineralization, and the suppression of plant diseases [5].

Advances in molecular techniques, particularly metabarcoding, have revolutionized our capacity to study nematode diversity and ecology. The approach offers the possibility of detecting rare species and provides higher taxonomic resolution than traditional morphology-based methods, which are often time-consuming and require taxonomic expertise [6–10]. As a result, metabarcoding has become increasingly attractive and widely adopted tool for assessing soil biodiversity and health [6,9,10].

However, both the DNA source and primer choice can heavily influence nematode metabarcoding results. The most common approach involves isolating nematodes from soil prior to DNA extraction, hereafter referred to as nematode DNA extraction [10,11]. Although nematode DNA extraction has become a standard protocol, it is labour-intensive, often requires large volumes of soil (ideally >200 g), and currently lacks commercial kits specifically optimized for nematode-rich samples [12–14].

Alternatively, extracting DNA directly from soil can reduce processing time, minimize species detection bias associated with traditional extraction protocols, and enable the identification of low-abundance nematodes [15]. Soil DNA extraction is already a popular and well-established approach in studies of fungi and bacterial ecology; however, it has been less widely adopted for nematode metabarcoding [16,17]. The main constraints limiting its acceptance in nematology are the relatively small volumes of soil that most extraction kits can handle and the lack of nematode-specific primers [13,18,19].

The development of metabarcoding has resulted in the design of several primers aimed at characterizing nematode communities [20,21]. Among these, the small-subunit 18S rRNA primer pair NF1–18Sr2b is one of the most widely used markers for profiling nematode communities [6,18,22]. However, the NF1–18Sr2b (NF1) primer set has notable limitations when applied to soil DNA extraction, as it tends to co-amplify non-target sequences from plants, fungi, protists, and other eukaryotes [13,19]. It has been shown to amplify only 4–10% nematode sequences in soil DNA extractions [13,19,23]. The consistently poor nematode recovery from direct soil DNA has impeded the broader adoption of this method for nematode metabarcoding studies.

To optimize the NF1–18Sr2b primer, Sapkota and Nicolaisen (23) designed the NemF–18Sr2b (NemF) primer set by modifying the NF1 primer to better match conserved nematode sites while excluding mismatches at the 3′ end. Subsequently, Waeyenberge, de Sutter (14) introduced degenerate bases into the NemF–18Sr2b primer to improve GC content to develop the NemFopt–18Sr2bRopt (NemFopt) primer set. These optimized primers have demonstrated improved performance, with NemF reported to recover over 74% of nematode sequences from soil-derived DNA samples [23,24].

Despite these advances, no study has directly compared the performance of NF1, NemF, and NemFopt using both nematode DNA and soil DNA extracted from agricultural soils. As a result, it remains unclear whether optimized primers make soil DNA extraction a viable alternative to nematode DNA extraction for community profiling. Prior assessments of soil DNA extraction for nematode metabarcoding have relied largely on a single primer set, which may underestimate the method’s full potential. Primer choice can strongly influence nematode detectability, and extraction efficiency may be further improved through protocol optimization. The development and standardization of soil DNA extraction protocols will improve the comparability and reproducibility of nematode community assessments across ecological studies and facilitate the implementation of global soil biodiversity monitoring programmes. Therefore, we hypothesized that an optimized soil DNA extraction method, when combined with suitable primer sets, could yield nematode community profiles comparable to those obtained from nematode extraction regardless of soil input volume. To address this gap, the present research aims to (i) evaluate the specificity of the NF1, NemF, and NemFopt primers in characterizing nematode communities, and (ii) compare community profiles derived from nematode and soil DNA extractions using different soil types and volumes. The study hypothesizes that the NemF and NemFopt primers will improve nematode detection in soil DNA extractions, and that the best-performing primer will yield community profiles comparable between nematode DNA extraction (NE) and soil DNA extraction (SE). The results of this study are intended to provide guidance on the optimal selection of primers and DNA extraction methods for metabarcoding analyses of nematode communities in agricultural soils.

For nematode ecological studies, soil DNA extraction has historically been underutilized, primarily due to low nematode sequence recovery [11,19]. The commonly used NF1–18Sr2b (NF1) primer targets the 18S rRNA V6–V8 region and has been widely applied, but it is not nematode-specific and co-amplifies non-target organisms [6,13]. Consequently, NF1 often yields only ~4–10% nematode sequences in soil DNA extractions, compared to over 80% from nematode DNA extraction [13,18,19]. This poor recovery has contributed to the perception that small soil subsamples cannot capture complete nematode communities. For example, Griffiths et al., [18] reported higher nematode richness from pre-isolated samples than from bulk soil, although this method is labor-intensive and may introduce extraction-related biases. However, recent large-scale studies suggest that these limitations are methodological rather than fundamental [19]. They showed that nematode communities recovered using soil-extracted DNA were broadly comparable to those obtained from nematode-sample DNA across a continent-scale survey. The key challenges in soil DNA approaches are now understood to be the use of non-specific primers and small soil input volumes. The former causes most reads to derive from non-nematode soil biota, while the latter limits the detection of rare nematode taxa. For instance, Kawanobe et al., [34] demonstrated that conventional 18S primers amplify abundant non-nematode sequences and that small soil quantities reduce sensitivity to rare taxa, advocating the need for nematode-specific primers to overcome these issues. Therefore, improving primer design and increasing soil mass can substantially improve nematode recovery.

To evaluate this further, two nematode-targeted degenerate primer pairs, NemF/18Sr2b (NemF) and NemF/18Sr2bRopt (NemFopt), were tested using both nematode-enriched DNA extraction (NE) and direct soil DNA extraction (SE) methods. As expected, NE samples yielded a higher proportion of nematode reads than SE samples, however, primer choice had a stronger effect than DNA source. For example, the NF1 primer performed poorly, detecting only 2–4% of nematode sequences in SE samples, confirming previous studies [13,23]. In contrast, the degenerate NemF and NemFopt primers greatly improved nematode detection from SE. Using NemF, the majority of reads in 1.25 g soil extractions were nematode-derived, aligning with findings by Sikder et al., [24], who obtained ~74% of nematode sequences from SE extracts. The NemFopt primer yielded the highest fractions of nematode reads (>70% of total sequences) in SE, outperforming the reference NF1 primer. While Waeyenberge et al., [14], had cautioned that NemFopt might co-amplify non-target organisms and suggested it for broad eukaryotic profiling, our results indicate otherwise. NemFopt was highly selective for nematodes, yielding 100% of nematode reads in NE and over 70% in SE samples. These findings confirm our hypothesis that nematode-specific primers substantially enhance both the sensitivity and specificity of soil DNA metabarcoding, supporting the use of soil DNA extraction as a reliable alternative to nematode isolation [13,21,23].

Differences in DNA extraction methods had a pronounced effect on nematode detection. The SE outperformed NE in capturing family-level diversity, and uniquely detected four families (Bunonematidae, Rhabdolaimidae, Aphelenchidae, Tylenchidae) that NE missed. In contrast, only a single family (Panagrolaimidae) was unique to NE. NE-derived samples showed high relative abundance of certain families, such as Nygolaimidae and Plectidae, suggesting that extraction method can bias community composition. Similar results have been reported in prior comparisons of bulk nematode DNA versus soil DNA methods [18,19]. More importantly, direct soil extractions tend to underrepresent some predators while overrepresenting other groups. This shows the importance of choosing an appropriate extraction method for accurate nematode biodiversity estimates.

Likewise, primer choice greatly influenced which nematodes were detected, signifying primer-driven biases in metabarcoding. In this study, the NemF and NemFopt consistently amplified taxa such as Anaplectus sp., Ditylenchus sp., and Oxydirus nethus across all extraction methods, whereas NF1 failed to recover these taxa. This discrepancy likely results from primer–template mismatches. For example, Anaplectus sp. went undetected by NF1 in SE samples, even though Ahmed et al., (11) successfully recovered it from a mock community using the same primer. The difference is that in our soil communities, the NF1 binding site in Anaplectus probably contained nucleotide mismatches, preventing amplification, whereas in controlled mock communities the primer perfectly matched the target sequences. In contrast, NemF and NemFopt target a more conserved region of the 18S rRNA, enabling broader taxonomic coverage. The NemF primer was specifically designed for enhanced nematode specificity and coverage, allowing amplification of a wide range of nematodes while avoiding non-target DNA. The success of NemF and NemFopt in detecting Anaplectus and other taxa missed by NF1underscores the importance of using updated, nematode-tailored primers for accurate metabarcoding studies.

Overall, the NE method using NemF and NemFopt primer pairs yielded fewer taxa than the SE approach. This is likely due to the limited extraction efficiency of the DNeasy Blood and Tissue Kit. Despite pre-treatment with liquid nitrogen [14], intact nematodes were still observed microscopically, suggesting incomplete lysis. In contrast, the DNeasy PowerMax Soil Kit and DNeasy PowerSoil Pro Kit produced higher DNA quality and concentrations (25 ng/µL and 35 ng/µL, respectively) compared to the DNeasy Blood and Tissue Kit (4 ng/µL). These results suggest that future nematode DNA extraction protocols could benefit from incorporating soil DNA extraction kits, as supported by a recent study [35]. Furthermore, nematode DNA extraction remains widely accepted in metabarcoding studies [6,11], and continues to serve as a benchmark for evaluating primer performance in high-throughput sequencing. Our results confirm that a highly specific primer should be able to detect similar nematode communities across both SE and NE methods. This was validated through NMDS and cluster analyses, where NemFopt, particularly in SE.10, identified nematode assemblages comparable to those obtained from NE samples. These findings suggest that NemFopt enhances nematode detection in soil DNA extracts and yields community profiles closely matching those derived from isolated nematodes. Moreover, current soil sampling practices are considered inefficient, as large amounts of soil, often exceeding 10 kg per site, are collected from the field, while only a small fraction (typically <200 g) is used for analysis [6,35]. Validating and optimizing soil DNA extraction methods could therefore reduce resource waste, streamline workflows, and improve the efficiency and accuracy of nematode metabarcoding studies.

In the present study, nematode diversity patterns further supported the effects of primer choice and extraction method. OTU richness was higher in the SE approach than in the NE when using the NemF and NemFopt primers. Conversely, NF1 yielded higher richness in NE across all DNA sources. The finding of NF1 is consistent with Donhauser et al (19), but the authors attributed the lower richness in soil DNA extractions to the small soil volumes used (10 g). However, our results with the NemF and NemFopt indicate that 10 g of dry soil was sufficient to capture the local nematode community when nematode-specific primers were used.

A key ecological insight from this study is that primer choice can shape perceived nematode community structure without fundamentally altering higher‐order ecological interpretations. Across both soil types, the ecological indices were largely stable between NE and SE, and this stability persisted regardless of the soil input volume used for SE (1.25 vs 10 g of dry soil). For both the MI and EI, no significant differences were detected between SE (at either soil volume) and NE, indicating that ecological inferences based on these indices are robust to DNA source and within the tested range of soil input masses. These findings directly support our hypothesis that comparable community-level ecological signals can be obtained from different extraction inputs and soil types and further suggest that increasing soil mass for soil DNA extraction does not necessarily provide additional ecological information, at least as reflected by MI and EI. In contrast, the SI was the only index that showed sensitivity to methodological differences, with a significant effect of extraction method (DNA source) observed with NF1. This likely reflects the strong dependence of SI on accurate recovery of higher trophic levels (e.g., omnivores and predators), which are typically rare and therefore more prone to under-detection or under-representation [6,10]. This was particularly evident with NF1, which underrepresented these groups in clay soils.

The absence of soil volume effects on community composition and ecological indices suggests that increasing soil input does not necessarily enhance nematode metabarcoding outcomes, supporting previous work demonstrating consistent performance using ≤0.25 g of soil [24]. This finding highlights the importance of protocol efficiency and primer optimization over bulk soil quantity and has clear implications for methodological standardization across laboratories. The adoption of optimized primers such as NemFopt, along with harmonized DNA extraction protocols, consistent soil input masses, and benchmarking using mock communities, would substantially improve cross-study comparability. We acknowledge that this study was limited to two soil types and field sites, which may constrain generalization to other soil contexts. Future work should evaluate a broader range of soil systems, larger or replicated soil inputs to improve detection of rare taxa, and alternative 18S rRNA regions to further refine primer performance and establish robust standards for nematode metabarcoding.

Community analyses consistently indicate that DNA source and primer choice interact to influence perceived nematode community composition, but that this effect is minimized when communities are recovered using NemFopt. NMDS ordinations showed that only NemFopt produced similar community structures across DNA sources, irrespective of soil input volume, indicating strong methodological consistency. In contrast, NF1 and NemF displayed greater separation among DNA sources, suggesting higher sensitivity to extraction-related variation. Principal coordinates analysis further demonstrated that primer choice structured community composition more strongly than soil type, with distinct clustering among primer sets, however, NemFopt-derived communities remained broadly comparable to those detected by both NF1 and NemF. Heatmap clustering reinforced these patterns, showing that NE communities clustered together regardless of primer set, consistent with previous reports that NE yields more homogeneous community structure than SE [19]. NemFopt-derived SE communities clustered closely with NE, indicating that this primer mitigates extraction-related bias and enhances cross-method comparability. Finally, downstream choices such as sequencing pipelines, reference databases, and bioinformatics filters may also influence nematode metabarcoding results. Therefore, there is a need for standardization of nematode metabarcoding workflows, including testing of new primers, DNA extraction protocols and reference databases, to ensure reproducibility and improve cross-study comparability. While the PR2 database was employed for taxonomic assignment and primer validation because of its broad coverage of eukaryotic diversity, we acknowledge that it may not provide the same level of curation for nematodes as specialized databases such as NemaBase or NemaTaxa. Consequently, taxonomic resolution at lower ranks (e.g., genus or species) may be limited and potentially prone to misassignment.

Metabarcoding has become an essential tool in nematode ecological studies, however, challenges remain that underscore the need for methodological standardization. Although several primers have been designed, further evaluation is necessary to determine the most suitable primers for nematode metabarcoding, similar to standardization efforts already undertaken in other fields such as microbial and fungal ecology. Comparative studies across different soil types, primer sets, and extraction methods are crucial for improving reliability and promoting broader adoption of this technique.

Our findings demonstrate that primer choice strongly influences nematode identification, feeding group composition, and diversity indices. Degenerate primers such as NemFopt provided greater taxonomic coverage in both soil and nematode DNA extractions compared to the widely used NF1 primer. Contrary to concerns that soil DNA extraction may not generate reliable nematode profiles due to limited sample volumes, NemFopt applied to SE.10 successfully captured communities comparable to those obtained through nematode DNA extraction. Our study further demonstrates that soil DNA input volume within the tested range does not substantially influence nematode community inference. Ecological indices were generally consistent between nematode- and soil-extracted DNA, except for SI. Community composition was sensitive to primer choice, but NemFopt reduced extraction-related variability. The study was limited to two soil types, two field sites, and three primer sets. Further work should assess a broader range of soil conditions, larger or replicated soil inputs, and alternative 18S rRNA regions. The findings in this study highlight the importance of rigorous primer evaluation using field-derived soil samples and support ongoing efforts to standardize primer selection and soil DNA extraction protocols. Adoption of such standardized approaches will enhance the reliability, reproducibility and interpretability of nematode ecological studies and soil health assessments.