Spiders play a crucial role in agricultural ecosystems (Benamu 2020; Maloney et al. 2003; Sarma et al. 2013). Because spiders capture and eat small insects, they are particularly effective in controlling pests around crops (Marc et al. 1999; Menalled et al. 2007; Nyffeler 1999). This can reduce pest damage to crops and enable the implementation of pest control activities that are more environmentally and human friendly (Korenko et al. 2016; Scherr and McNeely 2008). Spiders also play a crucial role in maintaining the biodiversity in agricultural ecosystems (Gogoi and Ningthoujam 2021).

Spider webs have strong adhesive properties, making it difficult for small insects to escape once they are caught, resulting in their capture (Murakami 1983; Wilgers et al. 2016). In this way, the webs produced by spiders play a crucial role in the survival of spiders as they serve as a means of capturing food and providing shelter from predators (Craig and Freeman 1991; Manicom et al. 2008). Moreover, by analyzing the types and numbers of small insects caught in spider webs, it is possible to understand the biodiversity and ecological health condition of the agricultural ecosystem. In addition, during the process of capturing small insects with spider webs, the genes of arthropods along with those of spiders feeding from the web can be present on the web in the form of eDNA (Astrin et al. 2006; Barrett and Hebert 2005; Blagoev et al. 2013; Thomsen and Willerslev 2015).

One way to determine a spider’s food source is by determining the type and number of small insects that are caught in its web (Blackledge 2011; Ludwig et al. 2018). However, spiders use digestive fluid to dissolve and extract food from the inside (Nentwig et al. 2022). During this process, the food is clumped into a round shape and transformed into a small lump, leaving only the shell behind, which makes it difficult to determine the exact source of food without dissection. In addition, when spiders are not hungry, they anesthetize and paralyze the prey caught in their webs before encasing it with spider webs for storage; thus, identifying the spider’s food source is challenging, as it can only be determined when the food is caught in the web or when the spider consumes it (Eberhard et al. 2006). Moreover, even if there are traces of food present in the spider web, it is difficult to determine the species of the spider without collecting spider individuals.

The environmental DNA (eDNA) of arthropods present in spider webs can provide crucial information for the study of spider distribution and ecology, yet research in this area is still notably limited. The concept of eDNA in spider webs was first introduced in 2015 (Xu et al. 2015), and studies on this topic remain relatively scarce. Existing literature primarily focuses on airborne eDNA attached to spider webs, with a limited exploration into using spider web eDNA for investigating spider prey sources (Corse et al. 2019; Gregorič et al. 2022).

This study aims to explore the potential of conducting ecological research on spiders by using eDNA to identify spider species and determine the types of food available through the analysis of eDNA found in spider webs.

Sampling spider webs and spiders in the field

The spider webs were collected in January 2023 from Nonsan City, Chungnam Province, South Korea, which has a high proportion of farmland. Five spider webs located adjacently (approximately 50 m radius) in the same region were sampled. The collection environment for the spider webs is characterized by being surrounded by hills and having small-scale rice paddies nearby (36°14´6.71˝, 127°13´0.19˝). At the same time, adult spiders were collected from spider webs existing near the agricultural environment, and eDNA was extracted from the spider webs. The collected spiders were stored in 95% ethanol, and then, their morphological characteristics were compared to those described in the literature (Im and Kim 2000; Kong 2013). The eDNA of the spider web was collected by attaching the web to a GF/F filter (Fig. 1). The tip of a pair of forceps was grabbed and swiped along the entire spider web to attach the silk to a GF/F filter membrane. Each strand of spider silk is affixed to a single filter membrane, and the filter membrane with attached spider silk was folded in half and placed into a Salivette tube.

Figure 1. Process of spider web sampling by GF/F filter paper. eDNA: environmental DNA; COI: mitochondrial cytochrome C oxidase subunit I.

The filter was autoclaved before collection and sealed in a petri dish before analysis. One filter paper was used for each spider web, and a total of five spider webs that were close by, but were not intertwined with each other, were collected. Of these five webs, four had no food in them, while one had food in it (Fig. 2). The filter collected from the spider web was placed in a Salivette tube and transported to the laboratory under refrigerated condition.

Figure 2. The sampled spider web in this study. In S1, we found three spider egg sacs, and in S2 we found an exuviated spider trace. In S4, we found a trace of food storage in the spider web. There were no noticeable traces on the spider webs of S3 and S5.

DNA and eDNA extraction and spider DNA barcode gene analysis

From the web samples transported to the laboratory, the eDNA present within the web was extracted using the extraction method presented in the environmental DNA sampling and experience manual version 2.1 (Minamoto et al. 2021). The DNA of captured spider individuals from spider webs, where eDNA was collected, was extracted using a genomic DNA extraction kit (Blood and Tissue kit; Qiagen Co., Hilden, Germany). Before extraction, cells were homogenized to facilitate the DNA extraction process. The mitochondrial cytochrome C oxidase subunit I (COI) gene was amplified from the DNA extracted from both the spider and spider web (BF2 + BR2; amplicon size 461 bp) (Elbrecht and Leese 2017), and the species was identified through BLAST based on Genbank database (Schoch et al. 2020; Tatusova et al. 2016). The polymerase chain reaction (PCR) conditions for amplifying the spider’s COI gene were as follows: initial denaturation at 95°C for 5 minutes, followed by 35 cycles of denaturation at 95°C for 30 seconds, annealing at 50°C for 30 seconds, and extension at 72°C for 50 seconds. The final extension was performed at 72°C for 5 minutes. In addition, the COI gene lineage analysis of spiders was performed based on the COI gene sequence identified in both the spiders and spider webs. MEGA 11 (molecular evolutionary genetics analysis version 11.0.11) was used to perform phylogenetic analysis, and a dendrogram was constructed using the maximum likelihood method.

Meta-barcoding of the COI gene

The genetic contigs of spiders analyzed through metabarcoding in spider webs were subjected to phylogenetic analysis based on gene nucleotide sequences to elucidate that they share genetic characteristics with spiders inhabiting the webs. To conduct the phylogenetic analysis, genetic data of spiders belonging to the Linyphiidae family, including the funnel weaver spider, were downloaded from GenBank for species found in Korea. Subsequently, using MEGA11, the data were aligned, and Maximum Likelihood phylogenetic analysis was performed employing the Tamura-Nei model (Stecher et al. 2020; Tamura and Nei 1993; Tamura et al. 2021). The root branch of the phylogenetic tree was established by incorporating the COI gene of the Brown planthopper (Nilaparvata lugens) into the phylogenetic analysis.

In order to analyze the genes of spiders and small insects, which serve as food sources for spiders, the COI gene specific to invertebrates was amplified using primers (Elbrecht and Leese 2017; Elbrecht et al. 2018) and Master mix Taq polymerase (GainBlue^TM Hot Start Max Master Mix PCR Master Mix; Gainbio ltd., Daejeon, Korea). The product amplified by primary PCR was attached to a barcode index (Illumina Nextera kit V2) through secondary PCR. The conditions for the second PCR were as follows: initial denaturation at 95°C for 3 minutes, followed by 9 cycles of denaturation at 95°C for 30 seconds, annealing at 55°C for 30 seconds, and extension at 72°C for 30 seconds. The final extension was performed at 72°C for 5 minutes. The primers utilized in the first-round PCR were identical to those employed in spider DNA barcoding analysis (BF2 + BR2). Additionally, Nextera consensus and sequencing adaptors were appended to the existing primer sequences for use. Samples with the barcode index were analyzed for metabarcoding by a company specializing in analysis (Ubix bioinnovations Co. Ltd, Waterloo, ON, Canada) using the Illumina MiSeq system, and the species was identified based on COI gene sequencing using the DECIPHER package’s (Wright 2016) IdTaxa classifier (Murali et al. 2018).

Spider COI gene in the web eDNA

DNA from arthropods was found in all five spider webs, and no other biological DNA was found. All five spider webs displayed the COI gene of Parasteatoda tepidariorum (common name: Glasshouse spider), and COI gene was not detected in the other spiders. While collecting spider webs, a single P. tepidariorum individual was found near the web (Fig. 3), and no other spider species were detected. Moreover, some spider webs contained egg sacs, and spiders; exhibited the same morphological characters of P. tepidariorum. The genes extracted directly from the P. tepidariorum individual collected at the site formed a lineage with the COI gene of P. tepidariorum found in the spider web; and accordingly, the spider web collected at the survey site could be identified as the one created by P. tepidariorum (Fig. 3A).

Figure 3. (A) Phylogenetic reconstruction of spiders within the family Theridiidae based on the mitochondrial cytochrome C oxidase subunit I (COI) gene sequence. The evolutionary lineage was elucidated using the maximum likelihood method with the Tamura-Nei model. The percentage of trees demonstrating the coalescence of respective taxa is indicated adjacent to the branches. The red color denotes DNA sequences analyzed in this investigation. The placement of individual spider COI genes on the phylogenetic tree corresponds to those captured in spider webs, while the spider COI genes represent DNA contigs amplified from spider web environmental DNA. (B) A visual representation of a spider individual within its web is provided. The morphological attributes of the spider individual closely resemble those of Parasteatoda tepidariorum.

COI gene of other Arthropoda in the spider web eDNA

Of the five spider webs collected, only one had traces of food storage, making it difficult to determine the taxa of the prey as they were completely covered with spider webs. Arthropod taxa genes other than those of spiders were found solely in the web where traces of food were present. In addition to P. tepidariorum, the genes of arthropods found in spider webs were two species belonging to the family Chironomidae and one species belonging to the family Berytidae, resulting in a total of three taxa, and genes of invertebrates other than arthropods were not found (Table 1). Species belonging to the family Chironomidae were Orthocladius tamarutilus and Tanytarsus tamagotoi, while only one species was identified for the family Berytidae, Yemma exilis.

Table 1 . Number of contigs of each spider web environmental DNA metabarcoding analysis.

	S1	S2	S3	S4	S5
Parasteatoda tepidariorum	531	254	458	41	56
Orthocladius tamarutilus	-	-	-	115	-
Tanytarsus tamagotoi	-	-	-	37	-
Yemma exilis	-	-	-	45	-

Arthropod DNA contigs were only found in the spider web that remains a storage trace of an insect.

Parasteatoda tepidariorum is a globally distributed species known for constructing webs beneath roofs or in corners, presenting an ostensibly irregular appearance, yet in reality, revealing a specific and streamlined architectural structure (Bradley 2012; Edwards 2004). Parasteatoda tepidariorum forms three to four cloddish egg sacs in the center of the spider web, and adults diligently care for the eggs around the egg sacs (Oda and Akiyama-Oda 2020). In this study, the spider webs had an extremely irregular shape, and three egg sacs were observed, attached side by side to the spider web. Based on these results, it can be inferred that eDNA may be generated in spider webs as spiders inhabit them. Consequently, it is conceivable that spider species can be identified without the need for spider collection by utilizing the eDNA present in spider webs that spiders leave behind.

Both families Chironomidae and Berytidae are found worldwide and typically travel short distances. In particular, Y. exilis found in the spider webs feeds on arthropods, such as the families Chironomidae and Cyclorrhapha, and mainly inhabits plants, such as fallen leaves in mountainous areas or shady locations (Kohno and Hirose 1997; NIBR 2016). The time of collection of spider webs was January, during which spiders remain inactive due to low temperatures and it is also a season of inactivity for spider prey. Thus, there may be only a few genes remaining for arthropods that were caught in spider webs. Common species like P. tepidariorum, may consume various types of food based on the surrounding environment, and they do not exhibit selective eating of specific species (Blasco-Aróstegui and Calatayud-Mascarell 2023; Goyal et al. 2017; Milne and Milne 1980). The environment in which spider webs were collected is a mixture of paddy fields, forests, and nearby residences, with various arthropods, including the families Chironomidae, Berytidae, Cyclorrhapha, Culicidae, and Heterocera. In particular, arthropods related to the agricultural environment, such as the families Chironomidae and Berytidae, are known as spider food in the agricultural environment (Graf et al. 2020; Park and Lee 2006), which suggests that in this study, the genes of three arthropods species, O. tamarutilus, T. tamagotoi, and Y. exilis, discovered in the spider web might be food for P. tepidariorum.

While some literature reports a diverse array of arthropod genes found in spider webs, our study identified a minimal quantity of arthropod genes in the spider webs. In general, the eDNA’s lasting time for analysis varies widely based on environmental conditions, ranging from a few hours to more than several years (Bairoliya et al. 2022; Harrison et al. 2019). In the case of eDNA in arthropods, PCR amplification for analysis is possible for about two to three months, but eDNA exposed externally for more than three months becomes fragmented and unreliable for proving the existence of the target taxa (Valentin et al. 2021). Accordingly, it has been determined that arthropod genes other than those of spiders were found exclusively in the spider webs where food was present, and thus, a higher diversity of arthropod genes is expected to be discovered between June and September, during the period of active feeding by spiders. Moreover, the analysis of spider prey using eDNA, specifically in orb-web spider species that capture and consume prey within their webs, is applicable exclusively to situations where the prey’s eDNA can be distinctly differentiated from the web material. Particularly during the active summer when organisms release airborne eDNA (eDNA in the air) that can adhere to spider webs, discerning spider prey amidst a diverse array of biological traces becomes exceedingly challenging (Newton et al. 2024). Therefore, the utilization of spider web eDNA for prey analysis is deemed feasible only for spider species that are made of spider-web, and meaningful results can be obtained solely from webs where prey is stored.

In conclusion, this study found that spider webs are not simply threads of protein components produced by spiders, but rather an important source for analysis that provides information on the ecological activities of various spiders. In particular, spider species that form spider webs were identified solely based on the genes of spiders found in the webs without directly collecting spider individuals, and the possibility of identifying spider food sources was confirmed through genetic analysis of spider webs without the need for collecting and analyzing traces of food within the webs. This can help facilitate efficient ecological research on spiders, enabling researchers to study spiders in diverse regions and at different times. However, in cases where spiders do not rely on spider webs to hunt for food, such as in the case of wandering spiders, it is impossible to determine their food sources through the genes present in spider webs, as shown in this study. Therefore, this study is considered useful for ecological studies focusing on settling spiders or webbing spiders (e.g., Theridiidae, Araneidae, and Gnaphosidae), which primarily rely on spider webs as a food source.

Not applicable.

eDNA: Environmental DNA

COI: Mitochondrial cytochrome C oxidase subunit I

KK and STK contributed to the study conception and design. KK wrote the first draft and contributed to material preparation, data collection and analysis, and interpretation of results. STK identified spiders based on morphological characteristics. All authors read and approved the final manuscript.

This work was carried out with the support of “Cooperative Research Program for Agriculture Science & Technology Development (Project No. RS-2020-RD009339)” Rural Development Administration, Republic of Korea.

Not applicable.

Not applicable.

Not applicable.

The authors declare that they have no competing interests.