Environmental DNA Reveals Hidden Biodiversity and Invasive Threats in Georgia’s Rioni River

Due to anthropogenic influences, habitat degradation and a continuous loss of biodiversity in freshwater ecosystems are occurring on a large scale. These invaluable natural resources require rigorous and modern monitoring to effectively guide conservation strategies. In the Caucasus biodiversity hotspot, the Republic of Georgia harbors unique relic flora and fauna, yet its biodiversity remains largely poorly characterized, necessitating urgent protective measures. Central to this ecological importance is the Rioni River, one of the largest rivers in the Black Sea Basin. This river holds global significance as it sustains the last remaining eastern Black Sea breeding populations of at least three sturgeon species, making it a critical focus for the conservation of this most endangered vertebrate group. Furthermore, the Rioni and its tributaries are home to numerous endemic fish species originating from the Colchic refugium, such as *Rhodeus colchicus* and *Oxynoemacheilus phasicus*, which encompass major parts of their known distribution. Despite this immense ecological importance, the river is subjected to heavy anthropogenic pressures, including pollution, mining, poaching, and substantial hydro-power development, exemplified by the Vartsikhe Hydropower Plant, which has reduced the historical spawning area of Black Sea sturgeons from approximately 90 kilometers to only 9 kilometers downstream. Given these severe threats, ongoing species monitoring in the Rioni watershed is essential, yet the last comprehensive assessment of the Rioni River fish community dates back to 1956, leaving a significant gap in ecological knowledge.

To address this critical information deficit, researchers employed a cutting-edge environmental DNA (eDNA) metabarcoding approach, a technique that has emerged as a promising tool for biodiversity monitoring in both fresh and marine waters. This methodology allows for the evaluation of species richness and the monitoring of rare or threatened species, crucially enabling the detection of taxa that are otherwise difficult to find using traditional sampling methods. eDNA methods are recognized as being among the most cost-effective approaches for biodiversity monitoring. The study aimed to set a precedent in the Caucasus biodiversity hotspot by utilizing these modern techniques to inventory fish species diversity in the Rioni River and to evaluate the effectiveness of eDNA sampling in this region. The researchers compared the eDNA-based taxonomic composition against historical records from the 20th century and the updated species list of fishes of Georgia, demonstrating the utility of this technology even in a region where DNA barcode reference libraries for native species are sparse.

The study involved collecting 12 water samples from the Rioni River between September 2018 and March 2019. The Rioni River itself is the longest river in Georgia, spanning 327 kilometers, and possesses an annual discharge of 13.37 km³. The sampling locations were distributed along the river, and water samples, each up to 0.8-liter volume, were collected using NatureMetrics eDNA filter kits. The filters, made of polyethersulfone with a 0.8 µm pore size, were used to capture DNA, which was then preserved according to manufacturer protocols. The volume of water filtered varied, dictated by turbidity, ranging from a minimum of 150 ml to a maximum of 800 ml. These samples were subsequently shipped to and processed by NatureMetrics using the specialized “eDNA Survey – Fish” pipeline. The DNA processing involved DNA extraction using a DNeasy Blood and Tissue Kit and the removal of PCR inhibitors using a DNeasy PowerClean Pro Cleanup Kit. For amplification, a hypervariable 12S rRNA gene fragment was targeted using vertebrate primers, aiming for an amplicon size between 140 and 200 base pairs.

The sequencing and bioinformatics phase was highly detailed. The index PCR products were pooled into a final library and sequenced on an Illumina MiSeq V.2 kit. The subsequent data processing utilized a custom bioinformatics pipeline developed at NatureMetrics, involving quality filtering, dereplication, denoising, and taxonomic assignment. Specific tools such as bcl2fastq for demultiplexing, USEARCH for merging paired-end reads, and cutadapt for primer removal were employed. Sequences were quality filtered using USEARCH, dereplicated, and then unique reads were denoised using UNOISE. Taxonomy was assigned by mapping the dereplicated reads to ZOTU representative sequences at a 97% identity threshold, followed by BLAST searches against the NCBI nucleotide database and a custom in-house taxonomic database of 12S fish sequences. For species-level identification, a sequence identity cut-off of 97% was used, with confident species IDs requiring a minimum of 98% identity. Low abundance detections, defined as less than 0.05% or fewer than 10 reads per sample, were excluded from the final analysis.

The results of the eDNA survey revealed a total of 34 distinct fish taxa across the 12 samples, excluding non-metazoan and contaminant taxa. Of these, 22 taxa could be confidently identified to the species level. The average species richness per sample was 20, with a range spanning from 1 to 24. The most abundant taxon, in terms of DNA sequences, was a nase species belonging to the genus *Chondrostoma*, possibly *C. colchicum* or *C. cyri*, which accounted for 18% of the total sequence reads. This high level of coverage, achieved despite the sparsity of the regional DNA barcode reference library, represents a significant leap forward in updating the fish status of the river. The study also provided evidence of at least nine additional species, including three invasive alien species: *Carassius gibelio*, *Pseudorasbora parva*, and *Rhinogobius lindbergi*.

The detection of these invasive species carries significant ecological implications for the Rioni River Basin. *Rhinogobius lindbergi*, a cryptic invader, was found in five of the 12 sampling locations, indicating it is already widely established in the basin, despite being a recent introduction to western Georgia. While previous reports suggested these species were subject to regular stocking and not yet established, the DNA evidence found in six, nine, and ten sampling locations, respectively, suggests a significant presence within the study area. This finding underscores the high risks associated with these non-native species establishing self-sustaining populations and poses a threat to the native fauna and the delicate ecosystem of the Rioni River.

Perhaps the most noteworthy biological finding was the detection of the DNA sequence for *Pungitius pungitius*, the ninespined stickleback. This species is typically associated with northern regions of Eurasia and America. Its detection in the Rioni River represents a new country record for Georgia. While the study cannot definitively rule out the possibility that the sequences belong to a related species, *P. platygaster*, known from the northern Black Sea and Azov Sea regions, further research is required to resolve the exact species identity. Additionally, the detection of three marine species—Atlantic mackerel (*Scomber scombrus*), anchovy (*Engraulis encrasicolus*), and Mediterranean horse mackerel (*Trachurus mediterraneus*)—is not entirely surprising. These species are commonly found in market fish along the river settlements and could potentially be introduced via wastewater effluent, although their frequent migration in the lower reaches of the Rioni River suggests a possible actual presence at the sampling localities near the river mouth.

The study also highlighted the critical challenges inherent in regional biodiversity assessments. The identification ambiguity for the 17 taxa that could not be assigned to the species level is primarily attributed to gaps in the DNA barcode reference library. The current CO1 barcode library for Georgian fresh and brackish water fishes only includes 52% of known species, and the 12S marker library is likely even less complete. Despite these limitations, the successful recovery of over 70% of known fish taxa, alongside the detection of new invasive and market species, demonstrates the immense potential of eDNA technology for assessing fish assemblages in the Rioni River. However, the study itself noted limitations, including the lack of true field replicates and field controls, as well as the small volumes of water filtered per sample, which may have influenced the observed diversity. Nonetheless, the findings strongly advocate for continued eDNA-based research on aquatic ecosystems within the Caucasus biodiversity hotspot to better understand and protect this crucial habitat.

Subject of Research: eDNA metabarcoding of fish diversity in the Rioni River, Georgia

Article Title: Fish diversity assessed by eDNA detection methods in the Rioni River

Article References: Beridze, T., Japoshvili, B., Edisherashvili, T., Anderson, C., & Mumladze, L. (2023). Fish diversity assessed by eDNA detection methods in the Rioni River. Metabarcoding and Metagenomics, 7, Article e96780. https://doi.org/10.3897/mbmg.7.96780

Image Credits: AI Generated

DOI: 10.3897/mbmg.7.96780

Keywords: eDNA, Metabarcoding, Rioni River, Freshwater Ecology, Invasive Species, Georgia Biodiversity, Ichthyology, Fish, diversity, assessed, detection, methods