The Chienti is one of the main rivers of the Marche region, in central Italy. It rises in the Umbria-Marche Apennines, in Serravalle di Chienti, and extends for about 91 kms until it flows into the Adriatic Sea. This river, which is fed by several tributaries, including the Fiastrone and the Fornace stream, has been modified by artificial reservoirs, such as Lake Polverina, Lake Caccamo, and Lake Le Grazie used for generating electricity by hydroelectric power plants. The Chienti river has significant economic and cultural relevance for the region, thanks to its fertile valley (Aringoli et al., 2025).

Samplings were performed at two sites of the Chienti river: Pontelatrave (x: 2363793; y: 4771352 located near the source) and Montecosaro (x: 2409404; y: 4791483; approx. 8 km upstream the river mouth). Pontelatrave site was selected due to its location in the upper Chienti basin, an area of high ecological quality and relatively low anthropogenic pressure, predominantly characterized by agricultural land use and surrounded by vegetation. In contrast, the Montecosaro site was chosen for its proximity to urban wastewater discharges, industrial, and agricultural areas, representing a downstream section with higher anthropogenic impact (Agenzia Regionale per la Protezione Ambientale delle Marche, 2021).

Meso- and MPs were collected in December 2022 and April 2023 from the river water using a Manta-net (Hydro-Bios™) with a mouth opening 30 x 15 cm, a length of the net bag of 200 cm and a mesh size of 300 microns, equipped with a mechanical flowmeter (Hydro Bios™). This equipment, anchored with a rope, was maintained in a fixed position on the water surface for 20 min to collect samples (duplicate). Recorded water volumes were in the range of 300 to 600 m³. The net was rinsed three times using river water, and the retained materials were collected into a glass jar and immediately transferred to the laboratory. A blank control was included, consisting of a glass jar filled with distilled water and kept open during the sampling and extraction process.

The collected plastic samples were pretreated using three cycles of density separation using a saturated sodium chloride solution in water. The recovered materials floating on the surface were then digested with 35% hydrogen peroxide solution to remove organic matter (Rani et al., 2023). The resulting solution was filtered on filter paper (Whatman 3MM™) and the meso-plastics were identified by visual inspection, while putative MPs particles were identified under the stereomicroscope (Olympus™ SZ61, Optika) based on the physical properties of plastic (color, size, and shape). Plastic particles were characterized by FT-IR ATR analysis, performed with a PerkinElmer FT-IR spectrometer Spectrum Two UATR, equipped with a ZnSe crystal. The measurements were conducted in the 400–4,000 cm⁻¹ range at a resolution of 2 cm⁻¹ with 4 scans and processed using the PerkinElmer™ data manager (Spectrum) (Cocci et al., 2022).

One liter of water from Chienti river was collected weekly for a month from the two sites (Pontelatrave and Montecosaro) described in Section 4.1 and from an additional site located in between (San Claudio: x: 2399783; y: 4792022). Water samples intended for the isolation of antibiotic-resistant bacteria were collected into sterile glass bottles, transported and stored at 4 °C, and processed within 6 h of collection.

River water was cultured to isolate cefotaxime-resistant coliforms and E. coli using the membrane filtration method. Fixed volumes (10–100 ml) of river water were filtered through 0.45 μm nitrocellulose membranes (Whatman™). The membranes were then transferred into Coliform Chromogenic Agar (CCA) (Thermo Fischer Scientific™) supplemented with Cefotaxime (CTX) (4 mg/L). Plates were incubated at 37 °C for 24 h, after which colonies were counted and isolated. Cefotaxime-resistant isolates were stored in 20% glycerol stocks at−80 °C for further analysis. Isolates resistant to CTX were biochemically identified using API 20 E^TM strips (Biomerieux™) for the detection of Enterobacteriaceae. Identification was performed using APIWEB software, Biomerieux. The antibiotic susceptibility profiles of CTX-resistant Enterobacteriaceae strains were determined using the disk diffusion method. Antibiotic tested were CTX 5 μg, ceftazidime (CAZ) 10 μg, gentamicin (CN) 10 μg, levofloxacin (LEV) 5 μg, sulfamethoxazole/trimethoprim (SXT) 1.25/23.75 μg, meropenem (MEM) 10 μg, amoxicillin/clavulanic acid (AMC) 20/10 μg, and tigecycline (TGC) 15 μg. After incubation at 37 °C for 18 ± 2 h, the diameter of the inhibition zones was measured and interpreted according to EUCAST guidelines.

To isolate vancomycin-resistant fecal enterococci, 100 ml of river water was filtered, and the membranes were placed on Slanetz and Bartley (SB) medium (Thermo Fischer Scientific™) supplemented with vancomycin (8 μg/ml). To confirm the identification, filters were transferred to the surface of Aesculin Azide Bile Agar (Thermo Fischer Scientific™) plates.

The simultaneous presence of MPs and ARB in freshwater may represents a combination of factors promoting the spread of antibiotic resistance (McCormick et al., 2014). In this study, we assess the level of plastic pollution and characterize the antimicrobial resistance of clinically significant bacteria in the Chienti river. Plastic pollution was studied by quantifying the plastic debris dispersed in river water and by identifying the most abundant polymer types. We have adapted a seawater samplings method involving the use of the Manta-net (Figure 1A) to recover plastic debris from river water.

The observation of the filters recovered from the last step of the extraction process, revealed the presence of meso- and MPs different in size, shape, and color (Figure 1B). A total of 99 MPs were quantified and characterized using FTIR spectroscopy. Over two sampling campaigns, 56 MPs were found at Pontelatrave and 43 were recovered at the Montecosaro site. Specifically, at Pontelatrave, 29 fragments were recovered in December 2022 with a density of 0.060 fragments/m³, and 27 fragments during the sampling in April 2023 corresponding to a density of 0.046 fragments/m³. At Montecosaro, 22 MPs were detected in December 2022 resulting in a density of 0.049 fragments/m³, while 21 were recovered in April 2023 corresponding to 0.051 fragments/m³. As a result of the two sampling campaigns, only one meso-plastic fragment was recovered in December 2022, at Pontelatrave site. The observed densities of MPs are lower compared to those reported for other rivers. For instance, the Ofanto River in Italy exhibits MPs concentrations ranging from 0.5 to 18 fragments/m3 (Campanale et al., 2020). Similarly, the Po River, Italy's largest river, displayed MP concentrations ranging from 0.29 to 3.47 fragments/m3 (Munari et al., 2021). In addition, the MP densities in the Chienti river are considerably lower than those observed in the Ticino River in northern Italy, where concentrations averaged 33 ± 20 fragments/m3. The average concentration obtained at Pontelatrave site (0.053 ± 0.007 fragment/m³) the closest to the river's source, are comparable to those obtained at the Montecosaro (0.050 ± 0.001 fragment/m³). This is somehow surprising considering that the plastic materials usually accumulate following a gradient of concentration, which increases traveling toward the river mouth (Emmerik and van Schwarz, 2020; Lebreton et al., 2017). A possible role in the mitigation of this type of pollution, could be played by the three artificial reservoirs located along the river Chienti. Reservoirs influence the transport of MPs, since the reduced flow velocity, prolonged water residence time, and enhanced sedimentation, favor their deposition and accumulation. As a result, reservoirs may function as temporary sinks, attenuating the downstream flux of plastic particles (Watkins et al., 2019; Cheng et al., 2024).

Similarly to our results, the Ticino study reported the lack of a gradient of MP concentrations moving along the river course (Winkler et al., 2022). Findings closer to the values recorded in this study, come from an analysis conducted in Rhine River in Switzerland, where MP concentrations ranged from 0.04 to 9.97 fragments/m3 (Mani and Burkhardt-Holm, 2020).

Chemical characterization of MPs by FT-IR shows that the PE was the most abundant polymer detected, accounting for over 60%. The other plastic polymers identified were 11% polypropylene (PP), 10% polyester (PES) and almost 7% polyamide (PA). Only 2 fragments of polystyrene (PS) and one fragment of ethylene vinyl acetate (EVA) were detected at Montecosaro (Figure 1C). The predominance of PE, PP, PES, and PA correlates with the widespread use of these plastic products in our daily life (Rodrigues et al., 2019).

Moreover, we studied the AMR profile of the bacteria isolated from the river water, focusing on 3GC-resistant Enterobacteriaceae and vancomycin-resistant enterococci. The water samples have been collected in June 2022 along the Chienti River, under medium flow conditions and at least 5 days after a rain event. The three water samples were analyzed in CCA supplemented with CTX 4 mg/L, and in Slanetz and Bartley medium with vancomycin 8 mg/L; then isolated colonies were biochemically identified. Out of the 94 bacterial colonies isolated, no vancomycin-resistant fecal enterococci were identified, while 66 oxidase-negative strains were detected on CCA+CTX. Specifically, the isolated enterobacteria were E. coli, Enterobacter cloacae, Klebsiella pneumoniae, K. oxytoca, Salmonella enterica, Serratia odorifera, Kluyvera spp, Citrobacter freundii, and C. koseri. In an analogous study carried out on three rivers of central Italy, a partially overlapping set of bacterial species has been identified (Piccirilli et al., 2019). In northern Italy, 3GC-resistant Enterobacteriaceae were detected from surface and groundwater sources of the Ticino River, confirming their widely pervasive presence (Piazza et al., 2025). Twenty-five strains were selected and characterized by antibiotic susceptibility test (AST). The panel of antibiotics used includes CAZ, CN, LEV, SXT, MEM, AMC, and TGC, according to EUCAST guidelines. Overall, these isolates exhibited a high percentage of resistance with the 44% (11/25) being MDR bacteria. Specifically, they showed resistances to: CTX: 100 %, CAZ: 96 %, LEV: 56 %, AMC: 52 %, SXT: 40 %. Three resistant strains to CN and one to TGC, respectively, were detected, while no resistant strains to MEM were found (Figure 2).

The findings of this study demonstrate the co-presence of plastic pollution and critical resistant bacteria in the Chienti River. Although this study did not directly assess the interaction between MPs and ARB, their simultaneous presence in the Chienti river, suggests that rivers may have a role in the ARGs transport. MPs as mobile substrates, may spread MDR bacteria and associated ARGs along the river from contaminated hotspots, altering local microbial communities, and introducing new resistance traits. Given the potential of MPs to act as vehicle of resistant pathogens, their presence in rivers systems represents a real public health concern (Ferheen et al., 2024; Yu et al., 2022). This concern is directly linked to the risk of infections caused by critical priority bacteria that may colonize these particles. Humans and animals can be exposed through recreational water activities, consumption of contaminated water or food, or direct contact with polluted environments. Therefore, the presence of MPs and 3GC-multidrug-resistant Enterobacteriaceae in river systems highlight the need to introduce standardized environmental monitoring and microbiological surveillance to understand their combined impacts on public health.