Fate and Distribution of Steroid Drugs and Conjugates in Wastewater Treatment Plants and River:Application of LC-HRMS based suspect screening
Article information
Abstract
Steroid drugs(SDs) have emerged as significant environmental pollutants due to their strong bioactivity and potential to disrupt endocrine systems. In this study, the occurrence of steroid drugs and their conjugates was investigated in wastewater treatment plants and natural river waters. A total of 57 steroid compounds, including parent drugs and conjugated forms, were screened using a suspect screening approach based on liquid chromatography-high-resolution mass spectrometry (LC-HRMS). Sampling was conducted in the Nakdong River around the Daegu and Busan regions of South Korea, as well as five wastewater treatment plants in 5 different regions.Among the screened compounds, testosterone sulfate and nandrolone acetate were identified with high confidence and were frequently detected in influent samples, exhibiting strong signal intensities. These compounds were effectively removed during the wastewater treatment processes, particularly after the bioreactor and nitrification-denitrification stages. However, no high-confidence detections were observed in river water samples, suggesting that these substances were either effectively removed or diluted to concentrations below the detection limit. This study demonstrates the effectiveness of LC-HRMS-based suspect screening for tracking steroid contaminants in aquatic environments. It also underscores the importance of monitoring not only parent steroid drugs but also their conjugated and potentially bioactive metabolites to fully assess environmental risks.
Introduction
Steroids are a broad class of biologically active compounds sharing a core structure of four fused rings,three six-membered and one five-membered.Steroids play two key roles in biological systems: as structural components of cell membranes, influencing fluidity and protein activity, and as signaling molecules, especially hormones, that regulate metabolism, immunity, and reproduction via receptor interactions. The predominant application of steroids is as steroid drugs for both human and veterinary medicine. SDs are either derived from endogenous steroid hormones or synthesized through chemical modification to enhance pharmacological activity. Owing to the potent biological activity and regulatory functions in multiple physiological and biochemical processes, steroids are widely used in clinical settings for treating inflammation, hormonal deficiencies, contraception, immunosuppression, and cancer [1,2,3 ,4,5,6].The broad therapeutic applications of steroid drugs underscore the significance in modern medicine. Following administration, steroid drugs undergo metabolic transformations in the liver, where most steroid conjugates (SCs) are formed to enhance water solubility and facilitate excretion. The conjugates are typically linked to sulfate or glucuronide groups, such as estradiol sulfate and testosterone glucuronide. [7,8]. However, conjugated forms can retain biological activity or be hydrolyzed back to the parent compounds under environmental or physiological conditions, thereby contributing to persistent environmental contamination and posing risks similar to or greater than the original compounds. [9,10,11]
SDs have been increasingly identified as contaminants of emerging concern and were reported and ubiquitously present in WWTP effluents and surface waters.[12,13,14].Likewise, the conjugated forms, such as glucuronides and sulfates are frequently found in aquatic systems, including surface water, groundwater, water sources, suspended solids, riverine and marine sediments, as well as influents and effluents of wastewater treatment plants.[15,16] SDs and SCs enter the environment through a variety of anthropogenic and natural pathways. The sources are diverse and include both point and non-point origins, each contributing differently to environmental contamination.Point sources, such as municipal wastewater treatment plants (WWTPs), pharmaceutical manufacturing facilities, and hospital effluents, where SDs and the SCs enter the environment in a concentrated form.[17.18.19].Non-point sources, are diffuse and include agricultural runoff from livestock operations treated with veterinary steroids, aquaculture effluents, and urban stormwater carrying residues from human excretion.[20,21,22,23]. In aquatic environments, microorganisms decompose steroids and their conjugates through enzymatic action, and photochemical reactions and redox reactions triggered by light also promote their structural destruction.[24,25] Additionally, steroids can undergo physical processes such as adsorption, deposition, dilution and volatilization in water bodies, affecting their migration and bioavailability.[26] Likewise, conjugated steroids may be converted into active free states by hydrolases in the environment, leading to secondary pollution. These complex degradation and transformation processes jointly determine the fate of steroids in the environment and their ecological risks. Current research on the degradation of steroid drugs and their conjugates mainly focuses on the following scenarios: wastewater treatment plant, sewers, and natural environments.
The conventional activated sludge (CAS) process degrades SDs via enzymatic reactions. However, removal efficiency varies with enzyme specificity, substrate structure, and operational conditions.[27,28]In A2/O and O/A/O, anaerobic tanks mainly promote steroid degradation through microbial reduction processes and dissociation reactions under oxygen-limited conditions, while aerobic tanks enhance removal through oxidative biodegradation driven by oxygen-dependent microorganisms.[29,30,31] .And Advanced Oxidation Processes (AOPs) use strong oxidizing groups to effectively remove both natural and synthetic steroids. [32] The Membrane The membrane bioreactor (MBR), integrating membrane separation with activated sludge technology, achieved a removal efficiency of over 95% for natural steroid estrogens, outperforming the conventional activated sludge process.33 In contrast, relatively fewer studies have focused on steroid conjugates, but Some studies have shown that the dissociation priority follows the order of glucuronidated steroids > sulfated steroids.[34]
In environmental monitoring, steroid drugs are increasingly recognized as environmental pollutants, particularly as a subclass of endocrine-disrupting chemicals (EDCs).[39] The characteristics of steroid drugs as environmental pollutants can be summarized into the following three aspects:Chronic Toxicity and Transgenerational Effects.SDs can cause harmful effects when exposure is long-term, disrupting development and reproduction in organisms, particularly during sensitive stages.[40,41,42] This toxic impact not only occurs in one generation, but also affects future generations. [43]. Meanwhile, Steroid hormones exhibit exceptionally high biological activity even at extremely low environmental concentrations,often at concentrations as low as nanograms per liter. [44,45] Steroid drugs can cause additive or synergistic effects due to the similar structures. When multiple steroids coexist in the environment, they may target the same receptors, resulting in stronger endocrine disruption than individual compounds alone, thereby amplifying their overall impact.[46,47]
This study comprehensively screens and monitors the occurrence of steroid drugs (SDs) and their conjugates (SCs) in aquatic samples, particularly from five wastewater treatment plants in South Korea and rivers around the Daegu and Busan regions, using suspect screening method based on liquid chromatography–high-resolution mass spectrometry (LC-HRMS). By applying confidence levels to guide compound identification, the study addresses existing gaps in understanding the environmental distribution of these compounds and demonstrates the utility of LC-HRMS suspect screening for monitoring both parent compounds and their conjugates in aquatic environments.
Materials and Methods
Chemical and materials
Throughout the study, high-performance liquid chromatography (HPLC) grade solvents were consistently used. DI water was obtained from J.T. Baker, ethanol from Thermo Fisher (Waltham, MA, USA), and methanol from Honeywell. For cleaning purposes, ultrapure water was used.. All vials of samples were stored at −4 °C until analysis. Cellulose acetate membrane filters (0.22 μm) were supplied by Advantece®.
Sample sites and collection
Wastewater treatment plant
Five regions were selected for this study,including Okjeong (OJ), Jeungpyeong (JP), Poseung (PS) ,Noksan (NS) and Onsan (OS) .Sampling was conducted three times in 2024: First sampling: March 28 to April 5; Second sampling: July 23 to 30; Third sampling: September 11 to 26.Samples were collected from various treatment process at each WWTPs. Grab samples of 1 L water were collected in pre-cleaned brown glass bottles, which had been rinsed twice with ultrapure water and methanol to eliminate potential contaminants and to prevent photodegradation of light-sensitive compounds.After collection, the bottles were tightly sealed, transported to the laboratory under dark and low temperature conditions, and stored at 4 °C refrigerator until analysis to preserve analyte integrity.
Nakdong river
As the largest river and major source of drinking water in South Korea, the Nakdong River flows through several densely populated and highly developed cities, before discharging into the sea at its estuary. Sampling sites were primarily selected along the main stream and key tributaries in Daegu,where major sewage treatment plant outfalls are located,and at the estuary area in Busan. Grab sampling was conducted four times in March 2025 at one-week intervals. For each event, 2 L of river water was collected in brown glass bottles, transported to the laboratory under cooled conditions (4 °C), and processed within 24 hours.
Pretreatment processing
Pretreatment for sewage treatment plant samples
Sample pretreatment was conducted prior to analysis according to a previously established method. [48] Each 1L WWTP sample were filtered through 0.45 μm glass fiber filters and subsequently spiked with 80 μL of citric acid buffer.Solid-phase extraction (SPE) was carried out using multi-layer cartridges packed with a mixture of Oasis HLB (200 mg), Strata-XCW (100 mg), Strata-XAW (100 mg), and Isolute ENV+ (150 mg). Prior to sample loading, the cartridges were conditioned sequentially with 10 mL of methanol and 10 mL of deionized water. The samples were subsequently loaded onto the cartridges at a flow rate of 5–10 mL/min over approximately 1.5 hours to ensure sufficient adsorption.After drying the cartridges with nitrogen gas, the analytes were sequentially eluted with 6 mL of an alkaline solution (ethyl acetate/methanol, 50:50, containing 0.5% ammonia) followed by 3 mL of an acidic solution (ethyl acetate/methanol, 50:50, containing 1.7% formic acid). The combined eluates were transferred into glass tubes and evaporated to dryness under a gentle stream of nitrogen at 35 °C.The dry residues were reconstituted in 100 μL of methanol and 900 μL of water, filtered through a 0.2 μm cellulose acetate membrane filter, and transferred into 1.5ml glass vials. The filtered extracts were stored in a freezer until further analysis by LC-HRMS.
Pretreatment for Nakdong river samples
The same extraction procedure was applied to river water samples, with the only difference being that the sample volume was adjusted to 2 L after filtration through 0.45 μm glass fiber filters. All subsequent steps, including buffer addition, SPE cartridge composition and conditioning, sample loading, elution, evaporation, reconstitution, filtration, and storage, were conducted following the exact same protocol as described above for WWTPs samples.
Instrumental Information
Instrument Information and setting
The LC-HRMS system consisted of a Q Exactive Plus Orbitrap mass spectrometer (Thermo Fisher Scientific, Waltham, USA) coupled with an Ultimate 3000 ultrahigh performance liquid chromatography (UHPLC) system, and was equipped with a heated electrospray ionization (HESI) source. This setup enabled high-resolution and accurate-mass detection, which is essential for analyzing trace levels of steroid pollutants in environmental matrices.To enhance structural elucidation and identification of both known and suspected compounds, the system operated in data-dependent acquisition (DDA) mode. In this mode, the instrument automatically selected precursor ions for fragmentation based on predefined criteria, allowing for efficient collection of MS/MS spectra for suspect screening purposes.The analysis was carried out in both positive and negative ion modes, depending on the ionization behavior of different analytes, thereby improving the overall detection coverage and sensitivity across a wide range of steroidal substances. All LC-HRMS operating conditions, including chromatographic and mass spectrometric parameters,such as mobile phases, flow rate, ion source settings, scan range, and injection volume are provided in Table SI-A5.
Quality Assurance and Quality Control
Before sample analysis, strict quality assurance and control procedures were implemented throughout the entire pretreatment and instrumental analysis process to minimize potential contamination and ensure data reliability. Details are described in the Text S1-A1.
Suspect screening
Suspect screening processing
Suspect screening is a widely adopted approach in environmental chemistry that enables the efficient detection and identification of compounds with prior molecular information available, without the immediate need for quantitative analysis using reference standards.Data interpretation and suspect screening were performed using TraceFinder 4.1 and Xcalibur 4.1 software (Thermo Fisher Scientific, Waltham, MA, USA), which enabled effective browsing, identification, and fragment ion analysis based on accurate mass data.Suspected screening analysis in this study was carried out using TraceFinder 4.1 software (Thermo Fisher Scientific), which integrates both compound identification and quantification workflows. The analysis was based on a data-dependent MS/MS (DDA) acquisition strategy, which was performed on a precompiled list of suspected steroid drugs. This approach was combined with liquid chromatography-high-resolution mass spectrometry (LC-HRMS) in full-scan mode, allowing for high-sensitivity detection and structure elucidation of unknown or partially characterized compounds in complex environmental matrices.In the screening process, the LC-HRMS platform first conducted a full MS scan to detect all ionizable compounds within the m/z range of interest. Upon detection of ions meeting preset criteria (e.g., intensity threshold and charge state), data-dependent MS2 acquisition was triggered, generating fragmentation spectra that were used to confirm compound identity. Suspicious peaks were preliminarily identified through a series of screening filters, including accurate mass matching, isotope pattern recognition, retention time consistency, and MS/MS spectral library matching. These steps significantly improved the confidence level of annotation and minimized the rate of false positives. The detailed procedure is shown in Figure SI-A3.
Trace finder 4.1 software setting information
The software settings and workflow parameters related to the use of TraceFinder 4.1 for conducting suspect screening analysis are comprehensively summarized in Table SI-A6. This table outlines key operational details, including data acquisition settings, compound identification criteria, and matching thresholds used for peak detection and annotation. Parameters such as mass accuracy tolerance, isotope pattern matching rules, retention time windows, MS/MS library match scores, and screening filters are included to provide transparency and reproducibility in the suspect screening process. These settings were optimized to ensure high-confidence identification of steroidal compounds in environmental water samples and to minimize the occurrence of false positives during data interpretation.
Confidence level
The identification results in the suspect screening process are categorized into five levels of confidence, following commonly accepted criteria in non-target and suspect screening methodologies. [49]Level 1:Confirmed identification;Level 2:Probable structure;Level 3:Tentative candidate; Level 4:Exact mass (m/z) and isotopic pattern match; Level 5:Mass detection only. These levels reflect the degree of certainty in compound identification based on available analytical evidence. The detailed of classification and requirements of confidence levels is shown in Figure SI-A4.
Results and Discussion
Steroid drugs in wastewater treatment plants via suspect screening
WWTPs
Two compounds were identified with high confidence (Level 2) based on accurate mass measurements, isotope pattern matching, and comparison of fragmentation spectra with reference libraries such as mzCloud and MassBank. In positive ion mode, nandrolone acetate was reliably identified, a commonly used anabolic steroid, suggesting possible anthropogenic or veterinary sources such as livestock operations or pharmaceutical discharge. In negative ion mode, testosterone sulfate, a typical phase II conjugate metabolite, was also confirmed. Detailed identification results with high-confidence compound information listed in Table 1, and MS/MS spectral comparisons shown in Figure 2.
Tentative identification of (a) nandrolone acetate;(b)testosterone sulfate with high resolved accurate mass, isotopic pattern and MS/MS spectrum evaluation with mzCloud database.
In order to comprehensively characterize the detection frequency of candidate compounds in the collected wastewater treatment plant influent samples, as well as their signal intensity (expressed as peak area), this study constructed a visual chart, as shown in Figure 3. Among them, there are 15 positive samples and 5 negative samples from wastewater treatment plants’ influents.The chart intuitively illustrates the spatial distribution characteristics and relative abundance of candidate compounds across all influent samples from the investigated sewage treatment plants. Specifically, the chart integrates two key parameters for each candidate compound with a confidence level of 4 or higher: (1) the occurrence frequency(OF) of the compound across sampling points. (2) the corresponding peak area observed, which serves as a proxy for its relative concentration or signal strength in the sample matrix.
Distribution and Abundance Profile of Candidate Steroid Compounds in Influent Samples from Multiple Sewage Treatment Plants.(a)postive;(b)negative.(OF: Occurrence Frequency)
This dual-indicator approach not only provides a comprehensive visual summary of the occurrence and abundance of these compounds, but also facilitates the identification of pollution hotspots and possible point sources. Importantly, two confidently identified steroid compounds (confidence level 2), namely nandrolone acetate and testosterone sulfate, are highlighted in the figure to underscore their significance and to distinguish them from other lower-confidence candidates.This visualization method serves as a powerful tool for rapid assessment of the environmental behavior of suspect steroid pollutants. It also supports source tracking and priority ranking of contaminants . Ultimately, the outcomes of this analysis establish a solid foundation for the subsequent development of a pollutant scoring and prioritization framework, as well as for comparing the effectiveness of different wastewater treatment processes in removing or degrading these bioactive compounds.
Preliminary analysis of the removal efficiency of steroid drugs and their conjugates by different wastewater treatment technologies
After preliminary screening of steroid drugs in water samples at the inlet of the sewage treatment plant, this study further expanded the sampling scope to all major wastewater treatment units. Due to the diversity and transformation complexity of steroid substances in the environment, the targeted screening method combined with standard substances often cannot fully cover all possible compounds, especially in the absence of standard substances or the unknown structure of the transformation products.Therefore, a suspect screening strategy based on high-resolution mass spectrometry (LC-HRMS) was adopted to screen from the perspective of suspect analysis through multidimensional features such as accurate mass matching, isotope distribution, and fragment ion information. Although such methods are currently unable to accurately quantify steroid drugs and their metabolites, their sensitivity and structural recognition capabilities can effectively assist in revealing the relative abundance trend of steroid pollutants in each treatment unit.
By comparing the detection of suspected steroid substances in water samples at different treatment stages, the removal efficiency and migration and transformation behavior of these pollutants in the sewage treatment process can be reflected to a certain extent, providing basic data support for evaluating the effectiveness of wastewater treatment plants in removing steroid drugs and conjugates. The peak areas of nandrolone acetate and testosterone sulfate, two substances identified with high confidence, after each wastewater treatment stage are shown in Figure 4. The treatment stages are indicated according to Table SI-A2, and the results from all three sampling events are presented in full.
In treated wastewater samples from different processing stages, both steroid drugs and their conjugates exhibited a reduction in peak area by more than three orders of magnitude after passing through the biological reactor and the nitrification–denitrification processes. This substantial decrease highlights the pivotal role of these treatment stages in removing steroidal compounds via biodegradation and nitrogen transformation pathways. In the biological reactor, diverse microbial communities enzymatically degrade complex steroid structures, breaking them down into simpler, less bioactive molecules. During nitrification, ammonia is oxidized to nitrate under aerobic conditions, while denitrification reduces nitrate to nitrogen gas under anoxic conditions. These alternating redox conditions foster dynamic microbial activity, further enhancing the degradation of both parent steroid drugs and their conjugated metabolites. Consequently, the biological reactor and nitrification–denitrification processes are likely to contribute substantially to the reduction of steroidal pollution in wastewater.
Steroid drugs in Nakdong river via suspect screening
In order to comprehensively characterize the detection frequency and signal intensity (expressed as peak area) of candidate compounds in samples collected from key locations along the Nakdong River and its estuary, this study constructed a visualization chart, as shown in Figure 5. The chart includes only those compounds identified with low confidence level , ensuring a focus on substances with moderate identification certainty. Specifically, the figure displays two primary indicators for each compound: (1) the number of detection events across different sampling sites, representing detection frequency, and (2) the corresponding peak area observed, serving as a proxy for the compound’s relative abundance in the aquatic environment.
The results demonstrated that, due to the extreme dilution of steroid compounds in river water, the available evidence was insufficient to support confident identification based solely on suspect screening, highlighting the need for complementary targeted analysis or sample enrichment techniques.
Conclusions
In this study, a suspect screening approach based on high-resolution mass spectrometry (LC-HRMS) was applied to investigate the occurrence and distribution of steroids drugs in wastewater samples collected from the influents and different treatment stages of municipal sewage treatment plants. A total of 2 high confidence level steroids drugs and conjugates features were tentatively identified from 57 suspect compounds.Notably, testosterone sulfate and nandrolone acetate are detected with relatively high frequency and signal intensities in influent samples, highlighting their potential persistence and relevance as emerging contaminants. Among them, testosterone sulfate ranked highest, reflecting its widespread presence, possibly derived from human excretion. In contrast, nandrolone acetate—an anabolic steroid not endogenously produced—suggests potential anthropogenic or veterinary inputs, with origins likely associated with pharmaceutical use, illegal enhancement practices, or livestock applications.Comparative analysis of different wastewater treatment technologies revealed that steroid drugs and conjugates are partially degraded after passing through biological reactors and the nitrification–denitrification stages. This finding highlights the significant contribution of biological treatment in the initial removal of these micropollutants. It suggests that microbial activity and nitrification–denitrification processes play a crucial role in transforming or breaking down steroidal substances prior to further treatment steps, thereby reducing their environmental load and potential ecological risks at an early stage of the treatment process.It is worth noting that no compounds were identified with high confidence (levels 1–2) in river water samples, indicating either effective removal during wastewater treatment or a significant dilution effect in the receiving water body. This also suggests that the environmental concentration of these steroidal compounds may fall below the current detection threshold outside the sewage system.Although quantification was limited , the suspect screening results provided a valuable overview of the relative abundance trends of steroidal contaminants throughout the treatment process. The findings also demonstrate the utility of combining accurate mass screening, isotope pattern matching, and MS/MS verification for environmental monitoring of semi-persistent pharmaceutical residues. Further targeted quantification and ecotoxicological assessment are recommended to better evaluate the risks associated with these compounds in aquatic environments.
Notes
Acknowledgement
This research was supported by Changwon National University in 2025~2026.
Conflict of interest
The author declare there is no conflict of interest in this study. The author certifies that they have no known competing financial interests or personal ties that could have seemed to affect the work disclosed in this paper.
CRediT author statement
Longzihao Ding:Software,Validation Writing - original draft, Data curation. Chari Jeong:Data curation. Kailin Shen: Software .Junho Jeon: Conceptualization, Supervision, Validation, Writing - review & editing.
Supplementary Material
Add short descriptions of supplementary material. This material is available online at www.eaht.org.
