Physicochemical and ecotoxicological approaches for Moknine Continental Sebkha in Tunisia

Contamination of water resources has reached an alarming level in many developing countries. Human activities, particularly those related to coastal development and industrialization, continuously introduce toxic contaminants and increase the inputs of nutrients (cultural eutrophication) into the aquatic environment, threatening the living organisms and affecting food security as well as drinking water quality. In this context, the Tunisian coast is under increasing pressure due to socio-economic development and inefficient management of coastal environment and associated waste. Traditionally, the environmental impact of the anthropogenic pressures on aquatic ecosystems has been mainly assessed by monitoring consistently general water quality parameters such as physicochemical (e.g., heavy metal, organic matter, hydrocarbon contents, etc.) and microbiological-related parameters (e.g., faecal indicators of water contamination) (Ilter Turkdogan Aydinol et al. 2012; Riani et al. 2014; Zrelli et al. 2018). Although the physicochemical approach provides insight into the overall pollutant load of a water matrix, it may fail to make hazard investigations and predictions of possible toxic potentials and to understand the complex interactions between structurally different chemicals and their transformation products as well as their chemical bioavailability occurring in water samples (Tavakoly Sany et al. 2015). Therefore, physicochemical approaches should be combined with toxicological bio-tools in order to deeply evaluate the overall potency of the micropollutant burden and to identify the main pathways of their influence in multi-stressed water bodies (Anderson & Phillips 2016).

In this context, ecotoxicological tests with species from different trophic links in the food chain may be used to biomonitor the water quality as they offer more knowledge of the complexity of various environmental factors and the chemicals dynamic in water matrices and they may help to understand the hazards of natural stressors and of emerging pollutant mixtures considering the combined interactions between stressors (Dummee et al. 2012; Serpa et al. 2014; De Souza Celente et al. 2020). In addition, standardized toxicity bioassays help to target sites with environmental vulnerability without any preknowledge of what contaminants might be present. To date, different species of bacteria, fish, crustaceans, algae and plants have been frequently used for this purpose, namely for marine and estuarine water quality assessments (Anderson & Phillips 2016). In fact, the luminescence inhibition biotest with marine photobacterium Vibrio fischeri has often been applied as a simple, quick and reliable biotool for evaluating the acute toxicity of numerous chemicals, surface water and wastewater (Lechuga et al. 2016; Methneni et al. 2021a, 2021b).

A unicellular freshwater algal bioassay (for example the use of Selenastrum capricornutum, renamed as Pseudokirchneriella subcapitata), which represents a significant primary food producer in aquatic environment, had also been applied in water pollution biomonitoring (Silva et al. 2009).

Among several bioassays using higher plants, the Lepidium sativum phytotoxicity test is frequently used for testing toxicity of chemical compounds (Favier et al. 2019) or water matrices (Žaltauskaitė & Čypaitė 2008; Methneni et al. 2021b) due to their fast seed germination and growth (roots and shoots), low cost and sensitivity to low pollutant levels (Favier et al. 2019). Thus, phytotoxicity bioassays constitute an essential element of ecotoxicological investigation because it represents the action of a toxic chemical at the first interface of the developing and growing plant (seed) and its environment (Lyu et al. 2018).

The central Moknine Continental Sebkha (MCS) is one of the most important resources of water and biodiversity. However, it is becoming vulnerable and experiencing noticeable deterioration of ecosystem health as it continuously receives massive volumes of treated or partially treated effluents dumped by a nearby wastewater treatment plant (WWTP) and by industrial sectors such as dairy, medical and fish industries, in addition to agricultural drainage water (Archiplan-Dgat 2019; Jebara et al. 2021c). Many publications have reported the spatial and seasonal distribution and sources of phthalate acid esters (PAEs), non-phthalate plasticizers (NPPs), mercury (Hg), pharmaceutically active compounds, paraben preservatives and UV filters in the coastal area of Tunisia (Afsa et al. 2020; Jebara et al. 2021a, 2021b; Fenni et al. 2022). The occurrence of these toxic chemicals in aquatic environments can adversely affect the aquatic ecosystem health and the food web equilibrium.

Therefore, the main goals of this research study are to (i) biomonitor the water quality of the MCS coast through the analysis of physicochemical properties combined with an ecotoxicological screening using a battery of bioassays with multiple trophic levels test systems (V. fischeri, S. capricornutum, and L. sativum), (ii) determine whether ecotoxicity is related to different sampling sites, and to (iii) investigate the contribution of some contaminants to the ecotoxicity of the water samples. To date no attempt has been made to assess the ecotoxicological profile and the environmental scenario of this area. This research study will contribute to a debate on future strategies for biomonitoring and sustainable management of pollution hotspots in water ecosystems MCS.

Sampling was carried out monthly (at the first week of each month) for 9 months (from March 2022 to November 2022) using the composite sampling methodology. Once collected, water samples (10 L) were immediately transported to the laboratory by cooler boxes, prior to measuring the physicochemical properties, and/or kept at −20 °C until toxicological testing. For toxicity assessment, water samples representing each site were mixed.

Physicochemical indicators of water quality including chemical oxygen demand (COD), biochemical oxygen demand (BOD), total organic carbon (TOC), total suspended solids (TSS), nitrate (⁠⁠) and adsorbable organic halogens (AOX) were measured using a portable UV analyzer (Pastel UV, Secomam, Alès, France). A turbidimeter AQUALITIC^® (Dortmund, Germany) was used to measure turbidity parameters and conductivity was determined using conductimeter WTW 315i apparatus. The pH measurements of water samples were carried out using a WTW pH meter.

The procedure for carrying out the bioluminescence assay was based on Microtox^® protocols (UNE-EN ISO 11348-2 2009). In this bioassay, the evaluated endpoint was the decrease in luminescence of the marine bacteria V. fischeri strain NRRL-B-11177, before and after incubation in different dilutions of seawater samples of CSM (100, 50, 25, 12.5, 6.2, and 3.1%). Bacteria were reconstituted from a lyophilized state by adding a reconstitution solution provided by Dr Lange (Dr Bruno Lange GmbH Co., Dusseldorf, Germany). Negative control samples (i.e., bacterial suspensions to which NaCl (2%) had been added instead of tested samples) were treated parallel to tested samples. The V. fischeri luminescence was measured after 15 and 30 minutes of incubation at 15 °C. Tests on all water samples were conducted in duplicate. The averages of V. fischeri bioluminescence were calculated and then used to determine the concentration capable of inhibiting half of the bacterial bioluminescence (EC₅₀), which was calculated as a previous study by Jurado et al. (2012).

S. capricornutum was used to evaluate the chronic toxicity of seawater samples according to the UNE-EN ISO 8692 (2012) standard (UNE-EN ISO 8692 2012). Six test sample dilutions (100, 50, 25, 12.5, 6.25, and 3.1%) in triplicate were prepared. S. capricornutum in the flask was adjusted to 10⁴ cells/mL by dilution with the ISO freshwater algal test medium. 25 mL aliquots of the diluted water samples were transferred to extra cells that were incubated at 23 ± 1 °C and under continuous illumination. Optical density of S. capricornutum at a wavelength of 670 nm was determined at the beginning of the bioassay and every 24 h until 72 h. The averages of S. capricornutum growth rates were determined and then used to calculate the effective concentration capable of inhibiting 50% of algal growth (EC₅₀), which was calculated as previous studies of Jurado et al. (2012) and Fernández-Serrano et al. (2014).

Phytotoxicity bioassay was performed following the method outlined by Methneni et al. (2021a). Seawater samples were diluted to the concentrations: 100, 50, 25, 12.5, 6.25, and 3.1%. Tests were carried out in three replicates. A total of 25 healthy seeds of the garden cress (L. sativum) plant were put on filter paper that was placed into Petri dishes and treated with 5 mL of the test solution. After exposure for 72 h, root and shoot lengths (cm) were measured. The inhibition of the root and shoot elongation was calculated compared to the control and was used to calculate the EC₅₀ values, which represent the concentrations capable of inhibiting 50% of the root and shoot growth. EC₅₀ values were determined using a calculation method outlined by Jurado et al. (2012) and Fernández-Serrano et al. (2014).

In this study, the short-term toxicity of seawater samples from the MCS coastal areas was measured by the Microtox^® assay. Bioluminescence is determined in V. fischeri after exposure to various concentration percentages of the tested seawater samples for 15 and 30 min. Toxic effects (bioluminescence inhibition) were not observed at all studied coastal sites (results not shown).

The chronic toxicity of the MCS coastal water samples for L. sativum is presented in Figure 3. Inhibition of L. sativum root and shoot elongation was observed in all tested seawater samples. A strong inhibition of root growth (96%) was observed after exposure to 100% of seawater concentration in S1, meanwhile, the higher inhibition of shoot growth (93%) was shown in S2.

Table 3 presents the results expressed as EC₅₀ of all performed bioassays. The inhibition of S. capricornutum growth and the inhibition of the root and shoot elongation of L. sativum were more pronounced in S5 than the other areas. The shoot growth rate of L. sativum was the most sensitive bioassay when assessing the toxicity of water samples from the MCS coast (CMS).

The Moknine Sebkha (Tunisia) is a highly saline aquatic system located in the Sahel of Tunisia. Many studies are taken in this system: sedimentology, organic geochemistry of sediment and geochemical organic and mineral of the water. All studies show that this system is oligotrophy and is subdivided into microareas governed by the topography, the climatology and the hydrology (Chairi 2018).

This environment constitutes one of the most coveted wetlands in the Sahel, undergoing different forms of degradation, but at the same time presenting potential for sustainable development. It is time to design Sebkha as a support for economic development and not a dumping ground for waste and clearance for harmful activities. This cannot succeed without the contribution of water and land planning stakeholders to find a balance between socio-economic issues and environmental objectives.

In general, the levels of physicochemical characteristics, namely organic pollution-related indicators, gradually decreased from the outer areas (S1–S5) of the MCS coastline. The spatial variation of physicochemical parameters may be explained by (1) the proximity to maritime and industrial activities, which constitute the main contributory activities to pollutant emissions and (2) the speed of movement of water masses and the atmospheric wind flow variability (Kim et al. 2020). The station of MCS (S1), sheltering one of the most important fishing and trade ports in Tunisia with its own food canning, mechanical, metallurgical and textile industries, revealed a higher pollution level as a result of pronounced anthropogenic pressures. Station S4 benefits from an active fishing port with 725 fishing units participating in 40% of the total governorate production and numerous seafood-canning industries. Station S3 has a fishing port; nevertheless, it contributes by a low total fish production (0.4%) of the governorate, as well as a much lower degree of industrialization. A ‘dilution effect’ induced by the speed of circulation of coastal currents can contribute to the low contamination degree observed in S3 and S4 stations (Jebara et al. 2021a). In addition, intertidal movements have a significant impact on the hydrodynamic pattern of existing chemical compounds in the water, a factor that can elucidate the variability of the obtained results (Gurgel et al. 2016). In regard to the Tunisian authorities (NT.09.11/1983), we observed that the majority of parameters in seawater samples exceeded the permitted concentrations in the seawater for SM, COD, BOD, turbidity.

When it occurs, the deterioration of the water body cannot be explained uniquely based on a physicochemical screening as these analyses do not give an insight into the ecological and biological effects of a water matrix. Evaluation of the toxicity of water samples can only be done by determining the effect on multiple model species. In this study, the toxicity of seawater samples from the MCS coastal areas was assessed using three organisms belonging to multiple trophic levels: the S. capricornutum, V. fischeri, and the L. sativum. Obvious toxicological effects were found on the two species: S. capricornutum and L. sativum after exposure to the seawater samples, whereas V. fisheri was the only species that was not affected at all.

The toxic effect of our tested samples against the model species was compared to that of other aquatic environments worldwide (based on the EC50 values). Our seawater samples showed effects similar to those of the surface water rivers in Morocco which have been receiving untreated industrial wastewater from textile, tanneries, canneries, oil and pottery (10–40% inhibition for S. capricornutum and absence of toxic effect for V. fischeriKoukala et al. 2004). Our seawater samples were found severely toxic to the algae S. capricornutum, higher than that from the surface water of the Portuguese river (96.6 to >100%), continuously receiving runoff water and untreated sewage from livestock (Serpa et al. 2014).

The toxic effect of our water samples on L. sativum (if we consider shoot growth rate as a study parameter) in this study was higher than that of the surface water from Khniss and Hamdoun rivers in Tunisia (37 and 42.1%), heavily contaminated by heavy metals and toxic textile dyes and showing signs of poor water quality (Methneni et al. 2021b). The different responses between the biomarkers observed in this study may be due to the different levels of sensitivity to environmental stress and the different biological levels and time scales at which these biomarkers act (Handy et al. 2003). It is worth noting that inhibition of growth of L. sativum (considering the shoot growth rate as an endpoint) was the most sensitive parameter towards tested seawater samples.

A targeted chemical screening of the MCS coastline revealed the presence of total mercury (Hg), four PAEs (DiBP, DEP, DEHP, and DBP) and one NPP (DEHT), with obvious quantity differences and under different temporal and spatial conditions (Jebara et al. 2021a, 2021b), a reason that could explain the obtained toxic effects.

The mercury has been reported to induce toxic effects to marine and freshwater algae (Microcystis aeruginosa, Nannochloropsis oculata, and Chlamydomonas reinhardtii) by inhibiting their cell growth and photosynthesis and by inducing physiological alterations (Samadani & Dewez 2018; Zamani-Ahmadmahmoodi et al. 2020; Tang et al. 2023). Even at low applied concentrations, this metal may exercise toxic effects to a wide range of plants by inducing growth retardation (Ahammad et al. 2018), inhibiting photosynthesis activity (Assad et al. 2016), generating reactive oxygen species (ROS) (Kim et al. 2017), oxidizing lipid content (Zhou et al. 2008) and inducing genotoxic responses (Azevedo et al. 2018; Sun et al. 2018).

The acute toxicity of PAEs to S. capricornutum by inhibiting their growth and reducing their pigments is well documented in the literature (Gledhill et al. 1980; Suggatt & Foote 1981; Bionomics 1984; Jonsson & Baun 2010). Suggatt & Foote (1981) have exposed the algae S. capricornutum to the DEP compound during 96 h and based on the chlorophyll a content and the cell growth inhibition rate they found EC₅₀ values of 85.6–90.3 mg/L. Bionomics (1984) measured effects on S. capricornutum upon exposure to DBP and DEP between days 6 and 10 and recorded EC₅₀ values of 7 and 132 mg/L, for both compounds, respectively.

PAEs may act as an environmental stressor for terrestrial and aquatic plants by inducing various morphological, physiological, metabolic and cellular alterations in plants. Indeed, PAEs are reported to affect seed germination, roots elongation, seedlings growth, biomass, photosynthetic pigments and metabolic activities of different plant species, namely perennial ryegrass, radish, wheat, alfalfa, oat, cucumber and onion (Ma et al. 2015; Zhang et al. 2016; Ge et al. 2020). Nevertheless, the toxicity of DEHT is rarely documented. DEHT is known to have toxic effects on rats (Barber & Topping 1995; Ball et al. 2012).

Previous studies have proven the contamination of MCS coast by pharmaceutically active compounds, UV filters and paraben preservatives (Afsa et al. 2020; Fenni et al. 2022). Excessive exposure to these compounds would have adverse toxicological effects on vertebrates, invertebrates and plants (Li & Randak 2009; Prakash & Anbumani 2021; Maia et al. 2023).

Nevertheless, mixtures of toxic chemicals present in a water matrix (even at trace concentrations) can exert joint effects on living organisms, which can be additive, synergistic or antagonistic. The contamination of MCS is undoubtedly the consequence of intense industrial activity. However, the solution does not lie in the closure of these industries, which constitute a driving force not only of the economy of the Monastir city region but also of the Tunisian economy. The solution then lies in the development of wastewater treatment techniques initially by adopting environmentally friendly processes and therefore working in harmony with sustainable development objectives. This will also avoid the growing concern of the population living very intensely around the Sebkha, especially with the bad smell and the massive death of migratory birds. The diversity in plant groups in MCS undoubtedly contributes to the diversification of habitats and landscapes. They represent a varied food source for wildlife: herbivorous or carnivorous waterbirds, domestic or wild herbivorous mammals. The existence of a few mammalian species is reported, including the rat, the jerboa, the hedgehog (Paraechinus aethiopicus) and the hare (Lepus kabilicus); species of amphibians (Rana saharica and Buffo viridis) and reptiles. The flora is also quite diverse. There were 52 species in the area. Some have an appreciable economic value: Fhragmites communis and Juncus maritimus. Others are characterized by their abundance: Arthrocnmum indicum, Solanum sodamaeum, and Salicornia arabica (Landolsi & Rejeb 2014). This diversity is today threatened by pollution by these emerging contaminants which risks disrupting the aquatic ecosystem and in certain cases the massive death of living beings.

This study clearly showed that biomonitoring based on biological indicators and markers is effective in predicting contamination and toxicity of the coast of MCS. Assessment was performed using three species representing different trophic levels, namely V. fischeri, S. capricornutum and L. sativum. It also turned out that the coast of MCS is heavily contaminated with plasticizers. A targeted chemical screening of the MCS coast previously performed revealed the presence of total mercury, four PAEs and one non-phthalate plasticizer, a fact that could explain the observed ecotoxicological effects and therefore might harm the bioma and the human health of the surrounding population. It is important to implement a physicochemical treatment of the coagulation flocculation type in each industry and at the outlet of the WWTPs, because this process makes it possible to reduce turbidity and therefore the pollutant load including plasticizers, textile dyes pharmaceutical residues, etc.

In this study, the limited volume of samples analyzed as well as the limited coverage of the study area, which only covers the coasts of Sebkha with a short sampling periodicity, leaves this study preliminary. For this, we propose in the perspective of carrying out a complete study by expanding the study areas and to vary the periods of sample collection to have a more correct overview of the water quality and the risks incurred.

Dr H.B.M. thanks the Spanish government for his Maria Zambrano grant.

None declared.

A.B. did data analysis and investigation; wrote the original draft; K.B. did the investigation; M.S. and A.B. did the statistical analysis and guided the study; H.B.M. conceptualised, supervised, and validated the study.

Data cannot be made publicly available; readers should contact the corresponding author for details.

The authors declare there is no conflict.