Toxicant Responses and Culturing Characteristics of Long‐Term Laboratory‐Reared and Field Populations of Ceriodaphnia dubia

Ceriodaphnia dubia is a standardized test organism for regulatory toxicity testing of surface waters and commercial chemicals because of its simplicity to culture and responsiveness to toxicants. For testing convenience, C. dubia is often cultured for extended periods in the laboratory with little knowledge of the impact on subsequent generations. Extended laboratory rearing could impact how they respond to stressors and decrease the accuracy of test results. The present study investigated if C. dubia cultured for an extended period were representative of three recently collected field populations by comparing their culturing characteristics and sensitivities to toxicants. For culturing characteristics, the field cultures were more challenging because they had shorter body lengths, fewer neonates, and higher mortality rates than the laboratory culture. Comparative chronic toxicity tests with sodium chloride and the neonicotinoid insecticide thiamethoxam indicated that the laboratory and field organisms did not differ much in their toxicological responses but did differ in the variability of responses (percentage of coefficient of variation). The differences between the laboratory and field cultures found in the present study highlight the challenges of addressing discrepancies between laboratory and field applications in existing standardized methodologies. Environ Toxicol Chem 2024;43:159–169. © 2023 The Authors. Environmental Toxicology and Chemistry published by Wiley Periodicals LLC on behalf of SETAC.


INTRODUCTION
Standard toxicological methods, developed by organizations such as the US Environmental Protection Agency (USEPA) and the International Organization for Standardization, are widely used in National Pollutant Discharge Elimination System Whole Effluent Toxicity testing, instream assessments, as well as environmental risk assessments and management (reviewed in Connors et al., 2022).These methods typically specify the species to be used for testing based on organismal characteristics.A group of organisms that is used extensively in toxicological studies is cladocerans, one of the main groups of freshwater zooplankton.The preference for the use of cladocerans in aquatic toxicology is due to their small size, high sensitivity to a wide range of toxicants, clonal reproduction via cyclic parthenogenesis, rapid life cycle with many broods of neonates, wide geographic distribution, and cultures that are relatively easy to maintain in the laboratory (Freitas & Rocha, 2011;Sarma & Nandini, 2006).These characteristics have led to standardized toxicological methods being developed with cladocerans, resulting in them being used frequently for regulatory purposes (International Organization for Standardization, 2017;USEPA, 2002aUSEPA, , 2002b)).
The three USEPA-recommended cladocerans for freshwater toxicity testing are Ceriodaphnia dubia, Daphnia magna, and Daphnia pulex (USEPA, 2002a(USEPA, , 2002b)).Of these species, C. dubia are the smallest, with a diameter of <1 mm, followed by D. pulex, and then D. magna as the largest at approximately 5 mm (USEPA, 2002b).The larger size of D. magna and D. pulex means that it takes them considerably longer to grow, mature, and reproduce than C. dubia.Methods for chronic toxicity testing demonstrate this, with test duration ranging from 7 days for C. dubia to 21 days for D. magna (USEPA, 2002a).In addition, it has been shown that D. magna and C. dubia overall are equisensitive in both chronic and acute testing to a wide range of chemicals, including metals, organics, pesticides, and salts (Connors et al., 2022).Other references suggest that smaller cladocerans can also be more sensitive to metals and cholinesterase-inhibitor insecticides than their larger counterparts (Hayasaka et al., 2012;Keithly et al., 2004).This makes C. dubia a logical choice for most toxicity testing because they provide similar or slightly more protective results while using less time and resources.
Toxicological works published in North America have been found to primarily utilize C. dubia, while European works continue to predominantly use D. magna (Sarma & Nandini, 2006).C. dubia also has the widest geographic distribution of the USEPA-recommended cladocerans because it is found in Europe, Asia, and North America, making it a more geographically diverse representative of freshwater invertebrates (USEPA, 2002b).

Laboratory versus field organisms
Standardized laboratory methods allow comparability between tests and purposefully minimize confounding factors but often oversimplify the complexity of natural systems.Validating a connection between laboratory and field applications is essential for quality research in environmental risk assessment (Vignati et al., 2007).One approach to accomplishing this is to demonstrate that laboratory test organisms are representative of those found in natural systems in their sensitivities to toxicants.
Organisms used in toxicity testing are often cultured in the laboratory for years to decades without the natural stressors that native organisms experience.Laboratories do typically replenish their original cultures periodically from suppliers or other facilities but rarely start new cultures from the field due to the time requirement and difficulty.Some of the stressors native organisms experience that laboratory cultures do not include seasonal temperature fluctuations, predation, difficulty finding food, and exposure to multiple toxicants.The lack of stressors found in a laboratory environment can change how organisms respond to stressors over time.
Comparisons of the sensitivities of laboratory and field organisms to toxicants have also been an issue.Some studies have reported no difference in the sensitivity of laboratory and field cultures.For example, Bossuyt and Janssen (2005) found no significant difference in the acute copper sensitivity between a laboratory culture of C. dubia and that of several Ceriodaphnia species that were collected from the field.Another study found that there was similar variation in sensitivity values between laboratory and field cultures of D. magna to lambda-cyhalothrin and cadmium (Barata et al., 2002).
Despite the results of these papers, most literature in this area indicates that there are sensitivity differences between laboratory cultures of cladoceran species and field organisms.Laboratory D. magna cultures were found to be less sensitive to copper than some of the field populations, indicating that D. magna can adapt to tolerate certain levels of copper (Bossuyt & Janssen, 2005).Laboratory clones of D. magna and C. dubia were both found to be less sensitive than fieldcollected cladoceran species to acute zinc exposure (Muyssen et al., 2005).Another study examined the differences in responses between one field culture and two laboratory cultures of D. magna to cadmium and ethyl parathion (Barata et al., 2000).The results indicated that the field population had more genetic variation, allowing for more phenotypic plasticity in their response to toxicants than laboratory cultures.These studies suggest that genetically isolated cultures of cladocerans are not necessarily the best indicators of the impacts of toxicants on freshwater invertebrates.
Variation in response between laboratory cultures and field organisms is seen in other common test organisms as well.Field cultures of Hyalella azteca had more variation in their response to toxicants, and this variation was directly linked to both agricultural and urban land use activities near the sites where the field cultures were collected (Clark et al., 2015).The present study also found that field organisms were as much as two orders of magnitude less sensitive to pyrethroids than laboratory-cultured organisms.Tigriopus, a genus of marine copepods commonly used in toxicological studies, became more sensitive to tributyltin oxide and less sensitive to copper over time as they were continually cultured.This indicates that sensitivity to chemicals can change depending on the toxicant under extended laboratory culturing (Sun et al., 2018).
Based on previous research, it appears that there are potential effects of extended laboratory culturing on commonly used toxicological test organisms.Culturing organisms in the laboratory for extended periods allows them to adapt to favorable conditions and could impact the accuracy of toxicological test results calculated in the laboratory.This is of particular concern for species such as C. dubia that are continually cultured for extended periods of time.The aim of the present study was therefore to determine if a long-term laboratory-reared C. dubia culture was representative of field organisms in their culturing characteristics and response to toxicants.

Starting field cultures
During the summer of 2021, live zooplankton were collected from three lakes in northeast Arkansas (USA), Lakes Charles (36.072614, −91.145804), Frierson (35.973607, −90.722621), and Hogue (35.581414, −90.960318), to start laboratory cultures of field organisms at the Arkansas State University Ecotoxicology Research Facility (ASUERF).Sampling was completed at the center and shore of each site from a kayak using 63-µm plankton tow nets.The organisms were put into 1-L Nalgene TM bottles and transported back to ASUERF for identification.Thermometers and air pumps were utilized to ensure that the temperature and dissolved oxygen remained constant to avoid unnecessary stress on the organisms.
Water samples were collected from each lake at the time of C. dubia sampling for toxicant analysis using an adapted existing method (Schaafsma et al., 2015).Four 250-ml samples were taken throughout the sampling area and collected in Nalgene TM bottles wrapped in dark plastic to protect the samples from the light.These water samples were put into a cooler with freezer packs for transport back to the laboratory.
They were then stored in a −20 °C freezer in complete darkness until processed.Samples were sent to AGQ Labs USA (Oxnard, CA) for a multi-residue screen of 576 toxicants.These water samples were run using both a gas chromatograph-tandem mass spectrometer and a liquid chromatograph-tandem mass spectrometer at a limit of quantitation of 0.010 ppm for all chemicals tested.
Organisms collected in the field were transported to ASUERF for identification.Individuals that resembled C. dubia under a microscope were initially sorted then isolated in 30-ml plastic cups of moderately hard synthetic water and allowed to reproduce.Each cup was fed 0.4 ml of freshwater green algae, Raphidocelis subcapitata, and 0.2 ml of a combined yeast-Cerophyl-trout chow (YCT) purchased daily from Aquatic Biosystems (USEPA, 2002a(USEPA, , 2002b)).The moderately hard synthetic water was made up biweekly for culturing and testing according to the USEPA method for making standard synthetic freshwater (USEPA, 2002a(USEPA, , 2002b)).
Of the organisms isolated in cups, only female parthenogenic organisms that produced more than eight offspring were dyed using a 2 to 1 mixture of chlorazol black and rose Bengal in ethanol for 48 h.After being dyed, organisms were identified based on morphological characteristics using a microscope and an image-based key for zooplankton (Aliberti et al., 2013).Special care was taken to examine the postabdominal claw of organisms to distinguish between the closely related species C. dubia and C. reticulata (USEPA, 1986).If the organism was positively identified as C. dubia, it was then mounted on a permanent slide using Euparal Mounting Medium as a reference for the start of the culture (Mugnai & Serpa-Filho, 2015).

Culturing
Offspring from a single positively identified C. dubia mother were placed in a 1-L beaker of moderately hard synthetic water with 7 ml of algae and 7 ml of YCT fed daily.Cultures were allowed to stabilize for approximately 1 month before beginning culturing on weekly changeover boards and toxicity testing.The long-term laboratory-reared ASUERF C. dubia culture, which was compared with the recently started field cultures, was started from the USEPA Environmental Monitoring and Support Laboratory culture in June 1999.All C. dubia cultures were kept under a 16:8-h light:dark growth light photoperiod, at a temperature of 22 to 25 °C, with weekly water changes and daily feeding (USEPA, 2002a).
Each C. dubia population was also kept on weekly culture boards to monitor culturing characteristics related to the survival and reproduction of the organisms as well as to provide organisms to initiate the toxicity studies.Average brood size per individual was calculated as the total number of neonates in 14 days divided by the number of broods in 14 days.Due to the high mortality rate of the field cultures, mothers that died were replaced with healthy adults from the mass cultures daily.The mortality rate of the ASUERF laboratory organisms was so low that dead mothers did not have to be replaced from the mass cultures to ensure adequate neonates for testing.

Measuring C. dubia body length
Ceriodaphnia dubia that were <24 h in age from each population were kept on weekly changeover boards with fresh food and water daily.When all organisms were 7 days old (±24 h) adult individuals were saved from each population to have their body length measured to note any physical differences among the populations.Each C. dubia was put in a single concave microscope slide with minimal water to limit movement.Each slide was then placed under a Zeiss SteREO Discovery.V8 Stereomicroscope with a camera to take images of the C. dubia.Using Zen 2.6 Imaging Software (Blue Edition), a line was drawn from the top of the eye to the apical base of the spine using the graphics tool (Agatz et al., 2015;Zeiss, 2018).

Chronic toxicity testing
Static renewal three-brood (8-day) chronic toxicity tests were completed following USEPA methods on four populations of C. dubia: three harvested field populations from the lakes mentioned and the ASUERF laboratory culture (USEPA, 2002a).Toxicity testing on the four C. dubia populations was completed in September 2021, January 2022, and September 2022 to see if culturing the field organisms changed how they responded to the toxicants over time.Testing was also attempted in May 2022, but the field cultures were not producing enough neonates to begin testing.
During each testing period, tests were conducted on each of the four populations of C. dubia using the standard USEPA reference toxicant sodium chloride (NaCl, purity 99.9%, CAS #7647-14-5; Macron Fine Chemicals TM ) with concentrations of a control of moderately hard synthetic water and 0.62, 0.89, 1.27, 1.82, and 2.60 g/L NaCl (USEPA, 2002a).Testing was also completed with the neonicotinoid insecticide thiamethoxam.A commercial brand of thiamethoxam, Centric ® 40WG (40% active ingredient, CAS #153719-23-4; Syngenta Crop Protection) was used for the present study because it is a product actively applied for pest control on cotton in northeast Arkansas.The concentrations used for testing were a moderately hard synthetic water control and 96, 137, 196, 280, and 400 mg/L Centric ® 40WG based on previous acute testing done with Centric ® 40WG on C. dubia (Rosado-Berrios, 2018).For all testing, replicates were fed 0.4 ml of algae and 0.2 ml of YCT daily, double the USEPA recommended amount, to support the field organisms that did not survive well at recommended food densities (USEPA, 2002a).

Confirmation of test concentrations
For the NaCl tests, concentrations were measured using a calibration curve on days 0, 3, and 6 based on a combination of a high-range Hach chloride test kit and American Public Health Assocoation 4500-Cl -(B) methods adapted to be read on a UV-1280 spectrophotometer (Shimadzu Corporation) at 480 nm (American Public Health Association, 2005;Hach, 2021).On days 0, 3, and 6 of the thiamethoxam tests, the water concentrations were analyzed on the spectrophotometer, along with a calibration curve, at 253 nm, which is the peak wavelength for absorbance of thiamethoxam (Fabunmi, 2019).The spectrophotometer was only able to detect thiamethoxam concentrations between 10 and 80 mg/L Centric ® 40WG (4-32 mg/L thiamethoxam), so all test concentrations were diluted into this range before measurements were taken.All test solutions had percent differences from the calibration curves of ≤10%, indicating that the test concentrations were close to the intended values.To confirm the thiamethoxam method accurately measured test concentrations, thiamethoxam solutions throughout the calibration curve range were run on a liquid chromatographtandem mass spectrometer at AGQ Labs (Oxnard, CA) and this confirmed that the spectrophotometer accurately measured test concentrations with percent difference values of ≤10%.

Statistical analysis
Comprehensive Environmental Toxicology Information System software was used to calculate toxicity test endpoints (median lethal concentration [LC50] and 25% inhibition concentration [IC25]; Tidepool Scientific, 2015).Depending on the amount of mortality, a linear regression (general linear model [GLM]), nonlinear regression, linear interpolation (inductively coupled plasma), or trimmed Spearman-Kärber method was used to generate an LC50 value.For reproduction, the IC25 endpoint estimate was calculated through linear interpolation.
Statistical comparisons of the culturing characteristics and toxicological endpoints of the C. dubia populations over time were completed with a two-way analysis of variance (ANOVA; α = 0.05) and Tukey's post hoc test.For the parameters that did not meet the parametric assumptions of normality or homogeneity of variances, two-way ANOVAs using GLMs (family = Poisson, Gaussian, quasiGaussian, or Binomial) were used to process the data.A one-way ANOVA with Tukey's post hoc test was used to compare the overall culturing characteristics and body lengths, as well as to compare how a single C. dubia population varied over time.If the assumptions for the one-way ANOVA were not met, a Kruskal-Wallis test with a Dunn's post hoc test were used to determine which values were significantly different from each other.These statistical analyses were completed in Program R (R Foundation for Statistical Computing, 2021).

Culturing characteristics
There were many differences between the culturing characteristics of the long-term laboratory-reared and field cultures of C. dubia.Average brood size, the average number of broods a mother produced in 14 days, the percentage of individuals that reached a third brood in 8 days and the average percent mortality per day on the culturing boards are presented in Table 1 for each population at culturing time points in September 2021, January 2022, May 2022, and September 2022.For each time point, weekly culture boards were maintained for 1 to 2 months to generate these results.
Overall mean brood size for the laboratory culture (10.9 ± 0.20 neonates) was significantly greater than for Lakes There was no significant difference among the three field cultures in their brood size at any time point (p = 0.222).The mean number of broods in 14 days in the laboratory population (6.94 ± 0.12 broods) was greater at all time points than the populations from Lakes Charles (0.82 ± 0.07 broods), Frierson (1.31 ± 0.07 broods), and Hogue (1.80 ± 0.14 broods; p < 0.001).In May 2022, the populations from Lakes Charles, Frierson, and Hogue had statistically significant decreases in the number of broods per 14 days (p = 0.048).
The overall percentage of C. dubia that reached a third brood in 8 days for the laboratory culture (91.1 ± 5.50%) was significantly greater than for the populations of C. dubia from Lakes Charles (24.1 ± 4.10%), Frierson (18.4 ± 2.80%), and Hogue (25.0 ± 5.00%; P < 0.001).Changes over time for all populations did not improve the model (p = 0.314).Percent mortality, defined as the percentage of C. dubia mothers that die on a culturing board daily, in the laboratory population (0.37 ± 0.11%) was lower at all time points than in the populations from Lakes Charles (8.20 ± 1.09%), Frierson (8.54 ± 0.96%), and Hogue (8.55 ± 1.05%; p < 0.001).Percent mortality increased over time in the field populations, with the May and September 2022 time points having significantly greater mortality rates than the September 2021 and January 2022 time points (p = 0.010).

Body length measurements
Overall C. dubia body length for the laboratory culture (689.20 ± 10.65 µm) was significantly longer than for the Lakes Charles (540.74 ± 9.55 µm), Frierson (562.57± 10.05 µm), and Hogue (539.49± 12.74 µm; p < 0.001) cultures.There were no significant differences in the body lengths of the C. dubia populations between Lakes Charles, Frierson, and Hogue (p = 0.080).When compared with the field populations, the laboratory culture had longer body lengths at all time points (p < 0.001; Figure 1).In addition, body length values appeared to become shorter over time in the field populations as they were continually cultured (p = 0.002).There were no significant differences among the body length populations from Lakes Charles, Frierson, and Hogue at each time point, indicating that the size of the field cultures fluctuated in similar ways (p = 0.106).

Feeding, behavior, and coloration
Some other notable differences were observed between the laboratory and field C. dubia during culturing.Based on modified USEPA methods, the laboratory culture boards at the ASUERF received 0.2 ml of algae and 0.1 ml of YCT daily, which is sufficient to support the organisms (USEPA, 2002a(USEPA, , 2002b)).The field organisms died off in mass when cultured on boards with these quantities of food, so the amount fed to these organisms was doubled to 0.4 ml of algae and 0.2 ml of YCT throughout all culturing.The culturing data from before the feeding amount was increased were excluded from the results.Increasing the density of food helped to decrease the percent mortality on the boards dramatically.The feeding for the field mass cultures was able to be maintained at the recommended USEPA quantities (USEPA, 2002a).
Another main difference observed between the field and long-term laboratory-reared C. dubia populations was in their behavior when being cultured.The laboratory culture organisms generally spent more time stationary or slowly moving around.The field cultures were much more active, even after being maintained in the laboratory for over a year, frequently darting around the vessels.This made the field cultures more challenging to manage because they were harder to transfer due to their hyperactivity.Another main difference that made the field cultures more challenging to work with was with their coloration.The laboratory C. dubia were significantly darker than the field populations (Figure 2).The field organisms commonly looked semitransparent during culturing, making them difficult to see and transfer.

Multiresidue pesticide screen
Four water samples collected from each lake, Charles, Frierson, and Hogue, were sent to AGQ Labs USA (Oxnard, CA) for a multiresidue screen of 576 pesticides.This included both legacy pesticides such as DDT, as well as those that are popular currently such as pyrethroids, neonicotinoids, and organophosphates.The results of the multiresidue screens found no samples from any of the lakes that were above the detection limit of 0.010 ppm for any of the toxicants.

Toxicity tests
The endpoints for the chronic NaCl toxicity tests are shown in Figure 3A,B.The overall LC50 (n = 3, mean ± SE) of the laboratory population (2.11 ± 0.07 g/L) was significantly greater than those of the Lakes Charles (1.65 ± 0.22 g/L), Frierson (1.51 ± 0.18 g/L), and Hogue (1.77 ± 0.17 g/L; p = 0.010) populations.Despite the differences in LC50 values among populations being significant, the overall LC50 values for all four populations were within the upper and lower confidence levels (±2SD) for the ASUERF monthly reference toxicity tests with NaCl.
The IC25 estimates (n = 3, mean ± SE) were significantly greater for the laboratory population (0.87 ± 0.18 g/L) than for the Lakes Charles (0.47 ± 0.16 g/L) and Frierson (0.40 ± 0.12 g/L; p = 0.008) populations.The IC25 values for Lake Hogue (0.59 ± 0.16 g/L) and the laboratory population did not significantly differ (p = 0.067).Similar to the LC50 values, all the IC25 values were within the upper and lower confidence levels (±2SD) for the ASUERF monthly reference toxicity tests with NaCl.This indicates that although there were significant differences between the toxicity endpoints of some of the C. dubia populations, these values were still reasonably close to the laboratory endpoints.

Mean control neonates
The mean number of control neonates per mother during testing, after males were removed, responded similarly to the culturing data, with the laboratory culture (27.8 ± 1.9 neonates) having significantly more neonates than the Lakes Charles (12.3 ± 1.5 neonates), Frierson (11.5 ± 1.4 neonates), and Hogue (Table 2; 12.0 ± 1.9 neonates; p < 0.001) cultures.Most of the chronic toxicity tests with the field populations did not meet the USEPA requirements of 60% of control organisms producing three broods with a mean total of at least 15 neonates in the 8-day test duration (USEPA, 2002a).This is because individuals in the field cultures reproduced unpredictably, often going 4 to 5 days between broods.The variability in the reproduction is also indicated by the much greater percentage of coefficient of variation of control neonates in the field populations than in the laboratory culture during testing (Table 3).

DISCUSSION
One of the primary concerns when working with standardized test organisms is whether or not they are representative of their field counterparts because connections between laboratory and field applications are essential for quality environmental research (Vignati et al., 2007).It is often more costeffective for laboratories that complete regulatory toxicity testing to continually culture standardized test organisms instead of collecting them from the field or purchasing them from vendors who also continually culture organisms.This has resulted in toxicity testing being completed on organisms that have been cultured in laboratories for years to decades with little knowledge of how this impacts test results.
The present study found that recently field-collected C. dubia were more similar to each other than to the longterm laboratory-reared culture in their culturing outcomes.Laboratory organisms had significantly more neonates and broods per individual as well as a lower mortality rate than all three field cultures.There were also differences in physical characteristics because individuals from the laboratory culture were larger, less active, darker in color, and thrived on lower densities of food than the field cultures.The similarity of the field cultures, as well as the absence of toxicants and no large differences in water quality (dissolved oxygen, pH, conductivity, alkalinity, hardness, chlorophyll a, turbidity, total suspended solids, and total and dissolved nutrients) at the field sites, indicated that nearby land use from the field sites did not impact the characteristics of the organisms collected.Despite finding distinct differences in the characteristics of the laboratory and field C. dubia cultures, the challenges associated with culturing field organisms in the laboratory using existing methods, including the use of standardized water and food with less diversity of algae and microorganisms, cannot be ignored as potentially influential factors for why the field organisms performed so poorly (Boersma & Vijverberg, 1995;Sison-Mangus et al., 2015).
The results of the chronic toxicity tests indicated that the field populations were more sensitive to NaCl, although all endpoints were within ASUERF reference testing confidence intervals.The laboratory and field populations did not differ in their sensitivities to thiamethoxam.The results for all time points indicated that the laboratory culture responded more consistently throughout, while the field cultures had more variability in response, shown by their infrequent and unpredictable reproduction as well as greater percentage of coefficient of variation values.Overall, these results indicate that long-term laboratory-reared cultures can be semirepresentative of field organisms in their direct toxicological responses, but not in terms of response variability.It is important to consider that these comparisons are limited because the field-collected cultures were stressed under laboratory culturing and therefore were at a disadvantage to any additional stressors, such as toxicant exposure.
The results of the present study indicate that long-term laboratory-reared cultures do not accurately represent the variability in the response of field populations in regulatory testing (Barata et al., 2000).This contradicts a previous study that found that D. magna laboratory clones are representative of field organisms in their responses to cadmium (Messiaen et al., 2013).However, Barata et al. (2000) found that fieldcollected D. magna had more variation in their responses as toxicant concentrations increased than in laboratory cultures.Decreased response variability in long-term laboratory-reared cultures limits the usefulness of the testing when trying to anticipate toxicity effects and extrapolate from the laboratory to the field.
The poor performance in the culturing characteristics of the field organisms has implications for their usefulness in standardized regulatory toxicity testing methods.It was challenging to get the number of neonates required to complete testing from the weekly field culture boards due to the small brood sizes, few number of broods, and high mortality rates.This presents a greater problem for acute testing because they require almost twice as many neonates for a standardized test.In addition, very few test organisms met the USEPA requirements of being from at least a third brood of at least eight neonates to ensure test organisms were from healthy mothers.The high mortality rates also meant that test boards struggled to meet the 80% control survival requirement.As well as the issues with the field cultures not meeting USEPA testing requirements, the cultures appeared to decline over time with continual culturing.The poor performance seen in the present study by the field cultures may have resulted from the field organisms being adapted to water and food sources that were not present when cultured in the laboratory under standardized methods.This makes renewing laboratory cultures with organisms collected from the field periodically an even less feasible alternative for addressing discrepancies between laboratory and field applications in environmental monitoring when utilizing existing methodologies.

Recommendations for future research
Maintaining field-collected organisms under standardized laboratory conditions has many challenges.Laboratory conditions have less diversity of food and microbes, which can negatively impact organisms (Boersma & Vijverberg, 1995;Sison-Mangus et al., 2015).Microbial diversity also has been found to increase dramatically when organisms are reintroduced into the field, which could impact how organisms respond to stressors such as toxicants (Hegg et al., 2021).In addition, laboratory experiments regularly fail to represent the complexity of environmental systems when translating from the laboratory to the field (Vignati et al., 2007).One potential solution being explored to address these issues with zooplankton is field cages.Field cages are a type of microcosm deployed in the field to study the impacts of toxicants in the natural environment.They have been found to not inhibit food flow (O'Connor et al., 2021) and have had success in previous toxicological studies examining highly agricultural sites with extensive contamination (Barata et al., 2007;Schulz, 2003).
Toxicity testing in deployed field cages can be expensive and time-consuming to complete, and there is typically no regulation of the concentrations of toxicants in the systems.A potential alternative that still accounts for some environmental variability is to run standard USEPA chronic toxicity testing methods on field and laboratory cultures with site water from where the organisms were collected.This would help address some of the questions concerning acclimation to laboratory conditions as noted in the present study.However, ignoring the time requirements and costs, the best option to get a holistic viewpoint is to run concurrent laboratory and field studies.Stoeckel et al. (2008) did this with Daphnia species and reported that laboratory testing resulted in no effect of herbicide pulses, whereas field studies did measure an effect.Completing parallel laboratory and field studies offers a comprehensive examination for researchers to better understand what variables are likely causing the given results.
The field of ecotoxicogenomics is rapidly growing and is useful for identifying modes of action, increasing the sensitivity of biomarkers, and comparing toxicological endpoints between species (Kim et al., 2015;Merrick & Bruno, 2004;Shaw et al., 2008).Integrating genomics and toxicology is particularly useful for studying sublethal concentrations of toxicants that are often more environmentally realistic through changes in gene expression (Shaw et al., 2008).For example, DNA microarrays have been used to investigate Daphnia response to cadmium exposure through gene expression (Connon et al., 2008).Utilizing genomic techniques, more detailed comparisons of laboratory and field Daphnia could be made by looking at changes in gene expression to help researchers better understand the relevance of standardized laboratory toxicity testing in the environment.
Another potential improvement that could be made to current toxicity testing protocols with cladocerans is to use a variety of species during testing.A wide range of sensitivities to toxicants can be found among different species of cladocerans.For example, Ceriodaphnia species are consistently equisensitive or more sensitive than other common cladoceran test species and therefore are more sensitive bioindicators than the historically popular D. magna (Connors et al., 2022;Hayasaka et al., 2012;Keithly et al., 2004).It is also important to consider using regionally specific organisms that are more geographically realistic representatives than the few USEPArecommended species.Harmon et al. (2003) found that D. ambigua worked in existing USEPA methods and was a more realistic representative of invertebrates found in the low alkalinity, low pH streams of the southeastern United States.A review paper by Sarma and Nandini (2006) suggested a balance between using traditional methods so that governmental requirements are met, and that the data are publishable internationally, while also incorporating regionally specific methods, including regional test organisms.This balance would allow for a continued improvement of regionally specific toxicological methods that more accurately depict local toxicity while also continuing to contribute to international research.
It has been suggested that "the actual relevance of one's work in the context of complex environmental issues becomes more important than sticking to the standard approaches and methodologies of a specific discipline" (Vignati et al., 2007(Vignati et al., , p. 1069)).However, researchers cannot ignore standard methods when completing regulatory testing.The current USEPA methodology for regulatory toxicity testing is dated and neglects to address many issues reported in the literature in recent years.Standardized methods can be improved to increase environmental relevance by making changes that incorporate both field and laboratory methods simultaneously as well as using a combination of standardized and regionally specific test organisms.Incorporation of genomic techniques to better understand mode of actions, detoxification mechanisms, and changes in gene expression would also help to minimize the gap when trying to validate a connection between laboratory and field applications in environmental research (Shaw et al., 2008).Although these alternative approaches are more comprehensive than existing protocols, it is likely that implementing many of them would be challenging, therefore more research is needed to determine their potential usefulness and applicability prior to amending the current protocols.This is a complex problem that is only going to get more complicated as the impacts of climate change, salinization, and continued habitat degradation become more prominent in the future (Hintz et al., 2022).

CONCLUSIONS
Overall, the present study provided information regarding how representative long-term laboratory-reared cultures of C. dubia are of naturally occurring invertebrates in regulatory testing.The results indicated that there were similarities in the direct toxicological endpoints between laboratory and field organisms, but there were also some notable differences in their culturing responses and the variability of their toxicant responses.Based on this, using recently collected field organisms in existing USEPA culturing and testing methods is not a feasible approach for addressing discrepancies between laboratory and field applications in environmental research, and alternative methods should be pursued.
Supporting Information-The Supporting Information is available on the Wiley Online Library at https://doi.org/10.1002/etc.5772.

FIGURE 1 :
FIGURE 1: Mean Ceriodaphnia dubia body length (µm) in September 2021, January 2022, May 2022, and September 2022 from (A) the Arkansas State University Ecotoxicology Research Facility laboratory culture and the field cultures from Lakes (B) Charles, (C) Frierson, and (D) Hogue.Each box plot represents the minimum, lower quartile (Q1), median, upper quartile (Q3), and maximum values.Letters represent significant differences among populations.

FIGURE 2 :
FIGURE 2: Ceriodaphnia dubia from each of the four populations photographed in January 2022: (A) Arkansas State University Ecotoxicology Research Facility laboratory culture, (B) Lake Charles, (C) Lake Frierson, and (D) Lake Hogue.Individuals A and D are gravid while B and C are not gravid in the images.

FIGURE 3 :
FIGURE 3: Estimated endpoints (mean ± SE) for the chronic sodium chloride (NaCl) and thiamethoxam toxicity tests completed in September 2021, January 2022, and September 2022: (A) NaCl median lethal concentration (LC50), (B) NaCl 25% inhibition concentration (IC25), (C) thiamethoxam LC50, and (D) thiamethoxam IC25.Endpoints are shown for the Arkansas State University Ecotoxicology Research Facility laboratory culture of Ceriodaphnia dubia and the field cultures from Lakes Charles, Frierson, and Hogue.Letters represent significant differences among populations.

TABLE 2 :
Neonates (mean ± SE) from the Arkansas State University Ecotoxicology Research Facility laboratory culture and the field cultures from Lakes Charles, Frierson, and Hogue after males were removed for the chronic toxicity tests completed with NaCl and thiamethoxam in September 2021, January 2022, and September 2022 Tests were completed on Ceriodaphnia dubia populations from the Arkansas State University Ecotoxicology Research Facility laboratory culture as well as populations from Lakes Charles, Frierson, and Hogue during September 2021, January 2022, and September 2022.