Effects of the antihistamine diphenhydramine on selected aquatic organisms

Environmental Toxicology and Chemistry, Vol. 30, No. 9, pp. 2065–2072, 2011 EFFECTS OF THE ANTIHISTAMINE DIPHENHYDRAMINE ON SELECTED JASON P. BERNINGER,*y BOWEN DU,z KRISTIN A. CONNORS,y STEPHANIE A. EYTCHESON,§ MARK A. KOLKMEIER,§ KRISTA N. PROSSER,§ THEODORE W. VALENTI JR.,z C. KEVIN CHAMBLISS,z§k and BRYAN W. BROOKSyz§ yInstitute of Biomedical Studies, Baylor University, Waco, Texas, USA zInstitute of Ecological, Earth, and Environmental Sciences, Baylor University, Waco, Texas, USA §Department of Environmental Science, Center for Reservoir and Aquatic Systems Research, Baylor University, Waco, Texas, USA kDepartment of Chemistry and Biochemistry, Baylor University, Waco, Texas, USA (Submitted 2 February 2011; Returned for Revision 6 April 2011; Accepted 11 May 2011) Abstract—In recent years pharmaceuticals have been detected in aquatic systems receiving discharges of municipal and industrialeffluents. Although diphenhydramine (DPH) has been reported in water, sediment, and fish tissue, an understanding of its impacts onaquatic organisms is lacking. Diphenhydramine has multiple modes of action (MOA) targeting the histamine H1, acetylcholine (ACh),and 5-HT reuptake transporter receptors, and as such is used in hundreds of pharmaceutical formulations. The primary objective of thisstudy was to develop a baseline aquatic toxicological understanding of DPH using standard acute and subchronic methodologies withcommon aquatic plant, invertebrate, and fish models. A secondary objective was to test the utility of leveraging mammalianpharmacology information to predict aquatic toxicity thresholds. The plant model, Lemna gibba, was not adversely affected atexposures as high as 10 mg/L. In the fish model, Pimephales promelas, pH affected acute toxicity thresholds and feeding behavior wasmore sensitive (no-observed-effect concentration ¼ 2.8 mg/L) than standardized survival or growth endpoints. This response thresholdwas slightly underpredicted using a novel plasma partitioning approach and a mammalian pharmacological potency model. Interest-ingly, results from both acute mortality and subchronic reproduction studies indicated that the model aquatic invertebrate, Daphniamagna, was more sensitive to DPH than the fish model. These responses suggest that DPH may exert toxicity in Daphnia through AChand histamine MOAs. The D. magna reproduction no-observed-effect concentration of 0.8 mg/L is environmentally relevant andsuggests that additional studies of more potent antihistamines and antihistamine mixtures are warranted. Environ. Toxicol. Chem.
2011;30:2065–2072. # 2011 SETAC of drug classes are fairly well characterized, such as hormones, Pharmaceuticals and personal care products (PPCPs) are analgesics, antidepressants, beta blockers, and antibiotics [5].
found in most aquatic systems that receive large amounts of The problem now becomes identifying which of the hundreds of municipal effluent discharges, especially in areas where effluent active pharmaceutical ingredients (APIs) should be the focus of makes up the majority of water entering the receiving system [1]. Although PPCPs have likely been present in the environ- Beyond the need for a harmonized hazard prioritization ment at low concentrations for some time, it is only over the last approach that incorporates both effects and exposure elements 20 years that advances in analytical techniques have allowed [3,4], the most obvious need for analysis are those drugs that scientists to detect them [2]. Pharmaceuticals and personal care have been identified in field studies. One drug in particular, the products are typically present at low levels (<1 mg/L), which antihistamine diphenhydramine (DPH), has been specifically historically represent concentrations of minimal concern for identified in several major environmental compartments (water, most environmental contaminants. However, pharmaceuticals sediment, tissue). In streams receiving significant discharges of are biologically active molecules developed to have specific treated municipal effluent, DPH has been detected in the water effects at low concentrations. Although substantial work has at concentrations ranging from 0.01 to 0.10 mg/L [6,7]. In the examined potential PPCP exposure, comparatively less work sediment, DPH concentrations were much higher (20–50 mg/ has been done on understanding the adverse effects to aquatic kg) [7], two and three orders of magnitude higher than asso- life. Assessing the ecotoxicological impacts of these PPCPs is ciated water concentrations. Perhaps most important, DPH has one of the primary needs identified by several authors [3,4] in been found in the tissues of fish. Ramirez et al. [8] found DPH in addition to the U.S. Environmental Protection Agency the muscle tissue of fish living downstream of a North Texas (U.S. EPA) white paper on PPCPs (http://www.epa.gov/water- municipal effluent outflow at a mean concentration of approx- science/criteria/library/sab-emergingconcerns.pdf). In fact, the imately 1 mg/kg. Furthermore, a U.S. EPA pilot study, con- scientific literature has few examples of well-characterized ducted by the same group, found DPH in the muscle and liver ecotoxicological effects of drugs, and of the available informa- (1–10 mg/kg) of fish residing near multiple large metropolitan tion most is limited to acute toxicity data [3,5]. Only a handful areas in the USA [2]. Another study found 0.03 to 0.08 mg/kg offree DPH, which are those molecules unbound to protein, in fishtissue just downstream of an effluent outflow [9]. Actual DPH * To whom correspondence may be addressed muscle concentrations might be as high as 0.2 to 8.0 mg/kg if the percent DPH bound to protein in fish is similar to the 86 and Published online 3 June 2011 in Wiley Online Library 99% protein binding reported in humans [10,11].
The quantification of DPH in surface waters may be partially barrier [15]. In humans it has both antihistamine and sedative explained because it is fairly stable in the environment [12], MOAs, which are reflected in the over-the-counter formulations although, like many drugs, it is subject to photodegradation that function either to reduce allergic reactions and motion [13]. In general, antihistamines, and likely DPH, are removed sickness or serve as sleep aids. Table 1 summarizes the general poorly through most wastewater processes [14]. With 2 to 15% physical, pharmacokinetic, and pharmacodynamics properties of DPH excreted as unmetabolized by humans, it is likely of DPH. Mechanistically, DPH targets a number of different continually discharged to receiving systems, resulting in poten- receptors, although its primary target is the H1 histamine tial life-cycle exposures, particularly in effluent-dominated receptor [19]. Histamine, released from mast cells (a component streams [1]. An additional influx of DPH may come from the of mammalian innate immune system) in response to an sewage treatment process where polar metabolites (e.g., diphen- allergic trigger, targets the H1 receptors in the smooth muscles hydramine N-glucuronide [11]) are cleaved back to the parent in the vasculature causing them to then dilate. This reaction compound, although this has not been studied directly [15].
allows blood and other immune cells to move into the affected Although studies have seldom examined seasonal differences in area, causing the swelling and redness associated with an environmental exposures, it is possible that DPH usage, and allergic reaction. This same mechanism is responsible for small consequently regional environmental loading, increases season- localized reaction and larger systemic responses (e.g., anaphy- ally to coincide with seasonal allergy responses in human lactic shock). Diphenhydramine competitively binds the H1 populations. Based on the relatively high log KOW (Log P) receptors and reduces the allergic response by preventing of 3.27 (Table 1) and the empirical information summarized histamine binding and allowing smooth muscle contraction.
above, it appears likely that DPH will partition to the sediment Diphenhydramine also targets the 5-HT reuptake transporter and tissue matrices. Although DPH is present in multiple (SERT), preventing the reuptake of serotonin at the presynaptic matrices in field samples, little work has been done to character- nerve cleft [20]. In general, this MOA adds to the sedation ize its potential ecological effects [16].
response associated with DPH. Interestingly, discovery of As with many pharmaceuticals, it is possible that chronic this MOA led directly to the development of fluoxetine, aquatic risks of DPH exposure are related to the potential for the first selective serotonin reuptake inhibitor (SSRI) antide- therapeutic mechanism or mode of action (MOA) specific pressant, which exerts its therapeutic effect through the same outcomes [3,5], rather than nonspecific narcosis responses mechanism, albeit with much greater specificity [20]. Further- typically seen with industrial chemicals [17]. Understanding more, DPH acts as an anticholinergic agent by competitively mammalian pharmacological properties may help predict antagonizing the acetylcholine receptor [19]. This reaction potential effects in nontarget species based on the conservation reduces the signal sent by the acetylcholine neurotransmitter, of critical drug receptors [18]. Diphenhydramine is a first- and as such has been suggested as a remedy for organophos- generation antihistamine drug found in many common over- phate poisoning [21] and in alleviating the symptoms of the-counter formulations (Table 1) and crosses the blood–brain Table 1. Information on the antihistamine diphenhydramine (DPH), including physical, pharmacokinetics, and pharmacodynamics properties Common brands: Benadryl1 McNeil-PPC, Unisom1 Chattem, Sominex 1 GlaxoSmithKlineDrugs commonly in mixture with DPH: Ibuprofen, acetaminophen, dextromethorphan, pseudoephedrine, benzocaine, ammonium chloride, codeineUsage categories: Hypnotics and sedatives, antiemetics, antiparkinson agents, antidyskinetics, antipruritics, anti-allergic agents, histamine H1 antagonists, anesthetics – local, antitussives, anticholinergic a ACR ¼ acute to chronic ratio; ATR ¼ acute to therapeutic ratio.
Unfortunately, the consequences of DPH exposure are each concentration for each of the three replicate studies prior poorly understood in nontarget organisms. This data gap is especially disconcerting for aquatic species, as many may be Standardized chronic study. A 7-d subchronic study was exposed to DPH by way of multiple routes. Thus, the objective conducted following slightly modified U.S. EPA protocols of this study was to develop a baseline aquatic ecotoxicological [26,27,29]. Four replicates of eight concentrations and a control understanding of diphenhydramine by using a number of stand- were prepared. Treatment levels for the fish subchronic study ardized toxicity test protocols with several species. In addition, were selected based on acute response thresholds, a prediction we also explored the utility of leveraging mammalian pharma- of acute to chronic ratio (ACR) response using slope and cological information to understand thresholds of adverse intercept (0.254 and 0.788, respectively) of the regression between a mammalian margin of safety parameter (the acuteto therapeutic ratio [ATR]; Table 1) and known ACR values The following experimental conditions described apply to all studies except where noted within individual methods. Recon- and predictions of plasma concentrations in fish [22,30].
stituted hard water (RHW), formulated according to U.S. EPA Specifically, Fitzsimmons et al. [30] provided an empirical methods [23], was used as control and dilution water for relationship for nonionic chemical bioaccumulation and parti- invertebrate and fish studies. All experiments were performed tioning to fish plasma (blood:water partition coefficients; PBW), in controlled environmental chambers at 25 Æ 18C under a which was previously recommended for pharmaceutical priori- 16:8 h light:dark regime. Water quality was monitored accord- tization [22]. Here we modified another Fitzsimmons et al. [30] ing to standard methods [24]. Water quality parameters were equation (Eqn. 2), which is more appropriate for drugs with measured daily and mean (Æstandard deviation [SD]) values apparent log P values less than 3 [16], and substituted log D [31] were well within acceptability criteria [23,25,26]: dissolved at the study pH (8.5) for log P (Eqn. 3).
oxygen, 8.3 (Æ0.2) mg/L (YSI Model 55); conductivity, 580 (Æ4.6) mS /cm (YSI Model 30); alkalinity, 116 (Æ4) mg/L as CaCO3; and hardness, 172.5 (Æ3.4) mg/L as CaCO3.
The pH of each study solution was measured (Thermo BW ¼ ð100:73log D ðpH 8:5 Á 0:16Þ þ 0:84 Orion 720A pH/ISE meter) and recorded separately for each We then conceptually applied the plasma model approach test conducted. A potential for shifts exists in the ionization recommended by Huggett et al. [22], where the fish plasma state of DPH (pKa 8.9; Table 1) resulting from slight dif- concentration (FPC) is determined by multiplying the aqueous ferences in pH, which could influence toxicological responses concentration (Aq) of a drug by its PBW (Eqn. 4). The model [27]. All tests were generally conducted at higher pH (8.4– considers an effect likely to occur any time the FPC is greater 8.7) to approximate worst-case scenarios and realistic pH than the human plasma therapeutic dose (Cmax) and the point at values for many effluent dominated streams in semiarid regions FPC is considered an effect threshold (ET).
Because Cmax and PBW are constants, it is then possible to Diphenhydramine hydrochloride (CAS 147-24-0) was solve for the aqueous concentration at the effect threshold obtained from Sigma-Aldrich. Concentrations used in prelimi- (AqET) (Eqn. 5) [32], and to derive Equation 6, which predicts nary range finding testing were developed from U.S. EPA EPI the concentration of DPH in water necessary to result in plasma Suite software [28] (96-h P. promelas median lethal concen- accumulation equal to a human Cmax value: tration [LC50] ¼ 13.7 mg/L; 48-h Daphnid LC50 ¼ 1.2 mg/L),then adjusted based on preliminary results (not reported). All DPH concentrations were analytically verified following meth-ods described below.
Standardized acute studies. Standardized fathead minnow (Pimephales promelas) acute studies were conducted accordingto U.S. EPA acute toxicity protocols [23] with slight modifi- Consistent with the acute studies, experimental units were cations [27,29]. Tests were run three times each at two different 600-ml beakers filled with 500 ml of test solution and loaded nominal pH levels, 6.5 and 8.5. To ensure test concentrations with 10 <24-h-old P. promelas. This was a static renewal were the same across both pH treatments, a large volume (8 L) experiment with feeding of brine shrimp nauplii twice daily.
of each test solution at higher pH (8.5) was prepared, then The test solution was renewed daily 2 h after the morning subdivided into two 4-L aliquots, of which one was adjusted to feeding with 80 to 85% renewal [25]. Stock solutions for each the target pH 6.5 using 1.5 to 2.1 ml of 1N HCl. The higher pH exposure concentration were made fresh daily and analytically study utilized five concentrations, while the lower pH required verified. Tests were monitored daily for survival. At the com- three additional (eight total) higher concentrations to establish pletion of the 7-d study, three fish from each replicate were the LC50. At each treatment level and control, four replicates of randomly selected for a feeding trial (see Discussion). The 600-ml glass beakers were loaded with 10 larval P. promelas remaining seven fish were euthanized according to standard (<24 h old). Prior to initiating the study, fish were fed brine methods [25] and placed in aluminum weigh pans. Weigh pans shrimp nauplii but were not fed during the test. To reduce the with fish were then placed into an 808C drying oven for 48 h.
likelihood of pH drift each replicate was covered tightly with Pans and fish were allowed to come to room temperature in a parafilm for the entire 48-h test period. Survival was assessed at desiccation chamber for 1 h. Fish were then weighed on a 24 and 48 h. Samples for analytical verification were taken at Mettler Toledo Model MX5 microbalance.
Feeding behavior. Three randomly selected fish from each replicate were placed in 100-ml glass beakers filled with fresh Exposure concentrations of DPH were verified in each stock exposure media of the appropriate concentration and held for solution and all experiments by way of liquid chromatography- 24 h without food. Experiments were conducted according to tandem mass spectrometry. Instrumentation consisted of a the approach outlined in Stanley et al. [29] with the modifica- Varian model 410 autosampler, ProStar model 212 binary tions suggested by Valenti et al. [27]. The trial started by adding pumping system, and model 1200L triple quadrupole mass 40 brine shrimp nauplii to the beaker containing a single fish.
analyzer. Fifty ml of a 10-ppm solution of the isotopically Fish were given 15 min to feed, after which time the fish was labeled internal standard (DPH-d3) was added to all samples removed and the remaining nauplii counted.
and calibration standards. To ensure that analyte concentrationsfell within the calibrated range of the instrument, sample aliquots were diluted with 95:5 0.1% (v/v) aqueous formic Acute study. A 48-h static acute study for D. magna was conducted according to established U.S. EPA protocols [23]. It Analyses were carried out using a 15 cm  2.1 mm (5 mm, 80 was conducted at a single pH, 8.59 (Æ0.05). Four replicates ˚ ) Extend-C18 analytical column (Agilent Technologies) and were used for each of five concentrations and a control. Each ˚ ) guard cartridge connected in series.
replicate was loaded with five D. magna. All D. magna used A binary gradient consisting of 0.1% (v/v) formic acid in water were <24 h old and hatched within a single 4-h window. This and 100% methanol was employed to promote elution of target acute test design was performed three times. Water samples for analytes within 6 min. Additional chromatographic parameters analytical verification were taken from each concentration prior were as follows: injection volume, 10 ml; column temperature, 308C; flow rate, 350 ml/min. Analytes were ionized using Subchronic study. A 10-d D. magna subchronic toxicity test positive electrospray ionization and monitored using the fol- was performed following standard protocols [33] with slight lowing optimized MS/MS transitions: m/z 256 > 167 and modifications [34,35]. The endpoints assessed were immobili- 259 > 167 for DPH and DPH-d3, respectively. Internal standard zation (mortality) and reproduction (young per female). Daph- calibration curves were constructed using linear or quadratic nia magna used to initiate the study were <24 h old and hatched regression, as appropriate (R2 !0.998) used to determine DPH within a 4-h period. Eight concentrations and a control were concentrations in all analyzed samples. During analysis, one used in this study, with 10 replicates per treatment level. The continuing calibration verification sample was analyzed every experiment was static renewal with daily renewal. To ensure 6th injection with an acceptability criterion of Æ20%.
consistency in renewal concentrations a 4-L stock solution ofeach concentration was made at test initiation. Stock solutions were analytically verified three times: day 0, day 5, and day 8.
An a ¼ 0.05 was used in evaluating response variables for all Experimental units were 30-ml disposable plastic cups with a experiments. The LC50 values were calculated using U.S. EPA test volume of 30 ml. Each replicate was fed 0.6 ml per day of a Toxstat. The probit method was used if data met assumptions; mixture of Pseudokirchneriella subcapitata and cereal grass otherwise, the trimmed Spearman–Karber method was applied media [23,36]. Neonates were counted and removed daily [23]. The LC50 values were calculated based on analytically verified concentrations for individual test. No-observable-effectconcentration (NOECs) and lowest-observable-effect concen- trations (LOECs) were calculated using analysis of variance Diphenhydramine toxicity to a model aquatic plant was with Dunnett’s post-hoc test, as suggested by U.S. EPA pro- assessed by exposing L. gibba (a duckweed) to five concen- trations (10, 5, 2.5, 1.25, 0.63 mg/L DPH, nominal) and acontrol and measuring effects on frond number, wet weight, and growth rate after 7 d. Lemna gibba G-3 culture was obtained Analytical confirmation of DPH concentrations from the Canadian Phycological Culture Center and maintainedin Hunter’s media, as described by Brain and Solomon [37].
Table 2 provides analytical verified concentrations of DPH Prior to experimentation, plants were acclimatized to test media for each treatment level of the acute and subchronic experi- (Hunter’s media) for one week before the study was initiated.
ments with the various model organisms. For acute studies The 7-d static renewal experiments were conducted according (Table 2) concentration reported are mean (n ¼ 3; ÆSD) values to the standardized protocol outlined in Brain and Solomon [37]. After the acclimatization period, two Lemna plants, eachwith four fronds, were transferred from the acclimatized mass culture into a 250-ml Erlenmeyer flask containing 100 ml Control survival was >95% for all P. promelas tests (acute sterilized test solution. Test solutions were created through and chronic). Mean (ÆSD) pH treatment levels for the acute serial dilutions. Flasks were arranged in a randomized complete studies were 6.45 (Æ0.03) and 8.52 (Æ0.02). Acute studies block design and maintained in a growth chamber (258C) under showed clear dose-dependent responses to DPH exposure, constant cool white fluorescent light (6800 lux). Frond number although mortality occurred at a much higher concentrations and fresh weight were measured on day 7. The number of in acute studies at lower pH (6.5; Table 3). The mean LC50 for P. promelas acute toxicity studies was 2.09 (Æ0.41) mg/L at pH8.5 and 59.28 (Æ6.64) mg/L at pH 6.5. The responses for P. promelas growth and feeding trials were similarly dose- where Ft is the number of fronds at time, t; F0 is the number of dependent (Fig. 1). Subchronic exposure survival was 100% fronds at time zero, is divided by the total exposure time (t) to except at the highest concentration tested in this study. The LOEC for growth and behavioral (feeding) responses were Table 2. Analytically verified mean (Æ standard deviation) diphenhydra- mine concentrations for acute and subchronic studies (mg/L) a Acute studies samples were taken from each replicate (n ¼ 3).
b Subchronic studies multiple samples were taken for Daphnia magna (n ¼ 3) and Pimephales promelas (n ¼ 7).
measured at much lower concentrations: 49.1 and 5.6 mg/L forgrowth and behavioral endpoints, respectively (Table 3). Acute Fig. 1. Mean (Æstandard error) growth (mg dry wt per fish; n ¼ 7 per to chronic ratios for growth and behavior endpoints were replicate) and behavioral responses (Artemia consumed per min; n ¼ 3 per calculated at 85 and 746, respectively (Table 3).
replicate) of larval fathead minnows (Pimephales promelas) following 7-ddiphenhydramine study. ÃSignificantly different from control ( p 0.05).
Control survival was >95% for both acute and chronic experiments. Acute tests showed dose-dependent responseswith a mean (n ¼ 3) LC50 of 0.37 (Æ0.14) mg/L. The 10-d The primary objective of this study was to establish a studies also exhibited a dose-dependent pattern. Survival in the baseline understanding of aquatic toxicological effects of a control and lower concentrations was 100% through the 10-d drug commonly reported in various environmental compart- exposure, while 100% mortality occurred at concentrations ments (tissue, sediment, water) [7,8]. Here we observed that an 27.8, 46.1, and 273.4 mg/L by days 7, 5, and 4, respectively.
aquatic plant model was insensitive to DPH, even at very high Reproduction LOEC and NOEC values were determined at 3.4 exposure levels (>10 mg/L). Such an observation is consistent and 0.8 mg/l, respectively (Fig. 2, Table 3). The corresponding with previous reports for several other classes of pharmaceut- ACR value for D. magna was 467.5 (Table 3).
icals (e.g., nonsteroidal antiinflammatory drugs, SSRIs, lipidlowering agents, beta-blockers) [38], likely because the hista- mine-H1, SERT, and muscarinic ACh receptors targeted by No statistically significant ( p > 0.05) effects of DPH on DPH were not present in either plant or algae models analyzed L. gibba responses were observed (Table 3). For example, for homologs [18]. However, significant acute and subchronic mean (ÆSD) growth rate for all plants was 0.358 (Æ0.014), effects of DPH were observed to a model fish and an inverte- compared to a mean growth rate in the highest concentration of 0.357 (Æ0.014) and 0.345 (Æ0.015) in control. No significant A second objective of this study was to employ approaches differences were observed among any of the various parameters previously proposed [3,5,22] to leverage mammalian pharma- measured (e.g., frond number, wet wt, growth rate). Because no cological information to understand aquatic hazards of phar- treatment level adversely affected this plant model, only the maceuticals. Fish are known to possess some degree of genetic highest concentration was confirmed analytically at 10.75 mg/L homology for the three critical DPH targets (histamine-H1, SERT, muscarinic ACh receptor), although the percent sim- Table 3. Toxicological thresholds of mean acute (n ¼ 3; Æ standard deviation) and subchronic endpoints of select organisms exposed to diphenhydramine and LC50 ¼ median lethal concentration; LOEC ¼ lowest observed effect concentration; NOEC ¼ no observed effect concentration.
[44]. Meinertz et al. [44] recently evaluated effects of DPH onD. magna over 21 d, but only at three widely separated con- centrations, resulting in an NOEC of 0.12 mg/L and LOEC of 70 mg/L. Subsequently, Meinertz et al. [44] were unable toreport differences between concentrations affecting survival and reproduction, as all D. magna above reported NOEC died and did not reproduce. In the present study, a reproduction NOEC value of 0.8 mg DPH /L for D. magna is in general agreement with this previous research, although we detected reproductive effects at an order of magnitude lower concen- tration than a survival NOEC of 27.8 mg/L (Table 3). One interesting observation in the Meinertz et al. [44] study was that even at the highest concentration tested (620 mg/L, reported as diphenhydramine hydrochloride) D. magna generally survived for about 10 d, whereas in the present study Daphniawere only able to survive for up to 7 d at the lowest lethal concentration (28 mg/L). It is possible the observed differences in time to death resulted from the ionization of DPH, as we demonstrated here with P. promelas (Table 3) and was observed previously for sertraline [27]. Meinertz et al. [44] reported a pHrange between 7.2 and 7.6, whereas pH was 8.63 (Æ0.05) in Fig. 2. Percent survival and mean (Æstandard deviation) Daphnia magna the present study. With a pKa of 8.98 DPH and other weak fecundity (neonate per female) following 10-d diphenhydramine study bases would be expected to shift ionization states within (n ¼ 10). ÃSignificantly different from control ( p 0.05).
environmental relevant pH ranges [27]. In this study, at apH closer to the pKa value, DPH was more un-ionizedand more toxic to D. magna than in the Meinertz et al. study.
ilarity is reported to vary between 40 to 70% [18]. When Thus, based on the information from the present study and observations of the present study are compared to similar others [27,39], it appears important to consider pKa during studies with the SSRIs fluoxetine [29] and sertraline [27], the environmental assessment of ionizable pharmaceuticals in DPH potency was very similar to these SSRIs, exerting sub- chronic toxicity on growth and feeding behavior with compa- The differences in D. magna response thresholds for DPH rable NOEC values (%10 mg/L). However, DPH was found to (Table 3) compared to SSRIs are likely related to other MOAs be much less effective in producing mortality in the 48-h and 7- of DPH and conservation of relevant targets in invertebrates.
d studies (Table 3) than comparable mortality thresholds for Although SSRIs were derived based on the SERT activity of sertraline [27] and fluoxetine [29]. Similar to observations DPH, SSRIs have been designed to more specifically target the previously reported for sertraline [27] and fluoxetine [39], this SERT, while DPH also has histamine and cholinergic targets.
study demonstrated that pH is a critically important factor Invertebrate physiology and neurochemistry is highly reliant on influencing aquatic toxicity of ionizable weak bases, because both histamine and acetylcholine as neurotransmitters. For a 28-fold higher DPH LC50 value was observed for P. promelas example, organophosphate (OP) pesticides are much more effective in invertebrates. Whereas OPs target acetylcholinees- In the present study the standardized growth endpoint in the terase, DPH and other antiacetylcholinergics (e.g., atropine) P. promelas model was not the most sensitive fish response to bind to the ACh receptor, preventing ACh neurotransmission DPH (Fig. 1, Table 3); rather, a behavioral response was more [45]. This binding is generally reversible, and over the short sensitive than the standardized growth endpoint. For example, term less toxic, but given continuous exposure and the like- the 5.6 and 24.5 mg/L DPH treatment levels significantly sup- lihood for bioaccumulation, particularly in effluent-dominated pressed feeding behavior but not growth (Fig. 1, Table 3).
streams [1], the probability of deleterious effects can increase.
Feeding behavior was examined here and in previous studies Thus, DPH may have exerted its toxicity to D. magna in the with the SSRIs sertraline [27] and fluoxetine [29] because it present study through an ACh MOA, which resulted in greater represents an alternative sublethal endpoint that may be plau- toxicity than previously reported for SSRIs. It may have also sibly related to the drug MOA (e.g., targeting the SERT). For been that an antihistamine MOA played a role in the observed example, previous work by Gould et al. [40] demonstrated that toxicity to cladocerans, because DPH also targets histamine ion SSRIs target the SERT in fish with similar binding kinetics as channel transporters in invertebrates [46]. It is important to note observed in mammals. Such MOA-related responses are rec- that DPH is not even the most potent antihistamine. For ognized as critical for pharmaceutical effects on aquatic organ- example, Berninger and Brooks [5] recently ranked deslorata- isms because therapeutic-related responses are often observed dine and loratadine much higher than DPH. Both of these drugs at much lower levels than traditional standardized survival and are also known to be much more potent at histamine H1 and ACh receptors [47]. Clearly these findings deserve additional Although similarities were found between DPH and sertra- line and fluoxetine potencies to the P. promelas model in the When we selected treatment levels for the subchronic fish present study, DPH toxicity to cladocerans differed drastically study, an ACR value of 2,100 was predicted for DPH, based on from previous studies of SSRIs. The responses of D. magna to mammalian margin of safety information presented in DPH exposure were two to three orders of magnitude lower than Equation 1 [5]. Based on results from the P. promelas feeding SSRI thresholds [29,41–43]. The only other study available on behavior study an ACR value of 746 was calculated (Table 3); the aquatic toxicology of DPH found similar results in D. magna an order of magnitude higher than previously reported feeding behavior ACR values for sertraline (ACR ¼ %15) [27] and fluoxetine (ACR ¼ 22) [29]. Although the observed ACR value 1. Brooks BW, Riley TM, Taylor RD. 2006. Water quality of effluent- was lower than predicted by Equation 1, a DPH ACR value of dominated ecosystems: ecotoxicological, hydrological, and manage- 746 is an order of magnitude higher than ACR values for 90% of ment considerations. Hydrobiologia 556:365–379.
all industrial chemicals [48]. Such an observation highlights the 2. Ramirez AJ, Brain RA, Usenko S, Mottaleb MA, O’Donnell JG, Stahl LL, Wathen JB, Snyder BD, Pitt JL, Perez-Hurtado P, Dobbins LL, importance to pharmaceutical risk assessment of understanding Brooks BW, Chambliss CK. 2009. Occurrence of pharmaceuticals and a priori pharmacological potency and if pharmacological targets personal care products in fish: Results of a national pilot study in the are present and maintain physiologically important functions United States. Environ Toxicol Chem 28:2587–2597.
in nontarget organisms [3–5,19,22]. Furthermore, we also 3. Ankley GT, Brooks BW, Huggett DB, Sumpter JP. 2007. Repeating history: Pharmaceuticals in the environment. Environ Sci Technol 41: employed a plasma model approach modified from that pre- sented by Huggett et al. [22] and advanced by Fick et al. [32].
4. Brooks BW, Huggett DB, Boxall ABA. 2009. Pharmaceuticals and We employed a partitioning equation (Eqn. 3) more appropriate personal care products: Research needs for the next decade. Environ for chemicals with apparent log P values less than 3. Addi- tionally, due to the appreciable effects of lowering pH on acute 5. Berninger JP, Brooks BW. 2010. Leveraging mammalian pharmaceut- ical toxicology and pharmacology data to predict chronic fish responses toxicity to fish (Table 3) log D was substituted at the study pH to pharmaceuticals. Toxicol Lett 193:69–78.
(8.5) for log P using Equation 3. Then, using Equation 6, at an 6. Stackelberg PE, Furlong ET, Meyer MT, Zaugg SD, Henderson HK, aqueous exposure concentration it was predicted that an AqET Reissman DB. 2004. Persistence of pharmaceutical compounds and of 2.53 mg/L would be required to potentially result in a fish other organic wastewater contaminants in a conventional drinking-watertreatment plant. Sci Total Environ 329:99–113.
plasma concentration equaling the human therapeutic dose for 7. Ferrer I, Heine CE, Thurman EM. 2004. Combination of LC/TOF-MS 50 ng/ml). As noted above, NOEC values for fish and LC/ion trap MS/MS for the identification of diphenhydramine in growth (24.5 mg/L) were not as sensitive as behavioral sediment samples. Anal Chem 76:1437–1444.
8. Ramirez AJ, Mottaleb MA, Brooks BW, Chambliss CK. 2007. Analysis Although plasma measurement of DPH was not possible due of pharmaceuticals in fish using liquid chromatography-tandem massspectrometry. Anal Chem 79:3155–3163.
to the size of P. promelas employed, this plasma model 9. Zhou SN, Oakes KD, Servos MR, Pawliszyn J. 2008. Application of approach, when the effects of log D were considered, appears solid-phase microextraction for in vivo laboratory and field sampling of useful for predicting thresholds related to the therapeutic MOA pharmaceuticals in fish. Environ Sci Technol 42:6073–6079.
of DPH because the NOEC value of 2.8 mg/L approximated the 10. Au-Yeung SCS, Rurak DW, Gruber N, Riggs KW. 2006. A pharmacokinetic study of diphenhydramine transport across the predicted threshold of 2.53 mg/L. If log D was not considered in blood-brain barrier in adult sheep: Potential involvement of a carrier- Equation 3, and instead Equation 2 was used, a slightly lower mediated mechanism. Drug Metab Dispos 34:955–960.
potential threshold value of 1.25 mg/L was predicted. Thus, the 11. Knox C, Law V, Jewison T, Liu P, Ly S, Frolkis A, Pon A, Banco K, Mak observations in the present study generally support use of a C, Neveu V, Djoumbou Y, Eisner R, Guo AC, Wishart DS. 2011.
plasma model approach for fish in further definitive studies, DrugBank 3.0: a comprehensive resource for ‘omics’ research on drugs.
Nucleic Acids Res 39:D1035–1041.
particularly when sublethal responses are plausibly linked to 12. Beijersbergen van Henegouwen GMJ, Van de Zijde HJ, Van de Griend J, therapeutic MOAs and plasma concentrations can be measured.
De Vries H. 1987. Photochemical decomposition of diphenhydramine inwater. Int J Pharm 35:259–262.
13. Boreen AL, Arnold WA, McNeill K. 2003. Photodegradation of pharmaceuticals in the aquatic environment: A review. Aquat Sci Observations in the present study highlight the importance of 14. Kosonen J, Kronberg K. 2009. The occurrence of antihistamines in carefully selecting study organisms and endpoints for pharma- sewage waters and in recipient rivers. Environ Sci Pollut Res Int 16:555–564.
ceuticals that possess multiple MOAs. Because standardized 15. Heberer T. 2002. Occurrence, fate, and removal of pharmaceutical toxicity testing methodologies may not account for specific residues in the aquatic environment: A review of recent research data.
aquatic MOAs of pharmaceuticals, environmental risks may be underestimated by current testing approaches [3–5]. Here we 16. Daughton CG, Brooks BW. 2011. Active pharmaceuticals ingredients and aquatic organisms. In Meador J, Beyer N, eds, Environmental demonstrated that an alternative behavioral endpoint was more Contaminants in Wildlife: Interpreting Tissue Concentrations, 2nd ed.
sensitive in the P. promelas model than survival or growth CRC, Boca Raton, FL, USA, pp 281–341.
responses, which is consistent with previous studies of the 17. van Wezel AP, Opperhuizen A. 1995. Narcosis due to environmental SSRIs fluoxetine [29] and sertraline [27], which possess a pollutants in aquatic organisms: Residue-based toxicity, mechanisms, common MOA as DPH (e.g., the SERT). Such alternative and membrane burdens. Crit Rev Toxicol 25:255–279.
18. Gunnarsson L, Jauhiainen A, Kristiansson E, Nerman O, Larsson DGJ.
endpoints that may be related to a specific therapeutic MOA 2008. Evolutionary conservation of human drug targets in organisms (e.g., the SERT) and are relevant to organismal and population used for environmental risk assessments. Environ Sci Technol 42:5807– level consequences are necessary to appropriately characterize environmental risks [3,4]. It is also important to note that 19. Brown NJ, Roberts LJ II. 2001. Histamine, bradykinin, and their antagonists. In Hardman JG, Limbird LE, eds, Goodman and Gilman’s responses might be related to another DPH MOA, ACh activity, Pharmacological Basis of Therapeutics, 10th ed. McGraw Hill, New which appeared to be appropriately characterized by the D. magna model. Thus, employing a priori knowledge of 20. Wong DT, Perry KW, Bymaster FP. 2005. The discovery of fluoxetine comparative pharmacology among target and nontarget organ- hydrochloride (Prozac). Natl Rev Drug Disc 4:764–774.
isms remains critical during environmental hazard and risk 21. Bird SB, Gaspari RJ, Lee WJ, Dickson EW. 2002. Diphenhydramine as a protective agent in rat model of acute, lethal organophosphate poisoning.
assessments of pharmaceuticals in the environment [3–5].
22. Huggett DB, Cook JC, Ericson JF, Williams RT. 2003. A theoretical Acknowledgement—This research was supported by a Glasscock Fund for model for utilizing mammalian pharmacological and safety data to Excellence in Environmental Science grant to J.P. Berninger and the Baylor prioritize potential impacts of human pharmaceuticals to fish. Hum Ecol University Department of Environmental Science. The Center for Reservoir and Aquatic Systems Research and Institute of Biomedical Studies at Baylor 23. U.S. Environmental Protection Agency. 2002. Methods for measuring University provided general support.
the acute toxicity of effluents and receiving waters to freshwater and marine organisms, 5th ed. EPA-821-R- 020-012. Office of Research and a treatment wetland using Pimephales promelas, Ceriodaphnia dubia, and Vibrio fischeri. Arch Environ Contam Toxicol 42:9–16.
24. American Public Health Association, American Water Works Associ- 37. Brain RA, Solomon KR. 2007. A protocol for conducting 7-day daily ation, Water Environment Foundation. 1998. Standard Methods for the renewal tests with Lemna gibba. Nat Protocols 2:979–987.
Examination of Water and Wastewater, 20th ed. American Public Health 38. Brain RA, Hanson ML, Solomon KR, Brooks BW. 2008. Aquatic plants exposed to pharmaceuticals: Effects and risks. Rev Environ Contam 25. U.S. Environmental Protection Agency. 2002. Short-term methods for estimating the chronic toxicity of effluents and receiving waters to 39. Nakamura Y, Yamamoto H, Sekizawa J, Kondo T, Hirai N, Tatarazako freshwater organisms. EPA-821-R-02-013. Office of Research and N. 2008. The effects of pH on fluoxetine in Japanese medaka (Oryzias latipes): Acute toxicity in fish larvae and bioaccumulation in juvenile 26. Organization for Economic Co-operation and Development. 2008.
Daphnia magna Reproduction Test. Test 211. In OECD Guidelines for 40. Gould GG, Brooks BW, Frazer A. 2007. [3H] citalopram binding to the Testing of Chemicals, Section 2: Effects on Biotic Systems. Paris, serotonin transporter sites in minnow brains. Basic Clin Pharmacol 27. Valenti TW, Perez-Hurtado P, Chambliss CK, Brooks BW. 2009.
41. Minagh E, Hernan R, O’Rourke K, Lyng FM, Davoren M. 2009. Aquatic Aquatic toxicity of sertraline to Pimephales promelas at environmentally ecotoxicity of the selective serotonin reuptake inhibitor sertraline relevant surface water pH. Environ Toxicol Chem 28:2685–2694.
hydrochloride in a battery of freshwater test species. Ecotoxicol Environ 28. U.S. Environmental Protection Agency. 2009. Estimation Programs Interface SuiteTM for Microsoft1 Windows, ver 4.00. Washington, 42. Brooks BW, Foran CM, Richards SM, Weston J, Turner PK, Stanley JK, Solomon KR, Slattery M, La Point TW. 2003. Aquatic ecotoxicology of 29. Stanley JK, Ramirez AJ, Chambliss CK, Brooks BW. 2007.
fluoxetine. Toxicol Lett 142:169–183.
Enantiospecific sublethal effects of the antidepressant fluoxetine to a 43. Oakes K, Coors A, Escher B, Fenner K, Garric J, Gust M, Knacker T, model aquatic vertebrate and invertebrate. Chemosphere 69:9–16.
Kuster A, Kussatz C, Metcalfe C, Monteiro S, Moon T, Parrott J, Pery A, 30. Fitzsimmons PN, Fernadez JD, Hoffman AD, Butterworth BC, Nichols Ramil M, Tarazona JV, Sanchez Arguello P, Ternes T, Trudeau V, Van JW. 2001. Branchial elimination of superhydrophobic organic com- Der Kraak G, Servos M. 2010. An environmental risk assessment for the pounds by rainbow trout (Oncorhynchus mykiss). Aquat Toxicol 5: serotonin re-uptake inhibitor fluoxetine: A case study utilizing the European risk assessment framework. Integr Environ Assess Manag 31. Scherrer RA, Howard SM. 1977. Use of distribution coefficients in quantitative structure-activity relationships. J Med Chem 20:53–58.
44. Meinertz JR, Schreier TM, Bernardy JA, Franz JL. 2010. Chronic 32. Fick J, Lindberg RH, Tysklind M, Larsson DGJ. 2010. Predicted critical toxicity of diphenhydramine hydrochloride and erythromycin thiocya- environmental concentrations for 500 pharmaceuticals. Reg Toxicol nate to daphnia, Daphnia magna, in a continuous exposure test system.
Bull Environ Contam Toxicol 85:447–451.
33. U.S. Environmental Protection Agency. 1994. 10-day chronic toxicity 45. Carvalho FD, Machado I, Marty´nez MS, Soares A, Guilhermino L. 2003.
test using Daphnia magna or Daphnia pulex. Compendium of ERT Use of atropine-treated Daphnia magna survival for detection of Standard Operating Protocols, SOP 2028. Office of Solid Waste and environmental contamination by acetylcholinesterase inhibitors. Eco- 34. Dzialowski EM, Turner PK, Brooks BW. 2006. Physiological and 46. Haas HL, Sergeeva OA, Selbach O. 2008. Histamine in the nervous reproductive effects of beta adrenergic receptor antagonists on Daphnia magna. Arch Environ Contam Toxicol 50:503–510.
47. Orzechowski RF, Currie DS, Valancius CA. 2005. Comparative 35. Stanley JK, Ramirez AJ, Mottaleb M, Chambliss CK, Brooks BW. 2006.
anticholinergic activities of 10 histamine H1 receptor antagonists in Enantio-specific toxicity of the b-blocker propranolol to Daphnia magna two functional models. Eur J Pharmacol 506:257–264.
and Pimephales promelas. Environ Toxicol Chem 25:1780–1786.
48. Raimondo S, Montague BJ, Barron MG. 2007. Determinants of the 36. Hemming JM, Turner PK, Brooks BW, Waller WT, La Point TW. 2002.
variability in acute to chronic toxicity ratios for aquatic invertebrates and Assessment of toxicity reduction in wastewater effluent flowing through fish. Environ Toxicol Chem 26:2019–2023.

Source: http://jasonberninger.com/wp-content/uploads/2013/11/Berninger-2012-Diphenhydramine-Ecotox.pdf