Diazepam, metformin, omeprazole and simvastatin: a full discussion
of individual and mixture acute toxicity
Raquel Sampaio Jacob 1,2 ●
Lucilaine Valéria de Souza Santos1,3 ●
Mirna d’Auriol1 ●
Yuri Abner Rocha Lebron 1 ●
Victor Rezende Moreira 1 ●
Liséte Celina Lange 1
Accepted: 6 June 2020
© Springer Science+Business Media, LLC, part of Springer Nature 2020
High consumption of drugs, combined with their presence in the environment, raises concerns about its consequences. Even
though researches are often engaged in analyzing substances separately, that is not the environmental reality. Therefore, the
aim of this study was to investigate the acute toxicity of the pharmaceuticals simvastatin, metformin, omeprazole and
diazepam, and all possible mixtures between them, to the organism Aliivibrio fischeri, verifying possible synergistic or
antagonistic effects and assessing byproducts formation. In terms of individual toxicity, omeprazole is the most toxic of the
active ingredients, followed by simvastatin, diazepam and, finally, metformin. When the toxicity of mixtures was tested,
synergism, antagonism and hormesis were perceived, most probably generated due to byproducts formation. Moreover, it
was observed that even when compounds are at concentrations below the non-observed effect concentration (NOEC), there
may be toxicity to the mixture. Hence, this work points to the urgent need for more studies involving mixtures, since
chemicals are subject to interactions and modifications, can mix, and potentiate or nullify the toxic effect of each other
Keywords Acute toxicity ●
Aliivibrio fischeri ●
Emerging contaminants ●
Mixtures toxicity ●
The rapid improvement of analytical instrumentation and
methods in the late twentieth century led to a large number
of substances, previously undetectable, to emerge as
environmental contaminants (Taylor and Senac 2014;
Nilsen et al. 2019). Among this class of so-called emerging
contaminants, there are drugs which are extensively used in
human and veterinary medicine, considerably prolonging
the life of living beings. A previous study (Reis et al. 2019)
evaluated the occurrence of 28 pharmaceuticals in different
drinking water treatment plants located in Brazil. These
authors found that the occurrence in the environment are at
trace concentration (ranging from ng/L up to μg/L).
Antipyretics, analgesics, lipid regulators, antibiotics,
antidepressants, chemotherapy agents, antidiabetics, gastric
pH regulators and contraceptive drugs are among the most
consumed pharmaceuticals. In the present study, four different drugs were chosen based on their world wild consumption and previous monitoring researches. They are
diazepam, metformin, omeprazole and simvastatin, which
their main physicochemical properties are presented in
Table 1. These drugs were recurrent in Brazilian waters
according to previous studies (Foureaux et al. 2019; Reis
et al. 2019) and composes the list of medicines provided
free of charge by the Brazilian public health system (SUS),
which corroborates with these drugs consumption and
occurrence in the environment. These substances are not
only present in the sewers, but also in water bodies in
concentrations ranging from 8–6323 ng/L (Tambosi et al.
2010; Reis et al. 2019). Reis et al. (2019) found metformin
in surface waters in Brazil in concentrations ranging from 8
* Raquel Sampaio Jacob
1 Sanitation and Environmental Engineering Department, School of
Engineering, Federal University of Minas Gerais, Avenue Antônio
Carlos, 6627, Campus Pampulha, MG, Brazil
2 Civil Engineering Department, Pontifical Catholic University of
Minas Gerais, Belo Horizonte, MG ZIP 30.535-901, Brazil
3 Chemical Engineering Department, Pontifical Catholic University
of Minas Gerais, Belo Horizonte, MG ZIP 30.535-901, Brazil
Supplementary information The online version of this article (https://
doi.org/10.1007/s10646-020-02239-8) contains supplementary
material, which is available to authorized users.
to 203 ng/L. Diazepam was also detected in surface water in
Brazil at concentrations up to 763 ng/L (Böger et al. 2018).
High medicines consumption, combined with their presence in the environment, raises concerns about its consequences since these compounds are bioactive and
therefore capable of causing effects in living systems
(Ginebreda et al. 2010; Nilsen et al. 2019; Patel et al. 2019;
Peña-Guzmán et al. 2019). Moreover, many of these substances are designed to exhibit persistence in organisms
(Fent et al. 2006; Patel et al. 2019), exacerbating the possible consequences of their environmental presence.
There are several organisms’ classes that can be used to
assess toxicity, but the use of Aliivibrio fischeri, a luminescent marine bacterium, is highlighted. Acute toxicity
testing with this organism is considered an effective alternative due to the correlation with other bioassays using fish
and invertebrates and the speed of results obtainment
(Kaiser 1998). Correlation studies of Aliivibrio fischeri
toxicity results and other aquatic organisms are quite
numerous, which provides greater confidence in the use of
this micro-organism in ecotoxicological tests (Kaiser and
Palabrica 1991; Zhao et al. 1993; Kaiser et al. 1994; Dong
et al. 2019; Baek et al. 2019; Zuriaga et al. 2019).
The evaluation of mixtures toxicity in the environment is
important because although researches usually analyze a
substance separately, that is not the environmental reality
(Gomez-Eyles et al. 2009; Lindim et al. 2019; Ukić et al.
2019). Chemicals are subject to interactions and modifications, they can mix, and are able to potentiate or nullify the
toxic effect of another.
Researches on toxicity of mixtures have shown increasing
reach in the scientific community. These studies include
in vitro and in vivo studies (Cedergreen et al. 2012; Coors et al.
2012; Boyd et al. 2013), evaluation of toxicity using combined
effects models (Moser et al. 2012; Crépet et al. 2013; Hertzberg
et al. 2013; Webster 2013), environmental impacts analysis
(Løkke 2010; Allan et al. 2012), risk assessment studies
(Johnson et al. 2013; Løkke et al. 2013; Meek 2013) and
examination of chemical reactivity in the context of complex
mixtures (Goel et al. 2013). While the specifics of these surveys vary, their common goal is to improve our ability to
predict the effects of exposure to chemicals’ mixtures.
According to Cleuvers (2005), two different concepts are
used to predict the mixtures toxicity, called concentration
addition (CA) and independent action (IA) models. The CA
model assumes that components of a chemical mixture
share a common action mechanism, i.e.: each component
has the same specific interaction with a molecular target in
the test organism (Berenbaum 1985). By contrast, the IA
model assumes different, not similar, action mechanisms
among the mixture components, i.e.: the toxics interact with
different molecular targets, resulting in a common toxicological response by different reactions chains in an
organism (Faust et al. 2003; Cleuvers 2003). Such concepts,
for Faust et al. (2003), represent different hypotheses about
the functional relationship between the toxicity of substances in individual action and combined action, and may
lead to possible synergism and antagonisms.
A previous work exemplified the synergistic and antagonistic toxicity impact of drugs. Yang et al. (2008) tested
Table 1 Physicochemical properties of the studied drugs, diazepam, metformin, omeprazole and simvastatin
Pharmaceutical Class Chemical formula
(Mol. wt. in g/mol)
Structure pKa LogKow Solubility
Simvastatin Lipid regulator C25H28O5 (418.574) – – – 4.68 0.07
Metformin Antidiabetic C4H11N5 (129.167) 3.46E−09 101 12.4 −2.64 1063.24
Diazepam Antianxiety C16H13ClN2O (284.743) – – 3.4 2.90 69.35
Omeprazole Antacid C17H19N3O3S (345.417) 3.62E−06 252 1.2; 7.4 2.23 0.34
Reference: Patel et al. (2019)
R. S. Jacob et al.
toxicity of twelve antibiotics: triclosan, triclocarban, roxithromycin, clarithromycin, tylosin, tetracycline, chlortetracycline, norfloxacin, sulfamethoxazole, ciprofloxacin,
sulfamethazine and trimethoprim; here arranged in ascending toxicity order. Analyzes were performed with the algae
Pseudokirchneriella subcapitata in order to assess growth
inhibition. Among the results, it was noted the important
antagonistic effect of the triclosan and norfloxacin combination, where the mixture was less toxic than its own
components individually considered.
Cleuvers (2003) conducted tests with Daphnia magna,
Desmodesmus subspicatus and Lemna minor organisms
exposed to clofibric acid, carbamazepine, propranolol,
metoprolol, ibuprofen, diclofenac, naproxen, captopril and
metformin. In this research, analysis with various pharmaceuticals’ combinations revealed more toxic effects than
expected by only measuring the drugs individual toxicity
(synergism), leading to the fact that it is difficult to predict
toxicity of mixtures from single compound toxicity data.
Given this scenario, this research aimed at the investigation of the acute toxicity of simvastatin, metformin, diazepam, omeprazole and all their possible mixtures, to
Aliivibrio fischeri. It is also intended to verify if the mixture
of the aforementioned substances generates synergistic or
antagonistic toxic effects. It should be mentioned that discussion in this article will also happen based in by-products
formation in mixtures, which might modify samples’
Materials and methods
Solutions containing the studied drugs diluted in Milli-Q
water were prepared. The active ingredients were purchased
from Sigma-Aldrich and have purity levels higher than
99%. It should be noted that the second and third stages
(corresponding to mixture toxicity assays; see “Toxicity
tests”) were performed in triplicate.
Individual pharmaceuticals analysis
The objective at this stage was to prepare saturated solutions
(initial concentration: simvastatin: 0.07 mg/mL; metformin:
1063.24 mg/mL; diazepam: 69.35 mg/mL; and omeprazole:
0.34 mg/mL; corresponding to their maximum solubility in
water at 25 °C) of each drug in Milli-Q water and immediately make dilutions in series (from solubility up to
0.001 μg/mL; dilution factor: 1:2) that would correspond to
the effective concentration (EC) or toxicity values. The
pharmaceutical compounds were quantified after analysis
with a HPLC-MS system (see “HPLC-MS analysis”). The
stock solutions were stored in the absence of light, at
<−20 °C and their concentrations were verified prior to
each toxicity test in order to assure their preservation.
Analysis with mixtures of pharmaceuticals
Aiming to test possible synergistic or antagonistic effects
between the drugs mixtures, ecotoxicological tests with all
possible mixtures of the active ingredients studied were
performed. There were 11 possible mixtures to be formed
between the four active ingredients evaluated, corresponding to six possibilities of two-drugs mixtures, four possibilities of three-drugs mixtures and a unique possibility to
mix the four-drugs together (all combinations represented in
“Toxicity of the pharmaceutical’s mixtures,” Table 4).
Possible antagonism or synergism phenomena were
assessed taking into account both the active ingredient
individual EC10 (concentration capable of causing effect in
10% of tested population; established as described in
“Individual pharmaceuticals analysis”) and the number of
compounds in the mixture (Cleuvers 2005). That means,
that to establish the mixture concentration between two
drugs, EC10/2 corresponding to each of those drugs were
used for the test organism. The same occurred with mixtures
of more compounds, in which the concentration adopted for
mixing was EC10/3, in the case of three components mixture, and EC10/4 in case when all four compounds were
used. Following the concept of concentration addition, the
mixture effect should add up to a total effect 10%. As
described previously, dilutions were performed in series
using Milli-Q water and EC10 determined by linear
Furthermore, to adequately assess the effect of the mixtures, toxicological tests were carried out with the substances separately in their EC10/2, EC10/3 and EC10/4
Acute ecotoxicological tests were performed with the
luminescent marine bacterium Aliivibrio fischeri provided
by SDI using a MICROTOX® model 500 Analyzer (SDI),
following the ABNT NBR 15411-3 (ABNT 2012) and
MICROTOX® Omni Software standard protocols. Analyzes
were divided into three steps: the first one was designed to
meet the toxicity of the drugs individually to the bacteria,
the second was with equitoxic mixtures containing each of
the drugs in their EC10/2, EC10/3 and EC10/4 and the last
with all possible mixtures of compounds. Firstly, acute
toxicity was determined from nine dilutions of the initial
solution in measurements of bacteria luminescence in
30 min. To determine toxic effect, the software performs a
comparison based on emitted sample light from its various
Diazepam, metformin, omeprazole and simvastatin: a full discussion of individual and mixture acute. . .
dilutions and the control solution. In the second and third
step, the procedure was different, since the objective was
not to find a effective concentration, but the effect caused to
the bacteria in certain individual or mixture concentrations.
In this case, there was an adaptation of the recommended
procedure by NBR 15411-3 (ABNT 2012), using the three
following equations. The first one was used to calculate the
correction factor (fkt) from the measured light output, correcting the initial values of all samples, before using them as
reference values to determine the decrease in luminescence
caused by the sample, as follows:
fkt ¼ Ikt=I0 ð1Þ
Ikt means the luminescence intensity of the control after the
exposure period and I0 is the intensity of control
luminescence, immediately before addition of the diluent,
in relative luminescence units. The second equation ensures
the usage of the corrected values for each reading, as
Ict ¼ Ikt fkt ð2Þ
fkt is the average of the controls’ fkt and Ict represents the
corrected values of I0. Finally, the sample effect on the
bacteria’s luminescence can be calculated from the following equation:
Et ¼ ð Þ Ict Ikt=Ict 100 ð3Þ
Et is the inhibitory effect of the bacteria suspension after the
exposure period, expressed as a percentage (%). The ECx%
for each compound was determined through concentrationeffect curves.
In all the above steps, before carrying out the tests, the
samples were subjected to pH adjustment between an
acceptable range of 6.5–7.5 with HCl or NaOH and dilution
with a 2% NaCl solution, called diluent. These possible
dilutions of the compounds were considered in the effect’s
The salinity of the samples was checked with a highresolution refractometer (RTS-101ATC, Instrutherm).
Those samples which have salinity values below 22% must
receive the addition of osmotic-adjusting solution for test
execution. In this study, all samples showed salinity values
Mixtures’ effect: synergism and antagonism
The two models used in this study to predict mixtures
toxicity were concentration addition (CA) and independent
action (IA). The concept of CA can be described
mathematically by Eq. (4) (Berenbaum 1985):
ð Þ i¼1
ci=ECx;i ¼ 1 ð4Þ
ci is the concentration of each substance from the mixture
and ECx,i is the substance concentration that causes x%
effect on the tested population.
Contrastingly, the equation that describes the combined
effect that acts upon the independent action model for a
mixture is given by:
EC;Mix ¼ 1 Yn
1 ECi ð5Þ
ECi is the effects of the individual substances and EC,Mix is
the total mixture effect.
The described models provide an initial mathematical
basis for the prediction of toxicity effects of the mixtures,
but interactions among compounds may occur, which may
result with deviations of mixture toxicity from the model(s)
applied. If mixture results in toxicity greater than the one
predicted by the model, the mixture components act
synergistically. On the other hand, if the combined action of
mixture components s result with toxicity smaller than the
one predicted by the model, the components acts antagonistically. The methodology for identifying possible
synergisms and antagonisms was proposed by Cedergreen
et al. (2007).
Finally, in order to compare deviation of observed mixtures toxicities from the applied models, the method of
effect residual ratio (ERR), proposed by Wang et al. (2010)
was used. The equation that describes the ERR model is:
ERR ¼ Eprd Eobs=Eobs 100 ð6Þ
Eprd and Eobs are, respectively, the effect values predicted by
the IA or CA models and the observed effects values to a
given concentration level.
The analytical determinations were performed using an
HPLC (LC 20A, Shimadzu) coupled to a mass spectrometer
MicroTOF QII (Bruker), with an electronic electrospray
ionization (ESI), at a resolution of 12,000 m/z. The chromatographic separation was achieved by a reverse phase C18
column (Shim-pack XR-ODS). The mobile phases used
were water (A), methanol (B) and formic acid. Furthermore,
the solvent composition started with 10% of B, increased to
70% in 3 min, then again increased to 95% over 6 min and
remaining stable for 7 min. After, it was decreased to 10%
of B in 10 min, and finally remained at 10% of B for 15 min.
The injected sample volume was 20 μL at 20 °C.
R. S. Jacob et al.
The drugs quantification consisted in the concoction of
an external calibration curve from the active ingredients,
using different concentrations (0.50, 1.00, 2.00, 10.00 and
20.00 mg/L) in methanol and water in a 1:1 ratio. The peak
area generated by the sample was compared to the external
calibration curve to determine the exact drug concentration.
Results and discussion
Individual toxicity of pharmaceuticals
Toxicity was observed for all drugs tested in a 30-min test,
as shown in Table 2.
Omeprazole was observed to be the most toxic among
the active ingredients, followed by simvastatin, diazepam
and finally metformin. Furthermore, the only effective
concentration identified in all drugs was EC10. For this
reason, this parameter was selected as the basis for the next
stage, analysis of mixtures toxicity. It should be mentioned
that only one of the drugs has the EC50 value, which is a
more commonly referred parameter in the literature. This
pharmaceutical is omeprazole, with a value of EC50 of
0.015 mg/L. The disparity in the acute toxicity observed
among the pharmaceutical compounds is most probably
caused by the effects derived from the different substituents
in their molecule structure. The following order of contribution to toxicity was reported Dong et al. (2019): –NO2
> -Cl > -CH3 > -NH2 > -OH, most of them possible to be
formed by omeprazole decomposition.
Ortiz de Garcia et al. (2014) reported that the toxicity of
omeprazole (EC50) for Aliivibrio fischeri was 1.76 mg/L in
the 30-min test. Another study (Zuriaga et al. 2019),
reported that the omeprazole EC50 for Aliivibrio fischeri was
3.7 mg/L (30-min test). It is noteworthy that although both
of these papers have observed high toxicity of the mentioned compound for Aliivibrio fischeri, the values are different probably due to omeprazole speciation. Both standard
procedures adopted by Ortiz de Garcia et al. (2014) and
Zuriaga et al. (2019) are different from this study, and the
difference in the test pH directly affects the omeprazole
speciation and bioavailability. Under higher pH conditions,
the pharmaceutical is considered more bioavailable (Prichard et al. 1985) and different ionic species may be presented, leading to higher inhibitory effects.
The second most toxic to Aliivibrio fischeri was simvastatin, and although there was no EC50 value for this
compound in the current study, some researches discuss the
consequences of this anti-hypertensive for the reproduction
and survival of amphibians (Neuparth et al. 2014), damage
caused to fish cells (Ellesat et al. 2011; Ribeiro et al. 2015)
and chronic toxicity to copepods (Dahl et al. 2006).
In contrast, about diazepam, some articles identify the
impact on living beings’ life cycle (Muñoz et al. 2008) and
acute toxicity to cnidarian (Pascoe et al. 2003). According
to Nunes et al. (2005), diazepam has EC50 of 12.7, 12.2 and
16.5 mg/L for Gambusia holbrooki, Artemia parthenogenetica and Tetraselmis chuii, respectively. It is noteworthy
that these are higher toxicity values than those identified for
Aliivibrio fischeri in this article.
It is also noteworthy that hydrophobic compounds pose a
higher toxicity compared to hydrophilic compounds, as they
present higher propensity for interacting with the cell
membrane. A similar trend was noticed in this study, in
which metformin, considered a hydrophilic compound,
presented the lowest luminescence inhibition. Furthermore,
excluding simvastatin, a significant (p value < 0.05)
increase in toxic effect (lower EC10) was observed when
increasing the logKow. The most hydrophobic compound
(simvastatin) did not show the highest luminescence inhibition. This may be due to the other factors such as drugs
physical-chemical characteristics, mode of action, functional groups among others. Baillie (2008) showed that an
interplay of drug-metabolizing enzymes and drug transporters can represent a critical determinant of drug disposition, drug interactions, and toxicity in animals and
Finally, it should be mentioned a relevant research that
dealt with acute toxicity of metformin combined with other
drugs for three test organisms (Cleuvers 2003). In this study,
metformin hydrochloride showed EC50 64, 320 and 110 mg/
L for Daphnia magna, Desmodesmus subspicatus and
Lemna minor, respectively. Again, these are higher values
than those identified for Aliivibrio fischeri in this article.
Toxicity of the pharmaceutical’s mixtures
The results of mixtures toxicity, regarding EC10/2, EC10/3
EC10/4, are presented in Table 3.
Some of the tests indicated a phenomenon called hormesis, which means a positive deviation observed in the test
organism in presence of the contaminant. Furthermore,
hormesis can be characterized by low-dose stimulation and
high-dose inhibition. Several reports have shown phenomenon of hormesis in toxicity assays as well as in the natural
Table 2 Acute toxicity to Aliivibrio fischeri (30 min) of omeprazole,
simvastatin, diazepam and metformin
EC10 EC20 EC50
Metformin 870.79 mg/L NAa NAa
Diazepam 8.69 mg/L 28.40 mg/L NAa
Simvastatin 9.30 μg/L 29.00 μg/L NAa
Omeprazole 5.30 μg/L 6.50 μg/L 15.00 μg/L
Not available: there was no value for this sample, which means that
the toxicity found was lower than this parameter
Diazepam, metformin, omeprazole and simvastatin: a full discussion of individual and mixture acute. . .
environment (Calabrese and Baldwin 2003; Calabrese
2008), especially when organisms are exposed to mixtures,
rather than single chemicals. It is emphasized that hormesis
detection is not a sign that the contaminant is beneficial to
the organism, on the contrary, it could be a case of a contaminant that is toxic in chronic toxicity tests or an acute
toxicity tests in higher concentration. Also according to
Calabrese (2008), hormesis is considered an evolutionary
mechanism, because it is a compensation or an adaptive
response of organisms to overcome some imbalance in
order to prevent the extinction of the species.
After the preliminary tests were completed, it was possible to develop the mixtures analyzes, identify the
responses provided by CA and IA models, as well as verify
any discrepancies between the models and reality (ERR).
The results are compiled in Table 4.
Regarding the binary mixtures, all of those containing
metformin generate the phenomenon called hormesis, discussed above. Furthermore, this pharmaceutical alone, at
EC10/2, promotes the same effect when in binary mixtures.
In this sense, because the result is below the predicted by
mathematical models CA and IA, it is said that there is
antagonism. Godoy et al. (2015) observed that another
binary mixture with pharmaceuticals, propranolol hydrochloride and losartan potassium, behaved similarly in tests
with the macrophyte Lemna minor.
The other binary mixtures (simvastatin + omeprazole,
simvastatin + diazepam and diazepam + omeprazole) show
toxicity, and in a higher value than expected by both CA and
IA models, featuring synergism. Similarly, Zou et al. (2012)
found a synergistic effect when testing mixtures of sulfonamides in contact with the luminescent bacterium Photobacterium phosphoreum. It must be emphasized that the
mixture of simvastatin and diazepam is the one that generated
the highest toxicity effects on Aliivibrio fischeri in this work.
About the ternary mixtures, all are toxic, with effective
concentration close to those predicted by the CA model. It
should be noted that the mixture of metformin + simvastatin +
diazepam has the highest value among the ternary mixtures,
again bringing out specifically the interaction between simvastatin and diazepam, which in a binary mixture generated a
high toxicity effect. The referred ternary mixture, as well as the
sample containing metformin, simvastatin and diazepam, present a synergistic behavior for both models. Regarding the
combination between metformin, simvastatin and omeprazole,
it is synergistic to the IA model and corresponds to the CA
model prediction. The last ternary mixture, simvastatin +
omeprazole + diazepam, behaves similarly to that provided by
both mathematical models. As in this article, Phyu et al. (2011),
tested the toxicity of pesticide mixtures for Ceriodaphnia
Table 4 Identified effect on acute toxicity tests with Aliivibrio fischeri (30 min) with pharmaceutical mixing samples, inhibition provided by the
CA and IA models and values of residual ratios effect (% ERR) calculated for differences between the real effects and those predicted by the CA
and IA model
Mixtures Effect observed in
the toxicity test (%)
by CA model (%)
ERR (%) from
the CA model
by IA model (%)
ERR (%) from
the IA model
Metformin + simvastatin Hormesis 10.0 100.0 2.7 100.0
Metformin + diazepam Hormesis 10.0 100.0 6.5 100.0
Metformin + omeprazole Hormesis 10.0 100.0 8.3 100.0
Simvastatin + diazepam 17.1 10.0 41.6 9.0 47.5
Simvastatin + omeprazole 11.4 10.0 12.7 10.8 6.0
Diazepam + omeprazole 14.9 10.0 33.1 14.3 4.7
Metformin + simvastatin + omeprazole 9.7 10.0 2.5 5.2 4.6
Simvastatin + omeprazole + diazepam 9.9 10.0 0.8 9.4 5.5
Metformin + omeprazole + diazepam 10.7 10.0 7.0 9.4 12.9
Metformin + simvastatin + diazepam 14.0 10.0 28.8 4.4 68.7
Simvastatin + omeprazole + diazepam + metformin 16.2 10.0 38.3 0.0 100.0
Table 3 Observed effect (%) for Aliivibrio fischeri organism in a
30-min test at various concentrations
Concentration Effect (%) Concentration Effect (%)
Metformin 870.79 mg/L 10.0 435.40 mg/L Hormesis
Diazepam 8.69 mg/L 10.0 4.35 mg/L 6.5
Simvastatin 9.30 μg/L 10.0 5.00 μg/L 2.7
Omeprazole 5.30 μg/L 10.0 3.00 μg/L 8.3
Metformin 290.26 mg/L Hormesis 217.70 mg/L Hormesis
Diazepam 2.90 mg/L 4.4 2.17 mg/L Hormesis
Simvastatin 3.10 μg/L Hormesis 2.30 μg/L Hormesis
Omeprazole 1.80 μg/L 5.2 1.30 μg/L Hormesis
R. S. Jacob et al.
dubia, founding better match to the CA model compared to IA.
Other than that, the referred authors encouraged the use of this
method whenever there is similarity of action of the compounds tested.
The quaternary mixtures generated one of the highest
toxicity effects (16.23%—Table 4), even though the four
compounds, individually, are in concentrations below the
one where there is no effect (NOEC). While this mixture
generates a synergistic effect compared to both models, this
behavior was predicted by Cleuvers (2005) when discussing
the conditions of the CA model and reports that substances
applied below its concentration of no observed effect may,
however, contribute to the total effect of the mixture. A
similar result was found by Backhaus et al. (2011), when
investigating the toxic potential for periphyton of the
pharmaceuticals fluoxetine, propranolol, triclosan, zinc
pyrithione, clotrimazole and its mixtures. In the cited study,
clear effects were identified on the organism when in contact with the mixture of pharmaceuticals, even when all five
components were present at non-observed effect concentration (NOEC).
About the adequacy of the findings to those provided
by CA and IA models (ERR), the values are in most
cases close to the predicted values, except those results
where the mixture generates hormesis or where the
mixture is quaternary. Data analysis clarifies that the
clear majority of the values presents a discrepancy of 0.8
to 12.92% to that suggested by CA and AI models. This
denotes a closer proximity to these models than those
found by other authors, such as Godoy et al. (2015),
where ERR reaches 228%. It can also be said that the
data in this article is closer to the predictions of the CA
model compared to IA. The CA model proposes that the
components of a chemical mixture share a common
mechanism of action, which means that each component
has the same specific interaction with a molecular target
in the test organism (Berenbaum 1985).
Still, it must be noted that the models’ adjustment to
reality is not precise, which means that it is difficult to
accurately predict the toxicity of mixtures without actually
performing tests with real samples. That is because other
factors, such as the byproducts formation, can affect the
mixtures toxicity, as described in the literature (Cleuvers
2003; Yang et al. 2008; Godoy et al. 2015) and will be
Identification of transformation products (TPs)
formation in the mixtures and their relationship
The analysis of transformation products was carried out
right after the drugs mixture. If not, the mixture was stored
in 1.5 mL amber vials at <−20 °C to avoid degradation. The
formation of these new products could be due to the interactions that can be established between the drugs molecules,
for example hydrogen bonding interactions between
hydroxyl and amino groups, as well as many other intermolecular interactions at lower energies (Zuriaga et al.
2019). In Table S1 are shown the transformation products
found after mixing the pharmaceuticals. It was observed the
existence of byproducts in the mixtures under analysis,
many of these compounds are common to several mixtures.
Examples are C9H20NO, C6H11N6O, C16H9N2O, C4H7N12,
and especially C6H18N3O3 and C14H42N3O2. This similarity
is expected, since the precursors of the eleven mixtures
were the same four drugs. Based on this result, it is expected
homogeneous toxicity results, which did not happen.
One possible explanation for the highly heterogeneous
toxicity results are the presence of infrequent TPs, which
occurred in only one or two samples. As can be seen, it was
found 13 compounds that did not replicate in any of the other
samples. For example, it was observed that the mixtures
simvastatin + diazepam and metformin + omeprazole + diazepam presented an increase in their toxicity compared to the
isolated substances. This may be due to the presence of TP
C13H25N2O, which showed significative presence in both
samples and only in them.
As already discussed, the mixture containing metformin
+ simvastatin + diazepam showed the highest toxicity
results. In this sample, it was found the compounds C8H13,
C8H15, C6H11N6O and C16H9N2O. It should also be highlighted that the degradation product C8H13 was only found
in this mixture, suggesting that the reported toxicity is
related to the presence of these TPs.
Finally, it is discussed the quaternary mixture, which has
one of the highest toxic effects (16.23%). In this mixture, it
was noted a large quantity of TP C9H11N6O, which was
only identified in this sample. Therefore, it is believed that
this byproduct may be related to the toxicity of the mentioned mixture.
It is known that more tests are necessary for the effective
confirmation of the reasons why some mixtures are more
toxic than others. Despite this, the identification of byproducts confirmed by this work clarifies that when two chemicals are combined, other compounds are generated,
which interferes in the survival of living organisms. Other
authors (De Souza Santos et al. 2014; Hassold and Backhaus 2014; Long et al. 2016) do the same discussion,
pointing to the toxic effects of byproducts formed in mixtures or by degradation processes.
Acute toxicity effects of diazepam, metformin, omeprazole
and simvastatin were measured for the organism Aliivibrio
Diazepam, metformin, omeprazole and simvastatin: a full discussion of individual and mixture acute. . .
fischeri, both individually and in mixtures. In terms of
individual toxicity, omeprazole (EC10: 0.0053 mg/L) shows
the higher toxicity among the active ingredients, followed
by simvastatin (EC10: 0.0093 mg/L), diazepam (EC10:
8.69 mg/L) and, finally, metformin (EC10: 870.79 mg/L). It
was also noticed a significant increase (p value < 0.05) in
toxicity effect while increasing the compound hydrophobicity, as they present higher propensity for interacting
with the cell membrane. Synergism, antagonism, and
hormesis phenomena were found when mixture toxicity was
tested. It was emphasized that hormesis detection is not a
sign that the contaminant is beneficial to the organism, on
the contrary, it could be a case of a contaminant that is toxic
in chronic toxicity tests or an acute toxicity tests in higher
concentration. Furthermore, a series of transformation products was found in these solutions, which may directly
affect their toxicities. These results indicate that although
mathematical models which predict mixture toxicity are
important tools for environmental management, they may
fail to address many possible phenomena. This work points
to the urgent need for more studies about mixtures’ behavior, since chemicals are subject to interactions and modifications, can mix and are able to potentiate or nullify the
toxic effects of each other.
The research data for this article is not available. Data can
be made available on request.
Funding This work was supported by CNPq, CAPES, FAPEMIG,
Pontifical Catholic University of Minas Gerais and Federal University
of Minas Gerais.
Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of
Ethical approval This article does not contain any studies with human
participants or animals performed by any of the authors.
Publisher’s note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
ABNT (2012) Aquatic ecotoxicology — Determination of the inhibitory effect of aqueous samples on the light emission of Vibrio
fischeri (luminescent bacteria test) Part 2: Method using liquiddried bacteria (ABNT NBR 15411-2:2012). 2nd edn. 21 pp.
Avaliable at https://www.abntcatalogo.com.br/norma.aspx?ID=
Allan SE, Smith BW, Tanguay RL, Anderson KA (2012) Bridging
environmental mixtures and toxic effects. Environ Toxicol Chem
Backhaus T, Porsbring T, Arrhenius Å et al. (2011) Single-substance
and mixture toxicity of five pharmaceuticals and personal care
products to marine periphyton communities. Environ Toxicol
Chem 30:2030–2040. https://doi.org/10.1002/etc.586
Baek I-H, Kim Y, Baik S, Kim J (2019) Investigation of the synergistic toxicity of binary mixtures of pesticides and pharmaceuticals on Aliivibrio fischeri in Major River Basins in South Korea.
Int J Environ Res Public Health 16:208. https://doi.org/10.3390/
Baillie TA (2008) Metabolism and toxicity of drugs. Two decades of
progress in industrial drug metabolism. Chem Res Toxicol
Berenbaum MC (1985) The expected effect of a combination of
agents: the general solution. J Theor Biol 114:413–431. https://
Böger B, Do Amaral B, Estevão PL, Da S et al. (2018) Determination
of carbamazepine and diazepam by SPE-HPLC-DAD in Belém
River water, Curitiba-PR/Brazil. Ambient e Agua 13:1. https://
Boyd J, Vrana JA, Williams HN (2013) In vitro approach to predict
post-translational phosphorylation response to mixtures. Toxicology 313:113–121. https://doi.org/10.1016/j.tox.2012.10.010
Calabrese EJ (2008) Hormesis: why it is important to toxicology and
toxicologists. Environ Toxicol Chem 27:1451–1474. https://doi.
Calabrese EJ, Baldwin LA (2003) Toxicology rethinks its central
belief. Nature 421:691–692. https://doi.org/10.1038/421691a
Cedergreen N, Kudsk P, Mathiassen SK et al. (2007) Reproducibility
of binary-mixture toxicity studies. Environ Toxicol Chem
Cedergreen N, Sørensen H, Svendsen C (2012) Can the joint effect of
ternary mixtures be predicted from binary mixture toxicity
results? Sci Total Environ 427–428:229–237. https://doi.org/10.
Cleuvers M (2005) Initial risk assessment for three β-blockers found in
the aquatic environment. Chemosphere 59:199–205. https://doi.
Cleuvers M (2003) Aquatic ecotoxicity of pharmaceuticals including
the assessment of combination effects. Toxicol Lett 142:185–194.
Coors A, Dobrick J, Möder M, Kehrer A (2012) Mixture toxicity of
wood preservative products in the fish embryo toxicity test.
Environ Toxicol Chem 31:1239–1248. https://doi.org/10.1002/
Crépet A, Héraud F, Béchaux C et al. (2013) The PERICLES research
program: an integrated approach to characterize the combined
effects of mixtures of pesticide residues to which the French
population is exposed. Toxicology 313:83–93. https://doi.org/10.
Dahl U, Gorokhova E, Breitholtz M (2006) Application of growthrelated sublethal endpoints in ecotoxicological assessments using
a harpacticoid copepod. Aquat Toxicol 77:433–438. https://doi.
De Souza Santos LV, Teixeira DC, Jacob RS et al. (2014) Evaluation
of the aerobic and anaerobic biodegradability of the antibiotic
norfloxacin. Water Sci Technol 70:265–271. https://doi.org/10.
Dong Y, Fang Z, Xu Y et al. (2019) The toxic effects of three active
pharmaceutical ingredients (APIs) with different efficacy to
Vibrio fischeri. Emerg Contam 5:297–302. https://doi.org/10.
Ellesat KS, Yazdani M, Holth TF, Hylland K (2011) Species-dependent
sensitivity to contaminants: an approach using primary hepatocyte
cultures with three marine fish species. Mar Environ Res
R. S. Jacob et al.
Faust M, Altenburger R, Backhaus T et al. (2003) Joint algal toxicity
of 16 dissimilarly acting chemicals is predictable by the concept
of independent action. Aquat Toxicol 63:43–63. https://doi.org/
Fent K, Weston AA, Caminada D (2006) Ecotoxicology of human
pharmaceuticals. Aquat Toxicol 76:122–159. https://doi.org/10.
Foureaux AFS, Reis EO, Lebron Y et al. (2019) Rejection of pharmaceutical compounds from surface water by nanofiltration and
reverse osmosis. Sep Purif Technol 212:171–179. https://doi.org/
Ginebreda A, Muñoz I, de Alda ML et al. (2010) Environmental risk
assessment of pharmaceuticals in rivers: relationships between
hazard indexes and aquatic macroinvertebrate diversity indexes in
the Llobregat River (NE Spain). Environ Int 36:153–162. https://
Godoy AA, Kummrow F, Pamplin PAZ (2015) Ecotoxicological
evaluation of propranolol hydrochloride and losartan potassium
to Lemna minor L. (1753) individually and in binary mixtures.
Ecotoxicology 24:1112–1123. https://doi.org/10.1007/s10646-
Goel S, Evans-Johnson JA, Georgieva NI, Boysen G (2013) Exposure
profiling of reactive compounds in complex mixtures. Toxicology
Gomez-Eyles JL, Svendsen C, Lister L et al. (2009) Measuring and
modelling mixture toxicity of imidacloprid and thiacloprid on
Caenorhabditis elegans and Eisenia fetida. Ecotoxicol Environ
Saf 72:71–79. https://doi.org/10.1016/j.ecoenv.2008.07.006
Hassold E, Backhaus T (2014) The predictability of mixture toxicity of
demethylase inhibiting fungicides to Daphnia magna depends on
life-cycle parameters. Aquat Toxicol 152:205–214. https://doi.
Hertzberg RC, Pan Y, Li R et al. (2013) A four-step approach to
evaluate mixtures for consistency with dose addition. Toxicology
Johnson LA, Welch B, Whitfield SM (2013) Interactive effects of
pesticide mixtures, predators, and environmental regimes on the
toxicity of two pesticides to red-eyed tree frog larvae. Environ
Toxicol Chem 32:2379–2386. https://doi.org/10.1002/etc.2319
Kaiser KLE (1998) Correlations of Vibrio fischeri bacteria test data
with bioassay data for other organisms. Environ Health Perspect
Kaiser KLE, McKinnon MB, Fort FL (1994) Interspecies toxicity
correlations of rat, mouse and Photobacterium phosphoreum.
Environ Toxicol Chem 13:1599–1606. https://doi.org/10.1002/
Kaiser KLE, Palabrica VS (1991) Photobacterium phosphoreum
toxicity data index. Water Pollut Res J Can 26:361–431. https://
Lindim C, de Zwart D, Cousins IT et al. (2019) Exposure and ecotoxicological risk assessment of mixtures of top prescribed pharmaceuticals in Swedish freshwaters. Chemosphere 220:344–352.
Løkke H (2010) Novel methods for integrated risk assessment of cumulative
stressors—results from the NoMiracle project. Sci Total Environ
Løkke H, Ragas AMJ, Holmstrup M (2013) Tools and perspectives for
assessing chemical mixtures and multiple stressors. Toxicology
Long X, Wang D, Lin Z et al. (2016) The mixture toxicity of environmental contaminants containing sulfonamides and other antibiotics in Escherichia coli: differences in both the special target
proteins of individual chemicals and their effective combined
concentration. Chemosphere 158:193–203. https://doi.org/10.
Meek MEB (2013) International experience in addressing combined
exposures: increasing the efficiency of assessment. Toxicology
Moser VC, Padilla S, Simmons JE et al. (2012) Impact of chemical
proportions on the acute neurotoxicity of a mixture of seven
carbamates in preweanling and adult rats. Toxicol Sci
Muñoz I, José Gómez M, Molina-Díaz A et al. (2008) Ranking
potential impacts of priority and emerging pollutants in urban
wastewater through life cycle impact assessment. Chemosphere
Neuparth T, Martins C, Santos CBdl et al. (2014) Hypocholesterolaemic pharmaceutical simvastatin disrupts reproduction and
population growth of the amphipod Gammarus locusta at the ng/
L range. Aquat Toxicol 155:337–347. https://doi.org/10.1016/j.
Nilsen E, Smalling KL, Ahrens L et al. (2019) Critical review: grand
challenges in assessing the adverse effects of contaminants of
emerging concern on aquatic food webs. Environ Toxicol Chem
Nunes B, Carvalho F, Guilhermino L (2005) Acute toxicity of widely used
pharmaceuticals in aquatic species: Gambusia holbrooki, Artemia
parthenogenetica and Tetraselmis chuii. Ecotoxicol Environ Saf
Ortiz de García SA, Pinto Pinto G, García-Encina PA, Irusta-Mata R
(2014) Ecotoxicity and environmental risk assessment of pharmaceuticals and personal care products in aquatic environments
and wastewater treatment plants. Ecotoxicology 23:1517–1533.
Pascoe D, Karntanut W, Müller CT (2003) Do pharmaceuticals affect
freshwater invertebrates? A study with the cnidarian Hydra vulgaris. Chemosphere 51:521–528. https://doi.org/10.1016/S0045-
Patel M, Kumar R, Kishor K et al. (2019) Pharmaceuticals of emerging
concern in aquatic systems: chemistry, occurrence, effects, and
removal methods. Chem Rev 119:3510–3673. https://doi.org/10.
Peña-Guzmán C, Ulloa-Sánchez S, Mora K et al. (2019) Emerging
pollutants in the urban water cycle in Latin America: a review of
the current literature. J Environ Manag 237:408–423. https://doi.
Phyu YL, Palmer CG, Warne MSJ et al. (2011) A comparison of Lovastatin
mixture toxicity assessment: examining the chronic toxicity of
atrazine, permethrin and chlorothalonil in mixtures to Ceriodaphnia cf. Dubia. Chemosphere 85:1568–1573. https://doi.
Prichard PJ, Yeomans ND, Mihaly GW et al. (1985) Omeprazole: a
study of its inhibition of gastric pH and oral pharmacokinetics
after morning or evening dosage. Gastroenterology 88:64–69.
Reis EO, Foureaux AFS, Rodrigues JS et al. (2019) Occurrence,
removal and seasonal variation of pharmaceuticals in Brasilian
drinking water treatment plants. Environ Pollut 250:773–781.
Ribeiro S, Torres T, Martins R, Santos MM (2015) Toxicity screening of
diclofenac, propranolol, sertraline and simvastatin using danio rerio
and paracentrotus lividus embryo bioassays. Ecotoxicol Environ
Saf 114:67–74. https://doi.org/10.1016/j.ecoenv.2015.01.008
Tambosi JL, de Sena RF, Favier M et al. (2010) Removal of pharmaceutical compounds in membrane bioreactors (MBR) applying
submerged membranes. Desalination 261:148–156. https://doi.
Taylor D, Senac T (2014) Human pharmaceutical products in the
environment—the “problem” in perspective. Chemosphere
Diazepam, metformin, omeprazole and simvastatin: a full discussion of individual and mixture acute. . .
Ukić Š, Sigurnjak M, Cvetnić M et al. (2019) Toxicity of pharmaceuticals in binary mixtures: assessment by additive and nonadditive toxicity models. Ecotoxicol Environ Saf 185:109696.
Wang LJ, Liu SS, Zhang J, Li WY (2010) A new effect residual ratio
(ERR) method for the validation of the concentration addition and
independent action models. Environ Sci Pollut Res
Webster TF (2013) Mixtures of endocrine disruptors: how similar
must mechanisms be for concentration addition to apply?
Toxicology 313:129–133. https://doi.org/10.1016/j.tox.2013.
Yang LH, Ying GG, Su HC et al. (2008) Growth-inhibiting effects of
12 antibacterial agents and their mixtures on the freshwater
microalga Pseudokirchneriella subcapitata. Environ Toxicol
Chem 27:1201–1208. https://doi.org/10.1897/07-471.1
Zhao YH, Wang LS, Gao H, Zhang Z (1993) Quantitative structureactivity relationships—relationship between toxicity of organic chemicals to fish and to Photobacterium phosphoreum. Chemosphere
Zou X, Lin Z, Deng Z et al. (2012) The joint effects of sulfonamides and
their potentiator on Photobacterium phosphoreum: differences
between the acute and chronic mixture toxicity mechanisms.
Chemosphere 86:30–35. https://doi.org/10.1016/j.chemosphere.
Zuriaga E, Lomba L, German B et al. (2019) Ecotoxicity in Aliivibrio
fischeri of ibuprofen, omeprazole and their mixtures. Chem Ecol
R. S. Jacob et al.