Environ Sci Pollut Res

Supplementary Material

Approach for detecting mutagenicity of biodegraded and ozonated pharmaceuticals, metabolites and transformation products from a drinking water perspective

Stefan Gartiser • Christoph Hafner • Kerstin Kronenberger-Schäfer • Oliver Happel • Christoph Trautwein • Klaus Kümmerer

1 Selection of the parent compounds and metabolites considered in the study

For selecting the target compounds to be investigated a tiered approach was applied. There are several thousand active pharmaceutical ingredients on the market (2,385 in Germany in 2008) and even more transformation products. First data on usage were collected and the tonnage of active pharmaceutical ingredients (API) was calculated from the defined daily dosages (DDD) according to WHO. For those APIs with a consumption of > 0.5 tonnage per year the excretion rate was taken into account. Furthermore a literature survey on transformation products and metabolites was performed. Quantitative structure activity relationship models (QSARs) were applied using the software packages EPI Suite (Estimation Program Interface, US EPA) and MultiCase (Computer Automated Structure Evaluation, MultiCASE Inc., Beachwood). The mobility in the aquatic environment was assessed through the surrogate parameters logPow and biodegradability potential. Genotoxic and mutagenic properties were estimated from the structural elements of the molecules such as nitro-compounds or nitogen-oxides. Additional, a review of the toxicological literature of the candidate compounds has been performed with respect to their metabolites and their toxic and fate properties. The commercial availability was checked in order to allow for experimental testing. The results of this pre-selection procedure are described in detail by Kümmerer et al. (2009).

The followingparent compounds and metabolites have been included in this study:

Metamizol [CAS 50567-35-6] is used as analgesic, antipyretic and antirheumatic pro-drug. After oral administration it is rapidly hydrolyzed to the active form 4-N-Methylaminoantipyrine in the stomach, which is further metabolised to 4-aminoantipyrine, 4-Formylaminoantipyrine (FAA) [CAS 1672-58-8], and 4-Acetylaminoantipyrine (AAA) [CAS 83-15-8] (Vlahov et al., 1990). The last two metabolites have been detected in the outflow of STPs and in surface water (up to 0.9 and 0.8 µg L1, ARGE 2003) and therefore have been included in our study. Sulfasalazin [CAS 599-79-1] is used as an anti-inflammatory agent in the treatment of inflammatory bowel diseases (Crohn's disease, ulcerative colitis) and for rheumatoid arthritis. In the gastrointestinal tract the pro-drug is metabolised to Sulfapyridine [CAS 144-83-2], which has antibacterial properties and has been detected in the outflow of STPs (AWEL 2005). This metabolite has therefore been considered in the study. Metformin [CAS 657-24-9] is an oral anti-diabetic drug that is not metabolised in the body. Also Piracetam [CAS 7491-74-9], which is applied against neurological dysfunction such as Alzheimer’s disease and that is completely resorbed in the gastrointestinal tract but not metabolised in the body. Biodegradation testing of Metformin resulted in that the compound is not completely mineralised, leading to the recalcitrant TP Guanylurea [CAS 141-83-3] (Trautwein and Kümmerer 2011) that also has been considered in this study.

2 Inherent and ultimate biodegradability

The biodegradability extents determined in the inherent CO2/DOC-combination test and the Zahn-Wellens test are summarised summarised in table 1S.

Table 1S: Inherent biodegradability

compound / DOC-elimination
/ % / CO2-evolution
/ %
Diethylene glycol / 100.9 (vessel 1)
100.7 (vessel 2)
99.7 (ZWT) / 95.4 (vessel 1)
94.0 (vessel 2)
n.d.
4-N-Formylaminoantipyrine (FAA) / 6.9 (vessel 1)
8.4 (vessel 2)
0.5 (ZWT) / 2.5 (vessel 1)
-2,4 (vessel 2)
n.d.
4-N-Acetylaminoantipyrine (AAA) / 55.3 (vessel 1)
24.5 (vessel 2)
18.8 (ZWT) / 40.1 (vessel 1)
20.3 (vessel 2)
n.d.
Sulfapyridine / 62.7 (vessel 1)
50.3 (vessel 2)
50.4 (ZWT) / 49.3 (vessel 1)
44.5 (vessel 2)
n.d.
Metformin / 49.4 (vessel 1)
51.6 (vessel 2)
47.9 (ZWT) / 43.8 (vessel 1)
40.2 (vessel 2)
n.d.
Piracetam / 100.6 (vessel 1)
100.9 (vessel 2) / 91.2 (vessel 1)
95.5 (vessel 2)

Vessel 1 and 2: CO2/DOC-combination test

ZWT: Zahn-Wellens test

n.d.: not determined

The reference compound diethylene glycol was ultimately biodegraded in 7 days and indicates the reliability of the test design (fig. 1S).

Fig. 1S: DOC-elimination and CO2-evolution of diethylene glycol

3. Chemical Analysis

Specifications of the chemical analysis using a HPLC (1100 series LC and 1200 series LC, Agilent Technologies) and a mass spectrometer, equipped with positive electro spray ionization (+ESI) (API 2000, Applied Biosystems):

The analysis of Metformin and Piracetam was performed by using a Luna C18 column (250 mm x 2 mm, 5 µm; Phenomenex). The composition of the mobile phase (A) was of 20 mmolL1 ammonia formiate in water and mobile phase (B) 20 mmolL1 ammonia formiate in acetonitril-methanol 2:1 (v:v). The elution profile for the mobile phase (B) was 20% to 50% in 0 to 8 min, 50% to 100% in 1 min and 100% for 6 min with a flow rate of 0.2 mL min1. The injection volume was 12 µL. The precursor ion and the most intense and specific fragment ions in the multiple reaction monitoring mode are m/z = 129.9 > 59.9 and 70.9 for Metformin, m/z = 142.9 > 97.9 and 125.8 for Piracetam, respectively.

The analysis of 4-Formylaminoantipyrine (FAA) and 4-Acetylaminoantipyrine (AAA) was performed by using a Nucleosil column (250 mm x 2 mm, 3 µm; Macherey-Nagel) and the composition of the mobile phase (A) was 20 mmolL1 ammonia acetate in water and mobile phase (B) 20 mmolL1 ammonia acetate in acetonitril-methanol 2:1 (v:v). The chromatographic separation method was as described above. The precursor ion and the most intense and specific fragment ions in the multiple reaction monitoring mode are m/z = 246.0 > 83.3 and 104.0 for AAA, m/z = 231.9 > 82.9 and 103.9 for FAA, respectively.

The analysis of Sulfapyridine was performed by using a MZ Aqua Perfect column (250 mm x 2.1 mm, 5 µm; MZ Analysentechnik GmbH) and the composition of the mobile phase (A) was 20 mmolL1 ammonia acetate in water adjusted to pH 5.1 with acetic acid and mobile phase (B) 20 mmol L-1 ammonia acetate in acetonitril-methanol 2:1 (v:v). The elution profile for the mobile phase (B) was 20% to 100% in 0 to 10 min and 100% for 8 min with a flow rate of 0.2 mL min1. The injection volume was 15 µL. The precursor ion and the most intense and specific fragment ions in the multiple reaction monitoring modes are m/z = 250.1 > 156.2 and 92.2 for Sulfapyridine. The results of the accompanying analysis of AAA is shown in figure 8a of the main part of the publication, the graphs for the other test items are shown in figures 4aS to 8aS (see chapter 3).

Analysis of the very polar Guanylurea was performed on a HPLC system (Agilent Technologies 1100 series, Waldbronn, Germany) coupled to an atmospheric pressure electrospray ionization (AP-ESI) ion trap mass spectrometer (Bruker Esquire6000+, Bruker Daltonik GmbH, Bremen, Germany).

Chromatographic separation was performed on a Nucleodur® CN-RP cyanopropyl modified high purity silica gel column (125 mm x 4 mm, 5 µm; Macherey-Nagel). Separation was performed with an isocratic flow of 10 mmol L1ammonium acetate in water. Flow rate was 0.6 mL min1 and total run time 6 min. The sample injection volume was 5 μL. Retention time of Guanylurea was 2.9 min. In the mass spectrometer the precursor ion of Guanylurea was found at m/z= 103.4 and gave product ions at m/z = 60.7 and m/z = 86.4 in the MS/MS mode (fig. 2S).

As a result of the high starting concentrations of the compoundsand the good sensitivity of the detection system direct quantification without an enrichment step was possible in all methods.

Fig. 2S: Elimination of Guanlyurea (m/z = 103.4) by ozonation and formation of a sole transformation product at m/z = 116.3: Extracted ion currents (EIC) of both compounds

4. Ozonation

The ozone stock solution was generated in a bubble column reactor containing 5 L demineralized water with a phosphate buffer (66 mg L1) for maintaining a pH of 5 to 6. An ozone generator (Ozomat COM-AD-02, Anseros, Tübingen, Germany) was used to produce ozone from oxygen by dielectric barrier discharge. The gas flow was passed through the reactor, which was maintained at 5 °C. The ozone waste of the gas flow was destroyed by a washing bottle containing thiosulfate solution. After approximately1 hour an ozone saturation of 20 mg L1 to 30 mg L1 was reached in the stock solution, as photometrically determined via the extinction of the dye Indigotrisulfonate (DIN 38408-3). This stock solution was used for the ozonation of the different dissolved compounds.

The bubble column reactor for preparing the ozone stock solution by means of an ozone generator is shown in figure 3.

Fig. 3: Ozone generation by dielectric barrier discharge, production of the ozone stock solution in the bubble column, and setup for batch experiments

The Elimination of AAA by ozonation is shown in figure 8b of the main part of the publication. The corresponding graphs for the other test items are shown in figures 4bS to 9S.

Fig. 4aS: Primary degradation of FAA / Fig. 4bS: Elimination of FAA by ozonation
Fig. 5aS: Primary degradation of Sulfapyridine / Fig. 5bS: Elimination of Sulfapyridine by ozonation
Fig. 6S: Elimination of Sulfapyridine by ozonation at higher concentration
Fig. 7aS: Primary degradation of Metformin / Fig. 7bS: Elimination of Meformin by ozonation
Fig. 8aS: Primary degradation of Piracetam / Fig. 8bS: Elimination of Piractam by ozonation
Fig. 9S: Elimination of Guanylurea by ozonation

5. Mutagenicity testing

The Ames test was carried out according to ISO 16240 (2005) with Salmonella typhimurium strain TA98, which detects frameshift mutagens, and strain TA100 which is susceptible for base pair substitution mutagens (point mutations). The number of reverted bacteria (revertants) in relation to the spontaneous mutation revertants is expressed as induction rate (IR) and provides a measure of the mutagenic potential. In higher organisms certain mutagens are first activated by metabolic processes (pro-mutagens) or become inactivated. Because the bacteria lack of these the test was carried out with and without Aroclor induced rat liver extract S9 (Moltox, USA) as an exogenous metabolic activation system. Each concentration has been tested with 5 replicates. The following reference substances were tested in each testing series: 2-aminoanthracene (2 μg per plate for TA98 and 2.5 μg per plate for TA100 both with S9), 2-nitrofluoren (1.5 μg per plate for TA98 without S9), and sodium azide (0.5 μg per plate for TA100 without S9).

The detailed data of the Ames test are documented in tables 2 to 8. None of the Ames tests revealed a significant mutagenic effect. The results of the umu test are shown in table 9.

Table 2: Ames test with the blank controls after biodegradation and ozonation

Table 3: Ames test with FAA after biodegradation and ozonation

Table 4: Ames test with AAA after biodegradation and ozonation

Table 5: Ames test with Sulfapyridine after biodegradation and ozonation

Table 6: Ames test with Metformin after biodegradation and ozonation

Table 7: Ames test with Guanylurea after ozonation

Table 8: Ames test with Piracetam after ozonation

Table 9: umu test with Sulfapyidine after ozonation

5. References

ARGE, Arbeitsgemeinschaft für die Reinhaltung der Elbe (2003) Arzneistoffe in Elbe und Saale.

AWEL, Amt für Abfall, Wasser, Energie und Luft (2005) Organische Spurenstoffe im Grundwasser des Limmattales / Ergebnisse der Untersuchungskampagne 2004. In Zusammenarbeit mit Eawag – Das Wasserforschungs-Institut des ETH-Bereichs und Kantona-les Labor Zürich.

DIN 38408–3 (1993) German standards for the examination of water,wastewater, and sludge. Determination of ozone (G3). Wiley-VCH, Weinheim (in German)

Kümmerer K, Schuster A, Längin A, Happel O, Thoma A, Schneider, K, Hassauer M, Gartiser S, Hafner C (2009) Identification and assessment of selected pharmaceuticals and their metabolites (degradation and transformation products) in the water cycle. Final Report FKZ 206 61 202 for the German Federal Environment Agency. Freiburg (in German)

Trautwein C, Kümmerer K (2011) Incomplete degradation of the antidiabetic Metformin and identification of the microbial dead-end transformation product Guanylurea. Chemosphere 85(5), 765–773

Vlahov V, Badian M, Verho M, Bacracheva N (1990) Pharmacokinetics of metamizol metabolites in healthy subjects after a single oral dose of metamizol sodium. European Journal of Clinical Pharmcology 38(1), 61–65 990

Page 1 of 10