ars.els-cdn.com · Web viewFigure S 10: Time trends of lipid-normalized concentrations [ng/g lipid weight, lw] of UV-350 in bream liver samples from the sampling site Koblenz (river

Supporting informationTable S 1: Overview of sediment (SD), suspended particulate matter (SPM) and bream liver (BL) samples from

German rivers used for the monitoring of BUVSs. Samples from adjacent sites at the same river were assigned

with the same code.

River Code Sampling site River km Average discharge

[m3/s]a

Catchment size

[km2]a

Matrix Month/Year

TOCb

[%]LCc

[%]

Rhine

R1 Iffezheim 333.6 1250d 50,000d SD 2004 3.1 n.a.R2 Koblenz

(harbor)591.4 1700 110,000 SD 09/2013 4.3 n.a.

R2 Koblenz 590.3 1700 110,000 SPM 03/2013 5.3 n.a.R2 Koblenz 590-591.4 1700 110,000 BL 2011 n.a. 42

Elbe

E1 Meissen 83.3 330e 54,000 SD 03/2012 5.7 n.a.E1 Zehren 83-98 330e 54,000 BL 2010 n.a. 23E2 Wittenberg

(harbor)216.4 360 62,000 SD 10/2013 4.8 n.a.

E2 Wittenberg 216.6 360 62,000 SPM 11/2013 6.7 n.a.E3 Barby (harbor) 295.6 560f 100,000f SD 10/2013 4.9 n.a.E4 Cumlosen

(harbor)469.6 710g 125,000 SD 07/2012 5.6 n.a.

E4 Cumlosen 469.9 710g 125,000 SPM 06/2013 10.3 n.a.

SaaleSL1 Wettin 70.4-86.5 100h 18,000h BL 2011 n.a. 37SL2 Rosenburg 2.0 115 24,000 SD 01/2012 3.8 n.a.

Saar

SR Rehlingen 54.2 75i 7000i SPM 2013 5.2 n.a.SR Dillingen

(harbor)56.4 75i 7000i SD 07/2013 5.7 n.a.

SR Rehlingen 50.7- 54.2 75i 7000i BL 2011 n.a. 28Moselle M Koblenz 2 310 28,000 SPM 11/2013 4.6 n.a.

n.a.: not applicable; a Data provided by the German Working Group on water issues (Bund/Länder-Arbeitsgemeinschaft

Wasser, LAWA) and the online information platform Undine (http://undine.bafg.de/servlet/is/8606/, assessed 29.11.2015);b TOC: total organic carbon; c LC: lipid content (dry weight basis); d water-level gauge Karlsruhe-Maxau (km 362.3); e water-

level gauge Dresden (km 55.6); f water-level gauge Magdeburg (km 326.6); g water-level gauge Neu-Darchau (km

536.5); h water-level gauge Halle-Trotha (km 89.1); i water-level gauge Fremersdorf (km 48.5).

1

1

2

3

4

5

6789

10

11

12

13

14

15

16

17

12

Table S 2: Overview of the two mass transitions used for quantification (MRM 1) and confirmation (MRM 2) and

their optimized mass spectrometric parameters.

Substance Mass MRM 1 [m/z] MRM 2 [m/z] DP [V]CE [eV]

(MRM 1/MRM 2)

CXP [V]

(MRM 1/MRM 2)

UV-326 316.2 316.2/260.2 316.2/107.1 50 30/35 15/19

UV-320 323.2 324.2/268.2 324.2/212.2 90 30/38 12/12

UV-329 323.2 324.2/212.2 324.2/57.1 50 34/49 10/10

UV-350 323.2 324.2/268.2 324.2/212.2 60 29/35 10/10

UV-328 351.2 352.2/282.2 352.2/212.2 60 33/43 12/12

UV-327 357.2 (35Cl) 358.2/302.2 358.2/246.2 60 30/40 13/13

UV-928 441.3 442.3/364.3 442.3/252.2 70 37/46 15/15

UV-234 447.2 448.2/370.2 448.2/119.2 50 28/44 15/15

UV-360 658.4 659.4/336.2 659.4/224.1 30 35/50 20/20

DP: Declustering potential, CE: Collision energy, CXP: Cell exit potential

Analysis of the lipid content in bream livers

Briefly, 0.5 g of the freeze-dried liver samples were weighed into a 15 mL centrifuge tube followed

by the addition of 1.6 mL propan-2-ol, 2 mL cyclohexane and 2 mL ultrapure water. After mixing with

an Ultra Turrax (T8, IKA-Werke GmbH & Co. KG, Staufen, Germany) for 2 min at 25000 rpm, the

sample was centrifuged for 10 min at 2000 rpm. The upper organic phase was transferred to a pre-

weighed aluminum pan. The extraction was repeated with 2 mL of a mixture of cyclohexane and

propan-2-ol (87/13; v/v). The upper organic phase was combined with the first extract and dried to

constant weight. The lipid content was calculated as the difference between the intake weight and

the mass of the residue.

Selection of GPC method

The selection of the final GPC procedure was based on the results of an assessment of two

different GPC methods. For both methods a GPC column filled with Bio Beads S X3 material (bed

length 320 mm, inner diameter 25 mm, 200-400 mesh) from LCTech (Dorfen, Germany) was used.

The mobile phase of method 1 consisted of a mixture of cyclohexane and acetone (3/1; v/v), while a

mixture of cyclohexane and ethyl acetate (1/1; v/v) was used as mobile phase for method 2. Based

on pre-tests with standard solutions a collection time of 20-36 min and 16-32 min (target fractions)

was determined for method 1 and 2, respectively. Applying both methods to spiked bream liver

extracts revealed a visible breakthrough of lipids to the target fraction of method 1. Moreover, about

20 % of UV-360 already eluted in fraction 1 prior to the target fraction (Figure S 1a) leading to a low

2

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

34

recovery of only 60% (Figure S 1c). Hence, starting collection earlier in order to increase the removal

of lipids would even have led to higher losses of UV-360. In contrast, using method 2 the amount of

target compounds eluting prior or after the target fraction were negligible (Figure S 1b), recoveries in

the target fraction were >80% (Figure S 1c) and the target fractions were free of any visible lipids

residuals. Therefore, method 2 based on elution with cyclohexane and ethyl acetate was chosen as

the final GPC procedure.

Figure S 1: Ratios [% of total amount detected] of BUVSs in the different GPC fractions of spiked bream liver

extract using a) GPC method 1 and b) GPC method 2. c) Recoveries [%] of BUVs in the target fraction 2 of the

extracts. The analyses were performed in triplicates. The error bars represent the standard deviation. CyH:

cyclohexane, Ac: acetone, EtAc: ethyl acetate.

3

42

43

44

45

46

47

48

49

50

51

52

53

54

56

Figure S 1: LC-MS/MS chromatogram of a 5 ng/mL standard solution of BUVSs ( 1) UV-329, 2) UV-350, 3) UV-

326, 4) UV-320, 5) UV-234, 6) UV-328, 7) UV-327, 8) UV-928, 9) UV-360). The zoomed area of the

chromatogram shows that interferences of the BUVS isomers UV-320, UV-350 and UV-329 exhibiting the same

precursor and product ions can be excluded due to their sufficient chromatographic separation.

Figure S 2: Average percentage ratio of BUVS concentrations in bream liver (n=3) and SPM (n=17) determined

by external calibration (cext.cal.) and by applying the standard addition method (cstd.add.). The error bars represent

the standard deviations. Ratios for bream liver samples could not be determined for UV-326, UV-329, UV-928

and UV-360, since their concentrations were <LOQ.

4

55

56

57

58

59

60

6162

63

64

65

78

Figure S 4: Average TOC normalized concentrations [ng/g TOC] of BUVSs in a) sediments and b) suspended particulate matter and lipid normalized concentrations [ng/g lw] in c)

bream liver samples from different sampling sites along German rivers. Capped bars represent the maximum and minimum values of duplicates. For the bream liver sample

from site R2 only a single sample existed (marked by a plus symbol). Concentrations <LOQ are indicated by an asterisk. Where no bar and no asterisk are drawn, no sample was

available.

5

1

2

34

5

6

7

910

Figure S 5: Pearson correlation matrix of BUVS concentrations in sediments. Correlation diagrams are shown

together with the corresponding correlation coefficients and p-values

Figure S 6: Time trends of the total organic carbon [%] of SPM samples from the river Rhine in Koblenz (river km

590.3) and the river Saar in Rehlingen (river km 54.2). The black lines represent the results of the linear

regression (solid for significant linear trend, dashed for not significant) and the dashed blue line the result of

the LOESS smoother (no significant non-linear trend). The grey-shaded areas mark the 95 % confidence

intervals of the LOESS function.

6

12

3

4

5

6

7

8

9

10

1112

7

11

1314

8

12

13

14

15

1516

9

16

17

18

19

20

21

22

1718

Figure S 7: Time trends of the non-normalized concentrations [ng/g dw] of BUVSs in SPM and bream liver

samples from the sampling site Koblenz (river Rhine, km 590.3) as well as in SPM from Rehlingen (river Saar, km

54.2). The lines represent the results of the linear regression (black; solid for significant linear trend, dashed for

not significant) and the LOESS smoother (blue; solid for significant non-linear trend, dashed for not significant).

The grey-shaded areas mark the 95 % confidence intervals of the LOESS function. In general, measurements

were performed in duplicate and the non-capped bars represent the maximum and minimum values.

Frequently, concentrations in SPM were determined by multiple measurements (n≥3) as indicated by the

capped error bars which represent the corresponding 95% confidence intervals. For the sum of BUVSs only

average values are plotted.

Figure S 8: Time trends of the average annual flow rate [m3/s] of the river Rhine in Koblenz (river km 590.3) and

the river Saar in Rehlingen (river km 54.2). The dashed black lines represent the result of the linear regression

(no significant linear trend) and the dashed blue line the result of the LOESS smoother (no significant non-linear

trend). The grey-shaded areas mark the 95 % confidence intervals of the LOESS function.

Figure S 9: Time trend of the lipid content [% wet weight, n=2] of annual composite samples of bream liver

collected between 1995 and 2013 at the sampling site Koblenz (site R2, river Rhine, km 590.0- 591.4). The solid

black line represents the result of the linear regression (significant linear trend, p<0.01) and the dashed blue

10

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

3940

41

42

1920

line the result of the LOESS smoother (no significant non-linear trend). The grey-shaded areas mark the 95 %

confidence intervals of the LOESS function.

Figure S 10: Time trends of lipid-normalized concentrations [ng/g lipid weight, lw] of UV-350 in bream liver

samples from the sampling site Koblenz (river Rhine, km 590.3) considering a time period from 1995 to 2005

(left) and from 2005 to 2013 (right). The solid black line represents the result of the linear regression

(significant linear trend, p<0.01) and the dashed blue line the result of the LOESS smoother (no significant non-

linear trend). The grey-shaded areas mark the 95 % confidence intervals of the LOESS function. The non-capped

bars represent the maximum and minimum values of duplicate measurements.

11

43

44

45

46

47

48

49

50

51

52

2122

Figure S 11: Average relative concentrations [% of c0] of BUVSs in sediment-water systems incubated for 100 d

at 20 ± 1 °C in a climate cabinet (n=3). The initial spike concentration in the supernatant was 5 µg/L. The error

bars represent the standard deviation. Results shown for UV-360 are from native concentrations in sediments

of the non-spiked controls (n=3), since the replicates of the spiked samples exhibited unacceptable high

variations.

12

53

54

55

56

57

58

59

60

2324