Microbial binding of per- and polyfluorinated alkyl substances (PFASs) : - Analysis of PFASs in microbes with ultra-performance liquid chromatography – tandem mass spectrometry (UPLC-MS/MS)

(1)

Degree project

Microbial binding of per- and polyfluorinated alkyl

substances (PFASs)

- Analysis of PFASs in microbes with ultra-performance liquid

chromatography – tandem mass spectrometry (UPLC-MS/MS)

by: Karolina Majdak

Supervisor: Ingrid Ericson Jogsten Assistant supervisor: Jana Jass Examinator: Leo Yeung

School of Science and Technology

Analytical Chemistry with focus on Forensics Project work in Chemistry, 15hp

(2)

1. Abbreviations

PFAS Per- and polyfluorinated alkyl compounds PFBS Perfluorobutane sulfonic acid

PFHxS Perfluorhexane sulfonic acid PFOS Perfluorooctane sulfonic acid PFBA Perfluorobutanoic acid PFPeA Perfluoropentanoic acid PFHxA Perfluorohexanoic acid PFHpA Perfluoroheptanoic acid PFOA Perfluorooctanoic acid PFNA Perfluorononanoic acid PFDA Perfluorodecanoic acid 4:2 FTSA 4:2 fluorotelomer sulfonic acid 6:2 FTSA 6:2 fluorotelomer sulfonic acid 8:2 FTSA 8:2 fluorotelomer sulfonic acid AFFFs Aqueous film-forming foams LC Liquid chromatography NFA National food agency MS Mass spectrometry MS/MS Tandem mass spectrometry POPs Persistent organic pollutants

(3)

Abstract

Per- and polyfluorinated alkyl substances (PFASs) belong to a large group of man-made chemicals that pollute the environment. Perfluorooctane sulfonic acid (PFOS) and perfluorooctanoic acid (PFOA) are the most commonly found PFASs. The pollution of PFASs can be caused among others by using of aqueous fire-fighting foams (AFFFs). PFASs are persistent compounds; that can travel long distances and bioaccumulate in biota. There are several exposure routes for PFASs, but the most common are via food and drinking water. A possible way for PFASs to enter the food chain is by adsorption to microbes. In this project, binding of PFASs to three gram-negative bacteria, Eschericha coli, Acidovorax delafieldii and Pseudomonas nitroreducens, was assessed. Microbes were exposed for fluorinated compounds in environmental water samples and a PFAS-11 solution with 11 PFAS substances prepared in the laboratory. The binding seems to be preferential to the most abundant compounds, PFOS, since the second most abundant compound in the samples was PFHxS with concentrations at one third of the PFOS concentration but nonetheless PFHxS was not detected in any of the samples.

The binding of mainly one PFAS was identified; PFOS was bound at highest concentrations in E. coli treated with both environmental water sample and a PFAS-11 solution. Low concentrations of FOSA and PFDoDS were identified in E. coli and PFNA in A. delafieldii. Only PFOS was detected in P. nitroreducens. The concentrations of other PFASs were below their respective method detection limits.

(4)

7. Conclusion ... 18 8. Acknowledgements ... 19 9. References ... 20 10. Appendix ... 22 Ändrad fältkod Ändrad fältkod Ändrad fältkod Ändrad fältkod Ändrad fältkod Ändrad fältkod Ändrad fältkod Ändrad fältkod Ändrad fältkod Ändrad fältkod Ändrad fältkod Ändrad fältkod Ändrad fältkod Ändrad fältkod

(5)

2. Introduction

Per- and polyfluorinated alkyl substances (PFASs) are highly persistent chemicals; some of them can bioaccumulate in humans and animals (Conder et al, 2008). Food (Fromme et al, 2009) and drinking water (Skutlarek, Exner and Färber, 2006) are believed to be the main source of contamination of PFASs for humans. Uptake of pollutants by microorganism is considered to be a way of entering of the pollutants into the food chain (Morrison et al, 1998). Uptake of organic pollutants by microbes may also have an important role in degradation of pollutants in the environment.

Bacteria play an important role in the environment (Ford, 1994). They produce carbon dioxide, water and inorganic salts out of organic matter. Because of fast cell growth and division, they must adapt to the environment in order to survive (Ayangbenro and Babalola, 2017). Therefore, microorganisms have evolved mechanisms of resistance to pollutants due to their degradative enzymes. There have been no studies on microbial binding of fluorinated compounds previously published, as far as we know. However, it has been shown that microorganisms have developed resistance for heavy metals (Ayangbenro and Babalola, 2017), antibiotics, and other environmental pollutants (Ford, 1994).

In this project, three different Gram-negative bacteria, Escherichia coli, Acidovorax delafieldii and Pseudomonas nitroreducens were exposed to PFASs in order to determine the adsorption of the different substances. The bacteria were exposed to PFASs in different solutions – surface water from a fire-fighting training site, which is an environmental sample; a PFAS-11 mixture (containing 11 different PFAS substances) prepared in the laboratory with the same concentrations of PFASs as in the environmental water; and K-media, which does not contain any PFASs and was used as a control.

2.1 Aim

The purpose of this project was to assess the binding capacity of per- and polyfluorinated compounds in E. coli, A. delafieldii and P. nitroreducens. In this study, the binding capacity of these environmental pollutants to various species of bacteria was, to the best of our knowledge, assessed for the first time. Also, to compare the binding capacity of different species of bacteria exposed for environmental water and PFAS-11 solution.

2.2 Background

PFASs is a large group of man-made chemicals with highly fluorinated structures. (Naturvårdsverket, 2018). Perfluorinated compounds are organic compounds with all hydrogen atoms on the carbon chain replaced with fluorine atoms. The chemical bond between fluorine and carbon is one if the strongest bonds existing and is the reason for the high stability of PFASs. Polyfluorinated compounds, on the other hand, have only some of the hydrogen atoms replaced by fluorine. Therefore, these are less stable than perfluorinated compounds and can be degraded to perfluorinated compounds in the environment. Examples of per- and polyfluorinated compounds are shown in Figures 1 and 2.

Figure 1: Polyfluorinated compound (6:2 FTSA - Figure 2: Perfluorinated compound (PFOS-

(6)

Perfluorooctanoic acid (PFOA) and perfluorooctanesulfonic acid (PFOS) are the most commonly found PFASs in the environment (Zhao et al, 2016). They are also believed to be the final product of degradation of other PFASs. In 2002, the 3M company, a world leading manufacturer of PFASs, banned the use of perfluorooctyl sulfonyl fluoride (POSF) in manufacturing processes. This led to replacement of longer chain PFASs by compounds with shorter chain lengths (Wang et al, 2015). Short chain PFASs are defined as C6 and higher for perfluorosulfonic acids (PFSAs) and C8 and higher for perfluorocarboxylic acids (PFCAs) (Oecd.org, 2018). For example, PFBS (C4) was used for replacing PFOS (C8). However, the persistence of short-chain PFASs has been shown to be similar to the long-chain compounds. Smaller compounds are more mobile than larger ones, therefore, the mobility of short-chain PFASs is higher in long-distance transport. Short-chain compounds can be found in the environment at concentrations of similar magnitude as the longer-chain ones. This was caused by using compounds with shorter chain lengths for replacement of those with long chains (Zhao et al, 2016).

Poly- and perfluorinated substances have been industrially used since 1950’s. (Borg and Ivarsson, 2018). Due to their high stability and unique properties such as water, dirt and grease repellence, film forming ability and temperature resistance, they have many of areas of use. The most common usages are as surface agents, in electronic and printing products, impregnation agents in paper and textiles and as pesticides. They can stick to surfaces via sorption of the polar end of the structure, or the non-polar carbon chain. (Dorrance, Kellogg and Love, 2017). Other areas of use are in cosmetics, medical equipment, smart phones and solar cells. The main source of pollution by PFASs is from the use of fire-fighting surfactant foams (Dorrance, Kellogg and Love, 2017), where they form an aqueous film on the burning surface. Because of the stability of the compounds they can be transported long distances without being degraded in the environment. The lipophobic and hydrophilic characteristics allow long-distance transport of PFASs (Dorrance, Kellogg and Love, 2017). They can also bioaccumulate in living organisms and be transferred between organisms through the food chain. All of these chemical properties led to the addition of PFOS, to the list of persistent organic pollutants (POPs) (Wang et al, 2009).

Presence of PFASs was identified in aquatic environments both in Sweden and other countries in the world (Livsmedelsverket, 2018). In a national survey, possible drinking water contamination was assessed and new recommendations for PFASs in drinking water, including 11 compounds (PFAS-11) that should be controlled for, have been set by The National Food Agency (Livsmedelsverket). In general, no PFASs should be present in drinking water. In order to retain some safety margin, a limit of 90 ng/L was set for PFAS-11 (Livsmedelsverket, 2018).

Table 1: Substances included in PFAS-11, recommended for monitoring by the National food agency.

Compound Name

PFBS Perfluorobutane sulfonic acid

PFHxS Perfluorhexane sulfonic acid

PFOS Perfluorooctane sulfonic acid

6:2 FTSA 6:2 fluorotelomer sulfonic acid

PFBA Perfluorobutanoic acid

PFPeA Perfluoropentanoic acid

PFHxA Perfluorohexanoic acid

PFHpA Perfluoroheptanoic acid

PFOA Perfluorooctanoic acid

PFNA Perfluorononanoic acid

(7)

Environmental contamination of PFASs has also led to exposure of PFASs to humans and wildlife. Studies have shown that the microbes can bind environmental pollutants, however there are no studies of binding of PFASs to the bacteria. In theory PFASs may enter the food web by microbial uptake or binding to the surface of the bacterium. Uptake of POPs by microorganisms is a two-step passive diffusion. First, the compounds adsorb to the surface of microbes and diffuse through the membrane into the intracellular matrix (Del Vento and Dachs, 2002). In the case of heavy metals adsorption, physical parameters also need to be considered, including temperature and pH (Ayangbenro and Babalola, 2017). Another way of uptake of pollutants by microorganisms is via active uptake, called bioaccumulation. A way of removing toxic compounds from the environment is by using bacteria as biosorbents. Some of bacteria with biosorption ability also have biodegradation activity. Bacteria owe their ability of uptake of contaminants to their size, flexibility to adapt to environmental conditions and ubiquity (Ayangbenro and Babalola, 2017). Biosorption is a passive mechanism of uptake by binding of the pollutants to the outer layers and cell wall of the bacterium (Fomina and Gadd, 2014). This process can be carried out by dead microorganisms as well as the living ones.

3. Materials

3.1 Chemicals and solutions

Chemicals used for preparation of 0,2 M sodium hydroxide in methanol, sodium hydroxide pellets (laboratory reagent grade) and methanol (HPLC grade) were purchased from Fisher Scientific UK (Loughborough, UK). A 2 mM ammonium acetate (aq) solution was used as mobile phase in the samples.

For exposure of microbes to PFAS, a water sample from a contaminated site in Västerås, PFAS-11 cocktail and K-media (control) were used. PFAS-PFAS-11 stock solution with final concentrations as listed in Table 2, was prepared from chemicals bought form Honeywell Fluka (PFBA, PFOA and PFHxS), Sigma Aldrich (PFPeA, PFHpA, PFNA, PFDA, PFBS and PFOS), Matrix Scientific (6:2 FTSA) and Tokyo Chemical industry (PFHxA). Sodium chloride (52 mM) and Potassium chloride (32 mM) for preparation of K-medium were purchased from Duchefa Biochemie and Scharlab, respectively. Glacial acetic acid (PharmaGrade) was bought from Sigma-Aldrich Fine Chemicals (Arklow, Ireland). EnviCarb SPE Bulk was purchased from Sigma-Aldrich(St. Louis, USA).

Table 2: Chemicals and their concentrations (ug/L) in PFAS-11 solution and environmental water.

Compound Final conc. (ug/L)

PFBA 0,031 PFPeA 0,175 PFHxA 0,286 PFHpA 0,100 PFOA 0,077 PFNA 0,010 PFDA 0,010 PFBS 0,149 PFHxS 3,460 PFOS 12,000 6:2 FTSA 0,050

(8)

Labeled internal standards (13_{C), including internal and recovery standards, and calibration} standards (12_{C) were purchased from Wellington Laboratories (Guelph, Canada). The internal} and recovery standards contained 13_{C-mass-labelled compounds, all in concentration of 0,2} ng/L. The compounds in internal standards are the following M8-FOSA-M, PFHxDA, M2-PFTeDA, M2-6:2FTSA, M2-8:2FTSA, M3-PFBS, M3-PFPeA, PFBA, M2-PFHxA, M4-PFOA, M5-PFNA, M2-PFDA, M2-PFunDA, M2-PFDoA, 18O2-PFHxS and M4-PFOS. The recovery standard contained the following M3-PFBA, M3-PFHxS, M8-PFOS, M8-PFOA, M9-PFNA, M6-PFDA, M7-PFUndA, M2-4:2FTSA, M5-PFHxA and M3-PFPeA. Concentrations of native PFASs compounds used for quantification of results in CS standards also have similar concentrations to the IS and RS. Exact concentrations are listed in Table 6 in the Appendix.

3.2 Materials

Glass pipettes and beakers were ethanol washed and burned in the oven to prevent adsorption of analytes to the surface. All materials were washed with methanol before use. Syringes for IS (50 uL), CS (15 uL) and RS (15 uL) previously washed at least eight times with methanol. For filtration of environmental water, Whatman microfiber glass filters with 1 um pore size were used. Filtration of purified samples was performed with 2 mL syringes purchased from Norm-Ject (Tuttlingen, Germany), and Acrodisc syringe filters with 0,2 um pore size from Water Corporation (Milford, USA).

3.3 Instrumentation

Analysis of PFASs is currently performed using liquid chromatography (LC) (Jahnke and Berger, 2009). Different types of detectors are used depending on the type of the sample. However, mass spectrometric (MS) detection methods are considered as reference methods. LC/MS methods give us the possibility to analyse the samples both quantitatively and qualitatively. The results of the analysis depend on the configurations of the instrument and detectors. Usually electrospray ionization (ESI) is used to ionize the samples.

Analysis was performed using an Acquity UPLC (Waters Corporation, Milford, USA) with a Xevo TQ-S tandem mass spectrometer (Waters Corporation, Milford, USA). An Acquity UPLC BEH C18 1,7 um, 2,1 x 100 mm column (50°C) was used for chromatographic separation. Mobile phase A contained 2 mM ammonium acetate (aq) in 30 % methanol and phase B 2 mM ammonium acetate in methanol. A volume of 10 uL injections were made, with gradient separation from 30 % 2 mM ammonium acetate (aq) in methanol to 100 % methanol over 17 minutes. MS-parameters used for detection of different PFASs are listed in Table 3 below and Table 6 in Appendix.

Table 3: Mass spectrometric parameters used for analysis.

Source temperature 150°C

Desolvation temperature 400°C

Cone gas flow 150 L/hour

Desolvation gas flow 800 L/hour

Collision gas Argon

3.4 Quality control and quality assurance (QC/QA)

The quality of the results was assured by performing a method test before analysis of real samples in order to analyse the recovery of the Internal standard. Blanks containing microbes and growth medium were analysed. Also, procedure blanks without any bacteria were prepared. All blanks were treated the same way as the samples with microbes exposed for PFASs. All

(9)

samples were analysed in triplicates. Recovery standards were used for monitoring the extraction procedure.

Results from analysis of K-medium blanks were used for calculation of method detection limit (MDL). The detection limit was set as 3 x the average concentrations of PFASs in K-medium blanks. Also, the ratio between the signals of the compounds in K-medium blanks were divided by the signals in the samples. The samples, where the ratio was above 33 % (the signal in blank was less than 1/3 of that in the sample) were reported as above the method detection limit.

4. Method

4.1 Exposure of bacteria to PFASs

A volume of 15 mL of microbes from an overnight culture was transferred into 15 mL tubes. Three biological replicates were prepared. The tubes were centrifuged (2 500 rpm) until pellets were formed. The supernatant was discarded, the tubes were filled with approximately 5 mL of K-media to wash the microbes and then the samples were vortexed. Another centrifugation was performed until the pellets were solid and the supernatant was discarded. Washing was performed two times.

After the washing, the tubes were dried upside down, and 5 mL of each condition (K-media, PFAS-11 mixture and environmental water) were added to the respective microbe. The samples were vortexed until all pellets were re-suspended in order to expose all microbes to PFAS compounds. Exposure was made on a shaker (250 rpm) in an incubator in 37°C for 1 hour. The tubes were centrifuged and washed twice with K-media. The supernatant was discarded, and the pellets were centrifuged again. The rest of the supernatant present at the bottom of the tubes was removed with a pipette. The final product was weighed and recorded for normalisation of concentrations against pellet weight.

4.2 Alkaline digestion

A volume of 0,5 mL of 0,2 M sodium hydroxide in methanol was added to respective sample after the exposure for PFASs. The tubes were stored in the freezer overnight. The microbes were vortexed and ultrasonicated for 30 minutes. The samples that were difficult to dissolve were mechanically processes using a small spatula. A portion of 1,5 mL of 0,2 M sodium hydroxide in methanol was added. The samples were then vortexed and left in the freezer for further extraction.

The samples were stored in the freezer one week. They were ultrasonicated for 15 minutes and shaken for 15 minutes. The tubes were centrifuged for 10 minutes (6 000 rpm). Additional centrifugation was performed if the pellet was not solid. All of the supernatant was transferred into an 8 mL glass vial previously washed with methanol. The remaining pellet was re-dissolved in 2 mL of methanol, performing a repeated extraction. Samples were vortexed and left to soak for 30 minutes. Microbes were ultrasonicated for 15 minutes and left on a shaker for 15 minutes. They were centrifuged until pellets was formed. The rest of the supernatant was transferred into the same vials as previously. Internal standards were added to monitor the extraction procedure. Spiking experiments were performed prior to extraction of real samples.

Recovery standards (RS) was added to the LC-vials and 200 uL of the samples were transferred into respective LC-vial. Before analysis with UPLC-MS/MS 300 uL of 2 mM ammonium acetate (aq) was added to the vials.

(10)

4.3 Concentration of the samples

The vials with the samples saved from the extraction were ultrasonicated for 10 minutes. A volume of 1 mL of each sample was transferred into LC-vials and placed under the evaporation unit. When most of the sample was evaporated, an additional 1 mL of the samples was added to the vials and the evaporation was continued until the final volume of 200 uL was reached. Final extracts in LC vials also contained 5 uL of RS and 300 uL of ammonium acetate (aq). The vials were stored in the freezer till further analysis. Upon analysis it was noticed that these samples had become like a gel and further clean-up was thus necessary.

4.4 Clean-up with EnviCarb

A method by Powley et al. was followed. Polypropylene tubes were filled with 50 mg of EnviCarb and 100 uL of glacial acetic acid. The samples were transferred from LC vials into pp-tubes and the vials were rinsed with 0,5 mL of methanol. The tubes were shaken, vortexed for 30 seconds and evaporated until the final volume of <1 mL. Syringe filters and syringes were washed twice with methanol. The content of the tubes was transferred into the syringes and filtered into LC vials. The tubes were rinsed with 0,5 mL methanol. Methanol was filtered into LC vials. The vials were placed on the evaporation unit and the samples were evaporated until the final volume of 500 uL.

5. Results

5.1 Method test and recovery

Before extraction of real samples, a method test was performed. Three of the samples in method test contained both internal standard and calibration standard 2 ng/L added before the extraction procedure started, and three others contained only internal standard. Results from method test are presented in Figure 12below. Most of the samples which contained calibration standard contained similar concentrations of PFASs.

Figure 12: Relationship between concentrations of PFASs in samples with calibration standard (Vik37A, Vik51A and Vik40A) and samples without CS (Vik37B, Vik51B and Vik40B).

0% 50% 100% 150% 200% 250%

(11)

Figure 13: Relationship between concentrations of PFASs isomers in samples with calibration standard (Vik37A, Vik51A and Vik40A) and samples without calibration standard (Vik37B, Vik51B and Vik40B).

The recovery of IS for PFASs compounds and PFASs isomers was between 51 % and 113 % for most of the compounds. This indicates that the method works and can be used for analysis of real samples. Recovery of RS for PFASs isomers was above 200 %. The results above or below the range for good recovery could be caused by possible blank issues or loss of internal standards during the analysis.

5.2 PFASs in bacterial pellets

Since all bacterial samples were analysed in triplicates the results presented below are average concentrations of analysed PFASs. Without the additional concentration step, only some PFASs could be detected in the samples. All three bacteria bound PFOS in the highest concentrations, see Figures 3-5. Other compounds that showed binding capacity to microbes were PFOcDA and PFDA. Very low concentrations of other analytes were identified, while some compounds could not be detected in any sample prior to the concentration step. Figures 3-5 present binding of PFOS by microbes. Concentrations detected in samples treated with environmental water with PFASs concentrations listed in Table 2, were 96000 pg/g, 62000 pg/g and 35000 pg/g in E. coli, A. delafieldii and P. nitroreducens. Total PFOS from PFAS-11 solution was bound in concentrations of 44000 pg/g, 19000 pg/g and 19000 pg/g to respective bacterium.

dimetyl

80/99 PFOS80/99 PFOS169/80 419/169PFOS PFOS 99/169 PFHxS 80/99 PFHxS 319/99 PFHxS 80/99 0% 20% 40% 60% 80% 100% 120% 140%

(12)

Figure 3: Concentrations of different PFASs in bacterial pellets in pg/g.

PF BA PF Pe A PF BS PF HxA PF HpA _PFPeS PF HxS PF HpS _PFOA _PFNA _FOSA _PFOS _PFDA PF Un DA PF NS PF DS PF Do DA PF Tr DA PF Do DS PF TDA PF Hx DA PF O cDA 4: 2 F TS A 6: 2 F TS A 8: 2 F TS A 0 20000 40000 60000 80000 100000 120000 co nc. p g/ g

E. coli

Environmental PFAS-11 PF BA PF Pe A PF BS PF HxA PF HpA _PFPeS PF HxS PF HpS _PFOA _PFNA _FOSA _PFOS _PFDA PF Un DA PF NS PF DS PF Do DA PF Tr DA PF Do DS PF TDA PF Hx DA PF O cDA 4: 2 F TS A 6: 2 F TS A 8: 2 F TS A 0 10000 20000 30000 40000 50000 60000 70000 co nc. p g/ g

A. delafieldii

Environmental PFAS-11

(13)

Below, results from confirmatory analysis of concentrated samples are presented in Figures 6-8. E. coli, A. delafieldii and P. nitroreducens bound following concentrations of PFOS 97000 pg/g (190 pmol), 56000 pg/g (110 pmol) and 36000 pg/g (72 pmol) in environmental water samples, and 48000 pg/g (96 pmol), 18000 pg/g (36 pmol) and 19000 pg/g (38 pmol) in PFAS-11 samples.

Figure 6: Concentrations of PFASs in E. coli samples in pg/g.

PF BA PF Pe A PF BS PF HxA PF HpA _PFPeS PF HxS PF HpS _PFOA _PFNA _FOSA _PFOS _PFDA PF Un DA PF NS PF DS PF Do DA PF Tr DA PF Do DS PF TDA PF Hx DA PF O cDA 4: 2 F TS A 6: 2 F TS A 8: 2 F TS A 0 5000 10000 15000 20000 25000 30000 35000 40000 co nc. p g/ g

P. nitroreducens

Environmental PFAS-11 PF BA PF Pe A PF BS PF HxA PF HpA _PFPeS PF HxS PF HpS _PFOA _PFNA _FOSA _PFOS _PFDA PF Un DA PF NS PF DS PF Do DA PF Tr DA PF Do DS PF TDA PF Hx DA PF O cDA 4: 2F TS A 6: 2 F TS A 8: 2 F TS A 0 20000 40000 60000 80000 100000 120000 co nc. p g/ g

E. coli

(14)

Figure 7: Concentrations of PFASs in A. delafieldii samples in pg/g.

Figure 8: Concentrations of PFASs in P. nitroreducens samples in pg/g.

The amounts of PFOS bound to bacteria in both cases are similar. E. coli bound the highest concentrations of PFOS. Also, it showed highest binding capacity in samples treated with environmental water.

5.3 PFASs isomers

During analysis of the results it was observed that isomers of some compounds were present in the samples and could possibly have been detected. Therefore, an additional analysis of isomers of PFOS and PFHxS was performed. Figures 9-11 below show concentrations of isomers detected in the samples. Isomers detected in the highest concentrations in all bacteria are the linear isomers of PFOS (L-PFOS). All PFHxS isomers bound to microbes were below method

PF BA PF Pe A PF BS PF HxA PF HpA PF Pe S PF HxS PF HpS _PFOA _PFNA _FOSA _PFOS _PFDA PF Un DA PF NS PF DS PF Do DA PF Tr DA PF Do DS PF TDA PF Hx DA PF O cDA 4: 2F TS A 6: 2 F TS A 8: 2 F TS A 0 10000 20000 30000 40000 50000 60000 Co nc . ( pg /g )

A. delafieldii

Environmental PFAS-11 PF BA PF Pe A PF BS PF HxA PF HpA PF Pe S PF HxS PF HpS _PFOA _PFNA _FOSA _PFOS _PFDA PF Un DA PF NS PF DS PF Do DA PF Tr DA PF Do DS PF TDA PF Hx DA PF O cDA 4: 2F TS A 6: 2 F TS A 8: 2 F TS A 0 5000 10000 15000 20000 25000 30000 35000 40000 co nc. p g/ g

P. nitroreducens

(15)

detection limit. Although PFHxS was present in the environmental water sample at a concentration of one third of the PFOS concentrations. This indicate on preferential binding for PFOS, the compound present in the highest concentration in the water samples used for treatment of bacteria. With similar binding capacity for PFHxS the concentrations in microbes should have been seen in the range of 5000 -36000 pg/g. No dimethyl-PFOS and 2-/3-/4-PFHxS could be quantified, because of the poor chromatography.

Figure 9: Concentrations of PFOS and PFHxS isomers detected in E. coli samples.

Figure 10: Concentrations of PFOS and PFHxS isomers detected in A. delafieldii samples.

dimetyl 3/4/5-PFOS 6/2-PFOS 1-PFOS L-PFOS 2/3/4-PFHxS 1-PFHxS L-PFHxS 0 20000 40000 60000 80000 100000 120000 140000 co nc. p g/ g

E. coli

dimetyl 3/4/5-PFOS 6/2-PFOS 1-PFOS L-PFOS 2/3/4-PFHxS 1-PFHxS L-PFHxS 0 10000 20000 30000 40000 50000 60000 70000 80000 90000 co nc. p g/ g

A. delafieldii

(16)

Figure 11: Concentrations of PFOS and PFHxS isomers detected in P. nitroreducens samples.

6. Discussion

6.1 Microbial binding of PFASs

Binding of more than one compound to bacteria was expected, however PFOS was the only PFAS to be adsorbed by all three microbes. The reasons for this are unknown. We could speculate that it can depend on the chemical properties of the surface structures on the microbes. We do not know the binding capacity of other bacteria such as Gram-positive bacteria with different surface properties for PFASs since only Gram-negative bacteria were analysed. A possible explanation for this phenomenon is that the microbial binding is concentration dependent and the microbes adsorb the most available compound in the first place.

Concentrations of 11 PFAS from NFA’s list of compounds that should be monitored, in the environmental water and PFAS-11 cocktail were the same, yet microbes treated with environmental water bound higher concentrations of PFOS than the ones treated with PFAS-11 mixture. The reason why the microbes preferred binding of PFASs in environmental samples more than in PFAS-11 could be impact of other contents of the environmental waters, that weren’t present in the PFAS-11 mixture. Still, this is only speculations since the real reason is not known.

The concentration of PFOS was of 12 ug/L, which is the highest of all eleven compounds monitored. It was expected that concentrations of PFOS in the samples would be higher than of the other PFASs. However, it was remarkable that there was no binding of other PFASs. The other compound that was expected to be in high concentrations was PFHxS with second highest concentration of 3,5 ug/L in PFAS-11 and environmental samples, which is approximately 30% of PFOS. Both PFOS and PFHxS have similar structures and the same functional group. Based on this information, assuming similar binding capacity for these compounds, it was expected that the amount of PFHxS bound by microbes would be about 30% of that of amount PFOS. Concentrations of PFOS bound to E. coli, A. delafieldii and P. nitroreducens in the environmental water samples were 190 pmol, 110 pmol and 72 pmol, and 96 pmol, 36 pmol

dimetyl 3/4/5-PFOS 6/2-PFOS 1-PFOS L-PFOS 2/3/4-PFHxS 1-PFHxS L-PFHxS 0 5000 10000 15000 20000 25000 30000 35000 40000 co nc. p g/ g

P. nitroreducens

(17)

and 38 pmol in the samples with PFAS-11 solution. Assuming that the binding capacity of microbes to PFHxS were the same as to the PFOS, the molar concentration of PFHxS would be as follow 72 pmol, 43 pmol and 28 pmol in environmental samples and 35 pmol, 35 pmol and 14 pmol in PFAS-11 samples for E. coli, A. delafieldii and P. nitroreducens respectively (for calculations see Appendix). Then, all PFHxS concentrations would be above the MDL, which for PFHxS is 100 pg/g.

As mentioned previously, the microbes preferred binding of the compound present in highest concentrations. It could be hypothesized that the binding of other PFASs would be higher in microbes exposed for solutions that do not contain PFOS or where the concentration of this compound is lower than of the other compounds.

According to studies on effects of pollutants on microorganism, bacteria may have developed different ways of rapid adaptation to environmental stress (Ford, 1994). Methods of reduction of toxicity of pollutants by microbes mentioned in the study are among others binding of pollutants by extracellular polysaccharides and transformation to less toxic forms. Considering that PFOS is one of the most commonly detected PFASs in the environment, it can be speculated that the exposure of microbes to this compound is higher than for other PFASs in the natural environment. Assuming that the microbes have been exposed for high concentration of PFOS for long periods of time, they could have developed mechanisms of adaptation for this compound, that is encoded in the genetic material. In this way, the microbes did not need to be exposed for PFOS before analysis. They might already have the mechanism of adaptation, if the older generations have been exposed for PFOS in their natural environment. Further experiment is needed to confirm this hypothesis.

It's possible that the microbes could have been exposed for PFASs before analysis and have already have some concentrations of PFOS bound to them. Also, perfluoroalkyl substances, like PFOS, are often the final product of degradation of other PFOS-based substances (e.g., FOSA). Studies have shown that some microbes in anaerobic conditions can employ reductive dechlorination as a source of energy (Sàez, de Voogt and Parsons, 2008). We could hypothesize that the process of defluorination also can be used as a source of energy by microorganisms and the degradation of polyfluorinated compounds to perfluorinated compounds has started before the extraction process. Again, further experiment is needed to test this hypothesis.

The environmental water was sampled in an area used for fire-fighting training. Such area is exposed to aqueous fire-fighting foams (AFFFs) more often than other areas. Therefore, higher concentrations of PFASs bound to bacteria exposed for that water were expected. Two of the dominant PFASs used for production of AFFFs are PFOS and PFHxS (Hu et al, 2016). The amount of these two PFASs in the surface water sample were highest of all eleven with known concentrations. Also, for this reason concentrations of PFHxS in the microbes were expected to be detected above MDL. Considering that the concentrations of PFOS were almost three times higher than of PFHxS in the water samples, it can be assumed that PFOS is the main component of AFFFs used in this area.

Microorganisms can be used as biosorbents for removal of toxic pollutants. They could be used as a way of clean-up of water contaminated with PFASs. As the microbes analysed in this project only bound PFOS, they may be effective biosorbents for this particular compound. It should be considered that the exposure to PFASs containing solutions in the lab was made for only one hour, in order to observe the random binding of PFASs, which include both specific

(18)

and unspecific binding. Therefore, it’s difficult to speculate if the binding of other compounds would be higher with increased exposure time.

A percentage binding of PFOS indicates that only 2 – 10 % of available PFOS has been bound by different microbes, treated with different solutions. A possible explanation to that is that the microbes have reached the maximum limit of uptake per cell and binding of more PFOS was not possible. Decreasing the volume of solutions used for exposure or increasing of the amount bacterial cells could possibly lead to binding of more of the available compounds.

6.3 Recommendations for future studies

Only the results where the concentrations of analytes in the samples are more than three times higher than concentrations in the K-medium blanks. Although procedural blanks without any microbes were also analysed, they could not be used for calculation of detection limit since no concentration per gram bacterium could be calculated, but the signal could be compared in the same way. The signal responses from the blanks compared against KM and bacterial samples were lower than the KM. MDL calculated from K-media samples confirmed that the concentrations of most of the compounds were below detection limits and could not be reported. The results of this study show that only PFOS was bound by the microbes. Further studies on this topic should be done, as we see that the PFASs can enter the food chain in early stages, on microbial level. Here, only Gram-negative microbes were exposed for PFASs. What can be done is a comparison of binding to Gram-negative microbes against Gram-positive microbes, in order to see if the structure of microbes has impact on how much can be bound. Another recommendation is to analyse the PFASs content in the solutions before and after the exposure. In this way it can be seen if the microbes have bound more compounds and they started to degrade to other forms and an expected binding capacity could be calculated. Also, if the bacteria were exposed for PFASs in their natural environment, possible higher concentrations in the samples would be observed. Since no other compounds could be detected above the detection limit, the microbes could be exposed for different PFASs separately, to see if they then bind other compounds.

Also, different concentrations of chemicals were used in this experiment. In result, the number of molecules available for binding was not equal. To obtain more accurate results and binding capacities, an equal amount of substance (mol) could be used for exposure.

7. Conclusion

In conclusion, only one PFAS was adsorbed by all three microbes. Highest concentrations of PFOS were detected in E. coli samples. Some other compounds were identified in smaller amounts in E. coli and A. delafieldii samples. Yet, PFOS was the only compound identified above the MDL in P. nitroreducens samples. The major PFOS isomer present in the samples was linear PFAS. The highest binding capacity for PFOS was shown in E. coli treated with environmental water samples.

(19)

8. Acknowledgements

Hereby, I want to thank my supervisor, Ingrid Ericson Jogsten, for all support and help with experimental and report writing, and my assistant supervisor Jana Jass. A big thanks to Maria Björnsdotter for help in the lab and with the analysis. Marios Stylianou for helping me with the microbiology part of the project and Petra Ståhl for preparing the microbes and the exposure.

(20)

9. References

Ayangbenro, A. and Babalola, O. (2017). A New Strategy for Heavy Metal Polluted Environments: A Review of Microbial Biosorbents. International Journal of Environmental Research and Public Health, 14(1), p.94.

Borg, D. and Ivarsson, J. (2018). Analysis of PFASs and TOF in products. [online] DIVA.

Available at: http://www.bing.com/cr?IG=6E3336F88744433CAAA9F6F5B430E81A&CID=28EBF52E2 023609E1691FEC2218C6149&rd=1&h=XDhj0czYaqqF4ozddetzjEknmS5xlaDCNvroENq4i A4&v=1&r=http%3a%2f%2fnorden.diva-portal.org%2fsmash%2frecord.jsf%3fpid%3ddiva2%3a1118439&p=DevEx.LB.1,5080.1 [Accessed 4 May 2018].

Conder, J., Hoke, R., Wolf, W., Russell, M. and Buck, R. (2008). Are PFCAs

Bioaccumulative? A Critical Review and Comparison with Regulatory Criteria and Persistent Lipophilic Compounds. Environmental Science & Technology, 42(4), pp.995-1003.

Del Vento, S. and Dachs, J. (2002). PREDICTION OF UPTAKE DYNAMICS OF PERSISTENT ORGANIC POLLUTANTS BY BACTERIA AND PHYTOPLANKTON. Environmental Toxicology and Chemistry, 21(10), p.2099.

Dorrance, L., Kellogg, S. and Love, A. (2017). What You Should Know About Per- and Polyfluoroalkyl Substances (PFAS) for Environmental Claims. Environmental Claims Journal, 29(4), pp.290-304.

Fomina, M. and Gadd, G. (2014). Biosorption: current perspectives on concept, definition and application. Bioresource Technology, 160, pp.3-14.

Ford, T. (1994). Pollutant effects on the microbial ecosystem. Environmental Health Perspectives, 102(Suppl 12), pp.45-48.

Fromme, H., Tittlemier, S., Völkel, W., Wilhelm, M. and Twardella, D. (2009). Perfluorinated compounds – Exposure assessment for the general population in western countries. International Journal of Hygiene and Environmental Health, 212(3), pp.239-270. Hu, X., Andrews, D., Lindstrom, A., Bruton, T., Schaider, L., Grandjean, P., Lohmann, R., Carignan, C., Blum, A., Balan, S., Higgins, C. and Sunderland, E. (2016). Detection of Poly- and Perfluoroalkyl Substances (PFASs) in U.S. Drinking Water Linked to Industrial Sites, Military Fire Training Areas, and Wastewater Treatment Plants. Environmental Science & Technology Letters, 3(10), pp.344-350.

Jahnke, A. and Berger, U. (2009). Trace analysis of per- and polyfluorinated alkyl substances in various matrices—How do current methods perform?. Journal of Chromatography A, 1216(3), pp.410-421.

Livsmedelsverket.se. (2018). Riskhantering - PFAS i dricksvatten och fisk. [online] Available at: https://www.livsmedelsverket.se/livsmedel-och-innehall/oonskade-amnen/miljogifter/pfas-poly-och-perfluorerade-alkylsubstanser/riskhantering-pfaa-i-dricksvatten [Accessed 17 May 2018].

(21)

Morrison, H., Gobas, F., Lazar, R. and Haffner, G. (1998). Development and Verification of a Benthic/Pelagic Food Web Bioaccumulation Model for PCB Congeners in Western Lake Erie. Environmental Science & Technology, 32(15), pp.2356-2356.

Naturvårdsverket. (2018). Perfluorerade ämnen i miljön. [online] Available at:

https://www.naturvardsverket.se/Sa-mar-miljon/Manniska/Miljogifter/Organiska-miljogifter/Perfluorerade-amnen/# [Accessed 2 May 2018].

Powley, C., George, S., Ryan, T. and Buck, R. (2005). Matrix Effect-Free Analytical Methods for Determination of Perfluorinated Carboxylic Acids in Environmental Matrixes. Analytical Chemistry, 77(19), pp.6353-6358.

Oecd.org. (2018). About PFASs - OECD Portal on Per and Poly Fluorinated Chemicals. [online] Available at: http://www.oecd.org/chemicalsafety/portal-perfluorinated-chemicals/aboutpfass/ [Accessed 8 Jun. 2018].

Skutlarek, D., Exner, M. and Färber, H. (2006). Perfluorinated Surfactants in Surface and Drinking Waters. Environmental Science and Pollution Research - International, 13(5), pp.299-307.

Wang, T., Wang, Y., Liao, C., Cai, Y. and Jiang, G. (2009). Perspectives on the Inclusion of Perfluorooctane Sulfonate into the Stockholm Convention on Persistent Organic

Pollutants1. Environmental Science & Technology, 43(14), pp.5171-5175.

Wang, Z., Cousins, I., Scheringer, M. and Hungerbuehler, K. (2015). Hazard assessment of fluorinated alternatives to long-chain perfluoroalkyl acids (PFAAs) and their precursors: Status quo, ongoing challenges and possible solutions. Environment International, 75, pp.172-179.

Zhao, P., Xia, X., Dong, J., Xia, N., Jiang, X., Li, Y. and Zhu, Y. (2016). Short- and long-chain perfluoroalkyl substances in the water, suspended particulate matter, and surface sediment of a turbid river. Science of The Total Environment, 568, pp.57-65.

(22)

10. Appendix

10.1 Preparation of solutions Filtration of environmental water

Whatman glass microfibre filter with 0.1 um pore size and 150 mm in diameter was used. The filter was washed three times with methanol and three times with MQ-water before filtration of 50 mL of water (previously sonicated) into a 50 mL pp-tube.

0.2M NaOH in MeOH

A mass 0.8299 g of sodium hydroxide pellets was weighed and transferred into a volumetric flask. Small portions of methanol were added, and the flask was swirled to dissolve the pellets. When all sodium hydroxide was dissolved, the flask was filled with methanol up to 100 mL line. The obtained concentration was 0.207475 M. New calculations were made, and the solution was diluted to 0.2 M. 3.6 mL were taken out from the flask and 3.6 mL of methanol were added. The final concentration was 0.2 M.

Blanks

One blank sample for each batch of microbes was prepared. The blanks were prepared of 2 mL of 0.2 M sodium hydroxide in methanol and 2 mL methanol. A volume of 50 uL of IS was added to each 8 mL glass vial with blanks. All blanks were treated the same way as the samples with microbes.

Batch Standard

Batch Standard was prepared in an LC-vial containing 5 uL recovery standard (RS), 5 uL internal standard (IS) and 5 uL calibration standard-1 (CS) added to the vial with a syringe designed for each standard. A volume of 185 uL methanol and 300 uL of 2 mM ammonium acetate were added.

10.2 Calculations of relationship between PFASs bound to microbes

The following formula was used for calculations of relationships between PFOS and PFHxS in the samples and K-medium blanks.

[PFOS]solutions = 12 ng/L [PFHxS]solutions = 3,5 ng/L 3,5/12 = 0,29 = 29 %

Assuming that the binding of PFHxS would be approx. 30% of that of PFOS, the theoretical binding was calculated by using following formula: [PFHxS] = [PFOS] * 0,3. Obtained concentrations are as follow:

E. coli [PFHxS]ENV = 29000 pg/g [PFHxS]PFAS-11 = 14000 pg/g [L-PFHxS]ENV = 36000 pg/g [L-PFHxS]PFAS-11 = 15000 pg/g A. delafieldii [PFHxS]ENV = 17000 pg/g [PFHxS]PFAS-11 = 5400 pg/g [L-PFHxS]ENV = 23000 pg/g [L-PFHxS]PFAS-11 = 6900 pg/g P. nitroreducens [PFHxS]ENV = 11000 pg/g [PFHxS]PFAS-11 = 5700 pg/g

(23)

[L-PFHxS]ENV = 11000 pg/g [L-PFHxS]PFAS-11 = 5700 pg/g MMPFHxS = 400 g/mol

n = m / MM

E. coli [PFHxS]ENV = 29000 / 400 = 72 pmol

[PFHxS]PFAS-11 = 14000 / 400 = 35 pmol [L-PFHxS]ENV = 90 pmol

[L-PFHxS]PFAS-11 = 38 pmol

A. delafieldii [PFHxS]ENV = 43 pmol

[PFHxS]PFAS-11 = 35 pmol [L-PFHxS]ENV = 58 pmol [L-PFHxS]PFAS-11 = 17 pmol

P. nitroreducens [PFHxS]ENV = 28 pmol

[PFHxS]PFAS-11 = 14 pmol [L-PFHxS]ENV = 28 pmol [L-PFHxS]PFAS-11 = 14 pmol

10.3 Calculations of amount of PFOS in moles

MMPFOS = 499 g/mol n = m / MM

E. coli [PFOS]ENV = 97000 / 499 = 190 pmol

[PFOS]PFAS-11 = 48000 / 499 = 96 pmol

A. delafieldii [PFOS]ENV = 56000 / 499 = 110 pmol

[PFOS]PFAS-11= 18000 / 499 = 36 pmol

P. nitroreducens [PFOS]ENV = 36000 / 499 = 72 pmol

[PFOS]PFAS-11 = 19000 / 499 = 38 pmol

10.4 Calculations of percentage of bound PFOS

Exposure volume = 5 mL

[PFOS]ENV/PFAS-11 = 12000 ng/L x 0,005 L = 60 ng

Microbes were exposed to a concentration of PFOS of 60 ng.

E. coli ENV = 2761 x 2 = 5522 pg ~ 5,5 pg 5,5 / 60 = 9,2 % PFAS-11 = 1087 x 2 = 2174 pg ~ 2,2 ng 2,2 / 60 = 3,7 % A. Delafieldii ENV = 1838 x 2 = 3676 pg ~ 3,7 ng 3,7 / 60 = 6,2 % PFAS-11 = 580 x 2 = 1160 pg ~ 1,2 ng 1,2 / 60 = 2 % P. nitroreducens ENV = 1961 x 2 = 3923 pg ~ 3,9 ng 3,9 / 60 = 6,5 % PFAS-11 = 1070 x 2 = 2140 pg ~ 2,1 ng 2,1 / 60 = 3,5 %

(24)

10.5 Data

Table 4: Calculated weights of microbes after exposure for PFASs.

Table 5: Concentration (ng/L) of native PFASs in CS.

CS1 mix 205 PFCA PFSA FTSA PFOSA Target conc. 0,200 ng/L

Solvent MeOH

Compound Final conc. (ng/L)

PFBA 0,200 PFPeA 0,200 PFHxA 0,200 PFHpA 0,200 PFOA 0,200 PFNA 0,200 PFDA 0,200 PFUdA 0,200 PFDoA 0,200 PFTrDA 0,200 PFTeDA 0,200 PFHxDA 0,200 PFODA 0,200 L-PFBS 0,177 L-PFHxS 0,189 L-PFOS 0,191 L-PFDS 0,193 FOSA 0,200 4:2FTS 0,187 6:2FTS 0,190 8:2FTS 0,192 L-PFHpS 0,190 L-PFDoS 0,194 L-PFNS 0,192 L-PFPeS 0,188

Sample Weight (mg) Sample Weight (mg) Sample Weight (mg)

KM1/EC 25 KM1/AC 61,9 KM1/PN 81,9

KM2/EC 38 KM2/AC 62,1 KM2/PN 116,6

KM3/EC 66,2 KM3/AC 64,9 KM3/PN 105,1

KM4/EC 57,6 KM4/AC 68,8 KM4/PN 107,6

ELD1/EC 52,1 ELD1/AC 68,4 ELD1/PN 89,7

ELD2/EC 55,9 ELD2/AC 61 ELD2/PN 117

ELD3/EC 63,3 ELD3/AC 68,1 ELD3/PN 113

PFAS1/EC 42 PFAS1/AC 54,8 PFAS1/PN 91,2

PFAS2/EC 41,8 PFAS2/AC 65,6 PFAS2/PN 125,8

(25)

Table 6: Mass spectrometric parameters for detection of particular PFASs.

Compound MRM Cone volt. Col. Energy Compound MRM Cone volt. Col. Energy PFBA 212,97 > 169,00 20 11 PFOSA 497,90 > 78,00 82 30

13C PFBA (RS) 215,97 > 172,00 20 11 PFOSA 497,90 > 168,96 82 28

13C PFBA (IS) 216,97 > 172,00 20 11 13C PFOSA (IS) 505,90 > 77,80 82 30

PFPeA 262,97 > 219,00 20 8 PFOS 498,97 > 79,96 20 44

13C PFPeA (IS) 265,97 > 222,00 20 8 PFOS 498,97 > 98,96 20 38

13C PFPeA (RS) 267,97 > 223,00 20 8 PFOS 498,97 > 169,03 20 34 PFBS 298,90 > 79,96 20 26 PFOS 498,97 > 419,00 20 35 PFBS 298,90 > 98,90 20 26 13C PFOS (IS) 502,97 > 98,06 20 38 13C PFBS (IS) 301,90 > 98,90 20 26 13C PFOS (RS) 506,97 > 98,96 20 38 PFHxA 312,97 > 118,95 20 26 PFDA 512,97 > 219,00 20 18 PFHxA 312,97 > 269,00 20 9 PFDA 512,97 > 469,00 20 11

13C PFHxA (IS) 314,97 > 270,00 20 9 13C PFDA (IS) 514,97 > 470,00 20 11

13C PFHxA (RS) 317,97 > 273,00 20 9 13C PDFA (RS) 518,97 > 474,00 20 11 4:2 FTS 327,00 > 81,00 20 28 8:2 FTS 527,00 > 80,00 20 28 4:2 FTS 327,00 > 307,00 20 20 8:2 FTS 527,00 > 507,00 20 20 13C 4:2 FTS (RS) 329,00 > 81,00 20 28 13C 8:2 FTS (IS) 529,00 > 509,00 20 20 PFPeS 348,90 > 79,96 20 30 PFNS 548,90 > 79,96 20 44 PFPeS 348,90 > 98,96 20 26 PFNS 548,90 > 98,96 20 38 PFHpA 362,97 > 168,97 20 16 PFUnDA 562,97 > 268,99 20 18 PFHpA 362,97 > 319,00 20 10 PFUnDA 562,97 > 519,00 20 12

13C PFHpA (IS) 366,97 > 322,00 20 10 13C PFUnDA (IS) 564,97 > 520,00 20 12

PFHxS 398,90 > 79,96 20 34 13C PFUnDA (RS) 569,97 > 525,00 20 12

PFHxS 398,90 > 98,90 20 30 PFDS 598,97 > 79,96 20 58

PFHxS 398,90 > 119,01 20 28 PFDS 598,97 > 98,90 20 42

PFHxS 398,90 > 319,00 20 35 PFDoDA 612,97 > 168,96 40 22

13C PFHxS (RS) 401,90 > 98,90 20 30 PFDoDA 612,97 > 569,00 34 14

13C PFHxS (IS) 402,90 > 102,90 20 30 13C PFDoDA (IS) 614,97 > 570,00 34 14

PFOA 412,97 > 118,93 20 30 PFTrDA 662,90 > 168,96 20 26

PFOA 412,97 > 168,97 20 18 PFTrDA 662,90 > 619,00 20 14

PFOA 412,97 > 219,00 20 14 PFDoDS 698,00 > 79,96 20 45

PFOA 412,97 > 372,00 20 10 PFDoDS 698,00 > 98,90 20 40

13C PFOA (IS) 416,97 > 372,0 20 10 PFTDA 712,90 > 168,97 20 28

13C PFOA (RS) 420,97 > 376,00 20 10 PFTDA 712,90 > 669,00 20 14 6:2 FTS 427,00 > 81,00 20 28 13C PFTDA (IS) 714,90 > 670,00 20 14 6:2 FTS 427,00 > 407,00 20 20 PFHxDA 812,90 > 168,96 30 32 13C 6:2FTS (IS) 429,00 > 409,00 20 20 PFHxDA 812,90 > 796,00 30 15 PFHpS 448,97 > 79,96 20 35 13C PFHxDA (IS) 814,90 > 770,00 30 15 PFHpS 448,97 > 98,90 20 30 PFOcDA 912,00 > 168,96 36 36 PFNA 462,99> 219,00 20 18 PFOcDA 912,90 > 869,00 36 15 PFNA 462,99 > 419,00 20 12 13C PFNA (IS) 462,99 > 423,00 20 12 13C PFNA (RS) 471,99 > 427,00 19 12

(26)

Table 7: Average concentrations (pg/g) of PFAS isomers in E. coli samples.

KM1 Environmental PFAS-11 LOD

dimetyl - - - - 3/4/5-PFOS 96 3300 4400 100 6/2-PFOS 130 4200 3000 140 1-PFOS 260 680 900 260 L-PFOS 130 120000 51000 2700 2/3/4-PFHxS - - - - 1-PFHxS 260 <LOD <LOD 540 L-PFHxS 1000 1200 550 110

Table 8: Average concentrations (pg/g) of PFAS isomers in A. delafieldii samples.

dimetyl - - - -

3/4/5-PFOS 55 1900 1600 100

6/2-PFOS 84 2500 1100 140

1-PFOS 87 <LOD <LOD 260

L-PFOS 700 78000 23000 2700

2/3/4-PFHxS - - - -

1-PFHxS 140 <LOD <LOD 400

L-PFHxS 560 930 500 110

Table 9: Average concentrations (pg/g) of PFAS isomers in P. nitroreducens samples.

dimetyl - - - - 3/4/5-PFOS 120 810 1500 100 6/2-PFOS 130 1100 1000 140 1-PFOS 17 <LOD 290 260 L-PFOS 1500 37000 19000 2700 2/3/4-PFHxS - - - - 1-PFHxS 140 <LOD <LOD 400 L-PFHxS 590 860 380 110

(27)

Table 10: Concentrations (pg/g) of different PFASs in E. coli samples from second analysis with UPLC-MS/MS.

PFBA 150 <LOD <LOD 1900

PFPeA 820 <LOD <LOD 1200

PFBS 160 <LOD <LOD 260

PFHxA 1400 <LOD <LOD 2500

PFHpA 430 <LOD <LOD 700

PFPeS - <LOD <LOD -

PFHxS 980 <LOD <LOD 2100

PFHpS 80 <LOD <LOD 150

PFOA 1100 <LOD <LOD 2000

PFNA 260 <LOD <LOD 400

FOSA 72 <LOD <LOD 120

PFOS 2300 97000 48000 4800

PFDA 3800 <LOD <LOD 8000

PFUnDA 970 <LOD <LOD 1600

PFNS 220 <LOD <LOD 390

PFDS 340 <LOD <LOD 570

PFDoDA 220 <LOD <LOD 400

PFTrDA 88 <LOD <LOD 200

PFDoDS 220 <LOD <LOD 220

PFTDA 690 <LOD <LOD 1000

PFHxDA 3600 <LOD <LOD 6200

PFOcDA - <LOD <LOD 57

4:2 FTSA 32 <LOD <LOD 51

6:2 FTSA 3200 <LOD <LOD 3600

8:2 FTSA 220 <LOD <LOD 340

Table 11: Concentrations (pg/g) of different PFASs in A. delafieldii samples from second analysis with UPLC-MS/MS.

PFBA 800 <LOD <LOD 1900

PFPeA 200 <LOD <LOD 1200

PFBS 61 <LOD <LOD 260

PFHxA 600 <LOD <LOD 2500

PFHxS 540 <LOD <LOD 2100

PFOA 510 <LOD <LOD 2000

PFNA 94 <LOD <LOD 400

PFOS 990 56000 18000 4800

PFDA 2500 <LOD <LOD 8000

PFNS 90 <LOD <LOD 390

PFDS 130 <LOD <LOD 570

PFDoDS - <LOD <LOD 220

PFTDA 220 <LOD <LOD 1000

PFOcDA 32 <LOD <LOD 57

4:2 FTSA 19 <LOD <LOD 51

6:2 FTSA 290 <LOD <LOD 3600

(28)

Table 12: Concentrations (pg/g) of different PFASs in P. nitroreducens samples from second analysis with UPLC-MS/MS.

PFBA 1000 <LOD <LOD 1900

PFPeA 150 <LOD <LOD 1200

PFBS 39 <LOD <LOD 260

PFHxA 520 <LOD <LOD 2500

PFHxS 620 <LOD <LOD 2100

PFOA 390 <LOD <LOD 2000

PFNA 54 <LOD <LOD 400

PFOS 1500 36000 19000 4800

PFDA 1800 <LOD <LOD 8000

PFNS 88 <LOD <LOD 390

PFDS 100 <LOD <LOD 570

PFDoDS - <LOD <LOD 220

PFTDA 140 <LOD <LOD 1000

PFOcDA 24 <LOD <LOD 57

4:2 FTSA 0 <LOD <LOD 51

6:2 FTSA 110 <LOD <LOD 3600

(29)

Table 13: Comparison on signal intensity in samples and K-medium used as blank in concentrated E. coli samples.

ENV1 EC ENV2 EC ENV3 EC PFAS-11 1 EC PFAS-11 2 EC PFAS-11 3 EC

PFBA - - - - PFPeA - - - - PFBS - - - - PFHxA 51% 76% 80% 57% 83% 73% PFHpA 50% 61% 67% 64% 49% 51% PFPeS - - - - PFHxS 36% 37% 44% 94% 87% 81% PFHpS - 22% - - - - PFOA 1% 53% 60% 81% 63% 67% PFNA 83% 57% 51% 52% 48% 57% FOSA - 19% 33% 75% 402% 402% PFOS 0% 1% 1% 4% 2% 4% PFDA 3% 73% 68% 66% 65% 66% PFUnDA 1% 58% 63% 70% 73% 65% PFNS 198% 12% 19% 118% 94% 77% PFDS 580% 24% 36% 71% 65% 62% PFDoDA 42% 74% 68% 139% 83% 58% PFTrDA 22% 84% 89% 107% 68% 82% PFDoDS 20616% 58% 57% 67% 56% 62% PFTDA 224% 120% 109% 72% 86% 74% PFHxDA 1% 51% 49% 66% 50% 47% PFOcDA 60% 98% 100% - - - 4:2 FTSA 904% 55% 57% 56% 96% 57% 6:2 FTSA 49% 102% 148% 214% 84% 63% 8:2 FTSA 16% 72% 44% 80% 88% 75%

(30)

Table 14: Comparison on signal intensity in samples and K-medium used as a blank in concentrated A. delafieldii samples.

ENV1 AC ENV2 AC ENV3 AC PFAS-11 1 AC PFAS-11 2 AC PFAS-11 3 AC

PFBA - - - - PFPeA - - - - PFBS - - - - PFHxA 94% 34% 76% 69% 61% 71% PFHpA 78% 57% 82% 67% 57% 69% PFPeS - - - - PFHxS 45% 47% 62% 88% 74% 137% PFHpS 46% - 52% - - - PFOA 80% 79% 77% 80% 71% 85% PFNA 74% 85% 95% 84% 118% 100% FOSA 38% 62% 27% 52% 233% 60% PFOS 2% 2% 2% 10% 5% 5% PFDA 74% 69% 81% 71% 67% 67% PFUnDA 71% 61% 79% 62% 73% 64% PFNS 29% 26% 35% 80% 71% 113% PFDS 29% 22% 33% 61% 60% 70% PFDoDA 64% 65% 93% 81% 97% 107% PFTrDA 120% 93% 113% 96% 82% 89% PFDoDS - - - - PFTDA 84% 91% 81% 61% 82% 87% PFHxDA 55% 66% 78% 52% 65% 66% PFOcDA 100% 102% - 149% 187% 126% 4:2 FTSA - - - - 6:2 FTSA 176% 230% 230% 59% 47% 169% 8:2 FTSA 60% 107% 200% 111% 111% 138%

Table 15: Comparison on signal intensity in samples and K-medium used as a blank in concentrated P. nitroreducens samples.

ENV1 PN ENV2 PN ENV3 PN PFAS-11 1 PN PFAS-11 2 PN PFAS-11 3 PN

PFBA - - - - PFPeA - - - - PFBS - - - - PFHxA 74% 80% 42% 26% 44% 52% PFHpA 48% 47% 67% 35% 56% 57% PFPeS - - - - PFHxS 53% 27% 39% 67% 74% 108% PFHpS - 52% 81% - - - PFOA 82% 89% 76% 84% 77% 76% PFNA 91% 65% 89% 117% 69% 85% FOSA 96% 81% 91% 704% 112% 139% PFOS 3% 2% 2% 4% 3% 4% PFDA 77% 84% 92% 78% 73% 88% PFUnDA 79% 85% 89% 80% 96% 86% PFNS 40% 19% 26% 108% 153% 131% PFDS 42% 29% 32% 72% 67% 88% PFDoDA 94% 76% 89% 84% 72% 80% PFTrDA 96% 85% 66% 133% 84% 91% PFDoDS - - - - PFTDA 79% 86% 187% 81% 160% 117% PFHxDA 77% 82% 79% 71% 91% 75% PFOcDA 152% - 123% 256% - 156% 4:2 FTSA - - - - 6:2 FTSA 150% 329% 356% 325% 188% 340% 8:2 FTSA 144% 146% 106% 82% 252% 161%

(31)

Table 16: Comparison on signal intensity in isomer samples and K-medium used as a blank in E. coli samples.

ENV1 EC ENV2 EC ENV3 EC PFAS-11 1 EC PFAS-11 2 EC PFAS-11 3 EC

dimetyl - - - - 3/4/5-PFOS 3% 2% 2% 3% 2% 3% 6/2-PFOS 3% 2% 2% 5% 3% 5% 1-PFOS 9% 6% 13% 8% 7% 16% L-PFOS 1% 1% 1% 4% 2% 3% 2/3/4-PFHxS - - - - 1-PFHxS 68% 71% 96% 70% 65% 75% L-PFHxS 38% 40% 43% 116% 89% 118%

Table 17: Comparison on signal intensity in isomer samples and K-medium used as a blank in A. delafieldii samples.

ENV1 AC ENV2 AC ENV3 AC PFAS-11 1 AC PFAS-11 2 AC PFAS-11 3 AC

Table 18: Comparison on signal intensity in isomer samples and K-medium used as a blank in P. nitroreducens samples.

ENV1 PN ENV2 PN ENV3 PN PFAS-11 1 PN PFAS-11 2 PN PFAS-11 3 PN

Microbial binding of per- and polyfluorinated alkyl substances (PFASs) : - Analysis of PFASs in microbes with ultra-performance liquid chromatography – tandem mass spectrometry (UPLC-MS/MS)

Degree project