Potassium 1,1,2,2,3,3,4,4,4-nonafluorobutane-1-sulphonate

Toxicity to other aquatic organisms

Link to relevant study record(s)

Reference

Reference 1

Endpoint:

toxicity to other aquatic vertebrates

Type of information:

experimental study

Adequacy of study:

key study

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

comparable to guideline study with acceptable restrictions

Justification for type of information:

Study was performed on the perfluorobutane sulfonate. Since PFBSK+ will dissociate to the perfluorobutane sulfonate anion and K+ cation in the test medium with pH 7.2 -7.4, toxicity testing on the perfluorobutane sulfonate is equivalent to testing on the potassium salt.

Qualifier:

equivalent or similar to guideline

Guideline:

other: ASTM E1439-98(2004)

GLP compliance:

not specified

Specific details on test material used for the study:

- Name of test material: Perfluorobutane sulfonate
- Source and lot/batch No.of test material: Tokyo KaSei Industry Co., Ltd, Japan.
- Purity:98%

Other substances tested:
Perfluorooctane sulfonate: (98%, Matrix Scientific, US)
Perfluorohexane sulfonate: (> 98.0%, Sigma-Aldrich, US)

Analytical monitoring:

no

Vehicle:

no

Details on test solutions:

PREPARATION AND APPLICATION OF TEST SOLUTION (especially for difficult test substances)
- Method: stir into test medium
- Controls: 10 mL FETAX solution
- Chemical name of vehicle (organic solvent, emulsifier or dispersant): None
- Concentration of vehicle in test medium (stock solution and final test solution(s) or suspension(s) including control(s)): None
- Evidence of undissolved material (e.g. precipitate, surface film, etc.): None

Aquatic vertebrate type (other than fish):

frog

Test organisms (species):

Xenopus laevis

Details on test organisms:

TEST ORGANISM
Adult male and female African clawed frogs (Xenopus laevis) were kept separately in aquaria filled with de-chlorinated tap water, the aquaria temperature was 22 ± 2 C, natural sunlight. The frogs were fed twice per week with feed containing liver: corn flour: soy flour at 60%: 30%: 10% ratio. Water was renewed 2 hrs after feeding.
The Male frogs were dissected after freezing for 10-15 min. The testes were gathered and added to the testes medium (47.0 mL of 1x Ringers solution added to 2.5 mL of fetal bovine serum and 25 mg of gentamicin sulfate) and stored at 4 °C. The dorsal lymph nodes of female claw frogs were injected with 300 - 400 IU of HCG (Human chorionic gonadotropin in 0.8% saline) to induce egg production. After approximately 12 hours, the claw frogs began to lay eggs. The eggs were collected in 150 mm Petri dishes. The rice-grain sized testes were added to 0.3 mL Ringer solution (6.6 g NaCl, 0.15 g KCl, 0.15 g CaCl2, 0.05 g NaHCO3 dissolved in 1L distilled water, pH 7.2 – 7.4) and crushed to release sperm. Immediately sperm and egg were mixed together, let sit for 3-5min, then added to 0.1x MMR solution (0.5844 g NaCl, 0.0149 g KCl, 0.012 g MgSO4,0.0294 g CaC12.2H2O and 0.1192 g HEPES, dissolved in 1L distilled water, pH 7.2 -7.4). After standing about 30 minutes, the fertilized embryo started to flip, the black animal pole up. Using 2% L(+)-Cysteine ( pH 7.8 - 8.0) to dejelly for 7 -8 min, then wash three times with tap water, two times with 0.1xMMR two times, the embryo was finally maintained in the 0.1x MMR solution. Embryos were held at room temperature until test initiation. Normal-growing embryos in blastula stage (NF8) to gastrula stage (NF11) were selected for the FETAX test.

Test type:

semi-static

Water media type:

freshwater

Limit test:

no

Total exposure duration:

96 h

Nominal and measured concentrations:

Nominal only: 0, 35, 42, 50, 60, 72, 86 and 100 mg/L;

Details on test conditions:

Reagent:
Human chorionic gonadotropin (HCG, Yantai Northern Pharmaceutical Co., Ltd.)
L(+)-Cysteine (Beijing Sola Po Technology Co., Ltd.),
HEPES ((4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid) (Sigma, USA),
Fetal Bovine Serum (HyClone, USA),
Gentamicin sulfate (Amresco Corporation, USA),
KCl, NaCl, CaCl2, NaHCO3, MgSO4, CaSO4.2H2O, NaOH, HCl


FETAX test procedure:
Twenty-five (25) normal-growing embryos in blastula stage (NF8) to gastrula stage (NF11) were selected and placed in each of 60 mm diameter petri dishes. Each dose group had 2 to 3 replicates dishes. The final volume was 10 mL in each petri dish. Blank contains 10 mL FETAX solution. Medium renewal at every 24 hours. At each 24 hours, the dead embryos were taken out and placed in 3% formalin solution, and the number of the dead embryos was recorded. After 96 hours exposure, all embryos were placed in 3 % formalin solution, observed under the microscope, and the number of dead, abnormal, and body length were recorded. The experiment was repeated 3 times with embryos from 3 pairs of frogs.

Deformities observed include spinal deformity, tail curvature, back and abdominal deformity, and enlargaed head etc.

Reference substance (positive control):

no

Key result

Duration:

96 h

Dose descriptor:

LC50

Effect conc.:

> 100 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

mortality

Duration:

96 h

Dose descriptor:

EC50

Effect conc.:

> 100 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

other: deformity

Remarks on result:

other: Aggregate malformation

Duration:

96 h

Dose descriptor:

NOEC

Effect conc.:

100 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

other: length

Remarks on result:

other: No effect observed at highest concentration

Details on results:

- Observations on Mortality: No significant embryo death was observed in embryo exposed to PFBS at 42 mg/L or lower. At concentration of 50 mg/L PFBS, embryo death was 13.78%. At 86 mg/L PFBS, embryo death was 20%, which was the highest mortality rate observed for embryo exposed to PFBS (See illustration)

- Observations on Embryo Deformity: The highest observed no-effect dose was 86 mg / L for PFBS, the highest deformity rate was 17.60% at concentration of 100 mg /L PFBS. Since the deformity rate did not reach 50%, it was not able to calculate the EC50.
- Observations on body length: PFBS exposure had no significant effect on embryo growth.

- Mortality and other adverse effects of control: The mean mortality rate in the blank control group was 4.57%; the deformity rate was 6.15%. The results were consistent with ASTM for FETAX (ASTM, 2004).

Results for other substance tested in this study:
PFOS: LC50 (mortality) = 51.46 mg/L (95 CI = 47.5 - 55.4 mg/L); EC50 (deformity) = 108.2 mg/L (95 CI = 85.16 - 193.33 mg/L); minimum concentration to inhibit growth (MGIG) = 35 mg/L

PFHS: LC50 (mortality) > 100 mg/L (No significant embryo death was observed in embryo exposed to PFHS at 42 mg/L or lower. At concentration of 50 mg/L PFHS, embryo death was 15.2%. At 100 mg/L PFHS, embryo death was 25.6%). Deformity: The highest observed no-effect dose was 60 mg / L for PFHS, the highest deformity rate was 21.71% at concentration of 100 mg /L PFHS. Since the deformity rate did not reach 50%, it was not able to calculate the EC50. Growth inhibition: PFHS exposure had no significant effect on embryo growth.

Validity criteria fulfilled:

not specified

Conclusions:

Perfluorooctane sulfonate (PFBS) had a 96h LC50 > 100 mg/Land a 96h EC50 > 100 mg/L (teratogenesis) in African clawed frog (Xenopus laevis) embryo (method similar to ASTM 1439). PFBS did not significantly affect the embryo growth at 100 mg/L.

Executive summary:

The developmental toxicity and teratogenicity of Perfluorooctane sulfonate (PFBS) was evaluated using a semi-static frog embryo teratogenisis assay-Xenopus (FETAX) test (method similar to ASTM 1439-98).

 

Twenty five healthy African clawed frog (Xenopus laevis) embryos in blastula stage (NF8) to gastrula stage (NF11) were exposed to 0, 35, 42, 50, 60, 72, 86 or 100 mg/L PFBS for 96 hours, with removal of any dead embryos and solution renewal every 24 hours. After 96 hours exposure, all embryos were fixed in 3 % formalin solution, and observed under the microscope. The total number of dead embryos, deformed embryos, and the body lengths were recorded. The highest mortality rate was 20%. Therefore, the LC50 was > 100 mg/L. The highest observed deformity rate was 17.60%. Therefore, the EC50 (teratogenesis) was >100 mg/L. PFBS exposure had no significant effect on embryo growth.

 

This study is similar to an accepted national test guideline and was well documented, but was not in accordance with GLP criteria. This study is considered reliable with restriction. It is suitable for use in Risk Assessment, Classification & Labeling, and PBT Analysis.

Reference 2

Endpoint:

toxicity to other aquatic vertebrates

Type of information:

experimental study

Adequacy of study:

other information

Reliability:

3 (not reliable)

Rationale for reliability incl. deficiencies:

unsuitable test system

Remarks:

chemometric analysis on cultured cell line

Justification for type of information:

The study report distinguishes between perfluorinated carboxylic acids and perfluoroalkyl sulfonates, but does not explicitly state that perfluorobutanesulfonate was tested as the potassium salt. However, the potassium salt is the likeliest test material given substance availability. Regardless of the material tested, at the low concentrations tested (1 nM to 10 µM) and pH range of the culture medium, and noting that concentration was expressed in molar units, testing is equivalent to PFBSK+.

Qualifier:

no guideline available

Principles of method if other than guideline:

- Principle of test: Chemometric analysis of FTIR spectra from fixed cell suspensions
- Short description of test conditions: Kidney epithelia cells were exposed at three points of differentiation to one of four perfluoroalkyl acid substances (PFAAs). Cells were harvested and fixed for analysis.
- Parameters analysed / observed: FTIR spectra for cell suspensions in the region of 1800 cm(-1) to 900 cm(-1)

GLP compliance:

no

Specific details on test material used for the study:

From Fluka (Austria), purity not specified

Analytical monitoring:

no

Vehicle:

yes

Remarks:

DMSO, 25 µL in 5 mL test medium

Details on test solutions:

PREPARATION AND APPLICATION OF TEST SOLUTION
- Method: Substances were dissolved in DMSO to make a stock solution and serially diluted in DMSO for addition to cell culture medium.
- Controls: DMSO (negative control)
- Concentration of solvent in test medium: 5 mL/L (note that this is 50X higher than the value stated in OECD bioconcentration and fish toxicity test guidelines, i.e. 0.1 mL/L.
Other substances in test:
Perfluorooctane sulfonate (unspecified salt)
Perfluorooctanoic acid
Perfluorononanoic acid

Aquatic vertebrate type (other than fish):

frog

Test organisms (species):

Xenopus laevis

Details on test organisms:

TEST ORGANISM
- Common name: African clawed frog, A6 kidney epithelial cells
- Strain: Accession ATCC CCL-102
- Source: American Type Culture Collection
- Age at study initiation: One or eight days
- Method of propagation: Cells were cultured in modified L15 medium (70% Leibovitz media, 19% filter-sterilized Milli-Q water, 10% fetal bovine serum (FBS), 1% penicillin/streptomycin) in T75 polystyrene flasks at 5% CO2 and 26 °C. Cells were trypsinized and resuspended in medium for seeding into fresh flasks.

Test type:

static

Water media type:

other: Cell culture medium

Limit test:

no

Total exposure duration:

24 h

Nominal and measured concentrations:

Nominal concentrations only
Experiment 1: 0 (solvent control), 1 nM, 10 nM, 100 nM, 1 µM, 10 µM
Experiment 2: 0 (solvent control), 1 nM, 10 nM, 10 µM
Experiment 3: 0 (solvent control), 10 nM, 10 µM

Details on test conditions:

TEST SYSTEM
- Test vessel: T25 polystyrene tissue culture flask containing 5 mL medium
- No. of organisms per vessel: 500,000 cells/flask
- No. of vessels per concentration (replicates): five (Exp 1&2) or two (Exp 3) each for the four PFAAs tested.
- No. of vessels per control (replicates): twenty (Exp 1&2) or eight (Exp 3), DMSO control only

TEST MEDIUM / WATER PARAMETERS
- Test medium: Article specifies modified L15 medium for propagation (above) but mentions only that a complete medium was used for the test. It is presumed that the same medium was used for both.

EXPOSURE CONDITIONS:
- Experiment 1: Cells were seeded at time zero. At 24 hours, test substance was added in 25 µl DMSO. At 48 hours, cells were disaggregated and fixed in 70% ethanol for further analysis
- Experiment 2: Cells were seeded at time zero. At 24 hours, test substance was added in 25 µl DMSO. At 48 hours, treated medium was removed by aspiration and replaced with untreated medium. Cells were then cultured for seven additional days, disaggregated and fixed in 70% ethanol for further analysis.
- Experiment 3: Cells were seeded at time zero. At eight days, test substance was added in 25 µl DMSO. After exposure of 24 hours, cells were disaggregated and fixed in 70% ethanol for further analysis.
By day seven cells grow to confluence and exhibit differentiation in the form of dome structures that form due to membrane assembly at the top and bottom of the cell monolayer, upon action of Na+ transporters on distribution of fluids on either side of the membranes. The exposure conditions thus examine undifferentiated cells (experiment 1), cells exposed and allowed to differentiate (experiment 2), and cells exposed after differentiation (experiment 3).

EFFECT PARAMETERS MEASURED :
Effects in these experiments were determined by chemometric analysis of Attenuated Total Reflectance - Fourier Transform Infrared (ATR-FTIR) spectrometry done on the fixed cell suspensions. After fixation, an aliquot of each test flask was applied to the surface of a slide and allowed to dry. IR spectra were obtained with a Bruker Vector 22 FTIR spectrometer with Helios ATR attachment containing an ≈250 μm × 250 μm aperture diamond crystal. 32 Scans were acquired for each spectrum, with a spectral resolution of 8 cm(-1) and a data spacing of 4 cm(-1). Ten spectra were acquired from each slide, for a total of 1200 spectra for experiment 1, 800 spectra for experiment 2, and 320 spectra for experiment 3.

The spectral region from 1800 cm(-1) to 900 cm(-1) is considered useful for biological fingerprinting. Accordingly, spectra were cropped to this region, baseline corrected, and normalized to the 1650 cm(-1) peak corresponding to Amide I. Spectra were then mean-centered (the average spectrum was subtracted to convert absorbance peaks to deviations from the average absorbance). Classes were then assigned to the spectra. The spectra were broken into 235 bands at a spacing of 4 cm(-1) for the analysis (because of mean centering, the average absorbance for each band over all the spectra was zero).

Two types of analysis were done to the data sets. Principal component analysis with Linear discriminant analysis (PCA-LDA) was used to examine dose-related effects for each substance within each of the three experiment (twelve total PCA-LDA models). ANOVA-Simultaneous Component Analysis (ASCA) was used to examine overall differences related to dose and substance for each of the three experiments.

PCA-LDA: PCA is a means of isolating variance in the original data by assembling independent linear combinations of variables in a way that concentrates the largest amount of variance into the first principal component, and afterwards concentrates the maximum remaining variance into each subsequent component. In this experiment, the first ten principal components captured 95% of the variance. LDA was then done. LDA is similar to PCA in applying a mathematical transformation to the data set, but depends on a priori assignment of classifications to the dataset. LDA then determines which factors contribute most strongly to the result. For each model, the results of the analysis were visualized through one-dimensional (1-D) scores plots, which depict the effect according to the most important discriminant. The scores plots were used to study dose−response effects of PFASs by examining the proximity in multivariate distance among exposed and control samples (Figure 1). The wavenumbers contributing to this discrimination were identified using cluster vectors plots. PCA-LDA was done using MATLAB 8.3.0 R2014a (The Math Works, Natick, MA, USA) and the IrootLab toolbox (http://irootlab.googlecode.com).

ASCA: ASCA is an multivariate extension of ANOVA that partitions the variation in a dataset (in matrix form) according to preselected factors, with a number of levels for each factor. The overall variation in the matrix of the data is the sum of the variation of in each partition matrix plus a matrix for residual variation. Simultaneous component analysis, which is equivalent to principal component analysis done on the partitions, allows an estimate of the contribution of each factor to variability. Significance is measured by randomly reassigning the levels for each factor and recalculating the overall result (a permutation test, in this case with 10,000 permutations). ASCA was done using PLS Toolbox 7.8 (eigenvector Research Inc., Wenatche, WA, USA) within in a MATLAB 8.3.0 R2014a environment.

Cell counts: In addition, multiple T25 flasks for each substance were seeded as above and used for cell counts. Duplicate counts were taken of three flasks at time zero using a hemocytometer. At 24 h and 48 h, triplicate flasks were washed, trypsinized, resuspended, counted. Cell number was normalized to average initial count and expressed as a percentage of initial (where T0 = 100%).

Key result

Duration:

48 h

Dose descriptor:

NOEC

Effect conc.:

> 10 µmol/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

other: cell count

Remarks on result:

other: No marked differences at highest concentration tested.

Details on results:

The majority of the publication discusses chemometric analysis of the effects of PFAAs, at low exposure level, in fixed whole-cell suspensions. Three of the twelve exposures and subsequent PCA-LDA analyses were done using PFBSK+. In all PCA-LDA analyses, the conclusions were based on degree of overall differentness among spectra. The first two analyses, corresponding to the exposure of undifferentiated cells without (experiment 1) and with (experiment 2) a post-exposure period for cell differentiation, showed the effect occuring at the 1 nM exposure level, with no or smaller overall difference at the next-higher exposure level. In contrast, the largest discrimination was seen at the highest dose (10 µM) when previously-differentiated cells were exposed (experiment 3). For other PFAAs, the greatest difference from control was at 1 nM concentration in experiment 1, however in experiment 2 PFOA showed greatest effect at 1 nM while PFOS and PFNAs had greatest effect at 10 µM. In experiment 3, PFOA showed greatest effect at 10 nM while PFOS and PFNA showed greatest effect at 10 µM. It should be noted that no attempt was made to establish the statistical significance of the analysis.

Cluster-vector plots were used to identify the wavebands associated with the previously identified differences, in order of impact on the difference. Specific spectral features contained within the band were then identified and used to make predictions about potential impacts at each stage of differentiation. These spectral features are presented in Table 1. This analysis is used to inform speculation that exposure to PFAAs may lead to genotoxic events at early exposure and subsequent alterations in protein expression and lipid metabolism. However, given the 4 cm(-1) waveband analyzed, the peak assignment is ambiguous. A biological relevance is difficult to assign.

The ASCA results suggest indicate that very little of the variability in the spectra (≤8%) is due to differences in the factors for chemical tested or exposure level (Table 2), and that most of the variability (≥85%) was partitioned into the residual matrix (i.e, the variability was not explained by an experimental factor). Perturbation analysis of the ASCA result indicates that the most factors had a statistically significant effect (P<0.05), but given the limited explanatory power of the experimental factors a biological relevance is difficult to assign.

Reported statistics and error estimates:

Statistical analysis is inherent to the experimental design and is discussed with experimental methods above.

Table 1, Principal segregating wavenumbers and associated biomolecular entities derived from cluster vectors plots of PCA-LDA

Experiment	Concentration with maximum discrimination for experiment	Peak associated with wavebands showing highest effects¹ (cm-1)	Biological fingerprint for Peak
Experiment 1
	PFBS 1 nM	1080	Stretching PO2- symmetric vibrations
		994	C-O ribose, C-C
		1201	PO2- asymmetric (phosphate I)
		1403	Symmetric -CH3 bending modes of the methyl groups of proteins
		1549	Amide II
		1698/9	C2=O guanine/ N-H thymine
		1489	In-plane CH bending vibration
	PFOS 1 nM	1581	Ring C-C stretching of phenyl
		1567	Ring base
		1137	Oligosaccharide C-OH stretching band
		996	C-O ribose, C-C
		1698/9	C2=O guanine/ N-H thymine
		1095	Stretching PO2-symmetric vibrations
		1717	C=O thymine; C=O stretching vibration of DNA and RNA; C=O stretching vibration of purine base
		1373	Stretching C-N cytosine, guanine
		1736	C=O stretching (lipids)
	PFOA 1 nM	1494	In-plane CH bending vibration
		1543	Amide II
		1020	DNA
		1524	Stretching C=N, C=C
		1555	Ring base
		1444	δ(CH2), lipids, fatty acids
		1250	Amide III
	PFNA 1 nM	1204	Vibrational modes of collagen proteins-Amide III
		1558	Ring base
		1624	Unassigned band
		1728	C=O band
		1089	Stretching PO2- symmetric in RNA
		1705	C=O thymine
		1408	Unassigned band
		1258	PO2- asymmetric (phosphate I)
Experiment 2
	PFBS 1 nM	1517	Amide II
		1510	In-plane CH bending vibration from the phenyl rings
		1616	Ring C-C stretching of phenyl
		1643	Amide I band (arises from C=O stretching vibrations)
		1559	Ring base
		1540	Protein Amide II absorption predominately β-sheet of Amide II
		1750	ν(C=C) lipids, fatty acids
	PFOS 10 µM	1543	Amide II
		1524	Stretching C=N, C=C
		1559	Ring base
		1652	Amide I
		1396	Symmetric -CH3 bending of the methyl groups of proteins
		1620	Peak of nucleic acids due to the base carbonyl stretching and ring breathing mode
	PFOA 1 nM	1647	Amide I in normal tissues
		1192	Unassigned band
		1535	Stretching C=N, C=C
		1620	Peak of nucleic acids due to the base arbonyl stretching and ring breathing mode
	PFNA 10 µM	1562	Unassigned band
		1146	Phosphate and oligosaccharides
		1026	Carbohydrates peak for solutions; vibrational frequency of CH2OH groups of carbohydrates (including glucose, fructose, glycogen, etc.); glycogen
		1254	Amide III
		1400	Symmetric stretching vibration of COO- group of fatty acids and amino acids
		1713	C=O thymine
Experiment 3	PFBS 10 µM	1698/9	C2=O guanine/ N-H thymine
		1620	Peak of nucleic acids due to the base carbonyl stretching and ring breathing mode
		1068	Stretching C-O ribose
		1639	Amide I
		1567	Ring base
		994	C-O ribose, C-C
		1119	Symmetric stretching P-O-C; C-O stretching mode
		1234	Composed of Amide III and phosphate vibration of nucleic acids
		1400	Symmetric stretching vibration of COO- group of fatty acids and amino acids
	PFOS 10 µM	1119	Symmetric stretching P-O-C; C-O stretching mode
		1504	In-plane CH bending from the phenyl rings
		1517	Amide II
		1396	Symmetric -CH3 bending of the methyl groups of proteins
		1659	Amide I
		1559	Ring base
		1539	Protein Amide II absorption predominately β–sheet of Amide II
		1724	C=O stretching band mode of the fatty acid ester
	PFOA 10 nM	1400	Symmetric stretching vibration of COO- group of fatty acids and amino acids
		1647	Amide I in normal tissues- for cancer is in lower frequencies
		1512	In-plane CH bending vibration from the phenyl rings
		1113	Symmetric stretching P-O-C
	PFNA 10 µM	1512	In-plane CH bending vibration from the phenyl rings
		1701	C=O guanine
		1651	Amide I
		1562	Unassigned band
		1393	Unassigned band
		1528	C=N guanine
		1748	ν(C=C) lipids, fatty acids
		1732	Absorption band of fatty acid ester; Fatty acid ester band

1, Wavenumbers for each model are listed from high to low impact on the analysis

Table 2. ASCA Modeling: Significance and Partitioning of the Total Variance into the Individual Terms Corresponding to Factors and Interaction.

Experiment	Factor	Percentage of variation¹	Significance (p-value)
1	Chemical	2	0.001*
	Dose	3	0.001*
	Chemical x Dose²	7	0.001*
	Residuals	89
2	Chemical	2	0.004*
	Dose	3	0.001*
	Chemical x Dose	3	0.3
	Residuals	93
3	Chemical	8	0.3
	Dose	8	0.08*
	Chemical x Dose	12	0.08*
	Residuals	85

1, Percentage of variation expressed as sums of squared deviations from the overall mean and not variances.

2, Interaction term for chemical and concentration

Validity criteria fulfilled:

not applicable

Conclusions:

Exposure to low concentrations (1 nM to 10 µM ) of PFBSK and other perfluoroalkyl acid substances (PFAAs) caused changes in infrared spectra of fixed, whole-cell suspensions of Xenopus laevis A6 kidney epithelial cells. The impact of PFAAs was statistically significant overall but explained very little of the variability. PFBSK+ up to 10 µM had no effect on cell counts.

Executive summary:

Impacts of low levels of exposure of PFBSK and three other perfluoroalkyl acid substances (PFAAs, being perfluorooctanesulfonate, perfluorooctanoic acid, perfluorononanoic acid) were examined in a chemometric approach using Xenopus laevis A6 kidney epithelial cells in cell culture. Cells were exposed for 24 hours to concentrations ranging from 1 nM to 10 µM of one of the four PFAAs a) before differentiation with immediate analysis, b) before differentiation with seven days allowed for differentiation before analysis, or c) after differentiation. Attenuated Total Reflectance-Fourier Transform Infrared spectroscopy was used for data gathering, with data analysis by Principal Component Analysis with Linear Discriminant Analysis (PCA-LDA), and by ANOVA-Simultaneous Components Analysis (ANOVA-SCA). The effect of the type of chemical, the concentration (independent of chemical) and the interaction between chemical and concentration were generally statistically significant after ANOVA-SCA, but in aggregate were unable to explain the majority of the effects seen in the experiments (≥85%). PCA-LDA analysis was used to identify spectral features associated with the greatest differences among exposure levels for each PFAA and experiment. For all PFAAs, the greatest difference from control was at 1 nM concentration in experiment 1 (undifferentiated cells analyzed immediately after exposure); however, in experiment 2 (undifferentiated cells exposed and allowed to differentiate before analysis), PFBSK+ (and PFOA) showed greatest effect at 1 nM while the other two PFAAs had greatest effect at 10 µM. In experiment 3 (differentiated cells analyzed immediately after exposure), PFBSK+ showed greatest effect at 10 µM. The spectral features associated with these differences varied among the treatments. No assessment of statistical significance of results was provided at this stage of the experiment. A biological significance cannot be assigned for the chemometric analysis at this time. In direct tests of cell proliferation, no effect was seen on cell counts up to 10 µM PFBSK+ (highest dose tested).

The paper used a chemometric approach to assess impacts of PFAAs. The relevance of this approach to ecotoxicology has not been established. Further, the biological implications of this approach for development of adverse outcome pathways is unclear. This study is assigned a Klimish 3 reliability score.

Reference 3

Endpoint:

toxicity to other aquatic vertebrates

Type of information:

experimental study

Adequacy of study:

weight of evidence

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

study well documented, meets generally accepted scientific principles, acceptable for assessment

Qualifier:

no guideline followed

Principles of method if other than guideline:

- Principle of test: Survival and histopathological endpoints
- Short description of test conditions: Tadpoles were exposed in aquaria from five days postfertilization until 2 months post-metamorphosis
- Parameters analysed / observed: Survival, growth (body weight), hepatosomatic index and hepatohistology, morphology and histology of gonads, sex ratio, expression of Estrogen Receptor and Androgen Receptor in brain and liver.

GLP compliance:

no

Specific details on test material used for the study:

Factor purity >98%

Analytical monitoring:

yes

Details on sampling:

Samples were taken at test initiation, on Days 61, 62, and 63, and at test termination. These sample days cover solutions at renewal and at one and two days post-renewal. The accompanying method paper describes prefiltration of 800-mL natural waters samples with 0.22 µm nylon filters, and storage up to seven days at 4 °C before extraction and analysis.

Vehicle:

no

Details on test solutions:

PREPARATION AND APPLICATION OF TEST SOLUTION (especially for difficult test substances)
- Method: Test substances were dissolved in DMSO (PFBS solutions were 100 mg/mL). DMSO concentration in all test chambers was 0.001% v/v (10 µL/L)
- Control: solvent control (DMSO) only, 17-beta-estradiol (E2), 5-alpha-androstan-17-beta-ol-3-one (DHT)

Aquatic vertebrate type (other than fish):

frog

Test organisms (species):

Xenopus laevis

Details on test organisms:

TEST ORGANISM
- Common name: African clawed frog
- Source: Internal laboratory colony, adult frogs obtained from Nasco (USA).
- Age at study initiation: 5 days post fertilization. Tadpoles were at stage NF 46/47
- Method of breeding: A pair of 3-year old X. laevis adults were injected by human chorionic gonadotropin (male 300 IU, female 700 IU) to induce breeding. Adults had been maintained in charcoal-filtered, dechlorinated tap at 22 ± 2 °C with a 12-hour light:dark cycle, and fed three times weekly with 1:1 amphibian feed:pork liver. Fertilized eggs were incubated in dechlorinated tap water at 22 ± 2 °C.
- Feeding during test: yes
- Food type: live Artemia
- Frequency: 3 times daily

Test type:

semi-static

Water media type:

freshwater

Limit test:

no

Total exposure duration:

13 wk

Remarks on exposure duration:

Duration was from development stage 46/47 to 2 months postmetamorphosis, a period of approximately 4 months (i.e, 13 weeks)

Hardness:

ca. 150 mg/L as CaCO3

Test temperature:

22 ± 2 °C

pH:

6.5 - 7.0

Dissolved oxygen:

>5 mg/L

Nominal and measured concentrations:

Nominal concentrations: Solvent control (DMSO only), 0.1 µg/L, 1 µg/L, 100 µg/L, 1000 µg/L. Perfluorooctanesulfonate was tested in separate chambers at the same concentrations.
Measured concentrations: see Table 1.

Details on test conditions:

TEST SYSTEM
- Test vessel: Aquaria, glass, with 18 L water.
- Renewal rate of test solution: Every second day.
- No. of organisms per vessel: 25
- No. of vessels per concentration (replicates): 3
- No. of vessels per control (replicates): 3 for E2, 1 for DHT
- No. of vessels per vehicle control (replicates): 3

TEST MEDIUM / WATER PARAMETERS
- Source/preparation of dilution water: Charcoal-filtered, dechlorinated tap water.
- Chlorine: <5.0 µg/L
- Iodine: 2.14 - 3.92 µg/L
- Culture medium different from test medium: no

OTHER TEST CONDITIONS
- Photoperiod: 12 hour light:dark
- Light intensity: fluorescent light to give 600 - 1000 lux at water surface

EFFECT PARAMETERS MEASURED (with observation intervals if applicable) : Mortality and growth daily, all other parameters after euthanization at end of exposure period. Total weight was recorded and the individual was dissected. Liver tissue of each frog was weighed to calculated hepatosomatic index (HSI). Liver samples from four males and four females from each tank (or all animals from the single DHT tank) were fixed, sectioned, and stained for histological examination. Gonadal morphology for each individual was examined by stereomicroscope to determine sex or intersex. Gonad and kidney from all individuals were fixed, sectioned and stained for histological examination. Additionally, liver subsamples from those as for liver histology, as well as brain tissues from the corresponding individuals, were treated with TRIzol reagent for RNA extraction and gene expression analysis by quantitative reverse-transcription-PCR (qRT-PCR). Genes of interest were androgen receptor (AR), estrogen receptor (ER), aromatase (ARO) and ribosomal protein L8. This last is a housekeeping gene constitutively expressed, and was used to normalize expression of other genes to general cell metabolism. qRT-PCR was done using the Quantscript RT Kit for reverse transcription, in the MX RT-PCR system (Stratagene MX3005P, USA) with SYBR Green Real Master Mix for PCR.

Reference substance (positive control):

yes

Remarks:

17-beta-estradiol (E2) and 5-alpha-androstan-17-beta-ol-3-one (DHT)

Key result

Duration:

13 wk

Dose descriptor:

NOEC

Effect conc.:

1 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

other: weight, mortality, hepatosomatic index, sex ratio, gonadal histology

Remarks on result:

other: No significant effect at highest concentration tested

Details on results:

- Behavioural abnormalities:
- Observations on body length and weight: Neither test substance nor either positive control caused any significant change in survival or body weight.
- Other biological observations: Neither test substance nor either positive control caused any significant change in hepatosomatic index, gonad histology, sex ratio/intersex, or expression of aromatase (involved in sex hormone synthesis) in brain or liver. It should be noted that one positive control, 17-beta-estradiol (E2), caused sharply altered sex ratios (ca 70% female, 25% intersex, 5% male) and induced intersexual gonads in both testes and ovaries. The other positive control, 5-alpha-androstan-17-beta-ol-3-one (DHT), had no effects on sex ratio or gonad morphology. Hepatocyte degeneration was observed at 0.1 mg/L, and hepatocyte hypertrophy also occurred at 1 mg/L. Neither observation was sex-related, and hepatocyte hypertrophy was not accompanied by an increase in HSI. These findings are not considered biologically relevant.

Results of Estrogen Receptor (ER) and Androgen Receptor (AR) expression studies are difficult to interpret (see Illustration). The ER positive control E2 showed no effect on ER expression in male or female frog brain or liver, while resulting in intersex and sharply altered sex ratios. The AR positive control DHT showed no effect on intersex or sex ratio, while showing significant increases in ER expression in male and female brain (but not liver), and significant increases in AR expression in male brain. While statistically significant, the factor increase is unexpectedly low for the AR positive control. More troubling is the lack of response to E2 despite obvious physiological effects. PFBSK+ showed a similar pattern in ER and AR to DHT. Because of the lack of relevant response to either positive control, the results cannot be interpreted at this time.

- Mortality of control: None
- Other adverse effects control: None

Reported statistics and error estimates:

Statistical analysis was performed using SPSS software version 16.0 (SPSS, USA). The data were checked for normal distribution (Kolmogorov-Smirnov test) and homogeneity of variance (Levene test). Post-hoc comparison was made using Dunnett’s test (variance homogeneity) or Dunnett’s T3 test (variance heterogeneity). Differences in body weight, HSI, and gene expression level were tested by one-way ANOVA. Differences in survival rate and sex ratio were evaluated with X² tests. Significance level was p < 0.05.

Validity criteria fulfilled:

not applicable

Conclusions:

X. laevis exposed from 5 days post-fertilization until 2 months past metamorphosis (ca. 13 weeks total) showed no effects up to 1 mg/L on survival and growth (weight), hepatosomatic index, sex ratio and intersex, or gonadal histology.

Executive summary:

Long-term effects of PFBSK+ on African clawed frog (Xenopus laevis) were addressed in a study of approximately 13 weeks' duration. Exposure began five days post-fertilization and continued two months past metamorphosis. Concentrations of 0.1 µg/L, 1 µg/L, 100 µg/L, and 1 mg/L were tested. Two controls, 17-ß-estradiol (E2) and 5 alpha-androstan-17-beta-ol-3-one (DHT), were used to verify endocrine-disruptive effects. No effects were seen for PFBSK+ on survival, growth, sex ratio, intersex, or gonadal histology. Hepatosomatic index (ratio of liver weight to total weight) was not affected.  Among positive controls, E2 had pronounced effects on sex ratio, intersex, or gonadal histology, whereas DHT did not. However, E2 had no significant effects on estrogen or androgen receptor transcript level, and DHT induction of transcript was modest at best. Owing to lack of appropriate response among positive controls, the genetic aspect of this experiment cannot be evaluated.

The study used scientifically valid principals, had analytical confirmation of exposure, and was published in a peer-reviewed scientific journal. Lack of appropriate response to positive controls impacts the genetic phase of the experiment only. The results of the other phases are considered reliable with restrictions, and acceptable for use in Risk Assessment, Classification & Labelling, and PBT Analysis.

Description of key information

Perfluorooctane sulfonate (PFBS) had a 96h LC50 > 100 mg/L and a 96h EC50 > 100 mg/L (teratogenesis) in African clawed frog (Xenopus laevis) embryo (method similar to ASTM 1439). PFBS did not significantly affect the embryo growth at 100 mg/L.

Additional information

Toxicology of PFBSK+ to the African clawed frog (Xenopus laevis) was examined in three studies. The key study was a semi-static frog embryo teratogenisis assay-Xenopus (FETAX) test, which is similar to ASTM 1439-98. Twenty five healthy African clawed frog embryos in blastula stage (NF8) to gastrula stage (NF11) were exposed to 0, 35, 42, 50, 60, 72, 86 or 100 mg/L PFBS for 96 hours, with removal of any dead embryos and solution renewal every 24 hours. After 96 hours exposure, all embryos were fixed in 3 % formalin solution, and observed under the microscope. The total number of dead embryos, deformed embryos, and the body lengths were recorded. The highest mortality rate was 20%. The highest observed deformity rate was 17.60%. PFBS exposure had no significant effect on embryo growth. This study is similar to an accepted national test guideline and was well documented, but was not in accordance with GLP criteria. This study is considered reliable with restriction. The second study was a publication on a study of approximately 13 weeks' duration. Exposure began five days post-fertilization and continued two months past metamorphosis. Concentrations of 0.1 μg/L, 1 μg/L, 100 μg/L, and 1 mg/L were tested. Two controls, 17-ß-estradiol (E2) and 5 alpha-androstan-17-beta-ol-3-one (DHT), were used to verify endocrine-disruptive effects. No effects were seen for PFBSK+ on survival, growth, sex ratio, intersex, or gonadal histology up to 1 mg/L. Hepatosomatic index (ratio of liver weight to total weight) was not affected. Among positive controls, E2 had pronounced effects on sex ratio, intersex, or gonadal histology, whereas DHT did not. However, E2 had no significant effects on estrogen or androgen receptor transcript level, and DHT induction of transcript was modest at best. Owing to lack of appropriate response among positive controls, the genetic aspect of this experiment cannot be evaluated. The study used scientifically valid principals, had analytical confirmation of exposure, and was published in a peer-reviewed scientific journal. Lack of appropriate response to positive controls impacts the genetic phase of the experiment only. The results of the other phases are considered reliable with restrictions, and acceptable for use in Risk Assessment, Classification & Labelling, and PBT Analysis. A third study examined impacts of low levels of exposure of PFBSK and three other perfluoroalkyl acid substances (perfluorooctanesulfonate, perfluorooctanoic acid, perfluorononanoic acid) in a chemometric approach using alterations of infrared spectra in lysates of  Xenopus laevis A6 kidney epithelial cells exposed to PFBSK+ in cell culture. The effect of the type of chemical, the concentration (independent of chemical) and the interaction between chemical and concentration were generally statistically significant after ANOVA-Simultaneous Components Analysis, but in aggregate were unable to explain the majority of the effects seen in the experiments (≥85%). Principal Component Analysis with Linear Discriminant Analysis was used to identify spectral features associated with the greatest differences among exposure levels for each PFAA and experiment.  The exposure level associated with the greatest difference from control varied among the three exposure schedules, as did the spectral features associated with these differences.  The statistical significance of these differences was not evaluated. A biological significance cannot be assigned to the chemometric analysis at this time. The relevance of this approach to ecotoxicology has not been established. Further, the biological implications of this approach in development of adverse outcome pathways are unclear. This study is assigned a Klimish 3 reliability score and is not considered further.