Assessing Ecotoxicity of Size-fractionated Airborne Particulate Matter

. Particulate matter (PM) is grouped as coarse, fine, and ultrafine particles (UFPs) with aerodynamic diameters of 2.5 to 10 μm (PM10), <2.5 μm (PM2.5), and <0.1 μm (PM0.1), respectively. The course and fine fractions have been well characterised from numerous aspects, including potential environmental hazard. However, more and more studies are targeted to the UFP fraction, as they bind relatively higher concentrations of potentially toxic materials and they might penetrate through cell biological barriers, posing higher risk to the biota. In our study, ecotoxic potential of size-fractionated urban aerosol was evaluated, using the kinetic version of the Vibrio fischeri bioluminescence inhibition bioassay. The kinetic protocol makes it possible to avoid false ecotoxicity readings which might appear in case of coloured and/or turbid samples. Our results showed that all PM fractions elucidated significant toxic response, highest toxicity was experienced in the range of 0.25/0.5μm and 0.5/1 μm (with the EC50s of 7.07 and 7.8%). Ecotoxicity in general followed the typical pattern of number size distributions of submicron particles experienced in Europe.

The deleterious effect of the PM2.5-10 fraction on human health has been widely studied (e.g. [6]), also, there are data on the ecotoxic potential of this particulate matter fraction (reviewed by [7]).
It is generally agreed that smaller particles pose higher hazard as they bind relatively higher concentrations of potentially toxic materials and they might penetrate through cell biological barriers [8]. As such, more and more research are targeted to assess the environmental risk posed by these particles. However, while human health problems associated with UFP have been long studied (e.g. [9]) practically no information is available on what ecological risk this fraction might pose on the nonhuman biota.
In order to gain a preliminary view about the ecotoxicity profile of the different fractions, sizefractionated urban dust sample was analysed. For ecotoxicity assessment, the test based on the bioluminescence inhibition of the marine bacterium Vibrio fischeri was used. The species was recently renamed Aliivibrio fischeri [10], however, as most standards still apply the V. fischeri name, it will be used hereinafter. This test is based on the inhibition of the NAD(P)H:FMN oxidoreductase and luciferase enzyme system, which is reflected in the rapid decrease of light emittance of the bacterium. The reduction of light intensity is proportional to the strength of the toxicant, therefore it gives a reliable and easy-to-measure end point ( Fig. 1.).  This bioassay was already used in the 1960's for assessing the deleterious effects of air pollutants [11,12]. Recent studies involve, among others, urban dust (e.g. [13][14][15]), traffic-related emissions (e.g. [16,17]) or fly ash (e.g. [18,19]). A detailed review is given by Kováts and Horváth, 2016 [20], including the application of genetically engineered bacterial strains.
In our laboratory, the Ascent Luminometer is used. This system is based on the kinetic version of the  [21,22]. Here, inhibition is calculated comparing the bioluminescence readings at the beginning and the end of the pre-set exposure, independently from the control. It is suitable to assess the ecotoxicity of solid and/or coloured samples, as false light output readings which might appear due to the physical effect of the presence of colour and/or suspended solids can be avoided. Accoring to our previous results, as dust samples are coloured and contain solid particles, this protocol provides the most reliable results, in comparison to non-kinetic protocols [23]. Light emittance is continuously recorded for the first 30 sec, as such, toxic effect can be immediately detected (Fig. 2).

Fig. 2.
Typical kinetic diagram of an aerosol sample. Light output is recorded in the first 30 sec interval. As the sample contains solid particles, physical interference occurs, which reduces light luminescence reading: the peak is lower than in the control. After the peak, toxicity causes a rapid reduction in the light output while light output remains even in the control. The two rows show the two replicates. G1-H1 (left): control. G2-H2: highest, G12-H12: lowest concentrations.
Particulates were collected onto aluminium substrate. Sampling took place between 20.02. and 24.02, 2017. Each day, sampling was continued for 23 hours. For further analyses, a composite sample was used.
Effective area of the Berner foils (the area which actually contained particulates) was cut into small pieces and placed in a 4 ml vial. Extraction was done by adding 3000 μL DMSO and stirring in Vortex at 750 rpm. The extract was filtered (0.45 μm pore size).
where KF is the correction factor, IC0 and IC30 are the luminescence intensities of the control at the beginning and after 30 min, IT0 and IT30 are the luminescence intensities of the sample at the beginning and after the 30 min contact time.
From the inhibition data of each concentration the software calculates Gamma using the equation below: and the inhibition what belongs to the Gamma=1 value gives the EC50.

Results and Discussion
Results clearly show that UFP has higher toxicity than coarse particles (PM1 and above). Highest toxicity was experienced in the range of 0.25/0.5μm and 0.5/1 μm (with the EC50s of 7.07 and 7.8%). However, it should be noted that all fractions elucidated significant ecotoxicity (Fig 3). Results are expressed as EC50 (calculated concentration of the sample which causes 50% of toxic effect, in this case 50% of bioluminescence reduction). Please note, that the lower the EC50 value, the higher the toxicity. Landkocz et al. (2017) demonstrated that UFP has greater cytotoxic and genotoxic potential than coarse particles [25]. In the study of Mesquita et al (2014), ecotoxicity of size fractionated aerosol samples was assessed on zebrafish embryos, using different endpoints: mortality and deformities. Results revealed that sub-micron fractions were mostly responsible for the toxic effects and putative toxic compounds, mainly PAHs, also concentrated in these fractions [26]. Valavanidis et al. (2006) also demonstrated that the fine particulate PAHs concentrations were higher than coarse particles. [27] An important mechanism for causing toxic effect is the generation of reactive oxygen species (ROS). Studies have already established that reactive oxygen species have the ability to elucidate toxic response in Vibrio fischeri (e.g. [28]).
It can be concluded that while all PM fractions elucidated significant toxic response in the test bacterium V. fischeri, ecotoxicity in general followed the typical pattern of number size distributions of submicron particles experienced in Europe [29]. It should be noted, however, that the average number concentrations of ultrafine (10-100 nm) particles can be higher in extremely polluted areas (e.g. [30]).
Present study rather intends to give a methodological overview about sample preparation and ecotoxicity testing of size-fractionated dust. The kinetic version of the V. fischeri assay as described here is able to assess ecotoxicity from an extremely low quantity of sample, which makes it very useful in aerosol research. However, in order to carry out comparative measurements (such as analytics), a more extended sampling campaign is required.