Response to the comments on the manuscript acp-2021-666 Are reactive oxygen species (ROS) a suitable metric to predict toxicity of carbonaceous aerosol particles?

Abstract. It is being suggested that particle-bound or particle-induced reactive oxygen species (ROS), which significantly contribute to the oxidative potential (OP) of aerosol particles, are a promising metric linking aerosol compositions to toxicity and adverse health effects. However, accurate ROS quantification remains challenging due to the reactive and short-lived nature of many ROS components and the lack of appropriate analytical methods for a reliable quantification. Consequently, it remains difficult to gauge their impact on human health, especially to identify how aerosol particle sources and atmospheric processes drive particle-bound ROS formation in a real-world urban environment. In this study, using a novel online particle-bound ROS instrument (OPROSI), we comprehensively characterized and compared the formation of ROS in secondary organic aerosols (SOA) generated from organic compounds that represent anthropogenic (naphthalene, SOANAP) and biogenic (β-pinene, SOAβPIN) precursors. The SOA mass was condensed onto soot particles (SP) under varied atmospherically relevant conditions (photochemical aging and humidity). We systematically analysed the ability of the aqueous extracts of the two aerosol types (SOANAP-SP and SOAβPIN-SP) to induce ROS production and OP. We further investigated cytotoxicity and cellular ROS production after exposing human lung epithelial cell cultures (A549) to extracts of the two aerosols. A significant finding of this study is that more than 90 % of all ROS components in both SOA types have a short lifetime, highlighting the need to develop online instruments for a meaningful quantification of ROS. Our results also show that photochemical aging promotes particle-bound ROS production and enhances the OP of the aerosols. Compared to SOAβPIN-SP, SOANAP-SP elicited a higher acellular and cellular ROS production, a higher OP and a lower cell viability. These consistent results between chemical-based and biological-based analyses indicate that particle-bound ROS quantification could be a feasible metric to predict aerosol particle toxicity and adverse human effects. Moreover, the cellular ROS production caused by SOA exposure not only depends on aerosol type, but is also affected by exposure dose, highlighting a need to mimic the process of particle deposition onto lung cells and their interactions as realistically as possible to avoid unknown biases.


and no significant OP compared to blank filters. Therefore, the sentence in lines 255-257 has been revised as "All the aqueous extracts of soot particles (fresh, aged, analysed online or offline) showed no detectable ROS concentrations and no significant OP compared to blank filters." Therefore, we think it is not necessary to add ROS control for the aqueous extracts of soot particles in Figure 2 due to no detectable ROS concentrations and no significant OP compared to blank filters. But we added a respective comment in the caption of Figure 2 for clarification: "ROS was also quantified in uncoated soot (fresh and aged) but no ROS content was measured in the aqueous extracts of uncoated soot, see text for details." Comment 5: Page 9, line 263-265. Is there any evidence during the experiments about hygroscopic aerosol growth when RH increased from 40% to 70%? Both naphthalene SOA and beta-pinene SOA are relatively less polar in general (Chhabra et al., 2010;Chen et al., 2015). If water is less partitioned onto aerosols, the impact of humidity on the aerosol phase reaction will be little, and humidity will have a small impact on aerosols. However, for other types of SOA, such as isoprene oxygenated products, it is relatively polar, and their SOA formation, as well as OP, could be potentially impacted by RH. Can the authors conclude the humidity effect on the SOA by using less polar organic matter?
Response: Thank you very much for your suggestion. We did not investigate aerosol hygroscopic property. But, we fully agree with your comment, and we have provided additional information in the revised manuscript (see lines 272-274) as follows: In addition, a change in RH from 40 to 70% showed marginal effects on ROS content in both aerosol types. This maybe because that naphthalene SOA and β-pinene SOA does not contain as highly oxidised compounds as other SOA types, which leads generally to a modest hygroscopicity (Chhabra et al., 2010;Chen et al., 2015) and thus the aerosol phase reactions might not be affected much by RH.
Comment 6: Page 11, line 313-319. Figure 3 shows that the carbon oxidation state for terpene SOA is different between the experiments under 40% RH and that under 70% RH (the highest number is 3 times different). Please provide the explanation for the higher carbon oxidation at the higher RH ? Is it due to the high OH radical concentration at the higher RH ? (high ozone production at the high RH). However, their observed ROS is very similar. It seems that the correlation between ROS and carbon oxidation state is only valid within the type of precursor but it is not sensitive to experimental conditions within the same precursor (i.e., SOA from different RH). This needs to be explained.

Response:
As described in lines 131-132, we generated SOA with a photochemical aging equivalent to 2, 3, 4, 5 and 9 days at ambient OH concentration of 1 x 10 6 molecules/cm 3 . OH concentrations were quantified using the D9-butanol approach as described in lines 129-131 and are therefore accounted for in the data presented here. Overlaying the data of Figure 3c and 3d into one graph (see Figure below) clearly indicates, that for -pinene SOA the higher RH is promoting higher and higher ROS concentrations, while this is not the case for naphthalene SOA.
Unfortunately, the current set of experiments does not allow to investigate the mechanisms driving the higher and higher ROS concentrations of b-pinene SOA at higher RH. This would be interesting to investigate but would be clearly beyond the focus of the current study. Response: The answer to each of your questions and comments is given below:

Q1:
Can the short-lived ROS decay during this period and cause the uncertainties in the analysis?
Response: As described in lines 361-364, a pervious study in our group showed that approximately 75% of the total ROS has a half-life of a few minutes when we compared offline and online ROS analyses in SOA formed from ozonolysis of oleic acid aerosol in a flow tube (Fuller et al., 2014 (2011) reported that the H2O2 yield (a significant component of ROS) from both α-pinene and β-pinene SOA was about 70% smaller after storing the filter samples in petri-dishes in the dark at room temperature for 20 hrs. Our current study found that more than 90% of all ROS components decay in both SOA types when we compared offline and online ROS analyses. All these studies together, clearly indicate that for conventional offline ROS analysis method, significant amount of ROS will be lost during sampling and filter storage. We argue that offline analysis is severely underestimating true ROS concentrations, even if there is only a delay of a few minutes or hours between aerosol sampling and analysis.
That is the main reason why we developed our online ROS instrument, i.e., to capture the most or all of the total ROS components, including short-and long-lived ROS. These key points are described in the manuscript (e.g., lines 339-341, lines 371-373). measurement included both short-lived and long-lived ROS compounds. In addition to the online ROS measurements, we also determined the long-lived ROS fraction only as discussed above, which we compared directly with the cell responses.
In a recent accompanying study, Offer et al. (2021), we used the same aerosols as discussed in this study to expose lung cells to particles directly at the air-liquid interface (ALI) without prior particle collection on filters and extraction. In the study by Offer et al., we compared the online ROS measurement directly with cell exposure results. These results indicated that SOANAP-SP causes higher negative biological responses than SOAβPIN-SP (i.e., higher DNA damage, and higher secretion of malondialdehyde and interleukin-8). The consistent results obtained from these different particle exposure methods (ALI cell exposure vs. filter extracts) may indicate that both types of cell exposure studies are indicative of particle toxicity, although more aerosol systems and multiple biological responses should be investigated to confirm these conclusions.
The above information is summarised in lines 567-575.
Comment 11: Page 23, line 527. Will the ROS products slowly decay in the cell medium in the absence of cell cultures? It may also be useful to test how SOA products decay in the cell culture buffer without cells within 24 hrs.

Response:
The purpose of cell exposure study is to investigate how water-soluble organic fraction causes cell responses. How ROS decay in the cell medium would be interesting to investigate but is beyond the scope of this study.
Comment 12: Table 1: Author increased the concentration of both soot particles and VOC, and Table 1 shows the changes in the oxidative characteristics of particles generated form the different concentration of soot particles and VOC. If there is a change in the concentration of VOC or soot particle only, are those oxidative characteristics influenced? Which one mainly cause this difference in oxidative characteristics?

Response:
In this study, we generated SOA with a photochemical aging equivalent to 2, 3, 4, 5 and 9 days at ambient OH concentration of 1 x 10 6 molecules/cm 3 . The number of OH radicals per VOC molecule available is lower at the higher VOC concentration, which lead to a lower overall degree of oxidation and thus a difference in ROS. Thus, we argue that oxidative characteristics of particles are mainly affected by VOC concentration. As for soot particles, they serve as seed for oxidised VOC reaction products during SOA formation, the change of which could also affect the participation of some ROS components from gas phase into particle phase. The explanation can be found in the first paragraph of section 3.4 (see lines 449-457).
Comment 13: Table 1: Carbon oxidation state of particle from SOAbpin-SP are negative values. What does the negative values of carbon oxidation state of particle mean? Response: Wall losses of gaseous organic oxidation products of the two VOC systems investigated here probably did occur, as it is the case for any laboratory SOA experiment. Effects of gaseous wall losses were not further investigated here but it seems not very likely the ROS components were selectively lost to reactor walls and thus a preferential ROS-specific wall loss effect is probably not significant.

Response
Comment 15: QC/QA: How many data points were used for the QC/QA? This information can improve the reliability of the QC/QA.

Response:
For the online ROS measurement, the ROS concentration under each test condition was continually monitored for about one hour. The additional information has been added in revised manuscript as (see lines 157-158) "The ROS concentration under each test condition was continually monitored for about one hour." For the offline ROS and OP analysis, the additional sentence has been added in lines 179-180 & 206-207, respectively, as "For each sample, three replicate analyses were conducted." For cellular ROS analysis, each sample was tested in six replicate wells. The information can be found in lines 216-217. Response: Thank you for pointing out this typographical error, which has been rectified.

Response:
The typographical error has been rectified. Figure 3 and 7. It would be better to add order number (e.g., a, b, c, and d) for each sub figures.

Response:
The order number has been added in Figures 3 & 7. Meanwhile, the sentence in lines 427-428 has been revised as "As can be seen from Figures 7(a) and (b), we further found that OP and offline ROS measurements for SOANAP-SP are clearly positively correlated, while there is a weaker correlation between OP and offline ROS for SOAβPIN-SP (Figures 7(c) and (d))." Comment 4: It would be also useful for readers to organize the experimental conditions in a Table (in main content or SI) for the experiments described in Section 3.1-3.3.

Response:
The experiments described in Section 3.1-3.3 include different types of tests, each of which has its unique experimental parameters. Therefore, a table would become very large and not straight forward to read. We think that all experimental conditions are clearly described and can be easily found in the main text, and we therefore would prefer not to summarise them in an additional table.

Anonymous Referee #2
Comment 1: Page 11, line 303: The authors attributed the differences in the content of organic peroxides vs. total ROS in naphthalene vs. pinene-derived SOA to the oxidation regimes. My 1st question: isn't there much more in the total ROS than simply the organic peroxide? And, if so, is this comparison valid? My 2nd question: if it is attributed to the oxidation conditions, then which regime is more atmospherically relevant (photo-oxidation vs. ozonolysis)?
Response: We agree that not only peroxides but likely also other compound classes contribute to ROS measured with the HRP/DCFH assay used here. But hydroperoxides, peroxy acids and H2O2 are some of the few compounds which are specifically known to react with HRP and which are also known to be abundant oxidation products of any organic SOA precursor. In section 3.1 we pointed out that quinones and semi-quinones in naphthalene-derived SOA also contribute to ROS. The sentence in lines 315-317 has been revised as "These differences might be explained by the oxidation regimes used (i.e. photo-oxidation vs. ozonolysis) in different studies, and also the different contributions of other known and unknown ROS species in SOA." In this study, we did not compare the ROS yield from photo-oxidation vs. ozonolysis. In our another recent study (paper is under preparation), we found that ROS generated from both biogenic and anthropogenic VOCs are affected by oxidation regime. For example, we found that SOA produced from the photooxidation of carene has a higher ROS/SOA than SOA generated from ozonolysis of carene.
Comment 2: Page 12, Line 333: Do you really have to store the filters for 6 months? It is kind of expected that most of the particle-bound ROS will be lost in that time-frame. A more relevant experiment could have been analysing the filters after couple of days (which is equivalent to ambient filter sampling for days), so that the effect of the integrated filter sampling, could have been better captured.
Response: As described in detail above in our replies to Anonymous Referee #1 (see the response to comment 7) and in lines 361-364, previous studies in our group showed that approximately 75% of the total ROS has a half-life of a few minutes to a few hours (Fuller et al., 2014;Gallimore et al., 201;Steimer et al., 2018). Similarly, Wang et al. (2011) reported that the H2O2 yield (a significant component of ROS) from both α-pinene and β-pinene SOA was about 70% smaller after storing the filter samples in petri-dishes in the dark at room temperature for 20 hrs. Based on these results, we believe that most particle-bound ROS has a very short life (mins-hrs). Therefore, analysing the filters after couple of days could not capture these short-lived ROS.
Comment 3: Section 3.4: I think the relevant discussion of this section actually starts from line 444. The discussion above that line does not fit under the heading of this section. Some rearrangement is warranted in this section.
Response: Thank you for your comment. The first paragraph of section 3.4 provided basic information on (1) how to generate SOA for biological studies, and (2) the relationship of SOA that used for chemical and biological studies, respectively. The paragraph serves as a connecting link between the chemical and biological analysis. We thus think the text is in the right place. However, the sentence in line 447-449 has been revised as below to make them fit well the text.
We also evaluated the oxidative characteristics of aerosols produced from the mixture of 1 mg/m 3 soot particles and 4 mg/m 3 precursor VOCs under 3 day-equivalent photochemical aging at 40% RH for the biological studies discussed here. Comment 4: Lines 450: The insignificant toxicity of fresh or aged-soot particles is surprising and inconsistent with the previous studies. I think it is related with water-insolubility of the soot particles. Did the authors make sure that soot particles remained suspended and are not lost?

Response:
We would like to clarify that we studied exclusively and specifically only the toxicity of aqueous extraction (i.e., water-soluble fraction) of fresh or aged-soot particles and of soot-SOA mixtures. We did not study the toxicity of soot particle, and we have highlighted the texts below, which can be found in the last paragraph of the manuscript (see lines 584-588): "Our findings that aqueous extracts of the soot particles (fresh, aged, analysed online or offline) showed no detectable ROS concentrations and no significant OP compared to blank filters do not necessarily mean that the insoluble part of soot particles are not a health-relevant part of ambient aerosol particles, but indicate that no significant water-soluble fraction exists in the soot produced here that lead to ROS production and OP. Therefore, more research is warranted to identify and quantity if and how the insoluble fraction of soot particles cause a change in cytotoxicity and cellular ROS production." Comment 5: The trend of carbon oxidation state vs. ROS content does not match in Table 1 vs. Figure 3. Figure 3 shows an increase in the ROS content with the carbon oxidation state while Table 1 shows the reverse trend (see top two rows). Can the authors provide an explanation?
Response: In Figure 3, the SOA generated from 0.25 mg/m 3 soot + 1 mg/m 3 precursor VOCs are given. We investigated how the ROS content in SOA are affected by aging conditions, and we found that the ROS content in the SOA increased with an increase in aging time. However, in Table 1, we compared the oxidative characteristics of SOA generated from different concentrations of soot and precursor VOCs under the same photooxidation aging condition. We have clearly explained the reasons of the discrepancy of the two SOA in the first paragraph of section 3.4.