https://orcid.org/0000-0003-3171-8958
Abstract
Bosnia and Herzegovina (B&H) is among the European countries with the highest rate of air pollution-related death cases and the poorest air quality. The main causes are solid fuel consumption, traffic, and the poorly developed or implemented air pollution reduction policies. In addition, the city of Sarajevo, the capital of B&H, suffers temperature inversion episodes in autumn/winter months, which sustain air pollution. Human biomonitoring studies may be confounded by the lifestyle of subjects or possible metabolic alterations. Therefore, this study aimed to evaluate Ligustrum vulgare L. as a model for air pollution monitoring by measuring DNA damage at one rural and two urban sites. DNA damage was measured as tail intensity (TI) in L. vulgare leaves, considering seasonal, sampling period, leaf position and staging, and spatial (urban versus rural) variation. Effects of COVID-19 lockdown on TI were assessed by periodical monitoring at one of the selected sites, while in-house grown L. vulgare plants were used to test differences between outdoor and indoor air pollution effects for the same sampling period. Significantly higher TI was generally observed in leaves collected in Campus in December 2020 and 2021 compared with March (P < 0.0001). Outer and adult leaves showed higher TI values, except for the rural site where no differences for these categories were found. Leaves collected in the proximity of the intensive traffic showed significantly higher TI values (P < 0.001), regardless of the sampling period and the stage of growth. In regards to the COVID-19 lockdown, higher TI (P < 0.001) was registered in December 2020, after the lockdown period, than in periods before COVID-19 outbreak or immediately after the lockdown in 2020. This also reflects mild air pollution conditions in summer. TI values for the in-house grown leaves were significantly lower compared to those in situ. Results showed that L. vulgare may present a consistent model for the air pollution biomonitoring but further studies are needed to establish the best association between L. vulgare physiology, air quality data, and air pollution effects.
Introduction
According to the World Health Organization, in 2016 air pollution in Bosnia and Herzegovina (B&H) was directly implicated in 159.3 deaths cases/100 000 people and, in winter months, Sarajevo was among the most polluted cities in Europe [1]. The prognoses are that the premature mortality due to air pollution exposure will double by 2050 worldwide [2], with vulnerable groups, children, and elderly adults, as the most adversely affected [3,4].
Sarajevo, the capital of B&H, is settled in the narrow valley of the Miljacka River. Average altitude is 550 m above the mean sea level (latitude 43° 51ʹ 27″N, longitude 18° 24ʹ 48″E). Surrounded by the mountains and forested hills, Sarajevo citizens frequently suffer from the effects of temperature inversions in winters [5] when high levels of pollution coincide with heavy fog resulting in prolonged periods of smog. The population is estimated at around half a million in the full metropolitan area. The Mediterranean influence is blocked by the mountains to the south, so the climate is continental with four well defined seasons: spring (April–June), summer (July–August), autumn (September–November), and winter (December–March). The coldest month is January (−1.3°C avg.), while the warmest is July (19.1°C avg.), and the average annual temperature is 9.5°C [6]. Average precipitation ranges from 64 mm in February to 91 mm in September.
The air quality is assessed based on the concentration of particulate matter (PM) of different sizes, and gaseous pollutants NO2, SO2, O3 [7]. The daily upper limits for PM10 of 50 μg/m3 with a tolerance value of 62.5 μg/m3 are defined by the Ordinance on the manner of performing air quality monitoring and defining the types of pollutants, upper limits, and other air quality standards in Federation of Bosnia and Herzegovina (FB&H) [8]. However, WHO recommends annual average of upper limits for PM2.5 and PM10 of 10 and 20 µg/m3, respectively. The upper limits at ground level for NO2, O3, and SO2 are 40, 100, and 20 µg/m3, respectively [9]. Polycyclic aromatic hydrocarbons (PAHs) are identified as pollutants with a high carcinogenic activity [10], which is especially true for PAHs with higher molecular weights [11]. Even though PAHs, in particular benzo(a)pyrene (BaP) as their main representative, are linked to PM, the PAHs are not routinely monitored in Sarajevo [12]. In the previous studies conducted to evaluate PAHs concentrations, higher average mass concentration of total PAHs has been reported for Sarajevo, B&H, than for Zagreb, Croatia [12] or other European cities [13]. The average BaP concentration in Sarajevo is five times higher than EU target annual value and contributes to the PAHs carcinogenic potency [12,13]. According to WHO guidelines, air pollution is mainly caused by PM2.5, PM10, ozone, nitrogen dioxide (NO2), sulphur dioxide (SO2), and carbon monoxide (CO), with the confirmed effects on human health [9]. NO2, CO, and SO2 can cause various complications and irritations of the respiratory system with lethal outcomes after prolonged exposure to high concentrations [14]. Ground-level ozone formed in photochemical reaction of anthropogenic volatile organic compounds, often leads to shortness of breath. However, in patients with previous history of pulmonary diseases and children, long exposure can lead to asthma [15]. PMs and PAHs are also associated with various adverse health effects. PMs the most severely affect cardiovascular and respiratory systems [16], while PAHs are being associated with the lung cancer [17] Numerous studies have also demonstrated positive correlation between the exposure to PMs in the polluted air and mortality rate [14,18,19].
In the relation to COVID-19 pandemic, several review papers have addressed positive impacts of COVID-19 partial or complete lockdowns on the air quality [20]. In Sarajevo, complete lockdown was in force from mid-March till mid-April 2020, followed by the partial lockdowns in 2020 and 2021. However, no studies have been conducted to evaluate whether lockdown-related reduction in traffic, as a dominant source of air pollution in Sarajevo, effected changes in the air quality or biomarkers frequencies in model organism.
Human biomonitoring is an important strategy for the studies of the biological effects of air pollutants; however, the results are often confounded by the lifestyle of subjects or their metabolic alterations. In order to obtain more reliable data, scientists are looking for other plant or animal models. For example, lichens and mosses are frequently used in Europe and worldwide as a reliable tool in air pollution monitoring [21–23].
As sessile organisms, plants are constantly exposed to numerous environmental factors. Evaluation of the genotoxic effects of pollutants in plants is crucial as plants comprise a major portion of our global ecosystem and constitute a vital link in the food chain [24]. A growing awareness of the effects of air pollution on the environment has promoted frequent use of plants as genetic models for screening and monitoring of environmental mutagens [25–27]. Pollutants originating from the atmosphere and soil can induce DNA damage and different plants are utilized in ecotoxicology through the monitoring of genetic damage [28–30]. The comet assay in plants has proven to be a quick and reliable diagnostic tool for assessing DNA damage induced by various genotoxic impacts (radiation, chemicals, phytocompounds, heavy metals, nanoparticles, or complex contaminants [31]) including air and soil pollution [30,32–34].
Higher plants in urban areas, due to the roughness and large contact area of their foliage, absorb significant amount of dust and air pollutants and are therefore considered as potent filters with the beneficial impact on air quality and human health [35]. As plants are continuously exposed to the mixture of pollutants they are recognized as bioindicators of DNA damage caused by urban environmental stressors. Therefore, different parts of higher plants are widely used for air pollution biomonitoring [36].
Comet assay has been successfully applied in different organs of plants such as Ginko biloba L., Epipremnum aureum (Linden & André) G.S.Bunting, Vinca rosea (L.) G.Don [28], Arabidopsis thaliana (L.) Heynh. [36], Ricinus communis L. [37], Tradescantia pallida (Rose) D.R.Hunt [33], Citrullus colocynthis (L.) Schrad, Anabasis setifera L., Prosopis juliflora (Sw.) DC., Prosopis farcta (Banks & Sol.) J.F.Macbr., Suaeda vermiculata Forssk. ex. J.F.Gmel., Ziziphus spina-christi (L.)Desf., Capparis spinosa L., Salsola vermiculata L. [30]. However, Ligustrum vulgare L. has not been used as a plant model in comet assay although it is widely distributed hedge plant with green parts available almost throughout the year. Additionally, air pollution biomonitoring data for Sarajevo and B&H are scarce [38–40].
Regarding the air pollution and the biomonitoring studies, the indoor air quality (IAQ) should be supplementary considered. However, the IAQ is mainly affected by the outdoor pollutants [41].
According to the above mentioned, this study aimed to evaluate reliability of the L. vulgare as a plant-based air pollution monitoring model for comet assay by assessing air pollution effects at three different locations (two urban and one rural) varying in the levels of air pollution. Seasonal and sampling period differences, leaf position and locational (urban versus rural) differences in DNA damage were compared. COVID-19 lockdown and seasonal effects have been evaluated by periodical sampling of L. vulgare leaves at one of the urban sites, while in-house model of L. vulgare was established to test differences between outdoor and indoor air pollution effects for the same sampling period.
Material and methods
The sampling sites and sample collection
The study spanned the period 2020–2022. The two sampling sites in the urban area were selected based on previously established levels of air pollution [6]: Campus of the University of Sarajevo (~200-m distance from the main road) and Pofalići site, heavy traffic area (~10-m distance from the main road). Both sites are at 528 m above mean sea level; latitude 43° 51ʹ 15″N; longitude 18° 23ʹ 32″N. The local recreation area Barice (950 m above mean sea level; latitude 43° 53ʹ 25″N, longitude 18° 26ʹ 25″E) was selected to represent rural area for its remoteness and vertical position above smog threshold (~650 m distance from the side road). Initially, samples were collected at Campus site in March, July, and December of 2020 to evaluate L. vulgare as a biomonitoring model for comet assay and to test TI variations potentially related to complete and partial COVID-19 lockdown. For the additional comparison of COVID-19 lockdown and seasonal effects related to air pollution, the samples were collected at Campus site in March and December of 2021 and March of 2022. Pofalići and Barice sites were included in the study with the first sampling in December of 2021 (beginning of winter season) followed by sampling in March of 2022 (the end of winter season). Sampling period variations in TI at each of the examined sites were compared between samples collected in December and those collected in March.
Campus and Pofalići site are located 366 and 673 m (calculated with Google Earth distance measurement), respectively, from the position of continuous PM2.5 monitor BAM-1020 (Met One Instruments Inc., Washington, USA) used for the US Embassy in Sarajevo AirNow—air quality monitoring project. PM2.5 concentrations for the month prior to sampling for Campus and Pofalići sites were collected at https://www.airnow.gov/international/us-embassies-andconsulates/#Bosnia_and_Herzegovina$Sarajevo. The sample collection sites are presented at Fig. 1, while the average concentration of PM2.5 for Campus and Pofalići sites through the sampling periods are presented at Fig. 2.
Sample collection sites in Sarajevo, Bosnia and Herzegovina. The map was created in Google Earth (https://earth.google.com/).
The average PM2.5 concentrations for sampling sites Campus and Pofalići. Source: https://www.airnow.gov/international/us-embassies-andconsulates/#Bosnia_and_Herzegovina$Sarajevo.
The in-house model for this study was set up in the controlled environment (plant chamber) at 23–25°C, 6 weeks prior to the commencement of the study, and evaluated in December of 2021. The plants were exposed to the vegetative growth light (16-h light phase) to resemble external conditions. The in-house model was established at Campus site at the Institute for Genetic Engineering and Biotechnology in order to compare tail intensity (TI) values (% of DNA in the tail of comets) between the in-house and in situ plants.
Plant material
Wild privet, L. vulgare semi-evergreen hedge species from family Oleaceae was chosen as a plant model due to its ubiquitous distribution and genome size suitable for the comet assay. Wild privet, also known as common privet or Europe privet [42] is native to Europe, the temperate zones of East Asia and North Africa [43]. It has small (3–6 cm long) and slightly leathery leaves. It tolerates a wide spectrum of soil conditions, and it is also resistant to drought [44].
The plant comet assay was applied on fully developed (l: 3 ± 0.5 cm; w: 1.5 ± 0.2 cm), and young (l: 1.5 ± 0.5 cm; w: 0.8 ± 0.2 cm) leaves, positioned in the inner and outer layers of hedge. Five randomly selected leaves, with no observable symptoms, per sampling were collected. For each period and sampling condition, one sampling was conducted.
Isolation of nuclei
Samples were transported using a portable icebox in an opaque container and processed within an hour from sampling. At room temperature of 20°C, an individual leaf of L. vulgare was placed in 60-mm glass Petri dish with 500 µl of cold extraction buffer [9-ml phosphate-buffered saline (PBS); 1 ml 200 mM Na2EDTA). The leaf was vigorously chopped with cooled razor blade for 30 s [45]. The procedure releases nuclei into the buffer. Nuclei from simultaneously processed leaves were pooled together for further processing.
Slides
Plant comet assay was performed according to Pourrut et al. [45] protocol with minor modifications. Briefly, nuclei suspension (75 µl) was collected and thoroughly suspended in 50 µl low-melting-point agarose (2%, dissolved in PBS), and two drops (50 µl) of mixture were dropped separately on slides coated with 1% normal-melting-point agarose, and cover-slipped. The slides were then placed on a chilled surface at the room temperature of 20°C for a minimum of 10 min. After solidification of agarose, the coverslips were gently removed. Positive control slides were immersed in a solution of 70 µM H2O2 at 4°C for 5 min and washed with dH2O [46]. Electrophoresis tank was filled with freshly made cold electrophoresis buffer (10 M NaOH, 200 mM Na2EDTA, pH 13), and slides were introduced horizontally for 15 min of DNA unfolding. Thereafter, the electrophoresis was conducted for 5 min at 1 V/cm in the same solution. In order to avoid light-induced damage, the whole procedure was conducted under LED red light [47].
Scoring and analysis
After electrophoresis, the slides were placed into a 50 ml beaker with dH2O for 1 min, neutralized twice with 1 ml PBS for 5 min, and then fixed with EtOH (95%) by immersion. The slides were rehydrated with PBS and stained with DAPI (1 µg/ml) and in each slide, 50 randomly chosen cells were analysed under the fluorescence microscope (U-MNU2; Olympus BX51, Tokyo, Japan). The TI was evaluated for each sample using Comet Assay IV (Instem, UK) software [48].
Statistical analysis
Analysis of variance (ANOVA), followed by Scheffe post hoc analysis, was performed on log-transformed data to compare DNA damage in leaves according to the sampling site, position in a shrub, and staging of the leaves for each sampling seasons. Differences in DNA damage between seasons, outer versus inner, and young versus adult leaves were compared with t-test for independent samples. The same analysis was also used to compare results between in situ and in-house model in December 2021 for the Campus site (MedCalc 20.1.4). Linear regression analysis was conducted to test association between PM2.5 concentrations and TI values at Campus site.
Results
Means and standard deviations for DNA damage measured as TI in all the collected samples are presented in Table 1.
Log-transformed TI values (mean ± SD) in leaves of L. vulgare collected at three sites in Sarajevo area in March of 2020, 2021, 2022 and December 2021
| Stage of growth | Young | Adult | ||||
|---|---|---|---|---|---|---|
| Leaves position | In | Out | In | Out | ||
| Sampling period—March | ||||||
| Sample sites and periods | 2020 | Campusu | — | — | 0.79 ± 0.23 | 1.43 ± 0.19b |
| 2021 | Campusu | 0.74 ± 0.31 | 0.77 ± 0.38 | 1.28 ± 0.33c | 1.36 ± 0.29b,c | |
| 2022 | Campusu | 1.15 ± 0.48a | 1.55 ± 0.17a,b | 1.57 ± 0.28a | 1.57 ± 0.26a,c | |
| Pofalićiu | 0.74 ± 0.3 | 0.57 ± 0.34 | 1.17 ± 0.45c | 1.64 ± 0.37b,c | ||
| Baricer | — | 0.65 ± 0.3 | — | 0.56 ± 0.29d | ||
| Stage of growth | Young | Adult | ||||
|---|---|---|---|---|---|---|
| Leaves position | In | Out | In | Out | ||
| Sampling period—March | ||||||
| Sample sites and periods | 2020 | Campusu | — | — | 0.79 ± 0.23 | 1.43 ± 0.19b |
| 2021 | Campusu | 0.74 ± 0.31 | 0.77 ± 0.38 | 1.28 ± 0.33c | 1.36 ± 0.29b,c | |
| 2022 | Campusu | 1.15 ± 0.48a | 1.55 ± 0.17a,b | 1.57 ± 0.28a | 1.57 ± 0.26a,c | |
| Pofalićiu | 0.74 ± 0.3 | 0.57 ± 0.34 | 1.17 ± 0.45c | 1.64 ± 0.37b,c | ||
| Baricer | — | 0.65 ± 0.3 | — | 0.56 ± 0.29d | ||
| Sampling period—December | ||||||
|---|---|---|---|---|---|---|
| Sample sites and periods | 2021 | Campusu | 0.96 ± 0.46 | 0.8 ± 0.38 | 1.11 ± 0.49c | 1.27 ± 0.52b,c |
| Pofalićiu | 1.65 ± 0.14 | 1.87 ± 0.09b | 1.75 ± 0.18c | 1.87 ± 0.08b | ||
| Baricer | 0.70 ± 0.33d | 0.76 ± 0.4d | 0.62 ± 0.35d | 0.68 ± 0.34d | ||
| Sampling period—December | ||||||
|---|---|---|---|---|---|---|
| Sample sites and periods | 2021 | Campusu | 0.96 ± 0.46 | 0.8 ± 0.38 | 1.11 ± 0.49c | 1.27 ± 0.52b,c |
| Pofalićiu | 1.65 ± 0.14 | 1.87 ± 0.09b | 1.75 ± 0.18c | 1.87 ± 0.08b | ||
| Baricer | 0.70 ± 0.33d | 0.76 ± 0.4d | 0.62 ± 0.35d | 0.68 ± 0.34d | ||
leaves; Out, outer leaves; u, urban area; r, rural area.
Significantly increased compared to previous year (same sampling period).
Significantly increased compared to inner leaves (same stage of growth).
Significantly increased compared to young leaves (same position).
Significantly different compared to urban sites (same sampling period).
Log-transformed TI values (mean ± SD) in leaves of L. vulgare collected at three sites in Sarajevo area in March of 2020, 2021, 2022 and December 2021
| Stage of growth | Young | Adult | ||||
|---|---|---|---|---|---|---|
| Leaves position | In | Out | In | Out | ||
| Sampling period—March | ||||||
| Sample sites and periods | 2020 | Campusu | — | — | 0.79 ± 0.23 | 1.43 ± 0.19b |
| 2021 | Campusu | 0.74 ± 0.31 | 0.77 ± 0.38 | 1.28 ± 0.33c | 1.36 ± 0.29b,c | |
| 2022 | Campusu | 1.15 ± 0.48a | 1.55 ± 0.17a,b | 1.57 ± 0.28a | 1.57 ± 0.26a,c | |
| Pofalićiu | 0.74 ± 0.3 | 0.57 ± 0.34 | 1.17 ± 0.45c | 1.64 ± 0.37b,c | ||
| Baricer | — | 0.65 ± 0.3 | — | 0.56 ± 0.29d | ||
| Stage of growth | Young | Adult | ||||
|---|---|---|---|---|---|---|
| Leaves position | In | Out | In | Out | ||
| Sampling period—March | ||||||
| Sample sites and periods | 2020 | Campusu | — | — | 0.79 ± 0.23 | 1.43 ± 0.19b |
| 2021 | Campusu | 0.74 ± 0.31 | 0.77 ± 0.38 | 1.28 ± 0.33c | 1.36 ± 0.29b,c | |
| 2022 | Campusu | 1.15 ± 0.48a | 1.55 ± 0.17a,b | 1.57 ± 0.28a | 1.57 ± 0.26a,c | |
| Pofalićiu | 0.74 ± 0.3 | 0.57 ± 0.34 | 1.17 ± 0.45c | 1.64 ± 0.37b,c | ||
| Baricer | — | 0.65 ± 0.3 | — | 0.56 ± 0.29d | ||
| Sampling period—December | ||||||
|---|---|---|---|---|---|---|
| Sample sites and periods | 2021 | Campusu | 0.96 ± 0.46 | 0.8 ± 0.38 | 1.11 ± 0.49c | 1.27 ± 0.52b,c |
| Pofalićiu | 1.65 ± 0.14 | 1.87 ± 0.09b | 1.75 ± 0.18c | 1.87 ± 0.08b | ||
| Baricer | 0.70 ± 0.33d | 0.76 ± 0.4d | 0.62 ± 0.35d | 0.68 ± 0.34d | ||
| Sampling period—December | ||||||
|---|---|---|---|---|---|---|
| Sample sites and periods | 2021 | Campusu | 0.96 ± 0.46 | 0.8 ± 0.38 | 1.11 ± 0.49c | 1.27 ± 0.52b,c |
| Pofalićiu | 1.65 ± 0.14 | 1.87 ± 0.09b | 1.75 ± 0.18c | 1.87 ± 0.08b | ||
| Baricer | 0.70 ± 0.33d | 0.76 ± 0.4d | 0.62 ± 0.35d | 0.68 ± 0.34d | ||
leaves; Out, outer leaves; u, urban area; r, rural area.
Significantly increased compared to previous year (same sampling period).
Significantly increased compared to inner leaves (same stage of growth).
Significantly increased compared to young leaves (same position).
Significantly different compared to urban sites (same sampling period).
The analysis of variance revealed significant increase in TI in leaves collected in December of 2021 at urban localities, compared to Barice—rural site (P < 0.001) for all of the observed categories: adult outer leaves; adult inner leaves; young outer; and young inner leaves. The same was observed for March of 2022, except for outer young leaves (Table 1). Effects of COVID-19 lockdown effects were evaluated by comparison of TI values among samples collected in March, July, and December of 2020 at Campus site (Fig. 3).
The comparison of TI on site Campus throughout 2020: (a) significantly increased compared to previous months; (b) significantly increased compared to inner leaves (same stage of growth); (c) significantly increased compared to young leaves (same position).
Both inner and outer adult leaves had the highest TI values in December, with the release of lockdown measures (Table 1). TI values were lower in outer leaves collected in March 2020, before COVID-19 outbreak but still in heating, polluted season, then in July 2020. Comparison of three sampling periods (March of 2020, 2021, and 2022) at Campus site, for both inner and outer leaves, revealed significantly higher TI (P < 0.001) for March of 2022 (Fig. 4). TI in samples collected in March of 2021, with the partial lockdown still in force, where significantly lower (P < 0.001) than in March of 2022, but not when compared with pre-covid sampling in March of 2020.
The comparison of TI on site Campus in March 2020/2021/2022; (a) significantly different compared to previous years; (b) significantly increased compared to inner leaves (same stage of growth); (c) significantly increased compared to young leaves (same position).
Linear regression analysis revealed positive association between average TI values and average PM2.5 concentration for the month before sampling at Campus site. The association was significant only for inner young leaves. Data for the other sites were not sufficient because small number of sampling periods.
The results showed significantly higher DNA damage in adult and outer leaves (P < 0.001) for Campus and Pofalići sites, except for inner young leaves that were significantly higher at Campus site (P = 0.0007) in December 2021 and at Pofalići site in March 2022 (P = 0.003). In March 2022, no significant differences were observed between adult inner and adult outer leaves, while no differences were found between outer adult and young leaves at Pofalići site in December 2021. Rural Barice site showed no differences in DNA damage in regards to either the position of leaves or the stage of growth.
The highest DNA damage observed in samples from Pofalići area in December of 2021 corresponds to the intensive traffic in the proximity of the collection point.
Comparison between the March and December revealed significantly higher DNA damage for December of 2020 and 2021 at Campus site (P < 0.001), except for outer adult leaves that showed higher TI in March then in December of 2021. March 2022 showed significantly higher TI when compared with December 2021. For the Pofalići site, all the categories of leaves had higher TI in December than in March. At Barice site, outer adult and outer young leaves had higher DNA damage in December, but only differences for outer adult leaves were significant (P = 0.03).
The in-house model of L. vulgare was established in winter season (November of 2021) and sampled after 6 weeks of growth (in December of 2021). The leaves from the in-house model were differentiated neither according to the stage of growth nor according to the position. TI for the in-house model was 0.62 ± 0.28 and significantly lower from TI in leaves collected in situ at Campus site in December 2021.
Discussion
The temperature inversion in Sarajevo is a phenomenon characteristic for winter season when it turns Sarajevo into one of the most polluted cities in continental Europe. Decline in green areas because of unplanned urbanization, coal-based heating, and low criteria for vehicles pollutants emission contribute to high air pollution. In Sarajevo, wood and coal PAHs were more evident, and total carcinogenic potency of PAHs, with the dominance of BaP, was significantly higher compared to Zagreb [12].
The presence of mutagens in the air was evaluated worldwide by short-term mutagenicity tests with bacteria, human, and plant cells. Genotoxic potential of organic air pollutants was analysed using bacterial models in Sarajevo and Tuzla, two main industrial cities in B&H. High pollution was detected in both cities with higher pollution, genotoxicity, and potential health hazard found in Tuzla, which features much stronger industrial activity than Sarajevo [49]. Lead pollution in soil was analyses by determination of lead content in Cichorium intybus L. in urban, suburban, and rural sites of Sarajevo area. The highest lead content was revealed in the soil from Pofalići site [38]. Epiphytic lichen Hypogymnia physodes (L.) Nyl. was used to monitor air pollution in Sarajevo [39]. Heavy metals content (Cd, Cr, Cu, Fe, Mn, Ni, Pb, and Zn) in lichen tissue was determined at 44 sites in B&H (urban and rural) in 3-month intervals over 1 year, but significant differences between urban and rural areas were not found [39]. Recent human biomonitoring study in Sarajevo revealed significantly higher air pollution induced DNA damage in winter using Buccal Micronucleus Cytome (BMNCyt) but the results were not supported by comet assay [40]. The latter report supported the effects of known, although lesser, air pollution in Sarajevo during spring–summer season, which is only less obvious than in winter because of the absence of temperature inversion [5,50].
Genetic damage caused by genotoxic substances from urban air in North Italy in winter was registered with all applied tests, including comet assay [51].
In our study, the leaves of L. vulgare were collected from plants growing in urban and rural areas in Sarajevo, B&H. The hedge leaves are the most affected when the wet and dry atmospheric deposition increase, and have an essential role in retention of the PMs [52]. DNA damage in leaves was compared between seasons and according to the stage of growth and position of the collected leaves. Significantly increased TI was observed in leaves collected in December at Pofalići site, while this was not the case for all comparisons at Campus site. Additionally, outer and adult leaves, in general, showed higher values of TI. The lack of differences between adult and young leaves and inner and outer leaves at Barice site indicates better air quality, as expected, although the monitoring data were not available.
Sporadically increased TI in young leaves may be related to the dynamics of leaf development. In dicotyledones, cell division and cell expansion occur simultaneously for the first half of the leaf development. Cell cycle is arrested earlier in the epidermis than in the mesophyll [53].
Sensitivity and specificity of budding leaves in air pollution monitoring should be further evaluated, while adult leaves of L. vulgare present reliable sample for plant comet assay. Because of the limited material, staging of leaves was not considered for L. vulgare in-house grown model that showed lower DNA damage than the closest in situ site—Campus. Although IAQ is mainly affected by the outdoor pollutants [41], this finding demonstrates that differences are still notable.
The results from the study of Rai et al. [54] showed remarkable differences in micromorphological features and growth parameters of leaves in dust-treated plants compared to control plants. Al Khateeb [30] also reported great variation in comet assay parameters among leaf samples from different sites with no observable differences among plants species collected at the same collection site in Jordan. No significant difference was observed in TI from Ricinus communis L. leaves from five different sites with high pollution levels [37]. Hattab et al. [55] reported that leaves are less sensitive than roots, while Reis et al. [56] demonstrated concordance in DNA damage between the results obtained from cells extracted from the root tips of Lactuca sativa L. and DNA damage in human leukocytes.
The reported findings are in line with the study of Sriussadaporn et al. [28] who also applied comet assay on leaf samples for urban air quality assessment. Using Euonymus japonicus Thunb. as a model for air pollution monitoring in Beijing, Li et al. [29] revealed positive correlation between roadside pollution and AQI (air quality index) values.
When association between available pollutants concentrations (PM2.5) and DNA damage data were compared, positive association was found. However, availability of additional air quality data would contribute to this analysis.
With regards to the concentration of PM10 particles in the period relevant for this study, Huremovic et al. [6] reported that the mean PM10 concentrations were above the defined WHO and local limits during the home heating seasons in Sarajevo (2010–2019), except in 2010 and 2020 when they were lower. Lower PM10 concentrations may correspond to the lower TI values that we evidenced in March 2020. Masic et al [50]. observed that during the period of strong urban pollution (2 December 2019–12 March 2020), the average value of PM2.5 and PM10 concentrations were82.9 and 95.5 µg/m3, respectively. Concentrations of PM2.5 and PM10 particles measured in the period of mild pollution (13 March 2020–4 May 2020) were lower, with the average concentration of PM2.5 and PM10 of 19.7 and 24.2 µg/m3, respectively. Lepers et al. [57] showed that PM samples collected in heating period are more genotoxic compared to spring–summer sample, likely due to the higher content of adsorbed organic compounds. Significant positive correlation was found for DNA damage in Nicotiana tabacum L. with the NO and PM10 concentrations [58].
Meteorological and environmental conditions, such as wind, rainfall duration and intensity, can largely affect the annual trend of PM accumulation in the leaves [59,60], therefore they should also be considered when applying plant comet assay for air pollution biomonitoring. Additionally, as recommended by Villarini et al. [58] in situ biomonitoring of airborne genotoxins using higher plants should be combined with chemical characterization of urban air.
COVID-19 lockdown effects were estimated by comparison of different sampling periods in 2020 and samplings in March of three different years. Significant increase in TI was observed for the March of 2021, with the partial lockdown still in force, than in March of 2022, but not when compared to pre-covid sampling in March of 2020. Significant increase in TI was registered in July 2020, after the lockdown loosening, then in the periods before COVID-19 outbreak or after lockdown in 2020. Although lower TI values in July are expected due the milder air pollution in summer, TI values were unexpectedly lower in outer leaves collected in March 2020, before COVID-19 outbreak and still in heating, polluted season, then in July 2020. Comparison of three sampling periods (March of 2020, 2021 and 2022) at Campus site, for both inner and outer leaves, revealed significantly higher TI (P < 0.001) for March of 2022. However, TI in samples collected in March of 2021, with the partial lockdown still in force, where significantly lower (P < 0.001) then in March of 2022, but not when compared to pre-covid sampling in March of 2020.
Our findings are in accordance with the systematic review of the impact of COVID-19 on air quality [61] that revealed general improvement in the air quality during the lockdown period compared to pre- and post-COVID period, especially in regards to the reduction of NO2 and PMs. That also shows that L. vulgare is good indicator of shifts in air pollution.
Despite the certain limitations of the study, related to the sampling replications, the lack of the air quality data from the rural site and limited air quality data for urban sites, this study addresses various valuable aspects of the air pollution biomonitoring using plants as biomonitors.
Conclusion
The ongoing concerns about air pollution as the world’s top problem emphasize the efforts to establish and continuously monitor biological effects of air pollutants. Ligustrum vulgare is a reliable plant model for the comet assay in biomonitoring studies. As we showed, DNA damage in leaves of L. vulgare depends on the sampling period, leaves position, and stage of growth. This model is sensitive to seasonal variations in air pollution levels, as DNA damage in L. vulgare leaves corresponds to the average PM2.5 levels, and indoor versus outdoor conditions. DNA damage in L. vulgare leaves collected in March of three successive years are well aligned with the COVID-19 lockdown periods.
Funding
This work was supported by the Environmental Protection Fund of the Federation of Bosnia and Herzegovina (grant no: 01-09-2-1469/2021).
Conflict of interest statement: None declared.
