Atmospheric Pollution Research xxx (2017) 1e12

H O S T E D BY Contents lists available at ScienceDirect

Atmospheric Pollution Research

journal homepage: http://www.journals.elsevier.com/locate/apr

Responses of Trichilia dregeana leaves to sulphur dioxide pollution: A

comparison of morphological, physiological and biochemical
biomarkers
Minoli Appalasamy a, Boby Varghese a, Riyad Ismail b, Sershen a, *
a
Plant Germplasm Conservation Research, School of Life Sciences, University of KwaZulu-Natal, Durban, South Africa
b
Department of Geography and Environmental Sciences, University of KwaZulu-Natal, Pietermaritzburg, South Africa

a r t i c l e i n f o a b s t r a c t

Article history: Industrial zones in eThekwini, South Africa, such as the South Durban Basin (SDB) are often characterised
Received 19 September 2016 by extremely poor air quality owing to industrial emissions. This study investigated the effects of SO2
Received in revised form pollution on selected stress biomarkers in the leaves of Trichilia dregeana to assess the utility of this
13 January 2017
species as a bioindicator of this pollutant. Leaves were sampled from trees growing at three industrial
Accepted 13 January 2017
Available online xxx
(treatment) sites (Prospecton, Ganges and Southern Works) within the SDB and from greenhouse-located
trees (ex situ control). Sampling accommodated for directional and seasonal wind effects and yielded a
sample size of n 24 for all four seasons. Ground-level SO2 concentrations ([SO2]) measured at each site
Keywords:
Air quality
were positively correlated with leaf sulphate contents and both [SO2] and leaf sulphate levels of the
Bioindicator control were lower than the treatment sites. Values for the various biomarkers did not differ signicantly
Industrial for leaves from different cardinal points within sites, but seasonal variations were evident. Except for leaf
Stress chlorophyll uorescence, all biomarkers could discriminate between treatment and control leaves.
Tree leaves Though seasonal data for a number of these biomarkers were signicantly correlated with leaf sulphate
content, none of the biomarkers were sensitive enough to reect differences in leaf sulphate levels across
the treatment sites. Furthermore, the results suggest that leaf area and relative chlorophyll content
should be measured in combination with each other. Leaf sulphate content is a reliable proxy for SO2
pollution. All biomarkers, except for chlorophyll uorescence, can be used to establish T. dregeana leaves
as a bioindicator of SO2 pollution.
2017 Turkish National Committee for Air Pollution Research and Control. Production and hosting by
Elsevier B.V. All rights reserved.

1. Introduction
Since the beginning of the industrial era, the greenhouse effect
has intensied due to the increase in atmospheric greenhouse gas
emissions from anthropogenic activities and the growing demand
for energy produced via the use of fossil fuels such as coal and oil
(Jain and Hayhoe, 2008). South Africa depends on the energy sector
which contributes to 15% of the country's GDP (Department of
Energy and Minerals, South Africa, 2008); 70% of the total pri-
mary energy comes from coal, and 93% of electricity from coal-red
stations (Menyah and Wolde-Rufael, 2010). Coal sectors are there-
fore the major contributors to atmospheric pollution in South
* Corresponding author.
E-mail address: naidoose@ukzn.ac.za (Sershen).
Peer review under responsibility of Turkish National Committee for Air Pollu-
tion Research and Control.
Africa, accounting for 96% of all atmospheric sulphur dioxide (SO2),
94% of all atmospheric nitrous oxides (NOx), and 87% of all atmo-
spheric carbon dioxide (CO2) (Menyah and Wolde-Rufael, 2010).
Globally, industries are the main contributors of air pollution
(Lamego et al., 2000) and consume the most energy. Along the east
coast of South Africa industrial zones such as the South Durban
Basin (SDB) are characterised by extremely poor air quality as a
result of industrial emissions (Matooane and Diab, 2001; Diab et al.,
2002). Atmospheric SO2 concentrations have been reported to be
exceptionally high (15.28 ppb) in this area and have been used as an
indicator of air quality within the SDB, as SO2 contributes to two
thirds of industrial emissions in the area (Diab and Motha, 2007).
Those authors also emphasised the need for air quality monitoring
in the SDB, given that SO2 levels are likely to rise with the expan-
sion of various industrial activities in the area.
In many parts of the world air quality data are usually based on

http://dx.doi.org/10.1016/j.apr.2017.01.005
1309-1042/ 2017 Turkish National Committee for Air Pollution Research and Control. Production and hosting by Elsevier B.V. All rights reserved.

Please cite this article in press as: Appalasamy, M., et al., Responses of Trichilia dregeana leaves to sulphur dioxide pollution: A comparison of
morphological, physiological and biochemical biomarkers, Atmospheric Pollution Research (2017), http://dx.doi.org/10.1016/j.apr.2017.01.005
2 M. Appalasamy et al. / Atmospheric Pollution Research xxx (2017) 1e12
one or a few monitoring stations at strategic sites across cities,
making average pollution levels unrepresentative of specic regions
(Moreno et al., 2003). This is often unavoidable since the installation
of multiple air quality monitoring stations is not always nancially
feasible, particularly in countries within the developing world. An
alternate approach is to supplement instrumental air quality moni-
toring with data collected using bioindicators, i.e. organisms that can
be used to identify and determine the effect that biotic and abiotic
stresses have on the environment (Conti and Cechetti, 2001). In this
regard, Novak et al. (2003) argued that it is best to use indigenous
species as bioindicators of air pollution, provided these species are
able to withstand high pollution levels, have a wide geographical
distribution, are abundant and easily accessible, and are negatively
impacted on by the pollution (Conti and Cechetti, 2001).
The use of plants and trees in particular, as bioindicators of air
pollution is a widespread strategy (Madejon et al., 2004) and the
identication of reliable bioindicators of air quality may be useful in
many countries across the developing world (Areington et al., 2015).
Several authors report the value of using morphological and
biochemical characteristics of leaves as biomarkers of environmental
pollution (Bakker et al., 1999 and other studies reviewed in Madejon
et al., 2004). This motivated the present study, which investigated the
utility of leaves of the indigenous tree species, Trichilia dregeana as a
bioindicator of SO2 pollution within selected industrial areas in
Durban (in eThekwini, KwaZulu-Natal), South Africa.
Vegetation represents passive sampling in biomonitoring and
has the advantage of high spatial and temporal resolution due to
widespread distribution of plants and low sampling costs (Sawidis
et al., 1997). One of the prerequisites of using tree leaves as bio-
indicators is that they have to be susceptible to the pollutants
investigated, but be tolerant enough to avoid mortality (Madejon
et al., 2004). Literature suggests that one can measure a number
of physiological, biochemical and morphological biomarkers of air
pollution in tree leaves (Sawidis et al., 2011). For instance, as a
consequence of air pollution and other environmental stressors,
plants increase the production of reactive oxygen species (ROS) in
their tissues (Vranova et al., 2002). The stress experienced by the
plants and damage incurred is determined by the equilibrium be-
tween oxidative stress and antioxidant activity (Gill and Tuteja,
2010). During non-stressful periods, antioxidant defence systems
are able to protect the organism from ROS (Gill and Tuteja, 2010);
however, excessive production of ROS can lead to damage such as
lipid peroxidation, protein oxidation and oxidative damage to
nucleic acids. Lipid peroxidation damages plant cell membranes,
the integrity of which can be measured in terms of leaf electrolyte
leakage (Santamaria and Martin, 1997). Other parameters such as
chlorophyll content and leaf area have also proven to be useful
biomarkers of pollution stress (Areington et al., 2015). Leaf chlo-
rophyll uorescence, though rarely applied in pollution research, is
a non-destructive measure of the efciency of photosystem II (PSII)
that can be used as an indicator of stress-induced damage in a
number of plant species (Maxwell and Johnson, 2000).
In light of the above, the aim of the present study was to asses
the utility of morphological (leaf area), physiological (relative leaf
chlorophyll content and uorescence) and biochemical (hydrogen
peroxide production, total aqueous antioxidants activity and elec-
trolyte leakage) parameters as biomarkers of SO2 pollution in
T. dregeana tree leaves.
2. Materials and methods
2.1. Study area: polluted (treatment) sites, ex situ control site,
location of trees and geography of area
The geographic location of the three SDB industrial treatment
Please cite this article in press as: Appalasamy, M., et al., Responses of Trichilia dregeana leaves to sulphur dioxide pollution: A comparison of
morphological, physiological and biochemical biomarkers, Atmospheric Pollution Research (2017), http://dx.doi.org/10.1016/j.apr.2017.01.005
sites, viz. Southern Works, Ganges High School; and Prospecton, is
shown in Fig. 1. At each site, four Trichilia dregeana Sond. trees were
selected for investigation, within 750 m of an air pollution moni-
toring station (the locations of which are also indicated in Fig. 1).
The selection of T. dregeana leaves is based on the fact that this
evergreen coastal forest tree grows naturally and is planted as a
street tree in Durban and the province of KwaZulu-Natal as a whole
(Pooley, 1993). Sampling sites were selected on the basis of the
presence of air pollution monitoring stations (owned and operated
by the eThekwini Municipality) that measured ground-level SO2
levels on an hourly basis. Trees housed in a greenhouse on the
grounds of the University of KwaZulu-Natal (Westville, Durban), a
month prior to and for the duration of the study, served as the ex
situ control. Recent studies (Areington et al., 2015; Appalasamy
et al., 2016) have shown the utility of greenhouse-located control
trees in studies of this nature, provided that the control and
treatment trees are exposed to comparable light levels, which was
ensured here as discussed below.
2.1.1. Measurement of ground-level SO2 concentrations ([SO2])
Ground-level [SO2] (in parts per billion [ppb]) were measured
hourly (by the eThekwini Municipality) at each of the monitoring
stations (Fig. 1) and the data was made available for use in this
study. The data used were collected in each of the four sampling
seasons, between 2014 and 2015. Prior to any analyses, the data
were cleaned (i.e. erroneous, zero, negative and blank values were
deleted) and a yearly average [SO2] and seasonal averages were
generated for each monitoring station. The air pollution monitoring
stations that provided the data, measure the ground-level con-
centrations of SO2, NO2 and PM (amongst others) via continuous
monitoring (using United States Environmental Protection Agency
(USEPA) designated analysers, North Carolina). Sulphur dioxide
levels were also measured within the greenhouse and at three
random points (on the university campus) less than 1 km from
where the greenhouse control site is located. These measurements
were carried out using a HORIBA PG-350E (HORIBA Europe GmbH,
Julius-Kronenberg-Strasse 9, Leichlingen, Germany) atmospheric
monitoring system. The SO2 detector functions with a cross-ow
modulation, non-dispersive infrared (NDIR) absorption method
(according to the European standard, DIN EN 15267-3, DIN EN
14181).
Data for NO2 and PM were only available for two sites, viz.
Ganges and Southern Works, and furthermore only collected at the
monitoring stations in spring and summer. Nevertheless, this
limited dataset was also analysed here in the interest of gaining
some insight into the levels of pollutants other than SO2 at the
treatment sites. The NO2 and PM results were not related to the
biomarkers directly but rather used to interpret some of the trends
observed at specic treatment sites.
2.1.2. Leaf sulphate content
Leaf material from three treatment sites and the control were
tested for sulphate content according to Lau and Luk (2001). Leaf
material was dried at 80  C for 48 h, weighed to l.0 g dry weight
(DW) for each of three replicates, and placed into a furnace at
500  C for 2 h. The ash produced was moistened with 1e2 drops of
distilled water, and 2 ml of 1:1 nitric acid was added. Subsequent to
evaporation on a hot plate, the residue was heated at 500  C for 1 h,
cooled and dissolved in 2 ml of warm 1:1 hydrochloric acid. The
solution was then ltered and diluted in a ask to reach a nal
volume of 25 ml (Lau and Ho, 1993). The sulphate concentration in
the solution was thereafter determined for each sample using a
tubidimetric method (American Public Health Association, 1992).
 M. Appalasamy et al. / Atmospheric Pollution Research xxx (2017) 1e12 3

Fig. 1. Study area showing industrial sites (Ganges, Southern Works and Prospecton) at which sampling was conducted. All sites are located within the South Durban Basin in
eThekwini, South Africa.

2.2. Leaf sampling design
Four trees at each of the three treatment sites and the control
were sampled as per Sawidis et al. (2011), with slight modications.
On any sampling day, which was always at least 24 h after the last
rain event, branches were removed from selected trees at all the
sites; leaves which showed signs of recent insect damage were not
selected. Trees were sampled randomly across all sites and in each
of the four seasons: autumn (1 March to 31 May), winter (1 June to
31 August), spring (1 September to 30 November) and summer (1
December to 28 February), between 2014 and 2015. On any
particular sampling day, a maximum of two trees were sampled at
each site (treatment and control). No individual tree was sampled
more than once a week, and within any particular season the trees
were sampled until the required number of replicates (n 24) was
achieved for all biomarkers. A random branch located at each car-
dinal point (north, east, south and west) was collected to accom-
modate for the effects of wind direction (Klumpp et al., 2003).
Control trees were subjected to the same sampling regime. The
time period between sampling and measurements was no longer

Please cite this article in press as: Appalasamy, M., et al., Responses of Trichilia dregeana leaves to sulphur dioxide pollution: A comparison of
morphological, physiological and biochemical biomarkers, Atmospheric Pollution Research (2017), http://dx.doi.org/10.1016/j.apr.2017.01.005
4 M. Appalasamy et al. / Atmospheric Pollution Research xxx (2017) 1e12
than 2 h. Upon arrival at the laboratory, the fresh weight (FW) of
individual leaves was recorded prior to any biochemical assays. For
physiological and morphological parameters, FW was recorded
after measurements, when the leaves were detached from the
branch. Given that the trees used in this study were street trees, it
was impossible to control for tree age. In light of this all parameters
were measured on leaves that were of a comparable age, in this case
the 3rd leaf from the top on side branches (for treatment and
control trees). This was based on the results of a pilot study, which
using methods prescribed by Chen et al. (2009), showed that the
3rd leaf from the top on side branches of T. dregeana trees to be
mature and in the asymptotic phase of growth.
2.2.1. Ex situ control and validation studies
When [SO2] was measured at the ex situ control site, levels
within the greenhouse in which the control trees were housed
were below the detectable limits of the instrument. The green-
house is constructed from clear 5 mm thick polycarbonate sheeting
with a light transmittance of 90% and this ensured that photo-
synthetically active radiation in the greenhouse
(1938.65 100.28 mm of photons m2 s1) was comparable with
that at the treatment sites (2027.65 124.24 mm of photons m2
s1). Light intensity was measured using a portable photosynthesis
system (Li-6400, Li-COR, Lincoln, Nebraska, USA) at midday on four
clear sunny days within the greenhouse and at the treatment sites.
Furthermore, to rule out the possibility of slight differences in light
intensity having a confounding effect on light-dependent bio-
markers, relative chlorophyll content and leaf area were also
measured for four naturally growing trees located within 1 km of
the greenhouse in which the control trees were housed. Relative
chlorophyll content and leaf area data were in turn related to the
SO2 measurements carried out on the university campus within
1 km of the greenhouse (as described above) to validate the trends
observed for these biomarkers. This validation process has been
used successfully in other bioindicator studies involving
greenhouse-located control trees (Areington et al., 2015;
Appalasamy et al., 2016). Additionally, leaf photosynthetic rates
(measured according to Naidoo, 2009) were comparable between
trees located within (7.86 2.64 mmol m2 s1) and outside
(8.23 2.45 mmol m2 s1) the greenhouse in which the control
trees were located. Leaf photosynthetic rates are extremely sensi-
tive to climatic changes and these data suggest that greenhouse
connement had limited effects on T. dregeana leaves.
2.3. Biomarker studies
All parameters described below were measured for 24 leaves,
per site, per season. For the physiological and morphological
studies, leaves remained attached to the branch. For biochemical
studies, all ne chemicals were supplied by Sigma-Aldrich (Ger-
many) unless otherwise stated.
2.3.1. Leaf area
Leaf area was measured for individual using a Leaf Area meter
(Licor CI-202 Area Meter, Lincoln, Nebraska, USA) as described by
Tiwari et al. (2006). Three measurements were taken for each leaf
and these were averaged for each leaf before further data analyses.
The values were expressed in cm2.
2.3.2. Relative chlorophyll content
Relative leaf chlorophyll content was measured for individual
leaves using a handheld Konica Minolta SPAD-502 Plus (Konica
Minolta Inc., Japan) as described by Lubbe and Rodda (2016).
Readings were taken at three random points for each leaf and these
were averaged for each leaf before further data analyses. The values
Please cite this article in press as: Appalasamy, M., et al., Responses of Trichilia dregeana leaves to sulphur dioxide pollution: A comparison of
morphological, physiological and biochemical biomarkers, Atmospheric Pollution Research (2017), http://dx.doi.org/10.1016/j.apr.2017.01.005
were expressed in SPAD units.
2.3.3. Leaf chlorophyll uorescence
Leaf chlorophyll uorescence was measured on individual
leaves, which were dark-adapted according to Netto et al. (2005),
prior to measurements. These were taken using a Leaf Chamber
Fluorometer (Li-Cor 6400-40, XT model, Li-Cor, Lincoln, Neb.), as
described by Pitterman and Sage (2000). The values were
expressed as Fv/Fm.
2.3.4. Intracellular hydrogen peroxide
Intracellular hydrogen peroxide (H2O2) levels were measured
according to Jana and Choudhuri (1981) with adjustments based on
Hung et al. (2008). Leaves were ground using LN in a pre-chilled
mortar and pestle, with 0.1 g of insoluble PVP and 4 ml of chilled
50 mM sodium phosphate buffer (pH 6.5) with 1 mM hydroxyl-
amine (to inhibit enzyme catalase). The homogenate was centri-
fuged at 4400 rpm for 30 min, after which the supernatant was
mixed with 0.1% titanium [III] chloride in 20% sulphuric acid. After
15 min incubation in the dark, the mixture was centrifuged at
4400 rpm for 30 min before the absorbance was read at 410 nm,
using a UV-2600 UV-VIS Spectrophotometer (Shimadzu, Japan).
The amount of intracellular H2O2 produced (expressed as mmol g1
FW) was calculated using the extinction coefcient
0.28 mmol1cm1.
2.3.5. Electrolyte leakage
Electrical leakage was measured according to Santamaria and
Martin (1997). Fresh leaves were cut into 1 cm2 discs, weighing
approximately 0.1 g. Leaf discs were then placed into test tubes
containing 20 ml of distilled water and then incubated in a water
bath (Grant, Duxford Cambridge, England) held at 30  C for 2 h.
Thereafter, 1.5 ml of leachate was pipetted into cells of a conduc-
tivity tray and electrolyte leakage was measured using a multi-cell
electrical conductivity meter (CM 100-2 Conductivity Meter, Reid
and Associates, South Africa). Electrolyte leakage was expressed as
S m1 g1 FW.
2.3.6. Total aqueous antioxidant activity
Total aqueous antioxidant activity (TAA) was measured ac-
cording to Re et al. (1999). Leaves were ground using LN in a pre-
chilled mortar and pestle, with 0.1 g of insoluble PVP and 4 ml of
chilled extraction buffer (50 mM sodium potassium phosphate
buffer containing 1 mM CaCl2, 1 mM KCl and 1 mM EDTA, pH 7.0).
Prior to centrifugation for 30 min at 4400 rpm, the homogenate
was transferred into 15 ml tubes and vortexed for 15 min at 5 min
intervals for extraction of total antioxidants. The resulting super-
natant was placed on ice for the TAA assay. Before the estimation of
TAA, the extract was diluted (1:1) to enable estimation in the
absorbance range recommended by Re et al. (1999). Total aqueous
antioxidant activity was measured in terms of the depletion of 2,2-
azino-bis-3-ethylbenzthiazoline-6-sulphonic acid (ABTS) radical, in
the presence of the antioxidant extract. Decolourisation of the
working solution (800 ml of ABTS dissolved in 50 ml of potassium
phosphate buffer) was measured at 0 s and 120 s after the addition
of 10 ml of the antioxidant extract. A standard curve was con-
structed using 0.1e1.5 M ()-6-hydroxy-2,5,7,8-
tetramethylchromane-2-carboxylic acid (Trolox, Sigma-Aldrich),
dissolved in extraction buffer, and used to express TAA in terms
of Trolox equivalents g1 FW.
2.4. Statistical analysis
The data were analysed using Predictive Analytics Software
(PASW) version 21 (SPSS Inc., Chicago, IL). All data were tested for
 M. Appalasamy et al. / Atmospheric Pollution Research xxx (2017) 1e12 5

Fig. 2. Ground-level SO2 concentration (ppb) and leaf sulphate contents (mg/g) measured at three industrial (treatment) sites within the SDB and the control: (A) Mean seasonal
[SO2]; (B) Mean seasonal leaf sulphate content; (C) Mean annual [SO2] and leaf sulphate content. Values represent means (n 725e1752 for seasonal SO2 data and n 3982e4962
for annual SO2 data). Bars labelled with different letters are signicantly different (p < 0.001, ANOVA) when compared across treatment sites and the control: lowercase for seasonal
SO2 data and uppercase for annual SO2 data; lowercase for both seasonal and annual leaf sulphate content. Standard deviations for [SO2] ranged from 2.24 to 6.65 for seasonal data
and from 4.09 to 5.29 for annual data, for leaf sulphate content those ranged from 0.001 to 0.008 for seasonal data and from 0.001 to 0.004 for annual data.

normality using the Shapiro-Wilks/Kolmogorov-Smirnov test.
There were no signicant differences across cardinal points for all
sites and seasons (p > 0.05, Analysis of Variance [ANOVA]) so data
for the different cardinal points were pooled within each site for
the subsequent analyses. Analyses of Variance were used to test for
signicant differences in all biomarkers at the following levels:
season main effects (comparisons within sites across seasons); site
main effects (comparisons within seasons across sites); interaction
between site and season (labelled site  season, henceforth); and
across sites with data for different seasons pooled (labelled annual
data, henceforth). Analyses of Variance was also used to test for
differences in [SO2] within seasons, across sites and across sites,
with seasonal data pooled (referred to as annual data henceforth).
Data for NO2 and PM were compared within spring and summer,
between Southern Works and Ganges only as data for other seasons
and sites were unavailable. Means were thereafter separated using
a Bonferroni post-hoc test at the 0.05 level of signicance. A Pear-
son's correlation test was used to determine whether there was a
relationship between average seasonal SO2 concentrations and leaf
sulphate content, and seasonal averages for individual biomarkers
and leaf sulphate content.
3. Results
3.1. Ground-level sulphur dioxide concentrations
Ground-level [SO2] differed signicantly (p < 0.001) when
compared across treatment sites, within seasons. Sulphur dioxide
levels for Southern Works were signicantly higher than Ganges
and Prospecton in autumn, spring and summer. Levels of SO2 at
Ganges were signicantly higher than Prospecton (Fig. 2A)
excluding spring, which showed levels between these two sites to
be comparable. In winter, SO2 levels were signicantly higher at
Ganges, followed by Prospecton and Southern Works. However,
given the wide range in [SO2] recorded (1e25 ppb), and the lack of
data for some days within specic seasons, seasonal data were
pooled to generate an annual average for individual sites (Fig. 2C).
Comparisons based on these averages revealed signicant differ-
ences (p < 0.001) across sites: [SO2] at Southern Works were
signicantly higher than Ganges and Prospecton, both of which
exhibited statistically comparable SO2 levels. When [SO2] was
measured at the ex situ control site, levels were below the detect-
able limits of the instrument within the greenhouse, while levels at
three random points within 1 km of the greenhouse (located on a
university campus), averaged 2.73 0.56 ppb. This meant that
T. dregeana trees growing outside the greenhouse on the university
campus were exposed to signicantly lower (p < 0.05, ANOVA) SO2
levels than the treatment sites, but higher than the control.
Additionally, from the limited NO2 and PM data available at the
time of this study it was evident that the levels of both these pol-
lutants at Ganges were signicantly higher than those at Southern
Works in summer and spring (Table 1). Data for NO2 ranged from
10.46 to 17.23 ppb in spring and 8.48e16.13 ppb in summer, with a
signicant difference between Ganges and Southern Works
(Table 1). Data for PM ranged from 11.07 to 17.12 mg/m3 in spring to
9.30e13.88 mg/m3 in summer, but unlike NO2, there were no

Please cite this article in press as: Appalasamy, M., et al., Responses of Trichilia dregeana leaves to sulphur dioxide pollution: A comparison of
morphological, physiological and biochemical biomarkers, Atmospheric Pollution Research (2017), http://dx.doi.org/10.1016/j.apr.2017.01.005
6 M. Appalasamy et al. / Atmospheric Pollution Research xxx (2017) 1e12
Table 1
Spring and summer ground-level nitrogen dioxide (NO2) and particulate matter (PM) concentrations at two industrial (treatment) sites at which sampling was conducted
between September 2014 and February 2015.
Spring NO2 (ppb)
PM (mg/m3)
Summer NO2 (ppb)
PM (mg/m3)
Values represent mean SD (n 4313 for PM and 8391 for NO2 for spring, and n 4274 for PM and 6527 for NO2 in summer); values followed by different letters are
signicantly different (p < 0.001, ANOVA) when compared within pollutants, between sites.
signicant differences between treatment sites (Table 1).
3.2. Leaf sulphate content
Leaf sulphate contents were higher than the control for all
treatment sites, irrespective of the season; these differences were
signicant for Southern Works and Ganges only (p < 0.001; Fig. 2B).
There was a trend for leaf sulphate contents at Southern Works and
Ganges to be relatively higher than Prospecton, but these differ-
ences were not signicant. When seasonal data were pooled to
generate annual average leaf sulphate contents (Fig. 2C), sulphate
levels in control leaves were signicantly (p < 0.001) lower than all
treatment sites. Also, leaf sulphate contents for Southern Works
and Ganges were signicantly higher than Prospecton.
Ground-level [SO2] and leaf sulphate contents were signicantly
positively correlated (r 0.820, p < 0.001). As per recommenda-
tions made by Dwivedi et al. (2008), this allowed for the use of leaf
sulphate content as a proxy for SO2 exposure and henceforth, all
biomarkers are related to leaf sulphate content rather than ground-
level [SO2] concentrations.
3.3. Morphological, physiological and biochemical markers
As mentioned earlier, when data for the various biomarkers
investigated were compared within sites, there were no signicant
differences (p > 0.05, ANOVA) across cardinal points, so data for the
different cardinal points were pooled within sites (including the
control) for all subsequent statistical analyses.
3.3.1. Leaf area
Site and season had a signicant (p < 0.001) effect on leaf area,
both independently and in combination (i.e. there was a signicant
interaction between site and season) (Fig. 3). Except for spring, in
which leaf area was signicantly lower at Ganges, this parameter
did not differ signicantly when compared within sites (treatments
and control) across seasons. Values for the treatment sites were
signicantly lower than the control, when compared within sea-
sons and in terms of pooled annual data. Across the treatment sites,
values at Ganges were signicantly higher in winter compared with
Prospecton, but signicantly lower in spring. When annual data
were compared across the treatment sites, leaf area was signi-
cantly higher at Southern Works and statistically comparable be-
tween Prospecton and Ganges. Annual leaf area data reected
differences in leaf sulphate content between the control and the
treatment sites, but did not appear to reect differences in leaf
sulphate content across the treatment sites (e.g. the highest annual
leaf area was found at the site with the highest annual [SO2]).
Seasonal leaf area levels were however, signicantly negatively
correlated with seasonal leaf sulphate content (r 0.689,
p < 0.05).
For validation purposes, leaf area was also measured for trees
growing within 1 km of the ex situ control. These trees, which were
exposed to lower SO2 levels than those at the treatment sites (but
Please cite this article in press as: Appalasamy, M., et al., Responses of Trichilia dregeana leaves to sulphur dioxide pollution: A comparison of
morphological, physiological and biochemical biomarkers, Atmospheric Pollution Research (2017), http://dx.doi.org/10.1016/j.apr.2017.01.005
Southern Works Ganges
10.46 6.31b 17.23 5.20a
11.07 5.90a 17.12 5.23a
8.48 5.22b 16.13 4.79a
9.30 4.79a 13.88 5.63a
higher than the control), exhibited signicantly (p < 0.001) higher
leaf areas relative to trees at all the treatment sites, but lower than
the control, in winter (25.24 8.02 cm2) and spring
(24.13 4.99 cm2) (data not shown; measured for two seasons
only).
3.3.2. Relative leaf chlorophyll content
Site and season had a signicant (p < 0.001) effect on relative
leaf chlorophyll content, both independently and in combination
(Fig. 4). Values for the treatment sites were signicantly higher
than the control, when compared within seasons and when
compared in terms of annual data. Across the treatment sites,
values at Ganges were signicantly higher in autumn and winter,
and when annual data were compared, values were signicantly
highest at Ganges and statistically comparable between Prospecton
and Southern Works. Annual relative leaf chlorophyll content could
therefore reect differences in leaf sulphate content between the
control and the treatment sites, but this was not true for differences
in leaf sulphate content across the treatment sites. Seasonal relative
leaf chlorophyll content was signicantly positively correlated with
seasonal leaf sulphate content (r 0.736, p < 0.05).
For validation purposes, relative leaf chlorophyll content was
also measured for trees growing within 1 km of the ex situ control.
Values measured for these trees were signicantly (p < 0.001)
higher than Prospecton, Southern Works and the control in winter
(67.36 6.14 SPAD units) and spring (66.02 3.41 SPAD units) (data
not shown; measured for two seasons only).
3.3.3. Chlorophyll uorescence
Site (in terms of seasonal data) and season had a signicant
(p < 0.001) effect on leaf chlorophyll uorescence, both indepen-
dently and in combination (data not shown). Except for summer, in
which values for Prospecton were signicantly lower than Ganges
and the control, chlorophyll uorescence did not differ signicantly
across sites when compared within seasons, and in terms of annual
data. Annual chlorophyll uorescence levels did not reect differ-
ences in leaf sulphate content between the control and the treat-
ment sites, nor across the treatment sites. Seasonal chlorophyll
uorescence levels were also not signicantly correlated with
seasonal leaf sulphate content (r 0.255, p 0.340).
3.3.4. Intracellular hydrogen peroxide (H2O2) production
Site and season had a signicant (p < 0.001) effect on intracel-
lular H2O2 production, both independently and in combination
(Fig. 5). Values for treatment sites were often higher than the
control in autumn, winter and spring, and in some cases these
differences were signicant. For example, H2O2 levels at Prospecton
were signicantly higher than the control in winter. When annual
H2O2 data were compared across sites, values for the control were
signicantly lower than the treatment sites, while across the
treatment sites values were signicantly higher at Ganges and
statistically comparable between Prospecton and Southern Works.
Annual H2O2 levels reected differences in leaf sulphate content
 M. Appalasamy et al. / Atmospheric Pollution Research xxx (2017) 1e12 7

Fig. 3. Leaf area for T. dregeana trees exposed to differing levels of SO2 at the treatment sites (Prospecton, Southern Works and Ganges) and an ex situ control, over four seasons.
Values represent the mean SD (n 24). Bars labelled with different lowercase letters are signicantly different (p < 0.001, ANOVA) when compared across various site  season
combinations. Uppercase letters indicate signicant differences (p < 0.001, ANOVA) across sites for annual data.

Fig. 4. Leaf chlorophyll content for T. dregeana leaves exposed to differing levels of SO2 at the treatment sites (Prospecton, Southern Works and Ganges) and an ex situ control, over
four seasons. Values represent the mean SD (n 24). Bars labelled with different lowercase letters are signicantly different (p < 0.001, ANOVA) when compared across various
site  season combinations. Uppercase letters indicate signicant differences (p < 0.001, ANOVA) across sites for annual data.

between the control and the treatment sites, but this was not true Seasonal H2O2 production levels were signicantly positively
for differences in leaf sulphate content across the treatment sites. correlated with seasonal leaf sulphate content (r 0.631, p < 0.05).

Please cite this article in press as: Appalasamy, M., et al., Responses of Trichilia dregeana leaves to sulphur dioxide pollution: A comparison of
morphological, physiological and biochemical biomarkers, Atmospheric Pollution Research (2017), http://dx.doi.org/10.1016/j.apr.2017.01.005
8 M. Appalasamy et al. / Atmospheric Pollution Research xxx (2017) 1e12

Fig. 5. Intracellular hydrogen peroxide production for T. dregeana leaves exposed to differing levels of SO2 at the treatment sites (Prospecton, Southern Works and Ganges) and an ex
situ control, over four seasons. Values represent the mean SD (n 24). Bars labelled with different lowercase letters are signicantly different (p < 0.001, ANOVA) when compared
across various site  season combinations. Uppercase letters indicate signicant differences (p < 0.001, ANOVA) across sites for annual data.

3.3.5. Electrolyte leakage
Site and season had a signicant (p < 0.001) effect on electrolyte
leakage, both independently and in combination (Fig. 6). In autumn
and summer, values at Southern Works and Ganges were signi-
cantly higher than the control; values at Ganges were also higher
than Prospecton in both these seasons. When annual data were
compared, values at Southern Works and Ganges were signicantly
higher than the control, but statistically comparable across the
treatment sites. Annual electrolyte leakage levels largely (except for
Prospecton) reected differences in leaf sulphate content between
the control and the treatment sites, but this did not apply to dif-
ferences in leaf sulphate content across the treatment sites.

Fig. 6. Electrolyte leakage for T. dregeana leaves exposed to differing levels of SO2 at the treatment sites (Prospecton, Southern Works and Ganges) and an ex situ control, over four
seasons. Values represent the mean SD (n 24). Bars labelled with different lowercase letters are signicantly different (p < 0.001, ANOVA) when compared across various
site  season combinations. Uppercase letters indicate signicant differences (p < 0.001, ANOVA) across sites for annual data.

Please cite this article in press as: Appalasamy, M., et al., Responses of Trichilia dregeana leaves to sulphur dioxide pollution: A comparison of
morphological, physiological and biochemical biomarkers, Atmospheric Pollution Research (2017), http://dx.doi.org/10.1016/j.apr.2017.01.005
 M. Appalasamy et al. / Atmospheric Pollution Research xxx (2017) 1e12 9

Seasonal electrolyte leakage levels were not signicantly related to
seasonal leaf sulphate content (r 0.456, p 0.076).
3.3.6. Total aqueous antioxidant (TAA) activity
Site and season had a signicant (p < 0.001) effect on TAA, both
independently and in combination (Fig. 7). Within the control,
values were signicantly lower in spring and summer. Values at the
treatment sites were signicantly lower than the control in autumn
and winter and signicantly higher than the control in the cases of
Prospecton and Southern Works in summer. When annual TAA data
were compared, values were signicantly highest in the control and
lowest at Ganges; values were, however, statistically comparable
between Prospecton and Southern Works. Annual TAA levels re-
ected differences in leaf sulphate content between the control and
the treatment sites, but this did not apply to differences in leaf
sulphate content across the treatment sites. Seasonal TAA values
were not signicantly correlated with seasonal leaf sulphate con-
tent (r 0.260, p 0.331).
4. Discussion
Biomonitoring of air quality using plant material has been
widely employed to detect the effects of pollution on ecosystems
(Sawidis et al., 2011). Trees, in particular, are efcient at taking in
and thereby reducing harmful pollutants in the atmosphere
(Sawidis et al., 2011). Individual pollutants have unique effects on
plants, and even though the present study focussed on SO2 pollu-
tion, it should be noted that pollutants such as NO2, O3, and SO2 act
together to compromise plant growth and performance (Tiwari
et al., 2006).
4.1. Ground-level SO2 and leaf sulphate levels
Sulphur dioxide is one of the major air pollutants focussed on in
a number of studies as it is the pollutant for which most monitoring
records exist, particularly for the SDB (Matooane and Diab, 2003).
Ground-level [SO2] in this study ranged from 2.18 to 9.45 ppb at the
treatments sites across the four seasons, and except for winter, was
highest at Southern Works, followed by Ganges and Prospecton
(Fig. 2A). Similarly, leaf sulphate levels were lowest in the control
and relatively higher at Southern Works and Ganges across the
treatment sites (Fig. 2C). Most importantly, leaf sulphate content
was signicantly positively correlated with [SO2]. Leaf sulphate
content data for other South African species/sites subject to SO2
pollution were not available at the time of this study, but other
research shows similar results for tree leaves (e.g. Ficus religiosa
[Dwivedi et al., 2008]). The results obtained here therefore support
suggestions that increasing levels of atmospheric SO2 can lead to
increasing leaf sulphate contents (Huang and Murray, 1993;
Prasanna et al., 2005; Dwivedi et al., 2008).
The SO2 and leaf sulphate data suggest that SO2 pollution can
vary, both temporally and spatially within the SDB, as reported by
other authors (Matooane and Diab, 2001, 2003; Diab et al., 2002).
Diab et al. (2002) also showed Southern Works, which is located at
the centre of the SDB, to have relatively higher SO2 levels than other
monitoring stations and attributed this to strong NE and SW winds.
Stack down-drifting together with strong winds lead to high [SO2]
at ground-level, while differences among sites is also due to the
location of monitoring stations in relation to air pollution sources
within the SDB (Diab et al., 2002). Within individual sites, the
highest SO2 and leaf sulphate levels were recorded in either
autumn or winter which may be explained by the fact that wind
speed is generally higher in spring and lower from autumn to
winter (Guastella and Knudsen, 2007). Lower wind speeds lead to

Fig. 7. Total aqueous antioxidant capacity for T. dregeana leaves exposed to differing levels of SO2 at the treatment sites (Prospecton, Southern Works and Ganges) and an ex situ
control, over four seasons. Values represent the mean SD (n 24). Bars labelled with different lowercase letters are signicantly different (p < 0.001, ANOVA) when compared
across various site  season combinations. Uppercase letters indicate signicant differences (p < 0.001, ANOVA) across sites for annual data.

Please cite this article in press as: Appalasamy, M., et al., Responses of Trichilia dregeana leaves to sulphur dioxide pollution: A comparison of
morphological, physiological and biochemical biomarkers, Atmospheric Pollution Research (2017), http://dx.doi.org/10.1016/j.apr.2017.01.005
10 M. Appalasamy et al. / Atmospheric Pollution Research xxx (2017) 1e12
poor vertical mixing and low horizontal transportation of pollution
out of the source area (Guastella and Knudsen, 2007). Other studies
have also attributed seasonal variation in SO2 levels (Carmichael
et al., 2003) to seasonal differences in wind patterns (Diab et al.,
2002). Annual [SO2] across all three sites ranged from 4.57 to
6.59 ppb, which is within the global SO2 threshold range
(3.8e11.4 ppb) where irreparable damage can be caused to agri-
cultural crops and natural vegetation (Carmichael et al.,. 2003;
Josipovic et al., 2010). These data also suggest that very little has
changed within the SDB with regards to SO2 pollution since Diab
et al.s 2002 study, which showed SO2 levels to be > 3.5 within
the SDB. It should also be noted that from the limited data available
for other pollutants, areas of high [SO2] may not necessarily overlap
with areas of high NO2 and PM pollution.
4.2. NO2 and PM pollution levels
As mentioned earlier, limited data on NO2 and PM levels were
available for two of the three treatment sites. Levels of NO2 were
not as high as other countries around the world (Emberson et al.,
2001); however, values for spring and summer at Ganges were
above the European threshold for vegetation protection (15.96 ppb)
(Baldasano et al., 2003; Josipovic et al., 2010). According to Josipovic
et al. (2010), high levels of NO2 are found at sites close to industrial
hubs with much seasonal variation. The range of PM values recor-
ded in the present study is below the European threshold (40 mg/
m3) (Baldasano et al., 2003).
4.3. Comparison of biomarkers
Changes in leaf morphology are the rst visible and most easily
identiable stress biomarkers (Hijano et al., 2005), as the surface of
a leaf is in direct contact with air pollution (Assadi et al., 2011;
Tanee and Albert, 2013). According to Pandey (2005), high levels
of SO2 can have adverse effects on plant growth such as reduced
leaf area, but the effects of air pollution on leaf area are species-
dependent and are based on a plant's ability to protect itself or
adapt to its surrounding environment (Wuytack et al., 2011). In this
study, T. dregeana leaf area was signicantly reduced relative to the
control with exposure to SO2, irrespective of the season (Fig. 3), and
was signicantly negatively correlated with sulphate content in
leaves (in terms of seasonal data). However, this biomarker was not
sensitive enough to reect differences in sulphate content in leaves
across the treatment sites. A number of authors have reported trees
in unpolluted locations to have higher leaf areas (Dineva, 2004;
Areington et al., 2015) and in species like white willow (Salix alba
L.), a decrease in leaf area is said to minimise the uptake of air
pollution (Wuytack et al., 2011).
In the present study, the decrease in leaf area across all the
treatment sites compared with the control was accompanied by a
signicant increase in relative leaf chlorophyll content (Fig. 4).
Based on the reasoning developed in other studies (Assadi et al.,
2011; Areington et al., 2015), the physiological link between leaf
area and relative chlorophyll content may have resulted in
T. dregeana trees decreasing leaf area to reduce the surface area
exposed to air pollution, while increasing relative chlorophyll
content to compensate for the reduction in the leaf area available
for harvesting photosynthetically active radiation (PAR). Relative
leaf chlorophyll content was signicantly positively correlated with
leaf sulphate content (in terms of seasonal data), but like leaf area,
was not sensitive enough as a biomarker to reect differences in
leaf sulphate content across the treatment sites. Trees growing
within 1 km of the ex situ control, exposed to signicantly lower
levels of SO2 than those at the treatments sites, also exhibited
signicantly higher leaf areas and relative chlorophyll contents
Please cite this article in press as: Appalasamy, M., et al., Responses of Trichilia dregeana leaves to sulphur dioxide pollution: A comparison of
morphological, physiological and biochemical biomarkers, Atmospheric Pollution Research (2017), http://dx.doi.org/10.1016/j.apr.2017.01.005
than the treatment sites, but lower than the control. This validates
the trends obtained for these biomarkers; however, the question
still exists as to why trees at Southern Works, which were exposed
to the highest SO2 levels, exhibited higher leaf areas than Ganges
and Prospecton (Fig. 3). Reference to the NO2 and PM levels at
Southern Works and Ganges may offer an explanation. The accu-
mulation of PM on leaves can block the opening of stomata,
reducing the gas exchange and consequentially leaf growth (Cui
et al., 2006), while increased NO2 levels, as alluded to earlier, can
affect plant growth negatively (Tiwari et al., 2006). Assadi et al.
(2011) stressed that the effects of atmospheric pollutants on
plants are cumulative. This suggests that while leaf sulphate con-
tent is a good proxy for SO2 pollution within the SDB, levels of other
pollutants (such as NO2 and PM) should also be considered when
using tree leaves as bioindicators of pollution, since the effects of
SO2 have been reported to be more severe when combined with
other atmospheric pollutants (Varshney et al., 1979).
Chlorophyll uorescence (Maxwell and Johnson, 2000) can give
insight into the ability of a plant to tolerate an environmental stress
or the extent of damage caused by the stress on the photosynthetic
system (Maxwell and Johnson, 2000). Air pollution can have sig-
nicant effects on the biochemistry of plants, often leading to
increased production of ROS and associated oxidative stress in
leaves (Dat et al., 2000). Air pollutants such as SO2, enter a leaf
through stomata causing damage to one of the main sources of ROS
in leaves, viz. chloroplasts (Dat et al., 2000; Seyyednejad et al.,
2013). Despite these possibilities, chlorophyll uorescence did not
appear to reect differences in leaf sulphate content between the
control and the treatment sites. Chloroplasts are also a major pro-
ducer of H2O2 (Quan et al., 2008), which plays a key role in plants
during environmental stress (Dat et al., 2000). Hydrogen peroxide
is also harmful when produced in excess owing to its relatively long
lifespan and high membrane permeability (Hung et al., 2008).
Studies have shown that exposure to high levels of air pollution can
lead to harmful levels of H2O2 in plants (Valavanidis et al., 2006;
Areington et al., 2015). Hydrogen peroxide levels in T. dregeana
leaves exposed to SO2 were often higher than the control in
autumn, winter and spring (Fig. 5), and was signicantly positively
correlated with sulphate content in leaves. However, H2O2 pro-
duction was not sensitive enough as a biomarker to reect differ-
ences in leaf sulphate contents across the treatment sites. This may
be partly due to the widely reported dual role (i.e. harmful versus
benecial) of H2O2 in plant tissues (Dat et al., 2000). Apart from
inicting damage in plant tissues, H2O2 can also be part of the
plant's overall adaptation to changes in the environment (Neill
et al., 2002), as it participates in activating defence responses to
stress (Quan et al., 2008).
Damage to plant tissues caused by excess levels of ROS can be
measured in terms of an increase in electrolyte leakage (Blokhina
et al., 2003). Given the fact that SO2 exposure led to heightened
levels of ROS in T. dregeana leaves (Fig. 5), it was not surprising that
annual electrolyte leakage levels at the treatment sites with the two
highest leaf sulphate levels (Southern Works and Ganges) were
signicantly higher than the control (Fig. 6). However, when one
looks at the seasonal values, this was only true for autumn and
summer, which may explain why electrolyte leakage levels were
not signicantly correlated with sulphate content in leaves. Elec-
trolyte leakage in the present study could be used to distinguish
between the control and the two sites with the highest leaf sul-
phate content, but it was not sensitive enough as a biomarker to
reect differences in sulphate content of leaves across the treat-
ment sites.
Plants have developed antioxidant (enzymic and non-enzymic)
defence mechanisms to combat oxidative stress. However, when
plants are exposed to high levels of environmental stress, the
 M. Appalasamy et al. / Atmospheric Pollution Research xxx (2017) 1e12
balance between the ROS produced and the quenching activity of
antioxidants is disturbed, leading to oxidative damage (Areington
et al., 2015). According to Tiwari et al. (2006), antioxidant levels
in plant tissues increase with an increase in pollution stress.
However, in the present study this was only true in summer, and
the difference between the control and treatment sites (Prospecton
and Southern Works) were signicant (Fig. 7). Interestingly, leaves
from the ex situ control had signicantly higher antioxidant levels
than the treatment sites in autumn and winter. This may be an
indication that control leaves (in autumn and winter, at least) were
healthier than those at the treatment sites; Novak et al. (2003)
suggest that high antioxidant levels are indicative of healthy
vigorous plants. In this regard, control leaves exhibited exception-
ally high TAA levels in autumn and winter when the highest leaf
sulphate levels were recorded across the treatment sites (Fig. 2B).
However, TAA was not signicantly correlated with leaf sulphate
content and the seasonal variability in TAA responses to SO2
exposure suggests that this parameter may not be a suitable
biomarker of air pollution in T. dregeana.
5. Conclusions
The use of plants as bioindicators of environmental stress rep-
resents an easy and inexpensive way of monitoring environmental
changes, including air pollution. From the six biomarkers compared
in this study, ve biomarkers (viz. leaf area, relative chlorophyll
content, intracellular H2O2, electrolyte leakage and TAA) were able
to discriminate between SO2 exposed and unexposed leaves.
However, of these ve biomarkers, only three were signicantly
correlated with leaf sulphate content (viz. leaf area, relative chlo-
rophyll content, and intracellular H2O2). These three biomarkers
were, however, not sensitive enough to reect differences in leaf
sulphate content across the treatment sites. The results suggest that
leaf sulphate content is a reliable proxy for SO2 pollution and that
biomarkers and leaf sulphate responses are strongly inuenced by
seasonal variations in [SO2]. Additionally, the study reinforced
previous reports of the physiological link between leaf area and
relative chlorophyll content (Assadi et al., 2011; Areington et al.,
2015), which suggests that these should be measured in conjunc-
tion with each other. In light of the above, all biomarkers, except for
chlorophyll uorescence, can be used to establish T. dregeana leaves
as a bioindicator of industrial SO2 pollution.
Acknowledgements
The lead author would like to express her gratitude to the
following individuals for their support throughout this research:
W.R. Woodenberg, B. Gijsbertsen, C.A. Areington, and P. Munien.
This study was funded by a Thuthuka grant awarded to Sershen.
The eThekwini Municipality is also acknowledged for providing the
air quality data used in this study.
References
American Public Health Association (APHA), 1992. In: Greenberg, A.E., Clesceri, L.S.,
Eaton, A.D. (Eds.), Standard Methods for the Examination of Water and
Wastewater. Washington, DC, 4500E.
Appalasamy, M., Varghese, B., Sershen, Ismail, R., 2016. Examining the utility of
hyperspectral remote sensing and partial least squares to predict plant stress
responses to sulphur dioxide pollution: a case study of Trichilia dregeana Sond.
South Afr. Geogr. J. 1e19.
Areington, C.A., Varghese, B., Ramdhani, S., Sershen, 2015. An assessment of
morphological, physiological and biochemical biomarkers of industrial air
pollution in the leaves of Brachylaena discolor. Water, Air Soil Pollut. 226 (9),
1e14.
Assadi, A., Pirbalouti, A.G., Malekpoor, F., Teimori, N., Assadi, L., 2011. Impact of air
pollution on physiological and morphological characteristics of Eucalyptus
camaldulensis Den. J. Food, Agric. Environ. 9 (2), 676e679.
Please cite this article in press as: Appalasamy, M., et al., Responses of Trichilia dregeana leaves to sulphur dioxide pollution: A comparison of
morphological, physiological and biochemical biomarkers, Atmospheric Pollution Research (2017), http://dx.doi.org/10.1016/j.apr.2017.01.005
11
Bakker, M.I., Vorenhout, M., Sijm, D.T.H.M., Kolloffel, C., 1999. Dry deposition at-
mospheric polycyclic aromatic hydrocarbons in three Plantago species. Environ.
Toxicol. Chem. 18, 2289e2294.
Baldasano, J.M., Valera, E., Jimenez, P., 2003. Air quality data from large cities. Sci.
Total Environ. 307 (1), 141e165.
Blokhina, O., Virolainen, E., Fagerstedt, K.V., 2003. Antioxidants, oxidative damage
and oxygen deprivation stress: a review. Ann. Bot. 91 (2), 179e194.
Carmichael, G.R., Ferm, M., Thongboonchoo, N., Woo, J.H., Chan, L.Y., Murano, K.,
Viet, P.H., Mossberg, C., Bala, R., Boonjawat, J., Upatum, P., 2003. Measurements
of sulfur dioxide, ozone and ammonia concentrations in Asia, Africa, and South
America using passive samplers. Atmos. Environ. 37 (9), 1293e1308.
Chen, C.C., Chen, H., Chen, Y.R., 2009. A new method to measure leaf age:Leaf
measuring-interval index. Am. J. Bot. 96 (7), 1313e1318.
Conti, M.E., Cechetti, G., 2001. Biological mointoring: lichens as bioindicators of air
pollution assessment - a review. Environ. Pollut. 114, 471e492.
Cui, H.X., Jiang, G.M., Niu, S.L., Jiang, C.D., Liu, M.Z., Yu, S.L., Gao, L.M., 2006.
Ecophysiological response of plants to combined pollution from heavy-duty
vehicles and industrial emissions in higher humidity. J. Integr. Plant Biol. 48
(12), 1391e1400.
Dat, J., Vandenabeele, S., Vranov
a, E., Van Montagu, M., Inze
, D., Van Breusegem, F.,
2000. Dual action of the active oxygen species during plant stress responses.
Cellular and Molecular Life Sciences. CMLS 57 (5), 779e795.
Department of Energy and Minerals, South Africa, 2008. Annual Report. http://
wwwdme.gov.za/pdfs/about_dme/AR%20pdf%20for%20web.pdf.
Diab, R., Prause, A., Bencherif, H., 2002. Analysis of SO2 pollution in the South
Durban industrial basin. South Afr. J. Sci. 98, 543e546.
Diab, R.D., Motha, A., 2007. An analysis of key institutional factors inuencing air
quality in South Durban using the DPSIR framework. South Afr. Geogr. J. 89 (1),
22e33.
Dineva, S.B., 2004. Comparative studies of the leaf morphology and structure of
white ash Fraxinus americana L. and London plane tree Platanus acerifolia Willd
growing in polluted area. Dendrobiology 52, 3e8.
Dwivedi, A.K., Tripathi, B.D., Shashi, 2008. Effect of ambient air sulphur dioxide on
sulphate accumulation in plants. J. Environ. Biol. 29 (3), 377e379.
Emberson, L.D., Ashmore, M.R., Murray, F., Kuylenstierna, J.C.I., Percy, K.E., Izuta, T.,
Zheng, Y., Shimizu, H., Sheu, B.H., Liu, C.P., Agrawal, M., 2001. Impacts of air
pollutants on vegetation in developing countries. Water, Air, Soil Pollut. 130
(1e4), 107e118.
Gill, S.S., Tuteja, N., 2010. Reactive oxygen species and antioxidant machinery in
abiotic stress tolerance in crop plants. Plant Physiology Biochem. 48, 909e930.
Guastella, L., Knudsen, S., 2007. South Durban Basin Multi-point Plan Case Study
Report. Air Quality Act Implementation: Air Quality Management Planning,
pp. 1e34. Book 12, series C.
Hijano, C.F., Domnguez, M.D.P., Gimnez, R.G., Snchez, P.H., Garca, I.S., 2005.
Higher plants as bioindicators of sulphur dioxide emissions in urban environ-
ments. Environ. Monit. Assess. 111 (1), 75e88.
Huang, B.L., Murray, F., 1993. Effect of sulphur dioxide fumigation on growth and
sulphur accumulation in wheat (Triticum aestivum cv. Wilgoyne) under salinity
stress. Agric. Ecosyst. Environ. 43 (3), 285e330.
Hung, K.T., Cheng, D.G., Hsu, Y.T., Kao, C.H., 2008. Abscisic acid-induced hydrogen
peroxide is required for antho cyanin accumulation in leaves of rice seedlings.
J. Plant Physiology 165, 1280e1287.
Jain, A.K., Hayhoe, K.A.S., 2008. Global air pollution problems. In: Hewitt, C.N.,
Jackson, A.V. (Eds.), Handbook of Atmospheric Sciences 2: Problems, Tools and
Applications. John Wiley & Sons, Oxford, pp. 339e375.
Jana, S., Choudhuri, M.A., 1981. Glycolate metabolism of three submersed aquatic
angiosperms: effect of heavy metals. Aquat. Bot. 11, 67e77.
Josipovic, M., Annegarn, H.J., Kneen, M.A., Pienaar, J.J., Piketh, S.J., 2010. Concen-
trations, distributions and critical level exceedance assessment of SO2, NO2 and
O3 in South Africa. Environ. Monit. Assess. 171 (1e4), 181e196.
Klumpp, A., Hintemann, T., Lima, J.S., Kandeler, E., 2003. Bioindication of air
pollution effects near a copper smelter in Brazil using mango trees and soil
microbiological properties. Environ. Pollut. 126, 313e321.
Lamego, M.M., de Souza Jr., P.A., Ferriera, E.P., Garg, V.K., 2000. Quantication of air
pollution sources. Revista Controle Automacao 11 (2), 77e81.
Lau, O.W., Ho, S.Y., 1993. Simultaneously determination of traces of iron, cobalt,
nickel, copper, mercury and lead in water by energy-dispersive X-ray uores-
cence spectrometry after pre-concentration aas their piperazino-1,4-bis-
dithiocarbamate complexes. Anal. Chim. Acta 290, 269e277.
Lau, O.W., Luk, S.F., 2001. Leaves of Bauhinia blakeana as indicators of atmospheric
pollution in Hong Kong. Atmos. Environ. 3113e3120.
Lubbe, E., Rodda, N., 2016. Effects of greywater irrigation on germination, growth
and photosynthetic characteristics in selected African leafy vegetables. Water
SA 42 (2), 203e212.
Madejon, P., Maranon, T., Murillo, J.M., Robinson, B., 2004. White poplar (Populus
alba) as a biomonitors of trace elements in contaminated riparian forests. En-
viron. Pollut. 132, 145e155.
Matooane, L., Diab, R.D., 2001. Air pollution carrying capacity in the South Durban
industrial basin. South Afr. J. Sci. 97 (11e12), 450e453.
Matooane, M., Diab, R., 2003. Health risk for sulphur dioxide pollution in South
Durban, South Africa. Archives Environ. Health An Int. J. 58 (12), 763e770.
Maxwell, K., Johnson, G.N., 2000. Chlorophyll uorescence-a practical guide. J. Exp.
Bot. 51 (345), 659e668.
Menyah, K.M., Wolde-Rufael, 2010. Energy consumption, pollutant emissions and
economic growth in South Africa. Energy Econ. 32, 1374e1382.
12 M. Appalasamy et al. / Atmospheric Pollution Research xxx (2017) 1e12
Moreno, E., Sagnotti, L., Dinares-Turell, Winkler, A., Cascella, A., 2003. Bio-
monitoring of trafc air pollution in Rome using magnetic properties of tree
leaves. Atmos. Environ. 37, 2967e2977.
Naidoo, G., 2009. Differential effects of nitrogen and phosphorus enrichment on
growth of dwarf Avicennia marina mangroves. Aquat. Bot. 90, 184e190.
Neill, S., Desikan, R., Hancock, J., 2002. Hydrogen peroxide signalling. Curr. Opin.
Plant Biol. 5 (5), 388e395.
Netto, A.T., Campostrini, E., de Oliveira, J.G., Bressan-Smith, R.E., 2005. Photosyn-
thetic pigments, nitrogen, chlorophyll a uorescence and SPAD-502 readings in
coffee leaves. Sci. Hortic. 104, 199e209.
Novak, K., Skelly, J.M., Schaub, M., Krauchi, N., Hug, C., Landolt, W., Bleuler, P., 2003.
Ozone air pollution and foliar injury development on native plants of
Switzerland. Environ. Pollut. 125 (1), 41e52.
Pandey, J., 2005. Evaluation of air pollution phytotoxicity downwind of a phosphate
fertilizer factory in India. Environ. Monit. Assess. 100 (1e3), 249e266.
Pitterman, J., Sage, R.F., 2000. The response of high altitude C4 grass Muhlenbergia
Montana (Nutt.) A.S. Hitchc. to long- and short- term chilling. J. Exp. Bot. 52,
829e838.
Pooley, E., 1993. The Complete Field Guide to Trees of Natal, Zululand and Transkei,
third ed. Natal ora publications trust, Durban, pp. 204e205.
Prasanna, R.T., Deshpande, U.K., Gawaid, U., 2005. Effect of sulphur dioxide on
protein and carbohydrate contents of some macrophytes. Pollut. Res. 24,
427e430.
Quan, L.J., Zhang, B., Shi, W.W., Li, H.Y., 2008. Hydrogen peroxide in plants: a ver-
satile molecule of the reactive oxygen species network. J. Integr. Plant Biol. 50
(1), 2e18.
Re, R., Pellegrini, N., Proteggente, A., Pannala, A., Yang, M., Andrice-Evans, C., 1999.
Antioxidant activity applying an improved ABTS radical cation decolorization
Please cite this article in press as: Appalasamy, M., et al., Responses of Trichilia dregeana leaves to sulphur dioxide pollution: A comparison of
morphological, physiological and biochemical biomarkers, Atmospheric Pollution Research (2017), http://dx.doi.org/10.1016/j.apr.2017.01.005
assay. Free Radic. Biol. Med. 26, 1231e1237.
Santamaria, J.M., Martin, A., 1997. Tree bark as a bioindicator of air pollution in
Navarra, Spain. Water, Air Soil Pollut. 98, 381e387.
Sawidis, T., Breuste, T., Mitrovic, M., Pavlovic, P., Tsigaridas, K., 2011. Trees as bio-
indicators of heavy metal pollution in three European countries. Environ. Pollut.
159, 3560e3570.
Sawidis, T., Heinrich, G., Chettri, M.K., 1997. Cesium-137 monitoring using lichens
from Macedonia, northern Greece. Can. J. Bot. 75 (12), 2216e2223.
Seyyednejad, S.M., Koochak, H., Vaezi, J., 2013. Changes in anti-oxidative enzymes
activity, protein content and ascorbic acid level in Prosopis juliora exposed to
industrial air pollution. J. Biol. 2 (10), 482e492.
Tanee, F.B.G., Albert, E., 2013. Air pollution tolerance indices of plants growing
around Umuebulu gas are station in rivers state, Nigeria. Afr. J. Environ. Sci.
Technol. 7 (1), 1e8.
Tiwari, S., Agrawal, M., Marshall, F.M., 2006. Evaluation of ambient air pollution
impact on carrot plants at a sub urban site using open top chambers. Environ.
Monit. Assess. 119, 15e30.
Valavanidis, A., Vlahogianni, T., Dassenakis, M., Scoullos, M., 2006. Molecular bio-
markers of oxidative stress in aquatic organisms in relation to toxic environ-
mental pollutants. Ecotoxicol. Environ. Saf. 64 (2), 178e189.
Varshney, C.K., Garg, J.K., Lauenroth, W.K., Heitschmidt, R.K., 1979. Plant responses
to sulfur dioxide pollution. Crit. Rev. Environ. Sci. Technol. 9 (1), 27e49.
Vranova, E., Inze, D., Van Breusegem, F., 2002. Signal transduction during oxidative
stress. J. Exp. Biol. 53 (372), 1227e1236.
Wuytack, T., Wuyts, K., Van Dongen, S., Baeten, L., Kardel, F., Verheyen, K.,
Samson, R., 2011. The effect of air pollution and other environmental stressors
on leaf uctuating asymmetry and specic leaf area of Salix alba L. Environ.
Pollut. 159, 2405e2411.