Response of roadside tree leaves in a tropical city to automobile pollution

One of the sources of air pollutants in the surrounding environment is the automobile emissions. Automobiles produce gaseous and particulate matters which are toxic and inflict damage to roadside plants. Roadside trees are notable for the absorption, sequestering of contaminants and the effective interceptor of airborne pollution. In view of this, the present work was based on investigating the macro-morphological and micro-morphological changes that boost the tolerance and continued existence of four roadside trees, namely Ficus platyphylla, Mangifera indica, Polyalthia longifolia and Terminalia cattapa in the incidence of vehicle exhaust emissions in Kumasi Metropolis, Ghana. Three arterial roads representing three different traffic volumes of extreme, heavy and severe were considered as observational sites. Kwame Nkrumah University of Science and Technology Campus was selected as the control site. The macro-morphological characteristics of the four tree species showed reduced leaf area, whilst the micro-morphological results revealed that stomata size, number and index were reduced at the arterial roadsides in all the four tree species. There was increased epidermal cell number and length and trichome length at the polluted arterial roadsides when compared to the control. These variations can be considered as pointers of environmental stress and could be used as indicators of urban air pollution.


Introduction
Exhausts from vehicles are one of the sources of air pollution that generates approximately 70% of all harmful emissions (Balat and Balat, 2009). Vehicles emit pollutants such as CO2, CO, oxides of nitrogen, SO2, hydrocarbons, unburnt petrol and carbon particles (Wang and Xie, 2009;Bhandarkar, 2013). These gaseous and particulate pollutants' arising from the vehicle exhaust are toxic and causes damage to roadside trees (Joshi and Swami, 2009). Roadside plants are noteworthy in the absorption, sequestering of pollutants and a proficient interceptor of airborne contaminants. Studies have recorded changes in plants caused by wide range of environmental pollutants, and the vast majority of these works allude to physiological modifications (Patra and Sharma, 2000). The morphological modifications that occurs in plants resulting from air pollution are changes in stomatal and epidermal cell size, lower recurrence, thickening of cell wall, epicuticular wax deposition alterations and chlorosis (Uka et al., 2017). Several studies in this area have been conducted to AcademicPres Notulae Scientia Biologicae investigate morphological, structural, physiological and biochemical variations in trees (Pourkhabbaz et al., 2010;Maheswari, 2012;Pandey et al., 2015;Lohe et al., 2015;Deepika and Haritash, 2016). Surface leaf characters, including stomata and epidermal cells, have been reported to be drastically altered in plant species growing along the roadside due to the stress of car exhaust pollution with high traffic intensity in urban areas (Rai and Mishra, 2013). A few authors consider foliar epidemis as a bioindicator of environmental quality (Alves et al., 2008;Balasooriya et al., 2009). Srivastava (1999) observed that general plant development was influenced with serious mutilations in foliar epidermal characters. It likewise brought up the significance of the cuticle and epidermal features in the determination of resistance/ sensitivity of each species to environmental pollutants. Gostin and Ivanescu (2007) observed structural and micro-morphological traits of the stomata in Salix alba leaves due to air pollution. Kapoor et al. (2014) reported that reduction in size of epidermal cell and stomata, and increment in the quantity of epidermal cells, stomata and trichomes occur on exposure to SO2 contamination. Platanus orientalis leaves subjected to automobile air pollution had reduced stomatal thickness and stomata widths than those leaves from the urban area (Pourkhabbaz et al., 2010). It was also observed that vehicular air pollution demonstrated checked modification in epidermal attributes, with decreased number of stomata and epidermal cells per unit zone, while length and width of stomata and epidermal cells increase were observed thus its usage as biomarkers of auto pollution (Verma and Chandra, 2014). The use of tree leaves as accumulative biomonitors of air pollution is of great environmental significance (Bargagli, 1998). The leaves go about as air pollution receptors and biological absorbers or filters of pollutants (Free-Smith et al., 2004), understanding responses that varies with different plants species to air pollution, would be better with utilising some of the selected tree species. As traffic and air pollution persists in the Kumasi Metropolis, attention is geared towards ameliorating the problems of environmental pollution. Morpho-anatomical adjustments are promising means to measure the air quality of the urban habitat (El-Khatib et al., 2011). Scientists have likewise reported that plants developed along roadsides showed impressive harm because of vehicular air pollution (Iqbal and Shafiq, 1999;Shafiq and Iqbal, 2003;Shafiq and Iqbal, 2005). In this study, leaves from Ficus platyphylla, Mangifera indica, Polyalthia longifolia and Terminalia catappa growing in the arterial road sites of Kumasi Metropolis with distinctive gridlock were collected to determine the adverse effect of vehicular pollutants on their macro-morphological and micro-morphological traits of their leaf epidermal structure.

Study area
Kumasi is the second biggest city in Ghana. It is situated around 270 km north of the national capital, Accra, 397 km south of Tamale (Northern Regional capital) and 120 km South-east of Sunyani (Brong Ahafo Regional capital). Kumasi is situated between latitude 6.35° -6.40° and longitude 1.30° -1.35° and has a land area of 254 km 2 . The minimum temperature in the area is around 21.5 °C with the maximum temperature of 33.7 °C. Kumasi-Accra, Kumasi-Mampong and Kumasi-Offinso roads were selected for sampling because these major roads experience extreme congestion using average vehicle speed as a parameter. These major arterial roads: Accra road I (Arterial road I); Offinso road (Arterial road II) and Mampong road (Arterial road III) representing three different traffic volumes, were considered as polluted areas, while Kwame Nkrumah University of Science and Technology Campus was selected as the control for the purposes of comparison (Table 1).
Tree species and collection of samples Four tree species (Terminalia catappa, Mangifera indica, Ficus platyphylla and Polyalthia longifolia) were selected for the study. The basis of their selection is anchored on easy recognition, high distribution and abundance along key arterial roads in the Kumasi Metropolis. Triplicate samples of leaves of Ficus platyphylla, Mangifera indica, Polyalthia longifolia and Terminalia cattapa were collected from each sampling site. Twenty (20) physiologically active fully expanded leaves, third from the tip of phyllotaxis position were collected from each of the four tree species (three individuals per species). The Total sample size of 240 leaves per sampling site was collected for the morphometry and micromorphometry analyses. Sampling intervals between each tree species replicate ranged from 2.0 to 4.8 km along each road. Leaf samples were collected between the months of August-November, 2015 before the onset of the harmattan season in December when trees shed their leaves.

Morphometry studies
The measurement of leaf length was from leaf tip along the midrib of the leaf lamina to the leaf base at the point of attachment of the lamina to the petiole. The leaf breadth was measured along the widest breadth across the lamina. The leaf area was determined using a graph paper. The margin of leaf samples was traced on graph sheets with 1mm 2 square cells. Leaf area was determined by counting the number of squares enclosed in each tracing (Ogoke et al., 2009).

Micromorphometry studies
The leaf samples were thoroughly washed with tap water to eliminate dust particles from their surfaces. The leaves were allowed to dry and leaf epidermal peel slides were made by the methods of lasting impressions (Rai and Mishra, 2013;Otuu et al., 2015). On the sampled leaves 1 cm 2 portion of the leaf abaxial and adaxial surfaces were painted with a thick layer of clear nail polish and left to dry for 10 minutes. Second and third coatings were successively applied at 5 minutes intervals and then left to completely dry for 30 mins. Epidermal strips of the leaf samples were scrape up gently using forceps and placed in drops of glycerine on clean microscopic slides. They were covered with cover slips and viewed under the light microscope at both low and high-power magnification. Photomicrographs were taken with Leica DM750 Microscope at different magnifications. Measurements were performed on ten observations. The stomata length and breadth, epidermal cell length and length of trichomes were measured with the aid of the stage ocular micrometer. Stomatal and epidermal cell numbers (densities) were observed and counted per microscopic field view diameter of 1mm 2 on the epidermal strips of sampled leaves. Stomatal size (SS) was calculated using the equation of Franco (1939). Stomata size = Length × breadth × K, where K (Franco´s constant = 0.78524). Stomatal index (SI), which is the number of stomata present in a unit area of leaf in percentage, was calculated according to Salisbury (1927) as follows: S.I% = SD/(SD+ECD)×100 where SD is stomatal density; ECD is epidermal cell density. The diurnal analysis of CO, NO2, SO2 and VOC was monitored in the sampling sites using Aeroqual Series 500 (S500) gas monitors (Aeroqual Limited, Auckland, New Zealand). The Aeroqual monitors were placed at 1.5 m elevation above the ground. The Aeroqual Monitors were programmed to record 5 min average concentrations of the monitored air quality parameters continuously for 8 hrs at three points in each sampling site. The 5 min averages were summed up to hourly means. The ambient air quality in each site was monitored for 6 days in week for three months.

Data analysis
Multiple regression analysis was carried out using Statistica software version 7.0 for the evaluation of relationship between air quality parameters and Micro-morphological features. Furthermore, One-way analysis of variance (ANOVA) was conducted to test for differences in plant morphological and among the different roads; this analysis was conducted to reinforce the regression analysis.

Results
Air quality of the studied sites The ambient CO, SO2, NO2 and VOC concentrations in the arterial roadsides and the control is as presented in Table 2. Morphological characteristics of the selected tree species in the study area All four plants, Terminalia catappa, Mangifera indica, Ficus platyphylla and Polyalthia longifolia exhibited a tree growth habit and were perennials. Except for Terminalia catappa which is semi deciduous the other tree species are evergreen. Mangifera indica and Ficus platyphylla had dense crown structure, whilst Terminalia catappa and Polyalthia longifolia had spreading and irregular crown structures respectively. The leaf surface textures of all species were leathery except for Polyalthia longifolia which had a smooth texture.
Leaves of all four tree species were hardy in nature (Table 3). Effect of vehicular air pollution on macro-morphological characteristics

Leaf area
The leaf area of all the tree species at the arterial road sites were lower and significantly different from those at the control except for Ficus platyphylla (P<0.05) ( Table 4). The highest leaf area was observed in leaves of the studied tree species growing in the Control, whilst the lowest leaf area was recorded at the arterial roads.
In Terminalia catappa the mean leaf area was highest at the control site and the lowest mean value of 61.18 mm 2 was recorded at the Arterial road I. There was no significant difference in the leaf area among the arterial road sites. However, there was significant difference (p= 0.02), when compared to the Control.  Effect of vehicular pollution on micro-morphological variations of the studied tree species in the Kumasi Metropolis Terminalia catappa In Terminalia catappa leaves, there was a general reduction in stomata size, stomata number and stomata index in all the arterial road sites when compared to the Control. There was no significant difference among the arterial road sites and also when compared to the Control (p>0.05) stomata size, stomata number and stomata index were lower in the arterial road sites, when compared to the control (Table 5). On the other hand, there was a general increase in epidermal cell number, epidermal cell length and trichomes length in all the arterial road sites when compared to the Control (Table 7). There was no significant difference among the arterial road sites and also when compared to the Control (p>0.05) ( Table 5). Stomata in the abaxial surface of leaves in the Control were observed to have little or no occludation (Plate 1a) whilst most of the stomata in abaxial surface of leaves in the arterial road sites were occluded (Plate 1b). The trichomes in the abaxial surface of leaves in the Control were observed to be shorter (Plate 1c) whilst those on abaxial surface of leaves in the arterial road sites were longer (Plate 1d)

Mangifera indica
In general, stomata size, stomata number and stomata index in the leaves of Mangifera indica were lower in all the arterial road sites when compared to the Control site. There was no significant difference among the arterial road sites and also when compared to the Control (p>0.05) ( Table 6). As observed in Terminalia catappa epidermal cell number and epidermal cell length in all the arterial roads were higher when compared to the Control site. There was no significant difference among the arterial road sites and also when compared to the Control (p>0.05) ( Ficus platyphylla In Ficus platyphylla leaves, stomata size, stomata number and stomata index were lower in all the arterial road sites when compared with the Control. There was no significant difference among the arterial road sites and also when compared to the Control (p>0.05) ( Table 7). Epidermal cell number, epidermal cell length and trichome length increased in all the arterial road sites. There was no significant difference among the arterial road sites and also when compared to the Control (p>0.05) ( Table 7). Stomata in the abaxial surface of leaves in the Control were observed to have no occludation (Plate 3a) whilst majority of the stomata in abaxial surface of leaves in the arterial road site were occluded (Plate 3b). The trichomes in the abaxial surface of the leaves in the Control were shorter (Plate 3c) than those on abaxial surface of leaves in the arterial road site (Plate 3d).  Polyalthia longifolia In the leaves of P. longifolia, stomata size, stomata number and stomata index were much lower in all the arterial road sites when compared to the Control. There was no significant difference among the arterial road sites and also when compared to the Control (p>0.05) ( Table 8). Epidermal cell number, epidermal cell length and trichome length increased in all the arterial road sites when compared to the control. There was no significant difference among the arterial road sites, however, there was statistical difference between Arterial road I and Control (p= 0.035) ( Table 8). Stomata in the abaxial surface of the leaves of P. Longifolia in the Control were observed to have no occludation (Plate 4a) whilst most of the stomata in the abaxial surface of the leaves in the experimental sites were occluded (Plate 4b). The trichomes length in the abaxial surface of leaves in the Control were observed to be shorter (Plate 4c) than those on the abaxial surface in the arterial road sites (Plate 4d).  Relationship between ambient air quality and micro-morphological variations The relationship between ambient air quality and micro-morphological variables of Terminalia catappa, Mangifera indica, Ficus platyphylla and Polyalthia longifolia was investigated using the multiple regression analysis. The result is presented using the Pareto chart of t-values for the regression coefficients (Figures 1-4). CO, SO2, NO2 and VOC were used as the independent or predictive variables, while stomata number, stomata size, epidermal cell number and trichome length were the response or dependent variables. In Terminalia catappa, CO was a significant predictive variable for epidermal cell number and stomata size, and SO2 was a significant correlate for epidermal cell number (Figure 1). SO2 was a significant predictive variable for epidermal cell number in Mangifera indica (Fig. 2b). In Ficus platyphylla, CO and NO2 were significant predictors of epidermal cell number (Figures 3a and c). SO2 was significant as a predictor of the epidermal cell number and stomata size in Ficus platyphylla (Figure 3b). It was also observed that none of the air pollutants acted as a significant predictive variable for stomata number, stomata size, epidermal cell number and trichome length in Polyalthia longifolia (Figure 4).

Discussion
In this study, the leaf area in all the four tree species (Terminalia catappa, Mangifera indica, Ficus platyphylla and Polyalthia longifolia) studied were lower in the arterial roads compared to the Control site, suggesting that vehicular emissions probably affected this morphological feature. Similar findings were reported by Leghari and Zaidi (2013) and Shafiq et al. (2009) that the leaves taken from the polluted sites showed decline in leaf area. The reduced leaf area was not only an indication of retarded growth, but also was a morphological adjustment to highly polluted condition, on the grounds that with the smaller leaf size, the lesser will be the absorption of obnoxious gas (Zarinkamar et al., 2013). The reduction in the leaf area may be due to more amount of damaging pollutants like SO2 and NOx emitted by the automobiles, which incidentally affects cell elongation mechanism and photosynthetic capacity of the leaves (Wagh et al., 2006). It was striking to notice that in the leaves undergoing vehicular pollution visible morphological injuries was not seen. This result is consistent with the findings of Agrawal et al. (1991) and Salgare and Thorat (1995). Perhaps, hidden injuries or physiological disturbance could have resulted in reduction in the leaf proportions in all the four tree species (Terminalia catappa, Mangifera indica, Ficus platyphylla and Polyalthia longifolia) studied. It has been documented that plants experience physiological changes before displaying visible damage to leaves when exposed to air pollutants (Dohmen et al., 1990). In this study, Mangifera indica and Polyalthia longifolia showed significant decrease in the leaf area among the sites, hence it appears that these tree plants are susceptible to these pollutants. However, Terminalia catappa, Ficus platyphylla tends to be more resilient to the pollutants for the fact these species were not significantly different from each other. This is in conformity of Honour et al. (2009) that plants response to air pollutants differs between species.
Stomata are the gateway for exchange of gases, but it can be overburdened by air pollutants resulting in variations in the micro-morphological characteristics of the leaf. Stomata size, stomata number and stomata index were observed to be lower in the arterial roads, whilst epidermal cell number, epidermal cell length and trichome length were higher when compared to the control, an indication that automobile air pollutants most likely altered the micro-morphological parameters of the studied tree species. It was also observed that SO2 pollution could possibly lead to stomata size reduction and increased epidermal cell number in Ficus platyphylla. NO2 pollution could lead to increased epidermal cell number in Ficus platyphylla. CO, SO2, NO2 and VOC did not relate significantly to the stomata size, stomata number, epidermal cell number and trichome of Polyalthia longifolia. This suggests that the air quality parameters had minimal impact on the Polyalthia longifolia. The decrease in the stomata size in the leaves of the studied tree species suggests that stomata size allows a reduced amount of gas exchange between the environment and the trees, thus shielding the trees from intake of toxic gases. The decrease in the stomata size could be an avoidance mechanism against inhibitory effect of a pollutant on physiological activities such as photosynthesis (Pathak and Pancholli, 2014). These changes can be used as indicators of environmental stress (Pawar, 2016).
The reduction of leaf area in the arterial road sites brings about growth retardation and reduced surface area. Therefore, reduced surface area of the leaves of Terminalia catappa, Mangifera indica, Ficus platyphylla and Polyalthia longifolia possibly have capacity for a smaller number of stomata in this manner toxic pollutants entering the leaves are reduced. The decrease in stomata number arising from vehicular air pollutants is an indication that the quantity of gaseous pollutants entering the leaves via the stomata is reduced and the plant becomes tolerant to the pollution. Larcher (2003) reported that a slight modification for general gas exchange control and pollutant entry through the stomata singularly is initiated through reduction of pollutant uptake by plants by merely decreasing their stomata number. Stomata index is the ratio of stomatal number and number of both epidermal and stomatal cell. In this case, the dividend is the stomata number, while the divisor is summation of epidermal cell number and stomata number. When compared with the control, it indicates variations in number of stomata and in number of epidermal cells. It has been opined that reduction in stomatal index could be considered as a favourable adaptation to air pollution, as it might help in reducing the absorption of gaseous pollutants (Chauhan et al., 2004).
There was increased epidermal cell number in all the four studied tree species at the arterial road sites. The increase in epidermal cells could be a favourable adaptation in plants found in arterial road sites for pollutant detoxification. The epidermis is the major site air pollutants are first confronted upon by free radical scavengers (Pawar, 2016). Epidermal cells increase ensures greater quantity of antioxidants, thereby enhancing the detoxification of pollutants (Shah et al., 2000). There was increase in epidermal cell length at arterial road sites when compared with the control. In view of the fact that, differentiation of stomata mother cells involves the division of epidermal cells, reduction in stomata number is followed by epidermal cell size increase (Sant'anna-Santos et al., 2008). Trichomes length at the arterial road sites when compared with the control was found highest. Consequently, the increase in trichomes for Terminalia catappa, Ficus platyphylla and Polyalthia longifolia appears to be an added adaptation to air pollution stress. Trichomes helps to trap particulate matter falling directly on the leaf surface which incidentally block the stomata pore and adversely affect gaseous exchange processes (Sharma, 1977).
In this study, clogging of stomata pores resulting from the effect of vehicular pollution were noticed in all the four tree species. This could be due to high concentrations of particulate matter arising from vehicular emissions that is subsequently deposited on the stomatal pores. According to Das and Pattanayak (1978), aerosols larger than the pore of stomata are largely accumulated on the pore opening and thus interfere with exchange of gas, photosynthesis and a reduction in plant growth. Verma and Chandra (2014) had similar observations regarding the clogging of stomata on the leaves of Sida cordifolia and Catharanthus roseus receiving the burden of auto pollution. Relatedly, stomata clogging in the leaves of Citrus medica attributable to diesel exhaust were observed by Kaur (2004).
It appears that they were tree species differences based on the response of the trees to vehicular pollution.
The tree specific differences in its air pollution absorption agrees with Zhang et al. (2013) assertion that plants as living entity varies individually in their adaptations to the environment and abilities to absorb pollutants. The difference in plants species ability to lessen air pollution is attributed to the change in the leaf surface features namely cuticle, epidermis, stomata and trichomes (Neinhus and Barthlott, 1998). It was also interesting to note that the air pollutants were the most predictive variable for the stomata size and epidermal cell number variations in the studied tree species. This is consistent with the report that air pollutants such as SO2 make entrance into leaf tissues through the stomata (Tripathi and Mukesh, 2007). More so, increased epidermal cells ensures greater quantity of antioxidants, thereby enhancing the detoxification of pollutants (Shah et al., 2000), hence epidermis is the major site air pollutants are first confronted upon by free radical scavengers (Pawar, 2016).
SO2 had more effect on the micromorphological variables in the tree species except Polyalthia longifolia than the other pollutants. This is in keeping with Hill (1971) and Bennet andHill (1973-1975) report that plants have more preference for SO2 and less preference for NO2 to be absorbed and metabolized in the plant tissue. They outlined the plants preference for air pollutants in the following order HF > SO2 > Cl > NO2> O3> PAN > NO > CO. The high of SO2 uptake by plants is due to its high solubility and rapid hydration in the aqueous phase in the plant (Pfanz et al., 1987;De Kok, 1990).
CO and NO are ineffectively taken up by plants. It has been reported that CO and NO which are insoluble are ineffectively taken up by plants (Williams, 1990). The major sink for CO is the soil organisms and in the case of NO, when slowly taken up is converted to other forms that may be taken up readily (Baby and Goel, 1995). None of the predictive variables related significantly to the prediction of VOCs in all the plants, this could be due to emission of VOC´s by plants (Niederbacher et al., 2015).

Conclusions
In this study, the tree species response to automobile air pollution was by adjusting its leaf macro-and micro-morphological characteristics and such there was reduction in leaf area and stomata size, stomata number and stomata index as well as an increased epidermal cell number, length, and trichome length at the arterial road sites. In this study, stomata clogging with occluded stomata pores resulting from the effect of automobile pollution were noticed in all the four tree species. These alterations can be considered as indicators of environmental stress for initial revelation of urban air pollution Authors' Contributions Conceptualization: UNU and EJDB; UNU; Investigation: UNU; Supervision: EJDB; Writingoriginal draft: UNU; Writing -review and editing: EJDB. All authors read and approved the final manuscript. Both authors read and approved the final manuscript.