Fractionation and Biotoxicity of Heavy Metals in Urban ... - Springer Link

Environmental Geochemistry and Health (1998), 20, 185±198

Fractionation and Biotoxicity of Heavy Metals in Urban Dusts Collected from Hong Kong and London W.H. Wang1, M.H. Wong2*, S. Leharne1 and B. Fisher1 1

Department of Earth and Environmental Sciences, University of Greenwich, Medway Campus, Chatham, Kent ME4 4TB, England 2 Department of Biology and Institute for Natural Resource and Waste Management, Hong Kong Baptist University, KowloonTong, Hong Kong

Samples of urban dusts, road site dusts and car park dusts, were collected at two selected sites each in Hong Kong and London. Sequential extraction was used to characterise the chemical compositions of these urban dusts. Copper, lead, zinc, pH, electrical conductivity and organic content were measured. Biotoxicity tests of urban dusts were conducted on higher plants (Brassica chinensis and Lolium perenne), a dinoflagellate green alga (Dunaliella tertiolecta) and luminescent bacteria (Photobacterium phosphoreum). A significant correlation was found between total lead (r ÿ0:70, p < 0:01) and zinc (r ÿ0:74, p < 0:05), and the 20min ÿ EC50 using P. phosphoreum. In addition, there was a significant correlation (r ÿ0:72, p < 0:01) between the exchangeable lead content in dust and the 48 h-EC30 using D. tertiolecta. No specific trend was obtained for higher plants. Total lead and zinc contents were higher in dusts from London while the percentage of exchangeable fraction of metal contents was higher in those from Hong Kong. Keywords: Bioassay tests, chemical characterisation, Cu, Pb, sequential extraction, street dust, Zn.

Introduction High levels of heavy metals, such as copper (Cu), lead (Pb) and zinc (Zn), in urban street dusts have been reported in many countries (Harrison, 1979; Harrison et al., 1981; Lau and Wong, 1983; Harrison, 1992; Leharne, 1992; Ogunsola et al., 1994; Wang et al., 1995; Stone and Marsalek, 1996). While Cu and Zn are both essential elements to living organisms, their existence in high concentrations causes toxic effects. For example, typical symptoms of Cu phytotoxicity include chlorosis, reduced shoot and root growth, abnormal root development and wilting, while excessive Zn tends to antagonise the absorption or metabolism of other metals, especially Cu. Many forms of lead (Pb) particles are harmful and poisonous to animals and vegetation. In particular, elevated Pb level in street dusts (Farmer and Lyon, 1977) poses a potential human health hazard, especially to small children living in urban areas (Lepow et al., 1975; Duggan and Williams, 1977; Sayre, 1981; Royal Commission on Environmental Pollution, 1983). On the other hand, metals bound in urban street dusts might be transported and redistributed by storm runoff to receiving water. Stone and Marsalek (1996) reported that metals bound to urban sediment represented a large proportion of total metals exported to the aquatic environment and adversely affected benthic organisms and benthic *To whom correspondence should be addressed. 0269±4042 # 1998 Chapman & Hall

community structure. Therefore, urban street dusts are potential pollution sources in both urban terrestrial and aquatic environments. The mobility and transport of metals bound to urban dusts in the environment is a function of the chemical form of the element, which is governed by the physicochemical and biological characteristics of the environmental system. The measurement of the total metal content does not provide enough information to estimate the bioavailable metals because the biochemical and ecotoxicological significance of a metallic element are associated with the specific chemical form of the element and environmental factors (e.g. pH and temperature). The sequential extraction procedure developed by Tessier et al. (1979) provides a useful analytical method for the partitioning of particulate trace metals into five geochemical phases including exchangeable, carbonate, organic, hydrous iron and manganese oxide and residual fraction. Traditional chemical tests cannot ensure that all toxic chemicals of consequence will be identified and measured in the sediment of interest; nor can they be used to estimate synergistic effects among compounds in a mixture. Biological toxicity testing has, therefore, become an important tool in supplementing the traditional environmental monitoring programmes based on chemical tests to characterise complex chemical mixtures in sediments and water. Biological toxicity testing is based on exposing test

186

organisms to all the bioavailable chemicals in a test sample and then noting the changes in biological activity (Ribo et al., 1985). Sediment bioassays have been developed recently (Ross et al, 1988; Chapman, 1988; Burton and Scott, 1992). The selection of organisms for a biological toxicity test depends on the chemicals present in the sample, the prediction to be made, the time required and cost involved to perform a test. Algae have been widely used as toxicity test organisms because they are at the base of food chains, and algal bioassays are relatively simple, rapid and inexpensive compared to those using fish and invertebrates (Wong and Coulture, 1986). Furthermore, the algal toxicity test is widely adopted by various environmental regulatory bodies in North America and OECD countries (Wang and Freemark, 1995). Plant toxicity tests, especially seed germination and root elongation tests, have also been successfully used (e.g. Wong and Bradshaw, 1982; Wong and Lau, 1985; Baker et al., 1994) to screen metal tolerance of plants and to study metal toxicity to plants. Traditional bioassays for testing sediment and soil samples are performed on water or solvent extracts. This might underestimate or overestimate the exposure routes (Ongley et al., 1988; Kwan and Dutka, 1990). A newly developed Microtox1 Solid-phase Test using luminescent bacteria may eliminate these problems (Brouwer et al., 1990; Kwan and Dutka, 1992; Kwan, 1993) and may indicate true bioavailability of the tested samples. The luminescent bacteria bioassays are widely used for acute toxicity tests primarily because they are easy to perform and results can be obtained within a short period of time (Nohava et al., 1995). The present study involves taking dust samples from two of the world's busiest cities, namely London and Hong Kong, and analysing their heavy metal contents and biotoxicity. Only Pb, Cu and Zn are studied here because they are the major types of heavy metals found on urban surfaces exposed to road-traffic. London and Hong Kong are both densely populated urban centres (6.7 million in London, 6 million in Hong Kong) with similar heavy road-traffic densities but different climates. London, with an area of 1579 km2 , is located in the temperate zone of Europe with average temperatures between 5.58C and 18.18C, and an annual rainfall of 597 mm in 1991. Hong Kong, with an area of 1078 km2 , is located within the sub-tropic zone of Asia, with an average temperature of 23.18C, and 2344 mm of annual rainfall (Census and Statistics Department, HK, 1994). The pH of rainwater is 4.4±4.5 in London (Warren Spring Lab, 1988) and 4.6 in Hong Kong (W.H. Wang, unpublished data, 1996). High levels of dust, combined with other air pollutants (e.g., sulphur dioxide, nitrogen oxides and carbon monoxide) have rendered Hong Kong one of the most polluted areas in terms of air pollution in the region, threatening the health of the urban population and city aes-

W.H. Wang et al.

thetics. In fact, dust has been reported to be Hong Kong's number one air pollution problem (Environmental Protection Department, HK, 1995). The aim of this study is to compare the chemical form of metal contained in different urban dust samples from Hong Kong and London and to assess the toxicity of these urban dust samples using different trophic level organisms, Dunaliella tertiolecta (a dinoflagellate green alga), Brassica chinensis (Chinese white cabbage), Lolium perenne (rye grass) and Photobacterium phosphoreum (luminescent bacterium). Correlations between chemical data and the results of toxicity tests are also studied. Materials and methods Study sites and sampling The study sites were located at: (1) the busy road junction between Creek Road (23 300 vehicles a day) and Deptford Church Street (16 700 vehicles a day), London; (2) a 40-year-old school car-park (used as Depford Power Station's car park since the 1950s), Deptford Campus, University of Greenwich, Gonson Street, Deptford, London; (3) a section of Waterloo Road (45 000 vehicles a day) between Junction Road and Suffolk Road, Kowloon, Hong Kong; (4) a 20-year-old car-park at The Chinese University of Hong Kong, Shatin, Hong Kong. Street dusts and dirt were collected by gently sweeping the road surface with a plastic brush into a dustpan, taking care to collect the smaller particles that might be lost by suspension into the atmosphere. Samples were then placed in plastic bags and taken for analysis. The dust samples were collected in September 1995 from London, and in March 1996 from Hong Kong. The samples were oven dried at 408C for 7 days and then sieved into three different particle size ranges: 0±125 m, 125±250 m and 250± 500 m. Code numbers were assigned to the samples as follows: HDR1, HDR2, HDR3: Hong Kong road dust 0±125 m, 125±250 m, 250±500 m, respectively; HDC1, HDC2, HDC3: Hong Kong car-park dust 0±125 m, 125±250 m, 250±500 m, respectively; LDR1, LDR2, LDR3: London road dust 0±125 m, 125±250 m, 250±500 m, respectively; LDC1, LDC2, LDC3: London car park dust 0±125 m, 125±250 m, 250±500 m, respectively.

Fractionation and biotoxicity of heavy metals in urban dusts from Hong Kong and London

Chemical analysis Partitioning of metal contents in dust samples was conducted by using a sequential extraction method developed by Tessier et al. (1979) (Figure 1). The Cu, Pb and Zn concentrations of the solution were determined with a flame atomic absorption spectrophotometer. The solutions of dust samples were

187

also analysed for their pH, electrical conductivity (Orion pH/ISE Meter Model 920A), and organic content (loss-on-ignition: Allen (1989) ). Bioassay methodologies To test the effect of urban dust on seed germination and root growth of the two plants B. chinensis and

Figure 1 Sequential extraction procedure (modified from Tessier et al. (1979).

188

W.H. Wang et al.

L. perenne, elutriates from the urban dusts of 5%, 10%, 20%, and 40% (w/v) were prepared using distilled water and shaking at 150 rpm for 24 h before filtering through a 5C filter paper. Twenty seeds were placed on filter papers in Petri dishes with 1 ml of elutriate added every 2 days. Distilled water was used as a control. All the treatments were triplicated. The seeds were placed in a growth chamber with temperature 22 38C and a light/dark cycle of 16 h/8 h and light intensity of 4000 500 lux (lumen mÿ2 ) (TES 1332 Difital Lux Meter). The percentage of seed germination and root length were recorded at days 2, 4 and 7 for B. chinensis and at days 4 and 7 for L. perenne. The toxic effects of elutriates were assessed by the algal bioassay method modified from the standard ASTM STP 988 (Ross et al., 1988). Dunaliella tertiolecta, an unicellular green microscopic alga belonging to the estuarine and marine plankton was chosen. The algal stock was cultured in 50 ml of MAV enrichment medium (Droop, 1969). Screw type culture tubes of 25 mm 150 mm, each containing 10 ml test elutriate of different concentrations were used. The algal cultures were kept in suspension by an orbital shaker (110 rpm) and the initial inoculum density was kept at 5 104 cells mlÿ1 . The temperature was maintained at 25 38C with a 16 h/8 h light/dark cycle and a light intensity of 4500 500 lux. Elutriates were prepared by mixing one part of dust samples and four parts of artificial seawater (prepared from commercially available salt: Marinemix, manufactured by Wiegandt Ltd., Krefeld, Germany) following the methods described by Ankley (1991) and five concentrations of each dust sample elutriate, 5%, 20%, 40%, 60% and 80% were selected. The control was 0% of dust elutriate. Artificial seawater (Marinemix) was used as a diluent. The preparations of different concentrations of elutriates are listed in Table 1. All samples were treated in triplicate. At the end of 48 h and 96 h, the cell numbers in each tube were determined by haemocytometry. The percentage increase or decrease in cell numbers at different elutriTable 1 The preparation of different percentage of dust elutriates. Dust elutriate (ml) (One part of dust sample mixed with four parts of artificial seawater) Artificial seawater (ml) Algal culture (ml) 10 modified Johnson medium (ml)

0

0.5

2

4

6

8

8 1

7.5 6 1 1

4 1

2 1

0 1

1

1

1

1

1

1

Final volume (ml)

10

10

10

10

10

10

Final percentage of dust elutriate

0% 5% 20% 40% 60% 80%

ate concentrations was calculated (Wong and Couture, 1986). The EC30 values were estimated by probit analysis using SAS computer software. EC30 was used instead of the more common EC50 for the two higher plants and the green alga because the toxic effect of the elutriate samples in different concentrations was more easily revealed and distinguished in this way. The Microtox8 Solid-phase Test based on the Detailed Solid-phase Test Protocol (Microtox1 model 500, Microbics Co.) was used to evaluate the toxicity of dust samples using a species of marine bacteria, P. phosphoreum. Bacteria are sensitive indicators of contaminant stress and they can respond rapidly. The end-point used was 20±min EC50 . Results and Discussion The results of the chemical analyses are presented in Table 2. Figure 2 further illustrates the distribution of Cu, Pb and Zn at different geochemical phases in urban dusts. The results show that Pb and Zn were predominantly associated with the hydrous iron and manganese oxide fractions, while the smallest amount was in the carbonate fraction from all the dust samples collected from London and Hong Kong. Copper was predominantly distributed in the organic fraction, followed by the hydrous iron and manganese oxide fractions. Cheam and Gamble (1974) suggested that organic matter in soil formed more stable complexes with Cu. The results were in line with other investigations (Harrison et al., 1981; Hamilton et al., 1984). High levels of Pb and Zn were associated with the smallest particles in road dusts from London and Hong Kong, and car-park dusts from London only; but Cu did not have such a trend. The total Cu, Pb and Zn contents in dust from London were higher than those from Hong Kong. However, the percentage of exchangeable fractions of metals from the Hong Kong car-park dust was 1±10 times higher than those from London, and the percentage distribution in the exchangeable fractions in car-park dusts was higher than those in road dusts in both London and Hong Kong. This showed that the metal content of the car-park dust from Hong Kong was more mobile than that from London and carpark dusts were more mobile than road dusts. The road dusts were slightly alkaline (pH 8.33±8.79) and car-park dusts were neutral (pH 6.87±7.36) from both London and Hong Kong. The percentages of organic content in dust samples from Hong Kong were 1.5±3 times higher than those from London. Table 3 compares the results of the present study with other surveys conducted in London and Hong Kong. The Pb levels in both London and Hong Kong gradually fell from the 1980s to 1990s. This was due to the Pb content of leaded petrol being reduced from around 0.34 g Pb Lÿ1 to 0.143 g Pb Lÿ1 in January, 1986 and also the increase in the

Table 2 Chemical speciation of copper, lead and zinc using sequential extraction and chemical parameters of urban dust samples. London Sep 95

Hong Kong March 96

Road dust

Lead (g gÿ1 ) Exchangeable Carbonate Fe±Mn oxide Organic Residual Total Copper (g gÿ1 ) Exchangeable Carbonate Fe±Mn oxide Organic Residual Total Zinc (g gÿ1 ) Exchangeable Carbonate Fe±Mn oxide Organic Residual Total Chemical parameter EC(ms cmÿ1 ) Organic content (%) pH

Car park dust

LDR1

LDR2

18.65fg 9.07b 1206.83a 202.53ab 199.42a 1636.50a

20.37fg 7.38ced 774.58b 148.55bcd 78.35c 1029.23b

5.25ced 4.12b 90.40a 338.53a 74.05a 512.35a

4.73de 3.20bcd 67.27a 178.67abc 38.10bcd 291.97c

LDR3

LDC1

LDC2

Road dust LDC3

HDR1

Car park dust HDR2

HDR3

HDC1

HDC2

HDC3

25.32ced 27.17bcd 22.07ef def 7.10 11.82a 8.75bc c abc 514.82 748.12 421.25cd a ab 228.12 164.18 46.78ef 55.25c 155.43b 44.75cd c b 830.60 1106.72 543.60d

25.47ced 8.42bcd 565.73c 98.33ced 69.43cd 767.38c

30.63b 5.73fg 566.85c 83.73def 67.68c 754.63c

15.88g 5.82fg 412.02cd 43.88ef 49.05cd 526.65d

22.83def 5.68fg 286.35cd 21.08f 43.48cd 379.43e

36.48a 6.37efg 146.92e 12.55e 27.68d 230.00f

28.85bc 7.40ced 144.85e 10.77e 23.93d 215.80f

28.13bc 5.45g 135.40e 10.50f 28.33d 207.82f

5.40ced 2.13de 66.45a 173.73abc 37.17bcd 284.88c

5.42ced 4.07b 23.53b 139.40bc 61.38ab 233.80cd

7.87bc 4.20b 70.42a 277.85ab 31.65bcd 391.90b

4.30e 3.03bcd 23.78b 165.30abc 42.33abcd 238.75cd

4.58e 2.77ced 30.22b 327.05a 20.12cd 384.73b

7.85bcd 2.10de 27.10b 42.92c 12.15d 92.12f

8.87ab 1.97e 22.52b 65.22c 13.90d 112.47e

11.00a 2.25ced 40.60b 66.57c 11.85d 132.27e

137.43b 18.15b 932.53g 162.05ab 413.75a 1663.92c

1.762h 0.78e 2152.62b 158.30ab 67.50d 2396.82b

27.93h 28.17h 95.92ced 2.68de 2.75ff 7.75ced c e 1845.08 1569.17 372.50j abc bcd 135.57 84.72 47.15cd d d 53.82 46.70 51.12d 2065.08b 1731.50c 574.43e

71.33ef 14.23bc 426.46i 57.30cd 53.87d 623.19e

37.32gh 10.52bcd 437.75i 53.82cd 36.18d 575.59e

637.00c 9.58de 7.10b

931.00b 17.40a 8.64a

123.40g 14.19b 6.87c

96.10h 14.32b 6.87c

9.22ab 3.53e 7.35a 3.40b b 36.58 33.27b bc 116.57 50.70c 46.40abc 20.20cd 216.12d 111.10e

108.20c 13.17bc 2790.00a 179.42a 266.67c 3357.45a

78.55def 12.78bc 1717.50d 57.70cd 57.47d 1924.00bc

56.75f 11.65bc 1143.96f 33.67d 34.68d 1280.70d

179.50a 29.90a 1750.83d 182.08a 349.17b 2491.48b

100.85cd 18.22b 740.63d 56.52bcd 71.28d 987.49d

177.50e 9.12de 8.65a

154.80f 7.54e 8.69a

127.70g 4.84f 8.79a

1703.00a 10.43d 7.29b

130.10g 4.77f 7.36b

688.00c 16.97a 8.47a

500.00d 14.32b 8.33a

145.30f 13.30bc 6.93c

Superscript letters indicate any significant difference between measurements in each row according to Duncan's Multiple Range Test; in each row, any two measurements whose superscripts share one or more letters indicate no significant difference at p < 0:05. HDR1, HDR2, HDR3, Hong Kong road dust 0±125 m, 125±250 m, 250±500 m, respectively; HDC1, HDC2, HDC3, Hong Kong car park dust 0±125 m, 125±250 m, 250±500 m, respectively; LDR1, LDR2, LDR3, London road dust 0±125 m, 125±250 m, 250±500 m, respectively; LDC1, LDC2, LDC3, London car park dust 0±125 m, 125±250 m, 250±500 m, respectively.

190

W.H. Wang et al.

Figure 2 Distribution (ggÿ1 and percentage of different fractions) of copper, lead and zinc in urban dust samples. Exchangeable; carbonate; Fe±Mn oxide; organic; residual.

sales of unleaded petrol to about 60% in UK. On the other hand, Hong Kong had a gradual reduction of Pb content in leaded petrol in the 1980s (0.84, 0.6, 0.4, 0.25 and 0.15 g Pb Lÿ1 of petrol in 1980, 1981, 1983, 1985 and 1987, respectively) (Chan et al., 1989), and the sale of unleaded petrol now has a take-up rate of 77% since its introduction on 1 April 1991. It is commonly agreed that automobile exhaust is the most important source of Pb pollution in urban areas (Cantwell et al., 1972; Dinman, 1972; Ter Haar et al., 1972; Huntzicker et al. 1975; Lau and Wong, 1983.)

Table 4 shows the results of toxicity tests using the different living organisms: B. chinensis, L. perenne, D. tertiolecta, and P. phosphoreum. Figures 3 and 4 show the percentage root response of B. chinensis and L. perenne. This is rated as the average reduction of root length exposed to the toxicant during the test period as compared to the control. Brassica chinensis (Chinese white cabbage) was used here because it is an important economic crop and has a fast germination rate, within 4 days. Lolium perenne (rye grass) is an internationally recognised species for toxicity testing (Wong and Bradshaw, 1982). The value of EC30


191

Table 3 Comparison of total mean concentrations g gÿ1 of Cu, Pb and Zn in the street dusts between Hong Kong and London. Location

No. of sites Pb

Hong Kong Hong Kong Hong Kong Commercial Industrial Residential Rural Hong Kong Main road Car-park London Outer urban street Rural area London Urban major road London Side road Busy trunk road London Main and residential London Main road Industry/scrap metal Residential Garage repairers London Main road Car-park

Cu

Zn

Reference Ho (1979) Lau and Wong (1983) Chan et al. (1989)

2006 1627

92

1886

993 376 658 333

220 716 334 96

1418 1616 1513 586

1 1

652 233

296 86

2305 629

18 7

920 35

1

1914

112

1 1

1045 2346

181 231

Hamiliton et al. (1984) London

65

1354

115

Thornton et al. (1985)

6 4 13 1

978 713 656 2241

143 151 140 140

Leharne (1992)

1 1

897 1344

300 384

14 30

(effective concentration to reduce the root length of the control by 30%) was calculated form Figures 3 and 4. B. chinensis had a higher toxic effect at 96 h, but there was no specific trend of 96 h and 168 h for L. perenne. Figure 5 shows the percentage response of D. tertiolecta cells to dust elutriates of different dilution. The percentage response is calculated by comparing average reduction in cell number of alga exposed to the toxicant during the test period to the control. The method of calculation followed Wong and Couture (1986). For D. tertiolecta, the EC30 ranges for 48 h and 96 h were from 1.09% to 32.70% and from 4.12% to 69.93%, respectively.

This study (1996) Duggan and Willians (1977)

571

1866 2372

Harrison (1979)

This study (1995)

The root elongation method (Wilkins, 1957) using higher plants has been commonly used and refined by many researchers as a simple, rapid and easy test for toxic concentrations of trace metals. The main pathway of trace metal entry into a plant is from soil to root. In soil, metal ions can reach plant root surfaces by diffusion, provided there is a concentration gradient; by mass-flow, which can be induced by water-depleting processes in the plant itself such as transpiration and evaporation; or ion exchange between clay and a root in contact with it. The direct harmful effects of the water extract of roadside soils on the seed germination and root growth of

Table 4 The results of bioassays: numbers are percentage dilution of the elutriates of dust samples. LDR1 LDR2 LDR3 LDC1 LDC2 LDC3 HDR1 HDR2 HDR3 HDC1 HDC2 HDC3 DT 48 h-EC30 (%) DT 96 h-EC30 (%) BC 48 h-EC30 (%) BC 96 h-EC30 (%) BC 168 h-EC30 (%) LP 96 h-EC30 (%) LP 168 h-EC30 (%) Microtox 20 min-EC50 (%)

32.70 4.12 64.81 31.43 59.78 42.25 8.83 1.96

15.35 8.16 NT 8.45 33.40 31.68 56.83 2.31

23.25 5.49 87.17 4.14 14.36 NT 77.60 2.20

22.92 38.06 26.86 15.39 14.73 76.10 11.45 0.50

27.88 21.30 NT 4.25 27.32 25.76 NT 1.99

23.21 39.67 41.77 1.58 12.05 NT 13.27 2.43

1.09 60.66 24.81 20.67 20.43 15.21 16.04 0.79

28.61 69.93 14.58 5.11 27.40 18.95 57.60 2.69

17.62 62.78 66.88 4.08 10.77 7.39 26.25 3.41

1.33 14.59 49.28 5.27 13.27 43.12 34.09 4.48

22.49 34.92 48.67 13.45 67.52 51.06 72.50 6.33

24.69 61.37 33.19 3.37 18.72 NT NT 7.00

For abbreviations in column headings, see Table 2. DT, Dunaliella tertiolecta; BC, Brassica chinensis; LP, Lolium perenne. NT, effective concentration > 100%.

192

W.H. Wang et al.

Figure 3 Response of Brassica chinensis to dust elutriates of different dilutions (%). A positive value indicates the percentage inhibition in root growth, a negative value the percentage increase in root growth, i.e. a stimulating effect. * 96 hours; & 168 hours.

B. chinensis have been shown by Wong and Lau (1983). Heavy metals in soluble form are highly toxic to plants, e.g. Wong and Bradshaw (1982) showed that 0.02, 1.7 and 1.6 g mlÿ1 of Cu, Pb and Zn, respectively, inhibited 50% of the normal root growth of L. perenne. Algae act as primary producers in transferring energy to the higher trophic levels (Christensen and Scherfig, 1979) and form the first link in the food web, and the algal bioassay test is also commonly used to evaluate effluent toxicity.

The Microtox1 Solid-phase Test detects the overall toxic response to all pollutants in the dust samples. Data obtained from the Microtox1 Solid-phase Test are reported as that dilution of a sample elutriate which produces a 20 min-EC50 effect (50% decrease in the bacterial light output). Kwan and Dutka (1995) interpreted the EC50 values as follows: (1) values equal to or less than 0.5% are classified as very toxic, (2) values greater than 0.5% and equal to or less than 1.0% are moderately toxic, and (3) values greater than 1.0% as not toxic. The results of the Microtox1 SPT


193

Figure 4 Response of Lolium perenne to dust elutriates of different dilutions (%). For details, see Figure 3.

test of the present study 20 min-EC50 listed in Table 4 show that the toxic effect was associated with the small dust samples and these coincided with the results of chemical analysis. According to Kwan and Dutka's interpretation, the 20 min-EC50 of the finest car-park dusts from London (0.5%) and the finest road dusts from Hong Kong (0.79%) were very toxic and moderately toxic, respectively. Table 5 shows the correlation coefficients between metal concentrations and bioassay toxicity data of the dust samples. Correlation coefficients between total metal content, exchangeable fraction of metal

content, pH, organic content, EC and effective concentration of bioassays were also calculated. In this study, bioassays using higher plants, B. chinensis and L. perenne, and the unicellular alga, D. tertiolecta, in particular, mainly respond to toxicological effects of water soluble chemicals leaching from dust samples. For B. chinensis, significant correlation (p < 0:05) was obtained between 96 h-EC30 and total content of Cu, Pb and Zn. As for D. tertiolecta, a significant negative correlation (p < 0:01) was observed between 48 h-EC30 and the exchangeable Pb. D. tertiolecta may have internal detoxifying mechanisms, which may result in a tolerance to heavy metals (Stee-

194

W.H. Wang et al.

Figure 5 Response of Duniatella tertiolecta cells to dust elutriates of different dilutions (%). A positive value indicates the percentage inhibition in cell growth, a negative response the percentage increase in cell number, i.e. a stimulating effect. * 48 hours; & 96 hours.

mann-Neilsen and Kamp-Neilsen, 1970; Davies, 1976; Heuillet et al., 1986). In addition, certain algae protect themselves from toxic metals by extracellular excretions. There are two mechanisms of extracellular excretion: (1) the cell wall acts as a barrier which prevents metal uptake, and (2) the cell excretes organic material, which chelates metals thus preventing metal uptake (Fogg and Westlake,1955; Butler et al., 1980; Kaplan et al.,1987). However, there are no other statistically significant correlations between toxic values and dissolved metal contents. For the

algal bioassay, the relationship between metal sorption to algal cells and toxicity is still not clear. The Microtox1 EC50 values show significant negative correlations with total Pb and Zn contents in the dust samples. Microtox1 has been demonstrated as the most sensitive assay for detecting of Cu, Zn and Hg (Dutka and Kwan, 1981, 1984; Dutka et al., 1983; Dutton et al., 1986; Codina et al., 1993). These results seem to suggest that Pb and Zn are two of the toxic pollutants in urban dusts causing the suppres-

Table 5. Correlation coefficients between chemical data and bioassay data

Cu(ex) Cu(t) Pb(ex) Pb(t) Zn(ex) Zn(t) E.C. Organic content pH DT 48 h-EC30 DT 96 h-EC30 BC 48 h-EC30 BC 96 h-EC30 BC 168 h-EC30 LP 96 h-EC30 LP 168 h-EC30 Microtox EC50

Cu(ex) Cu(t)

Pb(ex)

Pb(t)

Zn(ex) Zn(t)

EC

Organic C. pH

DT 48 h DT 96 h BC 48 h BC 96 h BC 168 h LP 96 h LP 168 h Microtox

ÿ1.00 ÿ0:36 ÿ0.68* ÿ0:30 ÿ0.06 ÿ0:30 ÿ0.21 ÿ0.43 ÿ0:56 ÿ0:27 ÿ0.25 ÿ0:54 ÿ0.14 ÿ0.01 ÿ0.31 ÿ0.11 ÿ0.48

ÿ1.00 ÿ0:46 ÿ0.08 ÿ0:51 ÿ0.04 ÿ0.24 ÿ0:56 ÿ0:72** ÿ0:02 ÿ0:31 ÿ0:09 ÿ0:27 ÿ0.19 ÿ0:04 ÿ0.31

ÿ1.00 ÿ0.45 ÿ0.86*** ÿ0.27 ÿ0:44 ÿ0.55 ÿ0.31 ÿ0:48 ÿ0.16 ÿ0.67* ÿ 0.23 ÿ0:05 ÿ0:45 ÿ0:70*

ÿ1.00 ÿ0.20 ÿ0.37 ÿ0:50 ÿ0:41 ÿ0.24 ÿ0:47 ÿ0.10 ÿ0.15 ÿ0.01 ÿ0.17 ÿ0:23 ÿ0:33

ÿ1.00 ÿ0.27 ÿ0.01 ÿ0:10 ÿ0.41 ÿ0:45 ÿ0.21 ÿ0:35 ÿ0:12 ÿ0:48 ÿ0:58*

ÿ1.00 ÿ0:10 ÿ0:38 ÿ0.79** ÿ0:72** ÿ0.10 ÿ0.00 ÿ0:30 ÿ0:24 ÿ0.32

ÿ1.00 ÿ0:07 ÿ0.11 ÿ0.04 ÿ0.37 ÿ0.25 ÿ0.31 ÿ0.04

ÿ1.00 ÿ0:45 ÿ0.71* ÿ0:17 ÿ0.84*** ÿ0.16 ÿ0.02 ÿ0.80* ÿ0.07 ÿ0:04 ÿ0:08 ÿ0.62* ÿ0.13 ÿ0:22 ÿ0:56 ÿ0:52

ÿ1.00 ÿ0.50 ÿ0.00 ÿ0.64* ÿ0.19 ÿ0:03 ÿ0:12 ÿ0.71* ÿ0.15 ÿ0:31 ÿ0:62* ÿ0:74**

ÿ1.00 ÿ0.01 ÿ0:11 ÿ0.19 ÿ0.33 ÿ0.04 ÿ0:29 ÿ0:34 ÿ0:56

ÿ1.00 ÿ0:57 ÿ0:22 ÿ0:29 ÿ0:08 ÿ0.05 ÿ0.18

ÿ1.00 ÿ0:17 ÿ0.11 ÿ0:20 ÿ0.40 ÿ0:17

ÿ1.00 ÿ0.60 ÿ0:38 ÿ0:43 ÿ0:33

DT, Dunaliella tertiolecta; BC, Brassica chinensis; LP, Lolium perrene; (ex), exchangeable fraction of metal content; (t), total metal content. Significant differences *p < 0:05, **p < 0:01 ***p < 0:001, according to Student's t-test.

ÿ1.00 ÿ0:31 ÿ0.00 ÿ0.24

1.00 0.25 0.26

1.00 0.47

1.00

196

sive effect on P. phosphoreum. There is no significant correlation between different bioassays. Ahlf et al. (1989) suggested that the comparability of bioassays is limited by the different test sample sensitivities of test organisms. Conclusions This study consisted of chemical speciation of the metals bound in urban dusts from London and Hong Kong and biological toxicity tests using an alga, two higher plants and a luminescent bacteria, and the correlation between chemical data and bioassay data. Lead and Zn were predominantly associated with the hydrous iron and manganese oxide fraction, while Cu was predominantly distributed in the organic fraction. In general, metal contaminant levels depended on the particle size, with smaller particles having a higher toxic effect. There was a higher metal content in London dust, but a high percentage of the exchangeable fraction was seen in Hong Kong car-park dust. The Microtox1 Solidphase Test showed a significant negative correlation with metal contents in dusts. There were no specific trends found in the alga and higher plants tests between different dust samples. Acknowledgments The authors thank Mr K C Pun, Mr W K Ip, Mr K W Chan and Mr K K Ma of Hong Kong Baptist University for technical assistance and Dr R Y H Cheung of City University of Hong Kong for his permission to use the Microtox1. The research project was financed by a Faculty Research Grant of Hong Kong Baptist University awarded to M. H. Wong. The senior author, W.H. Wang would like to thank the Hong Kong British Council for the award of a travel grant. This paper will form part of W.H. Wang's PhD thesis, registered at the University of Greenwich. References Ahlf, W., Calmano, W., Erhard, J. and Forstner, U. 1989. Comparison of five bioassay techniques for assessing sediment-bound contaminants. Hydrobiologia, 188/189, 285±289. Allen, S.E. (ed.) 1989. Chemical Analysis of Ecological Materials, 2nd Ed. Blackwall Scientific, Oxford. Ankley, G.T. 1991. Predicting the toxicity of bulk sediments to aquatic organisms with aqueous test fractions: pore water vs. elutriate. Environmental Toxicology and Chemistry, 10, 1359±1366. Baker, A.J.M., Reeves, R.D. and Hajar, A.S.M. 1994. Heavy metal accumulation and tolerance in British populations of the metallophyte Thlaspi caerulescens J. & C. Presl (Brassicaceae). New Phytologist, 127, 61±68. Brouwer, H., Murphy, T.P. and McArdle, L. 1990. A sediment-contact bioassay with Photobacterium phosphoreum. Environmental Toxicology and Chemistry, 9, 1353±1358.

W.H. Wang et al. Burton Jr, G. A. and Scott, K.J. 1992. Sediment toxicity evaluations: their niche in ecological assessments. Environmental Science & Technology, 26, 2068±2075. Butler, M., Haskew, A.E.J. and Young, M.M. 1980. Copper tolerance in the green alga, Chlorella vulgaris. Plant, Cell and Environment, 3, 119±126. Cantwell, E.N., Jacobs, E.S., Gunz, W.G. Jr and Liberi, V. E. 1972. Control of particulate lead emissions from automobiles. In: M.G. Curry and G.M. Giliott (eds.) Cycling and Control of Metals, pp. 95±107. USEPA, Cincinnati, Ohio. Census and Statistics Department, HK 1994. Hong Kong Annual Digest of Statistics. The Government Printer, Hong Kong. Chan, G.Y.S., Chui, V.W.D. and Wong, M.H. 1989. Lead concentration in Hong Kong roadside dust after reduction of lead level in petrol. Biomedical and Environmental Science, 2, 131±140. Chapman, P.M. 1988. Marine sediment toxicity tests. In: Chemical and Biological Characterization of Sludges, Sediments, Dredged Spoils and Drilling Muds, ASTM STP 976, pp. 391±402 American Society for Testing and Materials, Philadelphia. Cheam, V. and Gamble, D. S. 1974. Metal fulvic acid chelation equilibrium in aqueous NaNO3 solution: Hg(II), Cd(II) and Cu(II) fulvate complexes. Canadian Journal of Soil Science, 40, 413. Christensen, E.R. and Scherfig, J. 1979. Effects of manganese, copper and lead on Selenastrum capricormutum, Chlorella stigmatophora. Water Research, 13, 79. Codina, J.C., Perez-Garcia, A., Romero, P. and de Vicente, A. 1993. A comparison of microbial bioassays for the detection of metal toxicity. Archives of Environmental Contamination and Toxicology, 25, 250±254. Davies, A.G. 1976. An assessment of the basis of mercury tolerance in Dunaliella tertiolecta. Journal of the Marine Biological Association of the United Kingdom, 56, 39±57. Dinman, B.D. 1972. Airborne lead in perspective. In: Proceedings of the International Symposium, on Environmental Health Aspects of Lead, Amsterdam, V2, pp. 45±57. Droop, M.R. 1969. In: J.R. Norris and D.W. Ribbons (eds.), Methods in Microbiology SB, Academic Press, London. Duggan, M.J. and Williams, S. 1977. Lead in dust in city streets. Science of the Total Environment, 7, 91. Dutka, B.J. and Kwan, K.K. 1981. Comparison of three microbial toxicity screening tests with the Microtox test. Bulletin of Environmental Contamination and Toxicology, 27, 753±757. Dutka, B.J. and Kwan, K.K. 1984. Studies on a synthetic activated sludge toxicity screening procedure with comparison to three microbial toxicity tests. In: D. Liu, and Dutka, B.J. (eds), Toxicity Screening Procedures Using Bacterial Systems, pp. 125±138. Marcel Dekker, New York. Dutka, B.J., Nyholm, N. and Peterson, J. 1983. Comparison of several microbiological toxicity screening tests. Water Research, 17, 1363±1368. Dutton R.J., Bitton, G. and Koopman, B. 1986. Rapid test for toxicity in wastewater systems. Toxicity Assessment, 1, 147±158. Environmental Protection Department, HK 1995. Environment Hong Kong 1995. The Government Printer, Hong Kong.

Fractionation and biotoxicity of heavy metals in urban dusts from Hong Kong and London Farmer, J.G. and Lyon, T.D.B. 1977. Lead in Glasgow street dust and soil. Science of the Total Environment, 8, 83±89. Fogg, G.E. and Westlake, D.F. 1955. The importance of extracellular products of algae in freshwater. Verhandl Intern Ver Limnology, 12, 219±231. Hamilton, R.S., Revitt, D.M. and Warren, R.S. 1984. Levels and physico-chemical association of Cd, Cu, Pb and Zn in road site sediments. Science of the Total Environment, 33, 59±74. Harrison, R.M. 1979. Toxic metals in street and household dusts. Science of the Total Environment, 11, 89±97. Harrison, R.M., Laxen, D.P.H. and Wilson, S.J. 1981. Chemical associations of lead cadmium, copper and zinc in street dust and roadside soils. Environmental Science & Technology, 15, 1378±1381. Harrison, S.G. 1992. Sorption of trace metals by urban dust and dirt: measurement and characterization. MSc Thesis. Pennsylvania State University. Heuillet, E., Moreau, A., Halpern, S., Jeanne, N. and Puiseux-Dao, S. 1986. Cadmium binding to a biomolecule in vacuoles of Dunaliella tertiolecta contaminated with CdCl2 : electron probe microanalysis. Biology of the Cell, 58, 79±86. Ho, Y.B. 1979. Lead, copper and manganese in street dust in Hong Kong. Journal of Asia Ecology, 1, 95±101. Huntzicker, J.J., Friedlander, S.K. and Davidson, C.I. 1975. Material balance for automobile emitted lead in Los Angeles Basin. Environmental Science & Technology, 9, 448±456. Kaplan, D., Christiaen, D. and Arad, S. 1987. Chelating properties of extracellular polysaccharide from Chlorella spp. Applied and Environmental Microbiology, V53, 12, 2953±2956. Kwan, K.K. 1993. Technical methods section: direct solid phase toxicity testing procedure. Environmental Toxicology and Water Quality, 8, 345±350. Kwan, K.K. and Dutka, B.J. 1990. Simple two-step sediment extraction procedure for use in genotoxicity and toxicity bioassays. Toxicity Assessment, 5, 395±404. Kwan, K.K. and Dutka, B.J. 1992. Evaluation of toxichromotest direct sediment testing procedure and Microtox1 Solid-phase testing procedure. Bulletin of Environmental Contamination and Toxicology, 49, 656±662. Kwan, K.K. and Dutka, B.J. 1995. Comparative assessment of two solid-phase toxicity bioassay: the direct sediment toxicity testing procedure (DSTTP) and the Microtox1 Solid-phase Test (SPT). Bulletin of Environmental Contamination and Toxicology, 55, 338±346. Lau, W.M. and Wong, M.H, 1983. The effect of particle size and different extractants on the contents of heavy metals in roadside dusts. Environmental Research, 31, 229±242. Leharne, S. 1992. A survey of metal levels in street dusts in an inner London neighbourhood.. Environmental International, 18, 263±270. Lepow, M.L., Bruckman, L., Gillette, M., Markowitz, S., Robino, R. and Kapish, J. 1975. Investigation into sources of lead in the environment of urban children. Environmental Research, 10, 415±426. Nohava, M., Vogel, W.R. and Gaugitsch, H. 1995. Evaluation of the luminescent bacteria bioassay for the estimation of toxicological potential of effluent water samples ± comparison with data from chemical analysis. Environmental International, 21, 33±37. Ogunsola, O.J., Oluwole, A.F., Asubiojo, O.I., Olaniyi, H.B., Akeredolu, F.A., Akanle, O.A., Spyrou, N.M.,

197

Ward, N.I. and Ruck, W. 1994. Traffic pollution: preliminary elemental characterisation of roadside dust in Lagos, Nigeria. Science of the Total Environment 146/ 147, 175±184. Ongley, E.E., Birkholz, D.A., Carey, J.H. and Saoiloff, M.R. 1988. Is water a relevant sampling medium for toxic chemicals? An alternative environmental sensing strategy. Journal of Environmental Quality, 17, 391±401. Ribo, J.M., Zaruk, B.M., Hunter, H. and Kaiser, K. 1985. Microtox toxicity test results for water samples from the Detroit River. Journal of the Great Lake Research, 11, 297±304. Ross, P., Jarry, V. and Sloterdijk, H. 1988. A rapid bioassay using the green alga Selenastrum capricornutum to screen for toxicity in St. Lawrence river sediment elutriates. In: Functional Testing of Aquatic Biota for Estimating Hazards of Chemicals, J. Cairns and J.R. Pratt (eds.) pp. 68±73. American Society for Testing and Materials, Philadelphia. Royal Commission on Environmental Pollution 1983. Lead in the environment, nine report, HMSO, London. Sayre, M.D. 1981. Dust lead contribution to lead in children. In: D.R. Lyaman, L.G. Piantanids and J.F.Cole (eds.) Environmental Lead, New York: Academic Press. Steemann-Neilsen, E. and Kamp-Neilsen, L. 1970. Influence of deleterious concentration of copper on the growth of Chlorella pyrenoidosa. Physiologia, 22, 1121± 1131. Stone, M. and Marsalek, J. 1996. Trace metal composition in street sediment: Sault site. Marie, Canada. Water, Air and Soil Pollution, 87, 149±169. Ter Haar, G., Lenare, D.L., Hu, J.N. and Brandt, M. 1972. Composition, size and control of automotive exhaust particulates. Journal of Air Pollution Control Association, 22, 39±46. Tessier, A., Campbell, P.G.C. and Bisson, M. 1979. Sequential extraction procedure for the speciation of particulate trace metals. Analytical Chemistry, 51, 844±851. Thornton, I., Culbard, E.B., Moorcroft, S., Watt, J., Wheatley, M. Thompson, M., and Thomas, J.F.A. 1985. Metals in urban dusts and soils. Environmental Technology Letter, 6, 137. Wang, W. and Freemark, K. 1995. The use of plants for environmental monitoring and assessment. Ecotoxicology and Environmental Safety, 30, 289±301. Wang, W.H. 1996. The measurement of the pH of rainwater in Kowloon ( unpublished data). Wang, W.H., Leharne, S., and Fisher, B. 1995. The characterisation of urban dust and dirt. In: G.D. Fowler (ed.), Current Trends in Contaminated Land Research Proceedings of the SCI Symposium, London, December 95. Warren Spring Laboratory 1988. United Kingdom acid rain monitoring, Stevenage. Wilkins, D.A. 1957. A technique for measurement of lead tolerance in plant. Nature, 180, 37±38. Wong, M.H. and Bradshaw, A.D. 1982. A comparison of the toxicity of heavy metals, using root elongation of rye grass, Lolium perrenne. New Phytologist, 91, 255± 261. Wong, M.H. and Lau, W.M. 1983. Effects of roadside soil extracts on seed germination and root elongation of edible crops. Environmental Pollution, 31, 203±215. Wong, M.H. and Lau, W.M. 1985. Root growth of Cynodon dactylon and Eleusine indica collected from

198 motorways at different concentration of lead. Environmental Research, 36, 257±267. Wong, P. T. S. and Couture, P. 1986. Toxicity screening using phytoplankton. In: G. Bitton and B.J. Dutka,

W.H. Wang et al. (eds.) Toxicity Testing Using Microorganisms Vol. 2, pp. 79±100. CRC Press, Bica Raton, FL, USA. [Manuscript No. 457: Submitted January 20, 1997 and accepted after revision July 30, 1997.]