www.elsevier.com/locate/scitotenv

Science of the Total Environm

An ecotoxicity assessment of contaminated forest soils from

the Kola Peninsula

Graeme I. Patona, Ekaterina Viventsova (Ruth)b, Jurate Kumpenec,

Michael J. Wilsona, Hedda J. Weitza, Julian J.C. Dawsona,*

aSchool of Biological Sciences, University of Aberdeen, Cruickshank Building, St Machar Drive, Aberdeen, AB24 3UU, UKbDivision of Applied Geology, Department of Chemical Engineering and Geosciences,

Lulea University of Technology, SE-971 87 Lulea, SwedencDivision of Waste Science and Technology, Department of Civil and Environmental Engineering,

Lulea University of Technology, SE-971 87 Lulea, Sweden

Received 10 August 2004

Available online 1 June 2005

Abstract

Point source copper and nickel contamination emanating from smelters of the Kola Peninsula, NW Russia, has been

observed since the mid-1960s. Previous studies have concentrated on the spatial distribution of heavy metals and their effects on

forest ecology and indigenous mammals and birds. Soil is perceived as the major repository for the metal pollutants but there is

a need to link the soil concentration of pollutants on the Kola Peninsula with biological parameters. Many standard methods

currently used in soil ecotoxicology are developed and refined with artificial amendments and rarely modified for use in

historically contaminated environments. In this study, forest soils were sampled along a 34 km transect from the smelter and

analysed both chemically and with a range of ecologically relevant biological tests. Soil respiration, total nematode count,

microbial heterotrophic numbers and minimal inhibitory concentrations to copper and nickel were carried out on bulk soil. The

soil pore water was tested with bacterial and fungal bioluminescence-based biosensors. The heterotrophic numbers and their

inhibitory concentration showed strong correlation with heavy metal concentrations while decreasing biosensor luminescence

was related to increasing copper concentrations present in the pore waters. Overall, there were considerable impacts on some

microbial parameters but other measures including respiration and nematode populations were insensitive to pollutant levels.

While chemical analysis of heavy metals proved essential in defining the extent of contamination, environmentally relevant

ecotoxicological tests complemented these data by demonstrating pollutant impact. Ecotoxicological approaches that study both

the bulk soil and pore water may represent the key to understanding the fate of heavy metal in soils.

D 2005 Elsevier B.V. All rights reserved.

Keywords: Nickel; Copper; Microbial biosensors; Soil ecotoxicity; Kola Peninsula

0048-9697/$ - s

doi:10.1016/j.sc

* Correspondin

E-mail addre

ent 355 (2006) 106–117

ee front matter D 2005 Elsevier B.V. All rights reserved.

itotenv.2005.04.036

g author. Tel.: +44 1224 272259; fax: +44 1224 272703.

ss: j.j.dawson@abdn.ac.uk (J.J.C. Dawson).

G.I. Paton et al. / Science of the Total Environment 355 (2006) 106–117 107

1. Introduction

Forest soils of the Kola Peninsula in NW Russia

have historically been contaminated with copper and

nickel produced from smelters first established in the

1930s (Kozlov and Barcan, 2000; Perkiomaki et al.,

2003). However, it was not until the mid-1960s that

the effects of their contamination began to be studied

in detail. These studies focussed on mapping and

determining heavy metal (HM) concentrations in

soils, edible berries, mushrooms and snow, as well

as the effects on forests, small mammals and birds in

the surrounding area (Kozlov and Barcan, 2000). To

date, little work in the Kola Peninsula has been carried

out on interpreting the link between contaminant con-

centration and the effect on soil biology.

Traditional analytical approaches that measure the

total or extractable chemical concentrations of HM

provide information on their general distribution but

not on the impacts that they may have on a given

ecosystem. In isolation, these data may be of limited

use for assessing environmental significance. Bio-

availability of HM in soil can be determined using

chemical methods (Nolan et al., 2003) or bioassays

that use living organisms (Spurgeon et al., 1994;

Korthals et al., 1998, 2000; McGrath et al., 1999);

these procedures complement conventional chemical

quantification.

Soil toxicity tests are becoming widely adopted in

environmental monitoring. Some of these tests have

developed traditional soil techniques, while others are

based upon novel biotechnological methods. Howev-

er, it is important that the selected bioassays reflect the

ecology of the particular soil ecosystem under study.

Traditional methods, which use bulk soils, study

key soil nutrient cycling such as the mineralisation of

carbon, nitrogen, phosphorus or sulphur. These meth-

ods have been critically reviewed (van Beelen and

Doelman, 1997). Some of these techniques, which are

well suited to agricultural soils, may not respond to

the generally more conserved nutrient balance of for-

est soils (The Royal Society, 1983). However, it has

been reported that decreases in respiration rates may

indicate the presence of gross HM pollution in forest

soils (Bewley and Stotzky, 1983; Laskowski et al.,

1994; Fritze et al., 1996).

There are many methods for determining the mi-

crobial biomass or composition of soils, but each

technique represents a compromise between relevance

to the study, ease of application, sample throughput

and cost-effectiveness. Dehydrogenase-based meth-

ods have been identified as ecologically appropriate

for the testing of forest soils after pollution incidents

(Ajungla et al., 2003). For this reason, a simple tetra-

zolium-based method was adapted from the standard

assay for use in this study.

The prolonged exposure to a given HM may result

in the selection of a tolerant HM microbial population,

which can vary considerably from the original micro-

bial community derived over time in a previously

unpolluted environment. As many bacterial and fun-

gal species have evolved different mechanisms to

survive in HM contaminated soils, a measurement of

the minimum inhibitory concentration (MIC) of a

community can indicate the presence and extent of

this population (Trevors et al., 1985; Klerks and Weis,

1987; Cervantes and Guiterrez-Corona, 1994). This

can be key to understanding the spatial and temporal

impact of HM.

Earthworms, which are commonly used in soil

ecotoxicology studies, are not present in the higher

latitude podzolic soils of this study, hence nematodes

were used as an appropriate meiofuanal group. Previ-

ous studies have reported the relative sensitivity of

nematode numbers to HM contamination (Siciliano

and Roy, 1999; Korthals et al., 2000; Ellis et al.,

2002). Indirect effects on both number and type of

nematodes can be a response to an HM-mediated

microbial community or biomass changes.

The tests considered so far employ the use of bulk

soil. Recently, there has been increasing use made of

soil water extracts for toxicity assessments. Previous

research has shown that single species assays using

microbial-based biosensors are applicable to a wide

range of soil types (Campbell et al., 2001). Moreover,

the responses of both bacterial and fungal biosensors

correspond to the bioavailable fraction of the contam-

inant present (McGrath et al., 1999; Tiensing et al.,

2001; Weitz et al., 2002).

The aim of this work was to select a historically

contaminated soil transect on the Kola Peninsula and

to chemically characterise the soils. Once charac-

terised, the performance of a range of soil ecotoxico-

logical tests was compared to enable an estimation of

the effect of the HM pollution to be made. The final

aim was to recommend the most suitable tests for

G.I. Paton et al. / Science of the Total Environment 355 (2006) 106–117108

appraising the impact of HM pollutant on the Kola

Peninsula.

2. Materials and methods

2.1. Study area

The dSeveronickelT nickel-smelting complex, offi-

cially opened in 1937, is located in the central part of

the Kola Peninsula, NW Russia, south of Monche-

gorsk City (Kozlov and Barcan, 2000). It is recog-

nised that the Kola Peninsula is a significant source of

environmental pollution with annual emissions of sul-

phur dioxide and HM, mainly copper and nickel,

amongst the highest in Europe (Arctic Pollution,

2002; Ruth-Balaganskaya and Kudrjavtseva, 2002).

According to the report of the Environmental Com-

mittee of Murmansk Region (2000), emissions from

industries in the area in 2000 were 274,000 t for

sulphur dioxide, 1080 t for copper and 1570 t for

nickel.

The Kola Peninsula ecosystem belongs to the

northern taiga zone. The original (undisturbed) veg-

etation type in the area is spruce (Picea abies) and

pine (Pinus sylvestris) forest with Vaccinium species

(V. vitis-idaea and V. myrtillus) and lichens in the

ground layer. The predominant soil type is Podzolic

as characterised by a distinct profile with a ca. 5–10

cm deep organic horizon (Ah) with black–brown raw

organic material (pH 5.5), a 6–10 cm light grey

eluvial (E) horizon (pH 4.5) and a 30–40 cm red–

brown sandy illuvial (Bh) horizon (pH 5.0).

Sampling took place during the autumn of 2001.

The sampling sites were located at distances of 2, 10,

11, 17, 22, 27 and 34 km south of the dSeveronickelTsmelting complex, along the predominant wind di-

rection. Sampling sites were about 5 m in diameter

and were selected to have a homogeneous topogra-

phy. Three 1-kg samples were taken at each sam-

pling area and analysed independently. Soil was

initially sieved to 2 mm and stored at 4 8C until

analysis within a month of sampling. Soil samples

were collected from the Ah horizon on all the sam-

pling sites except for those nearest to the smelter

complex, where samples were taken from the upper

3 cm because there was no pedogenic evidence

beneath this depth.

2.2. Chemical analysis

2.2.1. Copper and nickel concentrations

Soil samples for determining total concentrations

of copper and nickel were ground to pass through a

0.25 mm sieve. The HM content was determined in

air-dried samples by X-ray fluorescence spectrometry

(Niton 700 XRF).

Pore water was extracted in a 1:2 soil:water mix-

ture by adding 20 ml of deionised water to 10 g dry

weight of soil (Flynn et al., 2002). The mixtures were

then shaken for 2 h on an end-over-end shaker at 50

rpm. To extract the water, samples were centrifuged

(Coolspin 2 RR/1061) for 10 min at 1050�g at 4 8C.A 10 ml aliquot of the supernatant was retained for

biosensor testing, while the HM concentrations were

determined by flame atomic absorption spectropho-

tometry (Perkin Elmer AAnalyst 100) following acid-

ification with 2% HNO3.

2.2.2. Soil water pH and total organic matter

Soil pH values were measured in water with a

combined glass/reference electrode. Total organic

matter (OM) content was determined using a dry

combustion method; 0.5–1 g of air-dried samples

were pressed into pellets and burned under an O2

pressure of 20 bar in a Gallenkamp Autobomb

calorimeter.

2.3. Ecotoxicological analysis

2.3.1. Respiration

Replicate 1.5-g soil samples were placed in 30 ml

sealed vials. The vials were incubated for 24 h at 15 8Cand the headspace analysed for CO2 concentrations

using a gas chromatograph (Chrompack 9001) with

attached methanizer and flame ionisation detector

(FID) following calibration with certified gas stan-

dards (Linde Gases, Aberdeen). A sample of the head-

space was removed from the vial using a syringe and

injected into the GC flow injection loop (250 Al)system with a nitrogen carrier gas (20 ml min�1)

onto a 80/100 mesh Poropak Q column (2 m�1 /

8WOD�2 mm). The oven temperature remained con-

stant at 100 8C and the FID temperature was set at

250 8C. Respiration values were determined on a mg

CO2 g soil�1day�1 basis following subtraction of a

blank vial containing atmospheric CO2 only.

G.I. Paton et al. / Science of the Total Environment 355 (2006) 106–117 109

2.3.2. Total microbial heterotrophic numbers and min-

imum inhibitory concentration (MIC) values

The total number of aerobic heterotrophs was

enumerated by the most probable number (MPN)

method modified from that described by Braddock

and Catterall (1999). Tryptone soya broth was added

as a carbon source (Lawlor et al., 2000) to a minimal

salt medium supplemented with a trace elements

solution (Dorn et al., 1974), prior to the addition

of 0.2% (w/v) p-iodonitrotetrazolium violet (INT)

dissolved in 0.5% N-tris[hydroxymethyl]methyl-2-

aminoethane-sulfonic acid (TES) buffer. Soil (10 g)

was added to 100 ml of deionised water in 250 ml

conical flasks, shaken for 10 min on a wrist-action

shaker and 50 Al aliquots used to inoculate 96-well

microtitre plates containing 100 Al of the media with

the carbon source, INT and TES. Following serial

dilutions, the plates were sealed and incubated un-

disturbed at 22 8C for 14 days. After incubation,

positive wells, indicated by a purple colouration,

were used to determine the MPN of microorganisms

ml�1. Calculations were performed using the MPN

calculator (Version 4.0) computer program (Klee,

1993). The data has been optimised and standardized

against a known cfu count (Paton et al., 2003).

To assess the MIC values of copper and nickel, the

same MPN method was applied but the wells were

supplemented with a gradient of HM amendments and

only the dilution corresponding to 106 MPN was used

as the inoculum. Copper and nickel were added as

NO3� salts to give final well concentrations of 0.005,

0.01, 0.05, 0.1, 0.5, 1.0, 5.0, 10 and 50 mM. Sterile

control samples verified the absence of false positive

results.

2.3.3. Total nematode count

Extraction of nematodes from the soil samples was

based on methods outlined in Hooper (1986). The soil

(20 g) was spread uniformly onto sieves covered with

tissue paper and placed over a funnel. A rubber tube

was fitted to the narrow end of each funnel and a

clamp connected to each tube. Water was added to

each funnel to soak but not flood the soil and left for

48 h. The nematodes settled at the bottom of the

rubber tubing at the clamp. After 48 h, 20 ml of the

water was collected in a large Petri dish from each

funnel by loosening the clamp on the rubber tube and

the nematodes were counted under a microscope using

a tally counter. Data were then square root trans-

formed to normalise the variance.

2.3.4. Bioluminescence-based bacterial bioassay

Cultures of Escherichia coli HB101 pUCD607 and

Pseudomonas fluorescens 10586r pUCD607 were

freeze-dried using standard protocols and stored at

�20 8C (Sousa et al., 1998). Freeze-dried cultures

were resuscitated by resuspending the cells in 10 ml of

0.1 M KCl and 10 ml of Luria Bertani broth (10 g

tryptone, 5 g yeast extract and 5 g NaCl in 1 l) and

incubated at 25 8C for 1 h on an orbital shaker at 200

rpm. After resuscitation, E. coli cells were used im-

mediately. The suspension of P. fluorescens was

centrifuged, supernatant discarded and resuspended

in 5 ml of 0.1 M KCl. In each luminometer cuvette,

a 100 Al aliquot of the resuscitated cells was added to

900 Al of pore water test solution and mixed at 15-s

intervals. The luminescence of the samples was mea-

sured after 30 min of exposure on a Bio-Jade lumin-

ometer (Labtech International, Uckfield, UK). The

assays were carried out as three independent replicates

and luminescence expressed as a percentage of the

luminescence of the control (least contaminated soil)

sample. Bioluminescence response may be sensitive

to extreme pH ranges and consequently, the samples

were tested as extracted and after the pH had been

adjusted (normalised) to a pH value of 5.5. By nor-

malising the pH values, the bioassay could be per-

formed consistently to still assess the relative HM

toxicity.

2.3.5. Bioluminescence-based fungal bioassay

Fungal bioassays were carried out using the natu-

rally bioluminescent fungus Armillaria mellea (ATCC

1113) as described by Weitz et al. (2002). A. mellea

cultures grown on Malt Extract (ME) agar (Oxoid)

were used to inoculate 100 ml ME broth (adjusted to

pH 5.0F0.1 using 20% phosphoric acid) in 250 ml

Erlenmeyer flasks. The batch cultures were incubated

at 22 8C in an orbital-shaking incubator at 150 rpm in

the dark. After 8 days of incubation, the globular

mycelia formed were used in the pore water solution

bioassays. After 60 min of exposure to the samples,

luminescence was measured using a TD 20/20 lumin-

ometer (R.S. Aqua Ltd., Alton, UK) and recorded as

relative light units. Bioluminescence response was

calculated from the mean of triplicate flasks for each

G.I. Paton et al. / Science of the Total Environment 355 (2006) 106–117110

sample and the percentage luminescence relative to the

control (least contaminated soil) sample calculated.

2.4. Data analysis

Data from each ecotoxicity test were correlated

against both copper and nickel concentrations, de-

rived from the chemical analyses of either total

bulk soil or water-soluble fractions. The results

for those assays that were performed in the bulk

soil have been plotted against total soil HM con-

centration, whereas those that use pore water ex-

tract (biosensors) have been plotted against water-

soluble HM concentration. Any strong relationships

were noted and subjected to more detailed analysis

using non-linear regression function of Sigma Plot

2001. Sigmoidal, Gompertz model curves were fit-

ted to the data relationships between total hetero-

trophic bacterial numbers and HM concentrations in

bulk soil. Linear relationships with 95% c.i. fitted

to relationships between MIC and HM concentra-

tions in bulk soil and an exponential decay model

for bacterial biosensors to water-soluble copper

concentrations only.

Table 1

Chemical characterisations of soils, sampled along the 34 km transect fro

Sample Distance

(km)

pH Total Cu

(mg kg�1)

Total N

(mg k

1.1 34 5.28 66.3 138

1.2 34 3.51 97.8 174

1.3 34 3.55 66.7 142

2.1 27 3.79 271 390

2.2 27 4.10 576 669

2.3 27 3.73 488 743

3.1 22 4.04 1380 1570

3.2 22 4.03 2080 1490

3.3 22 3.92 1360 1570

4.1 17 3.79 1170 1040

4.2 17 4.14 265 400

4.3 17 4.09 308 413

5.1 11 4.08 1920 1979

5.2 11 3.78 1240 1450

5.3 11 3.79 3040 2210

6.1 10 3.97 519 761

6.2 10 4.05 725 981

6.3 10 3.92 1360 1819

7.1 2 4.44 845 1979

7.2 2 4.55 1330 4589

7.3 2 4.25 1739 4227

3. Results

3.1. Chemical analysis

The results of the analyses for total and water-

soluble HM concentrations, pH and percentage OM

of each sample are given in Table 1. Both the copper

(66.3–3040 mg kg�1) and nickel (138–4589 mg

kg�1) concentrations in the bulk soil tended to de-

crease with distance from the smelter. The HM con-

centration in the water-soluble fractions was

commonly 3 orders of magnitude lower than those

found in their respective bulk soil samples. Typical of

many forest soils, pH values were low (3.51–5.28)

with a high percentage of OM in most of the soils

(16.19–85.91%): the exception being the soils closest

to the smelter.

3.2. Ecotoxicological analysis—bulk soil

3.2.1. Respiration

There was no clear relationship between respiration

and concentration of either copper or nickel (Fig. 1).

When the data were interpreted by considering certain

m the Severonickel smelter

i

g�1)

Water-soluble

Cu (mg l�1)

Water-soluble

Ni (mg l�1)

Total

OM (%)

b0.01 b0.01 85.91

0.12 b0.01 62.23

0.14 b0.01 73.70

0.13 0.12 68.29

b0.01 0.21 60.62

0.09 0.09 68.63

0.09 0.01 67.57

0.30 0.61 67.80

0.09 0.70 83.24

0.01 0.02 74.46

0.02 0.01 74.66

0.01 0.01 43.98

0.08 0.03 59.63

0.67 0.41 35.27

0.81 0.10 83.20

b0.01 0.07 16.19

1.04 0.87 15.38

0.28 0.02 23.84

1.20 0.20 0.60

0.89 0.19 38.04

1.01 1.10 16.09

Total soil [HM] (Log10 mg kg-1)

1.5 2.0 2.5 3.0 3.5 4.00.0

0.2

0.4

0.6

0.8CuNi

CO

2 pr

oduc

ed (

mg

CO

2 g-

1 so

il da

y-1 )

Fig. 1. Relationship between basal respiration and HM concentra-

tion for soil samples taken along the 34 km transect. .—[Cu],

o—[Ni] (n =21).

10

15

20

25

30

35Cu r2 = 0.69595 % c.i.

G.I. Paton et al. / Science of the Total Environment 355 (2006) 106–117 111

soil conditions (OM, MPN, pH) at individual sites,

there was still no clear pattern. Neither total HM

concentrations nor low pH values could be attributed

to causing differences in respiration.

3.2.2. Total microbial heterotrophic numbers

The decline in microbial cell numbers did not

correspond to a standard dose response but at levels

of about 1000 mg kg�1 for both copper and nickel,

there was a major decrease in the total number of

microbial heterotrophs present (Fig. 2). The only sam-

ple that did not respond in this manner was 5.3, which

had a more elevated concentration of copper than the

Total soil [HM] (Log10 mg kg-1)

1.5 2.0 2.5 3.0 3.5 4.06.0

6.5

7.0

7.5

8.0

8.5

9.0

9.5

Cur2 = 0.526Nir2 = 0.898

Tot

al m

icro

bial

het

erot

roph

s (L

og10

cel

ls g

-1 s

oil)

Fig. 2. Relationship between the microbial heterotrophic number

and HM concentration for soil samples taken along the 34 km

transect. .—[Cu], o—[Ni] (n =20). Site 5.3 n/5 removed from

trendline data.

two locations beside it. The lowest numbers of micro-

bial heterotrophs were found at the 3 sites situated 2

km from the smelter.

3.2.3. MIC values

As the concentration increased for both copper and

nickel, the MIC value also increased in a linear fash-

ion ( p b0.001) (Fig. 3). The exception to this was the

soils nearest the smelter where the MPN and respira-

tion values were also lowest. The correlation between

the total HM concentration in soil and MIC value was

higher for nickel than copper and the MIC itself also

occurred at a higher concentration for nickel.

3.2.4. Total nematode count

As the concentration of copper and nickel in-

creased, the total nematode numbers tended to decline

but the relationship was not strong for either HM (Fig.

4). The decline was similar to that of a dose response

relationship accentuated by Site 4.1 which had signif-

MIC

val

ues

(mM

)

0

5

Total soil [HM] (Log10 mg kg-1)

1.5 2.0 2.5 3.0 3.5 4.00

20

40

60

80

100 Ni r2 = 0.84395 % c.i.

Fig. 3. Relationship between the MIC and HM concentration for soi

samples taken along the 34 km transect. .—[Cu], o—[Ni

(n =18). Sites 7.1, 7.2 and 7.3 n/5 removed from trendline data

as heterotrophic numbers small.

l

]

Total soil [HM] (Log10 mg kg-1)

1.5 2.0 2.5 3.0 3.5 4.0

sqrt

nem

atod

e co

unt/

mas

s

0

1

2

3

4

5

6

CuNi

Fig. 4. Relationship between total nematode count and HM concen-

tration for soil samples taken along the 34 km transect. .—[Cu],

o—[Ni] (n =21).

G.I. Paton et al. / Science of the Total Environment 355 (2006) 106–117112

icantly higher nematode numbers than any of the

other sites sampled and relatively high concentrations

of copper and nickel (N1000 mg kg�1). Although at

0

20

40

60

80

100 E. coli r2 =0.915

Water-soluble [Cu] (mg l-1)

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

% lu

min

esce

nce

of le

ast

cont

amin

ated

sit

e

0

20

40

60

80

100 P. fluorescensr2 = 0.934

0

A

C

Fig. 5. Relationship between bacterial % luminescence compared to the le

from soil samples taken along the 34 km transect. (A) E. coli to [Cu], (B) E

(n =21).

this site, the water-soluble HM concentrations (0.01

mg l�1 Cu and 0.02 mg l�1 Ni) were among some of

the lowest concentrations recorded in the study.

3.3. Ecotoxicological analysis—soil pore waters

3.3.1. Bioluminescence-based bacterial bioassay

Prior to normalising the pH values of the samples,

there was little correlation between the water-soluble

HM concentration and the decline in bioluminescence

of the E. coli biosensor. The samples were found to all

have low levels of luminescence yet some soils had low

pore water HM concentrations; these soils had the

lowest pH values, which was close to the operating

limit of the sensor. However, when the pH value of

these samples was adjusted, the response changed sig-

nificantly (Fig. 5) as the absolute value of luminescence

increased in the non-HM contaminated samples. More-

over, there was a significant negative relationship

( p b0.001) between the concentration of copper and

E. coli

Water-soluble [Ni] (mg l-1)

.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

P. fluorescens

B

D

ast contaminated site (Site 1.1) and water-soluble HM concentration

. coli to [Ni], (C) P. fluorescens to [Cu] and (D) P. fluorescens to [Ni]

water-soluble [HM] mg l-10.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

% lu

min

esce

nce

of le

ast

cont

amin

ated

sit

e

0

20

40

60

80

100

120

140

CuNi

Fig. 6. Relationship between fungal % luminescence compared to

the least contaminated site (Site 1.1) and water-soluble HM con-

centration from soil samples taken along the 34 km transect (n =16).

G.I. Paton et al. / Science of the Total Environment 355 (2006) 106–117 113

the luminescence values recorded but not with nickel.

The response of the P. fluorescens biosensor also be-

came more evident after the pH adjustment was carried

out (Fig. 5). Once again, the main reason for the decline

in luminescence was due to copper.

3.3.2. Bioluminescence-based fungal bioassay

Generally, fungal bioluminescence was reduced

when the water-soluble concentration of either copper

or nickel was greater than 0.15 mg l�1, as lumines-

cence declined sharply to approximately 50% of the

control sample (HM b0.01 mg l�1) (Fig. 6). Although

some samples showed luminescence inhibition at

lower HM concentrations, overall 0.15 mg l�1 pre-

sented a threshold level to which the A. mellea was

severely impacted by the water-soluble HM content.

4. Discussion

The absolute HM concentrations in the soil and

their relative decline with distance from the point

source corresponded to those reported in other recent

studies of smelters on the Kola Peninsula. Ayras and

Kashulina (2000) found that for the Monchegorsk

smelter, the copper and nickel concentrations in sur-

face soils declined rapidly depending upon the wind

direction and topographical features. Reimann et al.

(1999) reported elevated HM concentrations from

smelters in vegetation in a pattern similar to that

reported here. The transect selected in this study

corresponded closely with the findings of the north

to south transect of Ayras and Kashulina (2000) with

maximum nickel concentrations of 3000 mg kg�1

declining to less than 4 mg kg�1 over a similar

distance. These trends were further verified by data

from the Pechenganikel smelter studied by Koptsik et

al. (2003).

For both copper and nickel, the water-soluble HM

concentration was considerably lower than the total

soil concentration. This observation has been reported

by Vulkan et al. (2000) and Tom-Petersen et al. (2004)

for agricultural soils, in which the water extractable

fraction was considerably lower than that removed by

a saline or weak acid extraction. Furthermore, Tipping

et al. (2003) demonstrated that soils with horizons

high in OM had greater complexing ability. These

laboratory and field findings demonstrate the funda-

mental soil and pollutant physico-chemical parameters

that drive the processes of bioavailability. According-

ly, the low concentrations of copper and nickel in the

soil water phase in this study are to be expected. In

particular, the high OM content and the Podzolic Ah

horizon enhance the binding of the HM causing only a

small proportion to pass to the aqueous phase.

With respect to the soil ecotoxicity assays, there are

few comparable data in the literature for such tests

conducted on the Kola Peninsula. The selected eco-

toxicity tests reflect different aspects of the soil biol-

ogy and have been refined for the specific ecological

conditions of this study. The nematode assay is an

ecological measure of the in situ population, which

would be dominated by species that would feed on

fungi and bacteria. Copper, zinc, cadmium and nickel

at concentrations that inhibit earthworm activity in

agricultural studies have been reported to have a

significant effect on total nematode numbers (Korthals

et al., 1996a,b). From an ecological perspective, a

dose of copper at 250 mg kg�1 has been found to

reduce the number of bacterial-feeding nematodes yet

have no impact on the hyphal feeders (Korthals et al.,

1996b). Omnivorous and predator nematodes have

been reported as the most sensitive to copper while

plant-feeding nematodes showed the largest difference

in abundance demonstrating the impact of copper on

primary production (Dennis et al., 1997; Korthals et

al., 1998). Conditioning to the presence of pollutant

may have meant that the nematodes were relatively

insensitive to the pollutant doses. In the case of this

G.I. Paton et al. / Science of the Total Environment 355 (2006) 106–117114

study, other perturbations such as the change in pH

value and the source and form of carbon could explain

the differences in nematode numbers better than the

HM concentrations.

The extent of inhibition of soil respiration in forest

soils has been extensively reviewed (Komulainen and

Mikola, 1995; Niklinska et al., 1998; Schloter et al.,

2003). It is acknowledged that the nature and compo-

sition of soil litter and the microbial population as

well as laboratory preparation will govern the sensi-

tivity of this assay. The respiration results in this study

support the findings of Komulainen and Mikola

(1995) who stated that basal respiration was less

sensitive to HM pollutants than invertebrate assays

in forest soils. In addition, studies mimicking the HM

deposition of the Kola Peninsula onto a similar Finn-

ish boreal forest area showed no effect on basal

respiration either (Perkiomaki et al., 2003). Previous

studies that have reported significant impacts on soil

respiration in the presence of HM have used sites

polluted by an order of magnitude higher than those

reported in this study for the Kola Peninsula (Las-

kowski et al., 1994; Fritze et al., 1996).

It is acknowledged that when soil microbial com-

munities are exposed to HM polluted environments,

two responses occur: firstly, a reduction in species’

numbers and diversity and secondly, the development

of HM resistant or tolerant microbial populations

(Cervantes and Guiterrez-Corona, 1994; Davis et al.,

2004). These responses were assessed with a tetrazo-

lium-based assay. Obbard (2001) reported the unsuit-

ability of using copper with INT because it could

produce false negatives when binding to HM ligands.

However, in this study, there was no evidence of this.

In the case of the total MPN count, this may be

because the assay was carried out in a pH buffered

system, and as such, the residual copper from the

inoculum (which was below the levels tested by

Obbard (2001) may have complexed with the TES

and carbon source rather than the INT.

This study shows that the microbial heterotrophic

numbers dropped significantly at soil HM concentra-

tions of about 1000 mg kg�1. It is difficult to relate

the HM concentrations in these soils to that of agri-

cultural soils, where most ecotoxicity assays have

been conducted. If a comparison of the total HM

concentration and the water-soluble HM concentration

is made, then the results reported here exhibit a similar

response to those reported by Aoyama and Nagumo

(1997), Kelly and Tate (1998), Ellis et al. (2001) and

Stuczynski et al. (2003). However, conducting a total

measure of biomass activity or diversity indices as a

means of studying the impact of pollutants does not

give the same amount of information as using species

composition (Baath, 1989). It may be concluded that

HM concentrations of 1000 mg kg�1 impact on the

activity of the system, but changes in the microbial

community and species composition require the ap-

plication of more rigorous techniques.

With respect to the MIC tests, the actual HM

concentration in the test sample should be interpreted

as a relative concentration to enable discrimination

between samples rather than absolute values. This is

because a proportion of the copper and nickel would

be unavailable due to the pH of the sample and the

effect of complexation sources. The relative concen-

tration is valid for the soils in this test but cannot be

translated to other soils where alternative procedures

have been performed.

Prolonged exposure of soil to elevated HM con-

centrations will facilitate the development of a HM

tolerant or resistant population (Trevors et al., 1985).

Without the use of specific molecular probes, there are

few readily available routine tests to assess these

changes. The change in MIC values with increasing

pollutant load in this study were similar to those

reported by Baath (1992) and Diaz-Ravina et al.

(1994). The presence and indeed the extent of this

physiological response is a key process to understand-

ing the spatial and temporal impact of the pollutant.

Biosensors that are relevant for soil testing have

been developed to enable a rapid and ecologically

appropriate evaluation of pollutants to be made. In

the case of HM effects, their successful application

has been widely reported (Paton et al., 1997; McGrath

et al., 1999; Campbell et al., 2001). However, in this

study, the low pH values of the sample meant that for

the bacterial assay the results were variable and re-

quired all of the samples to be adjusted to the same

pH. Once this was done, dose response curves in-

ferred that copper (and not nickel) was the reason

for the decline in luminescence. The relative sensitiv-

ity of the assay and the nature of the dose response

curve were similar to that reported by Vulkan et al.

(2000) for both the E. coli and P. fluorescens biosen-

sors. The fungal biosensor A. mellea has been found

G.I. Paton et al. / Science of the Total Environment 355 (2006) 106–117 115

to be sensitive to copper (EC50 value b1 mg l�1) (as

previously shown by Weitz et al., 2002) but insensi-

tive to nickel (EC50 value N100 mg l�1). Hence,

changes in bioluminescence by A. mellea in this

study could only be caused by copper. The relative

differences in sensitivities observed between the bac-

terial and fungal biosensors agree with recent work by

Rajapaksha et al. (2004), who also noted diverse

responses to soil HM loads between bacteria and

fungi. In addition, this shows the importance in uti-

lizing a number of relevant soil microbiological tests

when investigating ecotoxicological response to con-

taminated soils.

Soil pollution studies cannot rely solely on chemical

analysis. Each of the assays performed here evaluated

the effect to varying degrees that HM have on a mea-

sured parameter. The development of microbial toler-

ance to copper and nickel was the most sensitive

procedure carried out but this cannot be directly related

to ecological functionality. The change in nematode

numbers was not solely attributable to the HM concen-

tration, while the biosensors responded to the water-

soluble HM concentration. To place these results in

context, a measure of the soil process must be made

and correlated with a soil ecotoxicity test.

Samples 7.1, 7.2 and 7.3 were greatly impacted by

the HM loads and the proximity to the smelter. These

sites had elevated HM concentrations, but due to the

deposition of other base cations, their pH values were

not the most extreme. Regardless of the pH values at

this location, the pedological processes had ceased to

function and measurements of the soil showed it to be

greatly impacted. Samples 5 and 6 (11 km and 10 km,

respectively) were the sites that had the highest HM

loads present, the highest development of HM toler-

ance, low microbial heterotrophic numbers and high

toxicity as interpreted by the biosensor responses. The

pollution from the smelter significantly impacted the

biological parameters of these locations.

To assess the relative impact of the remaining sites

is more difficult. Some locations (4.2, 3.1, 3.2 and

3.3) had concentrations of copper and nickel exceed-

ing 1000 mg kg�1 yet relatively low aqueous extract-

able metals. With respect to the biological tests, the

responses of the biosensor suggested that these sites

had low acute toxicity and that nematode numbers

remained unaffected. The microbial heterotrophic

count and HM tolerance values suggested that the

sites were impacted by the HM concentration

recorded. The HM may be bound in the OM, partially

complexed due to the pH value or immobilised by

exchange sites, hence the acute toxicity was low but

long-term exposure tests revealed a response.

5. Conclusions

Chemical characterisation in isolation is an inef-

fective method to assess the impact of soil pollution.

Although there are a wide range of soil biological tests

available, they must be selected to represent the key

attributes associated with the area under investigation

and the pollutant present. The Kola Peninsula, a high

latitude and fragile ecosystem, has elevated HM con-

centrations derived from point sources. Key soil bio-

logical analyses suggest that the system has become

conditioned to these pollutant burdens despite a com-

ponent of the HM being in a bioavailable form.

Acknowledgements

Ekaterina Viventsova (Ruth) wishes to thank the

Swedish Foundation for International Co-operation in

Research and Higher Education, STINT for funding

her visit to Aberdeen.

References

Ajungla T, Sharma GD, Dkhar MS. Heavy metal toxicity on dehy-

drogenase activity on rhizospheric soil of ectomycorrhizal pine

seedlings in field condition. J Environ Biol 2003;24:461–3.

AMAP, Arctic Monitoring and Assessment Programme. Arctic

pollution. Oslo.

Aoyama M, Nagumo T. Effects of heavy metal accumulation in

apple orchard soils on microbial biomass and microbial activi-

ties. Soil Sci Plant Nutr 1997;43:601–12.

Ayras M, Kashulina G. Regional patterns of element contents in the

organic horizon of podzols in the central part of the Barents

region (Finland, Norway and Russia) with special reference to

heavy metals (Co, Cr, Cu, Fe, Ni, Pb, V and Zn) and sulphur as

indicators of airborne pollution. J Geochem Explor 2000;68:

127–44.

Baath E. Effects of heavy-metals in soil on microbial processes and

populations (a review). Water Air Soil Pollut 1989;47:335–79.

Baath E. Measurement of heavy-metal tolerance of soil bacteria

using thymidine incorporation into bacteria extracted after

homogenization centrifugation. Soil Biol Biochem 1992;24:

1167–72.

G.I. Paton et al. / Science of the Total Environment 355 (2006) 106–117116

Bewley RJF, Stotzky G. Effects of cadmium and zinc on microbial

activity in soil-influence of clay minerals: I. Metals added

individually. Sci Total Environ 1983;31:41–55.

Braddock JF, Catterall PH. A simple method for enumerating

gasoline- and diesel-degrading microorganisms. Bioremediat J

1999;3:81–4.

Campbell CD, Paton GI, Towers W, Paterson E, Dawson JJC,

Cameron CM, et al. A biological classification scheme to assess

the sensitivity of Scottish and Northern Ireland soils to heavy

metals. SNIFFER Report No. SR (00) 09, 2001.

Cervantes C, Guiterrez-Corona F. Copper resistance mechanisms in

bacteria and fungi. FEMS Microbiol Rev 1994;14:121–37.

Davis MRH, Zhao F-J, McGrath SP. Pollution-induced community

tolerance of soil microbes in response to a zinc gradient. Envi-

ron Toxicol Chem 2004;23(11):2665–72.

Dennis JL, Mutwakil MHAZ, Lowe KC, de Pomerai DI. Effects of

metal ions in combination with a non-ionic surfactant on stress

responses in a transgenic nematode. Aquat Toxicol 1997;40:

37–50.

Dıaz-Ravina M, Baath E, Frostegard A. Multiple heavy-metal tol-

erance of soil bacterial communities and its measurement by a

thymidine incorporation technique. Appl Environ Microbiol

1994;60:2238–47.

Dorn E, Helweg M, Reineke W, Knackmass HJ. Isolation and

characterisation of a 3-chlorobenzoate degrading pseudomonad.

Arch Microbiol 1974;99:61–70.

Ellis RJ, Neish B, Trett MW, Best JG, Weightman AJ, Morgan P,

et al. Comparison of microbial and meiofaunal community

analyses for determining impact of heavy metal contamination.

J Microbiol Methods 2001;45:171–85.

Ellis RJ, Best JG, Fry JC, Morgan P, Neish B, Trett MW, et al.

Similarity of microbial and meiofaunal community analyses for

mapping ecological effects of heavy metal contamination in soil.

FEMS Microbiol Ecol 2002;40:113–22.

Environmental Committee of the Murmansk Region, Ministry of

Natural Resources of Russian Federation, Murmansk. Environ-

mental situation in the Murmansk Region in 2000 [Sostojanii

okruzhajuschej prirodnoj sredy v Murmanskoj oblasti v 2000

godu] (in Russian).

Flynn HC, McMahon V, Diaz GC, Demergasso CS, Corbisier P,

Meharg AA, et al. Assessment of bioavailable arsenic and

copper in soils and sediments from the Antofagasta region of

Northern Chile. Sci Total Environ 2002;286:51–9.

Fritze H, Vanhala P, Pietikainen J, Malkonen E. Vitality fertilization

of Scots pine stands growing along a gradient of heavy metal

pollution: short-term effects on microbial biomass and respira-

tion rate of the humus layer. Fresenius’ J Anal Chem 1996;

354:750–5.

Hooper DJ. Extraction of free-living stages from soil. In: Southey

JF, editor. Laboratory methods for work with plant and soil

nematodes. London7 HMSO; 1986. p. 5–31.

Kelly JJ, Tate RL. Effects of heavy metal contamination and reme-

diation on soil microbial communities in the vicinity of a zinc

smelter. J Environ Qual 1998;27:609–17.

Klee AJ. A computer program for the determination of most prob-

able number and its confidence limits. J Microbiol Methods

1993;18:91–8.

Klerks PL, Weis JS. Genetic adaptation to heavy metals in aquatic

organisms—a review. Environ Pollut 1987;45:173–205.

Komulainen M, Mikola M. Soil processes as influenced by heavy

metals and the composition of soil fauna. J Appl Ecol 1995;32:

234–41.

Koptsik S, Koptsik G, Livantsova S, Eruslankina L, Zhmelkova

T, Vologdina Z. Heavy metals in soils near the nickel

smelter: chemistry, spatial variation, and impacts on plant

diversity. J Environ Monit 2003;5:441–50.

Korthals GW,AlexievAD, LexmondTM,Kammenga JE, Bongers T.

Long-term effects of copper and pH on the nematode community

in an agroecosystem. Environ Toxicol Chem 1996a;15:979–85.

Korthals GW, van de Ende A, van Megen H, Lexmond TM, Kam-

menga JE, Bongers T. Short-term effects of cadmium, copper,

nickel and zinc on soil nematodes from different feeding and life-

history strategy groups. Appl Soil Ecol 1996b;4:107–17.

Korthals GW, Popovici I, Iliev I, Lexmond TM. Influence of

perennial ryegrass on a copper and zinc affected terrestrial

nematode community. Appl Soil Ecol 1998;10:73–85.

Korthals GW, Bongers M, Fokkema A, Dueck TA, Lexmond TM.

Joint toxicity of copper and zinc to a terrestrial nematode com-

munity in an acid sandy soil. Ecotoxicology 2000;9:219–28.

Kozlov MV, Barcan V. Environmental contamination in the central

part of the Kola Peninsula: history, documentation, and percep-

tion. AMBIO 2000;29:512–7.

Laskowski R, Maryanski M, Niklinska M. Effect of heavy-metals

and mineral nutrients on forest litter respiration rate. Environ

Pollut 1994;84:97–102.

Lawlor K, Knight BP, Barbosa-Jefferson VL, Lane PW, Lilley AK,

Paton GI, et al. Comparison of methods to investigate microbial

populations in soils under different agricultural management.

FEMS Microbiol Ecol 2000;33:129–37.

McGrath SP, Knight B, Killham K, Preston S, Paton GI. Assessment

of the toxicity of metals in soils amended with sewage sludge

using a chemical speciation technique and a lux-based biosen-

sor. Environ Toxicol Chem 1999;18:659–63.

Niklinska M, Laskowski R, Maryanski M. Effect of heavy metals

and storage time on two types of forest litter: basal respiration

rate and exchangeable metals. Ecotoxicol Environ Saf 1998;41:

8–18.

Nolan AL, Lombi E, McLaughlin MJ. Metal bioaccumulation and

toxicity in soils—why bother with speciation? Aust J Chem

2003;56:77–91.

Obbard JP. Measurement of dehydrogenase activity using 2-p-iodo-

phenyl-3-p-nitrophenyl-5-phenyltetrazolium chloride (INT) in

the presence of copper. Biol Fertil Soils 2001;33:328–30.

Paton GI, Rattray EAS, Campbell CD, Cresser MS, Glover LA,

Meeussen JCL, et al. Use of genetically modified microbial

biosensors for soil ecotoxicity testing. In: Pankhurst C, Doube

B, Gupta V, editors. Biological indicators of soil health and

sustainable productivity. Wallingford7 CAB International; 1997.

p. 397–418.

Paton GI, Iroegbu CO, Dawson JJC. Microbiological characterisa-

tion of a diesel contaminated beach site. Mar Pollut Bull

2003;46:903–17.

Perkiomaki J, Tom-Petersen A, Nybroe O, Fritze H. Boreal forest

microbial community after long-term field exposure to acid and

G.I. Paton et al. / Science of the Total Environment 355 (2006) 106–117 117

metal pollution and its potential remediation by using wood ash.

Soil Biol Biochem 2003;35:1517–26.

Rajapaksha RMCP, Tobor-Kaplon MA, Baath E. Metal toxicity

affects fungal and bacterial activities in soil differently. Appl

Environ Microbiol 2004;70:2966–73.

Reimann C, Halleraker JH, Kashulina G, Bogatyrev I. Comparison

of plant and precipitation chemistry in catchments with different

levels of pollution on the Kola Peninsula, Russia. Sci Total

Environ 1999;244:169–91.

Ruth-Balaganskaya E, Kudrjavtseva O. Sulphur migration in the

soil–plant system contaminated by deposits from nickel indus-

try: a field manipulation. Environ Pollut 2002;117:287–93.

Schloter M, Dilly O, Munch JC. Indicators for evaluating soil

quality. Agric Ecosyst Environ 2003;98:255–62.

Siciliano SD, Roy R. The role of soil microbial tests in ecological

risk assessment: differentiating between exposure and effect.

Hum Ecol Risk Assess 1999;5:671–82.

Sousa S, Duffy C, Weitz H, Glover LA, Bar E. Use of a lux-

modified bacterial, biosensor to identify constraints to bioreme-

diation of BTEX-contaminated sites. Environ Toxicol Chem

1998;17:1039–45.

Spurgeon DJ, Hopkin SP, Jones DT. Effects of cadmium, copper,

lead and zinc on growth, reproduction and survival of the

earthworm Eisenia-foetida (Savigny)—assessing the environ-

mental-impact of point-source metal contamination in terrestrial

ecosystems. Environ Pollut 1994;84:123–30.

Stuczynski TI, McCarty GW, Siebielec G. Response of soil micro-

biological activities to cadmium, lead, and zinc salt amend-

ments. J Environ Qual 2003;32:1346–55.

The Royal Society. The nitrogen cycle of the United Kingdom.

Report of a royal society study group. London UK7 The Royal

Society; 1983. p. 264.

Tiensing T, Preston S, Strachan N, Paton GI. Soil solution extraction

techniques for microbial ecotoxicity testing: a comparative eval-

uation. J Environ Monit 2001;3:91–6.

Tipping E, Rieuwerts J, Pan G, Ashmore MR, Lofts S, Hill MTR,

et al. The solid-solution partitioning of heavy metals (Cu, Zn,

Cd, Pb) in upland soils of England and Wales. Environ Pollut

2003;125:213–25.

Tom-Petersen A, Hansen HCB, Nybroe O. Time and moisture

effects on total and bioavailable copper in soil water extracts.

J Environ Qual 2004;33:505–12.

Trevors JT, Oddie KM, Belliveau BH. Metal resistance in bacteria.

FEMS Microbiol Rev 1985;32:39–54.

van Beelen P, Doelman P. Significance and application of microbial

toxicity tests in assessing ecotoxicological risks of contaminants

in soil and sediment. Chemosphere 1997;34:455–99.

Vulkan R, Zhao FJ, Barbosa-Jefferson V, Preston S, Paton GI,

Tipping E, et al. Copper speciation and impacts on bacterial

biosensors in the pore water of copper-contaminated soils.

Environ Sci Technol 2000;34:5115–21.

Weitz HJ, Campbell CD, Killham K. Development of a novel,

bioluminescence-based, fungal bioassay for toxicity testing.

Environ Microbiol 2002;4:422–9.