Environ Monit Assess (2010) 166:641–651DOI 10.1007/s10661-009-1029-z

Environmental impacts of anthropogenic activitieson the mineral uptake in Oreochromis mossambicusfrom Indus River in Pakistan

Farhat Jabeen · Abdul Shakoor Chaudhry

Received: 12 November 2008 / Accepted: 3 June 2009 / Published online: 16 June 2009© Springer Science + Business Media B.V. 2009

Abstract We examined the extent of mineral up-take in different tissues of Oreochromis mossam-bicus from Indus River which is claimed to bepolluted by human activities. Samples of waterand fish tissues were analysed from two sites(SK = upstream and CH = downstream) of IndusRiver. Whilst the water quality appeared to besuitable for aquatic life, significant differences be-tween fish tissues and sampling sites were ob-served for different mineral concentrations. Finsgenerally had the highest metal load followed bymuscles, gills, scales and skin. Na, Mg, Mn andZn concentrations in different fish tissues weregreater for CH than SK, whereas K, Ca, Pb,Cu, Fe, Hg and Cr were higher at SK than CH(P < 0.001). This variation in metal profiles ofdifferent locations of the Indus River was a reflec-tion of relevant mineral pollutions at these sites.It appeared that the pattern of metal uptake in

F. Jabeen · A. S. Chaudhry (B)School of Agriculture,Food and Rural Development,Newcastle University,Newcastle upon Tyne, England, UKe-mail: a.s.chaudhry@ncl.ac.uk

F. JabeenZoology Department,GC University Faisalabad,Faisalabad, Pakistane-mail: farjabeen2004@yahoo.co.in

fish tissues can be utilised as an indicator of envi-ronmental contamination of river water systems.These studies may help us plan strategies to alle-viate the ecotoxicological impacts of heavy metalsin freshwaters on fish and human populations.

Keywords Oreochromis mossambicus ·Mineral uptake · Indus River · Pollution ·Heavy metals · Bioaccumulation

Introduction

The accumulation of toxic metals in aquatic biotahas become a problem of increasing concern(Idodo-Umeh 2002). Excessive pollution of sur-face waters could lead to health hazards in humanbeings, either through drinking water and/or con-sumption of fish. The widespread contaminationof watercourses by heavy metals is problematicbecause of their toxicity, persistence and bioaccu-mulation in various biosystems including fish. Inthe fluvial environments, heavy metals are pro-duced from various natural and anthropogenicsources, such as atmospheric deposition, geologicweathering, agricultural activities and residentialand industrial products (Demirak et al. 2006). Theincreasing importance of fish as a protein sourceand the interest in understanding the accumu-lation of heavy metals at the trophic levels ofthe food chain extend the focus towards finfish

642 Environ Monit Assess (2010) 166:641–651

(Obasohan and Oronsaye 2004). Pollutants enterfish through five main routes (food or non-foodparticles, gills, water and skin), absorbed intoblood and then carried to either a storage pointor to the liver for its transformation or storage.Pollutants that are transformed and not stored inthe liver are excreted in bile or transported toother excretory organs such as gills or kidneysfor elimination or stored in fat (Nussey et al.2000). The pollutant concentration in any tissuetherefore depends on its absorption rate and thedynamic processes associated with its eliminationby the fish. Fish species are often the top con-sumers in aquatic ecosystems (Dallinger et al.1987), and thus, metal concentrations in fish couldindicate the environmental status (Jorgenson andPedersen 1994; Widianarko et al. 2000). As fishbioaccumulate metals, their use as biomonitorshas the advantage to compare metal concentra-tions amongst sites where water samples are be-low the detection limits of the most analyticaltechniques (Ramelow et al. 1989).

The Indus River is one of the key water re-sources for the economy of Pakistan—especiallythe breadbasket of Punjab province, which pro-vides most agricultural production and fisheries ofthe country. Therefore, the study was carried outin Mianwali District of Pakistan along the stretchof the Indus River to assess the bioaccumulationof metals in highly exposed organs (e.g. skin,scales, gills and fins) and muscles of freshwaterfish under natural conditions (Sultana and Rao1998). This information could be used in earlywarning systems to monitor freshwater metal pol-lutions and subsequently to adopt practices toreduce their impacts on the aquatic and humanpopulations.

Oreochromis mossambicus was selected as anexperimental model because it is hardy, tolerantand adaptable to high salinities in Indus Riverwhich flows through salty mountains. It is an ex-otic fish for Pakistan as it was originated fromAfrica. Its juveniles are omnivorous, whilst adultsfeed on detritus. It matures early and breedsthroughout the year. Moreover, the local inhabi-tants prefer this fish due to its taste, whilst fishingis one of the traditional occupations of this area.Therefore, it is vital for the river authorities tomaintain the fish health, production and meat

quality through its regular biomonitoring in andbioremediation of its inhabiting waters if needed.

Materials and methods

Study area and its importance

The study area of Mianwali District is locatedaround the River Indus. This district covers about5,840 km2 and contains nuclear power plants,Chashma Barrage and the Chashma Hydel powerplant. It is located 32◦34′60 N and 71◦32′60 E withan altitude of 211 m (695 ft). It is quite rich in min-erals, argillaceous clay, coal, dolomite, fire clay,gypsum, limestone, salts, silica sand and rockswhich are excavated in commercial quantities.The district has extreme hot and cold climatewhere temperature ranges between 51◦C in sum-mer and −2◦C in winter, and annual rainfall isabout 250 mm. There are about 259 cottages andother small to large industries including cement,penicillin, cotton ginning and pressing, drugs andpharmaceuticals, fertiliser, flour, oil and powergeneration. This study was conducted at two sitesalong the stretch of Indus River in Mianwali,which were 40 km apart from each other. TheIndus River flows from Shehbaz Khel (SK =upstream) to Chashma (CH = downstream). Dur-ing its course, some sediments are deposited andother pollutants from agricultural runoffs and do-mestic and municipal wastes enter into the IndusRiver at CH where water is stored for powergeneration. Therefore, the study was planned toinvestigate the effect of mineral pollution fromdifferent sources on the mineral profiles of waterand fish.

This study envisages the bioaccumulation pat-terns of different minerals in exposed parts (e.g.fins, gills, scales and skin) and muscular tissues ofthe fish which could be used as an early warningindicator of freshwater pollution in order to safe-guard the quality of aquatic life to promote fishproduction.

Water sampling

Representative samples of about 1 L water werecollected in clearly marked polypropylene bottles

Environ Monit Assess (2010) 166:641–651 643

with polyethylene caps that were thoroughlywashed with distilled water and then with the riverwater from the sampling sites. The water sampleswere collected in October 2007 at midday fromthree locations as three replicates from each of thetwo aforementioned sites at 30-cm depth.

Preservation of water samples

The collected water samples were filtered andpreserved in 5 mL 55% HNO3 per litre to preventmetal adsorption on the inner surface of the con-tainer and stored at 4◦C before their analyses forphysical, chemical and heavy metal parameters.For faecal coliform count, the water samples wereimmediately stored on ice before being processedafter their transportation to the laboratory within8 h post-sampling.

Fish sampling and measurements

Fishing was performed during late night with thehelp of professional local fishermen. Gill nets ofabout 1,200 cm long and 180 cm wide with a corkline at the top rope and the metal line with theground nylon rope were used for fishing. Four fish-ermen on two wooden boats operated a single gillnet. Motor-driven boats were not used to avoidfish disturbance due to their noisy engines. Nextmorning, the total fish catches were harvestedfrom three nets per site and the relevant samplesof live fish of similar size were transferred to largewater buckets. These fish were then humanelykilled using the concussive blow to the head (per-cussive stunning) of each selected fish. Twenty-seven samples of O. mossambicus by involvingnine fish per net as replicates were collected onice from each site when water samples were alsocollected for their analyses. The fish samples wereimmediately transported to the laboratory wheremorphometric measurements involving fresh deadweight (FDW), length and width of each of thesefish were carried out.

Fish dissection and preservation

After morphometric measurements, each fish wasdissected to collect different organs and tissues.

These organs were weighed individually, washedwith distilled water and transferred into markedsterilised polythene bags and stored in a freezer at−20◦C for further analysis.

Biophysical and chemical analysis

Water samples

Standard methods as described by the AmericanPublic Health Association (APHA), AmericanWater Works Association (AWWA) and WaterEnvironment Federation (WEF 2005) were fol-lowed for the determination of various biophysicaland chemical parameters of these water samples.All the reagents used during water analyses wereof analytical grade and purchased from Sigma.Temperature, pH and electrical conductivity ofwater samples were measured immediately using atemperature probe and a pH and conductivity me-ter (720WTW, Series 82362 Wellehein, Germany).

Turbidity

Turbidity was measured using 2130 B nephelo-metric method as described by APHA, AWWA,WEF (2005).

Total dissolved solids

One hundred millilitres of a well-mixed watersample was filtered through a glass fibre filterpaper before adding to a pre-weighed dish forits evaporation to dryness on a water bath anddried for 1 h in an oven at 105◦C. The dried masswas then weighed after cooling in a desiccator.The soluble salts were weighed, calculated andreported as milligrams per millilitre.

Faecal coliform bacteria

Faecal coliform densities were measured by 9222D faecal coliform membrane filter procedure asprovided by APHA, AWWA, WEF (2005).

Chlorine

Chlorine was measured by 4500-Cl G. DPD(N,N-diethyl-p-phenylene diamine) colorimetric

644 Environ Monit Assess (2010) 166:641–651

method as described by APHA, AWWA, WEF(2005).

Total hardness

EDTA titrimetric method was used for the esti-mation of total hardness. Well-mixed water sam-ple (25 mL) was diluted to 50 mL with distilledwater in a flask. To this flask, 2- to 4-mL bufferwith a pH of 10 (67.5 g NH4Cl in 570 mL con-centrated NH4OH and diluted to 1 L) and twoto three drops of Eriochrome Black-T [0.5 gsodium salt of 1-(1-hydroxy-2-naphthylazo)-5-nitro-2 naphthanol-4-sulfonic acid dye in 100 gtriethanolamine] indicator were added and slowlytitrated against 0.01 M EDTA with continuousstirring until the last reddish tinge colour changedto bluish purple.

Total hardness(mg

/L

) = Mg + Ca (as CaCO3)

= V × M × 100 × 1,000

Sample (ml)

where V is volume of EDTA used. M is molarityof EDTA (0.01 M) and 100 is the molecular weightof CaCO3.

Mineral analysis

The water samples were analysed for Na, K, Ca,Mg, Cu, Mn, Zn and Cr. For this purpose, each100 mL of filtered water sample was acidified with5 mL of HNO3 (55%) in a 250-mL volumetricflask and evaporated on a hot plate to about20 mL within a fume cupboard. After cooling ofthis evaporated digested solution, 5 mL of HNO3

(55%) and 10 mL of perchloric acid (70%) wereadded. The mixture was evaporated on a hot plateuntil the brown fumes converted into dense whitefumes of perchloric acid. The samples were re-moved from the hot plate, cooled and diluted to100 mL with distilled water in a 100-mL volumet-ric flask. The solutions were then aspirated intoan atomic absorption spectrophotometer (modelAA-660X VI42) using an air acetylene flame for

the determination of these minerals. Standardsolutions were prepared to construct standardcurves for comparison with the sample readingsto determine each metal concentration. Atomicabsorption spectrophotometry, due to its avail-ability at the University of Agriculture FaisalabadPakistan, was used to determine the mineral pro-files of water samples. However, the analysis offish samples was carried out using inductively cou-pled plasma optical emission spectroscopy (ICP-OES) due to its speed and availability at theNewcastle University.

Fish tissues

The frozen fish tissues were carried to the UK bythe prior authorisation of the Secretary of Statefor DEFRA under regulation 4 of Products ofAnimal Origin Regulation 2006 in October 2007.These samples were stored at −20◦C on arrivalbut freeze-dried and ground afterwards. Thesesamples were digested in concentrated HNO3 us-ing about 1 g of dried sample in 10 mL of con-centrated HNO3 (VWR, UK) in digestion blocksat 80◦C. Each sample was evaporated to about2 mL, cooled, diluted to 10 mL with distilled waterand filtered with Whatman filter 1. These sampleswere then analysed using Unicam 701 ICP-OESsystem. The machine was calibrated over the rel-evant concentrations using individually certifiedstandards obtained from Sigma-Aldrich, UK. Formineral analysis, fins, gills, scales, skin and mus-cles of each fish were selected to check the impactof the river environment on these tissues.

Statistical analysis

The data were statistically analysed using Minitabsoftware to compare the main effect of either onlythe sampling site on water quality and fish para-meters or the sampling site, fish tissues and theirsite × fish tissue interaction on the mineral profilesof these fish tissues. These effects were declaredsignificant if P < 0.05 and highly significant if P <

0.01. The Tukey’s test was used if there were morethan two means to compare at P < 0.05.

Environ Monit Assess (2010) 166:641–651 645

Results

Water quality

Table 1 presents the mean values of the biophys-ical and chemical composition of water samplesfrom two sites of Indus River. All parametersexcept temperature showed significant differencesbetween these sites. SK had significantly greaterpH, turbidity, total dissolved solids, conductivity,total hardness and chlorine than CH (P < 0.001).However, the coliform counts were significantlylower for SK than CH (P < 0.001).

Table 2 presents the mean values of macro andtrace elements at two sites of the Indus River.All macro elements showed significant differencesbetween sites. CH had greater concentration ofNa and K than SK, whereas SK had greater Caand Mg than CH (P < 0.001). Trace elements in-cluding Cu, Mn, Zn and Cr showed non-significantdifferences for the two sites (P > 0.05), althoughCr was numerically higher at SK than CH.

Fish parameters

Table 3 presents the mean values of length, width,FDW and weight as percentage of FDW of differ-ent organs in O. mossambicus from two sites of theIndus River. There was no significant difference in

Table 1 Physical and chemical parameters of water(means, SE and P values)

Parameters CH SK CH versus SK

Mean Mean SE P value

BiophysicalTemperature (◦C) 25.1 25 0.07 0.418pH 7.23 8.14 0.07 0.001Turbidity (NTU) 1.25 2.50 0.11 0.001Total dissolved solids 225.4 431 6.94 0.001

(mg/L)Coliform count 22 11.3 1.1 0.001

(no./100 mL)Electrical conductivity 288.1 319.7 3.38 0.001

(μS/cm)Chemical

Chlorine (mg/L) 11.7 16 0.54 0.001Total hardness (mg/L) 138.3 300.4 10.85 0.001

Table 2 Mineral profile of water samples (with mean, SEand P values)

Elements CH SK CH versus SK

Mean Mean SE P value

MacroNa (ppm) 38.4 13.0 1.95 0.001K (ppm) 23.9 9 0.71 0.001Ca (ppm) 54.1 84.7 3.84 0.001Mg (ppm) 15 34.6 2.37 0.001

TraceCu (ppm) 0.18 0.17 0.01 0.641Mn (ppm) 0.02 0.02 0.01 0.995Zn (ppm) 0.27 0.29 0.01 0.129Cr (ppm) 0.06 0.14 0.04 0.123

different parameters of fish at the two sites (P >

0.05), except for the weights as percent of FDWof fins (P < 0.01) and gills (P < 0.05) which weresignificantly greater for SK than CH. Although thefish fresh weight was numerically greater at CHthan SK, the difference between fresh fish weightswas non-significant (P > 0.05).

Table 4 shows the mean concentration of macroand trace elements in milligrams per kilogramof dry matter in the selected fish tissues at twoselected sites of the Indus River. While all macroand trace elements showed highly significantdifferences between fish organs (P < 0.001), thesite × organ interaction was also significant for

Table 3 Mean length, width, FDW and weight of differentorgans as percent of FDW in O. mossambicus from twosites of Indus River

Parameters CH SK CH versus SK

SE P value

Length (cm) 19.8 20.2 0.74 0.760Width (cm) 8 7.6 0.43 0.610Fish weight (g) 207 143 36.88 0.287Organs as percent of FDW

Muscles 60.3 61.3 1.80 0.761Scales 4.5 5.9 1.00 0.369Skin 7.2 7.3 1.58 0.935Fins 5.4 8.5 0.35 0.003Gills 2.4 3.2 0.18 0.034Intestine 3 2.2 0.45 0.246Liver 0.33 0.30 0.04 0.756

646 Environ Monit Assess (2010) 166:641–651

Tab

le4

Mea

nco

ncen

trat

ion

(mg/

kgD

M)

ofm

acro

and

trac

eel

emen

tsin

som

ese

lect

edti

ssue

sof

O.m

ossa

mbi

cus

from

two

site

sof

Indu

sR

iver

CH

SKSE

and

sign

ifica

nce

for

the

mai

nef

fect

sof

site

and

orga

nan

dth

eir

inte

ract

ion

Ele

men

tsF

ins

Gill

sSc

ales

Skin

Mus

cle

Fin

sG

ills

Scal

esSk

inM

uscl

eSi

teO

rgan

Site

×or

gan

Mac

roN

a32

419

921

315

51,

662

278

244

209

151

1534

30.4

948

.16∗

∗∗68

.20

K2,

758

3,10

21,

164

5,33

08,

046

223

5,18

41,

129

5,26

27,

226

32.7

7∗∗

51.7

6∗∗∗

73.3

1∗∗∗

Mg

1,61

549

51,

386

390

951

1,24

983

41,

239

474

799.

84.

96∗∗

∗7.

84∗∗

∗11

.10∗

∗∗C

a70

,650

14,5

4010

0,62

73,

691

20,0

9973

,410

25,3

0095

,860

6,22

712

,286

246.

939

0.12

∗∗∗

552.

49∗∗

∗T

race

Mn

7.33

14.7

811

.42.

956.

811

.74

9.79

8.43

3.78

6.94

0.29

0.47

∗∗∗

0.66

∗∗∗

Pb

6.39

2.44

5.8

2.95

3.6

6.30

2.87

5.49

2.22

2.87

0.05

∗∗∗

0.08

∗∗∗

0.12

∗∗∗

Cu

5.47

3.03

5.33

5.1

5.4

14.3

12.7

15.

086.

573.

550.

20∗∗

∗0.

31∗∗

∗0.

44∗∗

∗Z

n21

8.77

29.7

639

.23

82.2

66.2

79.4

53.0

531

.37

89.1

27.4

25.

23∗∗

∗8.

26∗∗

∗11

.70∗

∗∗F

e21

.08

233.

268

.20

59.0

62.1

42.5

222.

8329

.57

54.4

96.3

5.97

9.43

∗∗∗

13.3

5H

g2.

191.

681.

261.

502.

691.

872.

421.

232.

12.

240.

060.

09∗∗

∗0.

12∗∗

∗C

r2.

951.

181.

151.

4925

.42.

843.

600.

971.

3333

.73.

315.

23∗∗

∗7.

41

*P

<0.

05,*

*P

<0.

01,*

**P

<0.

001

most minerals (P < 0.001). Among macro ele-ments, Na and Mg in different fish tissues werehigher at CH than SK, whereas K and Ca weremore in fish tissues at SK than CH (P < 0.001).The order of Na in different tissues at CH wasmuscles>fins>scales>gills>skin and muscles>fins>gills>scales>skin at SK. K bioaccumulationpattern was similar at the two sites, i.e. muscles>skin>gills>fins>scales. In the present study,Na and K bioaccumulation pattern at the twosites was muscles>fins>gills>scales>skin andmuscles>skin>gills>fins>scales, respectively, andthere was a highly significant difference in thebioaccumulation of Na and K in different tissuesof O. mossambicus (P < 0.001; Fig. 1).

Mg bioaccumulation profile showed fins>scales>muscles>gills>skin at CH and fins>scales>gills>muscles>skin at SK. Ca bioaccumulationpattern showed scales>fins>muscles>gills>skinat CH and scales>fins>gills>muscles>skin at SK.Tukey’s test showed highly significant differencesin the bioaccumulation pattern of Mg and Ca indifferent tissues of O. mossambicus (P < 0.001;Fig. 1).

Mn, Pb and Zn in different fish tissues werehigher at CH than the SK site, whereas Cu, Fe,Hg and Cr were more at SK than the CH site(P < 0.001).

It appeared that fins were more susceptible tomineral contamination, followed by muscles, gills,scales and skin. Thus, the overall order of metalbioaccumulation in the tissues of O. mossambicuswas fins>muscles>gills> scales>skin.

Discussion

Water quality

The biophysical, chemical and mineral profilesof the water samples of this study comparedwell with the recommended standards of NEQS(1999), WHO (1985) and FEPA (2003), whichhighlighted the suitability of the Indus River waterfor aquatic life. However, chlorine concentrationexceeded the NEQS standards, so it may be pos-sible that it was a source of toxicity in Indus Riverdue to its affinity to react with certain organic

Environ Monit Assess (2010) 166:641–651 647

Fig. 1 Mean values ofmacro elements (Na, K,Mg and Ca as mg/kg DM)in selected tissues ofO. mossambicus (meansas columns showingdifferent letters differsignificantly at P < 0.05where standard errors ofeach mean are presentedas error bars)

Na

0

200

400

600

800

1000

1200

1400

1600

1800

Fins Gills Scales Skin Musclem

g/Kg

aa

b

a a

K

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

Fins Gills Scales Skin Muscle

mg/

Kg

a

b

c

d

e

Mg

0

200

400

600

800

1000

1200

1400

1600

Fins Gills Scales Skin Muscle

mg/

Kg

a

b

c

d

e

Ca

0

20000

40000

60000

80000

100000

120000

Fins Gills Scales Skin Muscle

mg/

Kg

a

b

c

d

e

compounds to form chlorinated products whichcould be toxic or carcinogenic for fish and otheraquatic animals. The low number of faecal col-iform count at SK might have been due to thehigh chlorine concentration which is known for itsdisinfecting properties (Tomar 1999). As chlorineaffects growth, reproduction and behaviour offish, it is vital to maintain chlorine within permis-sible recommended levels. Hence, high chlorinelevels are a cause of concern in relation to waterquality for aquatic life. However, the concentra-tion of most trace metals except Cr remainedcomparable for water from these two sites. Thesedata indicated that the flow of the Indus Riverin this region is under the influence of salt rangelithology. Indus River during its course bringsmany salts, metals, other solids and bicarbonatesfrom the rich mineral hills into SK. So some met-als, carbonates, bicarbonates, Ca and Mg formsediments along its course into upstream (SK)and other metals enter into downstream (CH).Therefore, the water quality variables show higherpH, turbidity, total dissolved solids, electrical con-ductivity, chlorine and total hardness at SK thanCH. Na and K are readily soluble in water, andhence, these were higher at CH than SK. This was

perhaps due to the increased accretion of salts asthe water ran from upstream to downstream of theIndus River in the study area.

Metal bioaccumulation in fish tissues

The metals in the fish tissues were several foldshigher than their corresponding values in the wa-ter. The metals also varied amongst different tis-sues of the same fish. When fish are exposed to theelevated metals in an aquatic environment, theycan absorb and so bioaccumulate the availablemetals directly from their surrounding environ-ment via the gills and skin or through the inges-tion of contaminated water and food (Ademoroti1996). Metals in fish are then transported by theblood stream to various organs and tissues wherefish can regulate metals to a certain extent, butthereafter, bioaccumulation can occur. Therefore,the ability of each tissue to either regulate or ac-cumulate metals can be directly related to the to-tal amount of metal accumulation in that specifictissue. Furthermore, physiological differences andthe position of each fish tissue within the aquaticenvironment can also influence the bioaccumula-tion of a particular metal (Kotze 1997).

648 Environ Monit Assess (2010) 166:641–651

Sodium and K are the most common non-toxicmetals of natural waters, as these are abundantin the earth’s crust. Due to their water solubility,sodium and potassium are leached out from soiland rocks into the neighbouring water coursessuch as the Indus River. Excessive sodium andpotassium can impart a bitter taste to drinkingwater and could be hazardous for people withcardiac and kidney ailments. There is no specificrecommended reference value for Na and K inthe edible part of the fish, and so it is difficultto comment on their values in the fish tissues ofthis study. However, as Na and K are the keyelements in physiological processes, the effect ofthe bioaccumulation of these elements needs to bedetermined in the fish of this area.

Mg and Ca occur naturally in the sediment andare the most common ions in freshwater causingwater hardness (USEPA 1999). In the presentstudy, fins have the highest and skin the least ten-dency for Mg bioaccumulation (Table 4). As Mg isnot potentially harmful to fish and wildlife (Swann2000) and it is unclear if its elevated levels infish tissues are harmful for fish itself, the humansand other wildlife species consuming the high Mgfish, the Mg levels of this study might not be ofmajor concern. The high Ca in scales and fins ofO. mossambicus may not be of concern for thehumans because these are not the edible tissuesof fish. However, their high levels in scales andfins confirmed their value in providing strength tothese vital components of living fish.

Mn levels (2.95–14.78 mg/kg DM) in fish sam-ples were higher when compared to 0.01-mg/kg(WHO 1985) and 0.05-mg/kg (FEPA 2003) stan-dards. The order of Mn concentrations varied indifferent tissues at both locations (Table 4), andthere was a highly significant difference in theMn bioaccumulation in different tissues of fish(P < 0.001). Mn has been reported to be taken updirectly through the gills or indirectly from foodand ingested sediments via gut (Bendell-Youngand Harvey 1986). High Mn in gills of this studyindicated possible uptake from gills. High Mn in-terferes with the central nervous system of verte-brates by inhibiting dopamine formation and alsoother metabolic pathways. Na regulation in fish isdisrupted by Mn and may ultimately cause death,whilst Mn can accumulate in the liver of fish.

High levels of Mn in exposed (non-edible) andedible fish tissues are a cause of concern, as Mn-contaminated fish can cause Mn-related disordersin the consumers.

Pb profile in the O. mossambicus variedamongst different tissues at two locations(Table 4),and the order of Pb accumulation agreed with thefindings that Pb in both aquatic and terrestrialvertebrates localised in hard tissues such as bonesand teeth (Kurey 1991), but these tissues were nottargeted for this study. It appeared, however, thatPb in O. mossambicus possessed a major affinityto reside in hard tissues like fins and scales.The Pb in different tissues varied from 2.218to 6.393 mg/kg DM (Fig. 2), which were higherthan the maximum allowable limit of 2 mg/kg forfood fish (WHO 1985; FEPA 2003). This valuewas also higher than the maximum contaminantlimit which was 0.05 mg/L for freshwater stan-dards. Acute Pb toxicity in fish causes renaldisorders which interfere with sugar metabolism.Lead disrupts haemoglobin synthesis and alsointerferes with the uptake of calcium and potas-sium through the gills. Fish affected by lead poi-soning become disoriented, and skin may peel offafter prolonged exposure to contaminated water(USEPA 1999). Consequently, it could be in-ferred that heavy load of lead in fish tissues,especially the edible parts, could induce healthhazards in fish as well as its consumers.

Copper affects growth, reproduction and be-haviour of fish. Fish affected by copper becomedarker, lethargic and indifferent to external stim-uli. If exposure persists, the fish become unco-ordinated and disoriented. The fish may becomeextremely colourful just before death since coppercauses the melanophores to relax. The behaviourof natural fish population is also affected by traceamount of copper. Sensitive fish population mayrestrict themselves to areas of stream where cop-per concentrations are lowest. Although this mayreduce exposure to copper, it limits spawning andinterferes with feeding habits. However, the fishsamples of this study did not show any apparentsigns of abnormality. This may be because theCu concentration in different tissues of this studyvaried from 3 to 14.3 mg/kg DM (Fig. 2), whichwere lower than the maximum recommendedstandards of 30 mg/kg in food fish (WHO 1985;

Environ Monit Assess (2010) 166:641–651 649

Fig. 2 Mean values oftrace elements (Mn, Pb,Cu, Zn, Fe, Hg and Cr asmg/kg DM) in selectedtissues of O. mossambicus(means as columnsshowing different lettersdiffer significantly atP < 0.05 where standarderrors of each mean arepresented as error bars)

Mn

02468

10121416

Fins Gills Scales Skin Muscle

ab

ca

d

Pb

01234567

Fins Gills Scales Skin Muscle

a

bc

d

b

Cu

02468

101214

Fins Gills Scales Skin Muscle

mg/K

gm

g/K

gm

g/K

ga

b

fefe

Zn

020406080

100120140160180200

Fins Gills Scales Skin Muscle

mg/K

gm

g/K

gm

g/K

g

a

bcc

b

e

Fe

0

50

100

150

200

250

Fins Gills Scales Skin Muscle

a

b

dad ad

Hg

00.5

11.5

22.5

3

Fins Gills Scales Skin Muscle

a a

b

a

c

Cr

0

510

15

2025

30

3540

45

Fins Gills Scales Skin Muscle

mg

/Kg

a

b

aa a

FEPA 2003; FAO 1983), and so the Cu in fishof Indus River was within safe limits for theconsumers.

The Zn contents in different tissues of thisstudy ranged from 27.4 to 218.8 mg/kg which werehigher than the recommended maximum limits of50 mg/kg in fish (WHO 1985; FAO 1983). Con-sequently, the consumption of fish of the IndusRiver may pose Zn-induced health hazards be-cause the fish muscles also contained high Zn to-gether with exposed parts. These findings agreedwith the studies on the Zn bioaccumulation infreshwater fish Channa punctatus, which indicatedthat accumulated Zn in gill tissues was lower thanthat in the liver and kidney (Murugan et al. 2008).

Lower Zn in gills suggested that Zn is excretedmore rapidly and reduced the body burden ofZn, which suggested that Zn was not accumulatedduring prolonged period in gill tissues. Studiesindicated that excessive Zn in muscles was trans-ferred to other organs in the fish being exposedto Zn-contaminated system (Madhusudan et al.2003). Fish had a tendency to push zinc burdenfrom muscles to other tissues like kidney duringmetallic stress, but this Zn metabolism in fish doesnot allow for excessive ambient metal in muscletissue to pose a threat to fish. This ability of de-loading of fish is advantageous to consumers whoare using fish muscles for food (Murugan et al.2008).

650 Environ Monit Assess (2010) 166:641–651

Biologically, Fe is an essential micronutrient re-quired by all living organisms, although at highconcentrations, it is toxic and inhibits enzymefunction. Since Fe is not readily absorbed throughthe gastrointestinal tract of vertebrates, it is notcommonly associated with toxicity when Fe-contaminated fish is consumed. Elevated levelsof Fe are also known to increase the susceptibil-ity of fish to infectious diseases. Fe in differenttissues varied from 21.1 to 233.2 mg/kg (Fig. 2),which agreed with the findings that Fe accumu-lated most in the gill tissues (Phillips and Russo1978). The Fe contamination cannot be judged asit was not considered in USEPA MCL or otherplans. Hg accumulated differently at two sites ofthis study where muscles>fins>gills>skin>scalesat CH and gills>muscles>skin>fins>scales at SK.The greater mercury in fish muscles of this studyagreed with the literature (Houserova et al. 2006;Dusek et al. 2005) where Hg was bioaccumulatedin fish mainly as methyl mercury. Hg in presentstudy varied from 1.23 to 2.7 mg/kg, which ex-ceeded the maximum allowable concentration of0.5 mg/kg for edible fish (Forstner and Wittmann1981). In lakes and other freshwater, small organ-isms convert naturally occurring inorganic mer-cury into organic methyl mercury. It is reportedthat methyl mercury binds with particles and sedi-ments eaten by smaller fish. Larger game fish preyon these smaller, mercury-contaminated fish. Asfish eliminate mercury at a much slower rate, itaccumulates in fish tissues and organs where itcannot be removed by filleting or cooking and soaccumulate in the skin and fat. As mercury affectsthe behaviour of vertebrates, inhibits enzyme ac-tivity and increases the abnormal cell division, itis vital to investigate mercury contamination infreshwater fish for the fish as well as consumer’shealth.

Cr levels (0.97 to 33.7 mg/kg) were high whencompared to standard limits of 0.05–0.15 mg/kg infood fish (WHO 1985; FEPA 2003). It has beenreported that enhanced metal levels in fish tissuesarise through biomagnification at each trophiclevel and carnivorous bottom feeders concen-trate higher metal levels (Forstner and Wittmann1981). As O. mossambicus is an omnivorous fishand adults mostly take detritus food, it bioaccu-mulated high Cr levels from the river sediments

and prey. In view of the higher levels of Cr, whencompared to WHO limits, it could be inferred thatconsumption of these fish could lead to health haz-ards in humans. High levels of Cr bioaccumulationin fish tissues could be due to chromite deposits inthe close vicinity of the study area of this paper.

Conclusions

The higher bioaccumulation of heavy metals atSK (upstream) than CH (downstream) may bedue to the fact that pH, turbidity, total dissolvedsolids, electrical conductivity, chlorine, total hard-ness, Ca, Mg, Zn and Cr levels were higher atSK than CH, and the heavy metals are known toaccumulate in the sediments, which act as sinksfor these pollutants. It therefore follows that O.mossambicus, known as detritus feeders, wouldrecord elevated levels in this study area. The highlevels of minerals in O. mossambicus give us thecause of concern for the community health issues,as the communities depend on fish as a majorprotein source. The high mineral levels in the fishtissues and especially in the edible part of fishof this study area would have detrimental effecton the health of rural community of Mianwali.Therefore, a very close identifying and monitoringof the source of metal loads in Indus River isneeded to minimise the possible risks to the healthof consumers. It is important to protect IndusRiver from anthropogenic sources of pollution toreduce environmental risks, and the study mayprovide preliminary database for future researchof this kind to examine metal levels in the fish ofIndus River.

Acknowledgements Thanks to the Higher EducationCommission of Pakistan for funding to support the post-doctoral research of Farhat Jabeen and Assistant DirectorFisheries and the fishermen of Mianwali District for theirhelp during sampling from the river.

References

Ademoroti, C. M. A. (1996). Standard methods for waterand effluent analysis (p. 182). Ibadan: Foludex.

APHA (American Public Health Association); AWWA(American Water Works Association); WEF (Water

Environ Monit Assess (2010) 166:641–651 651

Environment Federation) (2005). In A. D. Eaton,L. S. Clesceri, E. W. Rice, & A. E. Greenberg Standardmethods for the examination of water and wastewater(21st ed.). Washington, DC.

Bendell-Young, L., & Harvey, H. H. (1986). Uptakeand tissue distribution of manganese in the whitesucker (Catostomus commersoni) under conditions oflow pH. Hydrobiologia, 113, 117–125. doi:10.1007/BF00031861.

Dallinger, R., Prosi, F., Segner, H., & Black, H. (1987).Contaminated food and uptake of heavy metalsby rainbow trout (Salmo gairdneri): A field study.Oecologia, 73, 91–98. doi:10.1007/BF00376982.

Demirak, A., Yilmaz, F., Tuna, A. L., & Ozdemir, N.(2006). Heavy metals in water, sediments and tissuesof Leuciscus cephalus from a stream in southwest-ern Turkey. Chemosphere, 63, 1451–1458. doi:10.1016/j.chemosphere.2005.09.033.

Dusek, L., Svobodova, Z., Janouskova, D., Vykusova,B., Jarkovsky, J., Smid, R., et al. (2005). Bioaccu-mulation of mercury in muscle tissue of fish in theElbe River (Czech Republic): Multispecies moni-toring study 1991–1996. Ecotoxicology and Environ-mental Safety, 61, 256–267. doi:10.1016/j.ecoenv.2004.11.007.

FAO (Food and Agriculture Organization) (1983). Com-pilation of legal limits for hazardous substances in fishand fishery products (pp. 5–100). FAO Fishery Circu-lar no. 464.

FEPA (Federal Environmental Protection Agency) (2003).Guidelines and standards for environmental pollutioncontrol in Nigeria (p. 238).

Forstner, U., & Wittmann, G. T. W. (1981). Metal pollu-tion in the aquatic environment (p. 486). New York:Springer.

Houserova, P., Kuban, V., Spurny, P., & Habarta, P.(2006). Determination of total mercury and mercuryspecies in fish and aquatic ecosystems of Moravianrivers. Veterinarni Medicina, 51(3), 101–110.

Idodo-Umeh, G. (2002). Pollution assessments of Olomorowater bodies using physical, chemical and biological in-dices (p. 485). PhD thesis, University of Benin, BeninCity, Nigeria.

Jorgenson, L. A., & Pedersen, B. (1994). Trace metals infish used for time trend analysis and as environmentalindicators. Marine Pollution Bulletin, 28(4), 235–243.doi:10.1016/0025-326X(94)90099-X.

Kotze, P. J. (1997). Aspects of water quality, metal con-tamination of sediment and fish in the Olifants River,Mpumalanga. MSc thesis Africa University, SouthAfrica.

Kurey, W. J. (1991). New Madrid Refuge ContaminantsReport. US Fish Wild Survey, 695O-H AmericanParkway, Reynoldsburg, Ohio; 43068, 614-469-6923.

Madhusudan, S., Fatma, L., & Nadim, C. (2003). Bioac-cumulation of zinc and cadmium in freshwater fishes.Indian Journal of Fisheries, 50(1), 53–65.

Murugan, S. S., Karuppasamy, R., Poongodi, K., &Puvaneswari, S. (2008). Bioaccumulation pattern ofzinc in freshwater fish Channa punctatus (Bloch.) afterchronic exposure. Turkish Journal of Fish and AquaticScience, 8, 55–59.

NEQS (National Environmental Quality Standards)(1999). Pakistan Environmental Protection Council,Government of Pakistan.

Nussey, G., Van Vuren, J. H. J., & Du Preez, H. H. (2000).Bioaccumulation of chromium, manganese, nickel andlead in the tissues of the moggel, Labeo umbra-tus (Cyprinidae), from Witbank Dam, Mpumalanga.Water S.A., 26(2), 269–284.

Obasohan, E. E., & Oronsaye, J. A. O. (2004). Bioaccu-mulation of heavy metals by some cichlids from OgbaRiver, Benin City, Nigeria. Niger. Annals of NaturalScience, 5(2), 11–27.

Phillips, G. R., & Russo, R. C. (1978). Metal bioaccumula-tion in fisheries and aquatic invertebrates: A literaturereview. Environmental research laboratory, Duluth,MN. EPA; 600/3-78-103.

Ramelow, G. J., Webre, C. L., Mueller, C. L., Beck, J. N.,Young, J. C., & Langley, M. P. (1989). Variationsof heavy metals and arsenic in fish and other organ-isms from the Calcasien River and Lake, Louisiana.Archives of Environmental Contamination and Toxi-cology, 18, 804–818. doi:10.1007/BF01160294.

Sultana, R., & Rao, D. P. (1998). Bioaccumulation pat-tern of zinc, copper, lead and cadmium in greyMullet, Mugil Cephalus (L.), from harbour waters ofVisakhapatnam, India. Bulletin of EnvironmentalContamination and Toxicology, 60, 949–955. doi:10.1007/s001289900720.

Swann, L. (2000). A fish farmer’s guide to understand-ing water quality. Illinois–Indiana Sea grant program,Purdue University. http://ag.ansc.purdue.edu/aquanic/publicat/state/il-in/as-503.htm.

Tomar, M. (1999). Quality assessment of water and wastewater (pp. 110–113). Boca Raton: Lewis.

US Environmental Protection Agency (1999). Nationalrecommended water quality criteria-correction. EPA822/Z-99-001.Washington, DC.

W.H.O. (World Health Organization) (1985). Guidelinesfor drinking water quality (Vol. 1, p. 130). Recommen-dation WHO, Geneva.

Widianarko, B., Van Gestel, C. A. M., Verweij, R. A.,& Van Straalen, N. M. (2000). Associations betweentrace metals in sediment, water and guppy, Poeciliareticulata (Peters), from urban streams of Semarang,Indonesia. Ecotoxicology and Environmental Safety,46, 101–107. doi:10.1006/eesa.1999.1879.