Analysis In Marine Sediments Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.


Elemental (C, N, N/C) and isotopic (δ13C, δ15N) signatures were used as proxies to identify changes in the proportions of sedimentary organic matter in the Mtoni estuary. Sandy particles dominated the mangrove sediments, with less than 6% of organic matter. No general trends in the levels of TOC and TON with depth were observed in all the stations, indicating mixed biogeochemical processes that cause little change with time in carbon and nitrogen in the estuary. Low N/C ratios (0.03 - 0.11) and lower δ13C values (-27.4‰ to -24.7‰) suggest that organic matter in Mtoni estuary sediment originated mainly from terrestrial source. General increase in the ratios towards the ocean in Kizinga and Mzinga streams in the estuary is indication of greater contribution of organic matter from the streams draining into the estuary and selective degradation of autochthonous organic matter resulting in dominant preservation of stable terrestrial organic carbon. Whereas, upstream stations has been categorised by principal component analysis as having relatively high elemental carbon and nitrogen (TPC and TOC for carbon, and TPN and TON for nitrogen), lower N/C and more depleted δ13C and δ15N, other stations were categorised as having relatively high N/C, low elemental carbon and nitrogen as well as more enriched δ13C and δ15N. The δ15N values (3.4 to 8.5‰) detected in the Mtoni estuary sediments were indicative of the terrigenous origin. Lower δ15N indicated that there is additional organic matter in these areas which is fresher, less decomposed, or more labile and lower N/C is indicative of the decreased potential decomposition trend of terrestrial vegetation. Terrigenous contribution was > 63%, with relatively higher contribution from the Mzinga stream than the Kizinga stream.


Polychlorinated dibenzo-p-dioxins, PCDDs and polychlorinated dibenzofurans, PCDFs are tricyclic aromatic compounds, constituting of two benzene rings connected by oxygen atoms (Gevao et al., 2009; Srogi, 2008; El-Kady et al., 2007). Both groups of chemicals may have up to eight chlorine atoms attached at carbon atoms 1 to 4 and 6 to 9 giving rise to several isomers or congeners (Killops and Killops, 2005). As the two chemical families are closely related in structure, they are commonly known as dioxins (Gevao et al., 2009; Smith and Lopipero, 2001) and abbreviated as PCDD/Fs. Polychlorinated biphenyls, PCBs are structurally and chemically similar to the PCDD/Fs, and comprise of mixture of several individual PCB isomers for any degree of chlorination (de Souza et al., 2008; Killops and Killops, 2005; Smith and Lopipero, 2001). Twelve PCBs that have 4 to 8 chlorine atoms, including four non-ortho (IUPAC Nos. 77, 81, 126 and 169) and eight mono-ortho (IUPAC Nos. 105, 114, 118, 123, 156, 157, 167 and 189) are conformationally similar to the PCDD/Fs. These PCB congeners tend to elicit similar dioxin-specific biochemical and toxic responses (through a similar mechanistic action. In this regard, they are often referred to as dioxin-like PCBs (Pan, et al., 2010; Okay et al., 2009; Sanctorum et al., 2007b; Smith and Lopipero, 2001).

PCDD/Fs occur in the environment as unintended by-products of various technological processes and that, they have never been produced commercially (Pan et al., 2010; Roots, et al., 2004). They can be formed during industrial activities including metal industry and during manufacture of various chlorinated chemicals (El-Kady et al., 2007; Ryoo et al., 2005; Müller, et al., 2002) such as wood preservatives and pesticides like pentachlorophenol (PCP). PCDD/Fs originate either from natural combustion processes like bushfires and volcanoes (Birch et al., 2007) or during incomplete combustion processes (Pan et al., 2010) when chlorinated wastes containing chlorine and carbon, like polyvinyl chloride plastics, are incinerated (Terauchi et al., 2009; De Wolf and Rashid, 2008; Birch et al., 2007; Liu et al., 2006; Manahan, 2000). The formation of dioxins in such incinerators takes place due to the presence of both chlorine and catalytic metals (Manahan, 2000). PCBs were once produced worldwide as commercial chemicals (Srogi, 2008; Koistinen et al., 1997). First synthesis of PCBs was described in 1881 (Schmidt and Schultz, 1881), and from 1930 their industrial application started (Erickson, 1997). Since then, PCBs have been found in many industrial and consumer products (Wang et al., 2007; Liu et al., 2006) including anti-corrosion materials, coolants and insulators in heat transfer systems like transformers (Shen et al., 2008; Srogi, 2008), electronic appliances and hydraulic fluids (Yang et al., 2009; Shen et al., 2008) and as capacitors in electrical industries (Pan et al., 2010). Well-known sources of PCBs include the use or disposal of industrial PCB products, by-products of waste incineration (Pan et al., 2010; Wang et al., 2007). Non-ortho PCBs, which do not originate solely from commercial PCB mixtures, can be formed during coal combustion and industrial waste incineration particularly when the combustion temperature is not sufficiently high for destruction i.e. less than 800 °C (Chi et al., 2007). Their commercial utility was based largely on their high chemical stability (low flammability), high thermal stability, electrical insulating properties (El-Kady et al., 2007), electrical resistance and low volatilities (Killops and Killops, 2005).

Marine disposal of sewage sludge and contaminated municipal and industrial wastewater can release a large amount of these anthropogenic compounds resulting into contamination of coastal marine environments (Gevao et al., 2009; Eljarrat, et al., 2001). Natural and anthropogenic sources continuously add various compounds to the aquatic ecosystem where they pose a serious threat because of their toxicity, long time persistence, bioaccumulation, and biomagnifications in the food chain (Kumar, et al., 2008). Increased anthropogenic activities contribute to their elevated levels in the environment (Kumar, et al., 2008). Elevated dioxin and dioxin-like concentrations are therefore associated with highly urbanised and industrialised areas (Müller, et al., 2002). These are the areas where municipal wastewater effluents (Moon et al., 2009), cement manufacture, cigarette smoke (Gevao et al., 2009), combustion processes from waste incinerators, power plants and automobile exhausts (Zhang et al., 2010) as well as industrial processes, such as pulp bleaching and metal refining/melting are found (Bruckmeier, 1997).

Chemically activated luciferase gene expression, CALUX®, is a reporter-gene-based cell bioassay, which uses genetically modified cells (hepatoma cells stably transfected) that respond to chemicals that activate the cytosolar aryl hydrocarbon receptor (AhR) through the induction of luciferase (Kroes et al., 2011). The AhR is a protein complex with graeat affinity and low capacity (one site per molecule) that binds to various polyhalogenated aromatic hydrocarbons (Van Langenhove et al., 2011). As a ligand-dependent transcription factor, the AhR not only binds and activated by dioxins and related chemicals but is also responsible for mediating the toxicity of these chemicals. The cell lines used in the bioassay were the recombinant mouse hepatoma cell lines H1L7.5c1 stably transfected with pGudLuc 7.5 and containing five dioxin responsive domains (DRDs) each with five dioxin response elements (DREs).

CALUX is a useful technique for the rapid screening of total concentrations of the dioxins, furans, and dioxin-like PCBs in sediments (Song et al., 2006). The approach is fast and more cost-effective and offers an alternative for the identification and quantification of the AhR agonist chemicals (Song et al., 2006; Schecter et al., 1999). Once dioxins, furans and dioxin like-PCBs interact with the cytosolar aryl hydrocarbon receptor (AhR), which is a chemical-responsive DNA binding protein (EPA, 2008; Schecter et al., 1999; Murk et al., 1996), they bind together forming an AhR complex that is activated and translocated to the nucleus. In the nucleus, the complex binds to the dioxin responsive elements (DRE) stimulating transcription of a luciferase responsive gene (Sanctorum et al., 2007a; Joung et al., 2007; Murk et al., 1996). The toxicity is produced as a change in gene expression mediated through the AhR, or by interference with other signalling pathways (Hurst, et al., 2004). The different congeners have a different CALUX response according to their relative equivalent potency, REP (Sanctorum et al., 2007a). The relative toxic potencies of different congeners are indicated by their corresponding toxic equivalency factors (TEFs). During chemical analysis, these TEFs in combination with the concentration of each congener are used to calculate the toxic equivalent (TEQ) concentration in a sample (Van Langenhove et al., 2011). The TEQs are then summed, assuming absence of interactions between compounds, similar toxicity mechanism (Keupers, 2010), and hence additive contribution of the PCDD/Fs and PCBs (Sanctorum et al., 2007b). Measurement of the level of activation of AhR-dependent gene expression by a chemical or chemical extract provides a measure by which to estimate the relative potency and toxic potential of these chemicals (EPA, 2008).

Coastal marine environments are known to receive large amounts of chemicals and are considered to be among the most sensitive areas for the accumulation of toxic compounds (Kumar, et al., 2008). They usually act as temporary or primary long-term sinks (Pan et al., 2010; Chi et al., 2007; Müller, et al., 1999) for PCDD/F and PCB contaminants and consequently act as the source of these substances to the ocean and biota (Guzzella et al., 2005). Being the principle reservoirs of environmental pollutants, sediments can be used to evaluate pollutant sources, historical trends, and fate processes of the contaminants (Moon et al., 2009; Lee et al., 2006; Müller, et al., 1999) since the amounts of these compounds in sediments reflect their regional or global discharges. In addition, contaminated sediments may constitute a particular threat to associated biota and other organisms in the marine environment (Zhao et al., 2010).

Although the usage of dioxins and dioxin-like PCBs in Tanzania is not known, the use of these substances in transformers, electrical equipments, ship painting and other industrial activities is common. In the coastal Tanzania, there are a lot of municipal, chemical and even hospital wastes that are directed into the Indian Ocean with incineration and burning being the main treatments. This probably increases the levels of the pollutants. Wood burning is a common source of household emissions as most households use either charcoal or firewood for cooking. In many local households plastics have become a common substance to lighting fire on charcoal when burnt. Vehicle emissions have become common recently due to traffics and importation of used and old cars. There are also public allegations that transformer fluids are used in commercial roasting potatoes due to their low volatility. Therefore, the study was conducted to assess the current levels of dioxins, furans and dioxin-like PCBs congeners in marine sediments to identify possible source and fates of the contaminants in the marine environment. The study also intended to identify the various compound profiles in a way to investigate the pollution sources, their historical inputs as well as the extent of their exposure to the mangrove ecosystem.


Study area

This study was conducted in Mtoni mangrove stand (Figure 1) which is located approximately 3 km south of Dar es Salaam. Mangrove trees are found on both sides of the creek, along a distance of approximately 1.5 km. The Mtoni mangrove stand which is composed of Sonneratia alba, Avicennia marina, Ceriops tagal and Rhizophora mucronata is impacted by the wastewater drainage systems from industrial and residential areas as well as charcoal burning and mangrove harvesting for residential places, salt mining, hotels, and agriculture (Taylor et al., 2002). The anthropogenically impacted Mtoni estuary is drained by two streams, the Kizinga and the Mzinga, which form the dominant creeks of the Mtoni mangrove stand (PUMSEA, 2007). The Kizinga stream is suspected to carry mixed wastes from household, agricultural, as well as industrial origin. The mangrove forest receives the discharges from the stream (Kruitwagen et al., 2008). The estuary further receives inputs from the Dar es Salaam harbour which is located at the mouth of the estuary during diurnal tides (up to 5m amplitude) and from the Mtoni solid waste dumping site located in between the two streams. The area also suffers from industrial wastes from Keko, Chang'ombe, Kurasini and Temeke that discharge various types of inorganic and organic wastes into the Mtoni estuary (Taylor et al., 2002). There is a rapid growth of human settlements along the Mzinga creek as a result of urbanisation and increased human population. It is presumed that sewage and other household wastes are also emptied into the creek.

Figure 1: Map of the Mtoni estuary showing the samping points (E1 and E2 in the Kizinga stream, E3-E5 at the confluence and E6 and E7 in the Mzinga stream


Sampling of sediment was conducted in the mangrove forests during low tides at Kizinga and Mzinga creeks (to show in the Map) of Mtoni estuary, coastal Dar es Salaam, Tanzania. Two sampling campaigns were conducted, one during the wet season (December - March 2011) and second during the dry season (July - September 2011). Samples were collected from exactly the same locations during both campaigns. Seven sampling stations were identified using a hand-held global positioning system (GPS): two in the Kizinga stream, two in the Mzinga stream and three at the confluence of the two streams (Figure 1).

Sediment sampling was done as described in EPA, (2001) using a hand corer (30 cm height, 6 cm internal diameter). The core was plunged in the mangrove sediments and cores were collected by gently pushing it into the sediments. Sediment samples were later extracted by twisting while pulling the core and then sectioned into three segments 0-3, 3-6 and 6-9 cm. Core fractions were then packed in prior labelled and zipped polyethylene bags, stored in iceboxes and later frozen to −20 °C. Sediment samples were then air-transported to the laboratory of the Department of Analytical and Environmental Chemistry, Free University of Brussels, Belgium for dioxins and dioxin-like compounds analyses.

Chemical reagents and standards

Acetone (Pesti-S grade, minimum 99.9%), hexane (minimum 96% assay) and toluene (minimum 99.8% assay) both dioxins and PCB grade, were purchased from Biosolve (The Netherlands). Ethyl acetate (Pestanal, 99.8% assay) was purchased from Sigma-Aldrich (Germany). Sulphuric acid (95 -97% w/w, ACS reagent) and Dimethylsulfoxide (DMSO) were obtained from Biosolve (The Netherlands). Glass fiber filters were purchased from Whatman (UK). Alpha-minimal essential medium (α-MEM) and 10% fetal bovine serum (FBS) were obtained from Gibco, UK. Trypsin (0.25%) and phosphate buffer saline were obtained from Ambion (UK). Luciferase assay substrate and buffer were purchased from Promega (USA). Anhydrous sodium sulphate was purchased from Boom (The Netherlands) and the X-CARB from Xenobiotic Diagnostics Syetems XDS, USA. The standard solution of 2,3,7,8-TCDD (50 ng/mL, purity 99%) was purchased from Campro Scientific (The Netherlands).

Determination of Particle Size and Total Organic Carbon

Lyophilised mangrove sediment sample (10 - 50 g) for grain size analysis was placed in a top sieve of an analytical vibratory sieve shaker (Retsch AS 200 Control, GmbH &Co, Germany) operating at an amplitude of 1.50 mm for 10 min. Particle were separated into fractions on the basis of particle size as < 63 µm, > 63 µm, > 125 µm, > 250 µm, > 500 µm and > 2000 µm. The different fractions obtained in every sieve were then weighed and the mass of each fraction expressed as a percentage of the total mangrove sediment. Total organic carbon (TOC) content in sediment was determined using a Flash 1112 EA Elemental Analyser (Thermo Finnigan, Italy) by analysis of sub-samples after removal of inorganic carbon in the sample by acidification with 5% HCl.

Sample preparation for dioxin and dioxin-like compounds analyses

Lyophilised sediment (2 g) was extracted using pressurised liquid extraction in an Accelerated Solvent Extractor (Dionex, USA) with toluene:methanol (4:1 v/v) solvent system and 33 mL extraction cells. The ASE extraction conditions were: oven temperature = 100 °C; Pressure 1500 psi (100 MPa); Static time = 10 min; Oven heat time = 6 min; Purge time 60 s; Flush volume = 60% of extraction cell volume and static cycles = 2. The extracts were then concentrated to dryness in a vacuum centrifuge and re-suspended in 5 mL hexane.

Activated Copper Column Preparation for Clean up

Activated copper column was prepared by filling a Pasteur pipette from bottom to top with glass wool and activated copper (making up 1 cm length) previously made by dissolving metallic copper in 20% hydrochloric acid. The copper column was then rinsed with Mill-Q water (3 x 1 mL) and then with acetone, toluene and hexane (each 3 x1 mL) in that order. Later, the activated copper column was stored submerged in hexane to avoid oxidation ready for use.

Acidified Silica Column Preparation for Clean up

A 10-mL Pyrex disposable column (Sigma Aldrich, Germany) was filled, from bottom to top, with glass wool, sodium sulphate (0.5 cm3 @ 0.7 g), acidified silica gel (4.5 cm3 (@ 3 g) of 33% (w/w) sulphuric acid and sodium sulphate (0.5 cm3 @ 0.7 g). The silica column was then rinsed with hexane (3 x 10 mL).

X-CARB Column Preparation for Clean up

A 10-mL Pyrex disposable column (Sigma Aldrich, Germany) was filled from bottom to top with glass wool, sodium sulphate (0.5 cm3 @ 0.7 g), 1% X-CARB from Xenobiotic Detection Systems Inc., USA (1 cm3 packed @ 0.34 g @ 2 cm length), sodium sulphate (0.5 cm3 @ 0.7 g) and glass wool, and rinsed sequentially with acetone (5 ml), toluene (20 ml) and hexane (10 ml) in the mentioned order.

PCDD/Fs and PCBs Clean Up and Fractionation

The acidified silica column was then placed above the X-CARB column and in between was the activated copper column. The sediment extract in hexane was first sonicated for 5 min and later concentrated sulphuric acid (2.5 mL) was added followed by vigorous vortexing. The resulting supernatant as well as the hexane vial rinses (3 x2 mL) were quantitatively loaded on the acidified silica column. The column was later eluted with hexane (3 x 5 mL). Then, the acidified silica column was rinsed with hexane (3 x 5 mL) and removed to dry in designated fume hood when the last solvent has passed. The activated copper column was also removed once the solvent has passed through. The remaining X-CARB column was rinsed with extra hexane (5 mL) and then eluted with mixture (8:1:1) of hexane:toluene:ethylacetate (3 x 5 mL) to get the fraction containing coplanar PCBs (i.e. PCB fraction). The fraction containing the PCDD/Fs (dioxin fraction) was afterwards eluted with toluene (3 x 5 mL) after flipping the X-CARB column. The PCB and dioxin fractions in borosilicate vials were later concentrated to dryness in a vacuum centrifuge and resuspended in hexane (4 mL) for CALUX analysis.

Preparation of the cells for the CALUX analysis

CALUX analysis for the Mtoni estuary sediments was done as described by Van Langenhove et al., (2011). Briefly, mouse hepatoma cells (H1L7.5c1 cell line) were cultured in α-MEM supplemented with 10% FBS at 37 °C and 5% CO2 in an atmosphere saturated with water (80% relative humidity). Prior to cell harvesting, the α-MEM was removed as it contains Ca2+ and Mg2+ that interefere with trypsin. The cells were further rinsed with phosphate buffer saline (5 mL). Cells were harvested by adding trypsin (0.05%, 1.5 mL) followed by gently swirling the plate to detach them. The cells were then incubated at similar environmental conditions for 4 min. The trypsin action was then stopped by adding α-MEM (20 mL) and the cells centrifuged at 1100 rpm for 3 min. After counting and dilution of the cells, the 96-well culture plates (Perkin Elmer, USA) for the CALUX bioassay were seeded with 100 µL of the resulting cell suspension in α-MEM at a density of 7.5 x104 cells/mL. The plates were then incubated at similar environmental conditions for 24 hours prior to dosing.

Dosing the cells

After 24 h, the media (α-MEM) was removed from the cells. TCDD standards were prepared by dilution of the intermediate standard (1.537 x 102 µM) forming 10 working standards with concentrations in DMSO (0.3 pM, 9.5 pM, 70.6 pM, 0.304 nM, 0.608 nM, 1.22 nM, 2.43 nM, 9.73 nM, 77.8 nM and 1.25 µM). Before dosing, DMSO (4 µL) were then added to TCDD standards and sample extracts in hexane and centrifuged under vacuum. Later, α-MEM were added to each standard and extract so that the concentration of DMSO becomes 1% v/v. The resulting standards and the sample extracts were then dosed to the cell lines in triplicate (100 µL/well) and incubated again for another 24 hours.

Reading the Plates

After 24 h of incubation, the media in the dosed plate was removed, The wells were later rinsed with PBS buffer (50 µL; pH 7.4) and the cells examined microscopically for obvious toxicity of the extract and any altered morphology. In absence of extract toxicity, 50 µL of cell culture lysis reagent (Promega, USA) were added to each well and the plate was shaken for 9 min at room temperature. White backing tape (Perkin Elmer) was applied to the bottom of the plate before being placed in the Glomax 96-well microplate luminometer (Promega, USA). Prior to the light output, 50 µL of reconstituted luciferase assay reagent (luciferin., Promega, USA) was automatically injected. The light output was then integrated after a delay time of 5.6 s, and the results expressed in relative light unit (RLU). The average RLU value measured for triplicate blanks (DMSO alone) was subtracted from all RLU values and an average RLU value for the triplicate wells with the same extract was calculated. The values were reported as percentage of maximum RLU induced by TCDD.

Statistical Analysis of Data

TCDD standards in DMSO were used to generate the calibration curve. To describe the response effect of a given dose of pollutant, a four-variable Hill equation fitting the calibration curve was used to produce a sigmoid curve of the standard solutions:

where [x] is the concentration or mass unit of the analyte (i.e TCDD), m is the maximum efficacy or limiting value of the RLU response as the concentration of TCDD increases, k is the dose corresponding to 50% of the maximum dose response, h is a parameter that defines the sigmoid shape of the curve, yo is as experimental background noise, yi ia a normalised response to the maximum efficacy of the respecive pollutant and ε is the residual term.

The equation was also used to convert the measured RLU values of the samples into toxic equivalency value (CALUX-TEQ) by comparing the sample response with the 10-point sigmoid dose-response curve. From the graph, the estimated model parameters (m, k, h, and yo) were used to determine the EC20, EC50 and EC80. Potency or biological equivalency (BEQ) was assessed as EC50 ratio (EC50 of TCDD/ EC50 of sample) for PCDD/Fs giving pgTCDD per g matrix.

Quality Control

For each batch of samples, a blank was introduced through the complete treatment procedure (extraction blank) to monitor the activity contributed by solvents and column matrices used in the sample treatment. In addition, another blank was introduced at the clean up (clean up blank) to monitor the background activity associated with extraction and clean up processes. Moreover, DMSO treatment and media blanks were added during dosing to detect contamination and experimental background level. All the blanks were in triplicate and were treated in a similar way as real samples. The results with p < 0.05 were considered statistically significant. Since a statistically significant result maynot be significant practically, correlation coeefficient was used to evaluate the practical significancy of the results.

The detection limit (LOD) was calculated from the DMSO blank plus 3 times the standard deviation of the mean DMSO blank signal. LOD for the PCDD/Fs calculated from 10 plates was 46.9 fg well-1, which corresponds to 0.24 pg CALUX-BEQ/g matrix. The LOD for the dioxin-like PCBs were 0.06 pg/well, corresponding to 0.24 pg CALUX-BEQ/g matrix. The data obtained were expressed in pg CALUX-BEQ/g sample.

Results and Discussion

Grain size Distribution

Sandy particles dominated the mangrove sediments in the study area, with fine sand (63 μm < x < 125 μm) contributing up to 53% of the weight. Unpaired Student' two tailed t-test was performed to determine if there was any significant difference between samples collected during two different seasons (wet and dry seasons). The results indicated that there was no significant difference in grain sizes between wet and dry season (t =-1.156, p > 0.05). The average grain size during the two seasons indicated that clay and silt content in the sampling sites were < 6%, indicating that sand is the main component of the sediments in the study area (Table 1). The high sandy fraction and low % mud (silt and clay) in the mangrove sediment samples has impact on assemblage and bioavailability of various micro-pollutants (Davies & Tawari, 2010). Low %TOC and high sand content imply that the capacity of these mangrove sediments to adsorb the dioxins and dioxin-like PCBs is low. Therefore, it is not surprising to find low PCDD/Fs and dioxin-like PCBs in these sediments.

The CALUX H1L7.5c1 Assay

The CALUX bioassay integrates the responses of every AhR ligand available in the analysed sample and because of this, it provides only the overall toxicity (Van Langenhove et al., 2011). The results presented hereunder assume the additivity principle holds as the sulphuric acid treatment and the clean up step of the samples eliminated the non-additivity caused by polychlorinated aromatic hydrocarbons (PAHs). In addition, separation of PCDDs and PCDFs eliminates the antagonistic effects that would further validate the non-additivity principle (Van Langenhove, et al., 2011).

Mean values of luciferase response/TCDDmax response measured in three replicate wells were used to generate the dose-response curves. The dose-response curve of TCDD standard was sigmoidal in appearance as shown in Figure CYZ

Figure CYZ: TCDD Standard calibration curve for mouse hepatoma H1L7.5c1 cell line with the working range (the linear range) indicated by arrows

Luciferase induction was reproducible with coefficients of variation (CV) less than 15%. Sample responses were expressed as percentage maximum induction to 2,3,7,8-TCDD (%TCDDmax). Based on the assumption that environmental samples cannot exhibit equal eficacy to TCDD (Villeneuve et al., 2000, Besselink et al.,2004)., the 50% TCDD max (EC50) of samples was used.

A total of 72 quality control experiments were used as QC in this study and the amount of TCDD was theoretically set to be 0.250 pg/g. When these experiments were plotted on a control chart (Figure XyXY), a mean of 0.241 ± 0.017 pg/g indicated that the procedures were reliable with coefficient of variation (CV) being 7.2%. Whereas the recoveries of dioxins ranged from 92.9% to 114.3%, recoveries of dioxin-like PCBs ranged between 85.8% and 119.5%

Figure XyXY: Quality Control (QC) scatter chart of analyses during the study. The 1st and 2nd dashed lines above and below the mean (0.2406) the 1st and 2nd standrad deviations, respectively

Determination of Biological Equivalency (BEQ) in Mtoni sediment samples

BEQ in Mtoni sediments was assessed using three methods: the Hill regression equation, Box - Cox transformation and slope ratio methods. The relationship between these methods indicated that they correlate well (r2 = > 0.81, Figure XXX), implying that either of the method can provide reliable results in this study. Precision of these methods were within 20% RSD. REQUEST VALUES FROM KERSTEN

Figure XXX: Relationship between PCDD/Fs Effective concentrations (EC50) in samples estimated by Hill regression and Box - Cox transformation (a), Hill equation and slope ratio method (b) and Box - Cox transformation and Slope ratio (c)

In this study, Hill regression BEQs were used in data presentation. To account for non-parallelism that usually exist between dose-response curves of the reference and the sample, relative effect potency (REP) range of PCDD/Fs in a sample was determined using effective concentrations (EC) eliciting 20% (EC20), and 80% (EC80) of TCDD. Table 1 gives the site properties as well as the REP of samples used in the study.

Table 1: Mtoni estuary sediment properties and the REP values for PCDD/Fs

Code and location

Section depth



(silt +clay)

TOC (%)

REP (EC20-EC80) pg/g




S 06°52.443

E 039°17.014

0-3 cm









3-6 cm









6-9 cm










S 06°52.357

E 039°17.099

0-3 cm









3-6 cm









6-9 cm











E 39°17.355

0-3 cm









3-6 cm









6-9 cm










S 06°52.090

E 039°17.501

0-3 cm









3-6 cm









6-9 cm










S 06°52.164

E 039°17.658

0-3 cm









3-6 cm









6-9 cm










S 06°52.882

E 039°18.391

0-3 cm









3-6 cm









6-9 cm










S 06°52.952

E 039°18.454

0-3 cm









3-6 cm









6-9 cm









Dioxin-like PCBs in samples did provide higher responses owing to the low induction. Their potency was therefore determined using inverse prediction assuming a sample is a diluted TCDD (Elskens et al., 2011).

Dioxin-like activity in Mtoni sediments

The results of the CALUX screening with the H1L7.5c1 assay of Mtoni sediment extracts are given in Figure XYZa. The dioxin concentrations based on the EC50 ranged from 2.82 ± 0.42 to 68.78 ± 4.91pg BEQ/g in wet season and between 8.52 ± 0.79 and 57.42 ±+ 3.84 pg BEQ/g in dry season. Higher levels of dioxins in both seasons were found in Kizinga stations (wet season mean 24.07 ± 1.12 pg BEQ/g and dry season mean 28.33 ± 1.11 pg BEQ/g) than in Mzinga (wet season mean 17.03 ± 0.98 pg BEQ/g and dry season mean 15.58 ± 0.89 pg BEQ/g) and confluence (wet season mean 22.64 ± 0.80 pg BEQ/g and dry season mean 26.25 ± 1.10 pg BEQ/g) stations. PCDD/Fs levels were decreasing towards the ocean in the Kizinga and increasing in the Mzinga. Higher PCDD/F levels at confluence station E3 in both seasons could be attributed to a localised source. This can be affirmed by the subsequent higher levels at the nearby station E4 in both seasons. No significant difference in dioxin levels was observed (t = 0.547 two tailed; p > 0.05) between the two seasons.

Response levels of dioxin-like PCBs (Figure XYZb) ranged from 0.14 ± 0.03 to 0.60 ± 0.02 pg BEQ/g in wet season and from 0.10 ± 0.03 to 1.03 ± 0.03 pg BEQ/g in dry season. Higher PCB levels in wet season were detected in Mzinga stations (mean 0.37 ± 0.12 pg BEQ/g) as compared to Kizinga (0.34 ± 0.19 pg BEQ/g) and confluence stations (0.31 ± 0.17 pg BEQ/g). Higher levels in dry season were detected in Kizinga stations (0.48 ± 0.22 pg BEQ/g) than Mzinga (0.26 ± 0.16 pg BEQ/g) and confluence (0.33 ± 0.12 pg BEQ/g) stations. Levels of dioxin-like PCBs were increasing towards the ocean during wet season and decreasing during dry season in the Kizinga stream. In the Mzinga however, the levels were slightly decreasing during the wet season and increasing in the dry season. PCB levels detected in Mtoni estuary showed no significant difference (t =0.270 two tailed; p > 0.050) between the seasons.

Higher values of the micro-pollutants in the study area can be explained by the position it occupies in relation to the habour. Dioxins and dioxin-like PCBs could be transported from the harbour during tides to the area. Moreover, domestic use of wood burning as source of energy in nearbouring human settlements can be a contributing factor to detected PCDD/FS and PCBs in the stations. In addition, availability of polymeric materials (such as household scraps, plastics, vehicle tires and electronic wastes) in domestic and industrial wastes subjected to open burning can lead to formation of these chemicals (Estrellan & Iino, 2010) and hence their transport and deposition in the study area.

The higher values detected in dry season in Kizinga and confluence stations could be attributed to the wind. The study area is affected by north easterly, south easterly and easterly winds. The easterly (April) and south easterly (May - September) winds could be responsible for transporting dioxins and dioxin-like PCBs coming from burnt dumps and domestic garbage into the areas. In addition, rains could have facilitated wet deposition and consequently elevated levels in wet season than dry season. Moreover, contributions from localised sources near the stations (i.e. small scale domestic garbage burning and the factory upstream of Kizinga stream could be significant. More or less similar levels detected in both seasons for Mzinga stations can be explained by an additional north easterly winds (October - March), which seem not to affect much the Kizinga and confluence stations. The higher levels in station E7 in the Mzinga stream could be explained by the upstream position it occupies and a localised source that is due to anthropogenic activities.

PCDD/Fs values in sediment samples were generally decreasing with increasing depth(Figure XYZc). Dioxin-like PCBs were also decreasing with increasing depth (Figure XYZd) except with the exceptions of station E1 in the Kizinga, station E4 at the confluence and station E7 in the Mzinga. When the levels were normalised to %TOC (Figure XYZe-f), PCDD/F levels were generally decreasing (Figure XYZe) as previously indicated except at station E2 where the levels were increasing.

Figure XYZ: Mean BEQ for PCDD/Fs (a) and PCBs (b) in Mtoni sediment and the corresponding depth profiles before ((c) and (d)) and after normalisation to % TOC ((e) and (f))

Dioxin-like PCBs were also decreasing with increasing depth (Figure XYZf) for all stations except stations E2 in Kizinga, E5 at the confluence and E7 in Mzinga where the levels were increasing with depth. Levels in station E1 in Kizinga were increasing and decreasing with depth. PCDD/Fs and PCBs are recalcitrant chemicals and that they will tend to acumulate with time in sediments. Therefore, deeper sediment layers were expected to have higher pollutants levels that upper ones. The findings have contradicted the common understanding probably due to the diluting efects of ocean (Pieters, 2007) as well as high hydrodynamics of the estuary that provide frequent sediment mixing and do not allow for further biogeochemical processes. This can be supported by the relatively low variation of pollutant levels with depth in almost every sampling point.

Correlation between pollutants and sediment geochemical characteristics

Spearman correlation coefficient was performed to pollutants levels as well as sediment geochemical characteristics (Table qAX). Significant correlation was observed between PCDD/Fs and dioxin-like PCBs (r2 = 0.453, p <0.05) and between PCDD/Fs and δ15N (r2 = 0.573, p < 0.05). However, no significant correlations (p > 0.05) were found for PCDD/Fs and TOC (r2 = -0.182,), PCDD/Fs and %mud (r2 = -0.159). No significant correlations were also observed between dioxin-like PCBs and %TOC (r2 = 0.012) and PCBs and %mud (r2 = -0.284). This is indicated that %TOC and %mud (silt + clay) have no direct relationship with PCDD/Fs and dioxin-like PCBs. Thus dioxin and dioxin-like PCB pollutant load was not related to sediment geochemical characteristics.

Table qAX: Correlation coefficients between pollutants and various geochemical parameters
































































* significant at α = 0.05

General knowledge shows that %TOC, %mud (silt + clay), PCDD/Fs and dioxin-like PCBs will be associated together, that is, well correlated because organic pollutants tend to preferably associate with organic fractions. The weak correlation between the pollutants and %TOC is contrary to the theoretical expectation similar to Hilscherova et al., (2003). However, the observation was contrary to those made by Koh et al., (2004, 2006) and Pieters, (2007). Good and significant correlation (p < 0.05) was observed between %TOC and %TN (r2 = 0.824), %TOC and %TC (r2 = 0.879) and between %TN and %TC (r2 = 0.936) as expected and previously observed (Chapter Two).

Pollutant Source analysis by Principal Component Analysis (PCA)

Multivariate analysis can be used to identify similarities and differences between pollutants in samples as a means to suggest for possible sources. Principal component analysis (PCA) was used to illustrate for any similarities and/or differences that exist between pollutants and the sediment geochemical parameters in sediments. Principal components (PCs) were regarded significant only when their eigenvalues were greater than 1. The pollutants (PCDD/Fs and dioxin-like PCBs) measured and geochemical properties (%mud (clay + silt), %TOC, %TN and %TC) were used as variables (total 6), with the mean concentrations of pollutants in the different sites as objects (total 21).

The results have indicated that the 21 variables can be represented by two new principal components that accounted for 81.724% of the total variance in the original data sets (Table PZYA). Based on the loading distribution of the variables, %TC, %TN, %TOC and %mud constituted one related group (PC1). Another group (PC2) constituted the pollutants PCDD/Fs and dioxin-like PCBs.

Table PZYA: Rotated Principal component (PC) loadings of pollutants and sediment parametersa

Principal Components

PC 1 (56.145%)

PC 2 (25.579%)



















a The percentage in brackets indicates the explained contribution of principal component to the total variance after varimax normalisation; bPercentage mud (<63 µm, silt + clay) in the marine sediment.

Association of variables in one PC is indicative of their source. Presence of all geochemical properties in one PC indicated their origin as observed and was therefore expected. Higher loading indicated by the pollutants in PC 2 clearly indicated that both pollutants originated from similar source, which is not related to geochemical parameters.

Comparison with Other Studies and with Sediment Quality Guidelines

Table YYYYZ: Comparison of PCDD/F and PCB levels (pg/g d.w) detected in other global marine environments

Study site




Mtoni estuary, Tanzania

5.72 - 39.86

0.22 - 0.59

This study1

Hong Kong2

71 - 6000


Terauchi et al., 2009

Daliao estuary, China2

11.3 - 133.2

1971.2 -37632.4

Zhao et al., 2011

Belgian coast

0.02 - 42.2


Sanctorum et al., 2007

Mtoni estuary, Tanzania2


3080 -3650

Kruitwagen et al., 2008

Shandong Peninsula, China2

47.7 - 147

21.3 - 54.4

Pan et al., 2010

Korea coast2

10.27 - 395.66


Lee et al., 2006

Changjiang estuary, China2

19.33 - 351.68


Sheng et al., 2008

River sediments, South Africa

0.82 - 45


Nieuwoudt et al., 2009

Yeongil Bay, Korea3

1.08 - 311


Koh et al., 2006

1mean BEQ (n =3); 2Values measured by chemo-analysis (GC-MS and GC-ECD); na = not analysed; 3Values indicate the ∑(PCDDs + PCDFs +PCBs)

An attempt was made to compare the normalised levels of PCDD/Fs and dioxin-like PCBs with the sediment quality guidelines. Since Tanzania lacks these guidelines, National Oceanic and Atmospheric Administration (NOAA), USA and Canadian Sediment quality guidelines were used to predict the toxicity of pollutants with the already set threshold effect level (TEL) and probable effect level (PEL). Normalised mangrove sediments from Mtoni estuary had higher dioxins (3.91 - 42.54 pg/g TOC) levels compared to NOAA apparent effect concentration (PEC) of 3.6 pg/g TOC. However, the levels were lower than the Canadian sediment quality guideline for TEL (0.85 pg/g), but slightly higher than the PEL (21.5 pg/g TOC). Since there were no available NOAA guidelines for PCBs, comparison to dioxin-like PCBs levels to sediment guidelines was done using only Canadian guidelines. PCB levels were lower (0.06 - 0.55 pg/g TOC) than the TEL (2.15 x 104 pg/g) and PEL (1.89 x 105 pg/g TOC).

The comparisons have shown that currently there is no risk associated with dioxin-like PCBs. However, there is a risk as regards to PCDD/Fs using the NOAA guideline. Because the toxicity of these chemicals can be additive, their presence in mangrove sediments may cause impairments particularly to sensitive benthic species. The effects can reach other organisms higher in the trophic level due to their bio-accumulation, bio-concentration and persistence properties. The higher levels of pollutants observed in the frequently exchanged upper (0 - 3 cm) layer can have impacts on the distribution and fate of pollutants to mangrove ecosystem and to higher organisms that use mangrove sediment organisms as food.


The present study is the first investigation using in vitro bioassay of dioxin-like compounds in mangrove sediments collected in this area. CALUX assay has been effectively used to screen for compounds with a dioxin-like response in marine sediment and has enabled to give a general overview of the mixtures of the compounds in the Mtoni estuatine sediments. Quantifiable PCDD/Fs and dioxin-like PCBs were detected in Mtoni estuary indicating the ecological and human risks that may emanate from these chemicals. Lower PCBs than PCDD/Fs is probably indicative of the level of industrialisation and also implying the banned use of the products. Higher PCDD/Fs levels may imply that these products are being produced in various industrial and domestic activities, taking into consideration the increased population and urban migration that could result to increased socio-economic activies and hence increased production of PCDD/Fs generating wastes.


This work was supported by the Belgian Technical Cooperation (BTC) under the Belgian Development Agency through a scholarship offer to MJ Mihale.