Available online at www.sciencedirect.com
Chemosphere 71 (2008) 447–455 www.elsevier.com/locate/chemosphere
Relationship between blood concentrations of heavy metals and cytogenetic and endocrine parameters among subjects involved in cleaning coastal areas affected by the ‘Prestige’ tanker oil spill Beatriz Pe´rez-Cadahı´a a,b, Blanca Laffon a,b,*, Miquel Porta c,d, Anunciacio´n Lafuente e, Teresa Cabaleiro e, Toma`s Lo´pez c, Ana Caride e, Jose´ Pumarega c, Alejandro Romero e, Eduardo Pa´saro a, Josefina Me´ndez b b
a Toxicology Unit, University of A Corun˜a, Edificio de Servicios Centrales de Investigacio´n, Campus Elvin˜a s/n, E-15071 A Corun˜a, Spain Department of Cell and Molecular Biology, University of A Corun˜a, Faculty of Sciences, Campus A Zapateira s/n, E-15071 A Corun˜a, Spain c Institut Municipal d’Investigacio´ Me`dica, Barcelona, and CIBERESP, Spain d School of Medicine, Universitat Auto`noma de Barcelona, Spain e Toxicology Laboratory, Faculty of Sciences, University of Vigo, Campus Ourense, Las Lagunas, E-32004 Ourense, Spain
Received 8 May 2007; received in revised form 18 October 2007; accepted 28 October 2007 Available online 25 January 2008
Abstract The sinking of the ‘Prestige’ oil tanker in front of the Galician coast (NW of Spain) in November 2002 offered a unique opportunity to analyze intermediate cytogenetic and endocrine effects among people exposed to the complex mixture of substances that oil constitutes, including several toxic heavy metals. In this work we evaluated the relationship between exposure to heavy metals (blood concentrations of aluminium, cadmium, nickel, lead and zinc) and genotoxic parameters (sister chromatid exchanges, micronucleus test and comet assay) or endocrine parameters (plasmatic concentrations of prolactin and cortisol) in subjects exposed to ‘Prestige’ oil during cleaning tasks developed after the spillage. Concentrations of lead were significantly related to the comet assay even after adjusting by age, sex and smoking. Cortisol concentrations were significantly influenced by aluminium, nickel (both, inversely) and cadmium (positively). Women had clearly higher concentrations of prolactin and cortisol, even when adjusting by age, smoking, cadmium, aluminium or nickel. Plasmatic cortisol was jointly influenced by gender, smoking and aluminium or nickel (all p < 0.05). In women there was a strong relationship between concentrations of cadmium and prolactin (b = 0.37, p = 0.031). When the effects of cadmium, aluminium and nickel on cortisol were simultaneously assessed, only the latter two metals remained statistically significant. Among parameters analysed, cortisol appeared to be the most sensitive to the effects of metal exposure. Plasma levels of cortisol deserve further evaluation as a potentially relevant biomarker to assess the effects of exposure to heavy metals. Ó 2007 Elsevier Ltd. All rights reserved. Keywords: Sister chromatid exchanges; Micronucleus test; Comet assay; Prolactin; Cortisol
1. Introduction Heavy metals constitute a wide and special group of chemical agents: while many of them are essential to * Corresponding author. Address: Toxicology Unit, University of A Corun˜a, Edificio de Servicios Centrales de Investigacio´n, Campus Elvin˜a s/n, E-15071 A Corun˜a, Spain. Tel.: +34 981 167000; fax: +34 981 167172. E-mail address: blaff
[email protected] (B. Laffon).
0045-6535/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2007.10.053
the development of a variety of physiological functions, others, as cadmium and lead, do not have well recognised roles or are toxic, even at low doses (Thompson et al., 1992; Merzenich et al., 2001). Furthermore, many heavy metals bioaccumulate in several compartments of the human body and have slow excretion rates (Elinder et al., 1994; Hayes, 1997). Hence, in some circumstances exposure to these agents may be a public health concern.
448
B. Pe´rez-Cadahı´a et al. / Chemosphere 71 (2008) 447–455
Among the large number of heavy metals present in the human environment several stand out because of their well known toxicity and noxious effects on living organisms: they include aluminium, cadmium and nickel, which are classified as human carcinogens (group 1) by the International Agency for Research on Cancer (IARC), and lead, classified as possible human carcinogen (group 2B). Besides interference at distinct steps of several DNA repair mechanisms (Hartwig and Schwerdtle, 2002), some of their carcinogenic effects are related to DNA–protein crosslinking (Kasprzak, 2002); the effects of metals on chromatin have been observed in vitro and in vivo. Some metals may also contribute to cause diseases of complex aetiology through indirect genotoxic and epigenetic mechanisms (Luch, 2005). In addition, some metals have been characterized as endocrine disruptors, and one of their targets is the hypothalamus-pituitary-gonadal (Lafuente et al., 2003, 2004a, 2005) and/or hypothalamus-pituitary-adrenal axis (Stejskal et al., 2006), which is crucial for vertebrates to cope with stressors (Lacroix and Hontela, 2004). Plasmatic prolactin and cortisol are especially sensitive to physical and psychical stress (Sobrinho, 2003; Dahlgren et al., 2005). Aluminium and its compounds are common food and water contaminants. While no biological function of this element has been identified, some aspects of aluminium toxicity are well described (Berthon, 1996; Corain et al., 1996; Reinke et al., 2003). Its relation to some disorders such as microcytic anaemia, and to neurodegenerative disorders as Alzheimer’s and Parkinson’s diseases have been assessed (Berthon, 1996; Corain et al., 1996; Yokel, 2002). It has been also reported that physiologically relevant amounts of aluminium are capable of inducing alterations in gene expression programs that may support downstream pathogenic responses and brain cell dysfunction (Alexandrov et al., 2005). Cadmium is one of the most toxic heavy metals, capable of generating mutagenic and genotoxic events (Waalkes, 2003). The carcinogenicity of cadmium and nickel have long been recognized; mechanisms include the induction of oxidative DNA damage (Kasprzak, 1995; Bal et al., 2000) and epigenetic alterations like gene silencing by changes in DNA methylation patterns (Klein and Costa, 1997; Luch, 2005). Because of its persistence in the environment and effects on humans, exposure to lead has long been of public health concern. In eukaryotic cells this metal is usually genotoxic, though mechanisms are only partly known; they possibly involve indirect damage to DNA affecting the stabilization of chromatin, or interaction with repair processes (Vaglenov et al., 2001). Zinc stands out for having some opposite effects to the other metals: it is co-released with neurotransmitters by neurons, and zinc supplements may delay the progression of Alzheimer’s disease (Potocnik et al., 1997). Protective effects of zinc against cadmium and lead toxicity have also been described (Leblond and Hontela, 1999; Nampoothiri et al., 2007).
The sinking of the ‘Prestige’ oil tanker in front of the Galician coast (NW of Spain) in November 2002 offered a unique opportunity to analyze intermediate cytogenetic and endocrine effects among people who were exposed to a variety of environmental agents while cleaning the coast (Porta et al., 2004; Gestal et al., 2004). The composition of ‘Prestige’ oil was determined by the Spanish National Research Council as 50% aromatic hydrocarbons, 22% saturated hydrocarbons and 28% resins and asphalthene (CSIC, 2003a). In addition, presence of different quantities of heavy metals in emulsified samples (with 54–59% water) of ‘Prestige’ oil was also confirmed (CSIC, 2003b; Albaige`s et al., 2006). In order to test the hypothesis that exposure to some heavy metals might be associated with genotoxicity events or several endocrine alterations, we evaluated the relationship between blood concentrations of aluminium, cadmium, nickel, lead and zinc and genotoxic or endocrine effects in a human population exposed to ‘Prestige’ oil during cleaning tasks developed after the spillage. The assays sister chromatid exchanges (SCE), micronucleus (MN) test, and single cell gel electrophoresis (comet) assay were selected as genotoxicity biomarkers, and plasmatic prolactin and cortisol as indicators of endocrine alterations. 2. Materials and methods 2.1. Study population The study population comprised 179 healthy individuals who participated in the cleaning of coastal areas affected by the ‘Prestige’ oil spill, which occurred on November 2002. Their mean age was 32 years and 49% were women (Table 1). Sixty individuals were college student volunteers (V) from Spanish universities engaged in the cleaning of the oil in beaches for five consecutive days (from Monday to Friday) during 4 h day 1. The other subjects were workers hired by the Galician government for the manual collection of the oil in beaches (MW, N = 59), and workers hired to clean the rocks using high-pressure machines (HPW, N = 60) during 6.5 h day 1. At the time of the study MW had been exposed to oil for 4 months and HPW for 3 months. Information on smoking was elicited through a self-administered questionnaire. Smoking was more common among males (Table 1); it was thus treated as a potential confounder in statistical analyses. None of the study participants had undergone diagnostic tests (e.g., X-rays) or received significant medical treatment in the 3 months prior to the study. Peripheral blood samples were collected between 24 March and 12 May 2003. Blood was drawn before the beginning of the working shift in the morning (08:00– 09:00 am), codified to ensure a blind study, and immediately transported refrigerated to the laboratory. Two samples were obtained from each donor, one in a lithium heparin container for determination of heavy metals, hormones, sister chromatid exchanges (SCE) and the micronu-
B. Pe´rez-Cadahı´a et al. / Chemosphere 71 (2008) 447–455 Table 1 Characteristics of the study population Total
Total number of individuals Age (years) Mean ± SD Median Exposure group College volunteers (V) Manual workers (MW) High-pressure machine workers (HPW) Smoking Non-smokers Smokers Cigarettes day 1e (median) Years smokinge (median) a b c d e
Gender
179
Males
Females
N (%)
N (%)
92 (51.4)
87 (48.6)
p-Valuea
32.4 ± 11.7 33.1 ± 11.4 31.7 ± 11.9 0.434b 28.5 31.0 26.0 0.293c 0.010d 60 (33.5) 23 (25.0) 37 (42.5) 59 (33.0) 39 (42.4) 20 (23.0) 60 (33.5) 30 (32.6) 30 (34.5)
60 (69.0) 27 (31.0) 10.0
0.034c
9.0
9.5
0.995c
8.0
Cells were stored at 80 °C in a NalgeneÒ Cryo 1 °C Freezing Container (Nalgene Nunc International) in frozen medium (40% RPMI 1640, 50% foetal bovine serum and 10% DMSO) and until their use. Viability was checked with trypan blue, being always P85%. Comet assay was performed following Singh et al. (1988) with minor modifications (Laffon et al., 2002b). QWIN Comet software (Leica Imaging Systems, Cambridge, UK) was used to capture and analyse one hundred cells per donor (50 per replicate slide). The percentage of DNA in the comet tail (%TDNA) was used as DNA damage parameter. 2.4. Hormone (prolactin and cortisol) measurement
0.033d 109 (60.9) 49 (53.3) 70 (39.1) 43 (46.7) 10.0 17.5
449
Plasma was obtained from blood samples by centrifugation at 3500 rpm for 15 min at 4 °C. Specific radioimmunoassays were used to determine plasmatic prolactin and cortisol levels using two commercial kits, previously tested: Prolactin IRMA DSL-4500 and Cortisol RIA DSL-2000 (Diagnostic Systems Laboratories, Webster, TX, USA), following manufacturer’s recommendations.
Males vs. females. Student’s t-test (two tail). Mann–Whitney’s U-test. Fisher’s exact test (two tail). Among smokers only.
2.5. Statistical analysis
cleus test (MN) test; and one in a special BD Vacutainer CPT for leukocyte isolation for the comet assay.
TM
TM
2.2. Heavy metals determination Blood concentrations of aluminium, cadmium, nickel, lead and zinc were analyzed by graphite furnace atomic absorption spectrophotometry (GFAAS) after microwave digestion (Lafuente et al., 2004a). 2.3. Genotoxicity biomarkers Sister chromatid exchanges (SCE): Aliquots of 0.5 ml of heparinised blood were cultured in duplicate and treated as described elsewhere (Laffon et al., 2002a). A total of 50 well-spread metaphases were examined by the same scorer for each donor, half from each duplicate culture, in a Nikon HFX-DX light microscope, to determine the number of SCE cell 1. Cytokinesis-block micronucleus test (MN): MN test was performed in duplicate as described by Laffon et al. (2006). A total number of 1000 binucleated cells, 500 from each duplicate culture, were examined for each individual in a Leica DM-RXA fluorescence microscope, equipped with a 100 W mercury lamp, to determine the number of binucleated cells with micronuclei (BNMN). Criteria proposed by Fenech et al. (2003) were followed for the identification of binucleated cytokinesis-blocked cells and MN. Comet assay: Mononuclear leukocyte isolation was made using BD Vacutainer CPT Cell Preparation Tubes (Becton Dickinson) following manufacturer’s instructions. TM
TM
Univariate statistics were computed as customary (Kleinbaum et al., 1998; Armitage et al., 2002). Fisher’s exact test for homogeneity or independence was applied to assess the relationship between categorical variables. Student’s t-test and Mann Whitney’s U-test were used to assess statistical differences between means and medians, respectively. To analyse variations in biological markers, general linear models were applied. In order to strictly fulfill the assumptions of general linear models, the variables BNMN, %TDNA and plasmatic prolactin were normalized by natural logarithmic transformation (Kleinbaum et al., 1998). The main effects of all predictors were independently explored in the base models. In accordance with the objective of the study, all individuals were jointly analyzed, independently of their exposure group. The level of statistical significance was set at 0.05, and all tests are two-tailed. Analyses were performed using SPSS, version 12.0 (SPSS Inc., Chicago, IL, 2003). 3. Results Blood concentrations of heavy metals and biological markers in the study individuals are summarized in Table 2. Median concentrations of lead and zinc were over 100 times higher than those of nickel, and over 50 times higher than those of cadmium. Concentrations of aluminium, cadmium and nickel were similar in males and females, whereas significantly higher levels of lead and zinc were observed in males, even after adjusting by age and smoking. In addition, significant, mostly positive correlations were obtained for each pair of metals, except for nickel
450
B. Pe´rez-Cadahı´a et al. / Chemosphere 71 (2008) 447–455
Table 2 Concentrations of heavy metals and biological markers Total
Gender Males
Females
N (%)
N (%)
p-Valuea
Aluminium (lg l 1) Mean ± SD Median
143 (79.9) 10.19 ± 7.14 7.11
70 (76.1) 9.94 ± 6.88 8.06
73 (83.9) 10.42 ± 7.42 6.84
0.686b 0.568c
Cadmium (lg l 1) Mean ± SD Median
158 (88.3) 0.85 ± 0.43 0.81
82 (89.1) 0.89 ± 0.43 1.08
76 (87.4) 0.82 ± 0.43 0.54
0.341d 0.324c
Nickel (lg l 1) Mean ± SD Median
153 (85.5) 1.69 ± 2.48 0.27
78 (84.8) 1.43 ± 2.28 0.27
75 (86.2) 1.95 ± 2.67 0.36
0.191d 0.327c
Lead (lg l 1) Mean ± SD Median
163 (91.1) 59.13 ± 26.43 58.50
84 (91.3) 65.95 ± 30.40 63.00
79 (90.8) 51.89 ± 19.09 49.50
0.001d 0.001c
Zinc (lg l 1) Mean ± SD Median
163 (91.1) 41.89 ± 15.51 41.04
84 (91.3) 46.02 ± 18.51 43.88
79 (90.8) 37.50 ± 9.85 38.07
