This study was conducted on the rural island communities of Ararca, Barú, Caño del Oro, and Tierra Bomba, which are traditional fishing communities located in the district of Cartagena de Indias, Colombia ( Figure 1). Ararca, with a population of 900 inhabitants residing in 219 homes, is situated 15 km away from the urban area of Cartagena and maintains direct interaction with the Bay of Cartagena and Dique Canal; and Barú is a peninsula inhabited by 1922 people living in 448 homes, located 34 km from the urban area of Cartagena. ¹⁷ Caño del Oro and Tierra Bomba are communities that are part of the island of Tierra Bomba, with populations of 1257 (326 homes) and 1472 inhabitants (298 homes), respectively.

The population analyzed in this study consisted of individuals residing in the communities of Ararca, Barú, Caño del Oro, and Tierra Bomba (5551 inhabitants). To determine the sample size and proportion for each site, the formula for sample estimation was used within a defined population, ^{18 , 19} where N represents the total population size (5551 inhabitants); p denotes the expected proportion of individuals with Hg and/or Cd-caused DNA damage, estimated at 50% (0.5); and d2 represents the maximum allowable error of 0.1 (10%).

The total estimated sample size was 94. Subsequently, this number was distributed among the four communities based on the contribution of each community to the total population size. Consequently, the numbers of individuals evaluated in each community were determined to be: 16 from Ararca, 31 from Barú, 22 from Caño del Oro, and 25 from Tierra Bomba, resulting in a combined sample size of 94. Additionally, a control group of 30 individuals residing in the city of Cartagena was included for comparative analysis.

Participants were registered using a comprehensive survey that collected information on various lifestyle factors, including alcohol and tobacco consumption, dietary habits, water source and treatment, occupational activities, duration of residence, and history of exposure to genotoxic agents, such as chemicals or radiation.

Based on the survey data, individuals included in the study were required to have resided in the study area for more than three years, with no history of alcoholism, smoking, cancer, or chronic degenerative diseases within the six months prior to the survey. Additionally, they should not have undergone any pharmacological treatment and should not have a history of prolonged exposure to any other known chemical or physical genotoxic agent. Failure to meet any of these criteria was considered exclusion criterion.

The inclusion criteria for individuals in the control group were the same as those in the study group except that they should not belong to any of the communities under study. Therefore, individuals residing within the urban area of Cartagena were consulted to ensure a match in sex and age proportions with the study group.

Every participant, without exception, provided their consent by signing an informed consent form. The form outlined the objectives of the study and authorized the use of their biological samples for scientific purposes.

The blood samples and buccal mucosal cells were collected from all participants included in the study. To analyze the levels of Hg and Cd, 3–6 mL of the peripheral blood was drawn into EDTA-containing tubes and stored on ice. For the BMCyt assay, buccal mucosal cell samples were obtained by gently swabbing the inner cheeks with a cotton swab. These cells were then suspended in a cold saline solution (0.9%). Before sample collection, the participants were instructed to rinse their mouths with bottled drinking water provided by the researchers. Subsequently, the samples were transported to the UNIMOL Laboratory at the University of Cartagena for further analysis.

Total mercury (T-Hg) and Cd levels in the blood were determined at the Laboratory of Toxicology and Environmental Management at the University of Córdoba within the first 48 h. Two milliliters of blood were transferred to a 100 ml Teflon vessel, to which 8 ml of concentrated nitric acid and 2 ml of hydrogen peroxide (30%) were subsequently added. The vessels were sealed and digested in an Ethos Touch microwave oven at a temperature of 280 °C, with a pressure of 80 bars and a power of 1,400 W for 30 minutes. Once cooled, the sample volumes were adjusted with deionized water to a final volume of 70 ml for T-Hg analysis and 500 μl of the sample adjusted to a final volume of 1 ml for Cd analysis. T-Hg analyses in the blood were performed using Cold Vapor Atomic Absorption Spectrophotometry (CVAAS; Thermo Scientific ^TM, model iCE, series 3000) at a wavelength of 253.7 nm and detection limit of 0.40 μg/L. ²⁰ Blood Cd analyses were conducted using Graphite Furnace Atomic Absorption Spectrometry (GFAAS, Thermo Scientific iCE 3000 Series equipped with a graphite furnace) with a detection limit of 0.20 μg/L. ²¹ Notably, Cd analysis was not possible for four individuals because of sample scarcity. Concentration units for both metals were determined in micrograms per liter. The exposure limits for T-Hg and Cd in the blood were established following the guidelines of the World Health Organization (WHO) ²² and the Centers for Disease Control and Prevention (CDC), ²³ respectively; both of which specify an allowable limit of up to 5 μg/L in the blood for these metals.

After buccal massage, the cells were suspended in 15-mL centrifuge tubes containing 5 mL of 0.9% saline solution (w/v). Upon arrival at the laboratory, samples were centrifuged at 1500 RPM for 10 min. The settled buccal cells were washed twice with saline solution and once with Carnoy’s fixative (methanol and glacial acetic acid, 3:1) under the same centrifugation conditions. Subsequently, 80–100 μL of cell suspension were carefully applied to angled slides to ensure optimal cell spreading. ² The fixed cells were stained using the Feulgen method, following the recommendations of Thomas et al. ⁷ In summary, the slides were first treated in a 50% and 20% ethanol solution for 1 minute at room temperature, followed by immersion in distilled water for 2 minutes. Next, the slides were exposed to 5 M HCl for 30 minutes. A slide not treated with HCl served as a negative control during this step. After this treatment, the slides were rinsed with running water for 3 minutes and then immersed in Schiff’s reagent (Sigma-Aldrich, St. Louis, MO, USA) for 1 hour in complete darkness at room temperature. Following this, the slides were rinsed for 5 minutes with running water and then with distilled water. Finally, the slides were immersed in 0.2% (w/v) Light Green solution (Sigma-Aldrich, St. Louis, MO, USA) for 20 to 30 seconds, rinsed with distilled water, and allowed to air dry for 15 minutes at room temperature.

Stained slides were examined using an optical microscope (Leica Microsystems, Germany), with 1000 cells/individual visualized and analyzed at 100× magnification. Nuclear abnormalities such as DNA damage (micronuclei and nuclear buds), cell death indicators (condensed chromatin, pyknosis, karyorrhexis, and karyolysis), and cytokinesis defects (binucleated cells) were assessed following the recommendations of Thomas et al. ⁷ :

Micronucleated cells: micronucleated cells possess a main nucleus as well as one or more micronuclei. Micronuclei are generally round or oval and exhibit a staining intensity comparable to that of the main nucleus. They typically measure between 1/3 and 1/16 the diameter of the main nucleus and are situated within the cellular cytoplasm ( Figure 2a).

Nuclear buds: the main nucleus exhibits a sharp constriction that forms a bud, which remains attached to it. This bud has a staining intensity similar to that of the main nucleus and its diameter ranges from one-quarter to one-half of the nuclear diameter ( Figure 2b).

Binuclead cell: Cells contain two main nuclei of similar size and staining intensity. The nuclei are usually very close and may touch each other and usually have the same morphology as that observed in normal cells ( Figure 2c).

Condensed chromatin: the nucleus shows areas of aggregated chromatin, with distinct regions that are more intensely stained. Additionally, the nucleus exhibits a striated pattern ( Figure 2d).

Karyorrhexis: The nucleus exhibits extensive aggregation of chromatin, which can be observed in substantial clusters. Additionally, nuclear fragmentation may be evident ( Figure 2e).

Karyolysis: the nucleus is depleted of DNA, as indicated by its lack of staining with the Feulgen method (the nucleous apparent as a ghost-like image) ( Figure 2f).

Pyknosis: the cell has a small, shrunken nucleus that is uniformly and intensely stained. The diameter of this nucleus is approximately 1/3 to 2/3 of the diameter of a normal nucleus ( Figure 2g).

The normality of the variables was assessed using the Kolmogorov–Smirnov test or Shapiro–Wilk test to determine whether parametric or non-parametric analysis was appropriate. The Shapiro-Wilk test is more appropriate for small sample sizes (N ≤ 50), while the Kolmogorov-Smirnov test is used for larger sample sizes (N > 50). Both the Shapiro-Wilk and the Kolmogorov-Smirnov tests compare the scores in the study sample with a normally distributed set of scores with the same mean and standard deviation; their null hypothesis is that sample distribution is normal. Therefore, if the test is significant (p < 0.05), the sample data distribution is non-normal. ²⁵

In our study, the variables did not follow a normal distribution (p < 0.05). Therefore, the results are reported as medians and interquartile ranges (IQR: 25 ^th–75 ^th percentile). Significant differences between groups were examined using the Mann–Whitney U test or Kruskal–Wallis test for multiple comparisons, followed by post-hoc Dunn’s test for each pairing. The Mann-Whitney U test and the Kruskal-Wallis test are non-parametric tests that assume a non-normal distribution of data. The Mann-Whitney U test can only deal with the comparison between two samples, whereas the Kruskal-Wallis test can handle multiple samples. Furthermore, post-hoc tests such as Dunn’s test enable us to identify which specific groups exhibit statistical differences. ²⁶

Finally, correlations among blood metal levels, quantitative demographic variables, and BMCyt parameters were assessed using Spearman’s coefficient (a non-parametric test for establishing correlations). All analyses were performed using the statistical software SPSS v.19.0 (IBM Corp, Armonk, NY, USA). Statistical significance was defined as p < 0.05.

Blood concentrations of T-Hg and Cd are presented in Table 2. The median T-Hg level was significantly higher in the study group than in the control group (p < 0.001). Furthermore, the averages exceeded the permissible limits set by the WHO for populations environmentally exposed to Hg, both within communities and overall. When comparing different communities, Barú exhibited significantly higher levels as compared to Ararca and Caño del Oro.

Analysis of blood Cd levels revealed no significant differences between the study and control groups. Notably, 77.7% (70/90) of the examined individuals exhibited values below the detection limit for Cd in the blood, according to the employed methodology (0.2 μg/L). Among the 20 individuals in whom Cd was detectable, the levels ranged from 0.2 to 2.99 μg/L (median = 0.43 μg/L). This suggests that none of the participants exceeded the Cd limit set by the CDC (5 μg/L).

Comparisons of demographic variables potentially influencing the blood levels of T-Hg and Cd within the study group, such as sex, age, duration of residence in the area, fish consumption, and source of drinking water, are presented in the Supplementary Material ( Tables S1 and S2). The analysis revealed that men and individuals consuming fish daily exhibited significantly higher blood T-Hg levels (p < 0.05) ( Table S1). Spearman’s correlation analyses indicated significant associations between blood T-Hg levels and age (Rho = 0.260; p = 0.012), duration of residence in the area (Rho = 0.374; p < 0.001), and fish consumption (Rho = 0.415; p < 0.001). No differences between the groups or correlations between different variables and blood Cd levels were observed (p > 0.05) ( Table S2).

Table 3 presents a comparison of the BMCyt biomarkers between the control and study groups. Significantly higher frequencies of all BMCyt assay biomarkers (except pyknosis) were observed in the study group than in the control group (p < 0.05). Site-specific comparisons revealed that the frequency of nuclear alterations indicative of cell death (condensed chromatin, karyorrhexis, and karyolysis) was generally lower in the Ararca community than in the other communities. Figure 2 outlines each of the reported biomarkers.

Table 4 shows the Spearman’s correlations between blood metal levels and BMCyt biomarkers. Significant positive correlations were observed between T-Hg levels and the frequencies of micronuclei, karyorrhexis, and karyolysis. Considering that sex and age might significantly influence the parameters measured by the BMN-cyt assay, we evaluated these factors as potential predictors of the frequency of these parameters. Our analysis found no differences between sexes and no significant correlation between age and the parameters tested by the BMCyt assay ( Table S3 and S4).

Limitations of this study include several factors that may have influenced the interpretation of results. Firstly, the sample size, although calculated based on statistical power, may not fully represent the diversity within the island communities of Cartagena. Secondly, while efforts were made to control for confounding variables such as age, sex, and some habits, other environmental and genetic factors could have influenced the observed genotoxic effects. Additionally, the BMCyt assay, while effective for assessing buccal cell damage, provides a snapshot of genetic damage that may not capture long-term effects or variations in different cell types. Furthermore, the study’s cross-sectional design limits the ability to establish causal relationships between heavy metal exposure and genotoxic outcomes. Longitudinal studies would provide more robust evidence of the cumulative effects of chronic exposure to mercury on genetic integrity in these vulnerable populations.