Although the exposure to Mn from drinking water by Bangladeshi population has been recently highlighted , studies on Mn intake from foods are limited. The first study to address Mn intake from foods in the Bangladeshi diet was reported by Zablotska et al. , although this study did not incorporate contributions from betel quids and water. The observation by Hafeman et al.  that betel quid chewers, who are likely to be consuming As and Mn contaminated drinking water, had higher incidence of skin lesions and tremor led us to determine the urinary levels of Mn in chewers compared to non-chewers and also to carry out a more comprehensive estimation of the dietary intake of Mn that took into consideration contributions from betel quid chewing, tea drinking etc.
Since the Bangladeshi community in the United Kingdom have a similar diet to people living in Bangladesh , we determined the level of Mn in urine from 37 volunteers in this community. Betel quid chewers have a statistically significant (P = 0.009) higher Mn levels compared to non-chewers, which is likely to be due to the high Mn content in betel quids. Urinary Mn levels in the Bangladeshi population residing in Bangladesh have been determined by Ljung et al. . They analysed urine from 388 Bangladeshi women and reported a mean concentration of 2.5 μg L-1 with a median value of 1.6 μg L-1. Interestingly, this is similar to Mn level in urine for the male betel quid chewers in our study (2.25 μg L-1), but is approximately 1.5-fold higher than that for the female chewers in our study (1.62 μg L-1). The average urinary Mn level in the UK Bangladeshi community (chewers and non-chewers combined, 1.04 μg L-1) is 3.4-fold higher compared to the general UK population (0.3 μg L-1)  and is also far higher than those reported for other countries, including USA, Germany and Japan [23–25]. Although the higher urinary Mn for population residing in Bangladesh may partly arise from Mn in drinking water, this is not the case for UK Bangladeshis where the Mn in drinking water is relatively low at < 50 μg L-1 . Thus, the similarity between the urinary Mn levels in UK Bangladeshi chewers and Bangladeshi females (residing in Bangladesh) , can be mainly attributed to Mn intake from betel quids although contribution from other food sources cannot be ruled out.
The lower urinary Mn levels in UK Bangladeshi females could be due to lower betel quid consumption in the UK compared to Bangladesh and other dietary differences. Thus, for example, our questionnaire data revealed 3.5 betel quids are consumed on average per day by Bangladeshis in the UK, compared to 5.7 - 6.3 in Bangladesh . Bangladeshis residing in Bangladesh have a greater proportion of vegetables in their diet compared to UK Bangladeshis who consume a higher quantity of animal products such as poultry, fish and meat. In addition, the rice intake for UK Bangladeshis is almost 50% lower compared to that of Bangladeshi's residing in Bangladesh. Thus, the higher Mn levels for the latter group can be attributed mainly to a higher intake of plant based products which are known to have greater Mn content compared to animal products. Of course Mn intake from consumption of Bangladeshi groundwater is another factor explaining the higher Mn levels for the Bangladeshi females. Ultimately, Mn levels in chewers and non-chewers residing in Bangladesh needs to be carried out to ascertain our conclusions based on studies on Bangladeshis residing in the UK.
Mn intake by pregnant Bangladeshi females is of particular concern as Mn can be easily transferred from the mother to the foetus via the placenta, and subsequently crossing the blood-brain barrier of the underdeveloped brain affecting neurodevelopment . In this context, it has been reported that children exposed to Mn through drinking contaminated water in Bangladesh displayed poor intellectual function .
We have shown that a positive correlation exists between the total daily intake of Mn and urinary excretion of Mn for different populations (Figure 3). This is an interesting finding and such a correlation may be used for prediction of either urinary Mn excretion or total daily intake of Mn for other population, provided data on one of these two parameters are available. The positive correlation also suggests that bioaccessibility and bioavailabilty of Mn from different dietary sources, including drinking water, are likely to be similar. Our study shows that the average intake of Mn for Bangladeshi population is 18.3 and 18.7 mg day-1 from foods when betel quids (6 quids) or water (2.7 litres), respectively, are included in the daily intake estimation. When both betel quid chewing and drinking water are included in the calculation, the total daily intake of Mn is 20.3 mg day-1 for all foods. These results are not too different from the daily Mn intake of 16.51 mg day-1 calculated by Zablotska et al. , despite the fact they did not include water and betel quid intake in their calculation.
From an international perspective, the total Mn intake by the Bangladeshi population is by far the highest compared to non-occupationally exposed groups in any other country for which data has been published so far (Table 4) [32–37].
The THQs of Mn estimated for the Bangladeshi population were 0.18, 1.55 and 0.21 for betel quids, food and water, respectively. Although the "food" category shows the highest THQ, the nature of the Mn species present in foods and its bioaccessibility and bioavailability may differ to that derived from betel quids and water. Surprisingly, the THQ of Mn resulting from water and betel quids are very similar, but again the bioaccessibility and bioavaliability could be very different. The issue of Mn species, Mn bioavailability and bioaccssibility needs to be investigated in the future. Nevertheless, the total THQ of Mn from Bangladeshi diet (including food, water and betel quids) should be considered as very high (1.94) as THQ values greater than one are considered to be of concern . As already pointed out, the daily intake of Mn by Bangladeshis is the highest compared to other communities thus far reported and the health impact of this requires further investigation.
Several studies have attributed betel quid consumption with the development of different diseases including oral cancer, diabetes, cardiovascular disease etc [38, 39]. In the UK, Asians have the highest incidence of head and neck cancer which has been attributed to smoking and betel quid chewing . As pointed out earlier, chewing betel quids was associated with a higher risk of skin lesions  and tremor  in populations exposed to high levels of As in their drinking water. Unfortunately, these studies linking betel quid chewing with certain health impacts in As exposed populations did not analyse betel quids. However, a recent study analysed the As content of tobacco used in betel quids . This study only focused on As content of tobacco rather than the other substances consumed as part of the chewing material such as Piper betel leaves, areca nut, lime etc.
The mechanism underlying Mn induced toxicity in people exposed to high levels of Mn is poorly understood. However, it has been suggested that Mn plays a role in the generation of ROS that may result in neurotoxicity . More recent studies have provided evidence suggesting that oxidative stress induced by Mn exposure can trigger apoptosis of neural stem cells . In light of these studies, we hypothesise that increased Mn exposure through a combination of diet, betel quid chewing and drinking water results in oxidative stress and cellular damage that may result in Mn induced neurotoxicity in certain sectors of the Bangladeshi population. Simultaneous exposure to high levels of both Mn and As may result in increased toxicity which may explain the observation of greater tremor and skin lesions in betel quid chewers. Although, several studies have reported modulating effect of areca nut in As induced skin lesions [9–11], only Pilsner et al.  and Lindberg et al.  attempted to provide an explanation for this observation. Pilsner et al.  suggested that ROS generated by arecoline, a key compound in areca nut, may be responsible for the higher incidence of tremor and skin lesions in betel quid chewers. Lindberg et al.  suggested poorer As metabolism in female chewers may responsible for their greater arsenic induced skin effects but were unable to attribute this to a particular substance in the betel quids. We suggest that the inorganic (Mn and As) and organic (arecoline) components of betel quids, are jointly responsible for the adverse health outcomes in betel quid chewers who drink contaminated groundwater. This needs to be investigated in the future.
A limitation of the present study is that the number of volunteers for the urinary Mn analysis was rather low (37 in total) and that we were only able to conduct studies with UK Bangladeshi population, rather than with participants residing in Bangladesh who drink groundwater that is often contaminated with Mn. Another limitation is that we analysed betel quid components that were sold in the UK market and we were not able to survey the wide variety of these materials that are available in Bangladesh. In addition, Mn levels in water, rice and Piper betel leaves, vegetables etc can vary from region to region within Bangladesh. Furthermore, the absence of information on Mn bioavailability and bioaccessibility from foods and betel quids makes it difficult to obtain an accurate estimation of exposure levels and risk assessments. Finally, we have assumed that the entire betel quid is ingested which is very common amongst the Bangladeshi community. However, this is not the case for all individuals as some spit out the juice accumulated in the mouth, whilst others spit out the fibrous portion of the betel quid after extensive chewing. Thus, the level of Mn exposure from betel quid chewing for the latter groups may prove to be lower. We plan to address these issues in future studies.