Due to various geological mechanisms or anthropogenic activities, an altered distribution of certain chemical elements occurs on the surface in some parts of the earth. The over- or under-abundance of certain chemical elements will result in an imbalance of element exchange between human bodies and the environment. If the exchange exceeds the normal range that the organism is able to buffer or to tolerate, certain kinds of health disorders will emerge. The major biogeochemical abnormality-related endemics known so far in China include endemic arseniasis (which covers two kinds of As exposure routes: one via As polluted drinking water sources and another one via indoor combustion of high As-content coal), fluorosis (via polluted water or via indoor air pollution from high fluorine-content coal burning and from brick tea), iodine deficiency disorders (IDD), and other diseases.

Several villages in Southwest Guizhou Bouyei (Buyi) and Hmong (Miao) Ethnic Minority Autonomous Prefecture, Southwest China represent a unique case of endemic arseniasis, which is related to indoor combustion of high arsenic-content coal, not with As polluted drinking water sources (Jin et al., 2003[7]; Zheng et al., 2005[20]; Liu et al., 2002[13]).

Since the early 1960s, as local woods and bushes have been depleted, villagers in the above mentioned area burned local high As containing coal in poorly or unventilated stoves (without chimneys) for cooking, heating and drying crop and food. The highest As concentration in local coal was once detected as 3.2-3.5 % (Ding et al., 2001[2]). Since the early 1970s thousands of arseniasis cases emerged. The area where the target village of the present investigation is located was the first one reported (Zhou et al., 1993[21]). Most of the cases diagnosed by the end of the millennium (1,386 out of 2,241 cases) are clustered in the township where the target village of the present investigation is located (Jin et al., 2003[7]).

Most of the work about this endemic population reported so far was focused on environmental causes and confirmed the causality between indoor burning of high arsenic-content coal and the excess prevalence of arseniasis cases in rural population (Jin et al., 2003[7]; Zheng et al., 2005[20]; Liu et al., 2002[13]; Zhou et al., 1993[21]). Variation in individual susceptibility to the chronic poisoning of inorganic arsenic exposure has been suggested (Vahter, 2000[16]). In our early field survey conducted in 2002, a remarkable ethnicity-dependent (or clan-dependent) difference of arseniasis prevalence has been observed in the residents with different ethnic origin and various clan relationships in the same village. This was surprising as the families of different ethnicity and of different clans have been living together in the same village for generations (Lin et al., 2003[8]) and were exposed likewise to indoor combustion of local high As coal for quite similar time duration (Lin et al., 2006[10]). An array of host factors has been proved to be related to the different impact of susceptibility to arseniasis risk under this unique As exposure scenario (Lin et al., 2006[10], 2007[11], 2010[12][9]).

New techniques for the in situ detection of micro-distribution and speciation of the element arsenic in specimens of exposed persons and in polluted environment are required for further expanding our knowledge of various risk factors. This may pave the way for a quantitative understanding of all the factors which might influence the excess risk of arseniasis, irrespective of whether or not exposure-related, host-related, or related to a combination of both. The utilizing of synchrotron radiation techniques, such as micro-beam X-ray fluorescence (μ-XRF) and X-ray absorption fine structure (XAFS), might be a promising option (Gault et al., 2008[5]). Both, μ-XRF and XAFS are non-destructive physical procedures and have been widely used to detect, quantify and map the element content and the speciation of samples in their natural state. In this work, distribution and speciation of the element arsenic within a single human hair or rice grain was determined using the well-established μ-XRF mapping procedure and XAFS technique in Shanghai Synchrotron Radiation Facility (SSRF, Shanghai, China).

This is the first report on a study applying synchrotron radiation techniques, previously applied by Finkelman and colleagues on high arsenic coals in this area (Finkelman et al., 1999[4]), on food and hair samples on the rare case of endemic arseniasis related to indoor exposure to As. μ-XRF and XAFS are non-destructive physical approaches which need no chemical preparation procedures prior to the assay. The non-destructive, in situ assays provided more new, stimulating data which cannot be gained by any other conventional, neither chemical nor biochemical technique. The present approach will also presumably promote the investigation of other population health problems triggered by biogeochemical abnormality, such as endemic thallotoxicosis (Xiao et al., 2004[19]) or mercury poisoning (Feng and Qiu, 2008[3]) observed in the Southwest Guizhou area.

Usually, arsenic levels in hair, especially the hair in females, can serve as a biomarker for short-term (<1 year) internal dose. In forensic medicine practice, variation of As levels in different sections of the same hair has been successfully used to recapitulate the chronic poisoning process and the approximate dosage applied. An example is the scientific deduction of possible historical events that happened in St Helena island during 1815-1821 when Napoléon Bonaparte was assumed to be gradually murdered by deliberate arsenic poisoning (Weider and Fournier, 1999[17]). In our case the As levels along the hair samples were found largely homogenous. It might be suggested that the exposure environment in this endemic village was, basically, stable around the time of sampling, so that any part of the hair samples would be suitable for the assay.

In our preliminary work in cooperation with SSRF, the X-ray fluorescence spectra of hair samples from the same endemic village suggested that there were no noticeable As peak area differences between two different groups of the residents (patients and their asymptomatic neighbors, preliminary data not shown). Furthermore, ICP-MS assays have been performed with all human hair samples collected, resulting in almost identical results for both groups of residents (F=0.001, P=0.981). This is totally different from observations reported so far from the arseniasis endemic areas related to As contaminated drinking water sources in this country and along the world as well. For final excluding of any possible interfering factors, we carefully checked the decades-long medical surveillance records archived in the Prefecture Centre for Disease Control (CDC) and made quite clear that the last medical dearsenization intervention with dimercaprol in this endemic population was conducted in 1990-1991 and covered only about 500 local people, only a small portion of the local population. Based on all the information available, it is reasonable to assume that the dearsenization intervention in part of the local residents in the early 1990s does not have a long-lasting impact on the As body burden in this exposed population.

The “hypernormal” phenomena observed in this investigation is well consistent with our previous observation. In our early work conducted in 2000 and in 2004 in the same population (Lin et al., 2003[8], 2006[10], 2010[12]), we found that As internal dosage expressed as total As level in hair and in urine in the subgroup of diagnosed arseniasis patients was not higher than in the subgroup of asymptomatic neighbors. The internal As dosage of ethnic Hmong subjects was found significantly higher than of their ethnic Han neighbors (p < 0.001 for As in hair and p < 0.01 for urine samples), though the arseniasis prevalence in Hmong farmers was significantly less profound than among the Han farmers (5.9 % vs. 32.7 %, OR: 0.12, 95 % CI: 0.06-0.27, P=3 x 10^-10) (Lin et al., 2006[10]).

Quite similar As levels in the hair samples collected from a thallotoxicosis-endemic village which is only about 10 km away from our target arseniasis-endemic village were confirmed by our current work. The exposure route for thallium is via locally produced food (Xiao et al., 2004[18]). Furthermore, these authors stated that local food is also the source for elevated urinary levels of arsenic and mercury. Thus, the impact of mercury should not be ignored. However, to our best knowledge no data on the local coal is available - in contrast to coals of the most eastern part of that province which is about 400 to 500 km from the investigated target village. However, chemical analysis of a coal sample used in another village (approximately 40 km from the investigated village) showed a mercury concentration of 55 ppm (Finkelman et al., 1999[4]). Interestingly, no typical arseniasis symptoms have ever been diagnosed in addition to the thallotoxicosis among the residents so far. All of these observations might suggest that the exposure to high level As in the environment or high As internal dosage is not inevitably related to an increased prevalence of arsenic-related skin symptoms.

The X-ray absorption fine structure (XAFS) analysis revealed the coexistence of both the As⁺³ and As⁺⁵ states in every hair sample tested. However, the ratio of As⁺³ / As⁺⁵ varied greatly from person to person. Generally, the diagnosed patients displayed a clear tendency of higher As⁺³ / As⁺⁵ ratios than the asymptomatic fellow villagers. This finding is in line with the basic concept of toxicological science that As⁺³ is more toxic than As⁺⁵.

A variance in the individual susceptibility to chronic arsenic poisoning has been suggested by Vahter (2000[16]). A remarkable clan aggregation was found in our preliminary survey in this village in April 2002 (Lin et al., 2003[8]). In a field investigation conducted in 2004, the ethnicity-dependent (or clan-dependent) variation in arseniasis prevalence was confirmed in this village, though all the families had been living together in the same village for generations and were exposed to similar indoor combustion of local high As coal for a quite similar time duration (Lin et al., 2006[10]). A few of the host factors, such as the polymorphisms at several loci of the genome were proved to be associated with the modulation of arseniasis risk in this exceptionally exposed rural population (Lin et al., 2006[10], 2007[11], 2010[12][9]). Recent data might add more power to support the pivot role of host factors.

For decades, the villagers in the endemic area have been taught persistently by local medical personnel and by local government officials that the grain or the vegetable must be water washed or rinsed thoroughly to remove the As before it can be cooked or ingested. The μ-XRF mapping displayed, for the first time, that the arsenic pollution of rice grains is not only restricted to the surface but has also penetrated deeply inside the grain and cumulated in the embryo of the grain, which is the major part of daily nutrition at a level which cannot be ignored any longer. Only a limited number of grain samples have been mapped by μ-XRF technique since the machine time of SSRF for the present study was limited. Therefore, our observation can only be considered as “preliminary”. More work is required. Our new data of micro-distribution of As deeply inside the grain may explain why the major item of the teaching program to the local residents, i.e. rinsing the grain with water several times, is largely ineffective. The results of this study should promote studies to investigate, whether rinsing with water remains also largely ineffective or not when applied to chili or corn, food which is very important for the arsenic load of the villagers in this area.

It must be mentioned that exposure conditions in this village have been improved in the last decades or so, since a series of countermeasures has been put forward in the endemic area by local authorities and by some other agencies as we have seen in our field investigation in April 2004 (Lin et al., 2006[10], 2010[12]). Our data of 2004 also witnessed a significant decrease of internal dose of total arsenic in the villagers. The hair As content in each subject obtained in the present study does no longer represent the real internal As dose of the subjects at the time of skin lesion onset or at the time of diagnosis. The 1991 survey, the first and the only overall field investigation ever held in the whole endemic area, including the target village of the present investigation was targeted only towards different exposure environments, not focused on specific individuals or subject subgroups. The retrospective assessment of As internal load in each individual remains impossible.

Synchrotron radiation techniques, such as micro-beam X-ray fluorescence (μ-XRF) and X-ray absorption fine structure (XAFS), are suitable techniques to determine arsenic species concentrations in different parts of hair and rice grain samples. When rice is stored over the open fire of high As-content coal for drying, arsenic penetrates the endosperm, the major edible part of the grain. Thus rinsing the grain with water several times will remain largely ineffective.

The authors declare no conflict of interest.

The data of the present study were obtained at the BL15U and BL14W1 beamline stations of the Shanghai Synchrotron Radiation Facility (SSRF). The authors would like to thank the staff of both beamline stations for their assistance and valuable suggestions.