NeuroToxicology 45 (2014) 159–167

Platelet hyperactivity, neurobehavioral symptoms and depressionamong Indian women chronically exposed to low level of arsenic

Bidisha Mukherjee a, Banani Bindhani a, Hirak Saha a, Dona Sinha b,*, Manas Ranjan Ray a

a Department of Experimental Hematology Unit, Chittaranjan National Cancer Institute, Kolkata-700 026, Indiab Department of Receptor Biology and Tumor Metastasis, Chittaranjan National Cancer Institute, Kolkata-700 026, India

A R T I C L E I N F O

Article history:

Received 19 April 2014

Accepted 21 October 2014

Available online 28 October 2014

Keywords:

Arsenic

Water

Neurobehavioral symptoms

Depression

Indian women

A B S T R A C T

The prevalence of neurobehavioral symptoms (NBS) and depression has been investigated in

premenopausal rural women of West Bengal, India enrolled from arsenic (As) endemic (groundwater

As 11–50 mg/L; n = 342) and control areas (As level � 10 mg/L; n = 312). The subjective symptoms

questionnaire and Beck’s 21-point depression inventory-II were used for the detection of NBS and

depression, respectively. Platelet P-selectin expression was measured by flow cytometry, plasma

neurotransmitter activity with high performance liquid chromatography and groundwater As level by

atomic absorption spectroscopy. The As level in groundwater was 2.72 � 1.18 mg/L in control and

28.3 � 13.51 mg/L in endemic areas (p < 0.0001). Women residing in endemic areas demonstrated a higher

prevalence of depressive symptoms (39.8 vs. 19.9%, p < 0.001) and anxiety (43.3 vs. 18.0% in control,

p < 0.001), fatigue (68.4 vs. 23.4%, p < 0.0001), reduced sense of taste (15.8 vs. 4.5%, p < 0.0001) and smell

(14.9 vs. 5.8%, p < 0.001); burning sensation (36.8 vs. 5.4%, p < 0.0001) and tingling or numbness in the

extremities (25.1 vs. 5.1%, p < 0.0001); and transient loss of memory (69.9 vs. 28.2%, p < 0.001). As-exposed

women had 1.6-times more plasma epinephrine and norepinephrine (p < 0.05), 1.8-times higher level

plasma serotonin with 28.9% lower intraplatelet serotonin (p < 0.05 for both), but their plasma dopamine

level was not significantly different (p > 0.05) from that of controls. Moreover, women from endemic areas

had 2.3-times more P-selectin-expressing platelets in their circulation (p < 0.001). After controlling the

potential confounders, chronic low level As (11–50 mg/L) exposure showed a positive association with the

prevalence of neurobehavioral symptoms and depression among Indian women in their child-bearing age.

� 2014 Elsevier Inc. All rights reserved.

Contents lists available at ScienceDirect

NeuroToxicology

1. Introduction

Several million people living in West Bengal, a state in easternIndia, and adjoining Bangladesh drink tube well water contami-nated with variable concentrations of inorganic arsenic (As). As isan environmental toxin, and chronic exposure to As can lead tocarcinogenic as well as non-carcinogenic health problems (Smithand Steinmaus, 2009). Epidemiological studies have shown thatpeople who consume As-contaminated water for long periods oftime develop specific skin lesions (Guha Mazumder et al., 1998). Inaddition, As-exposed persons suffer more from reduced lungfunction, chronic obstructive pulmonary disease (Smith et al.,2006; Chattopadhyay and Ghosh, 2010; Amster et al., 2011;

* Corresponding author at: Department of Receptor Biology and Tumor

Metastasis, Chittaranjan National Cancer Institute, 37, S.P. Mukherjee Road,

Kolkata 700 026, India. Tel.: +91 33 2476 5101; fax: +91 33 2475 7606.

E-mail address: donasinha2012@gmail.com (D. Sinha).

http://dx.doi.org/10.1016/j.neuro.2014.10.011

0161-813X/� 2014 Elsevier Inc. All rights reserved.

Nafees et al., 2011), impairment of liver function (GuhaMazumder,2005) anemia (Guha Mazumder and Dasgupta, 2011), hyperten-sion and cardiovascular diseases (Guha Mazumder and Dasgupta,2011). Moreover, As-exposed subjects have increased risk ofdeveloping cancers of the skin, lung, colon and the urinary bladder(Agusa et al., 2010; Aballay et al., 2012). Considering the toxiceffects of As on human health, the World Health Organization(WHO) has advocated that the As level in groundwater should notexceed 10 mg/L. In India, however, the legally enforceable Standardfor As in drinking water has remained 50 mg/L (Bureau of IndianStandards (BIS), 2009).

There has been a growing concern that the central nervoussystem (CNS) can be yet another target of groundwater As(Rodrıguez et al., 2003). Animal studies have shown that As cancross the blood brain barrier and accumulate in different regions ofthe brain including the striatum (Itoh et al., 1990). As aconsequence, the synthesis and release of the central neurotrans-mitters can be altered affecting the locomotor activity (Itoh et al.,1990; Rodrıguez et al., 2003). As affects the development of the

B. Mukherjee et al. / NeuroToxicology 45 (2014) 159–167160

CNS in neonatal rats and elicits neuronal death in adult rat brain(Chattopadhyay et al., 2002). As exposure has been shown to beassociated with damage of the neurobehavioral function of theMexican children (Calderon et al., 2001; Rosado et al., 2007), anddecrease in intelligence quotient and sensory and motor functionsin adults (Bardullas et al., 2009), adolescents (Tsai et al., 2003) andschool-aged children of Bangladesh (Wasserman et al., 2011) andMexico (Calderon et al., 2001). As affects the cognitive develop-ment of the children (Rosado et al., 2007), and chronic malnutritionpotentiates this activity (Calderon et al., 2001). Despite thesereports, the potential impact of As, especially the chronic, low doseexposure, on mental health of the people who are drinking As-contaminated water from their childhood is largely unknown.Mental illness including depression represents three of the tenleading causes of disease burden in low and middle-incomecountries (Anwar et al., 2011). Depression in women increases therisk of delivering low birth weight baby (Nasreen et al., 2010), andimpairs health-related quality of life with increase in social cost(Glied et al., 2010). Against this background, we conducted a studyon the prevalence of neurobehavioral symptoms and depressionamong pre-menopausal rural housewives in India who weredrinking groundwater contaminated with As level higher thanthe WHO guideline of 10 mg/L but within the Indian Standard of50 mg/L.

Although the biological mechanism(s) of neurobehavioralsymptoms and depression is incompletely understood, severalconditions like malnutrition (German et al., 2011; Kvamme et al.,2011), inflammation (Song et al., 2009), disruption of the circadianrhythm such as alterations in sleep/wake cycles and themelatonergic system (Srinivasan et al., 2009; McClung, 2011),and alteration of neurotransmission in the CNS (Emanuele et al.,2010) may contribute to these conditions. Because it is difficult toaccess the CNS, and considering the importance of platelets asperipheral markers of central serotonergic activity (Camacho andDimsdale, 2000), we have used platelet serotonin as a surrogate forcentral serotonergic neurons. Moreover, since enhanced plateletactivation with regard to P-selectin expression is a commonfinding among depressed subjects (Mendoza-Sotelo et al., 2010;Banerjee et al., 2012a,b), we compared the platelet P-selectinexpression between the control (who were drinking water withnot more than 10 mg/L of As) and age-matched exposed women(drinking water containing 11–50 mg/L of As) recruited from thesame geographical area.

2. Materials and methods

2.1. Study areas and participants

The study was conducted in Kalyani, Saguna and Naryanpurvillages of Haringhata block in Nadia district, and Dhaltita ofBasirhat block in North 24-Parganas district of West Bengal, a statein eastern India. Among these villages, Narayanpur and Dhaltitawere in As endemic zone (As level in groundwater, 11–50 mg/L)while the As concentrations in groundwater of remaining threevillages were within the WHO guideline of 10 mg/L. Womenparticipants were enrolled from all these five villages. Initially weapproached 701 women and 678 (96.7%) gave consent toparticipate, and 654 were finally enrolled (96.5% of those whogave consent) after considering the inclusion/exclusion criteria.Inclusion criteria were pre-menopausal married women who werenever smokers and non chewer of tobacco or betel quid, not toothin or obese (body mass index >15 and <40 kg/m2), not pregnantor breast feeding, having no past history of malignancy and womenwho were not under any medication at the time of enrolment.Exclusion criteria were women with As-related visible symptomsof arsenicosis such as melanosis on the skin and hyperkeratosis

and hard patches on the palms and soles, those using oralcontraceptive, underwent recent surgical operation, had previousor recent histories of drug addiction and had exposure toagricultural/home insecticides within last 1 month, as these couldinfluence neurotransmitter activity. The Ethics Committee ofChittaranjan National Cancer Institute approved the study proto-col. Written informed consent form was obtained from all theparticipants.

In all four villages automobiles were rare as cycle rickshaw wasthe principal mode of transport, and industrial emissions werenegligible due to the absence of air polluting industries within 5-km radius. Agricultural practices were the livelihood of themajority of the people. Drinking water sources in these villageswere tube well water. The person responsible for obtaining waterfor each family was established in a meeting with the head of thehousehold, and that person was then asked to walk to the tube wellthey used. Also they were asked if they ever used any other well.Through this procedure the identification of the tube wells used byeach family was established.

2.2. Collection of water samples and measurement of As

The hand pumps of the tube wells were pressed so as to flow thewater for 5 min and then the water samples were collected insterile polyethylene bottles with 0.1% hydrochloric acid (HCl). TheAs analysis in collected water samples was done by atomicabsorption spectroscopy (AAS) using vapour generation assemblyaccording to the procedure of Behari and Prakash, 2006. In brief, astock solution of 100 ppm (0.1 mg/L) As was prepared from Asstandard for AAS (Sigma–Aldrich) in milli Q water with 0.1% HCl.Subsequently a range of standards (1–50 mg/L) were prepared withdilution of the stock solution for calibration before use. The AtomicAbsorption Spectrophotometer Duo (Agilent Technologies)equipped with vapour generation assembly (Agilent VGA 77)was used for the estimation of As. The instrument was calibratedusing 5 M HCl in the acid channel and 0.6% NaBH4 and 0.5% NaOHin the reduction channel in the working range of 10–50 ppb Asstandard. For pre reduction experiment, the sample was preparedin 5 M HCl and 20% potassium iodide was added and allowed toreact up to 45 min at room temperature. For direct analysis, theunknown samples were diluted in varying concentrations and prereduced in the same way as the standards. A standard curve wasdrawn with the prepared range of standards and the concentra-tions of arsenic in the unknown samples were ascertained from thestandard curve with the help of the software Spectra AA. Thesamples were analyzed in duplicate and three readings were takenfor each sample. For quality control, inter-laboratory tests of thesame samples were done along with School of EnvironmentalStudies (SOES), Jadavpur University, Kolkata. The inter-laboratorydifferences were within 5–10%.

2.3. Measurement of particulate pollution in indoor air

The concentrations of particulate matter with a diameter of lessthan 10 and 2.5 mm (PM10 and PM2.5, respectively) were measuredby real-time aerosol monitor (DustTrak, model 8520, TSI Inc., MN,USA) following the procedure described earlier (Banerjee et al.,2012a,b). Monitoring was carried out in each household for threeconsecutive days, 8 h/day (07:00–15:00 h).

2.4. Collection of background data

Female research assistants interviewed each participantindividually using a structured questionnaire that asked aboutage (year of birth), education (years of schooling), substance use(betel quid chewing, alcohol), housing, year of marriage, husband’s

B. Mukherjee et al. / NeuroToxicology 45 (2014) 159–167 161

occupation and smoking habits, number of children, family sizeand income, cooking fuel (biomass/liquefied petroleum gas, LPG),menstrual cycle length and regularity, occurrence of leucorrhea,mastalgia, fetal loss, preterm baby and congenital defects in anybaby delivered in the past year, food habits (vegetarian/mixed),occupation (household/household plus agricultural/salaried job)and respiratory symptoms in the past 3 months. The anthropo-metric indicators used in this study were height (m) and weight(kg) measured with shoes off, body mass index (BMI; weight in kg/height in m2), and mid-upper arm circumference (MUAC) incentimeters. MUAC values of 23.0 cm in men and 22.0 cm inwomen are useful cut-off points for simple screening of nutritionalstate in developing countries like India (James et al., 1994).Changes in menstrual cycle characteristics in the last 12 months,e.g. short cycles (<24 days), long cycles (>36 days), too irregularcycles, missed periods for 6 weeks or more, and intermenstrualbleeding (bleeding or spotting between periods) were evaluated byself-report following the protocol of Farr et al. (2004). The presenceof breast pain (mastalgia) was identified using a modified McGillPain Questionnaire (Khan and Apkarian, 2002).

2.5. Questionnaire survey for neurobehavioral symptoms

The questionnaire was adapted from the subjective symptomquestionnaire accompanying the World Health OrganizationNeurobehavioral Core Test Battery [WHO-NCTB], 1986 (Johnsonet al., 1987). Because of the poor educational standard of theparticipants, we did not use WHO-NCTB proper as it is not suitablefor people with limited education (Anger, 2003; Farizwana et al.,2013). Instead, we selected some questions from the subjectivesymptoms questionnaire containing 37 questions which list themost common discomfort in behavior, feeling and sensations that aperson may experience prior to the commencement of the NCTBtests. The questionnaire focused mainly on symptoms like burningsensation in extremities (feeling of burn in distal and terminalportions of the body such as hand and foot), tingling (repetitivemoving pin prick-like sensation), numbness (temporary loss ofsensation), transient loss of memory (short-term memory loss,usually lasting for a few hours), anxiety (unpleasant emotionalstate in anticipation of imagined danger), vertigo (an illusionarysensation that the body or surrounding environment is revolving),and dizziness (sensation of unsteadiness with a feeling ofmovement within the head, giddiness). In order to elicit a betterresponse, a five point rating scale using simple and clear words like‘never’, ‘rarely’, ‘sometimes’, ‘frequently’ and ‘very frequent’ wasused in the questionnaire. Afterwards, answers like ‘never’ and‘rarely’ were classified as no such symptom, while responses like‘sometimes’ frequently’ and very frequently’ were recognized ashaving such symptoms (Ray et al., 2006).

2.6. Detection of depressive symptoms

Beck Depression Inventory Second Edition (BDI-II) was used todetect depression and its severity (Beck et al., 1996a,b; Arnau et al.,2001). The 21 questions included emotional, cognitive, andmotivational changes related to depression experienced in dailylife. We made a Bengali (local language) translation of the BDI II.Symptoms are rated on a 4-point scale (0–3) of how intensely theyhave been experienced over a 2-week period, with increasingscores indicating greater depression severity. Responses weresummed to yield a total score that ranges from 0 to 63. Usingstandard cut-off scores (Beck et al., 1996a,b), we created fourdepression categories: 0, without depressive symptoms (score 0–13); 1, mild depressive symptoms (score 14–19); 2, moderate(score 20–28); and 3, severe depressive symptoms (score 29–63).Earlier studies have shown that the versions of the BDI-II in Indian

languages are useful instruments for the assessment of depressivesymptoms in children, adolescents and adults (Bansal et al., 2009;Basker et al., 2007; Chadda et al., 2001; Kulkarni et al., 2011;Banerjee et al., 2012a,b) as it showed the same highly acceptablepsychometric properties as the original English version (Baskeret al., 2007). BDI-II and the Patient Health Questionnaire-9 (PHQ-9)perform similarly among low-income women in terms ofdepressive symptom severity measurement and classifying levelsof depressive symptoms, and do not vary across subgroups on thebasis of select demographics (Kneipp et al., 2010).

2.7. Collection of blood

The participants were requested to take rest in supine positionin a relaxed environment for 30 min before blood drawing. Venousblood (5 mL) was collected after informed consent in vacutainertubes (Becton Dickinson, USA) anticoagulated with EDTA (forplasma catecholamines) or 3.8% (w/v) trisodium citrate (forplatelet P-selectin, 5-HT) at a fixed time of the day (9.30–11.00 h).

2.8. Measurement of surface expression of P-selectin by platelets

The expression of P-selectin (P-sel; CD 62P) on platelet surfacewas measured in citrated whole blood by flow cytometry usingphycoerythrin (PE)-conjugated anti-human CD62P monoclonalantibody and fluorescein isothiocyanate (FITC)-conjugated anti-human CD41a (platelet finder) monoclonal antibodies (BD-Pharmingen, USA). About 10,000 events were acquired in a flowcytometer (FACSCalibur with sorter, Becton Dickinson, USA). Thepercentage of P-selectin expressing platelets was calculated usingstatistical package of Cell Quest software (Becton Dickinson, USA).

2.9. Measurement of platelet serotonin by ELISA

Citrated blood samples were centrifuged at 200 � g for 10 minat room temperature. The supernatant was collected as platelet-rich plasma (PRP), and 800 mL of physiological saline was added to200 mL PRP, and centrifuged at 4500 � g for 10 min at 4 8C. Thesupernatant was collected as platelet-rich plasma (PRP), and theplatelets were counted. To 200 mL PRP, 800 mL physiological salinewas added and centrifuged at 4500 � g for 10 min at 4 8C.The supernatant was separated as platelet-poor plasma (PPP).The platelet pellet was collected, and the cells were disrupted bysonication (Ultrasonic Liquid Processor, model 385; Heat SystemsUltrasonics Inc., Farmingdale, NY, USA) and serotonin content inplatelet lysate was measured by enzyme-linked immunosorbentassay (ELISA) using a commercially available kit (IBL, Hamburg,Germany) as per the manufacturer’s instructions.

2.10. Plasma catecholamine assay

The plasma concentrations of norepinephrine (NE), epinephrine(E) and dopamine (DA) were measured in high performance liquidchromatography with electrochemical detector (HPLC–ECD). Inbrief, 200 mL of 3,4-dihydroxybenzylamine prepared as 0.1 mMsolution in 0.1 M perchloric acid (containing 400 mM sodiummetabisulphite) was added as internal standard to 2 mL of plasmasamples. Then 400 mL of 0.5 M Tris–HCl (pH 8.6) and 20 mg ofactivated alumina were added and the contents of the tube shakengently for 15 min. Following centrifugation at 600 � g for 2 min,the supernatant was removed and the catecholamines were elutedfrom alumina into 50 mL of 0.6 M perchloric acid containing400 mM sodium metabisulphite. The mixture was centrifuged at800 � g for 30 min and 20 mL of the supernatant was injected into areverse phase high performance liquid chromatographic column(WATERS Novapak C 18 column, 3.9 nm � 150 nm, coupled into an

B. Mukherjee et al. / NeuroToxicology 45 (2014) 159–167162

electrochemical detector WATERS 464 pulsed ECD, Waters, USA).The mobile phase consisted of 780 mL of acetate–citrate buffer pH5.2, 220 mL of methanol and sodium octane-1 sulphonate (5 mM)filtered through 0.45 mM Millipore filter and degassed prior to useand the flow rate was set at 1.2 mL/min.

2.11. Statistical analysis

The obtained data were analyzed using SPSS statistical software(version 10.0; SPSS Inc., Chicago, IL, USA) and are presented asmean � standard deviation (SD) and median with interquartile range(IQR) in parenthesis. Demographic characteristics of control andexposed groups were compared by Chi-square test. Mann–Whitney U

test was used for comparison of laboratory data between the controland As-exposed women. Correlation of the changes in measurableparameters with groundwater As was done by Spearman’s rank,Pearson’s and Kendall’s correlation. Multivariate logistic regressionanalysis was done to establish whether a significant association existsbetween observed changes and As exposure after controlling theinfluence of potential confounders. p < 0.05 was considered statisti-cally significant.

3. Results

3.1. Social and economic characteristics of the participants

The participants were stratified on the basis of As content intube well water as control (below 10 mg/L) and exposed (above10 mg/L). Of the 654 participants, 342 women aged 22–47 yr(median 41 yr) were in the exposed group. The remaining312 women (median age 39 yr) were in the control group.Demographic and socio-economic characteristics of control andAs-exposed women are compared in Table 1. The mean age,education, BMI, MUAC, prevalence of leucorrhea, biomass ascooking fuel, years of cooking and average cooking hours per day,environmental tobacco smoking (ETS), food and number of familymembers and family income of the exposed and control groupswere comparable. However, the two groups differed in menstrualproblems, fetal loss, mastalgia and prevalence of respiratorysymptoms in past 3 months (p < 0.05; Table 1).

The participants were all housewives engaged in domesticchores like cooking family meals twice a day, care giving tochildren and elders in the family, attention to pets and cattle,washing the clothes of the entire family and collecting drinkingwater from community tube wells. None of the participants had a

Table 1Comparison of demographic and socio-economic characteristics of arsenic-exposed an

Variable Contro

Age in year, median (range) 39 (2

Body mass index in kg/m2, median (range) 22.5

Years of schooling, median (range) 6 (4

Use of unprocessed solid biomass as cooking fuel (%) 72.4

Cooking years, median (range) 19 (5

Cooking hours per day, median (range) 4 (3

Smoker in family (%) 40.06

Abnormal menstrual cycle length (%) 35.89

Fetal loss in past 1 year (%) 2.56

Delivered baby with congenital defect in past 1 year (%) 0.06

Delivered preterm baby in past 1 year (%) 2.4

Experienced stillbirth in past 1 year (%) 0.2

Prevalence of respiratory symptoms in past 3 months (%) 28.52

Leucorrhea (%) 20.19

Mastalgia (%) 5.12

Family income/month in US $ (mean � SD) 104 �

NS, not significant in Mann–Whitney test (median, range), Chi-square test (percentage* Significant in Chi-square test.

regular, time-bound salaried job. Motor vehicles were rare in thesevillages; bicycles, cycle rickshaws and bullock carts were theprincipal mode of transport. Electricity was present in all thevillages. Most of their husbands were farmers (85.25% in controland 86.54% in exposed, p > 0.05) and remaining were smallvendors, rickshaw puller or other professions. The averageduration of chronic As exposure for all women in the exposedgroup was >15 years. The lack of exposure to agricultural/homeinsecticides or pesticides for the last month has been establishedby self-report on a questionnaire.

3.2. Arsenic levels in drinking water

The As level in tube well water of the villages where the controlwomen resided was 2.72 � 1.18 (SD) mg/L; median 2.89 mg/L, range0.89–5.07 mg/L. In contrast, the mean As concentration in the tubewell water of As-endemic villages was 28.3 � 13.51 mg/L; median29.9 mg/L, range 11.5–43.1 mg/L. The differences in the mean andmedian As concentrations in drinking water of these two areas weresignificant in Student’s t-test (p < 0.0001) and Mann–Whitney U-test(p < 0.0001), respectively.

3.3. Indoor air pollution

The concentrations of PM10 and PM2.5 in indoor air of thehouseholds of control and As-exposed women were comparable(p > 0.05): PM10 level was 78 � 32 (SD) in control vs. 84 � 29 mg/m3

in exposed; PM2.5 level 42 � 19 in control vs. 48 � 21 mg/m3 inexposed.

3.4. Prevalence of neurobehavioral symptoms

As-exposed women had a higher prevalence of psychoneur-ological symptoms and adverse reactions of the cranial (abnormalsmell, taste, vision) and peripheral nerves (feet numbness), andmucous membrane irritations (cough, shortness of breath, eye andnose irritations) (Table 2). The common symptoms among As-exposed women were transient loss of memory (69.9% vs. 28.2% incontrols; p < 0.001) and fatigue (68.4% vs. 23.4% in control,p < 0.001). Also, exposed women had higher prevalence of burningsensation in the extremities, tingling and numbness (Table 2).However, differences in the prevalence of vertigo and dizzinessbetween control and exposed groups were not significant.

Controlling for age, education, cooking years, menstrual length,adverse reproductive outcome experienced in past 1 year and

d control women.

l (n = 312) As-exposed (n = 342) p value

3–45) 40 (24–45) NS

(19.6–23.7) 21.8 (19.7–23.4) NS

–12) 5 (3–12) NS

73.4 NS

–25) 20 (3–24) NS

–6) 4 (3–6) NS

38.59 NS

57.01 <0.05*

7.60 <0.05*

0.17 <0.05*

3.3 <0.05*

0.9 <0.05*

49.12 <0.05*

19.59 NS

10.52 <0.05*

19 98 � 22 NS

) and Students’ t-test (mean � SD).

Table 2Prevalence (%) of neurological symptoms.

Control

(n = 312)

As exposed

(n = 342)

p value*

Sleep disturbance 10.25 11.40 0.7194

Palpitation 14.10 27.77 0.0002

Anxiety 17.95 43.27 <0.0001

Drunken feeling 5.44 11.98 0.0050

Fatigue 23.39 68.42 <0.0001

Inability to concentrate 23.07 36.54 0.0050

Reduced sense of taste 4.48 15.78 <0.0001

Reduced sense of smell 5.76 14.91 0.0001

Abnormal smell 5.12 11.40 0.0055

Blurred vision 4.48 13.45 <0.0001

Burning sensation in extremities 5.44 36.84 <0.0001

Tingling/numbness 5.12 25.14 <0.0001

Transient loss of memory 28.20 69.88 <0.0001

Vertigo/dizziness 9.93 10.52 0.8514

Chronic cough 10.25 27.19 <0.0001

Eye irritation 7.37 9.64 0.4030

Nose irritation 10.89 15.78 0.1383

n, number of subjects.* p values in Chi-square test.

B. Mukherjee et al. / NeuroToxicology 45 (2014) 159–167 163

family income as possible confounders, the level of groundwaterAs was positively associated with transient loss of memory (oddsratio [OR] = 2.02; 95% confidence interval [95% CI], 1.17–3.12),burning sensation in extremities (OR = 4.16; 95% CI, 1.89–7.27),tingling or numbness (OR = 3.08; 95% CI, 2.10–7.35), anxiety(OR = 1.49, 95%CI 1.16–3.46), fatigue (OR = 1.92, 95%CI 1.49–4.57),reduced sense of taste (OR = 2.29, 95%CI 1.28–4.62) and smell(OR = 1.36, 95%CI 1.12–2.74).

3.5. Changes in plasma CA levels

Plasma catecholamines (CA) were measured in 40 control and40 As-exposed women. Control and exposed women were selectedin such a way that they were comparable with respect to year ofbirth and marriage (�1 year), BMI (�1 kg/m2), education (�1 year ofschooling) and monthly family income (�3 US$). As-exposed womenhad 1.6-times more E and NE levels in blood plasma (p < 0.05), butthey had 13% lower plasma DA than the controls which was notstatistically significant (p = 0.0971; Fig. 1). After controlling spouse’ssmoking habit, menstrual disturbances, and adverse childbirth

Fig. 1. Comparison of catecholamine levels (ng/mL) in control (As < 10 mg/L) and

chronically low As-exposed (11–50 mg/L) women. Epinephrine and norepinephrine

were significantly higher (*p < 0.05 in student’s t test) in exposed women than

control group. The decrease in dopamine level in the exposed women was not

significant with respect to control.

experiences as potential confounders, the level of As in water waspositively associated with elevated plasma E (OR = 1.27; 95% CI, 1.07–1.78), and NE (OR = 1.24 95% CI, 1.06–1.91) levels.

3.6. Prevalence of depressive symptoms

The mean BDI-II score of As-exposed women was 16.8 � 11.4(SD) which was significantly higher than the mean score of9.6 � 5.2 of controls (p < 0.0001). The median BDI-II score of exposedwomen (14.6 with a range of 5.8–55.7) was also higher than that ofcontrol women (median score 8.3, range 3.7–38.8), and the differencewas significant (p < 0.001 in Mann–Whitney U test). Depressivesymptoms were present in 39.8% of As-exposed women comparedwith 19.9% of controls (p < 0.0001). Exposed women also had agreater prevalence of moderate (15.5% vs. 6.4%, p = 0.003) and severe(4.1% vs. 1.6%, p = 0.0481) depressive symptoms (Table 3).

The prevalence of depression was significantly higher inwomen who consumed water with As level >10 mg/L, aged morethan 30 years, had lower education (less than 5 years of schooling)and family income (less than <40 USD/month), had poor nutrition(MUAC < 22 cm), exposed to environmental tobacco smoke due tohusband’s smoking habit, had breast pain, had abnormal menstru-al cycle length and experienced fetal loss or have delivered babywith congenital defect in past 1 year (Table 4). After controlling theinfluences of these confounders in multivariate logistic regressionanalysis, As in drinking water was found to be independentlyassociated with depression (OR = 1.37, 95% CI 1.11–1.97).

3.7. Platelet P-selectin expression, intraplatelet and plasma serotonin

In order to examine platelet activity in relation to depression,we measured platelet P-selectin expression by flow cytometry andthe level of intraplatelet and plasma serotonin by ELISA in fourgroups of women each consisting of 20 participants: (i) controlwithout depression; (ii) control with depression; (iii) As-exposedwithout depression; and (iv) As-exposed with depression. Therespective control and exposed groups were selected in such a waythat they were comparable with respect to year of birth andmarriage, BMI, education and monthly family income.

In general, As-exposed women had significantly higherpercentage of P-selectin-expressing platelets (5.2 � 0.7% vs.2.3 � 0.5%, p < 0.05; Fig. 2A), lower intraplatelet serotonin(221 � 20 vs. 311 � 26 ng/109 platelets, p < 0.05; Fig. 2B) and higherplasma serotonin (5.7 � 1.3 vs. 10.6 � 1.7 ng/mL, p < 0.05; Fig. 2C)than the control women. Depressed women had elevated levels ofactivated (P-selectin-expressing) platelets with depleted intraplateletserotonin but elevated plasma serotonin than their non-depressedcounterparts both in As-exposed and control groups. However, themagnitude of platelet activation was much greater in depressed As-exposed women (Fig. 2). We also found 35% increase in number ofplatelets in circulation of women who were chronically exposed togroundwater As (3.1 � 0.4 vs. 2.4 � 0.3 � 105 platelet/mL in control,

Table 3Prevalence of depressive symptoms among the participants as detected by Beck’s

depression inventory (BDI-II).

BDI score Depressive

symptoms

Control women

(n = 312)

As exposed

women (n = 342)

p value

0–13 Absent 250 (80.1%) 206 (60.2%) 0.0023

14–63 Present 62 (19.9%) 136 (39.8%) <0.0001

14–19 Mild 37 (11.9%) 69 (20.2%) 0.0161

20–28 Moderate 20 (6.4%) 53 (15.5%) 0.0003

29–63 Severe 5 (1.6%) 14 (4.1%) 0.0481

The results are expressed as total number of women in each category with the

percentages in parentheses.

Fig. 2. The comparison of platelet P-selectin expression, intraplatelet serotonin and

plasma serotonin with respect to depression in both control (As < 10 mg/L) and As-

exposed women (11–50 mg/L). Platelets expressing P-selectin (A), intraplatelet

serotonin (B) and plasma serotonin (C) have been depicted in both control and

exposed women. In all the graphs statistical significance was determined with

Student’s t test where ap < 0.0001 was the significance level compared with

corresponding non-depressed women in the same group and *p < 0.0001 was the

significance level compared with depressed women in control group.

Table 4Factors affecting depression among arsenic-exposed women of rural India.

Factors Odds ratio (95% CI)

1. Age (yr)

20–30 1.00a

31–39 1.19 (1.02–1.66)b

�40 1.54 (1.06–2.42)b

2. Years of schooling, 5–10 1.00a

<5 1.44 (1.12–1.97)b

3. Food habit, vegetarian 1.00a

Mixed 0.92 (0.67–1.67)

4. Mid upper arm circumference � 22 cm 1.00a

<22 cm 1.82 (1.33–3.71)b

5. Husband non-smoker 1.00a

Smoker 1.31 (1.16–1.87)

6. Menstrual cycle length, normal (28 � 2 d) 1.00a

Abnormal (�24 to �36 d) 1.28 (1.04–1.82)b

7. Mastalgia absent 1.00a

Present 2.02 (1.36–4.18)b

8. No experience of fetal loss in past 1 year 1.00a

Experienced spontaneous abortion in past 1 year 1.99 (1.40–3.88)b

9. Delivered normal baby in past 1 year 1.00a

Delivered preterm baby in past 1 year 1.08 (0.87–1.42)

Delivered baby with congenital defect in past 1 year 1.52 (1.26–2.46)b

Delivered stillbirth in past 1 year 1.47 (1.19–2.31)b

10. Family income �50 USD 1.00a

Less than �50 USD 1.66 (1.25–3.02)b

11. As in drinking water <10 mg/L 1.00

>10 mg/L 1.89 (1.36–3.52)

a Reference group.b Statistically significant compared with reference group.

B. Mukherjee et al. / NeuroToxicology 45 (2014) 159–167164

p < 0.05). Estimation of the absolute number of P-selectin-expressingplatelets in circulation by multiplying platelet count and thepercentage of P-selectin-expressing platelets revealed nearly three-fold increased number of activated platelets in circulation of As-exposed women (16.1 � 3.2 vs. 5.5 � 1.2 � 105/mL, p < 0.001).

The percentage of P-selectin-expressing platelets in circulationcorrelated significantly with BDI score (r = 0.745, p < 0.001,Spearman’s rho 0.857, p < 0.001). Similarly platelet serotoninshowed a strong negative correlation with BDI score (r = �0.686,p < 0.001, Spearman’s rho �0.824, p < 0.001). The results suggestthe possible involvement of platelet activation and consequentsecretion of alpha granules (e.g. P-selectin) and dense granulecontents (e.g. serotonin) in the pathophysiology of depressionamong As-exposed women.

4. Discussion

We found a higher prevalence of the neurobehavioralsymptoms and depression in pre-menopausal women of easternIndia who had been drinking water contaminated with a low levelof As (11–50 mg/L) for the past 15 years or more. Our findings areconsistent with a study conducted in the United States thatshowed a high rate of depression among 1185 people who wereexposed to groundwater As- contamination ranging from 0 to2389 mg/L with a median value of 2 mg/L (Zierold et al., 2004).

The control and As-exposed women in this study were recruitedfrom the same geographical area. The ambient PM10 in these areasfor the past few years were within the national ambient air qualitystandards for residential areas (60 mg/m3 annual average), and thelevel of indoor air pollution (PM10 and PM2.5) was comparableamong the groups. Furthermore, the control and exposed groupswere similar with respect to age, food habit, family size and maritalstatus. Therefore the differences in depression prevalence betweenthese two groups could not be attributed to air pollution,demography and lifestyle factors. The observed associationbetween lower education and family income and malnutrition(MUAC of <22 cm) with depression underscores the importance of

poverty and malnutrition with mental health. Poverty is wellrecognized as a stress that affects people’s well being through thedevelopment of symptoms of anxiety and depression (Palomar-Lever and Victorio-Estrada, 2012). Consistent with the presentfindings, reports from southern and western India showed anintimate relationship between lower education, economic hard-ship and depression (Poongothai et al., 2009; Kamble et al., 2009).

Compelling evidence suggests that inflammatory and neuro-degenerative pathways contribute to the development of depres-sion (Wolkowitz et al., 2011). Many depressed individuals havehigher levels of pro-inflammatory mediators such as interleukin-6(IL-6) and tumor necrosis factor-alpha (TNF-a), which appear tointeract with many of the pathophysiological domains of depres-sion, including neuroendocrine function, neurotransmitter metab-olism, and synaptic plasticity (Catena-Dell’Osso et al., 2011;

B. Mukherjee et al. / NeuroToxicology 45 (2014) 159–167 165

Lotrich et al., 2011). As can elicit neurobehavioral changes bydirectly affecting the CNS function, because it can cross the blood–brain barrier (Itoh et al., 1990) causing a decrease in DA and NE andan increase in serotonin in midbrain and diencephalons (Chatto-padhyay and Ghosh, 2010; Yadav et al., 2010; Itoh et al., 1990;Delgado et al., 2000; Bardullas et al., 2009; Liu et al., 2013). Evensub chronic exposure to As can reduce the concentrations of themonoamine neurotransmitters in the mouse brain (Zhang et al.,2013). The CNS toxicity of As was found more among females(Bardullas et al., 2009). Plasma NE is used as a measure ofsympathetic activity and emotional or physical stress can increaseplasma E and NE levels substantially (Villecco et al., 1997).Therefore, elevated plasma E and NE levels in As-exposed womenof this study suggest stress and stimulation of the sympatheticactivity. It has been hypothesized that depression could be due to astress-related increased production of pro-inflammatory cytokinesthat, in turn, would lead to increased oxidative and nitrosativebrain damage, indoleamine 2,3-dioxygenase induction, productionof tryptophan catabolites, and consequent reduced availability oftryptophan and serotonin (Catena-Dell’Osso et al., 2011). Based onthese reports, depression among As-exposed women can beattributed in part to inflammation. The observed up-regulationof platelet P-selectin might have played a role in mediatinginflammation among As-exposed women as it helps in neutrophiltransmigration (Lam et al., 2011). Also, the greater prevalence ofrespiratory symptoms observed among exposed women of thisstudy (Table 1) may be associated with up-regulation of P-selectinexpression, because P-selectin activates eosinophil a4b1 integrinthat correlates inversely with lung function especially forcedexpiratory volume in 1 s (Johansson and Mosher, 2011). Thesefindings suggest that targeting proinflammatory mediators andtheir signaling pathways may represent a therapeutic option totreat depression and related conditions in As-exposed subjects.

As-exposed women in general and those with depressivesymptoms in particular exhibited up-regulation of platelet P-selectin expression, depletion of platelet serotonin and concomi-tant rise in the level of plasma serotonin, suggesting plateletactivation. Several investigators have shown that increased andprolonged activation of platelets (Lederbogen et al., 2004;Musselman et al., 1996) with an increased number of P-selectin-expressing platelets in peripheral blood of depressedsubjects (Morel-Kopp et al., 2009). Moreover, the severity ofdepression correlated positively with the level of plateletactivation (Morel-Kopp et al., 2009). These findings suggest thatthe blood platelets are implicated in the occurrence of depression.Both increase in P-selectin expression (Piletz et al., 2000) anddepletion of platelet serotonin are considered biomarkers ofdepression (Maurer-Spurej et al., 2004). The lower level ofplatelet serotonin, as observed in women with depressivesymptoms in this study, can be due to decreased activity ofserotonin transporter or accelerated release from dense granulesfollowing platelet activation. Since platelet surface P-selectinexpression is highly correlated with serotonin transporter density(Frankhauser et al., 2008), the reduced platelet serotonin levelamong exposed women was perhaps not due to a deficiency inserotonin transporter. Change in hormonal profile, such asestrogen deficiency, is implicated in the regulation of moodand behavior. Although we have not measured the circulatingestrogen level, a change in estrogen level following cumulative Asexposure cannot be ruled out because As altered reproductivehormones in female rats (Guo et al., 2011) by diminishing ovariankey steroidogenic enzyme activities via changes in gonadotro-phins and estradiol signaling (Chattopadhyay and Ghosh, 2010).Also, alterations in the menstrual cycle can influence the onset ofdepressive disorders (Pinkerton and Stovall, 2010). In agreementwith this, we found an association between altered menstrual

cycle lengths with depression in bivariate analysis. Similarly, apositive association was found between leucorrhea, mastalgia,fetal loss and depression. The association between mastalgia anddepression is not surprising because frequent mastalgia isstrongly associated with depression, posttraumatic stress disor-der and other psychiatric conditions (Johnson et al., 2006).Similarly women suffer some level of depression following fetalloss especially after 20 weeks of gestation (Obi et al., 2009).

We found a rise in abnormal menstrual cycle length, fetal loss,congenital birth defect, preterm baby, and stillbirth with exposureto As through drinking water. Poverty and use of biomass fuel forcooking have been linked with adverse pregnancy outcomes inrural India (Wylie et al., 2014). However, the family income and thepractice of cooking with biomass were similar in As-exposed andcontrol women of this study. Therefore, it seems that the observedchanges in reproductive health were not due to poverty or indoorair pollution. Instead, chronic As exposure could be responsible forthese changes. Support for this argument comes from the studiesconducted in neighboring Bangladesh that showed impairment offetal development (Kippler et al., 2012) and increased risk ofstillbirth (Cherry et al., 2008) in association with chronic Asexposure through drinking water.

In summary, we found that even low level of As in drinkingwater which is within the permissible limit in India is harmful as itincreases the prevalence of depressive and neurobehavioralsymptoms in women. However, caution in interpretation of ourresults is warranted as this study has some limitations. First, beinga cross-sectional study, it limits our inference on the causaldirection. Second, we did not know the As levels in the participants’drinking water over a period of time so that how much arsenic awoman has actually ingested in the past 10 or 15 years is onlyspeculative. Third, we only measured groundwater As; we did notexplore the possibility of other co-minerals and metals in the watersamples contributing to health outcomes. Moreover sources of Asform food chain have not been measured in this study. Fourth,there are several other environmental pollutants (agriculturalpesticide, for example) that may cause depression. Like As,organophosphate and carbamate pesticides decrease acetylcho-linesterase in animals (Patlolla and Tchounwou, 2005) as well as inhumans subjects (Ali et al., 2010). Despite these shortcomings, oursample size was large enough to conclude that there wasconvincing evidence to support the association between chroniclow level As exposure and depression and neurobehavioralchanges in rural women. In essence, the findings of this studysuggest that the prevailing Indian Standard for As in drinking wateris too high for safety of the population, and that the WHOrecommended limit of 10 mg/L of As in drinking water should beimplemented in the country to protect public health.

Conflict of interest

The authors declare that there are no conflicts of interest.

Transparency document

The Transparency document associated with this article can befound in the online version.

Acknowledgments

The authors would like to acknowledge the funding agency,Central Pollution Control Board, Govt. of India, CNCI/CPCB/MR-4for providing the grant support for the project. The authors wouldalso like to thank Prof. (Dr.) Jaydip Biswas, Director, ChittaranjanNational Cancer Institute for providing the infrastructural facilities.

B. Mukherjee et al. / NeuroToxicology 45 (2014) 159–167166

References

Aballay LR, Dıaz Mdel P, Francisca FM, Munoz SE. Cancer incidence and pattern ofarsenic concentration in drinking water wells in Cordoba, Argentina. Int J EnvironHealth Res 2012;22:220–31. http://dx.doi.org/10.1080/09603123.2011.628792.

Agusa T, Kunito T, Kubota R, Inoue S, Fujihara J, Minh TB, et al. Exposure, metabolism,and health effects of arsenic in residents from arsenic-contaminated groundwa-ter areas of Vietnam and Cambodia: a review. Rev Environ Health 2010;25:193–220.

Arnau RC, Meagher MW, Norris MP, Bramson R. Psychometric evaluation of the beckdepression inventory-II with primary care medical patients. Health Psychol2001;20:112–9. http://dx.doi.org/10.1037/0278-6133.20.2.112.

Anger WK. Neurobehavioural tests and systems to assess neurotoxic exposures in theworkplace and community. Occup Environ Med 2003;60:531–8. http://dx.doi.org/10.1136/oem.60.7.531.

Anwar J, Mpofu E, Matthews LR, Shadoul AF, Brock KE. Reproductive health and accessto healthcare facilities: risk factors for depression and anxiety in women with anearthquake experience. BMC Public Health 2011;11:523. http://dx.doi.org/10.1186/1471-2458-11-523.

Amster ED, Cho JI, Christiani D. Urine arsenic concentration and obstructive pulmonarydisease in the U.S. population. J Toxicol Environ Health A 2011;74:716–27. http://dx.doi.org/10.1080/15287394.2011.556060.

Ali N, Hoque MA, Haque A, Salam KA, Karim MR, Rahman A, et al. Association betweenarsenic exposure and plasma cholinesterase activity: a population based studyin Bangladesh. Environ Health 2010;9:36. http://dx.doi.org/10.1186/1476-069X-9-36.

Banerjee A, Mondal NK, Das D, Ray MR. Neutrophilic inflammatory response andoxidative stress in premenopausal women chronically exposed to indoor airpollution from biomass burning. Inflammation 2012a;35:671e683. http://dx.doi.org/10.1007/s10753-011-9360-2.

Bansal V, Goyal S, Srivastava K. Study of prevalence of depression in adolescentstudents of a public school. Ind Psychiatry J 2009;18:43–6. http://dx.doi.org/10.4103/0972-6748.57859.

Basker M, Moses PD, Russell S, Russell PS. The psychometric properties of Beckdepression inventory for adolescent depression in a primary-care paediatricsetting in India. Child Adolesc Psychiatry Ment Health 2007;1:8. http://dx.doi.org/10.1186/1753-2000-1-8.

Banerjee M, Siddique S, Dutta A, Mukherjee B, Ranjan Ray M. Cooking with biomassincreases the risk of depression in pre-menopausal women in India. Soc Sci Med2012b;75:565–72. http://dx.doi.org/10.1016/j.socscimed.2012.03.021.

Bardullas U, Limon-Pacheco JH, Giordano M, Carrizales L, Mendoza-Trejo MS, Rodrı-guez VM. Chronic low-level arsenic exposure causes gender-specific alterations inlocomotor activity, dopaminergic systems, and thioredoxin expression in mice.Toxicol Appl Pharmacol 2009;239:169–77. http://dx.doi.org/10.1016/j.taap.2008.12.004.

Beck AT, Steer RA, Ball R, Ranieri WF. Comparison of beck depression inventories-IA and-II in psychiatric outpatients. J Pers Assess 1996a;67:588–97. http://dx.doi.org/10.1207/s15327752jpa6703_13.

Beck AT, Steer RA, Brown GK. Beck depression inventory manual. 2nd ed. San Antonio,TX: Psychological Corporation; 1996.

Behari JR, Prakash R. Determination of total arsenic content in water by atomicabsorption spectroscopy (AAS) using vapour generation assembly (VGA). Chemo-sphere 2006;63:17–21. http://dx.doi.org/10.1016/j.chemosphere.2005.07.073.

Bureau of Indian Standards [BIS]. Draft Indian standard. Drinking water specification(second revision of IS 10500). Doc: FAD 25 (2047) C. Last date for comments: 24/12/2009., 2009 http://bis.org.in/sf/fad/FAD25(2047)C.pdf.

Calderon J, Navarro ME, Jimenez-Capdeville ME, Santos-Diaz MA, Golden A, Rodriguez-Leyva I, et al. Exposure to arsenic and lead and neuropsychological developmentin Mexican children. Environ Res 2001;85:69–76. http://dx.doi.org/10.1006/enrs.2000.4106.

Chattopadhyay S, Ghosh D. Role of dietary GSH in the amelioration of sodium arsenite-induced ovarian and uterine disorders. Reprod Toxicol 2010;30:481–8. http://dx.doi.org/10.1016/j.reprotox.2010.05.002.

Chattopadhyay S, Bhaumik S, Nag Chaudhury A, Das Gupta S. Arsenic induced changesin growth development and apoptosis in neonatal and adult brain cells in vivo andin tissue culture. Toxicol Lett 2002;128:73–84. http://dx.doi.org/10.1016/S0378-4274 (01)00535-5.

Cherry N, Shaikh K, McDonald C, Chowdhury Z. Stillbirth in rural Bangladesh: arsenicexposure and other etiological factors: a report from Gonoshasthaya Kendra. BullWorld Health Organ 2008;86:172–7. http://dx.doi.org/10.2471/BLT.07.043083.

Camacho A, Dimsdale JE. Platelets and psychiatry: lessons learned from old and newstudies. Psychosom Med 2000;62:326–36.

Catena-Dell’Osso M, Bellantuono C, Consoli G, Baroni S, Rotella F, Marazziti D. Inflam-matory and neurodegenerative pathways in depression: a new avenue for antide-pressant development. Curr Med Chem 2011;18:245–55. http://dx.doi.org/10.2174/092986711794088353.

Chadda RK, Agarwal V, Singh MC, Raheja D. Help seeking behavior of psychiatricpatients before seeking care at a mental hospital. Int J Soc Psychiatry 2001;47:1–8.http://dx.doi.org/10.1177/002076400104700406.

Delgado JM, Dufour L, Grimaldo JI, Carrizales L, Rodrıguez VM, Jimenez-Capdeville ME.Effects of arsenite on central monoamines and plasmatic levels of adrenocortico-tropic hormone (ACTH) in mice. Toxicol Lett 2000;117:61–7. http://dx.doi.org/10.1016/S0378-4274(00)00240-X.

Emanuele E, Bertona M, Minoretti P, Geroldi D. An open-label trial of L-5-hydroxy-tryptophan in subjects with romantic stress. Neuro Endocrinol Lett 2010;31:663–6.

Farizwana S, Ridzwan M, Annual ZF, Sahani M, Ghazali AR. Neurobehavioral perfor-mance of estate residents with privately-treated water supply. Iran J Publ Health2013;42:1374–86.

Farr SL, Cooper GS, Cai J, Savitz DA, Sandler DP. Pesticide use and menstrual cyclecharacteristics among premenopausal women in the agricultural health study. AmJ Epidemiol 2004;160:1194–204. http://dx.doi.org/10.1093/aje/kwi006.

Frankhauser P, Baranyai R, Ahrens T, Schloss P, Deuschle M, Lederbogen F. Plateletsurface P-selectin expression is highly correlated with serotonin transporterdensity in human subjects. Thromb Haemost 2008;100:1201–3. http://dx.doi.org/10.1160/TH08-04-0227.

German L, Kahana C, Rosenfeld V, Zabrowsky I, Wiezer Z, Fraser D, et al. Depressivesymptoms are associated with food insufficiency and nutritional deficiencies inpoor community-dwelling elderly people. J Nutr Health Aging 2011;15:3–8. http://dx.doi.org/10.1007/s12603-011-0005-9.

GuhaMazumder DN. Effect of chronic intake of arsenic-contaminated water on liver.Toxicol Appl Pharmacol 2005;206:169–75. http://dx.doi.org/10.1016/j.taap.2004.08.025.

Guha Mazumder DN, Haque R, Gosh N, De BK, Santra A, Chakraborty D, et al. Arseniclevels in drinking water and the prevalence of skin lesions in West Bengal, India. IntJ Epidemiol 1998;27:871–7. http://dx.doi.org/10.1093/ije/27.5.871.

Guha Mazumder D, Dasgupta UB. Chronic arsenic toxicity: studies in West Bengal,India. Kaohsiung J Med Sci 2011;27:360–70. http://dx.doi.org/10.1016/j.kjms.2011.05.003.

Guo Z, Guo H, Xia Y. Effects on endocrine system of female rats exposed to chronicarsenic. Wei Sheng Yang Jiu 2011;40:178–9.

Glied S, Herzog K, Frank R. Review: the net benefits of depression management inprimary care. Med Care Res Rev 2010;67:251–74. http://dx.doi.org/10.1177/1077558709356357.

Itoh T, Zhang YF, Murai S, Saito H, Nagahama H, Miyate H, et al. The effect of arsenictrioxide on brain monoamine metabolism and locomotor activity of mice. ToxicolLett 1990;54:345–53.

James WP, Mascie-Taylor GC, Norgan NG, Bistrian BR, Shetty PS, Ferro-Luzzi A. Thevalue of arm circumference measurements in assessing chronic energy deficiencyin third world adults. Eur J Clin Nutr 1994;48:883–94.

Johansson MW, Mosher DF. Activation of beta1 integrins on blood eosinophils by P-selectin. Am J Respir Cell Mol Biol 2011;45:889–97. http://dx.doi.org/10.1165/rcmb.2010-0402OC.

Johnson BL, Baker EL, Batawi EI, MGilioli R, Hanninen H, Seppalainen AM, et al.Prevention of neurotoxic illness in working populations. New York: John Wiley &Sons; 1987.

Johnson KM, Bradley KA, Bush K, Gardella C, Dobie DJ, Laya MB. Frequency of mastalgiaamong women veterans. Association with psychiatric conditions and unexplainedpain syndromes. J Gen Intern Med 2006;21:S70–5. http://dx.doi.org/10.1111/j.1525-1497.2006.00378.x.

Khan SA, Apkarian AV. The characteristics of cyclical and noncyclical mastalgia: aprospective study using a modified McGill pain questionnaire. Breast Cancer ResTreat 2002;75:147–57.

Kneipp SM, Kairalla JA, Stacciarini JM, Pereira D, Miller MD. Comparison of depressivesymptom severity scores in low-income women. Nurs Res 2010;59:380–8. http://dx.doi.org/10.1097/NNR.0b013e3181f84ee9.

Kippler M, Wagatsuma Y, Rahman A, Nermell B, Persson LA. Raqib R, et al. Environ-mental exposure to arsenic and cadmium during pregnancy and fetal size: alongitudinal study in rural Bangladesh. Reprod Toxicol 2012;34:504–11. http://dx.doi.org/10.1016/j.reprotox.2012.08.002.

Kulkarni HS, Kulkarni KR, Mallampalli A, Parkar SR, Karnad DR, Guntupalli KK.Comparison of anxiety, depression, and post-traumatic stress symptoms in rela-tives of ICU patients in an American and an Indian public hospital. Indian J Crit CareMed 2011;15:147–56. http://dx.doi.org/10.4103/0972-5229.84891.

Kvamme JM, Grønli O, Florholmen J, Jacobsen BK. Risk of malnutrition is associatedwith mental health symptoms in community living elderly men and women: theTromsø study. BMC Psychiatry 2011;11:112. http://dx.doi.org/10.1186/1471-244X-11-112.

Kamble SV, Dhumale GB, Goyal RC, Phalke DB, Ghodke YD. Depression among elderlypersons in a primary health centre area in Ahmednagar, Maharashtra. Indian JPublic Health 2009;53:253–5.

Lam FW, Burns AR, Smith CW, Rumbaut RE. Platelets enhance neutrophil trans-endothelial migration via P-selectin glycoprotein ligand-1. Am J Physiol HeartCirc Physiol 2011;300:H468–75. http://dx.doi.org/10.1152/ajpheart.00491.2010.

Lotrich FE, El-Gabalawy H, Guenther LC, Ware CF. The role of inflammation in thepathophysiology of depression: different treatments and their effects. J RheumatolSuppl 2011;88:48–54. http://dx.doi.org/10.3899/jrheum.110903.

Lederbogen F, Baranyai R, Gilles M, Menart-Houtermans B, Tschoepe D, Deuschle M.Effect of mental and physical stress on platelet activation markers in depressedpatients and healthy subjects: a pilot study. Psychiatry Res 2004;127:55–64.http://dx.doi.org/10.1016/j.psychres.2004.03.008.

Liu X, Piao F, Li Y. Protective effect of taurine on the decreased biogenic amineneurotransmitter levels in the brain of mice exposed to arsenic. Adv Exp MedBiol 2013;776:277–87. http://dx.doi.org/10.1007/978-1-4614-6093-0_26.

Morel-Kopp MC, McLean L, Chen Q, Tofler GH, Tennant C, Maddison V, et al. Theassociation of depression with platelet activation: evidence for a treatment effect. JThromb Haemost 2009;7:573–81. http://dx.doi.org/10.1111/j.1538-7836.2009.03278.x.

Maurer-Spurej E, Pittendreigh C, Solomons K. The influence of selective serotoninreuptake inhibitors on human platelet serotonin. J Thromb Haemost 2004;91:119–28. http://dx.doi.org/10.1160/TH03-05-0330.

B. Mukherjee et al. / NeuroToxicology 45 (2014) 159–167 167

McClung CA. Circadian rhythms and mood regulation: insights from preclinical mod-els. Eur Neuropsychopharmacol 2011;21:S683–93. http://dx.doi.org/10.1016/j.euroneuro.2011.07.008.

Mendoza-Sotelo J, Torner C, Alvarado-Vasquez N, Lazo-Langner A, Lopez G, Arango I, etal. Ultrastructural changes and immunolocalization of P-selectin in platelets frompatients with major depression. Psychiatry Res 2010;176:179–82. http://dx.doi.org/10.1016/j.psychres.2009.07.021.

Musselman DL, Tomer A, Manatunga AK, Knight BT, Porter MR, Kasey S, et al.Exaggerated platelet reactivity in major depression. Am J Psychiatry1996;153:1313–7.

Nafees AA, Kazi A, Fatmi Z, Irfan M, Ali A, Kayama F. Lung function decrement witharsenic exposure to drinking groundwater along River Indus: a comparative cross-sectional study. Environ Geochem Health 2011;33:203–16. http://dx.doi.org/10.1007/s10653-010-9333-7.

Nasreen HE, Kabir ZN, Forsell Y, Edhborg M. Low birth weight in offspring of womenwith depressive and anxiety symptoms during pregnancy: results from a popula-tion based study in Bangladesh. BMC Public Health 2010;10:515. http://dx.doi.org/10.1186/1471-2458-10-515.

Obi SN, Onah HE, Okafor II. Depression among Nigerian women following pregnancyloss. Int J Gynaecol Obstet 2009;105:60–2. http://dx.doi.org/10.1016/j.ijgo.2008.11.036.

Palomar-Lever J, Victorio-Estrada A. Factors that influence emotional disturbance inadults living in extreme poverty. Scand J Psychol 2012;53:158–64. http://dx.doi.org/10.1111/j.1467-9450.2011.00921.x.

Patlolla AK, Tchounwou PB. Serum acetyl cholinesterase as a biomarker of arsenicinduced neurotoxicity in sprague-dawley rats. Int J Environ Res Public Health2005;2:80–3. http://dx.doi.org/10.3390/ijerph2005010080.

Pinkerton JV, Stovall DW. Reproductive aging, menopause, and health outcomes. Ann NY Acad Sci 2010;1204:169–78. http://dx.doi.org/10.1111/j.1749-6632.2010.05526.x.

Piletz JE, Zhu H, Madakasira S, Pazzaglia P, Lindsay DeVane C, Goldman N, et al.Elevated P-selectin on platelets in depression: response to bupropion. J PsychiatrRes 2000;34:397–404. http://dx.doi.org/10.1016/S0022-3956(00)00034-0.

Poongothai S, Pradeepa R, Ganesan A, Mohan V. Prevalence of depression in a largeurban South Indian population e the Chennai urban rural epidemiology study(CURES-70). PLoS ONE 2009;4:e7185. http://dx.doi.org/10.1371/journal.-pone.0007185.

Rodrıguez VM, Jimenez-Capdeville ME, Giordano M. The effects of arsenic exposure onthe nervous system. Toxicol Lett 2003;145:1–18. http://dx.doi.org/10.1016/S0378-4274(03)00262-5.

Ray MR, Basu C, Roychoudhury S, Banik S, Lahiri T. Plasma catecholamine levels andneurobehavioral problems in firefighters of India. J Occup Health 2006;48:210–5.http://dx.doi.org/10.1539/joh.48.210.

Rosado JL, Ronquillo D, Kordas K, Rojas O, Alatorre J, Lopez P, et al. Arsenic exposureand cognitive performance in Mexican schoolchildren. Environ Health Perspect2007;115:1371–5. http://dx.doi.org/10.1289/ehp.9961.

Smith NM, Lee R, Heitkemper DT, DeNicola Cafferky K, Haque A, Henderson AK. Inorganicarsenic in cooked rice and vegetables from Bangladeshi households. Sci Total Environ2006;370:294–301. http://dx.doi.org/10.1016/j.scitotenv.2006.06.010.

Smith AH, Steinmaus CM. Health effects of arsenic and chromium in drinking water:recent human findings. Annu Rev Public Health 2009;30:107–22. http://dx.doi.org/10.1146/annurev.publhealth.031308.100143.

Song C, Halbreich U, Han C, Leonard BE, Luo H. Imbalance between pro- and anti-inflammatory cytokines, and between Th1 and Th2 cytokines in depressedpatients: the effect of electroacupuncture or fluoxetine treatment. Pharmacopsy-chiatry 2009;42:182–8. http://dx.doi.org/10.1055/s-0029-1202263.

Srinivasan V, Pandi-Perumal SR, Trakht I, Spence DW, Hardeland R, Poeggeler B, et al.Pathophysiology of depression: role of sleep and the melatonergic system. Psy-chiatry Res 2009;165:201–14. http://dx.doi.org/10.1016/j.psychres.2007.11.020.

Tsai SY, Chou HY, The HW, Chen CM, Chen CJ. The effects of chronic arsenic exposure fromdrinking water on the neurobehavioral development in adolescence. Neurotoxicology2003;24:747–53. http://dx.doi.org/10.1016/S0161-813X(03)00029-9.

Villecco AS, de Aloysio D, Radi D, Sprovieri G, Bargossi AM, Grossi G, et al. Plasmacatecholamines in pre- and in postmenopausal women with mild to moderateessential hypertension. J Hum Hypertens 1997;11:157–162, doi. 10.1038/sj.jhh.1000411.

Wasserman GA, Liu X, Parvez F, Factor-Litvak P, Ahsan H, Levy D, et al. Arsenic andmanganese exposure and children’s intellectual function. Neurotoxicology2011;32:450–7. http://dx.doi.org/10.1016/j.neuro.2011.03.009.

Wolkowitz OM, Mellon SH, Epel ES, Lin J, Dhabhar FS, Su Y, et al. Leukocyte telomerelength in major depression: correlations with chronicity, inflammation and oxi-dative stress – preliminary findings. PLoS ONE 2011;23(6):e17837. http://dx.doi.org/10.1371/journal.pone.0017837.

Wylie BJ, Coull BA, Hamer DH, Singh MP, Jack D, Yeboah-Antwi K, et al. Impact ofbiomass fuels on pregnancy outcomes in central East India. Environ Health2014;13(1):1. http://dx.doi.org/10.1186/1476-069X-13-1.

Yadav RS, Shukla RK, Sankhwar ML, Patel DK, Ansari RW, Pant AB, et al. Neuropro-tective effect of curcumin in arsenic-induced neurotoxicity in rats. Neurotoxicol-ogy 2010;31:533–9. http://dx.doi.org/10.1016/j.neuro.2010.05.001.

Zhang J, Liu X, Zhao L, Hu S, Li S, Piao F. Subchronic exposure to arsenic disturbed thebiogenic amine neurotransmitter level and the mRNA expression of synthetase inmice brains. Neuroscience 2013;241:52–8. http://dx.doi.org/10.1016/j.neurosci-ence.2013.03.014.

Zierold KM, Knobeloch L, Anderson H. Prevalence of chronic diseases in adultsexposed to arsenic-contaminated drinking water. Am J Public Health2004;94:1936–7.