J. Munan, G. Janji, Influence of carcinogen compounds on hydrogen bonds in water.
Contemporary Materials, III1 (2012) Page 123 of 130
Original scientific papers UDK 502.36+504.45:[626.8
doi: 10.7251/COM1201123M
INFLUENCE OF CARCINOGEN COMPOUNDS
ON HYDROGEN BONDS IN WATER
J. Munan
1,*
, G. Janji
1
1
Biomedical Engineering, Faculty of Mechanical Engineering, University of Belgrade
Kraljice Marije 16, 11000 Belgrade, Serbia
Abstract: Citizens of the town of Zrenjanin Serbia have been exposed, to high
levels of arsenic in drinking water for long time. By looking at the structure of cate-
gories of leading causes of death, and registered diseases in adults and children, it is
evident that among them there are those associated with the exposure to arsenic.
In our previous studies we reported the effects of Zrenjanin tap water on tissues and
organism on the whole, using the method of opto-magnetic spectroscopy. Here, we
report our first results of Zrenjanin tap water analysis in comparison with other drin-
king waters that satisfy WHO recommendations, by using NIR spectroscopy and a new
conceptual approach to studying aqueous systems - quaphotomics.
Keywords: carcinogen, arsenic, water, near infrared spectroscopy,
aquaphotomics.
1. INTRODUCTION
Arsenic is class one human carcinogen [1].
The primary route of exposure to arsenic comes
from the drinking water, where it is mainly present
in inorganic species such as arsenate [As(V)] and/or
arsenite [As(III)].
The WHO recommended value for arsenic
concentration in drinking water is 10g/l and the
maximum concentration limit is 50g/l [2].
Multiple studies have confirmed that a long-
term exposure to arsenic leads to a number of vario-
us cancerous (skin cancer, lung cancer, bladder can-
cer etc.) and non-cancerous diseases (skin diseases,
vascular diseases, reproductive toxicity, diabetes
mellitus), some of which are very puzzling such
as diabetes mellitus [3]. Biological mechanisms by
which arsenic exerts its toxic and carcinogenic acti-
vities are not well understood.
The citizens of town Zrenjanin in Serbia have
been for a long time exposed to high levels of arse-
nic in drinking water. By looking at structures of
categories of leading causes of death (Fig.1a), or
registered diseases in adults (Fig.1b) and children
(data not shown, for further information see ref. [4])
it is evident that among them there are those associa-
ted with the exposure to arsenic. In our previous stu-
dies we reported the effects of Zrenjanin tap water
on tissues and the organism as a whole, using opto-
magnetic spectroscopy [5]. Here, we report our first
results of Zrenjanin tap water analysis in comparison
with other drinking waters that satisfy WHO
recommendations, by using Near infra red
spectroscopy and a new conceptual approach to
studying water systems Aquaphotomics [6].
2. MATERIAL
The samples investigated were Millipore pure
water (Milli Systems Co., USA), one kind of com-
mercial natural mineral water from Serbia (Aqua
Viva), two kinds of tap water from Serbian cities of
Belgrade (Belgrade tap water) and Zrenjanin (Zre-
njanin tap water). Commercial mineral water was
bought at a supermarket and stored in original pla-
stic bottle with a plastic screw cap. Samples of tap
waters were collected using polyethylene bottles
thoroughly rinsed with pure water prior to collec-
ting.
Basic information about particular water sample was
obtained from the manufacturer label on the bottle as
well as the data from the manufacturers website [7].
Physico-chemical properties of tap waters were
obtained from the last published data from Health
Institutions in Zrenjanin and Belgrade. These data
are summarized in Table 1.
*
Corresponding author: jmuncan@mas.bg.ac.rs
J. Munan, G. Janji, Influence of carcinogen compounds on hydrogen bonds in water.
Contemporary Materials, III1 (2012) Page 124 of 130
Figure 1. a) Leading causes of death categories in Zrenjanin county in 2007, b) Registered diseases in adults in Zrenja-
nin county in 2007 (based on data from ref. [4])
Table 1. Physico-chemical properties of investigated waters
pH EC
[S/cm]
As B Ca
2+
Fe
2+
K
+
Mg
2+
Na
+
HCO
3
-
Cl
-
SO
4
2-
[mg/L]
Zrenjanin 7.56 928 71.9 895 27.4 87.5 1.2 13.6 275 846 17.4 0.97
Belgrade 7.56 318 0.29 22.2 61.1 2.34 1.1 11.1 6.5 194 11.3 37.5
Aqua Viva 7.62 504 0.0006 0.1 90.1 0.05 2.1 13.3 13.7 341 20.6 28.1
3. METHOD
3.1 Near-infrared spectral analysis
NIR transmission spectra of water samples in
the 400-2500nm region were recorded using NIR
Systems 6500 spectrophotometer (FOSS-
NIRSystems). A quartz liquid sample cell was used
as a container.
The experiment was conducted 5 times, and
each time the ambient conditions were recorded.
These recorded data are presented in Table 2.
3.2 Multivariate analysis
All multivariate analysis was carried out by
Pirouette ver.4.0 (Infometrics, USA) software pro-
gram. Multivariate data analysis in the form of Prin-
cipal Component Analysis (PCA), Partial Least
Squares Regression (PLS) and Soft Modeling of
Class Analogies (SIMCA) was applied to obtain
quantitative and qualitative information from the
spectra. All multivariate spectral analysis was car-
ried out by Pirouette ver.4.0 (Infometrics, USA)
software program.
Principal component analysis is a method for
compressing the data by using orthogonal matrix
decomposition. Soft modeling of class analogies
employs principal components analysis of spectra
for the construction of mathematical models for each
class to be analyzed. The goal of partial least
squares regression is to predict or analyze a set of
dependent variables (for example concentration of
elements in water) from a set of independent predic-
tors ( e.g. IR spectra).
Before all analysis, the data were mean-
centered and a smoothing transformation (11 points)
was applied. The models were validated using cross-
validation (leave-three-out).
Only 1
st
overtone region of the water spectra
was used in the analysis (Fig. 2).
Table 2. Experimental conditions
Exp. No Date p [hPa] t [C] Humidity [%]
1 03/07/2011 998.9 24 19
2 03/08/2011 1000.3 23 15
3 03/09/2011 1001.9 24.4 9
4 03/10/2011 1003.7 23.8 13
5 03/16/2011 992.8 24.4 14
J. Munan, G. Janji, Influence of carcinogen compounds on hydrogen bonds in water.
Contemporary Materials, III1 (2012) Page 125 of 130
Figure 2. First overtone water absorbance region (1300-1600nm). Only this region of the water spectra was used in
the analysis.
3.3. Aquaphotomics
Aquaphotomics is a new term introduced to
describe a concept of approaching water as a multi-
element system that could be well described by its
multidimensional spectra [6]. In a series of
experiments, vis-NIR spectra were collected and
regression models of respective perturbations were
analyzed in order to see if there were common bands
in various biological systems and it was found that a
group of WAMACS (water matrix coordinates
characteristic water absorbance bands) repeatedly
occurred in different combinations in the spectral
models predicting the investigated perturbations [6].
It was found that in the area of the first over-
tone of water there were 12 such characteristic
wavelength ranges [6]. Therefore, characteristic
absorbance pattern for water or any aqueous system
can be described using this 12 WAMACS while the
graphical representation of this pattern is presented
in the form of aquagrams.
4. RESULTS AND DISCUSSION
In this study we applied NIR spectral analysis
to investigate the differences between 4 types of
water: Zrenjanin tap water known to have large
amounts of arsenic compounds proven to be a
human carcinogen; Belgrade tap water that satisfies
all WHO recommendations for drinking water, Aqua
Viva commercial mineral water with optimally
balanced mineral complex according to the produ-
cers marketing and Milli-Q pure water.
Raw spectra for all investigated waters recor-
ded in all experiments are presented in Fig. 3. It is
evident that spectra of different waters, or different
experiments are hard to recognize by visual inspec-
tion.
4.1 Aquaphotomics approach
Waters were investigated from the
Aquaphotomics point of view: a concept of approac-
hing the water as a multielement system. Thus, spe-
cial attention was paid to the area of 1
st
overtone of
water, where 12 characteristic wavelength ranges
were identified as especially important water bands
where the highest spectral variations were likely to
occur, with respect to biological systems functio-
ning.
Waters in different experiments showed
slightly different spectral signatures in the region of
1
st
overtone of water, which is reflected in water
aquagrams (Fig. 4 9).
J. Munan, G. Janji, Influence of carcinogen compounds on hydrogen bonds in water.
Contemporary Materials, III1 (2012) Page 126 of 130
Figure 3. IR spectra of all analyzed water samples recorded in all experiments.
Figure 4. Aquagram of analyzed waters in Experiment 1
Figure 5. Aquagram of analyzed waters in Experiment 2
J. Munan, G. Janji, Influence of carcinogen compounds on hydrogen bonds in water.
Contemporary Materials, III1 (2012) Page 127 of 130
Figure 6. Aquagram of analyzed waters in Experiment 3
Figure 7. Aquagram of analyzed waters in Experiment 4
Figure 8. Aquagram of analyzed waters in Experiment 5
J. Munan, G. Janji, Influence of carcinogen compounds on hydrogen bonds in water.
Contemporary Materials, III1 (2012) Page 128 of 130
When comparing these aquagrams, it is clear
that despite different experiment conditions, Zrenja-
nin tap water shows similar behavior. It has lower
absorbance at S0 (free molecule species), S1, S2, S3,
S4, H
5
O
2
(water molecules with i hydrogen bonds)
bands when compared to other waters, while
consistently higher absorbance at 1344 (3), 1364
(water shell), 1372 (1+ 3). The exception to this
happens in Exp. 3, but we believe that this exception
is rather a consequence of experiment conditions.
4.2 Soft modeling of class analogies
Spectra of the waters from different
experiments were analyzed separately. In each case
classes were well separated (Mahalanobis distance
>>3) and distinctive. The discrimination power in
cases of all experiments analysis was plotted in
order to observe which wavelengths are most influ-
ential in terms of separating waters into different
classes. The illustration of this process is given in
Fig. 9 for the waters spectral data recorded in the
first experiment. Wavelengths with the biggest
influence on discrimination between different waters
are summarized in Table 3.
4.3 Partial least squares regression
Spectra of waters from different experiments
were analyzed separately. Regression vectors for
each experiment model had a similar shape to the
one presented in the Fig. 10 which is a regression
vector from the results of partial least squares
regression applied on spectral data from the first
experiment.
Figure 9.Discriminating power plot from the SIMCA
analysis of the spectra of waters from the first experiment
Figure 10. Regression vector from the partial least squares
regression applied on the spectra of waters from the first
experiment. Regression vectors for all experiments had a
similar shape.
Table 3. Most influential wavelengths for separating different waters spectra into distinctive classes
Experiment No. 1 2 3 4 5
Wavelengths with highest discrimination power [nm] 1478 1462 1414 1456 1478
Table 4. Most prominent wavelengths in regression vectors from the regression models built on spectral data from dif-
ferent experiments
Experiment No. 1 2 3 4 5
Most prominent wavelengths [nm] 1454 1458 1442 1450 1466
5. CONCLUSION
The citizens of town Zrenjanin in Serbia have
been for a long time exposed to arsenic in drinking
water, and we previously investigated the effect of
drinking of this water in animal studies using opto-
magnetic spectroscopy [5]. Although arsenic is a
proven carcinogen, it is not well understood how the
exposure to arsenic from drinking water leads to
developing cancer; its association with diabetes mel-
litus is even more puzzling [8].
J. Munan, G. Janji, Influence of carcinogen compounds on hydrogen bonds in water.
Contemporary Materials, III1 (2012) Page 129 of 130
In this study, we applied NIR spectral analysis
to investigate the differences of Zrenjanin drinking
water compared to other waters which satisfy stan-
dard regulation requirements in order to understand
how the presence of arsenic influences the organiza-
tion of water molecules.
All waters showed different behavior depen-
ding on their composition, and small variations as a
result of differences in pressure and temperature,
which was expected given that water is very sensiti-
ve to changes in pressure and temperature [9]. We
believe that the different water dynamics observed
through repeated experiments is a result of macro
and micro heterogeneity of water samples.
However, some spectral features consistently
appeared in all experiments. By studying the absor-
bance pattern through aquagrams it is evident that
Zrenjanin water shows lower absorbance in bands of
free water molecules (S0) or water clusters (Si) with
i hydrogen bonds (i=1,2,3,4), and also at 1438nm
H2O-R where both lone pairs of the oxygen elec-
trons bound to water clusters. Also, this water
consistently showed higher absorbances at 1344nm
(3), 1364 nm (1
st
overtone OH-stretch [OH-
(H
2
O)
2
]), and 1372nm (1+ 3). This means that
Zrenjanin tap water has actually very few free water
molecules and water clusters available. It has been
particularly suggested that water molecules making
two hydrogen bonds might be of special importance
for water dynamics. All these findings imply that
Zrenjanin water may not have enough significant
water species for proper hydration of biological
structures in living systems.
The number of free water molecules and water
clusters proved to be a basis for discrimination
between water types (SIMCA analysis). Most pro-
minent wavelengths in regression vector having cor-
relation with arsenic are found to be in C8: 1450 nm
[1
st
overtone DDA symmetric stretch [OH-(H
2
O)
4,5
]
) water absorbance band and C9 : (S2) water band.
It is our intention to continue and expand this
research on skin tissues and blood in living orga-
nisms in order to find out how specific molecular
organization of water affects hydration of organisms.
We believe that this can lead to further understan-
ding of a specific mode of action that arsenic has in
causing cancer.
6. ACKNOWLEDGMENTS
This research has been partially funded by the
Ministry of Science and Technological Development
of the Republic of Serbia, through Projects III 41006
and III 45009.
Authors wish to express their sincere gratitude
to Biomeasurment laboratory, Faculty of Agricultu-
re, Kobe University, Kobe, Japan where all
experiments were performed, especially dr Roumia-
na Tsenkova for mentoring work and BSc Yutaro
Tsuchisaka for technical and laboratory work assi-
stance.
7. REFERENCES
[1] S. Tapio, B. Grosche, Arsenic in the
aetiology of cancer, Mutation Research 612 (2006)
215246.
[2] WHO, WHO Guidelines for Drinking-
water Quality, Chapter 12, Chemical Fact Sheets,
WHO, Draft, Geneva, Switzerland, 1998, 306308
[3] B. K. Mandal, K..T. Suzuki, Arsenic
round the world: a review, Talanta 58 (2002)
201235.
[4] J. Mojsin (ed.), State of Health in Popu-
lation of the Banat Middle Region in 2007 Report
(in Serbian), Zrenjanin Institute for Public Health,
Zrenjanin, 2007.
[5] Janjic G, Extracellular/Intracellular and
Free/Captured Water in Human Organism, Water,
Hydrogen Bonding Nanomaterials, Nanomedicine,
The Book of Abstracts, 2010.
[6] R. Tsenkova, Aquaphotomics: dynamic
spectroscopy of aqueous and biological systems
describes peculiarities of water, Journal of Near
Infrared Spectroscopy 17 (2009) 303314.
[7] www.knjaz.co.rs
[8] C. O. Abernathy, D. J. Thomas, R. L.
Calderon, Health effects and Risk assessment of
Arsenic, The Journal of Nutrition, 133 (2003)
15361538.
[9] R. Roy, W. A. Tiller, I. Bell, M. R. Hoo-
ver, The Structure of Liquid Water; Novel Insights
from Material Research; Potential Relevance to
Homeopathy, Material Research Innovations Online,
94 (2005) 577608.
J. Munan, G. Janji, Influence of carcinogen compounds on hydrogen bonds in water.
Contemporary Materials, III1 (2012) Page 130 of 130
:
. -
,
. ,
,
- . ,
,
-
- a.
: , , , -
, .
