Journal of Molecular Liquids 307 (2020) 112961

Contents lists available at ScienceDirect

Journal of Molecular Liquids

journal homepage: www.elsevier.com/locate/molliq

Saliva as a first-line diagnostic tool: A spectral challenge for identification

of cancer biomarkers
Czesława Paluszkiewicz a, Ewa Pięta a,⁎, Monika Woźniak a, Natalia Piergies a, Anna Koniewska b,
Wojciech Ścierski b, Maciej Misiołek b, Wojciech M. Kwiatek a
a
Institute of Nuclear Physics Polish Academy of Sciences, PL-31342 Krakow, Poland
b
Medical University of Silesia, Department of Otorhinolaryngology and Oncological Laryngology, PL-41800 Zabrze, Poland

a r t i c l e i n f o a b s t r a c t

Article history: The paper concentrates on the role of saliva in the early diagnosis of salivary gland tumor. Due to the still not fully
Received 15 October 2019 understood etiology of head and neck cancer, the fundamental objective of this study is to find tumor markers of
Received in revised form 5 March 2020 salivary gland tumor progression. This work demonstrates the feasibility of Attenuated Total Reflection Fourier
Accepted 20 March 2020
Transform Infrared spectroscopy (ATR-FTIR) to track spectral variations between saliva samples derived from
Available online 21 March 2020
healthy volunteers and from salivary gland tumor (tumor mixus, TM) patients. Furthermore, a Real-Time Poly-
Keywords:
merase Chain Reaction (RT-PCR) has been used to detect a selected genes expression associated with neoplasm
Attenuated total reflection Fourier-transform changes.
infrared spectroscopy (ATR-FTIR) The obtained results imply that spectral signals attributed to the amide I/II (secondary structure of protein), car-
Real-time polymerase chain reaction (RT-PCR) bohydrates and inorganic phosphates oscillations are the most sensitive to alterations associated with the sali-
Saliva biomarkers vary gland cancer progression. Several spectroscopic biomarkers have been indicated as potential predictors of
Tumor mixus (TM) salivary gland tumor development. Additionally, the RT-PCR results reveal the increased level of Bcl-2 factor in
Salivary gland tumor salivary gland tumor patients' samples which may be related to inhibition of apoptosis process and increasing un-
controlled cell proliferation. Such combination of physico-chemical methods is a unique approach towards better
understanding the tumor's etiology and early diagnosis problem. Based on the promising findings presented in
this article, it could be concluded that saliva fluid has a great potential to be used as a first-line diagnostic tool
in patients with suspicion of salivary gland tumor.
© 2020 Elsevier B.V. All rights reserved.

1. Introduction Multiple authors [7–9] emphasize a significant diagnostic role of ul-

trasonographic investigations (USG) in patients with parotid tumors.
Neoplasm is still the most common disease of civilization and repre- The procedure allows for precise evaluation of lesions in the parotid su-
sents one of the main causes of the morbidity and mortality worldwide. perficial lobe while assessment of the deep lobe appears much more dif-
The effective treatment of neoplasm is associated with the proper ficult. The USG allows for tumor localization within the gland,
choice of therapy, but it also depends on the early diagnosis. One of evaluation of it size and preliminary histology. The literature often as-
the most difficult type of tumor, especially due to the problematic loca- sumes such procedure as a diagnostic standard [10].
tion is a head and neck neoplasm [1]. Histological evaluation of a tumor is possible due to the fine needle
The biomolecular achievements of the neoplastic process have aspiration (FNA) biopsy, which is commonly carried out on every pa-
helped in understanding of this tumor mechanism and an early detec- tient with a parotid tumor for the further diagnosis. The risk of the
tion greatly increases the chances for successful treatment [2]. A tumor spread in tissues passed through by the needle is low, however
searching for new biomarkers can contribute to early detection of this FNA is associated with the risk of multiple false negative results [2].
disease [3]. In our study, we put a special attention on the salivary For instance, in the case of small tumors some of the morphological
glands disorders. Salivary gland tumors comprise approximately 3 to changes can be diagnosis as benign lesions, where there is a malignant
10% of all head and neck neoplasms and the global incidence is esti- lesion or alterations may be completely invisible [11].
mated on 0.4 to 13.5 per 100,000 persons annually [4]. Most of these Moreover, FNA can lead to some other defects. Firstly, it provides a
neoplasms are mixed tumors and are localized in parotid glands [5,6]. lot of stress and patients discomfort. Secondly, there are often incom-
patibilities between pre- and postoperative histological examinations,
⁎ Corresponding author. which depend only on the pathologist's experience [12]. Finally, biopsy
E-mail address: [email protected] (E. Pięta). involves a high-cost and time-consuming procedure, therefore we

https://doi.org/10.1016/j.molliq.2020.112961
0167-7322/© 2020 Elsevier B.V. All rights reserved.
2 C. Paluszkiewicz et al. / Journal of Molecular Liquids 307 (2020) 112961

postulate to find a better solution for analysis, which can be used for morning. The patients and volunteers were required not to consume
non-invasive monitoring of tumors development. A good direction any meals or drinks 8 h prior to saliva collection and rinse their mouth
seems to be using biochemical substances contained in body fluids only by water. Then, about 15 ml of whole saliva and internal cheeks
such as saliva. The early diagnosis is the basis for effective treatment swab from the oral cavity were collected into the sterile tubes. Subse-
of head and neck tumor and what is more, a proper cure may facilitate quently, all materials were frozen at −20 °C until analysis time. Shortly
the patient's return to social activity and reduce the chance of the recur- before measurement, salivary samples were thawed and centrifuged to
rence. One of the non-invasive and rapid method related with salivary obtain the sediment for Real-time PCR analysis or liquid supernatant for
gland disorders is a detection of salivary contents. In the recent several spectroscopic measurement. In ATR-FTIR method, the salivary samples
years, there has been a growing interest in application of saliva to bio- were additionally dried on optical slides.
medical research [13]. This body fluid plays an important role as a factor
responses for maintaining homeostasis in the oral cavity. Except diges- 2.2. ATR-FTIR measurements
tive and protective functions, the salivary fluid contains a lot of signifi-
cant factors which may help to predict many of diseases progression. ATR-FTIR spectra of saliva samples were recorded using the Nicolet
The main components of saliva is a water, but otherwise, it contains iS5 spectrometer with a iD5 ZnSe ATR accessory (Thermo Scientific).
also numerous of proteins, lipids, carbohydrates, ions and vitamins The deuterated triglycine sulfate (DTGS) detector was applied to regis-
[14]. The most useful biomarkers contained in saliva for biomedical di- ter spectral signal in the range from 4000 cm−1 to 700 cm−1. An appro-
agnosis are mucins, enzymes, immunoglobulins, hormones and other priate measurement procedure was established before a series of
factors such as EGF - an epidermal growth factor which plays an impor- measurement. Namely, 20 μl of fresh saliva fluid was deposited onto
tant role in the process of carcinogenesis [15,16]. the ATR crystal and allowed to dry at room temperature. The number
It seems interesting to measure a number of proteins, lipids or en- of scans was 64, and the spectral resolution was 4 cm−1. The spectra
zymes contained in salivary fluid as biomarkers of tumors. Moreover, were recorded three times for each freshly prepared sample and were
the genetic material such as DNA can be detected in saliva as well averaged to yield a single spectrum per patient. The spectra were nor-
[17]. Therefore, the analysis of proteins structure and gene expression malized to the amide I band (~1647 cm−1). No smoothing procedure
may be a good way to predict the early neoplasm changes on the prote- was used to improve visual quality of any spectra presented.
omic and genetic level. Due to the rich content, saliva is functionally
equivalent to blood, but what is important, it is a non-infectious fluid 2.2.1. Data reduction and treatment schemes
in contrast to blood. Moreover, saliva has even more advantages such Graphical interpretation, data reduction of the recorded spectra
as long durability, high bioavailability and cost-effectiveness [18]. The were performed using OMNIC 9.0 software product. The curve-fitting
abovementioned biomarkers contained in saliva may provide an impor- analysis (a Gaussian band shape) and the second derivatives
tant information about many diseases including tumors, thus saliva (Savitzky-Golay, 7 points) were carried out using the OMNIC 9.0
seems to be very useful in reflecting the physiological state of the software.
body. Collecting of saliva is a non-invasive technique with easy storage,
simple procedures of preparation, and most importantly, comfortable
2.3. Real-time PCR study
for the patients.
The etiology of head and neck tumors remains still not fully under-
The level of Bcl-2 gene expression, an antiapoptotic factor which is
stood. Thus, the main target of this research was to find tumor markers
strongly related with proliferation process was measured in saliva sam-
of salivary gland tumor based on the analysis of conformational changes
ples of patients with salivary gland tumor (tumor mixus) and control
of proteins, lipids and DNA by FTIR spectroscopy, which has a number of
volunteers. The analyses were performed in stages as follow.
medical applications [19–30]. FTIR spectroscopy has the potential to
provide information about changes occurring during cancer develop-
2.3.1. RNA isolation
ment but also brings new challenges connected with implementation
The total RNA was extracted from fresh salivary samples in spin col-
of this method in the clinical environment [21]. Nonetheless, applica-
umns with lysis buffers by RNeasy Mini Kit (Qiagen, Netherlands) ac-
tion and development of this method for medical purposes may be cru-
cording to the manufacturer's instructions. The quantity (ng/μl) and
cial for prevention and treatment processes.
quality of RNA (1.8–2.0) was determined by Spark 10 M spectropho-
In this work, we discuss spectral alterations between saliva samples
tometer (Tecan, Switzerland) at 260 and 280 wavelength.
derived from healthy volunteers and from salivary gland tumor pa-
tients. Additionally, we used a real - time polymerase chain reaction
(RT-PCR), a biomolecular technique to detect a selected genes expres- 2.3.2. Reverse transcription
sion associated with neoplasm changes. Tumor cells may be dependent Next step was applied with Precision nanoScript kit (PrimerDesign,
on Bcl-2 gene, which is an antiapoptotic factor, in order to survive. In re- UK) in 0.2 ml PCR tubes with 50 ng of RNA template and other reagents.
sponse to stress signals, malignant cells may express pro-apoptotic acti- Each sample was submitted to the annealing step in 65 °C for 5 min in
vators, but usually cancer cells overexpress Bcl-2, which can decreases thermocycler (Bio-rad, USA) and then incubated 20 min in 42 °C and
this pro-apoptotic response. Therefore, the measurement of this impor- 10 min in 75 °C for extension stage.
tant factor seems to be important in the analysis of early diagnosis of
neoplastic changes. 2.3.3. Real-time PCR
The combination of proposed physico-chemical methods allowed to The final step was performed by Precision PLUS qPCR protocol
get an interdisciplinary view on the tumor's etiology and early diagnosis (PrimerDesign, UK) using Master MIX SYBR Green (10 μl), forward
problem. and reverse primers (per 0.25 μM), cDNA template (25 ng) and
RNAse/DNAse free water (up to 20 μl). The temperature-time condi-
2. Material and methods tions of the reaction were determined as follows: 2 min in 95 °C for en-
zyme activation, then denaturation in 95 °C for 10 s and 1 min in 60 °C
2.1. Saliva samples preparation for primers annealing.

The investigated group included five healthy volunteers (control 2.3.4. Analysis
group) and five patients diagnosed with salivary gland tumor (tumor The relative expression of Bcl-2 gene was quantified using the com-
mixus, TM). All of salivary samples were freshly obtained in the parative threshold cycle (ΔΔCT) after normalizing to reference gene
 C. Paluszkiewicz et al. / Journal of Molecular Liquids 307 (2020) 112961 3

(GAPDH) by 2-ΔΔCT Livak's method. All experiments were performed in stretching modes of proteins, while strong band at ~3286 cm−1 is attrib-
triplicates. uted to amide A N–H stretching [33]. Another spectral feature appearing
at ~3073 cm−1 is due to amide B ν(NH)/ν(CH) vibrations in proteins (α-
3. Results and discussion amylase, albumin, cystains, mucins, proline-rich proteins, sIgA) [33–35].
On the other hand, the spectral region 3000–2800 cm−1 is rich in bands
3.1. ATR-FTIR study ascribed to symmetric and asymmetric CH2 and CH3 stretching oscilla-
tions of lipids [33,36–38].
Fig. 1 shows the ATR-FTIR spectra of saliva sample: (a) control (spec- It has to be noted that the most evident alterations occur in the re-
trum averaged from five healthy volunteers samples), and salivary gion between ~900 and 1300 cm−1. Nonetheless, some of bands in the
gland tumor (tumor mixus, TM) patients: (b) M70TM1, (c) K39TM2, fingerprint region may overlap. This spectral interval is regarded as a
(d) K45TM3, (e) K48TM4, and (f) M40TM5 (M – man, W – woman). highly diagnostic region in studying various cancer types
As is evident from the spectra, there are significant changes in bands po- [33,35–44,49–57]. Band arising at ~1078 cm−1 is related to the asym-
sitions and relative intensities, which may be associated with the devel- metric and symmetric PO− −
2 stretching from symmetric PO2 stretching
opment of a tumor. Nevertheless, there are a lot of overlapping bands in from inorganic phosphates [58] and phosphate group of phospholipids
the fingerprint region, especially in the 900–1200 cm−1 and amide I and [48]. It was proved that this spectral feature is associated with the role
II spectral regions. Therefore, the curve-fitting analysis of these areas is of phosphates during diseases [50]. Thus, remarkable increase in
necessary for more conclusive band assignments (please see Fig. 2a – d). ~1074 cm−1 band intensity in the spectrum of TM in relation to the
Furthermore, for a better comparison of the spectral alterations be- spectrum of control group may be correlated with the disease develop-
tween control group and patients diagnosed with cancer, the average ment (Fig. 2e) [50,54,55]. This band is especially pronounced in the case
spectra were imposed on each other (Fig. 2e). Table 1 lists the most of TM spectra. Moreover, a noticeable displacement of the band position
characteristic ATR-FTIR bands together with suggested band assign- (from 1078 cm−1 in control group to 1074 cm−1 for patients with sali-
ments for averaged spectra of control (healthy volunteers) and salivary vary gland tumor) confirms the participation of phosphates upon the
gland tumor patients. The vibrational analysis was based on the latest investigated disease development [50]. Apart from the aforementioned
literature reports and databases on the application of FTIR spectroscopy bands, vibration come from phosphate group is manifested in the spec-
in biological studies with particular emphasis on saliva analysis [31–48]. tra at ~1159, 1239 and 985, and 936 cm−1 [58]. Heise and Marbarch [48]
The high frequency range is dominated by bands assigned to protein have thoroughly studied human oral mucosa tissue from several pa-
and lipid constituents [31,32]. Briefly, peaks observed at the spectral in- tients under changing blood glucose concentrations by ATR-FTIR spec-
terval between 3600 and 3000 cm−1 originate from C–H, O–H and N–H troscopy. Interestingly, the obtained FTIR spectra of saliva dry film and

(f) M40TM5

(e) W48TM4

(d) W45TM3
Absorbance

(c) W39TM2

(b) M70TM1

(a) control

Wavenumber/cm-1

Fig. 1. ATR-FTIR spectra of saliva samples: (a) control (spectrum averaged from five healthy volunteers samples), and salivary gland tumor (tumor mixus, TM) patients: (b) M70TM1,
(c) K39TM2, (d) K45TM3, (e) K48TM4, and (f) M40TM5. M – man, W – woman.
4 C. Paluszkiewicz et al. / Journal of Molecular Liquids 307 (2020) 112961

(e) control
TM
Absorbance

(c) amide I and II – control (d) amide I and II – TM

Wavenumber/cm-1
(a) 900 – 1200 cm-1: control (b) 900 – 1200 cm-1: TM

1641 1653
1648
1078

1631

1664
1074

1631
1038
1119
1040

Absorbance

1546
Absorbance

1680
Absorbance

Absorbance

1527
1677
1543
1550

1619

1513
1615
1657
1159

1577

1690
1515
1120

985
1021
1163

1692

Wavenumber/cm-1 Wavenumber/cm-1 Wavenumber/cm-1 Wavenumber/cm-1

Fig. 2. Curve-fitting analysis of the 900–1200 cm−1 spectral interval (a – control group, b – TM patients) and amide I/II with second derivative spectra (c – control group, d – TM patients)
profiles together with (e) averaged ATR-FTIR spectra of saliva samples (black line – control group, red line – TM patients). Spectra were averaged from five healthy volunteers and five
salivary gland tumor patients spectra, respectively.

saliva difference spectrum received after subtraction of a scaled phos- intermolecular hydrogen bond in α-helical and β-sheet structures
phate buffer spectrum revealed significant contribution of inorganic [64]. Furthermore, the most intense band appearing at 1648 cm−1 at-
phosphate features on the spectral patterns [48]. tributed to ν(C=O), (CN) and δ(NH) vibrations from α-helix upon
Furthermore, peaks observed within the 1000 to 1200 cm−1 interval deconvolution split into two additional bands at 1664 and 1641 cm−1
may also be attributed to the C–O stretching vibrations from carbohy- due to the disordered structure-solvated [νs(C=O)] and unordered
drates. Consequently, bands at ~1021, 1040 and 1078 cm−1 originate structure [ν(C=O)], respectively (Fig. 2c and d). Another observation
from sugar moieties [41,44,48]. Nevertheless, due to the fact that most concerns the content of α-helical conformation, which decreased no-
of the salivary proteins are glycosylated, these peaks can be attributed ticeably for TM patients (Fig. 2c and d; Table 1). This may be related to
to glycosylated α-amylase, mucins or other sugar residues vibrations the formation of β-sheet structure [65]. It is also worth noting that the
[44,48,59–61]. It is also worth pointing out that ~1119 cm−1 spectral relative intensity of 1631 cm−1 band due to the β-sheet structure
signal due to the ν(C–O) and ν(C–O–C) of carbohydrates vibrations in [ν(C=O)/ν(C=C)] significantly decrease upon cancer development.
TM spectrum exhibits significant enhancement in comparison to the Conversely, the composition of 1615 cm−1 spectral signal ascribed to
non-cancerous spectrum [28,33]. The discussed oscillations are the β-sheet formation [ν(C=C)] increased considerably for TM spectral
regarded as spectral biomarkers to distinguish between normal from data [33–39,44]. It has also be noted that peaks arising at ~1403 [νs
cancer states [62], while the 1119 cm−1 peak is considered as a spectro- (COO−), ρb(CH3)] and 1450 cm−1 [ρb(CH3)/δ(CH2/CH3)] exhibit higher
scopic marker of salivary gland tumor [53]. Most of these motions reveal intensity for the TM patients. Another prominent spectral features
noticeable strengthened upon disease development (Figs. 1 and 2e). assigned to proteins can be observed at 1543 [ρb(NH), ν(CN), amide
These results are consistent with other studies which have shown that II], 1515 [trosine ring, α-amylase, albumin, cystains, mucins, proline-
FTIR spectroscopy can be successfully employed to distinguish normal rich proteins, sIgA] and 1315 cm−1 [ν(CN), ρb(NH), amide III (α-amy-
from cancer states, especially considering the spectral interval between lase, albumin, cystains, mucins, proline-rich proteins, sIgA)]. Additional
800 and 1300 cm−1 [53,57,62,63]. FTIR signal, which is absent in the case of the control group spectrum,
There are also significant alterations in secondary structure of pro- appears at 1527 cm−1 [ρb(NH), ν(C=N), ν(C=C), amide II]. The above
teins (Fig. 1) between ATR-FTIR spectrum of control and salivary observations clearly indicates that the secondary structure of proteins
gland tumor patients. It must be noted that the α-helix maximum in healthy volunteers samples differ significantly from that in TM pa-
peak frequency decreased for the TM spectrum (1634–1640 cm−1; tients samples.
Fig. 1b-f) in comparison to the spectrum of control group (1644 cm−1;
Fig. 1a). Simultaneously, the β-sheet maximum band frequency 3.2. Real-time PCR study
(1543 cm−1; Fig. 1a) increased for the salivary gland tumor patients
(Fig. 1b-d), except in the case of M40TM5 (Fig. 1f) [33–39,44]. It is con- To determinate the anti-apoptotic gene (Bcl-2) expression, a total
sistent with literature and may results from variations in the extent of RNA extraction and then cDNA strand from all of salivary samples
 C. Paluszkiewicz et al. / Journal of Molecular Liquids 307 (2020) 112961 5

Table 1
Wavenumber and suggested band assignments for averaged ATR-FTIR spectra of control (healthy volunteers) and salivary gland tumor (tumor mixus, TM) patients.a

FTIR bands [cm−1] Band assignment Component identification Reference number

Control TM

3286 3282 νas(NH) Amide A [31–38]

3073 3074 ν(NH)/ν(CH) Amide B/ring (α-amylase, albumin, cystains, mucins, proline-rich proteins, sIgA) [31–38]
2962 2961 νas(CH3) Lipids, protein side chains [31–38]
2936 2935 νas(CH2) Lipids [31–38]
2873 2874 νs(CH3) Lipids, protein side chains [31–38]
2850 2851 νs(CH2) Lipids (cholesterol and mono/diglycerides of fatty acids) [31–38]
1692 1690 ν(C=O), ν(CN), ρb(NH) Amide I: anti-parallel β-sheet [33–39,44]
1680 1677 ν(C=O) Amide I: unordered random coils and turns [33–39,44]
1664 νs(C=O) Amide I: disordered structure-solvated [33–39,44]
1648 1653 ν(C=O), δ(CN), δ(NH) Amide I: α-helix [33–39,44]
1641 ν(C=O) Amide I: unordered structure [33–39,44]
1631 1631 ν(C=O)/ν(C=C) Amide I: β-sheet structure [33–39,44]
1615 1619 ν(C=C) Amide I: β-sheet structure [33–39,44]
1543 1546 ρb(NH), ν(CN) Amide II [33–39,44]
1527 ρb(NH), ν(C=N), ν(C=C) Amide II [33–39,44]
1515 1513 tyrosine ring Proteins (α-amylase, albumin, cystains, mucins, proline-rich proteins, sIgA) [33–39,44]
1451 1450 ρb(CH3)/δ(CH2/CH3) Proteins/lipids [33,36,38]
1407 1403 νs(COO−) Fatty acids [33,35,38,40]
ρb(CH3) Proteins [33,36,40]
1341 1343 ρw(CH2) Phospholipids, fatty acid, triglyceride, amino acid side chains [33]
1315 1315 ν(CN), ρb(NH) Amide III (α-amylase, albumin, cystains, mucins, proline-rich proteins, sIgA) [33,35,38,40]
1239 1243 νas(PO−2 ) Phosphate group of phospholipids [33,36,38,48]
νas(PO−2 ) Inorganic phosphates [28,33,36]
1163 1159 νas(PO−2 ) Inorganic phosphates [58]
ν(CO), ν(C–O–H) Serine, threonine, tyrosine of proteins [33,35,40]
1120 1119 ν(C–O), ν(C–O–C) Carbohydrates [28,33]
1078 1074 νs(PO−2 ) Inorganic phosphates [33,48,58]
νs(PO− −
2 ), νas(PO2 ) Phosphate group of phospholipids [33,35–44,48]
ν(CO), ρb(C–O–H) Glycosylated α-amylase, mucins or other sugar residues [41,44,48]
1040 1038 ν(CO), ρb(C–O–H) Glycosylated α-amylase, mucins or other sugar residues [41,44,48]
1021 ν(CO), δ(C–O–H) Glycosylated α-amylase, mucins or other sugar residues [41,44,48]
975 985 ν(PO−2 ) Inorganic phosphates [58]
924 936 ν(PO−2 ) Inorganic phosphates [58]
δ(COH), δ(COC) Carbohydrates [33,40]
νas(CH3–N) Phospholipids [33,40]
a
Abbreviations: ν, stretching; δ, deformation; ρb, bending; as, asymmetric; s, symmetric.

were used as a template for Real-time PCR detection. Our research, characteristic for tumors development by inhibiting of programmed
based on comparative ΔΔCt analysis, indicates increased level of Bcl-2 cells death (apoptosis).
expression in saliva of patients with salivary gland tumor (tumor
mixus). Fig. 3 shows several times higher relative expression (RQ) of 4. Conclusions
Bcl-2 factor in all of 5 patients samples in compare to the averaged re-
sults from 5 healthy volunteers samples (control). The most significant From the research that has been carried out it possible to conclude
changes were noticed in M70TM1 and K39TM2 cases (6-folds higher that saliva can be regarded as a first-line diagnostic tool in patients
expression than control samples), where in all of other samples with suspicion of salivary gland tumor. ATR-FTIR spectroscopy can be
(K45TM3, K48TM4, and M40TM5) the Bcl-2 gene level was increased successfully applied for distinguishing between control group and sali-
approximately 3 times. The presented results illustrate changes vary gland tumor patients. It has been demonstrated that there are sig-
nificant changes in secondary structure of proteins upon cancer
development. We highlighted spectroscopic biomarkers associated
Bcl-2 expression
with salivary gland tumor development, in particular bands attributed
to proteins, carbohydrates and inorganic phosphates.
Moreover, our biomolecular studies indicated the differences in anti-
apoptotic gene expression between patients with salivary gland tumor
[2^-delta delta Ct]

control (tumor mixus) and control volunteers. Increased level of Bcl-2 factor in
M70TM1 patients' samples may be related to inhibition of apoptosis process
W39TM2 and increasing uncontrolled cell proliferation, which is strongly related
RQ

to neoplastic development. Therefore, we confirmed that salivary fluid

W45TM3
can be very useful and significant tool for early diagnosis based on bio-
W48TM4 markers contained in saliva. To summarize, our research may be a first
M40TM5 step for searching modern solutions for early detection of oral tumors
based on salivary samples.

CRediT authorship contribution statement

Fig. 3. Relative expression of antiapoptotic factor Bcl-2 in saliva of patients with salivary
gland tumor vs control volunteers. One-way ANOVA analysis followed by the Newman-
Czesława Paluszkiewicz: Conceptualization, Methodology, Supervi-
Keuls post hoc test. Statistical significance was assessed using GraphPad Prism 5.0 sion. Ewa Pięta: Investigation, Writing - original draft, Writing - review
software and P value b0.005 was considered statistically significant. & editing, Visualization, Formal analysis. Monika Woźniak:
6 C. Paluszkiewicz et al. / Journal of Molecular Liquids 307 (2020) 112961

Investigation, Visualization, Writing - original draft. Natalia Piergies: and Raman spectroscopy: state of play and future challenges, Analyst 143 (2018)
1735–1757.
Investigation, Visualization. Anna Koniewska: Resources. Wojciech [22] J. Wang, M. Sowa, H.H. Mantsch, A. Bittner, H.M. Heise, Comparison of different in-
Ścierski: Resources, Funding acquisition, Investigation. Maciej frared measurement techniques in the clinical analysis of biofluids, TRAC-Trend.
Misiołek: Resources, Funding acquisition. Wojciech M. Kwiatek: Anal. Chem. 75 (1996) 286–296.
[23] J. Ollesch, S.L. Drees, H.M. Heise, T. Behrens, T. Bruning, K. Gerwert, FTIR spectros-
Supervision. copy of biofluids revisited: an automated approach to spectral biomarker identifica-
tion, Analyst 138 (2013) 4092–4102.
[24] A.L. Mitchell, K.B. Gajjar, G. Theophilou, F.L. Martin, P.L. Martin-Hirsch, Vibrational
spectroscopy of biofluids for disease screening or diagnosis: translation from the
Declaration of competing interest
laboratory to a clinical setting, J. Biophotonics 7 (2014) 153–165.
[25] C. Hughes, M. Brown, G. Clemens, A. Henderson, G. Monjardez, N.W. Clarke, P.
The authors have no conflict of interests to declare. Gardner, Assessing the challenges of Fourier transform infrared spectroscopic anal-
ysis of blood serum, J. Biophotonics 7 (2014) 180–188.
[26] F.L. Martin, J.G. Kelly, V. Llabjani, P.L. Martin-Hirsch, I.I. Patel, J. Trevisan, N.J.
Acknowledgments Fullwood, M.J. Walsh, Distinguishing cell types or populations based on the compu-
tational analysis of their infrared spectra, Nat. Protoc. 5 (2010) 1748–1760.
This study was partially supported by the statuary research of Med- [27] M.J. Baker, J. Trevisan, P. Bassan, R. Bhargava, H.J. Butler, K.M. Dorling, P.R. Fielden,
S.W. Fogarty, N.J. Fullwood, K.A. Heys, C. Hughes, P. Lasch, P.L. Martin-Hirsch, B.
ical University of Silesia in Katowice, Poland, project No. KNW-1-159/N/ Obinaju, G.D. Sockalingum, J. Sulé-Suso, R.J. Strong, M.J. Walsh, B.R. Wood, P.
8/K/2018. The study was performed using equipment purchased in the Gardner, F.L. Martin, Using Fourier transform IR spectroscopy to analyze biological
frame of the project co–funded by the Małopolska Regional Operational materials, Nat. Protoc. 9 (2014) 1771–1791.
[28] M.J. Baker, S.R. Hussain, L. Lovergne, V. Untereiner, C. Hughes, R.A. Lukaszewski, G.
Program Measure 5.1 Krakow Metropolitan Area as an important hub of Thiefin, G.D. Sockalingum, Developing and understanding biofluid vibrational spec-
the European Research Area for 2007–2013, project No. troscopy: a critical review, Chem. Soc. Rev. 45 (2016) 1803–1818.
MRPO.05.01.00–12–013/15. The authors acknowledge Ms. Jolanta [29] F. Bonnier, H. Blasco, C. Wasselet, G. Brachet, R. Respaud, L.F.C.S. Carvalho, D.
Bertrand, M.J. Baker, H.J. Byrne, I. Chourpa, Ultra-filtration of human serum for im-
Adamczyk for her invaluable assistance with the measurements.
proved quantitative analysis of low molecular weight biomarkers using ATR-IR
spectroscopy, Analyst 142 (2017) 1285–1298.
References [30] H.J. Butler, B.R. Smith, R. Fritzsch, P. Radhakrishnan, D.S. Palmer, M.J. Baker,
Optimised spectral pre-processing for discrimination of biofluids via ATR-FTIR spec-
[1] M. Guzzo, L.D. Locati, F.J. Prott, G. Gatta, M. McGurk, L. Licitra, Major and minor sal- troscopy, Analyst 143 (2018) 6121–6134.
ivary gland tumors, Crit. Rev. Oncol. Hematol. 74 (2010) 134–148. [31] C. Petibois, G. Cazorla, A. Cassaigne, A. Perromat, G. Deleris, Plasma protein contents
[2] K. Awan, S. Patil, S. Islam, M. Jafer, Early detection of oral cancer – guidelines for den- determined by Fourier-transform infrared spectrometry, Clinical Chem 47 (2001)
tal practitioners, J. Int. Oral Health 8 (2016) 399–403. 730–738.
[3] K. Dahiya, R. Dhankhar, Updated overview of current biomarkers in head and neck [32] G. Deleris, C. Petibois, Application of FT-IR spectrometry to plasma contents analysis
carcinoma, World J. Methodol. 6 (1) (2016) 77–86. and monitoring, Vib. Spectrosc. 32 (2003) 129–136.
[4] I. Schwentner, P. Obrist, W. Thumfart, G. Sprinzl, Distant metastasis of parotid gland [33] Z. Movasaghi, S. Rehman, I. ur Rehman, Fourier transform infrared (FTIR) spectros-
tumors, Acta Otolaryngol. 126 (2006) 340–345. copy of biological tissues, Appl. Spectrosc. Rev. 43 (2008) 134–179.
[5] N. Sadeghi, S. Al-Dhahri, J.J. Manoukian, Transnasal endoscopic medial maxillectomy [34] A. Barth, Infrared spectroscopy of proteins, Biochim. Biophys. Acta 1767 (2007)
for inverting papilloma, Laryngoscope 113 (2003) 749–753. 1073–1101.
[6] W.S. Kim, D.W. Hyun, C.H. Kim, J.H. Yoon, Treatment outcomes of sinonasal inverted [35] S. Abbas, N.S. Ozek, S. Emri, D. Koksal, M. Severcan, F. Severcan, Diagnosis of malig-
papillomas according to surgical approaches, Acta Otolaryngol. 130 (2010) nant pleural mesothelioma from pleural fluid by Fourier transform-infrared spec-
493–497. troscopy coupled with chemometrics, J. Biomed. Opt. 23 (2018), 105003.
[7] H. Alphs, D. Eisele, W. Westra, The role of fine needle aspiration in the evaluation of [36] G. Bellisola, C. Sorio, Infrared spectroscopy and microscopy in cancer research and
parotid masses, Curr Opin Otolaryngol Head Neck Surg (2) (2006) 62–66. diagnosis, Am. J. Cancer Res. 2 (2012) 1–21.
[8] E. Matsuda, T. Fukuhara, R. Donishi, K. Kawamoto, Y. Hirooka, H. Takeuchi, Useful- [37] R.R. Sultana, S.N. Zafarullah, N.H. Kirubamani, Saliva signature of normal pregnant
ness of a novel ultrasonographic classification based on anechoic area patterns for women in each trimester as analyzed by FTIR spectroscopy, Indian J. Sci. Technol.
differentiating Warthin Tumors from pleomorphic adenomas of the parotid gland, 4 (2011) 481–486.
Yonago Acta Med 60 (2018) 220–226. [38] P.C. Caetano Jr., J.F. Strixino, L. Raniero, Analysis of saliva by Fourier transform infra-
[9] L. Luczewski, P. Golusinski, J. Pazdrowski, P. Pienkowski, M. Kordylewska, W. red spectroscopy for diagnosis of physiological stress in athletes, Res. Biomed. Eng.
Golusinski, The ultrasound examination in assessment of parotid gland tumours: 31 (2015) 116–124.
the novel graphic diagram, Eur. Arch. Otorhinolaryngol. 270 (2013) 2129–2133. [39] L.M. Rodrigues, T.D. Magrini Alva, H. da Silva Martinho, J.D. Almeida, Analysis of sa-
[10] M. Sahin, I. Tatar, A. Kurt, et al., Importance of sonoelastography in assessing non- liva composition in patients with burning mouth syndrome (BMS) by FTIR spectros-
thyroid neck masses, Turk. Arch. Otorhinolaryngol. 55 (2017) 10–16. copy, Vib. Spectrosc. 100 (2019) 195–201.
[11] K. Gajjar, A. Ahmadzai, G. Valasoulis, J. Trevisan, C. Founta, M. Nasioutziki, A. [40] R.P.C.B. Rodrigues, E.M.G. Aguiar, L. Cardoso-Sousa, D.C. Caixeta, C.C.F.V. Guedes,
Loufopoulos, M. Kyrgiou, S. Stasinou, P. Karakitsos, E. Paraskevaidis, B. Da Gama- W.L. Siqueira, Y.C. Paiva Maia, S.V. Cardoso, R.N. Sabino-Silva, Differential molecular
Rose, P. Martin-Hirsch, F. Martin, Histology verification demonstrates that signature of human saliva using ATR-FTIR spectroscopy for chronic kidney disease
biospectroscopy analysis of cervical cytology identifies underlying disease more ac- diagnosis, Braz. Dent. J. 30 (2019) 437–445.
curately than conventional screening: removing the confounder of discordance, [41] L.V. Bel'skaya, E.A. Sarf, N.A. Makarova, Use of Fourier transform IR spectroscopy for
PLoS One 9 (2014), e82416-e82416. the study of saliva composition, J. Appl. Spectrosc. 85 (2018) 445–451.
[12] M. Jandu, K. Webster, The role of operator experience in the fine needle aspiration [42] L.V. Bel'skaya, E.A. Sarf, I.A. Gundyrev, Study of the IR spectra of the saliva of cancer
cytology of head and neck masses, Int. J. Oral Maxillofac. Surg. 28 (1999) 441–444. patients, composition, J. Appl. Spectrosc. 85 (2019) 1076–1084.
[13] C. Miller, J. Foley, A. Bailey, C. Campell, R. Humphries, N. Christodoulides, P. Floriano, [43] B.H. Stuart, Biological applications, Infrared Spectroscopy: Fundamentals and Appli-
G. Simmons, B. Bhagwandin, J. Jacobson, S. Redding, J. Ebersole, J. McDevitt, Current cations, John Wiley & Sons, Ltd 2005, pp. 137–165 , (ISBN 9780470011140).
developments in salivary diagnostics, Biomark. Med 4 (1) (2010) 171–189. [44] C.M. Orphanou, The detection and discrimination of human body fluids, using ATR
[14] T. Pfaffe, J. Cooper-White, P. Beyerlein, K. Kostner, C. Punyadeera, Diagnostic poten- FT-IR spectroscopy, Forensic Sci. Int. 252 (2015) e10–e16.
tial of saliva: current state and future applications, Clin. Chem. 57 (5) (2011) [45] D. Naumann, H. Fabian, P. Lasch, FTIR spectroscopy of cells, tissues and body fluids,
675–687. Adv. Biomed. Spectrosc. 2 (2009) 312–354.
[15] K.E. Kaczor-Urbanowicz, F. Wei, S.L. Rao, J. Kim, H. Shin, J. Cheng, M. Tu, D.T.W. [46] C.E. Christersson, L. Lindh, T. Arnebrant, Film-forming properties and viscosities of
Wong, Y. Kim, Clinical validity of saliva and novel technology for cancer detection, saliva substitutes and human whole saliva, Eur. J. Oral Sci. 108 (2000) 418–425.
BBA – Rev. Cancer 1872 (2019) 49–59. [47] C. Petibois, K. Gionnet, M. Goncalves, A. Perromat, M. Moenner, G. Deleris, Ana-
[16] X. Wang, K.E. Kaczor-Urbanowicz, D.T.W. Wong, Salivary biomarkers in cancer de- lytical performances of FT-IR spectrometry and imaging for concentration mea-
tection, Med. Oncol. 34 (2017) 1–12. surements within biological fluids, cells, and tissues, Analyst 131 (2006)
[17] R. Pink, J. Simek, J. Vondrakova, E. Faber, P. Michl, J. Pazdera, K. Indrak, Saliva as a di- 640–647.
agnostic medium, Biomed. Pap. Med. 153 (2) (2009) 103–110. [48] H.M. Heise, R. Marbach, Human oral mucosa studies with varying blood glucose
[18] Y. Lee, D. Wong, Saliva: an emerging biofluid for early detection of diseases, Am. J. concentration by non-invasive ATR-FT-IR-Spectroscopy, Cell. Mol. Biol. 44 (1998)
Dent. 22 (4) (2009) 241–248. 899–912.
[19] L. Lovergne, G. Clemens, V. Untereiner, R.A. Lukaszweski, G.D. Sockalingum, M.J. [49] S. Argov, J. Ramesh, A. Salman, J. Goldstein, I. Sinelnikov, H. Guterman, S. Mordechai,
Baker, Investigating optimum sample preparation for infrared spectroscopic Inflammatory bowel diseases as an intermediate stage between normal and cancer:
serum diagnostics, Anal. Meth. 7 (2015) 7140–7149. a FTIR-microspectroscopy approach, J. Biomed Optics 7 (2002) 1–7.
[20] J.M. Cameron, H.J. Butler, D.S. Palmer, M.J. Baker, Biofluid spectroscopic disease diag- [50] S. Argov, R.K. Sahu, E. Bernshtain, A. Salman, G. Shohat, U. Zelig, S. Mordechai,
nostics: a review on the processes and spectral impact of drying, J. Biophotonics 11 Inflamatory bowel diseases as an intermediate stage between normal and cancer:
(2018), e201700299. a FTIR-microspectroscopy approach, Biopolymers 75 (2004) 384–392.
[21] M.J. Baker, H.J. Byrne, J. Chalmers, P. Gardner, R. Goodacre, A. Henderson, S.G. [51] D.C. Malins, N.L. Polissar, S.J. Gunselman, Models of DNA structure achieve almost
Kazarian, F.L. Martin, J. Moger, N. Stone, J. Sule-Suso, Clinical applications of infrared perfect discrimination between normal prostate, benign prostatic hyperplasia
 C. Paluszkiewicz et al. / Journal of Molecular Liquids 307 (2020) 112961 7

(BPH), and adenocarcinoma and have a high potential for predicting BPH and [59] A. Borg, D. Birkhed, Secretion of glucose in human parotid saliva after carbohydrate
prostatecancer, Proc. Natl. Acad. Sci. U. S. A. 94 (1997) 259–264. intake, Scand. J. Dent. Res. 96 (1988) 551–556.
[52] D.C. Malins, P.M. Johnson, E.A. Barker, N.L. Polissar, T.M. Wheeler, K.M. Anderson, [60] M. Khajehpour, J.L. Dashnau, J.M. Vanderkooi, Infrared spectroscopy used to evalu-
Cancer-related changes in prostate DNA as men age and early identification of me- ate glycosylation of proteins, Anal. Biochem. 348 (2006) 40–48.
tastasis in primary prostate tumors, Proc. Natl. Acad. Sci. U. S. A. 100 (2003) [61] P.D.V. de Almeida, A.M.T. Grégio, M.Â.N. Machado, A.A.S. de Lima, L.R. Azevedo, Sa-
5401–5406. liva composition and functions: a comprehensive review, J. Contemp. Dent. Pract.
[53] C. Petibois, G. Deleris, Chemical mapping of tumor progression by FT-IR imaging: to- 9 (2008) 072–080.
wards molecular histopathology, Trends Biotechnol. 24 (2006) 455–462. [62] W. Yang, X. Xiao, J. Tan, Q. Cai, In situ evaluation of breast cancer cell growth with 3D
[54] E. Giorgini, P. Balercia, C. Conti, P. Ferraris, S. Sabbatini, C. Rubini, G. Tosi, Insights on ATR-FTIR spectroscopy, Vib. Spectrosc. 49 (2009) 64–67.
diagnosis of oral cavity pathologies by infrared spectroscopy: a review, J. Mol. Struct. [63] D.C. Malins, V.M. Green, T.M. Wheeler, E.A. Barker, M.A. Vinson, M. Sayeeduddin, K.E.
1051 (2013) 226–232. Hellstrom, K.M. Anderson, Metastatic cancer DNA phenotype identified in normal
[55] S. Sabbatini, C. Conti, C. Rubini, V. Librando, G. Tosi, E. Giorgini, Infrared tissues surrounding metastasizing prostate carcinomas, Proc. Natl. Acad. Sci. U. S.
microspectroscopy of oral squamous cell carcinoma: spectral signatures of cancer A. 101 (2004) 11428–11431.
grading, Vib. Spectrosc. 68 (2013) 196–203. [64] R. Eckel, H. Huo, H.W. Guan, X. Hu, X. Che, W.D. Huang, Characteristic infrared spec-
[56] E. Lipiec, J. Kowalska, J. Lekki, A. Wiecheć, W.M. Kwiatek, FTIR microspectroscopy in troscopic patterns in the protein bands of human breast cancer tissue, Vib.
studies of DNA damage induced by proton microbeam in single PC-3 cells, Acta Spectrosc. 27 (2001) 165–173.
Phys. Pol. A 121 (2011) 506–509. [65] Y. Kim, C.A. Rose, Y. Liu, Y. Ozaki, G. Datta, A.T. Tu, FTIR and near-infrared FT-Raman
[57] G.I. Dovbeshko, V.I. Chegel, N.Y. Gridina, O.P. Repnytska, Y.M. Shirshov, V.P. studies of the secondary structure of insulinotropin in the solid state: α-helix to β-
Tryndiak, I.M. Todor, G.I. Solyanik, Surface enhanced IR absorption of nucleic acids sheet conversion induced by phenol and/or high shear force, J. Pharm. Sci. 83 (1994)
from tumor cells: FTIR reflectance study, Biopolymers 67 (6) (2002) 470–486. 1175–1180.
[58] H.M. Heise, L. Cocchieri, T. Vahlsing, D. Ihrig, J. Elm, Monitoring of interstitial buffer
systems using micro-dialysis and infrared spectrometry, Proc. SPIE 10072 (2017),
100720E–1–100720E–14.