Analytical methods used for the characterisation of specific features of biological tissues related with obesity: A review

INTRODUCTION

Obesity is defined as an unhealthy excess of body fat which increases the risk of medical illness and premature mortality, or as a body mass index (BMI) ≥25 kg/m² and ≥30 kg/m² for the overweight and obesity thresholds, respectively [1–4]. The obesity prevalence and severity is growing dramatically in many developed and developing countries. Obesity affects >650  million people in the  world, according to reports from the World Health Organization. In the European Union (EU), the levels of obesity and overweight have been rising as well in the last decades [5–7]. Obesity has been clearly associated with numerous pathophysiologic processes and comorbidities, such as metabolic syndrome (MetS), type 2 diabetes mellitus (T2DM), cardiovascular disease (CVD) and most types of cancer which is the  most common cause of death in the  western world [8–10]. Most of the adult obese population will have fatty liver, and approximately one-third will develop non-alcoholic fatty liver disease [11].

There are three types of adipocytes: white, brown and brite (brown in white, also known as beige) adipocytes. Briefly, white adipose tissue (WAT) is the main tissue for storing energy. Under conditions of energy surplus, lipids are stored as triglycerides and can then be released as fuel under conditions of negative energy balance. Brown adipose tissue (BAT) is responsible for non‐shivering thermogenesis, essentially converting stored lipids into heat through fatty acid oxidation. The function of brite adipocytes is less well characterised, but they have a phenotype intermediate between white and brown adipocytes. They differ in their appearance, morphology, function, and gene expression profile [12, 13]. The adipose tissue can also be classified according to its location in the human body (see Fig. 1 [14]). Subcutaneous adipose tissue (SAT) is the fat just under the skin. Preperitoneal adipose tissue (PAT) is located prior peritoneum. Visceral adipose tissue (VAT) surrounds organs interfering the function of organs. Moreover, there is conflicting information about differences between the fatty acid composition of subcutaneous adipose tissue, preperitoneal adipose tissue and visceral adipose tissue. Varying proportions of fatty acids from adipose tissue may be related to specific diseases as well [14].

Adipose tissue is generally considered as a storage depot for excess energy, which is stored as triglycerides. It was long considered as a passive organ, but the  adipose tissue has been described as an endocrine organ recently with important physiological roles  [15]. Adipose tissue, which is primarily composed of adipocytes as well as pre-adipocytes, macrophages, endothelial cells, fibroblasts and leucocytes, has been increasingly recognised as a major player of systemically metabolic regulation [16, 17]. There is conflicting information about differences between the  fatty acid composition of SAT and VAT. Varying proportions of fatty acids from adipose tissue may be related to atherosclerosis, MetS, T2DM, CVD and other diseases and might exert a direct influence on serum lipids that may differ depending on the adipose tissue region [11, 18]. During cancer development, loss of total adipose tissue occurs in most cancer patients  [19]. However, obesity related with metabolic disturbances varies widely among obese individuals. For example, as opposed to the extent of SAT, the increase in VAT is associated with increased metabolic disturbances and CVD. An individual with a normal BMI but with an increased VAT is at higher risk of developing metabolic disturbances than an obese person with less VAT [20]. However, loss of total adipose tissue did not reflect changes in SAT and VAT in the same direction or intensity. The adipose tissue distribution during pregnancy showed a tendency towards a decreased accumulation of subcutaneous and increased accumulation of preperitoneal adipose tissue [21]. The relationship between obesity, namely PAT and arterial stiffness in adults and in children, has been also reported [22–24]. It was found that PAT volume was increased in people with chronic major depressive disorder versus those with acute major depressive disorder or controls [25]. The hypothesis that the ratio of visceral to subcutaneous adipose tissue is associated with altered sepsis outcome was recently partially proved  [26]. The  relationship between VAT and sugar-sweetened beverage intake was also determined [27]. Several studies showed the strongest correlation between the  symptoms of different diseases and changes in VAT mass  [28–31]. For example, Brown et al. [32] determined that VAT is correlated with a variety of biomarkers related to glucose homeostasis, inflammation, and lipid metabolism. Moreover, the quality and quantity of VAT play significant roles in adipocyte function and are related to insulin resistance  [33]. It was also concluded about the existence of a strong correlation between glucose production and insulin sensitivity in SAT [34].

Although the development of obesity is easily attributed to excess intake of calories, the underlying reasons for the metabolic disturbances and health risks associated with obesity are still unclear. Though an association between obesity and health problems is well established, the link mayin reality-not be as simple. In fact, findings from experimental animal models and human studies collectively suggest that the  increased morbidity in obesity is mainly due to increased levels of circulating lipids resulting in ectopic lipid deposition and lipotoxicity, which in turn negatively impacts metabolism and tissue function, suggesting dyslipidemia as the  primary inducer of disease development [35–38]. To determine the link between obesity and metabolic syndromes, researchers are studying a  wide variety of adipose tissue features, changes of different parameters, and various physical properties in obese people; examine the  chemical composition of adipose tissue, the  change in composition depending on the progression of the disease and treatment [39–59]. The  abundance of various parameters is set so that it is not easy to see and capture the desired connection and it is not always possible. Not only physicians but also chemists, biochemists, physicists and materials scientists are actively involved in these studies. The following main methods are currently used to characterise adipose tissue: gas and liquid chromatography, FTIR, Raman and NMR spectroscopies, scanning electron microscopy (SEM), elemental analysis using sensors and other techniques. The main aim of this review is to introduce the scientific community with the possibilities of using these latest methods of instrumental analysis in the study of the peculiarities of human adipose tissue.

GAS AND LIQUID CHROMATOGRAPHY

The conventional methods for determining the composition of fats in the adipose tissue are gas and liquid chromatography [60–63]. Walters et al. [64] stated that the recent obesity epidemic in children also showed an increase in the prevalence of hypertension. As blood pressure (BP) is associated with (long-chain) polyunsaturated fatty acids (LC PUFA), the authors investigated the direct and indirect effects on the PUFA level, body mass index (BMI) and BP. Whole blood fatty acids were measured by a validated gas chromatographic method and recorded as percentage of weight of all fatty acids detected. Gas chromatographytandem mass spectrometry (GC-MS) was used for the determination of 209 polychlorinated biphenyl in the human adipose tissue [65]. For decades, these compounds are extensively used in industry, are very biostable, lipophilic and resistant to both chemical and biological degradation that tend to accumulate in different tissues of organisms. GCMS was also used to analyse the isomers of monounsaturated fatty acid having sixteen carbon atoms (16:1) in human adipose tissue (see Fig. 2) [66]. It was reported that human phagocytic cells also expressed low levels of a third 16:1 isomer, sapienic acid  [16:1n-10 (cis-6-hexadecenoic acid)]. Despite the common assumption that 16:1n-10 is restricted to humans, the authors found relatively high levels of this fatty acid in murine macrophages and also in murine macrophage-like cell lines. Poskus et al.  [67, 68] successfully applied a  silica-based silver ion solid-phase extraction (Ag⁺-SPE) for the  fractionation and subsequent gas chromatographic analysis of trans fatty acids in human adipose. In the work [69], the samples of adipose tissue and serum were obtained from 386 subjects undergoing non-cancer-related surgery and were extracted using validated methodologies. Residues of persistent organic pollutants were analysed by means of high-resolution gas chromatography with a mass spectrometry detector in a  tandem mode. The  possible association of concentrations of persistent organic pollutants in the human body with the prevalence of type 2 diabetes was determined in samples of adults from Southern Spain.

In the work [70], the total fatty acids extracted from red blood cells and arachidonic acid-to-eicosapentaenoic acid (AA-to-EPA) ratios were assessed using GC-MS assay in a  single-ion monitoring mode. The study was conducted on 64 male rats. It was concluded that anti-inflammatory drugs (e.g. celecoxib) and antioxidants (e.g. safranal) appeared as promising pharmaceutical and nutritional alternatives, respectively, that can prevent or reduce lipotoxicity-associated metabolic disturbances. The gas chromatography was also used to study the quality (oxidative stability), fatty acids profile and their distribution in triacylglycerol (TAG) molecules of fat obtained from the adipose tissue of wild boar, badger and wild goose [71]. The determination of fatty acids in milk using a new thin layer chromatography and gas chromatographic (TLC-GC) techniques was performed [72]. The results obtained confirmed the applicability of TLC-GC as a new method, not requiring blood sampling, that could be incorporated into the daily milking routine. The applicability of this method could allow the implementation of nutritional and management strategies in order to avoid the development of metabolic diseases.

Liquid chromatography techniques are also very important in lipidomics research [73]. For instance, antibiotic levels in the adipose samples were determined with reverse phase high-pressure liquid chromatography (HPLC) [74]. The aim of this study was to evaluate the concentration of cefazolin in adipose tissue of patients undergoing bariatric surgery. An inverse correlation between BMI and concentration of cefazolin in adipose tissue was observed. Bartosinska et al. [75] developed the HPLC method for the  quantitative determination of tocotrienols in human breast adipose tissue (Fig. 3). In the  work  [76], ultra-performance liquid chromatography (UPLC) coupled with electrospray ionisation (ESI) mass spectrometry (MS) were employed for the comprehensive qualification and quantification of triacylglycerols (TGs) with a specific fatty acid chain composition in human plasma and liver tissues including hepatocellular carcinoma and para-carcinoma tissues. Multiple MS detection modes from QTRAP MS and FT-ICR MS were utilised, and hundreds of TG species (including many oxidised TG species) with their specific fatty acid chain compositions have been qualified and quantified. It was concluded that the  developed analytical procedure is of significance for an accurate analysis of these lipids. Recently, liquid chromatography approaches were used to evaluate different obesity models by determining different analytes in adipose tissue of rats [77, 78]. The main information regarding application of gas and liquid chromatography for the analytical characterisation of biological tissues related with obesity is summarised in Table 1.

FTIR AND RAMAN SPECTROSCOPIES

Buckus et al.  [79] demonstrated that FTIR spectroscopy could be used for the analysis and direct qualitative characterisation of adipose tissue composition. According to the results of FTIR analysis it was concluded that the main functional groupings in the adipose tissue of different obese patients were the same; however, the real chemical composition of the samples obtained from subcutaneous, preperitoneal and visceral layers of adipose tissue from different patients was different. The representative spectra recorded for the different specimens are shown in Fig. 4. The FTIR spectra of adipose samples obtained from different patients and different tissues contain broad and sharp bands located at 2950–2800  cm^–1 as well as weaker bands centred at around 3150, 1350 and 900 cm^–1. These bands correspond to the n(C-H) stretching vibration of methylene and alkyl groups [79]. The sharp and intensive absorption band located at 1743– 1710 cm^–1 is attributed to the stretching vibrations of C=O in a triacylglycerol functional group [79]. FTIR spectra of all samples contain the same bands located at 1680–1600  cm^–1 and 1200–1000  cm^–1 which correspond to C=C and CO-O-C vibrations, respectively. These observations let us to state that unsaturated bonds in the  fatty acids and esters are present in the investigated samples. The broad bands centred at ~800–700  cm^–1 correspond to the –CH₂-CH₂– vibrations in long alkyl chains. Interestingly, the intensity of these bands is quite different in the spectra for different adipose samples. This only one slightly different observation is visible in all FTIR spectra confirming the presence of different fatty acids in the adipose tissues.

The similar conclusions were made by other authors as well. The FTIR results of the studies [80, 81] demonstrated that there were several alterations in the triglyceride spectral region of SAT and VAT of male obese mice. The authors [82] have used FTIR spectroscopy to investigate the composition of fatty acid of adipose tissue in men having the metabolic syndrome. The results from the patients with uronephrolithiasis differed from the  results from the  control group. Irrespective of MS diagnosis, all individuals with kidney stones had significantly higher percentages of monounsaturated fatty acids and lower percentages of polyunsaturated fatty acids than did healthy individuals. A very interesting observation was demonstrated by Ami et al. [83]. They showed that ATR-FTIR spectroscopy, coupled with multivariate analysis, can differentiate adipose tissue aspirates containing amyloid deposits from unaffected samples with a high sensitivity and specificity. Moreover, this approach was rapid and required no sample processing besides blood removal.

ATR-FTIR combined with chemometrics were used for the estimation of postmortem interval in the adipose tissue of mice [84]. This study showed that adipose tissue is a  promising option for estimating later stages in the postmortem interval but temperature has a great influence on the postmortem changes in adipose tissue. The  importance to use the cryostat technique to make thin sections of biological samples for FTIR microspectroscopy imaging was also shown by Liyanage et al. [85]. It was demonstrated that FTIR spectroscopy is a very useful tool in the development of alternative tissues for liver transplantation. In the procedure of the preparation of porcine livers, FTIR spectroscopy was used to control the presence of nuclear material and the detergent residue in the samples could be used for liver transplantation  [86]. FT-IR spectroscopy proved to be an accurate technique for detecting and quantifying the amount of lard taken from the adipose tissue of pigs and mixed with sunflower, canola, coconut, olive, and mustard oil. The  results were comparable to the  concentration of lard in binary mixtures with other oil. The correlation coefficient of 0.9577 was produced in the wavenumber region of 1078.01–1246.75 cm^–1 [87].

It was shown that Raman spectroscopy can be used as a  simple, fast and non-destructive method for the quantitative analysis of Omega-3 fatty acids in fish oil supplements [88, 89]. Raman spectroscopy is one of the vibrational spectroscopies. One of the great advantages of Raman spectroscopy is that it is less invasive; a sample can be studied in situ using laser light without pre-treatment [89]. These advantages make Raman spectroscopy possible for large-scale detection and analysis for fat content and fatty acids [90].

Raman spectroscopy for the  characterisation of adipose tissue samples obtained from human adipose tissue for the first time was used by Yanina et al. [91]. The enzymatic destruction of adipose tissue by encapsulation of lipase into the  polymeric microcapsules has been investigated. Raman spectra of adipose tissue before and after lipase treatment were acquired 100 and 40 times, respectively, from different spatial points of the tissue. Figure 5 shows the normalised averaged spectra of both tissue samples. The  development of Raman imaging probes for intraoperative cancer surgery guidance, the approach used in Refs. [92, 93], had a great potential to improve minimally invasive procedure performance with an endoscope or surgical robots. Both of those technologies are currently using an imaging bundle to recreate white-light images of the  tissue from which the surgeon guided himself through the tissue. The confocal Raman microspectroscopy was applied for the simultaneous visualisation, identification and quantification of the lipid accumulation in lipid droplets of aggressive human breast cancer cells derived from overweight and obese patients [94]. It was demonstrated that this technique enables imaging of cell compartments as well as quantification and monitoring of specific biomolecules and metabolic processes on a single cell level. Moreover, a significant increase in unsaturation, esterification and the lipid to protein ratio in lipid droplets in breast cancer cells incubated was identified using a Raman microspectrosopy method. The  same method was successfully applied to evaluate the effect of hyperglycemia on proliferating, differentiating and maturating human visceral pre/adipocytes [95].

The Raman spectroscopy was also used to investigate molecular compositions of lipid droplets changing in live hepatic cells stimulated with major fatty acids in the human body, i.e. palmitic, stearic, oleic and linoleic acids [96]. Majka et al. [97] developed a novel strategy of studying perivascular adipose tissue in human vessels with the application of fiberoptic Raman spectroscopy and spectral modelling. The study showed the results of the investigation of a group of 10 male patients of age 57–73 with advanced atherosclerosis. All patients were diagnosed with coronary artery disease and were subjected to coronary artery bypass surgery. All patients showed hypercholesterolemia and hypertension, and some of them suffered from other co-morbidities; in particular, 6 had confirmed type 2 diabetes mellitus. It was confirmed that the proposed method for the reliable and rapid evaluation of the phenotype of perivascular adipose tissue in the future can be adopted intraoperatively in hospital conditions.

The study  [98–101] demonstrated the  feasibility of using fiber-optic Raman spectroscopy in the quantitative analysis of specific fats in the body in a  totally non-invasive manner. Golden Syrian hamsters, different rat groups and pork were used as adipose tissue sources. Confocal Raman microscopy was used to record horse fat depth profiles measuring the  unsaturation degree  [102]. It was noted that the presence of proteins in the studied horse fat was found in the first layer of the tissue, that is, very close to the skin. The Raman spectroscopy was used to detect changes in adipose tissue associated with a  prolonged intake of desloratadine, a commonly used second generation antihistamine to treat allergic reactions [103]. White and brown adipose tissue samples were excised from adult rats. The Raman spectra indicated that antihistamine use reduces the protein-to-lipid ratio in the brown adipose tissue but not the white adipose tissue, indicating that the effect on adipose tissue is location-dependent.

It was recently showed that Raman spectroscopy results provided new insights into the  genetics of fatty acid traits in rainbow trout [104]. Many biomedical applications require measurements of Raman spectra of tissue under ambient lighting conditions. However, the background light often swamps the  weaker Raman signal. The  use of time-gated (TG) Raman spectroscopy based on a single photon avalanche diode (SPAD) operating in time-correlated single photon counting and near-infrared laser excitation was investigated for the acquisition of Raman spectra and spectral images of biological tissue in study [105]. Advances of Raman spectroscopy investigating metabolic syndrome was also observed in the review article [106]. Thus, Raman spectroscopy is a powerful analytical technique for the analysis of tissue and has been widely used for medical applications. Several Raman devices have also been tested in hospitals for improving diagnosis. Confocal Raman microscopy analyses might function as a  valuable diagnostic tool to identify metabolic alterations in biological samples which in turn could provide more detailed insights in the pathogenesis of breast cancer in association with obesity. The main information regarding the application of FTIR and Raman spectroscopies for the analytical characterisation of biological tissues related with obesity is summarised in Table 2.

NMR SPECTROSCOPY

The ¹H NMR spectroscopy analysis could also provide metabolic information indicating disturbed glucose metabolism, fatty acid metabolism, amino acid metabolism and activated inflammatory response in patients [107]. However, the ¹H NMR spectra of adipose tissue samples taken from different obese patients were rather similar  [79]. Thus, according to these results, the  disease symptoms observed for these patients might be also similar. However, the authors also identified some characteristic features in each ¹H NMR spectrum. Based on the integral values of the ¹H NMR spectra the composition of adipose tissue was estimated on two classes of fatty acids: unsaturated and saturated. It was suggested that patients with metabolic disorder had a  lower level of polyunsaturated and a high level of monounsaturated fatty acids in the adipose tissue.

A simple method to detect gluconeogenesis from glycerol in obese humans was developed [108]. For the first time in obese humans, a simple stable isotope technique to investigate in vivo the mechanisms underlying the effects of excess visceral adiposity on gluconeogenesis from glycerol was used. Glycerol-gluconeogenesis was evaluated by determining the fraction of ¹³C enrichment in blood glucose using the  NMR spectroscopic quantification of ¹³Clabelled glucose isotopomers. The findings provided a preliminary evidence that excess visceral fat disrupts multiple pathways in hepatic gluconeogenesis from glycerol. The differences in T2 relaxation times between the CH₂, CH₃ (non-omega-3) and n-3 CH₃ (omega-3) signals enabled the identification of wellresolved n-3 CH3 resonances, making the  developed method suitable for the direct detection of n-3 fatty acids at 7 T [109]. This work proposed a relatively simple approach for the reliable estimation of the n-3 FA fraction, in vivo, in a samples of healthy volunteers.

It was shown that the common shared lipid accumulation mechanism in lipoma is a reduction in lipolysis [110]. 15 patients (12 males and 3 females; age range 26–79 years; normal BMI) with single superficial subcutaneous lipomas were recruited. ¹H NMR spectra were corrected for baseline and phase distortions and calibrated using the residual proton resonance of methanol at 3.49 ppm. The lipid composition of the samples was determined by integrating the methyl (0.9 ppm), allylic (2.7 ppm), olefinic (5.3 ppm) and vicinal-olefinic (2.0 ppm) resonances. Superficial lipomas could thus be used as a model for lipid accumulation through altered lipolysis as found in obese patients. It is well known that the synthesis and turnover of triglyceride in adipose tissue involves enzymes with preferences for specific fatty acid classes and/or regioselectivity regarding the fatty acid position within the glycerol moiety. The focus of the study [111] was to characterise both the composition of fatty acids and their positional distribution in triglycerides of biopsied human subcutaneous adipose tissue, from subjects with wide ranges of the body mass index (BMI) and insulin sensitivity, using ¹H and ¹³C NMR spectroscopy. Chloroform extracts of the adipose tissue biopsies yielded ¹³C and ¹H NMR spectra with a good resolution of triglycerides fatty acid hydrogens and carbons, as shown by Fig. 6. The examination of the fatty acid ¹³C NMR signals from both human adipose tissue biopsies, as well as the triglycerides standard mixture, revealed that the ω-3 signals, whose chemical shifts are centered at 31.9 ppm, provided a superior resolution of the most abundant fatty acid species compared with the  ω-2 signals that were used previously. The ¹³C NMR spectra of individual glyceryl trimyristate, glyceryl tripalmitate and glyceryl tristearate standards showed identical chemical shift values for the ω-3 position. Returning to the human study cohort, the  data revealed that subtle alterations in the  distribution of MUFA and PUFA between the sn1,3 and sn2 sites of glycerol were associated with BMI, independent of their total abundance. These data were consistent with differences in fatty acid selectivity by one or more enzymes involved in triglycerides metabolism and/or different triglycerides-fatty acid cycling activities in the setting of obesity.

The variation in the metabolic profile of breast cancer focusing on lipids as triglycerides and free fatty acids that may alter in malignant breast tissues and lymph nodes from adjacent benign breast tissues was evaluated by ¹H NMR spectroscopy [112]. The ¹H NMR spectra were recorded on 173 tissue specimens comprising of breast tumour tissues, adjacent tissues, few lymph nodes and overlying skin tissues obtained from 67 patients suffering from breast cancer. Reduction in the  lipid content was observed in malignant breast tissues along with a higher fraction of free fatty acids. This study may help in future for a precise evaluation of the lipid classification including small molecules as a source of early diagnosis of invasive ductal carcinoma by employing in  vivo magnetic resonance spectroscopic methods.

Non-invasive ¹H NMR spectroscopy was used to monitor the hepatic lipid content and fatty acid composition in vivo [113]. Male (N = 10) and female (N = 10) mice were used for the animal experiments. The  authors detected higher amounts of omega‐3 in total liver lipids of female than male mice. Phenelzine-treated mice exhibited a  lower body fat content, subcutaneous adipose tissue mass and lipid content in skeletal muscles than the control, without the  decreased body weight gain or food consumption [114]. Mice were fed a standard chow and given phenelzine in drinking water for 12 weeks. The body composition was determined by NMR spectroscopy. ¹H, ¹³C and ³¹P NMR spectroscopies were used to investigate different health problems related with obesity performing animal experiments with male rats [115] and adipose tissue of mouse [116, 117]. High resolution magic angle spinning NMR spectroscopy was successfully used for the  qualitative study of the  highly specialised lipid tissues of cetaceans [118]. The HR-MAS NMR spectra of the representative samples are presented in Fig. 7. Assignment of the major peaks detected by HR-MAS NMR in the  blubber and melon of the harbour porpoise and long-finned pilot whale, and calculations used to determine peak area are presented in Table 3.

A new NMR-based method was developed, which allows, in addition to the determination of the relative fatty acyl composition, the assignment of unsaturated fatty acyl residues to the positions of the glycerol backbone [119]. Standard ¹H NMR as well as ¹³C NMR spectra were recorded from the solutions of pure triacylglycerols of known compositions and different vegetable oils (olive, garlic, sesame, corn, sunflower, walnut and linseed). However, it was concluded that this approach is not a suitable method for the trace analysis.

It was investigated how rapid weight loss due to a very low calorie diet affects the fecal microbiota and fecal bile acid composition  [120]. To support the explanation of obtained results, the urine metabolic phenotypes were measured by ¹H NMR spectroscopy. The investigations of the interaction between microbiota and adipose tissue physiology in humans occur very intensively. And ¹H NMR spectroscopy plays a very important role in these investigations [121, 122]. The main information regarding the application of NMR spectroscopy for the analytical characterisation of biological tissues related with obesity is summarised in Table 4.

ELEMENTAL ANALYSIS

Human body lipids serve as a harbour for various organic substances, they may absorb different metabolic products, and they have different calorific capacity depending on their location and forms. It was demonstrated that atherosclerotic plaque contains not only organic but also inorganic elements [123]. For the determination of sodium and calcium ions, atomic adsorptive spectrometry (AAS) was used. Figure 8 shows that the content of Ca is relatively higher in AP (loose) than AP (dense), and the ratio is a reverse for the Na content. For the determination of the amount of different metals in the adipose samples (Na, K, Mg, Ca, Cr, Mn, Fe, Cu, Zn and Ni) the  AAS analysis method was also applied  [79]. The  results for the  determination of selected elements in different layers of adipose tissue obtained from different patients are summarised in Table 5. Evidently, the concentration of sodium and potassium in adipose is much higher in comparison with other elements. The concentration of magnesium is higher in samples 1 and 3, the amount of calcium is very high in sample 1 (267 mg/g). On the other hand, a rather low concentration of calcium was determined in sample 2 (33 μg/g). Evidently, sample 4 contains a higher amount of copper. Interestingly, chromium was found only in the adipose of patients 2 (in all A, B and C layers) and 3 (only in layer A). The concentrations of Fe and Zn do not vary significantly in the  adipose tissue samples from different patients. Moreover, Mn and Ni were not detected in all analysed samples of adipose tissue. The  relative standard deviation (RSD) values obtained for the determination of metals in the adipose from the obese patients (6.4–10.3%) indicate a high degree of homogeneity, which could be expected for adipose samples. Moreover, the  values obtained are not unusual for such type of analysis and can be considered as suitable for routine analysis. No doubt, the results obtained show various distributions of different metals in the adipose tissue of patients with a  different metabolical state. According to the  results summarised in this study, the distribution of metal levels in adipose tissue layers is chaotic and does not serve very important information.

The elemental composition in 38 breast tissue samples (19 normal and 19 diseased) was analysed using polarized energy-dispersive X‐ray fluorescence (EDX) spectroscopy  [124]. The  measured elemental concentrations were used in the  principal component analysis study to determine the  variables that produced the  most differentiation between the normal and diseased tissues. The  results are summarised in the  EDX spectra presented in Fig. 9. Large differences can be seen in the  K, Fe and Zn peaks, with small apparent differences in the Ca and Cu peaks. The Ni and Ti peaks come from contaminants in the measuring system. The  adipose tissue from two groups of the  patients with obesity (with and without metabolic syndrome) was investigated by means of inductively coupled plasma optical emission spectrometry (ICP-OES) [125]. The main aim of this study was to reveal some differences and regularities in the  concentrations of trace elements in the adipose tissue between these two groups of the individuals. Moreover, different types of adipose tissue (subcutaneous, preperitoneal and visceral) were analysed separately in order to investigate distribution of metals between these types in different groups of obese people. Al, Ba, Ca, Co, Cu, Cr, Fe, K, Li, Mg, Mn, Na, Ni, Sr and Zn were selected for the quantitative determination. However, only 6 elements (Na, K, Ca, Fe, Mg and Zn) were determined in all analysed samples independently of the type of adipose tissue and the presence of metabolic disorder of the  patient. Concentrations of Ca, Fe, K and Na were found to be strongly dependent on the type of adipose tissue. Other elements in terms of detection frequency in the  adipose tissue samples were arranged in the  following sequence: Sr (94.7%), Cr (86.2%), Cu (24.6%), Li (18%), Ba (1.8%) and Co (0.05%). The concentrations of Al, Mn and Ni were found to be lower than the limit of detection (LOD) in all analysed samples. In general, there was no obvious difference in the concentrations of metals in the adipose tissue of the patients with and without the metabolic syndrome.

It was predicted that a  redox disproportionation develops in some cancers, leading to pools of both more reduced and more oxidised macromolecules compared to normal conditions [126]. This conclusion was made by analysing the average oxidation state of carbon and water demand per residue, calculated using elemental abundances and stoichiometric reactions to form proteins from basis species. The  authors  [127] demonstrated the ability to track 1% calcium variations between adipose and breast surrogates and 2% oxygen variations between the several tissue surrogates, and a logarithmic trend between the elemental concentration of the target and the prompt gamma production induced by the irradiation of such elements. The absolute measurements of total-body calcium using prompt gamma have been reported in renal patients. It was demonstrated that proton and ion beam spectroscopy is sensitive enough for the  analysis of calcification in metastases. The iodine value for the Italian heavy pig subcutaneous adipose tissue from fatty acid methyl ester profiles was calculated. The specific equations for the  calculation of lipids were developed [128]. Since only few studies of chemical analysis in the human adipose tissue are reported, and these results obtained are scattered, it is impossible to draw any general conclusions.

SCANNING ELECTRON MICROSCOPY (SEM)

The morphology by scanning electron microscopy (SEM) of the adipose samples was investigated in several studies [14, 79, 129, 130]. The adipose tissue was taken from volunteer obese patients from subcutaneous, preperitoneal and visceral layers of adipose tissue (Fig. 1). For example, the SEM micrographs of subcutaneous adipose tissue layer taken from obese patients having different metabolic diseases are presented in Fig. 10. These SEM images are slightly different in comparison with the ones taken from the SAT layers of people without metabolic diseases. It can be seen from Fig. 4 that the adipose tissue samples are composed of cubes, prisms and spherically shaped granules particles less than 0.5  μm in size. These different geometric shapes are connected by solid matrix. Moreover, pores and voids can also be seen, which result probably from the different natural condition of adipose tissue. The interesting morphological feature is rarely observed bright spots in all SEM micrographs of the  SAT adipose tissue picked up from different patients. These bright spots probably are associated with metabolic changes in the body of obese patients. However, only few samples contain rodand/or stick-like individual particles 0.5–2.0 μm in size which were seen in the SEM images of the SAT layers of healthy volunteers.

The distribution of different morphological features in different layers of the lyophilised adipose tissue was also investigated [14]. The typical SEM micrographs of the  adipose tissue samples obtained from the  preperitoneal (PAT), subcutaneous (SAT) and visceral (VAT) layers of obese patients with different metabolic diseases are shown in Fig.  11. The  obtained microstructural results clearly informed that the calculated specific area of thickening and scarring of the connective tissue for the obese patients with different metabolic diseases was much larger than for obese and healthy people (Fig. 12). Besides, three different techniques were tested to construct the 3D view from the SEM images of adipose tissue [130]. The 3D reconstructed model by VisualSFM and MeshLab free available software gave the  best and the  most promising results. The  surface was fully reconstructed and the pixel z dimension might be easily extrapolated (see Fig. 13). The reconstructed view also might be easily printed by a 3D printer which gives a new area of applications.

To study the microenvironment of bone marrow adipocyte a three-dimensional analysis of the ultrastructure with focused ion beam scanning electron microscopy (FIB-SEM) was also performed [131]. These images provided a unique opportunity to visualise the native bone marrow adipocyte within its skeletal niche and strongly implied that bone marrow adipocytes actively participate in physiologic energy partitioning and signalling within the skeletal microenvironments. The SEM results could provide an important information about the structural composition of adipose tissue layers in the human body, as well as the main microstructural features; however, the  sample preparation for SEM measurements should be performed with a special care. Recently it was demonstrated that a simple method of  thawing cryo-stored samples preserves ultrastructural features in electron microscopy [132]. In addition, transmission electron microscopy (TEM) was also applied to study the role of adipose tissue function in obesity [133]. Liu et al. [134] examined the morphological features of the adult and neonatal adipose tissue of mice by using TEM and SEM. They found that the neonatal adipose tissue contains a smaller number and size of lipid droplets, as well as more abundant mitochondria, compared with the adult samples. The dynamic morphological changes revealed that the number and size of lipid droplets increased, but the  number of mitochondria gradually decreased during the postdelivery development. Thus, careful morphological observations revealed the  individual surface morphology of some adipose tissue samples taken from different patients.

Additionally, few biosensing systems were developed for the  determination of different compounds presented in adipose tissue and for the  investigating obesity-associated inflammation [135–138].

CONCLUSIONS

In this review, for the first time, to the best of our know ledge, the  application of analytical techniques, such as gas and liquid chromatography, Fourier transform infrared (FTIR) spectroscopy, nuclear magnetic resonance (NMR) spectroscopy, elemental analysis and scanning electron microscopy (SEM) for the characterisation of adipose tissue from obese patients, was summarised. Gas and liquid chromatography are conventional methods for determining the composition of fats in the adipose tissue. Liquid chromatography techniques are also very important in lipidomics research. According to the FTIR analysis results, it was concluded that the main functional groupings in the adipose tissue taken from subcutaneous, preperitoneal and visceral layers of different obese patients were different. Therefore, FTIR spectroscopy was successfully used for the  analysis and direct qualitative characterisation of adipose tissue composition. The Raman spectroscopy showed even more possibilities for the analysis of the specific features of adipose tissue. It was demonstrated that the ¹H NMR spectra of adipose tissue samples could be used to predict the different pathologies of obese patients. The results of elemental analysis showed a various distribution of different metals in the adipose tissue of differently obese patients. It could be concluded that the  results of distribution of some metals in the adipose tissue layers of people with overweight are promising for further medical observation. Also, careful morphological observations using SEM and TEM measurements revealed the individual surface morphology of adipose tissue samples taken from different patients. From the  obtained results is was concluded that such characterisation of adipose tissue is an essential step for the possible prediction of the appearance of symptoms of different diseases, since recent studies have emphasised on a close relationship between the adipose tissue properties and development of fat distribution and metabolism, and different diseases. This may facilitate the future development of new prognostic tools useful for personalised treatment strategies that address problems in obesity and its complications. In summary, this review shows interesting findings, which may provide new insights into the treatment of metabolic diseases.

References

1. D.  T.  Villareal, C.  M.  Apovian, R.  F.  Kushner, S. Klein, Am. J. Clin. Nutr., 82, 923–934 (2005).

2. F.  J.  Ruiz-Ojeda, A.  Anguita-Ruiza, R.  Leis, C. M. Aguilera, Ann. Nutr. Metab., 73, 89–99 (2018).

3. E. Lespessailles, J. Paccou, R.-M. Javier, T. Thomas, B.  Cortet, J. Clin. Endocrinol. Metab., 104, 4756– 4768 (2019).

4. N. Hwalla, Z. Jaafar, Diagnostics, 11, 24 (2021).

5. H.  B.  Kitzinger, B.  Karle, Eur. Surg., 45, 80–82 (2013).

6. L. Basurto, L. Sanchez, A. Diaz, M. Valle, A. Robledo, C. Martinez-Murillo, Thrombosis Res., 180, 110–114 (2019).

7. A. Ortiz-Dosal, P. Rodil-Garcia, L. A. Salazar-Olivo, Biomarkers, 24, 499–509 (2019).

8. J. R. Larsen, L. Dima, C. U. Correll, P. Manu, Expert Rev. Clin. Pharmacol., 11, 397–410 (2018).

9. T. Grewal, C. Enrich, C. Rentero, C. Buechler, Int. J. Mol. Sci., 20, 3449 (2019).

10. C. Koliaki, S. Liatis, A. Kokkinos, Metabolism Clin. Exper., 92, 98–107 (2019).

11. H. Li, A. Chen, L. Shu, et al., Genes Cells, 19, 793– 802 (2014).

12. B. A. Evans, J. Merlin, T. Bengtsson, D. S. Hutchinson, Br. J. Pharmacol., 176, 2416–2432 (2019).

13. Q. Zhu, B. J. Glazier, B. C. Hinkel, et al., Int. J. Mol. Sci., 20, 2707 (2019).

14. G. Brimas, R. Skaudzius, V. Brimiene, R. Vaitkus, A. Kareiva, Medic. Hypotheses, 136, 109526 (2020).

15. V. Mohamed-Ali, J. H. Pinkney, S. W. Coppack, Int. J. Obes. Relat. Metab. Disord., 22, 1145–1158 (1998).

16. L. Luo, M. Liu, J. Endocrinol., 231, R77–R99 (2016).

17. S.  Russo, M.  Kwiatkowski, N.  Govorukhina, R.  Bischoff, B.  N.  Melgert, Front. Immunol., 12, 746151 (2021).

18. C. M. Ahlberg, L. N. Schiermiester, J. T. Howard, C. R. Calkins, M. L. Spangler, Meat Sci., 98, 804–814 (2014).

19. M. Ebadi, V. E. Baracos, O. F. Bathe, L. E. Robinson, V. C. Mazurak, Clin. Nutrit., 35, 1347–1353 (2016).

20. J. P. Bastard, B. Feve, Physiology and Physiopathology of Adipose Tissue, Springer, Verlag, France (2013).

21. A.  Selovic, J.  Sarac, S.  Missoni, J. Matern. Fetal Neonatal Med., 29, 2131–2137 (2016).

22. B. Hacihamdioglu, G. Ocal, M. Berberoglu, et al., Ultrasound Med. Biol., 40, 871–876 (2014).

23. A.  T.  Cote, A.  A.  Phillips, K.  C.  Harris, G. G. S. Sandor, C. Panagiotopoulos, A. M. Devlin, Arterioscler. Thromb. Vasc. Biol., 35, 1038–1044 (2015).

24. M. De Spiegeleer, E. De Paepe, L. Van Meulebroek, I. Gies, J. De Schepper, L. Vanhaecke, Mol. Med., 27, 145 (2021).

25. K.  G.  Kahl, J.  Herrmann, B.  Stubbs, et  al., Progr. Neuropsychopharmacol. Biolog. Psychiatry, 72, 30– 35 (2017).

26. C. Pisitsak, J. G. H. Lee, J. H. Boyd, H. O. Coxson, J.  A.  Russell, K.  R.  Walley, Critic. Care Med., 44, 1966–1973 (2016).

27. G. E. Shearrer, M. J. Daniels, C. M. Toledo-Corral, M.  J.  Weigensberg, D.  Spruijt-Metz, J.  N.  Davis, Physiol. Behav., 167, 188–193 (2016).

28. S. J. Yu, W. Kim, D. Kim, et al., Medicine, 94, e2159 (2015).

29. E. S. Wahyuni, I. W. A. Wiyasa, N. Nurdiana, Middle East Fertil. Soc. J., 21, 259–263 (2016).

30. M. Z. Zhang, T. Hu, S. Y. Zhang, L. Zhou, Sci. Rep., 5, 18495 (2015).

31. Y.-C.  Tsao, J.-Y.  Chen, W.-C.  Yeh, Y.-S.  Peng, W.-C. Li, BMJ Open, 0, e017117 (2017).

32. J. C. Brown, M. O. Harhay, M. N. Harhay, Insights, 12, 1–3 (2019).

33. T. Iwase, T. Sangai, H. Fujimoto, et al., Breast Cancer Res. Treatm., 179, 435–443 (2020).

34. S.  Casteras, A.  Abdul-Wahed, M.  Soty, et  al., Diabetologia, 59, 2645–2653 (2016).

35. D. H. Ipsen, P. Tveden-Nyborg, J. Lykkesfeldt, Curr. Obes. Rep., 5, 405–412 (2016).

36. S. Wang, X. Liu, Q. Chen, C. Liu, C. Huang, X. Fang, BMC Anesthesiol., 17, 118 (2017).

37. C. Aguiar, J. MacLeod, A. Yip, et al., BMJ Open, 9, e023418 (2019).

38. S.  Russo, M.  Kwiatkowski, N.  Govorukhina, R.  Bischoff, B.  N.  Melgert, Front. Immunol., 12, 746151 (2021).

39. A. Feola, S. Ricci, S. Kouidhi, et al., J. Cell. Physiol., 232, 69–77 (2017).

40. M.  A.  Kasinska, J.  Drzewoski, A.  Sliwinska, Arch. Med. Sci., 12, 1293–1301 (2016).

41. T. Y. Kim, A. L. Schafer, Curr. Osteoporos. Rep., 14, 337–344 (2016).

42. A.  Chirita-Emandi, M.  C.  Papa, L.  Abrudan, et  al., Rom. J. Morphol. Embryol., 58, 115–123 (2017).

43. J.  T.  Brenna, M.  Plourde, K.  D.  Stark, P.  J.  Jones, Y.-H. Lin, Am. J. Clin. Nutr., 108, 211–227 (2018).

44. L.  R.  S.  Abreu, M.  K.  Shirley, N.  P.  Castro, et  al., PLOS ONE, 14, e0221971 (2019).

45. F. C. Gabriel, G. Fantuzzi, Nutrition Res., 72, 18–35 (2019).

46. M. J. Trites, R. D. Clugston, Lipids Health Dis., 18, 204 (2019).

47. J.  Bliton, D.  Sussman, R.  M.  Summers, J.  H.  Yao, Current Med. Imag., 13, 364–382 (2017).

48. Z.  Bahadoran, S.  Jeddi, S.  Gheibi, P.  Mirmiran, K.  Kashfi, A.  Ghasemi, EXCLI J., 19, 972–983 (2020).

49. I.  Liakh, T.  Sledzinski, L.  Kaska, P.  Mozolewska, A. Mika, Molecules, 25, 5307 (2020).

50. A. Bar, A. Kieronska-Rudek, B. Proniewski, et al., J. Am. Heart Assoc., 9, e016929 (2020).

51. F.  Castriota, L.  Rieswijk, S.  Dahlberg, et  al., Env. Health Perspect., 128, 016001 (2020).

52. A. A. Tinkov, O. P. Ajsuvakova, T. Filippini, et al., Biomolecules, 10, 658 (2020).

53. A. V. G. Ramirez, D. R. Filho, L. B. P. C. de Sa, Curr. Diabetes Rev., 17, e081020184730 (2021).

54. M. Aslan, Mini Rev. Org. Chem., 18, 3–10 (2021).

55. Z.  Suna, B.  D.  Gepner, S.-H.  Leea, et  al., Acta Biomater., 129, 188–198 (2021).

56. S. Bamba, O. Inatomi, K. Takahashi, et al., Inflamm. Bowel Dis., 27, 1435–1442 (2021).

57. J.-B.  Bouillon-Minois, M.  Trousselard, et  al., Nutrients, 13, 3350 (2021).

58. S. Vedantham, L. Shi, A. Karellas, Phys. Med. Biol., 59, 6387–6400 (2014).

59. J. Liu, C. Zhang, B. Zhang, et al., Cells, 9, 201 (2020).

60. W. Stoffel, F. Chu, E. H. Ahrens, Anal. Chem., 31, 307–308 (1959).

61. N. Quinete, T. Schettgen, J. Bertram, T. Kraus, Anal. Bioanal. Chem., 406, 6151–6164 (2014).

62. F.  Bevalot, C.  Bottinelli, N.  Cartiser, L.  Fanton, J. Guitton, J. Anal. Toxicol., 38, 256–264 (2014).

63. D. Lu, C. Feng, Y. Lin, et al., Chemosphere, 114, 327– 336 (2014).

64. M. Wolters, C. Dering, A. Siani, et al., PLoS ONE, 12, e0181485 (2017).

65. Z. P. Yali, A. P. Jadid, L. A. Samin, Int. J. Environ. Sci. Technol., 14, 2357–2366 (2017).

66. A. M. Astudillo, C. Meana, C. Guijas, et al., J. Lipid Res., 59, 237–249 (2018).

67. V. Poškus, V. Vičkačkaitė, G. Brimas, Chemija, 27, 179–184 (2016).

68. V.  Poškus, V.  Vičkačkaitė, J.  Dargytė, G.  Brimas, Chemija, 29, 49–56 (2018).

69. J.  P.  Arrebola, J.  Pumarega, M.  Gasull, et  al., Environm. Res., 122, 31–37 (2013).

70. R. A. Ghazala, A. El Medney, A. Meleis, P. Mo hie El dien, H.  Samir, Biomed. Chromatogr., 34, e4743 (2020).

71. J.  Hamulka, J.  Brys, A.  Gorska, M.  JanaszekMankowska, M.  Gornicka, Appl. Sci., 11, 10029 (2021).

72. R. Tessari, E. Mazzotta, F. Blasi, et al., Large Anim. Rev., 27, 187–193 (2021).

73. M.  Lange, Z.  Ni, A.  Criscuolo, M.  Fedorova, 82, 77–100 (2019).

74. R. Anlicoara, A. A. B. Ferraz, K. da P. Coelho, et al., Obes. Surg., 24, 1487–1491 (2014).

75. E. Bartosinska, J. Jacyna, A. Borsuk-De Moor, et al., Talanta, 176, 108–115 (2018).

76. M. Guan, D. Dai, L. Li, et al., Talanta, 172, 206–214 (2017).

77. V.  C.  O.  Lima, A.  B.  S.  Luz, M.  do Socorro M.  Ama rante, et  al., Obes. Facts, 14, 357–369 (2021).

78. O. Awosemo, H. Neelakantan, S. Watowich, et al., Biomed. Anal., 204, 114255 (2021).

79. B. Buckus, G. Brimas, A. Stasinskas, et al., Chemija, 26, 98–106 (2015).

80. F. K. Baloglu, O. Baloglu, S. Heise, G. Brockmann, F.  Severcan, J. Biophotonics, 10, 1345–1355 (2017).

81. E. Aboualizadeh, O. T. Carmichael, P. He, D. C. Albarado, C.  D.  Morrison, C.  J.  Hirschmugl, Front. Endocrinol., 8, 121 (2017).

82. I. Bikulčienė, L. Vasiliauskaitė, Z. A. Kučinskienė, A. Kaminskas, V. Hendrixson, Med. Sci. Monit., 24, 818–826 (2018).

83. D. Ami, P. Mereghetti, A. Foli, et al., Anal. Chem., 91, 2894–2900 (2019).

84. K. Yu, H. Zhang, Y. Liu, et al., Microchem. J, 164, 105977 (2021).

85. S. Liyanage, R. S. Dassanayake, A. Bouyanfif, et al., MethodsX, 4, 118–127 (2017).

86. T. Ansari, A. Southgate, I. Obiri-Yeboa, et al., Stem Cells Develop., 29, 314–326 (2020).

87. F.  Munir, S.  G.  Musharraf, S.  T.  H.  Sherazi, M. I. Malik, M. I. Bhanger, Turk. J. Chem., 43, 1098– 1108 (2019).

88. M. Y. Bekhit, B. Grung, S. A. Mjos, Appl. Spectrosc., 68, 1190–1200 (2014).

89. M.  Motoyama, I.  Nakajima, H.  Ohmori, G.  Watanabe, K. Sasaki, JARQ, 52, 17–22 (2018).

90. F.  Tao, M.  Ngadi, Crit. Rev. Food Sci. Nutrit., 58, 1565–1593 (2018).

91. I. Y. Yanina, Y. I. Svenskaya, E. S. Prikhozhdenko, J. Biophotonics, 11, e201800058 (2018).

92. K. St-Arnaud, K. Aubertin, M. Strupler, et al., Med. Phys., 45, 328–339 (2018).

93. F.  Daoust, T.  Nguyen, P.  Orsini, et  al., J. Biomed. Optics, 26, 022911-1-18 (2021).

94. C. Blücher, C. Zilberfain, T. Venus, et al., Analyst, 144, 5558 (2019).

95. E.  Swiderska, M.  Podolska, J.  Strycharz, et  al., Nutrients, 11, 1835 (2019).

96. P. N. Paramitha, R. Zakaria, A. Maryani, et al., Mol. Sci., 22, 7378 (2021).

97. Z. Majka, K. Czamara, P. Wegrzyn, et al., Analyst, 146, 270 (2021).

98. P.  Meksiarun, B.  B.  Andriana, H.  Matsuyoshi, H. Sato, Sci. Rep., 6, 37068 (2016).

99. M.  Troyanova-Wood, C.  Gobbell, Z.  Meng, A.  A.  Gashev, V.  V.  Yakovlev, J. Biophotonics, 10, 1694–1702 (2017).

100. C. T. Kucha, L. Liu, M. O. Ngadi, Sensors, 18, 377 (2018).

101. G. Donjuan-Loredo, R. Espinosa-Tanguma, F. Leon Bejarano, et  al., Appl. Specrosc, 75, 1189–1197 (2021).

102. A. Zajac, L. Dyminska, J. Lorenc, M. Ptak, J. Hanuza, Spectroscopy Lett., 51, 81–88 (2018).

103. M. Troyanova-Wood, C. Gobbell, Z. Meng, O. Gasheva, A. A. Gashev, V. V. Yakovlev, J. Bio photonics, 14, e202000269 (2021).

104. C.  Blay, P.  Haffray, J.  D’Ambrosio, et  al., BMC Genomics, 22, 788 (2021).

105. C. Corden, R. Boitor, I. Notingher, J. Phys. D Appl. Phys., 54, 504003 (2021).

106. G.  Donjuan-Loredo, R.  Espinosa-Tanguma, M.  G.  Ra mirez-Elias, Appl. Spectrosc. Rev. (2021) [https://doi.org/10.1080/05704928.2021.1944175].

107. B. Jin, L. P. Liu, S. X. Zhang et al., Metabol. Syndr. Relat. Disord., 15, 439–449 (2017).

108. I. J. Neelanda, C. Hughes, C. R. Ayers, C. R. Malloy, E. S. Jin, Metabolism, 67, 80–89 (2017).

109. M. Gajdosik, L. Hinger, A. Skoch, et al., J. Magn. Reson. Imaging., 50, 71–82 (2019).

110. D. Le Duc, C.-C. Lin, Y. Popkova, et al., Int. J. Obes., 45, 565–576 (2021).

111. A. N. Torres, L. Tavares, M. J. Pereira, J. W. Eriksson, J. G. Jones, NMR Biomed., e4632 (2021).

112. A.  Paul, S.  Kumar, A.  Raj, Metabolomics, 14, 119 (2018).

113. A. F. Soares, J. Paz‐Montoya, H. Lei, M. Moniatte, R. Gruetter, NMR Biomed., 30, e3761 (2017).

114. C.  Carpene, J.  Mercader, S.  Le Gonidec, Br. J. Pharmacol., 175, 2428–2440 (2018).

115. S. O. Senera, U. Ozgen, S. Kanbolat, S. Afr. J. Bot., 141, 243–254 (2021).

116. C.  J.  Fallone, A.  G.  Tessier, C.  J.  Field, A.  Yahya, NMR Biomed., 34, e4455 (2021).

117. S. Furse, D. S. Fernandez-Twinn, B. Jenkins, Anal. Bioanal. Chem., 412, 2851–2862 (2020).

118. J.-L.  Jung, G.  Simon, E.  Alfonsi, PLoS One, 12, e0180597 (2017).

119. A. Meusel, Y. Popkova, D. Dannenberger, J. Schiller, Food Anal. Methods, 10, 2497–2506 (2017).

120. J. O. Aleman, N. A. Bokulich, J. R. Swann, J. Transl. Med., 16, 244 (2018).

121. J.  M.  Moreno-Navarrete, M.  Serino, V.  Blasco-Baque, et  al., Mol. Nutr. Food Res., 62, 1700721 (2018).

122. A. B. Granado-Serrano, M. Martin-Gari, V. Sanchez, et al., Sci. Rep., 9, 1772 (2019).

123. K.  Oshakbayev, B.  Dukenbayeva, G.  Togizbayeva, et al., BBA Clin., 8, 14–19 (2017).

124. E. M. Johnston, E. Dao, M. J. Farquharson, et al., X‐Ray Spectrom., 48, 432–437 (2019).

125. A.  Kizalaite, V.  Brimiene, G.  Brimas, et al., Trace Elem. Res., 189, 10–17 (2019).

126. J. M. Dick, PeerJ, 5, e3421 (2017).

127. P. M. Martins, R. D. Bello, B. Ackermann, et al., Sci. Rep., 10, 7007 (2020).

128. D. P. Lo Fiego, G. Minelli, L. A. Volpelli, A. Ulrici, P. Macchioni, Meat Sci., 122, 132–138 (2016).

129. R. Skaudžius, M. Misevičius, V. Brimienė, M. Beniu šė, G.  Brimas, A.  Kareiva, Chemija, 29, 67–80 (2018).

130. E.  Brimas, R.  Skaudžius, G.  Brimas, A.  Selskis, R.  Ra manauskas, A.  Kareiva, Chemija, 33, 7–11 (2022).

131. H. Robles, S. J. Park, M. S. Joens, J. A. J. Fitzpatrick, C.  S.  Craft, E.  L.  Scheller, Bone, 118, 89–98 (2019).

132. M. Galhuber, N. Kupper, G. Dohr, et al., Cell Biol., 155, 593–603 (2021).

133. F.  Wu, X.  Yang, M.  Hu, et  al., Phytomedicine, 76, 153258 (2020).

134. J. Liu, C. Zhang, B. Zhang, et al., Cells, 9, 201 (2020).

135. J. Zhu, J. He, M. Verano, et al., Lab. Chip., 18, 3550 (2018).

136. M. Jafari, M. Hasanzadeh, R. Karimian, N. Shadjou, Microchem. J., 147, 741–748 (2019).

137. K. L. Westmacott, A. P. Crew O. Doran, J. P. Hart, Biosens. Bioelectron., 150, 111837 (2020).

138. E. Santos, M. F. F. Vara, M. Ranciaro, W. Strasse, G.  N.  N.  Neto, P.  Nohama, J. Biomechan., 122, 110456 (2021).

Edvardas Brimas, Rimantas Raudonis, Aivaras Kareiva

ANALIZĖS METODAI, NAUDOJAMI SPECIFINĖMS SU NUTUKIMU SUSIJUSIŲ BIOLOGINIŲ AUDINIŲ YPATYBĖMS APIBŪDINTI: APŽVALGA