Milk Classification and Water Adulteration Detection by TD-NMR and Dielectric Spectroscopy

This study presents a fast, non-invasive method to determine the iodine value (IV) of edible oils using Time Domain Nuclear Magnetic Resonance (TD-NMR) and Magnetic Resonance Imaging (MRI) techniques. The IV, which quantifies the degree of unsaturation in fats and oils, is a key parameter in assessing oil quality and detecting potential adulteration. Different edible oils were used in this study (sunflower, soy, olive, sesame, and linseed). Statistically significant regression models (R2 > 0.92) were established between the IV derived from NMR spectra and the longitudinal (T1) and transverse (T2) relaxation times of the oils, which were obtained from MRI and TD-NMR analyses. The regression models obtained allow for the prediction of the IV from the T1 and T2 relaxation times across a range that includes predominantly mono- and polyunsaturated edible vegetable oils. The TD-NMR approach stands out for its speed (<2 min), lack of sample preparation (including direct analysis within the commercial packaging), and reproducibility, with a variability of only 0.62%. Meanwhile, the MRI technique allows for the simultaneous evaluation of multiple samples in a single acquisition. Together, these features make TD-NMR and MRI effective tools for the rapid and reliable analysis of the IV in edible oils.

The objective was to evaluate the effect of the type of milking and the environment in the quality of raw milk by identification the Somatic Cell Count - CCS in samples obtained from farms in the mesoregion of south goiano and correlate it with the different types of milk, and the milk components: fat, protein, lactose and total solids (ST). Furthermore, we sought to diagnose the prevalence of subclinical mastitis by CCS, evaluate the correlation between CCS and milking type with antimicrobial residues, and check the influence of season, temperature and humidity index (THI) and Equivalent Temperature Index (ITE) on the CCS. The research developed in farms with different types of milking, being divided into manual and mechanical milking. The datas were collected from January 2011 to December 2013, and in each property was collected a monthly sample in the expansion tank, a total of 8242 samples, which were later sent for analysis. The diagnosis of the milk quality held by electronic count of somatic cells, for the FossomaticTM® and the levels of the components were analyzed using the MilkoscanTM FT®, the detection of antibiotic residues was performed using the Delvotest-P and meteorological data were obtained from the National Institute of Meteorology - INMET in Catalão-GO station. Among the variables type of it milks and subclinical mastitis, there was dependent relationship (p <0.05). The milking mechanics was the one with a higher prevalence of subclinical mastitis with 58% (910) of the samples presented CCS> 300.000 cells/mL. Among the variables type of milking and CCS advocated by Instruction n°62/2011 (BRASIL, 2011), there was a significant relationship (p <0.05). Only 19,53% (1610) of the samples had CCS > 500.000 cells/mL. According to the statistical analysis, there was a positive correlation between CCS and fat and ST, however, there was a negative correlation between the CCS and the protein and lactose contents. The type of milking and the somatic cell count did not affect the presence or absence of antimicrobial residues. During the rainy season has become a CCS elevation compared to the dry season, but there was no correlation between CCS and environmental indices (ITE and ITU). Regarding environmental indices can be seen that with the increase of the ITU occurred a decrease in fat and an increase in ST and lactose. It can be concluded that the type of manual milking showed the lowest prevalence of subclinical mastitis in relation to milking and with elevated CCS is a reduction in the concentrations of protein and lactose and an increase in fat and total solids in the raw milk.

The aim of our study was to assess Staphylococcus aureus growth and the time of first detection of staphylococcal enterotoxins type A, B and C (SEA, SEB, SEC) in different type of milk, depending on the strain and storage conditions. Raw, pasteurized, and UHT milk were inoculated with three strains of S. aureus, and growth patterns were determined by the plate method in accordance with EN ISO 6888-1. Baird-Parker agar medium was used for the detection of S. aureus and the Enzyme Linked Fluorescent Assay (ELFA) used with a miniVIDAS analyzer tested the production of staphylococcal enterotoxins. The results of model experiments showed the dependence of the growth rate and subsequent production of staphylococcal enterotoxins on incubation (storage) temperature, S. aureus strain, and type of milk. A significant finding was that the growth of S. aureus and production of enterotoxins in raw milk was inhibited by natural microflora, and production of enterotoxins was therefore not detected in raw milk within 102 hours of storage either at 15 °C or 22 °C. The highest risk of SEs production is associated with secondary contamination of pasteurized and UHT milk when stored at room temperature, where production was first detected after 12 hours of incubation.

Although under certain circumstances it is necessary to express milk, there are not many recommendations about the ideal storage conditions for human milk. The objectives of this study were to analyze the effects on Dornic acidity of frozen storage at -20 °C in three types of milk: raw donor milk, mother's own raw milk, and pasteurized donor milk. Forty-three samples of raw donor milk, 40 samples of pasteurized donor milk, and 16 samples of mother's own milk were analyzed. Dornic acidity was measured at time 0, before freezing. The remaining aliquots were frozen and analyzed after 1, 2, 3, and 4 weeks and after 2 and 3 months. In raw donor milk, the median acidity at the start was 3 °D (interquartile range [IQR] 2-3 °D); after 3 months, it was 5 °D (IQR 3-7 °D), with a significant increase in acidity after the second week. In mother's own milk, the mean acidity at the start was 3 °D (IQR 2-4 °D) and 7 °D (IQR 4-8 °D) at 3 months. The increase was significant after the third week. In pasteurized donor milk, the mean acidity was 3 °D (IQR 2-3 °D) at the start and 2 °D (IQR 2-3 °D) at the end. When comparing the three types of milk, there were significant differences from the first week between the two types of raw milk and the pasteurized milk (p < 0.01), but not between the two raw milks (p = 0.77). Dornic acidity in unpasteurized milk significantly increases with the duration of freezing, probably due to the action of lipases, which is lost with pasteurization. It would be advisable to reduce the length of freezing time for unpasteurized milk.

Nuclear magnetic resonance (NMR) is a versatile technique that allows assessing the composition and structure of food samples. It provides information on molecular properties, which can be used for quantification of concentration and interactions of components within the food samples. Due to the fact that many foods are proton-rich (with protons originating from water, fat, carbohydrates, and proteins) and carbon-rich (originating from fat, carbohydrates, and proteins), 1H and 13C NMR have become the most common type of NMR in food analysis.1-3 NMR is used in a wide range of applications, included in the fields of food microbiology, food chemistry, food engineering and food packaging (see Table 1)4-23. Table 1 shows the multiple applications of NMR in food science. Analysis of food components is nowadays one of the main interests of authors. Water, proteins, and fat have been studied qualitatively, and in some cases, also quantitatively. Other aspects, such as processing, conformational analysis of certain molecules, and packaging, are also among the targets. In case of quality control, NMR is valued to certificate the quality of highly appreciated food products for consumers, for instance, olive oil. Within the field of food process engineering, two water-in-food issues widely studied with NMR techniques are the translation and location of water in food samples, as well as water-biopolymers interactions.24-27 Within all types of NMR, time domain nuclear magnetic resonance (TD-NMR) turns out to be a fast and accessible tool to study water populations present in food. Most of the TD-NMR studies on food samples use relaxometry methods. Spectral information seems to provide less information for water-related issues because water present in various physical environments does not give chemical differences (e.g., chemical shifts). The reason is that water is found in various physical environments, but chemical differences barely exist.26 The relaxometry analyses are based on differences in transverse relaxation times (the so-called T2, or its inverse R2 = 1/T2) of components. T2 values for water in food depend on the concentration of water in the food matrix, and usually range from microseconds to seconds. These T2 values can therefore be related to different populations of protons (1H) with various rotational mobilities, which can then be assigned to different groups of proton-bearing water populations contained in a proteinaceous sample.4, 26 The objective of this perspectives paper is, first, to collect studies on water binding to proteins and their interactions within proteinaceous food samples. Then, perspectives of future work on this topic are presented. In order to study water population within samples, Dekkers et al. (4) associated transverse relaxation times on the TD-NMR spectrum to proton-bearing components. As shown in Figure 1, the transverse decay curves were analysed with the Levenberge-Marquardt non-linear least squares algorithm. This algorithm fits a sum of exponential curves—the so-called components—to the decay, resulting in an amplitude A and transverse relaxation time T2 for each component. The longest relaxation time of the sample in the spectrum, which is shorter than that of pure water, would correspond to some relatively mobile water within interstices in the protein matrix. The fact that this relaxation time is lower than that of free water, in their conclusion, may be due to pore confinement and by fast proton exchange with dissolved components. When the concentration of soy protein isolate (SPI) rises, this water population disappears, which suggests stronger water-binding capacity of SPI at this concentration. The relaxation time of the averaged middle two components represented more than 90% of the signal intensity. This signal was assigned to absorbed water. The T2 of this water population decreases as a function of protein concentration, reflecting the decreasing mobility of this fraction. Finally, the component with the shortest relaxation time probably represents protein-bound protons. The intensity of this signal rises with concentration, which leads to think that it could represent the protein-bound proton population. The slight dispersion of relaxation rates of the middle fraction—which corresponds to second and third water populations, averaged on this spectrum—may be due to the heterogeneity within the individual phases. Another example of information that can be extracted from TD-NMR studies on water is the compared study of water-binding capacity (WBC) of different materials. Figure 2 shows the comparative study of four plant-based materials carried out by Peters et al.6. It can be observed that vital wheat gluten calibration curve increased more gradually than was found for SPI, pea protein isolate, and lupin protein concentrate. This behaviour might be explained as being related to isoelectric point of the proteins. The isoelectric point of gluten is around 7.5—whereas the value for the other proteins is around 4.5—which would lead gluten to less interact with water in these samples.28 Dekkers et al.5 focused on the WBC of phases in a blend of plant-based materials in order to better understand the rheological behaviour of that blend. The blend consisted of a phase-separated blend of an aqueous SPI phase and aqueous Wheat Gluten (WG) phase. The exact distribution of water over both phases determines the blend rheological properties. In past studies on rheology of biopolymer blends, the water distribution was often taken as a fit parameter. Now, NMR was used to determine the water distribution quantitatively, and the outcomes of NMR turned out to be fully in line with water distributions as predicted by rheology. The authors calculated a theoretical R2 value of water distributions in SPI-WG blends, then compared these values with experimental R2 values of the blends. They observed that experimental R2 values were lower than the calculated R2 values, which would mean that water was more mobile than expected. This could be explained by the fact that water was not evenly distributed among the SPI and WG phases. Peters et al.8 studied a dairy material for its WBC as well. They compared measurements of WBC on whey protein pellets by weighing the pellet obtained after centrifuging a dispersion with TD-NMR measurements. Previously, it was assumed that the interstitial water (water between particles) can be neglected, implying that all water was absorbed by dry matter in the pellet. However, this assumption turns out to be inaccurate to measure WBC. As a matter of fact, TD-NMR did allow differentiation between water inside and between the particles, which leads to a more accurate assessment of different populations of water inside the food sample. Figure 3 shows how TD-NMR can be potentially useful to study WBC in food samples. As stated in the introduction, a need exists for developing a new category of product, being dense protein food products. Partly as healthy foods,29 partly as products that can replace products from animal origin. A key element of interest is to better understand the sensory properties of those products. High-protein foods are often associated with hard and dry texture. Furthermore, plant-based products are often less juicy than the animal-based ones. TD-NMR is an attractive tool to investigate water binding in dense protein products. It allows to differentiate the different water populations in a product as well as to quantify them. Thus, a simple but versatile tool is available to compare differences in water binding of various products. We can therefore imagine that future research will focus TD-NMR studies on a broad range of samples, including plant-based food samples. For this purpose, both classic ingredients (soybean, lupin, wheat gluten) as well as novel plant-based ingredients (rapeseed, sunflower, pea, faba bean) will be studied. The first objective will be to identify water populations within the samples and their dependence with certain parameters, such as dry matter content, pH, or amount of added salt. Certain stages of processing, such as heating or shearing, can be included as well. A further step would be to compare the relative TD-NMR signals to carry out semi-quantification. Finally, we will aim to quantify these water populations in absolute units. The second objective will be the study of the biopolymers, which constitute the food matrix, these include fat, proteins, and carbohydrates as well as the interactions amongst them. Fat content determination is nowadays labour-intensive and invasive, thus a faster and non-invasive alternative is highly desirable. Both identification and quantification of these biopolymers will be aimed.

Epidemiological studies have shown an association between the consumption of raw farm milk and reduced incidence of allergy. In the present study, we fed untreated raw milk, gamma-sterilised milk, heat-treated milk or water to mice and compared their responses to allergen exposure and challenge treatment in a mouse model of gastrointestinal allergy. From weaning (3 weeks old), groups of BALB/c female mice (n 8) received raw milk, gamma-sterilised milk, heated milk or water via drink bottles, with the control group receiving water. All mice were fed a standard (dairy protein-free) rodent diet. At 6 and 8 weeks, groups were given intra-peritoneal injections with ovalbumin (OVA)/alum to sensitise them to the antigen. Controls were sham immunised. At week 10, mice were fasted and challenged four times on alternate days by intra-gastric administration with 50 mg OVA or saline. Levels of bacteria and milk proteins were assessed in milk samples. Mouse serum levels of specific IgE, IgG1 and IgG2a antibodies and mouse mast cell protease-1 (MMCP-1) were determined. Cytokine responses to 48 h activation with OVA were measured in cultured splenocytes from mice. Sterilised and heated milks contained no viable bacteria and reduced detectable levels of many milk proteins, in contrast to raw milk. Mice drinking raw milk had highest serum MMCP-1 and specific-OVA IgE responses. Cultured splenocytes from OVA-primed mice produced similar levels of IL-4 in response to the antigen; however, IL-10 levels were highest from mice drinking raw milk. Overall, the present study adds to the evidence that consuming different types of milk can affect allergic responses to a non-related dietary antigen.

The relationship between different types of milk and depression and anxiety remains unclear, with limited evidence from prospective cohort studies. This study aims to evaluate this relationship using data from the UK Biobank cohort and to explore its potential causality through Mendelian randomization (MR) analysis. Cox proportional hazards models were used to assess the association between different milk types and the risk of depression and anxiety among 357,568 UK Biobank participants free of these conditions at baseline. To further explore causality, a 2-sample MR analysis was conducted using data from the FinnGen study. During a median follow-up period of 13.5 years (interquartile range, 12.6-14.2 years), among 357,568 participants (mean [SD] age, 56.83 [8.06] years, 171,246 male individuals [47.9%]), a total of 13,065 and 13,339 participants were diagnosed with depression and anxiety, respectively. In the fully adjusted model (adjusted for sociodemographics characteristics, lifestyle behaviors and health indicators), full cream milk was related to a lower risk of anxiety (HR = 0.84, 95% CI: 0.75-0.94). Semi-skimmed milk had a lower risk of depression (HR = 0.88, 95% CI: 0.80-0.96) and anxiety (HR = 0.90, 95% CI: 0.82-0.98). No significant relationships were found between skimmed milk and depression/anxiety. Other types were related to an increased risk of depression (HR = 1.14, 95% CI: 1.02-1.28). After Bonferroni correction, the 2-sample MR analysis revealed a potential protective causal relationship between semi-skimmed milk and depression (OR = 0.83, 95% CI: 0.73-0.95, p = 0.006) and anxiety (OR = 0.71, 95% CI: 0.59-0.85, p < 0.001). These findings indicate that semi-skimmed milk consumption may be linked to a lower risk of depression and anxiety, potentially highlighting its role in dietary strategies to promote mental health.

Objectives: The presence of pathogenic bacteria in milk is the major public health concern resulting in food borne illness. The aim of this study is to determine the microbial quality of three different types of milk consumed in Kathmandu Valley with respect to the acceptable standard guideline and measure the antibiotic susceptibility pattern of the Escherichia coli and Staphylococcus aureus isolates.Methods: A total of 66 samples (16 pasteurized, 25 unpasteurized and 25 raw milk) were collected from various sites of Kathmandu Valley. Those samples were subjected for total plate count and total coliform count by pour plate method. Furthermore, identification was made for the presence of E. coli and S. aureus with biochemical tests.Results: The mean total plate count (TPC) of pasteurized, unpasteurized and raw milk was 1.2X106 cfu/ml, 2.3 X 107 cfu/ml and 2.0 X 107 cfu/ml respectively. And, the mean total coliform count (TCC) of pasteurized, unpasteurized and raw milk was 2.9 X 104cfu/ml, 6.3 X 105 cfu/ml and 1.6 X 105 cfu/ ml respectively. Coliforms were detected in 50%, 84% and 56% of the pasteurized, unpasteurized and raw milk sample respectively. E. coli and S. aureus were isolated from 18.8% and 12.5% of pasteurized, 40% and 16% of unpasteurized and 20% and 24% of the raw milk samples respectively. Among total E. coli isolates (n=18), 16.7% were susceptible to ampicillin whereas 100% isolates were susceptible to other tested antibiotics. Similarly, 33.3% and 66.7% of the isolated S. aureus were susceptible to penicillin and cefoxitin respectively, whereas all S. aureus isolates were sensitive to all other antibiotics.Conclusion: The mean value of TPC and TCC of pasteurized and raw milk exceed the standard guideline by FDA. Higher total plate count and presence of coliforms (also E. coli) and S. aureus in this study necessitates the close monitoring of the pasteurization process and post pasteurization process (packaging, transportation, storage etc.).

Improving the competitiveness of dairy products of Ukrainian producers in accordance with European standards

The concentration of aflatoxin M1 in raw and pasteurized milk: A worldwide systematic review and meta-analysis

Molecular mobility changes after high-temperature, short-time pasteurization: An extended time-domain nuclear magnetic resonance screening of ewe milk

Bouhezza is an Algerian traditional cheese made using different types of milk (goat, sheep and cow), ripened in a goat-skin or sheep-skin bag called Chekoua or Djeld of Bouhezza. The objective of the study was the characterization of Bouhezza cheese through ten cheese preparation trials using goat raw milk. The chemical and microbiological composition and proteolysis were monitored on ripened samples. The physicochemical characteristics showed that Bouhezza is acidic with a pH value of 3.95 and 2.36% for titratable acidity as lactic acid, 13.20% for protein and 12.74% for fat. At the end of ripening, Bouhezza was classified to be a ripened, soft and mid-fat cheese. Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) of Bouhezza samples showed low proteolysis after 30 days of ripening. Microbiological results showed a high content of lactic acid bacteria (LAB), 8 to 9 log cfu/ g, an absence of pathogenic bacteria, and a low level of contamination flora. This makes the cheese a healthy product and of an acceptable hygienic quality. Four proteolytic lactic acid bacteria were identified by 16S ribosomal deoxyribonucleic acid (rDNA) sequencing: two strains of Enterococcus faecalis, one E. faecium, and one Lactobacillus paracasei ssp. paracasei. Reversed phase-high performance liquid chromatography (RP–HPLC) of the soluble fraction at pH 4.4 showed changes occurred during ripening. This work highlights, characterizes, and improves the knowledge of cheese Bouhezza made with raw goat's milk. This product is rich in nutrients, such as proteins and minerals, that are useful to the consumer.

A longitudinal one-year study was conducted to determine aflatoxin M1 levels in different types of milk marketed in Pakistan. Processed and raw liquid milk from 21 sources, two milk powder and six tea whitener brands were sampled on monthly basis from Islamabad. The aflatoxin M1 levels in liquid milk were lower (p < 0.05) in summer (April to July) compared with the levels in winter (January, November and December). The mean aflatoxin M1 levels were 254.9, 939.5, and 1535.0 ng/L in UHT, pasteurized, and raw milk, respectively (differing at p < 0.001). The mean toxin level in powdered milk after reconstitution was 522.1 ng/L. Overall, 12.9, 41.0, 91.9 and 50.0% of the UHT, pasteurized, raw and powdered milk samples, respectively, exceeded the Codex maximum tolerable limit of 500 ng of aflatoxin M1/L. It was estimated that consumers of raw and processed milk were exposed to 11.9 and 4.5 ng aflatoxin M1, respectively, per kg of body weight daily. The study indicates potential aflatoxin M1 exposure risks for the consumers of raw milk in the country. The levels of the toxin though comparatively lower in milk powder, requires attention as this type of milk is consumed by infants.

Heat stability (HS) is substantial technology property of raw milk. Analysis of sources of HS variation and its regular monitoring can contribute to creating higher added value in the dairy industry. The goal of this analysis was to assess the practice sources of raw cow milk HS variability on the results of an extensive data set of bulk tank milk samples. There was implemented neither a compositional technology modification nor acidity adjustment of milk, just original raw milk was used for the analysis. A total 2634 HS analyses were performed, including other milk indicators, during three years of an experimental period. The log HS mean and standard deviation were 1.273654 ± 0.144189, equal to the HS geometric mean of 18.8 min. Explanation of the HS variability through the linear model used was 41.1% (p < 0.0001). According to the results of the variance analysis, the milk HS was influenced (p = 0.0033 and mostly <0.0001) by all the farm factors such as year; season; calendar month; altitude; total annual rainfall; herd size by the number of cows; milk yield; cow breed; type of milking; litter type in the stable; summer grazing application; farm effect. During the calendar months (p < 0.0001), milk HS values suggest similar seasonal dynamics with the somatic cell count, total count of mesophilic microorganisms, coli bacteria count and urea and lactose concentration and opposite configuration pattern to fat, crude protein, solids-not-fat and total solids content and milk freezing point depression. Here performed quantification of these effects by analyzing the variance may allow efficient raw milk selection to be processed into specific dairy products.

Detection and discrimination of various liquid substances is a very important problem with respect various security and industrial applications. The techniques, which provide a possibility of the chemical-specific identification of various materials, look prospective for practical implementation. Among various techniques of liquid identification, the time-domain nuclear magnetic resonance (TD-NMR) is considered as very promising for detection of energetic, flammable, and other dangerous liquids. TD NMR looks very attractive because of its high selectivity as well as relatively low price and maintenance expenses. However, for a reliable discrimination in a large set of various substances there is a need to obtain additional parameters for discriminations (e.g., the real and imaginary permittivities by dielectric spectroscopy, the 14N NMR signal, diffusion constant or T 1 /T 2 relaxation times distribution). Dielectric spectroscopy which can measure the dielectric constant and the loss tangent of liquids in the broad range of frequencies (in general, from DC to sub-THz) looks as especially prospective technique for the combined use with TD NMR. In this work, we review possible approaches in TD NMR as well as in MW detection based on probing dielectric permittivities (i.e., dielectric spectroscopy) to probe additional information about liquid content and to shorten the scanning time. Technical approaches for realization of MW detection device as well as construction of TD NMR instrument for obtaining versatile discrimination parameters of the tested liquids are also presented.