Prediction of fatty acid chain length and unsaturation of milk fat by mid-infrared milk analysis.

Our objective was to develop partial least squares (PLS) models to predict fatty acid chain length and total unsaturation of milk fat directly from a mid-infrared (MIR) spectra of milk at 40°C and then determine the feasibility of using those measures as correction factors to improve the accuracy of milk fat determination. A set of 268 milks (modified milks, farm bulk tank milks, and individual cow) were analyzed for fat, true protein, and anhydrous lactose with chemical reference methods, and in addition a MIR absorption spectra was collected for each milk. Fat was extracted from another portion of each milk, the fat was saponified to produce free fatty acids, and the free fatty acids were converted to methyl esters and quantified using gas-liquid chromatography. The PLS models for predicting the average chain length (carbons per fatty acid) and unsaturation (double bonds per fatty acid) of fatty acids in the fat portion of a milk sample from a MIR milk spectra were developed and validated. The validation performance of the prediction model for chain length and unsaturation had a relative standard deviation of 0.43 and 3.3%, respectively. These measures are unique in that they are fat concentration independent characteristics of fat structure that were predicted directly with transmission MIR analysis of milk. Next, the real-time data output from the MIR spectrophotometer for fatty acid chain length and unsaturation of milk were used to correct the fat A (C=O stretch) and fat B (C-H stretch) measures to improve accuracy of fat prediction. The accuracy validation was done over a period of 5 mo with 12 sets of 10 individual farm milks that were not a part of the PLS modeling population. The correction of a traditional fat B virtual filter result (C-H stretch) for sample-to-sample variation in unsaturation reduced the Euclidean distance for predicted fat from 0.034 to 0.025. The correction of a traditional fat A virtual filter result (C=O stretch) modified with additional information on sample-to-sample variation of chain length and unsaturation gave the largest improvement (reduced Euclidean distance from 0.072 to 0.016) and the best validation accuracy (i.e., lowest Euclidean distance) of all the fat prediction methods.

Preparation and stability of milk somatic cell reference materials.

Our objectives were to develop a method to produce milk somatic cell count (SCC) reference materials for calibration of electronic somatic cell count (ESCC) using gravity separation and to determine the effect of refrigerated storage (4°C) and freeze-thaw stability of the skim and whole milk SCC reference materials. Whole raw milk was high-temperature short-time pasteurized and split into 2 portions. One portion was gravity separated at 4°C for 22 h and the second portion was centrifugally separated to produce skim milk that was also gravity separated with somatic cells rising to the surface. After 22 h, stock solutions (low SCC skim milk, high SCC skim milk, high SCC whole milk) were prepared and preserved (bronopol). Two experiments were conducted, one to compare the shelf-life of skim and whole milk SCC standards at 4°C and one to determine the effect of freezing and thawing on SCC standards. Both experiments were replicated 3 times. Gravity separation was an effective approach to isolate and concentrate somatic cells from bovine milk and redistribute them in a skim or whole milk matrix to create a set of reference materials with a wider and more uniformly distributed range of SCC than current calibration sets. The liquid SCC reference materials stored using the common industry practice at 4°C were stable (i.e., fit for purpose, no large decrease in SCC) for a 2-wk period, whereas frozen and thawed reference materials may have a much longer useful life. A gradual decrease occurred in residual difference in ESCC (SCC × 1,000/mL) versus original assigned reference SCC over duration of refrigerated storage for both skim and whole milk SCC samples, indicating that milk ESCC of the preserved milks was gradually decreasing during 28 d of storage at 4°C by about 15,000 SCC/mL. No difference in the ESCC for skim milk was detected between refrigerated and frozen storage, whereas for whole milk the ESCC for frozen was lower than refrigerated samples. Future work is needed to determine the time and temperature of longer term frozen storage over which the SCC results are stable.

A proficiency test system to improve performance of milk analysis methods and produce reference values for component calibration samples for infrared milk analysis.

Our goal was to determine the feasibility of combining proficiency testing, analytical method quality-assurance system, and production of reference samples for calibration of infrared milk analyzers to achieve a more efficient use of resources and reduce costs while maximizing analytical accuracy within and among milk payment-testing laboratories. To achieve this, we developed and demonstrated a multilaboratory combined proficiency testing and analytical method quality-assurance system as an approach to evaluate and improve the analytical performance of methods. A set of modified milks was developed and optimized to serve multiple purposes (i.e., proficiency testing, quality-assurance and method improvement, and to provide reference materials for calibration of secondary testing methods). Over a period of years, the approach has enabled the group of laboratories to document improved analytical performance (i.e., reduced within- and between-laboratory variation) of chemical reference methods used as the primary reference for calibration of high-speed electronic milk-testing equipment. An annual meeting of the laboratory technicians allows for review of results and discussion of each method and provides a forum for communication of experience and techniques that are of value to new analysts in the group. The monthly proficiency testing sample exchanges have the added benefit of producing all-laboratory mean reference values for a set of 14 milks that can be used for calibration, evaluation, and troubleshooting of calibration adjustment issues on infrared milk analyzers.

Modification of the Kjeldahl noncasein nitrogen method to include bovine milk concentrates and milks from other species.

The objective of our research was to modify the current indirect casein method for bovine milk to enable it to be applied to bovine milk, bovine milk concentrates, and milks of other species that contain a protein concentration up to 9% (wt/wt). Our work used a series of bovine milk concentrates from about 3 to 9% protein with the same casein as a percentage of true protein to determine the amount of buffer required and pH of the noncasein nitrogen (NCN) filtrate to achieve consistent estimates of casein and casein as percent of true protein. As the concentration of protein in milk increased (either in bovine milk concentrates or in milks of other species), the amount of buffer needed for the NCN sample preparation method to achieve a filtrate pH of 4.6 increased. In the first part of the study using a series of bovine milk concentrates, it was demonstrated that the method gave more consistent predictions of casein as a percentage of true protein when the final NCN filtrate pH was between 4.5 and 4.6 at 38°C. When the amount of buffer added to the sample was not sufficient (i.e., the filtrate pH was too high), the filtrates were not clear. A polynomial equation was developed for prediction of the amount of acetic acid or sodium acetate buffer required to achieve pH 4.5 to 4.6 for milk protein concentrations from 3 to 9% protein using bovine milk and milk concentrates. When the equation developed using cow milk was applied to goat, sheep, and water buffalo milks, it correctly predicted the volume of reagents needed to achieve a final NCN filtrate pH of 4.6 at 38°C. We also verified as part of this work that the ability to measure NPN content of milk was not influenced by protein content of milk in the range from 3 to 9% protein. The results of this study will be used as the basis for proposed changes in the official methods for measurement of the casein content of milk to expand the scope of the method so it can be used to achieve accurate results for milk concentrates and milks of other species.

Effect of microbiological testing on subsequent mid-infrared milk component analysis of the same milk sample.

Our objectives were to determine if mixing and sampling of a raw milk sample at 4°C for determination of total bacteria count (TBC) and if incubation at 14°C for 18h and sampling for a preliminary incubation (PI) count influenced the accuracy of subsequent fat, protein, or lactose measurement by mid-infrared (IR) analysis of milk from the same sample container due to either nonrepresentative sampling or the presence of microbial metabolites produced by microbial growth in the milk from the incubation. Milks of 4 fat levels (2.2, 3, 4, and 5%) reflected the range of fat levels encountered in producer milks. If the portion of milk removed from a cold sample was not representative, then the effect on a milk component test would likely be larger as fat content increases. Within the milks at each fat level, 3 treatments were used: (1) 20 vials of the same milk sampled for testing TBC using a BactoScan FC and then used for a milk component test; (2) 20 vials for testing TBC plus PI count followed by component test; and (3) 20 vials to run for IR component test without a prior micro sampling and testing. This was repeated in 3 different weeks using a different batch of milk each week. No large effect on the accuracy of component milk testing [IR fat B (carbon hydrogen stretch) and fat A (carbonyl stretch)] due to the cold milk sample handling and mixing procedures used for TBC was detected, confirming the fact that the physical removal of milk from the vial by the BactoScan FC (Foss Electric, Hillerød, Denmark) was a representative portion of the milk. However, the representativeness of any other sampling procedure (manual or automated) of a cold milk sample before running milk component testing on the same container of milk should be demonstrated and verified periodically as a matter of routine laboratory quality assurance. Running TBC with a BactoScan FC first and then IR milk analysis after had a minimal effect on milk component tests by IR when milk bacteria counts were within pasteurized milk ordinance limits of <100,000 cfu/mL. Running raw milk PI counts (18h of incubation at 13-14°C) with the BactoScan FC before milk component testing by IR milk analysis had an effect on component tests. The effect was largest on fat test results and would decrease the accuracy of milk payment testing on individual producer milks. The effect was most likely due to the absorption of light by bacterial metabolites resulting from microbial growth or other chemical degradation processes occurring in the milk during the PI count incubation, not by the sampling procedure of the BactoScan. The direction of the effect on component test results will vary depending on the bacteria count and the type of bacteria that grew in the milk, and this could be different in every individual producer milk sample.

Development of the nasal chemosensory organs in two terrestrial anurans: the directly developing frog, Eleutherodactylus coqui (Anura: Leptodactylidae), and the metamorphosing toad, Bufo americanus (Anura: Bufonidae).

Nearly all vertebrates possess an olfactory organ but the vomeronasal organ is a synapomorphy for tetrapods. Nevertheless, it has been lost in several groups of tetrapods, including aquatic and marine animals. The present study examines the development of the olfactory and vomeronasal organs in two terrestrial anurans that exhibit different developmental modes. This study compares the development of the olfactory and vomeronasal organs in metamorphic anurans that exhibit an aquatic larva (Bufo americanus) and directly developing anurans that have eliminated the tadpole (Eleutherodactylus coqui). The olfactory epithelium in larval B. americanus is divided into dorsal and ventral branches in the rostral and mid-nasal regions. The larval olfactory pattern in E. coqui has been eliminated. Ontogeny of the olfactory system in E. coqui embryos starts to vary substantially from the larval pattern around the time of operculum development, the temporal period when the larval stage is hypothesized to have been eliminated. The nasal anatomy of the two frogs does not appear morphologically similar until the late stages of embryogenesis in E. coqui and the terminal portion of metamorphosis in B. americanus. Both species and their respective developing offspring, aquatic tadpoles and terrestrial egg/embryos, possess a vomeronasal organ. The vomeronasal organ develops at mid-embryogenesis in E. coqui and during the middle of the larval period in B. americanus, which is relatively late for neobatrachians. Development of the vomeronasal organ in both frogs is linked to the developmental pattern of the olfactory system. This study supports the hypothesis that the most recent common ancestor of tetrapods possessed a vomeronasal organ and was aquatic, and that the vomeronasal organ was retained in the Amphibia, but lost in some other groups of tetrapods, including aquatic and marine animals.