Early and advanced stages of Maillard reaction in infant formulas: Analysis of available lysine and carboxymethyl-lysine
Early and advanced stages of Maillard reaction in infant formulas: Analysis of available lysine and carboxymethyl-lysine
Kataneh AalaeiID 0 1
Ingegerd Sj o?holm 0 1
Marilyn Rayner 0 1
Cristina Teixeira 1
Eden Tareke 1
0 Department of Food Technology, Engineering and Nutrition, Lund University, Lund, Sweden, 2 Food for Health Science Centre, Lund University , Medicon Village, Lund , Sweden
1 Editor: Joseph Banoub, Fisheries and Oceans Canada , CANADA
Although the literature on the Maillard reaction in infant formulas is extensive, most studies have focused on model systems, and in only a few cases on real food systems. Therefore, the objective of the present study was to determine the status of the Maillard reaction, both the early and advanced phases, in a variety of commercial infant formulas available on the Swedish market. Ten powder and liquid milk-based infant formulas from three manufacturers were selected to determine available lysine and CML contents, the two established indicators of the reaction. The products were also characterized with respect to protein content, carbohydrates composition, water content and water activity. In order to be able to compare the impact of different processing steps applied on powder and liquid formulas, the solid formulas contained similar ingredients as their corresponding liquid ones. Our findings showed that powder and liquid formulas contained similar available lysine concentrations regardless of the manufacturer, showing 27.14-36.57% decrease in the available lysine, compared to the reference skim milk powder in this study. The CML concentrations were in a broad range of 68.77-507.99 mg / kg protein. In the case of one manufacturer, liquid infant formulas had significantly higher CML content, compared to the powder products (p < 0.05). The results from this study are a step taken towards better understanding of the extent of the Maillard reaction in real complex systems of infant formulas.
Data Availability Statement: All relevant data are
within the manuscript.
Funding: The authors greatly appreciate the
Swedish Research Council for funding this study.
Competing interests: The authors have declared
that no competing interests exist.
Infant formulas (IF) developed as substitutes for human milk are usually based on cow?s milk
and cow?s milk-derived ingredients. They are mainly composed of milk proteins (casein and
whey proteins), lactose, vegetable oils, starch, and cereal proteins such as wheat, oat, maize as
well as minerals and vitamins to meet the nutritional requirements of infants. The difference
between cows? and humans? milk is in the higher lactose content of the latter (7.1%), compared
to the corresponding value for cow?s milk (4.6%). Furthermore, the composition of human
milk protein is 40% casein and 60% whey proteins, compared to cow?s milk protein which
contains 80% casein and 20% whey proteins [
]. These differences are the main criteria taken into
consideration for the formulation of IFs.
To ensure bacteriological safety of the IFs, these products are exposed to a wide variety of
heat treatments including pasteurization, in bottle sterilization (110?C, 10 min), sterilization
through ultra-high temperature processing (UHT) (130?140?C, 3?6 s), spray-drying (inlet
temperature of 170?200?C) [
], and any combination there of. Therefore, due to their composition
and exposure to the rigorous thermal processing, these products are more prone to the chemical
reactions, among which non-enzymatic browning or the Maillard reaction has captured
considerable attention [
]. This reaction involves reducing sugars and amino acids in a complex
network of reaction pathways which leads to the formation of dietary advanced glycation end
products (AGEs) [
]. AGEs are a group of molecules that when binding with the receptor for
advanced glycation end products (RAGE), the main receptor of AGEs in the body, potentially
activate a pro-inflammatory status, which in long term may lead to the development of the
chronic inflammatory diseases such as diabetes, renal disease, and coronary heart disease [
There are studies showing that processing of the IFs, especially the application of
sterilization and drying induce the Maillard or the so called non-enzymatic glycosylation reaction in
these products [
]. To understand the occurrence and progression of these reactions, different
markers have been the focus of several studies. Lysine, the essential amino acid is one of the
main reactants which become unavailable or blocked in the beginning of the reaction via
interaction with lactose. Histidine with its imidazole group, and arginine with a guanidine group
are the other reactants [
]. Therefore available lysine, the established marker of early stage of
the Maillard reaction has been utilized in several studies [
There are also studies in the literature focusing on the advanced stage of the Maillard
reaction and formation of AGEs in IFs through analysis of the related indicators such as
carboxymethyl-lysine (CML) and pyrraline [
]. However, research on both the early and
advanced stages of the Maillard reaction in real complex systems of infant formulas are scarce.
The objective of the present study was thus, to determine the status of the Maillard reaction,
both the early and advanced phases, in selected commercial infant formulas available on the
market. Thereby, ten powder and liquid milk-based infant formulas from three manufacturers
on the Swedish market were chosen and were analyzed using our previously validated
] to determine both the available lysine and CML contents. The aim was to
compare powder and liquid infant formulas, and to improve our understanding of the impact of
different processing steps applied on powder and liquid IFs on the decrease of available lysine
and formation of CML as a well-characterized AGE molecule.
Materials and methods
The study was carried out using four powder and six liquid infant formulas from three
manufacturers A, B and C (Table 1). Powder samples 1?4 contained similar ingredients as liquid
samples 5?8 respectively which made it possible to compare the impact of processing on the
progression of the Maillard reaction in these products. The common ingredients were skim
milk powder, whey powder, vegetable oil, starch, maltodextrin, vitamins, minerals and either
oat or maize were the other protein source. All the products had been in the first third of their
best before time period at the time of purchase, and had significant time (6?9 months) until
the best before date written on the packages.
Determination of protein. Protein contents of the products were determined using a
protein analyzer, Thermo Fisher Scientific Flash EA 1112 N series (Waltham, MA, USA), which
operated according to the combustion technique, also known as the modified Dumas method.
2 / 10
a Powder infant formulas were vacuum packed after purchasing and placed into freezer -20?C until analysis.
bLiquid formulas were freeze-dried (Labconco, Missouri, USA), vacuum packed and put into freezer -20?C before further analysis.
The instrument determined the nitrogen concentration and the relative protein content. The
sample weight was 50 mg and the analysis was carried out in duplicate.
Determination of carbohydrates. Samples were analyzed for the determination of
glucose, fructose, lactose and maltose. The analysis was carried out in duplicate using AOAC
Determination of water content. Water content (or the dry matter) of the samples were
determined using the standard method of the International Dairy Federation [
samples were dried in an oven (Termaks, Bergen, Norway) at 102?C for 2 hours. Following
this, they were kept in a desiccator for one hour to reach the constant weight. 1 g of sample
was utilized for each replicate and this analysis was carried out with two replicates.
Measurement of water activity (aw). Water activity of the samples were measured using
Aqua-lab water activity meter Series 3TE (Decagon Devices Inc., Washington, USA). 1 g of
sample was used for each measurement and the analysis was performed with two replicates.
Determination of available lysine. The dye-binding method with Acid-orange 12 was
employed for the determinations of available lysine is fully described and validated in our
previous work [
] and is based on the work presented by Hurrell, Lerman & Carpenter, with slight
]. The method was successfully applied on a variety of skim milk powders in
our previous work [
] and verified to be suitable for quantification of available lysine.
Briefly, 300 mg of sample were mixed with 2 ml sodium acetate solution for 20 min on an
orbital lab shaker (type 3005, GFL, Gesellschaft fu?r Labortechnik, Burgwedel, Germany).
Then, 0.2 ml propionic anhydride were added to half of the flasks and the mixing was
continued for another 20 min. Subsequently, 40 ml of the dye solution (with concentration of 1.36
mg/ml) were added and mixed for 1h. The analysis continued with the centrifugation step
(5000 rpm, 10 min) by an Optima LE-80 k ultracentrifuge (Beckman Coulter, Bromma,
Sweden) and the absorbance at 475 nm of the supernatants were measured using a Varian Cary
50 UV-Vis spectrophotometer (Agilent Technologies, CA, USA) and concentration of the
available lysine was calculated using the equation from the calibration curve y = 47.37 x + 0.01
(R2 = 0.9999). The analysis was carried out with three replicates.
Determination of CML. The samples were prepared by hydrolyzing 0.3 g sample for 24 h
at 110?C using 2 ml 6M HCl, together with an isotope labelled d4-CML (Larodan Fine
3 / 10
Chemicals AB, Malmo? Sweden) as the internal standard. To protect samples from oxidation
during hydrolysis, the test tubes with samples in HCl were flushed with nitrogen (bubbling
nitrogen gas through the sample) and sealed before incubation. Following hydrolysis,
extraction of CML was performed using solid phase extraction (TelosneoPCX, Teknolab Sorbent
AB, Va?stra Fro?lunda, Sweden). The SPE extracts were then evaporated to dryness,
reconstituted in 200 ?l 5mM NFPA in H2O and analyzed using LC-MS/MS [
The quantification of CML was performed using high-pressure liquid chromatography
mass spectrometry (HPLC-MS/MS). Instruments used were Accela UHPLC pump with
autoinjector coupled to LTQ VelosPro Orbitrap massspectrometer (Thermo Scientific, Waltham,
USA). For data quantification and evaluation XcaliburTM 2.2 software (Thermo Scientific)
was used. The MS/MS was run in positive electrospray ionization ion trap mode, detecting two
Selected Reaction Monitoring (SRM) transitions for CML and two for the internal standard.
Solid phase extraction, chromatographic parameters, ion source parameter and the SRM
transitions are the same as described by Tareke et al., 2013.
The calibration curve was prepared with 0, 1, 2.5, 10, 50, 100 and 500 ?g/ml CML and fixed
amounts (100?l) of IS (d4-CML) with 100 ?g/ml concentration (R2 = 0.9991). Samples
containing higher concentrations of CML than the highest concentration in the calibration curve
were diluted to ensure the analysis was carried out over the calibrated concentration range.
Analysis of available lysine were conducted with three replicates, the determination of CML
was performed using two replicates, and the results are reported as mean ? standard error.
Student?s t-test (two samples assuming unequal variances) at 95% significance level was used to
assess statistical significance of the data.
Results and discussion
Protein contents of infant formulas
Protein contents of the products measured using protein analyzer, are presented in Table 2.
Composition of carbohydrates
In Table 3, results of sugar analysis are presented.
Among the analyzed carbohydrates there is a large variation on the content of each sugar
among samples (glucose: <0.04?2.67 g/100g, lactose: 10.8?31.8 g/100g, maltose: < 0.04?14.0
g/100g), independent of being a powder or liquid formulation, but seemingly more related
with the manufacturer. Within the sugars analyzed it was not possible to find a relation with
the content of CML. However, it does not disregard the possibility that free galactose could
contribute to the formation of CML, but this was not possible to access with the method of
analysis used and remains to be investigated.
Water content and water activity
The infant formulas were also analyzed with respect to water content and water activity. These data
were utilized in the further calculations regarding available lysine and CML contents (Table 4).
Water is a crucial factor in the progression of the Maillard reaction with its impact on glass
transition temperature. Glass transition is the phase in which the lactose is transformed from a
glassy and stable state to a rubbery state, where the reaction has the highest rate [
Furthermore, water may be produced with the progression of the Maillard reaction, and this leads to
the increase in the water activity and water content of the product [
]. It was shown that the
reaction has the highest rate at intermediate aw 0.5?0.7 [
Indication of early stage of Maillard reaction in available lysine
To evaluate whether the reaction is initiated, available lysine may provide reliable information.
Traditionally, the Maillard reaction is divided into 3 steps. The interaction of lactose with
5 / 10
Available lysine % in dry matter (n = 3) ? SD
0.95 ? 0.06
1.13 ? 0.06
0.96 ? 0.02
1.02 ? 0.02
Available lysine % in protein (n = 3) ? SD
8.13 ? 0.05
9.75 ? 0.49
7.74 ? 0.18
8.28 ? 0.16
lysine and the formation of the Amadori product is regarded as the early stage of the reaction.
Degradation of the Amadori product and the formation of molecules such as CML,
hydroxymethyl furfural (HMF), ?-pyranone, 3-furanone, reductones, ?-dicarbonyls, cyclopentenone,
galactosylisomaltol and acetylpyrrole is considered as the advanced stage of the reaction and,
finally, the generation of Melanoidins is the final stage [
]. In this study available lysine served
as an indicator; to compare powder and liquid infant formulas and to assess the effect of
different processing (spray-drying in case of powder and sterilization for the liquid formulas) on the
commencement of the reaction. It has been reported that sterilization and spray-drying induce
the Maillard reaction during the processing of infant formulas [
]. Table 5 provides the
results of available lysine, both presented based on dry matter and protein.
Since in our previously published works available lysine contents have been reported based
on dry matter [
], we use this unit for further discussion. Nevertheless, available lysine
contents are also reported based on protein content to make it easy to compare with other studies.
The conclusions can be drawn from these results is that powder and liquid infant formulas
contain similar available lysine concentrations regardless of the manufacturer. Furthermore,
considering that the concentrations of available lysine in these products are in the range 0.95?
1.28% in dry matter, the infant formulas show 27.14?36.57% decrease in the available lysine,
compared to the reference skim milk powder sample. The reference sample in this study was
skim milk powder which was not exposed to any heat-treatment, the skim milk was collected
after the fat separation step and freeze-dried. This sample contained 3.54% ? 0.09 available
lysine based on dry matter.
Birlouez-Aragon et al. also showed that there was no significant difference between the
liquid and powder infant formulas with respect to available lysine. They reported a decrease of
20% in the available lysine compared to raw cow?s milk [
]. Our results are slightly higher than
the results reported by Contreras-Calderon. They showed that the average of available lysine
in the infant formulas was 4.03?6.43% based on protein [
]. This can be explained by
difference in the composition and heat processing applied on the products as well as the method
that was employed for the determination of available lysine. In another study the available
lysine loss as a result of processing in infant cereals ranged between 14?29% [
Ferrer et al. reported a loss of 15.2?26.7% in the infant formulas, in comparison with the raw
cow?s milk [
]. What can be understood from the present and also the previous studies is that
infant formulas are more liable to the Maillard reaction and its consequences due to their
composition, the presence of other provoking ingredients such as iron and vitamin C, and the
application of rigorous heat treatments such as sterilization. In other words in order to mimic
6 / 10
Fig 1. Studying the advanced stage of the reaction in the infant formulas using the indicator CML. CML Concentration is Calculated in mg / kg Protein (n = 2). The
Error Bars Represent the Standard Errors. The Infant Formulas are from Three Manufacturers A, B and C. P Stands for Powder and L Means Liquid.
the composition of human milk (7.1% lactose, 40:60 casein to whey protein ratio) lactose and
whey powder, both Maillard potential stimulants, are added to cow?s milk. Furthermore, to
guarantee the microbiological safety of the products aimed for this sensitive group of
population the heat applied in the form of in-bottle sterilization or spray-drying is more rigorous [
]. The substantial effect of heat processing such as sterilization on the Maillard reaction has
been well discussed in the literature [
1, 10, 26
]. Addition of vitamin C (one of the proposed
pathways for formation of CML) and poly unsaturated fatty acids in the form of vegetable oils
may lead to further progression of the Maillard reaction .
Determination of CML
After studying the early phase of the reaction i.e., the study of the reaction precursors, it was
logical to try to understand how the reaction status was in the advanced stage. The importance
of this part of the study was not only to gain insight into the advanced phase of the Maillard
reaction, but also the role of CML as an identified AGE molecule, as AGEs have been
associated with risk factors for diabetes complications and other autoimmune diseases [
was analyzed using the procedure established by Tareke et al. 2013 with two replicates. The
analytical performance was assessed by considering the accuracy of the LC-MS/MS
measurements using a standard sample with 10 ?g/ml CML which was 115 ? 4%. The results of CML
concentration in the 10 infant formulas are presented in Fig 1.
Unlike analysis of available lysine in these products (section 3.3), where the available lysine
content of the infant formulas were quite similar, the CML concentration are in a broad range
of 68.77?507.99 mg / kg protein. In the case of manufacturer A, the liquid infant formulas had
7 / 10
significantly higher CML content compared to the powder products (p < 0.05). This may be
explained by difference in handling/processing procedure applied on the products. This
should be further investigated in the future.
Our results are in agreement with the previous studies. Hartkopf et al. reported that CML
concentration was 50?200 mg/kg protein in the infant formulas [
]. In another study it was
shown that 16 powder infant formulas contained 50?700 mg CML in kg protein. It was also
concluded that CML concentration in the infant formulas were 28?389 times higher than the
fresh human milk [
]. In another study by Dittrich et al. CML contents of 8 infant formulas
available on the German market was reported to be 514?11372 ng / ml and 35 times higher
than the CML concentration in the human milk. [
]. Similarly, Troise et al. reported the
CML concentration of 80?140 mg /kg protein regarding commercial powder infant formulas
]. As it was explained before, infant formulas show higher sensitivity towards the Maillard
reaction, and this is mainly due to their composition which mimics the composition of human
milk as well as the application of rigorous heat treatments. AGE content of these products may
increase even more during storage, considering that these products mostly have prolonged
shelf-life. It was shown in our previous study with skim milk powder that concentration of
CML increased several times during storage at realistic storage conditions [
]. With respect
to the spray-dried skim milk powder stored at 52% RH, 30?C after 30 days, the CML
concentration reached approximately 3000 mg/kg protein and 24500 mg/kg protein after 200 days.
This was, however, the accelerated storage condition. When the samples of skim milk powders
were stored at 33% RH, 30?C, concentration of the CML after 30 and 200 days storage was
around 400 and 2500 mg/kg protein respectively, indicating the importance of controlling
every individual parameter such as temperature and RH throughout the storage. This is
something that should be the priority of the future studies with respect to infant formulas.
In summary, status of the Maillard reaction both the early and advanced phases of the reaction,
were investigated in this study regarding commercial infant formulas available on the Swedish
market. Ten powder and liquid infant formulas were analyzed with respect to available lysine,
the indicator of early stage of the reaction, and CML as a well-characterized AGE molecule, a
marker of the advanced phase of the Maillard reaction. The powder and liquid formulas
contained similar available lysine contents in the range of 0.95?1.28% dry matter. Regarding
CML, the concentrations were in a wide range of 68.77?507.99 mg/kg protein. In other words
CML content of infant formulas available on the Swedish market can vary 7 fold, depending
on the physical form and the manufacturers. Due to the higher liability of infant formulas to
the Maillard reaction and sensitivity of the target group (infants), impact of the storage of
these products at realistic storage conditions should be the priority of future studies.
Conceptualization: Kataneh Aalaei, Ingegerd Sjo?holm, Marilyn Rayner.
Data curation: Kataneh Aalaei, Cristina Teixeira.
Formal analysis: Kataneh Aalaei.
Funding acquisition: Marilyn Rayner.
Investigation: Kataneh Aalaei, Cristina Teixeira.
Methodology: Kataneh Aalaei, Cristina Teixeira.
Project administration: Ingegerd Sjo?holm, Marilyn Rayner.
8 / 10
Resources: Ingegerd Sjo?holm.
Supervision: Ingegerd Sjo?holm, Marilyn Rayner, Eden Tareke.
Validation: Kataneh Aalaei, Ingegerd Sjo?holm, Marilyn Rayner, Eden Tareke.
Visualization: Kataneh Aalaei.
Writing ? original draft: Kataneh Aalaei.
Writing ? review & editing: Kataneh Aalaei, Ingegerd Sjo?holm, Marilyn Rayner, Eden
9 / 10
1. Pischetsrieder M , Henle T. Glycation products in infant formulas: chemical, analytical and physiological aspects . Amino Acids . 2012 ; 42 ( 4 ): 1111 - 8 . https://doi.org/10.1007/s00726-010 -0775-0 PMID: 20953645
2. Birlouez-Aragon I , Pischetsrieder M , Lecl?re J , Morales FJ , Hasenkopf K , Kientsch-Engel R , et al. Assessment of protein glycation markers in infant formulas . Food Chemistry . 2004 ; 87 ( 2 ): 253 - 9 . http:// dx.doi.org/10.1016/j.foodchem. 2003 . 11 .019.
3. Van Boekel MAJS. Effect of heating on Maillard reactions in milk . Food Chemistry . 1998 ; 62 ( 4 ): 403 - 14 . https://doi.org/10.1016/S0308- 8146 ( 98 ) 00075 - 2 .
4. Van Boekel MAJS. Kinetic aspects of the Maillard reaction: a critical review . Food / Nahrung . 2001 ; 45 ( 3 ): 150 - 9 . https://doi.org/10.1002/ 1521 - 3803 ( 20010601 )45: 3 < 150 : : AID -FOOD150>3.0.CO;2-9 PMID: 11455780
5. Poulsen MW , Hedegaard RV , Andersen JM , de Courten B , Bu?gel S , Nielsen J , et al. Advanced glycation endproducts in food and their effects on health . Food and Chemical Toxicology . 2013 ; 60 : 10 - 37 . https://doi.org/10.1016/j.fct. 2013 . 06 .052 PMID: 23867544
6. Aalaei K , Rayner M , Tareke E , Sjo ?holm I. Application of a dye-binding method for the determination of available lysine in skim milk powders . Food Chemistry . 2016 ; 196 : 815 - 20 . https://doi.org/10.1016/j. foodchem. 2015 . 10 .004 PMID: 26593559
7. Albala ? -Hurtado S , Bover-Cid S , Izquierdo-Pulido M , Veciana-Nogue?s MT , Vidal-Carou MC . Determination of available lysine in infant milk formulae by high-performance liquid chromatography . Journal of chromatography A . 1997 ; 778 ( 1-2 ): 235 - 41 . PMID: 9299737
8. Delgado-Andrade C , Rufia?n- Henares JA , Morales FJ . Lysine availability is diminished in commercial fibre-enriched breakfast cereals . Food Chemistry . 2007 ; 100 ( 2 ): 725 - 31 . http://dx.doi.org/10.1016/j. foodchem. 2005 . 10 .031.
9. Fernandez-Artigas P , Garcia-Villanova B , Guerra-Hernandez E . Blockage of available lysine at different stages of infant cereal production . Journal of the Science of Food and Agriculture . 1999 ; 79 ( 6 ): 851 - 4 . https://doi.org/10.1002/(sici) 1097 - 0010 ( 19990501 )79: 6 < 851 : :aid-jsfa298>3.0 .co;2- u
10. Ferrer E , Alegr??a A , Farre? R, Abella?n P , Romero F. Effects of Thermal Processing and Storage on Available Lysine and Furfural Compounds Contents of Infant Formulas . Journal of agricultural and food chemistry . 2000 ; 48 ( 5 ): 1817 - 22 . https://doi.org/10.1021/jf991197l PMID: 10820100
11. Mehta BM , Deeth HC . Blocked Lysine in Dairy Products: Formation, Occurrence, Analysis, and Nutritional Implications . Comprehensive Reviews in Food Science and Food Safety . 2016 ; 15 ( 1 ): 206 - 18 . https://doi.org/10.1111/ 1541 - 4337 . 12178
12. Ram ??rez-Jime?nez A, Garc??a- Villanova B , Guerra-Herna? ndez E. Effect of storage conditions and inclusion of milk on available lysine in infant cereals . Food Chemistry . 2004 ; 85 ( 2 ): 239 - 44 . http://dx.doi.org/ 10.1016/j.foodchem. 2003 . 06 .017.
13. Schmitz I , Gianfrancesco A , Kulozik U , Foerst P. Influence of temperature and the physical state on available lysine in powdered infant formula . Procedia Food Science . 2011 ; 1 ( 0 ): 1031 - 8 . http://dx.doi. org/10.1016/j.profoo. 2011 . 09 .154.
14. Dittrich R , Hoffmann I , Stahl P , Mu?ller A , Beckmann MW , Pischetsrieder M. Concentrations of N?-Carboxymethyllysine in Human Breast Milk, Infant Formulas, and Urine of Infants . Journal of Agricultural and Food Chemistry . 2006 ; 54 ( 18 ): 6924 - 8 . https://doi.org/10.1021/jf060905h PMID: 16939359
15. Drusch S , Faist V , Erbersdobler HF . Determination of N -carboxymethyllysine in milk products by a modified reversed-phase HPLC method . Food Chemistry . 1999 ; 65 ( 4 ): 547 - 53 . http://dx.doi.org/10. 1016/S0308- 8146 ( 98 ) 00244 - 1 .
16. Hull GLJ , Woodside JV , Ames JM , Cuskelly GJ . N?-(carboxymethyl)lysine content of foods commonly consumed in a Western style diet . Food Chemistry . 2012 ; 131 ( 1 ): 170 - 4 . http://dx.doi.org/10.1016/j. foodchem. 2011 . 08 .055.
17. Plaza M , O? stman E , Tareke E. Maillard Reaction Products in Powder Based Food for Infants and Toddlers . European Journal of Nutrition and Food Safety 2016 ; 6 ( 2 ): 65 - 74 .
18. Tareke E , Forslund A , Lindh CH , Fahlgren C , O?stman E. Isotope dilution ESI-LC-MS/MS for quantification of free and total N?-(1- Carboxymethyl ) -l-Lysine and free N?-(1- Carboxyethyl)- l-Lysine: Comparison of total N?-(1- Carboxymethyl ) -l-Lysine levels measured with new method to ELISA assay in gruel samples . Food Chemistry . 2013 ; 141 ( 4 ): 4253 - 9 . https://doi.org/10.1016/j.foodchem. 2013 . 07 .003 PMID: 23993613
19. Ru?ckold S , Grobecker KH , Isengard HD . Determination of the contents of water and moisture in milk powder . Fresenius' Journal of Analytical Chemistry . 2000 ; 368 ( 5 ): 522 - 7 . https://doi.org/10.1007/ s002160000511
20. Hurrell RF , Lerman P , Carpenter KJ . Reactive lysine in foodstuffs as measured by a rapid dye-binding procedure . Journal of Food Science . 1979 ; 44 ( 4 ): 1221 - 7 . https://doi.org/10.1111/j.1365- 2621 . 1979 . tb03485.x
21. Aalaei K , Rayner M , Sjo?holm I. Storage stability of freeze-dried, spray-dried and drum-dried skim milk powders evaluated by available lysine . LWT-Food Science and Technology . 2016 ; 73 : 675 - 82 . https:// doi.org/10.1016/j.lwt. 2016 . 07 .011.
22. Roos YH . Glass transition temperature and its relevance in food processing . Annual Review of Food Science and Technology . 2010 ; 1 : 469 - 96 . https://doi.org/10.1146/annurev.food. 102308 .124139 PMID: 22129345
23. O 'Brien J. Non -enzymatic degradation pathways of lactose and their significance in dairy products . Advanced dairy chemistry . 2009 : 231 - 94 .
24. Labuza TP , Saltmarch M. Kinetics of Browning and Protein Quality Loss in Whey Powders During Steady State and Nonsteady State Storage Conditions . Journal of Food Science . 1982 ; 47 ( 1 ): 92 - 6 . https://doi.org/10.1111/j.1365- 2621 . 1982 .tb11035.x
25. Contreras-Calder O?N J , Guerra-Hern A ? Ndez EJ , Garc?IA-Villanova B . Modification of Nutritional Value During Storage of Infant Formulas Elaborated with Different Intact and Partially Hydrolyzed Proteins and Carbohydrates . Vitae. 2015 ; 22 : 87 - 100 .
26. Nguyen HT , van der Fels-Klerx HJ , van Boekel MAJS. N -(carboxymethyl)lysine: A Review on Analytical Methods, Formation, and Occurrence in Processed Food, and Health Impact . Food Reviews International. 2014 ; 30 ( 1 ): 36 - 52 . https://doi.org/10.1080/87559129. 2013 .853774
27. Hillman M , Westro?m B , Aalaei K , Erlanson-Albertsson C , Wolinski J , Lozinska L , et al. Skim milk powder with high content of Maillard reaction products affect weight gain, organ development and intestinal inflammation in early life in rats . Food and Chemical Toxicology . 2019 ; 125 : 78 - 84 . https://doi.org/10. 1016/j.fct. 2018 . 12 .015 PMID: 30553875
28. Hartkopf J , Pahlke C , Lu?demann G , Erbersdobler HF . Determination of N? -carboxymethyllysine by a reversed-phase high-performance liquid chromatography method . Journal of Chromatography A . 1994 ; 672 ( 1 ): 242 - 6 . http://dx.doi.org/10.1016/ 0021 - 9673 ( 94 ) 80613 - 6 .
29. S?ebekova? K, Saavedra G , Zumpe C , Somoza V , Klenovicsova ? K, Birlouez-Aragon I . Plasma Concentration and Urinary Excretion of N?-(Carboxymethyl)lysine in Breast Milk-and Formula-fed Infants . Annals of the New York Academy of Sciences . 2008 ; 1126 ( 1 ): 177 - 80 . https://doi.org/10.1196/annals. 1433.049 PMID: 18448813
30. Troise AD , Fiore A , Wiltafsky M , Fogliano V . Quantification of N?-(2- Furoylmethyl ) -l-lysine (furosine), N?- ( Carboxymethyl ) -l-lysine (CML), N?- ( Carboxyethyl ) -l-lysine (CEL) and total lysine through stable isotope dilution assay and tandem mass spectrometry . Food Chemistry . 2015 ; 188 : 357 - 64 . https://doi. org/10.1016/j.foodchem. 2015 . 04 .137 PMID: 26041204
31. Aalaei K , Sjo?holm I, Rayner M , Tareke E. The Impact of Different Drying Techniques and Controlled Storage on the Development of Advanced Glycation End Products in Skim Milk Powders Using Isotope Dilution ESI-LC-MS/MS . Food and Bioprocess Technology . 2017 . https://doi.org/10.1007/s11947-017- 2022-0