Determination of Bovine Lactoferrin in Powdered Infant Formula and Adult Nutritionals by Heparin Affinity Extraction and Reverse-Phase High-Performance Liquid Chromatography/Ultraviolet Detection (HPLC/UV): Single-Laboratory Validation, First Action 2021.10

Abstract Background Infant formulas, and pediatric and adult nutritional products, are being fortified with bovine lactoferrin (bLF) due to its beneficial impacts on immune development and gut health. Lactoferrin supplementation into these products requires an analytical method to accurately quantify the concentrations of bLF to meet global regulatory and quality standards. Objective To develop and validate a lactoferrin method capable of meeting the AOAC INTERNATIONAL Standard Method Performance Requirements (SMPR®) 2020.005. Methods Powder formula samples are extracted using warm dibasic phosphate buffer, pH 8, then centrifuged at 4°C to remove insoluble proteins, fat, and other solids. The soluble fraction is further purified on a HiTrap heparin solid-phase extraction (SPE) column to isolate bLF from interferences. Samples are filtered, then analyzed by LC–UV using a protein BEH C4 analytical column and quantitated using an external calibrant. Results The LOQ (2 mg/100 g), repeatability (RSD: 2.0–4.8%), recovery (92.1–97.7%), and analytical range (4–193 mg/100 g) all meet the method requirements as stated in SMPR 2020.005 for lactoferrin. Conclusion The reported single-laboratory validation (SLV) results demonstrate the ability of this lactoferrin method to meet or exceed the method performance requirements to measure soluble, intact, non-denatured bLF in infant and adult nutritional powder formulas. Highlights The use of a heparin affinity column to isolate lactoferrin from bovine milk products combined with a selective analytical chromatographic column provides suitable analyte specificity without requiring proprietary equipment or reagents.

Lactoferrin is a human milk glycoprotein shown to have beneficial physiological effects, including modulating immune functions, protective activities against pathogens, and supporting intestinal health (1).Lactoferrin is also naturally present in bovine milk and some infant formulas, though innate concentrations are much lower than human milk.Since bLF has a reported bioactivity comparable to human lactoferrin (1, 2) and is commercially available as a purified ingredient, it could be used to fortify infant formula and adult nutritionals.For this reason, assays to detect and quantify bovine lactoferrin (bLF) have been developed utilizing an array of techniques, including ELISA (3, 4), UHPLC-MS/MS (5, 6), HPLC coupled with a heparin affinity column (7), and surface plasmon resonance (SPR) optical biosensor immunoassay (8, 9).The method described here has been optimized and validated to measure bLF in powdered infant formulas and adult nutritionals and is applicable to matrixes containing only lactoferrin of bovine origin.The previous version of this method was internally developed and validated at Reckitt/Mead Johnson Nutrition, Evansville, IN for milk-based infant formula matrixes.As new matrixes were introduced (e.g., partially hydrolyzed milk-based powder), modifications were made to maintain consistent bLF method performance and are summarized in Table 1.

Single-Laboratory Validation
A single-laboratory validation (SLV) was conducted to validate this bLF method against the SMPR 2020.005criteria listed in Table 2 (10).This SLV characterized the method performance for specificity, instrument response linearity, limits of detection and quantitation, repeatability, intermediate precision, and accuracy.Selectivity experiments to address heat-impacted (denatured) bLF and other milk proteins were also incorporated into the final round of SLV testing to completely evaluate SMPR 2020.005guidelines.
Due to the limited commercial availability of infant formula and adult nutritional matrixes fortified with bLF, internally sourced materials were utilized to assess method performance.
The validation samples are listed in Table 3.All sample results were analyzed as powder basis then converted to reconstituted values based on a reconstitution rate of 25 g powder þ 200 g diluent (dilution factor ¼ 9).

Specificity
(a) Chromatographic integrity.-Specificitywas determined by comparing chromatograms from a solvent blank, a lactoferrin standard, and at least one matrix blank and sample per matrix family.(b) Heat-impacted lactoferrin.-Toassess the ability to differentiate intact, soluble lactoferrin from heat-impacted lactoferrin, a 10.0 mg/mL solution of lactoferrin standard (Cerilliant, L-047) was heated to 90 � C for 60 min in duplicate on separate days.In each case, the heat-treated lactoferrin was spiked into a milk-based IF powder matrix blank at 56 mg/100g prior to sample preparation and analysis.These samples were compared to matrix blanks spiked at 56 mg/100g with unadulterated lactoferrin.(c) Other proteins.-Themolecular weights and isoelectric points for bovine milk proteins are calculated in Uniprot.These values show that bovine lactoperoxidase (LP, calculated average MW: 78.2 kDa, calculated isoelectric point (PI): 8.91) has similar physical properties to bLF (calculated average MW: 76.1 kDa, calculated PI: 8.67 ; 11, 12).The similarities suggest that lactoperoxidase may interfere with lactoferrin in the cleanup and chromatographic analysis.A solution of lactoperoxidase (Sigma-Aldrich, L2005) was analyzed both with and without the heparin column cleanup to assess the potential interference and was followed by a chromatographic evaluation.
Bovine lysozyme was also considered as a potential interference.
The lack of a commercially available standard inhibited collecting any experimental data.However, its molecular weight and isoelectric point (MW �18 kDa, PI 9.5), and relatively low content in bovine milk, suggest it is even less likely to cause any substantial interference (13)(14)(15).

Linearity
The linearity of the standard curve provides an assessment of the instrument response to the concentration of the analyte standards.The slope and intercept of the response define the standard curve and the correlation coefficient provides one assessment of the linearity.Six levels of standards were prepared and analyzed for slope, intercept, correlation, and percent deviation, to assess the accuracy of the calibration.

LOD
The practical LOD was set at one-tenth the concentration of the lowest level calibration standard and was assessed by diluting the lowest standard (10 mg/mL) by 10 and performing six replicate injections of the resulting solution.The ratio of bLF signal to baseline noise was determined for each injection.

LOQ
The LOQ for the analysis was set at the lowest standard level and assessed by performing six replicate injections of the lowest level standard.The ratio of bLF signal to baseline noise was determined for each injection.

Repeatability and Intermediate Precision
Repeatability values were obtained from six replicate data points in a single run using six independent sample preparations for each sample, and the intermediate precision of the method was determined from single data points obtained on six different days.

Method Equivalence Studies
To assess the impact of method modifications, equivalence studies were performed on two of the samples validated in the original SLV: milk-based infant formula powder and milk-based partially hydrolyzed infant formula powder.Since the matrixes had been fully validated using the same samples, the equivalency study was considered sufficient to demonstrate that the updated method produced equivalent results compared to the original version.Six Independent replicates for each matrix were prepared and analyzed using the method initially submitted to the Stakeholder Program on Infant Nutrition and Adult

Accuracy
Accuracy was tested by spiking each matrix with two levels of lactoferrin, roughly representing 50% and 100% overspike levels, using the certified reference standard as spiking agent.Recovery studies for each sample were analyzed in triplicate with individual recovery replicates performed on different days.

Analytical Range
The required analytical range coverage is 4-200 mg/100g, based on reconstituted concentrations.The analytical range of the lactoferrin method is evaluated based on the standard curve range, as well as the demonstrated repeatability, intermediate precision, and accuracy results.

A. Principle
Lactoferrin content is determined by reverse-phase (RP)-HPLC combined with a UV or a diode array detector (DAD) at 280 nm wavelength.Samples are dissolved in a sodium phosphate dibasic solution and then centrifuged to separate the liquid phase (whey) from fat and casein.bLF is then isolated and purified from the liquid phase using a HiTrap heparin affinity column as solidphase extraction (SPE).Eluates from the heparin affinity column are then filtered and collected for HPLC analysis using a protein BEH C4 analytical column.

Critical components.-(1)
A suitable bLF reference standard [see Note (2) in section E(a)] must be obtained and carefully weighed to avoid absorbed water weight.(2) The sample extract solution should be carefully handled to avoid any spills.(3) The HiTrap heparin column should not be allowed to run dry before the final bLF eluate collection.

E. Preparations of Standard Solutions
Note: Standard preparations may be scaled up or down as needed.
(a) bLF stock solution I (10 mg/mL).-Accuratelyweigh 90-120 mg bLF reference material into a 10 mL volumetric flask.Record the weight to 0.01 mg.Add 5-8 mL water and agitate gently to mix.Allow the solution to stand at room temperature for at least 1 h until dissolved.Dilute to volume and gently mix.Store at 4 � C in a sealed plastic vessel protected from light.Calculate the concentration (conc) to four significant figures as follows: Notes on bLF stock solution I: (1) bLF reference materials may be hygroscopic.Please ensure the reference standard is at room temperature before opening and work quickly when weighing out.
(2) Lactoferrin standards may be made from either certified reference standard material or from lactoferrin material that has been validated against a certified reference standard (assessed in at least triplicate over a minimum of 2 days by LC using a certified reference standard for external calibration and stored sealed in a desiccator to prevent degradation over the shelf life of the standard).(b) bLF stock solution II (�0.5 mg/mL).-Pipette1.25 mL bLF stock solution I and dilute to 25 mL with 0.05 M Na 2 HPO 4 -1 M NaCl buffer, pH 8.0 (eluting buffer) in a volumetric flask.Store at 4 � C in a sealed plastic vessel.Calculate the concentration to four significant figures as follows: where, Conc (WS) ¼ the concentration of the working standard; Conc (bLF stock II) ¼ the concentration of the bLF stock solution II; V (bLF stock II) ¼ the volume of the bLF stock solution II used to make each working standard; V (WS) ¼ the final volume of the working standard.(c) Sample purification (1) Ensure the extract solution from step F(a9) above is mixed well and then pipette 5 mL extract solution into the syringe barrel.Drain the solution through the column with vacuum and discard the eluent.Note: Ensure the total amount of lactoferrin being loaded onto the column is at least 0.1 mg.(2) Wash the column, using vacuum, with 10.0 mL 0.2 M sodium phosphate dibasic solution, pH 8.0 (loading buffer) and discard the eluent.(3) Remove the column assembly from the vacuum manifold.Elute the column with 4.50 mL 0.05 M Na 2 HPO 4 -1 M NaCl solution, pH 8.0 (eluting buffer), collecting the eluent into a 5.0 (or 10.0 mL) volumetric flask or equivalent container.(4) Dilute to volume with 0.05 M Na 2 HPO 4 -1 M NaCl solution, pH 8.0 (eluting buffer) and mix well.Note: The final volume (size of volumetric flask for collecting eluent) can be adjusted so that the bLF level falls on the standard curve.(5) Filter a portion of mixed solution through a 0.45 mm, 13 mm wwPTFE Mini syringe filter into a silanized HPLC vial for analysis.(6) After collecting the final bLF eluent from the purification process, wash the column using the vacuum manifold with 5 mL of 0.05 M Na 2 HPO 4 -2 M NaCl solution, pH 8.0 (cleanup buffer) and then 5 mL 20% ethanol.Discard eluents after washing and store the column in 20% ethanol at RT for reuse up to four times.Note: Discard column if sample matrix binds on column during purification step, indicated by significant restriction in flow.

(d) Care and use of HiTrap heparin columns as SPE
(1) Each column may be used up to four times.However, if there is difficulty loading, washing, and eluting a sample, do not use the column again.
(2) Columns are connected to the vacuum manifold using a threaded Luer adapter.Syringe barrels are connected to the tops of columns with a 1/16" male/Luer female connecter.One top connecter is included in each box of five columns.When suing a column for the first time, a plastic cap must be snapped off the bottom of the column.(3) Equilibrate the columns with 5 mL loading buffer.After loading, use 10 mL loading buffer to wash away interferences.Discard the eluent from equilibrating, loading and wash steps.(4) The recommended flow rate through the HiTrap heparin column is 0.1 to 1 mL/min for the 1 mL columns.(5) Do not let the heparin column run dry during loading and washing of the sample.If you do allow the loading buffer to run out, you can add another couple of milliliters of loading buffer and allow it to drain to the bottom of the column.(6) During loading, washing and eluting steps, the flow through the column may need to be assisted with the use of the plunger.(7) The minimum load for adequate recovery on these specific columns in this size (1 mL) is 100 mg.(8) When eluting lactoferrin, the column is removed from the vacuum manifold and eluent is collected into an appropriately sized volumetric flask.2021.10C for at least 30 min to equilibrate the system.(f) Injection sequence:-See Table 2021.10D.

Note:
The sample stability at RT has not been evaluated past 24 h.(g) End of run.-Afterall samples and standards have been analyzed, it is recommended to inject water three times: twice to clean out the UHPLC system and once to bring the instrument to standby mode.
H. System Suitability/Analytical QC (a) Calibration curve.-Everysequence must include a set of calibration standards covering the analytical range.The coefficient of determination (r 2 ) of the standard curve must be a minimum of 0.995.The measured concentration of each standard level must be within ±10% of the theoretical value.A 1/x weighting may be used if an excessive bias is apparent in the low standards.(b) Check standards.-Equilibrationinjections, followed by a set of six replicate injections of the check standard (middle standard) must be made prior to the calibration curve.The RSD, %, of the bLF peak areas for the six check standard injections must be ≤3.0%.The check standard must also be injected throughout the analytical set for bracketing, with the initial bracket standard run after the calibration standards and no more than 10 injections between bracketing check standards.The overall RSD of all the check standard peak areas must be ≤5%.Retention times of bracketing standards must be within ±5% agreement with the average retention time of the six replicate check standard injections.The retention time of the sample peaks must fall within ±10% agreement with the average check standard retention time.If any bracketing standard fails in the middle of a sequence, the data beyond the last successful check standard is to be considered invalid.

I. Quantification
(a) Plot a linear external calibration curve by using the standard concentration values in units of mg/mL versus the integrated area of each bLF peak.Display the regression trendline equation: where C ¼ concentration of bLF in the sample, (mg/mL); y ¼ area of the bLF peak in the injected standard solution; b ¼ y-intercept of the curve; and M ¼ slope of the curve.(b) Display the R 2 value for the curve.The R 2 value must be ≥0.995.(c) Calculate the bLF concentration (mg bLF/100 g) in the sample using the following equation:   In the injection sequence, an extra blank of water or 0.05 M Na 2 HPO 4 -1M NaCl solution can be added after an anticipated higher bLF sample injection to avoid carryover.demonstrating that the method measures only soluble, intact, non-denatured lactoferrin.An overlay of chromatograms including a matrix blank, the lowest level lactoferrin standard (LOQ), the matrix blank þ lactoferrin, and the matrix blank þ heat-treated lactoferrin is shown in Figure 5, demonstrating that heat-impacted bLF was not detected using this method.(c) Lactoperoxidase.-Asolution of lactoperoxidase (Sigma-Aldrich, L2005) was analyzed both with and without the heparin column cleanup to assess the potential interference, and was followed by a chromatographic evaluation of a mixed lactoferrin/lactoperoxidase sample solution.An overlay of these experiments is shown in Figure 6,   demonstrating that not only does the heparin column eliminate nearly all the lactoperoxidase, but the chromatographic column is also sufficiently selective to resolve the two components.

Linearity
Linearity data are based on a standard curve of 10 mg/mL to 250 mg/mL, corresponding to an analytical range of 2 mg/100g to 278 mg/100g.Six levels of standards were prepared and analyzed on three standard runs on three days and data are summarized in Table 4.

LOD
The practical LOD was made by diluting the lowest calibration standard (10 mg/mL) by 10, yielding a standard solution corresponding to 0.2 mg/100g.Six replicate injections of this solution were made, the ratio of bLF signal to baseline noise for each injection was determined, and the results are displayed in Table 5.
All individual S/N values were 8 or higher.

LOQ
The LOQ for the analysis is set at the lowest standard level of 10 mg/mL and determined through back calculation using the default weights and dilution factors to be 2 mg/100 g, reconstituted.The LOQ required by the SMPR for lactoferrin is 4 mg/100g, reconstituted.The peak area and S/N results of six replicate injections of the lowest standard from one analytical run are displayed in Table 6.

Repeatability (RSD r ) and Intermediate Precision (RSD Int )
The repeatability RSD r was obtained for seven different lactoferrin fortified, milk-based powder samples and is summarized in Tables 7 and 8.Both RSD r and RSD Int for all matrixes studied fall below the SMPR RSD r acceptance criteria.

Method Equivalence Studies
Equivalency studies were performed on two of the samples validated in the original SLV: milk-based infant formula powder and milk-based partially hydrolyzed infant formula powder.A summary of the results, demonstrating good agreement between the two methods, is shown in Table 9.

Accuracy
The accuracy results as determined by mean recoveries for the original method are summarized in Table 10, and in Table 11 for the updated method.Included is a matrix blank spiked at high bLF levels (97 mg/100g and 193 mg/100g) to assess recovery at the upper end of the SMPR analytical range.All average recoveries for each sample matrix fell within the accepted range of 90-110%.

Analytical Range
The analytical range of the lactoferrin method as determined from the range of the standard curve is 2-278 mg/100g.The sample concentrations assessed in this lactoferrin validation study ranged from 4-193 mg/100g, based on the repeatability, intermediate precision, and accuracy results.

Conclusions
The described method has met the performance requirements defined in SMPR 2020.005 and is suitable for the quantification of soluble, intact, non-denatured bLF in bovine milkbased infant formulas, and pediatric and adult nutritional products.

(a )
Sample hydration(1) Accurately weigh 0.5-1.50g sample, or enough to contain 0.5-10 mg lactoferrin, directly into a 15 mL centrifuge tube.Record the sample weight to 0.0001g.(2) Pipette 11.5 mL warm (40 � C) 0.2 M sodium phosphate dibasic solution, pH 8.0 (loading buffer) to the centrifuge tube in step (1), vortex and/or shake well until smooth.(3) Continue mixing the solution in step (2) by test tube rocker or shaker for at least 60 min, check the solution during the rocking, vortex mixing or shaking manually if needed.(4) Centrifuge for at least 20 min at 8000g at 4 � C. (5) Carefully transfer as much of the liquid layer between the fat (top) and solid (bottom) layers in the centrifuge tube as is practicable, while minimizing disturbance of other layers, to a 25 mL volumetric flask.(6) Pipette 11 mL warm (40 � C) 0.2 M sodium phosphate dibasic solution, pH 8.0 (loading buffer) into the same centrifuge tube in step (1) and use a vortex mixer to mix the contents for 10 s or until fully suspended.Continue mixing by rocking for at least 20 min on the test tube rocker or shaker.(7) Centrifuge the solution for 20 min at 8000g at 4 � C and then repeat step (5), using the same transfer pipette (if applicable) from step (5) to transfer the second liquid layer to the same 25 mL volumetric flask.(8) Dilute the solution to volume (25 mL) with 0.2 M sodium phosphate dibasic solution, pH 8.0 (loading buffer).(9) Thoroughly mix the combined solution in the volumetric flask.(b) Prepare the HiTrap heparin column [see section F(d) for full details on the use of the HiTrap columns] (1) Connect a 10 mL syringe barrel to the heparin affinity column using a syringe fitted to the Luer connector.(2) Remove the snap-off end (for a new HiTrap heparin column) or unscrew the cap (for used) at the column outlet.Tightly connect the column outlet to a vacuum manifold with a threaded Luer adapter.(3) Equilibrate the heparin affinity chromatography column by passing 5 mL 0.2 M sodium phosphate dibasic solution, pH 8.0 (loading buffer) through the column and discard the eluent.Notes about column usage can be found in section F(d).

( a )
Instruments.-AWaters Acquity I-Class UHPLC system with flow-through needle and UV detector or equivalent.(The HPLC system should be located in an area where temperature fluctuations will be minimal throughout the run.)(b) Column.-AWaters Bridge Protein BEH C4 column, 4.6 × 150 mm.Part No. 186004504 or equivalent.(c) Instrument settings.-SeeTable 2021.10B.(d) Gradient.-SeeTable 2021.10C.(e) Turn on the detector and pump mobile phase over the column at a flow rate of 0.5 mL/min with the initial gradient conditions from Table C ¼ bLF concentration derived from the standard calibration curve (mg/mL); m ¼ sample weight (g); V1 ¼ sample volume after centrifuge steps, combining both liquid layers and diluting to volume (25 mL); V2 ¼ volume loaded onto the HiTrap column (5 mL); and V3 ¼ final collection volume after HiTrap column (5 or 10 mL) Chromatographic integrity.-Examplechromatograms for the unfortified solvent blank, bLF standard, milk-based infant formula powder matrix blank, and milk-based infant formula powder (bLF ≤14 mg/100g) are shown in Figures 1-4.Lactoferrin was not detected in the blank or unfortified samples.Interference peaks were not identified in any of the injections.bLF retention times of samples were within ±10% of the standard.(b) Heat-impacted lactoferrin.-Allresults of heat-treated lactoferrin spiked into matrix blanks were <LOQ,

Table 1 .
bLF method development modifications

Official Method SM 2021.10 Bovine Lactoferrin in Powdered Infant Formula and Adult Nutritionals Heparin Affinity Extraction and Reverse-Phase High-Performance Liquid Chromatography/ Ultraviolet Detection (HPLC/UV) First Action 2021
Reagent preparations may be scaled up or down as needed.
a) Water.-18.2MΩ/cm, or equivalent.(b)Sodiumhydroxide (NaOH).-ACSgrade,≥98%,Fisher Scientific, Catalog No. RDC0530125, or equivalent.(e)Sodiumchloride.-CertifiedACSgrade,≥99.3%, Fisher Scientific, Stock No. S271500, or equivalent.(f)Ethylalcohol.-200proof(absolute),≥99.5%, Sigma-Aldrich, Stock No. 459836, or equivalent.(g)Trifluoroaceticacid (TFA).-≥99.9%,EMDChemicals,Stock No. MTX1276-6, or equivalent.(h)Acetonitrile.-HPLCgradeorbetter, 99.9%, Fisher Chemical, or equivalent.Frueh et al. | 695 (i) Lactoferrin (from bovine milk).-Certifiedreferencematerial,Cerilliant No. L-047, Supelco, or a standard that has been validated [see Note (2) in section E(a)].D. Reagent/Mobile Phase PreparationNote: -Weigh 3.55 g sodium phosphate dibasic and 29.2 g sodium chloride into a 500 ml beaker, add 450 mL water to dissolve, adjust pH to 8.0 ± 0.05 with o-phosphoric acid (or diluted) and 1 N sodium hydroxide solution, and then transfer the solution to a 500 mL volumetric flask or to an equivalent container, dilute to a final volume of where, Conc (bLFStock I) ¼ concentration of bLF stock solution I; Mass (Cert.Ref. Std) ¼ mass of the certified reference standard; Vol (Water) ¼ the volume of water added to the certified reference standard (10.0 mL).Alternatively, accurately weigh 80-120 mg bLF reference standard material into a 15 mL centrifuge tube.Record the weight to 0.01 mg.Calculate the net bLF mass by multiplying the reference standard weight by the purity (wt.%).Calculate and add the volume of water required to reach a concentration of 10.00 mg bLF/mL to the tube.Gently swirl every 10-15 min to dissolve and let the solution stand for at least 1 h at room temperature.Store at 4 � C in a sealed plastic vessel protected from light.Calculate the concentration to four significant figures as follows:

Table 2021 .
10A for each calibration standard.Alternatively, solutions may be made up in 10.0 mL volumetric flasks.Mix well and filter into a silanized HPLC vial through a 0.45 mm, 13 mm wwPTFE Mini syringe filter for analysis.Prepare fresh daily.Calculate the concentrations to four significant figures as follows:

Table 2021 .
10D.Injection sequence a a

Table 5 .
LOD assessment with bLF standard concentration at 1 mg/mL (corresponds to 0.2 mg/100g in samples)

Table 6 .
LOQ assessment with bLF standard concentration at 9.54 mg/mL (corresponds to 2 mg/100g in samples)