Abstract
Urea is used in biopharmaceutical manufacturing processes for the purification of therapeutic proteins, for cleaning columns, and for refolding proteins after purification. The urea used for such purposes is typically USP grade material obtained from commercial sources and further characterization is required prior to use, such as determination of purity and identity. For this purpose, a robust analytical method is needed that can characterize the known organic impurities of urea. However, the existing methods show high assay variability and are not able to resolve all known organic impurities as desired for accurate quantification. In the present manuscript we developed a new high-performance liquid chromatography method with UV detection for the separation of urea and its impurities (biuret, cyanuric acid, and triuret). The method performance characteristics evaluated for urea and biuret were specificity, linearity, accuracy, identity, precision, and robustness and the newly developed method met all predefined performance acceptance criteria.
Introduction
Urea is a chaotropic agent widely used for solubilization of therapeutic proteins expressed in Escherichia coli in the form of insoluble inclusion bodies and for the refolding of proteins to restore biological functions (1⇓⇓⇓–5). Urea solutions are also used for cleaning and regeneration of affinity chromatography columns such as protein A/G after purification of therapeutic monoclonal antibodies and Fc fusion proteins from cell culture harvest (6⇓⇓–9). The high price of these column resins necessitates reuse, and therefore extensive cleaning is required to prevent the impurities, such as host cell proteins and nucleic acids, from accumulating and leaching into subsequent process cycles (8). Cleaning restores column capacity, extends column life, and prevents column fouling; the latter may have a deleterious effect on the safety and efficacy of therapeutic proteins (10, 11).
Urea manufactured in the chemical industries varies in its purity level and contains several organic and inorganic contaminants and side products. The common urea organic impurities are biuret, cyanuric acid, and triuret, which are produced during the synthesis and processing of urea (12). Among the organic impurities, biuret is the major impurity and USP grade urea contains not more than 0.1% (w/w) (13). For manufacturing of therapeutic proteins, pharmaceutical-grade urea of defined purity is required to minimize contamination of therapeutic drugs caused by urea impurities. Therefore, the urea used in the manufacturing of therapeutic proteins is rigorously tested to ensure that all previously defined specifications are met, the impurities are well controlled, and the urea is compliant and fit for the intended purposes.
Several approaches have been used for the quantification of urea in a variety of samples. The derivatization of urea followed by a colorimetric analysis (14) and the enzymatic hydrolysis of urea to carbon dioxide and ammonia have been used (15⇓⇓–18). These methods have limited applications because of interference from sample matrices and a time-consuming derivatization process that may not be suitable for quantification of impurities (16, 19).
A variety of chromatographic approaches have been used for the analysis of urea in various matrices. Examples documented in the literature include a high-performance liquid chromatography method with UV detection (HPLC-UV) analysis of urea using hydrophilic interaction chromatography (HILIC) (20), liquid chromatography mass spectrometry (LC-MS) analysis using SeQuant Zic-HILIC (21) and Acclaim PA2 C18 columns (16). In our experience, the HILIC columns fail to resolve all impurities and C18 columns do not retain urea well. The USP Monograph Urea (13) employs an HPLC-UV analysis using a 150 × 4.6 mm HILIC Ascentis Express OH5 column with a gradient of 0.1% (v/v) formic acid in water and acetonitrile for identification and quantification of urea and biuret. Based on our experience, the USP method varies in success rate potentially because of the extensive column conditioning and equilibration required. Furthermore, the method cannot resolve all the organic impurities commonly present in urea. These deficiencies led us to develop a new method that reliably quantifies urea and biuret and can resolve all major common organic impurities.
In this paper we describe the development and validation of a new HPLC-UV method to establish the identity and purity of urea used at manufacturing facilities. The method resolves urea and all commonly known urea organic impurities, biuret, cyanuric acid, and triuret. It accurately quantifies urea and biuret in a raw material used for therapeutic protein manufacturing. The method has been validated and is used in a good manufacturing practice (GMP) environment for the urea raw material lot release. The method is robust, precise, and accurate and exceeds the USP assay acceptance criteria required for the urea raw materials used in manufacturing processes.
Materials and Method
Materials
Urea USP reference standard (urea RS), biuret, and cyanuric acid were obtained from Sigma-Aldrich (St. Louis, MO). Phosphoric acid (85%),10 N sodium hydroxide, and urea (J.T Baker, USP grade) were from Fisher Scientific (Waltham, MA). Triuret was from Enamine (Monmouth Jct., NJ), urea pellets were from Reliable Pharmaceutical (St. Louis, MO). Acclaim Mixed-Mode WAX-1 (4.6 mm × 150 mm, 5 µm) was from Fisher Scientific (Waltham, MA), Ascentis Express OH5 (150 mm × 4.6 mm, 2.7 µm) was from Sigma-Aldrich (St. Louis, MO). All other columns used during development are listed in Table S-I. Milli-Q (Millipore-Sigma, Burlington, MA) HPLC grade water was used for the mobile phase preparation. The HPLC system used was Agilent 1100 (Santa Clara, CA). Data acquisition and analysis were performed using Chromeleon SR4 Software (Thermo-Fisher Scientific, Waltham, MA). Fusion QbD software V 9.8.0 Build 921 (S-Matrix, Eureka CA) was used for design of experiment (DOE) in the robustness study.
Method
Preparation of Resolution Solution:
The resolution solution contained 3 mg/mL urea and 3 µg/mL biuret in water. For urea, a 6 mg/mL stock solution in water was prepared by dissolving 150 mg of urea reference standard (USP grade urea) in a 25 mL volumetric flask and stirring for 30 min using a magnetic stirrer. For biuret, an initial 300 µg/mL stock solution in water was prepared by dissolving 75 mg of USP reference material in a final volume of 250 mL water in a volumetric flask and stirring using a magnetic stirrer for 60 min or until all solid dissolved. The secondary biuret stock containing 6 µg/mL concentration in water was prepared by diluting 5 mL of 300 µg/mL solution to a final volume of 250 mL in a volumetric flask. The resolution solution was then obtained by mixing equal volumes of 6 mg/mL urea and 6 µg/mL biuret stocks.
Preparation of Working and Check Standards:
Both the working standard (3 mg/mL) and a check standard (3 mg/mL) were prepared independently by dissolving 75 mg of urea USP RS in water in a 25 mL volumetric flask. The mixture was stirred for 30 min at room temperature using a magnetic stirrer.
Preparation of Standards for Urea Linearity Determination:
A stock solution of 4 mg/mL urea in water was prepared by dissolving 200 mg urea in a 50 mL volumetric flask and stirring at room temperature for 30 min. A set of calibration standards were prepared by transferring 5, 6, 7.5, 9, and 10 mL of 4 mg/mL urea stock using class A precision volumetric pipettes to 10 mL volumetric flasks and filling to the mark with water. This gave five calibration standards with concentrations of 2.0, 2.4, 3.0, 3.6, and 4 mg/mL, respectively.
Preparation of Standards for Biuret Linearity Determination:
A biuret solution of 6 µg/mL in water was prepared from the 300 µg/mL biuret stock solution as described previously in the resolution solution preparation section and used as the highest concentration standard. Linearity standards of 4.5, 3.0, and 2.4 µg/mL biuret were prepared by diluting 7.5, 5, and 4 mL of the 6 µg/mL solution in 10 mL volumetric flasks, respectively, and linearity standard 1.5 µg/mL biuret was prepared by diluting 5 mL of the 6 µg/mL solution in a 20 mL volumetric flask.
Preparation of Accuracy and Precision Samples:
The method accuracy was determined by preparing urea pellets solutions in sextuplicate. For each of the six solutions, ∼75 mg of urea pellets was dissolved in water in a 25 mL volumetric flask to a target concentration of ∼3 mg/mL. In addition, the working and check standards were prepared as described in the previous section.
Similarly, biuret solutions for determining accuracy and intraday precision were prepared in sextuplicate from urea pellets and the biuret standard to target concentrations of approximately 3 mg/mL urea and 3 µg/mL biuret, respectively. The working and check standards were also prepared to a target concentration of 3 μg/mL biuret. The accuracy was determined by subtracting any biuret contribution from the urea pellets and comparing against the 3 µg/mL biuret working standards. Inter-day intermediate precision of the method for both urea and biuret was determined by conducting the experiment on different days with two analysts and two instruments. Urea pellets with a slightly higher level of biuret were used as a test sample at a nominal urea concentration of 3 mg/mL, and chromatographic analysis was conducted in duplicate for each day.
System Suitability:
The resolution solution of 3 mg/mL urea and 3 μg/mL biuret was used to evaluate system suitability parameters. New columns and columns that had been extensively used were compared to establish the resolution of urea and biuret that yields acceptable quantification results. The parameters evaluated were peak asymmetry (EP, USP tailing factor; acceptance criterion, <1.8), resolution factor (USP) of urea and biuret (acceptance criterion, ≥2.30), relative standard deviation (RSD) of six working standard injections (acceptance criterion, ≤0.85%), check standard recovery, and bracket standard recovery (acceptance criterion, 100% ± 1.5%). The working standard was injected and used as the bracket standard after every 10 samples.
Specificity–Matrix Interference and Carryover Evaluation:
Matrix interference was evaluated by comparing the urea and biuret signals of their respective check standards to those of preceding blank injections. Carryover was evaluated for both urea and biuret by comparing the signal of blanks analyzed right after their respective highest calibration standards.
Reporting Limit:
The biuret reporting limit (limit of quantification) was assessed by analyzing the 1.5 μg/mL (15 ng on column) biuret solution six times.
Robustness:
The method robustness is its ability to remain unaffected by small deliberate changes in analytical parameters (22). The experimental setup for robustness evaluation was carried out by a DOE approach using Fusion QbD software V 9.8.0 Build 921. The mobile phase pH was set at three pH values, pH 6.0, pH 6.2, and pH 6.4, respectively. The mobile phase flow rate was set at 0.45 mL/min, 0.5 mL/min, and 0.55 mL/min, respectively, and the column temperature was set at 28°C, 30°C, and 32°C, respectively. The solution containing 3 mg/mL urea and 3 µg/mL biuret was analyzed by HPLC under robustness experimental conditions to evaluate the effect of these small changes on peak area of urea and system suitability parameters including peak resolution between urea and biuret.
HPLC Analysis of Urea and Impurities:
The analysis of urea and its impurities, biuret, cyanuric acid, and triuret, was performed using Agilent 1100 HPLC-UV, at 200 nm on either a diode array detector (DAD) or variable wavelength detector (VWD) with the column thermostat at 30°C. Optimum separation of urea and its impurities was achieved on Acclaim Mixed-Mode WAX-1 (150 mm × 4.6 mm) using a mobile phase of 25 mM phosphate buffer (pH 6.2) at a flow rate of 0.5 mL/min over 30 min and injection of 10 µL sample for analysis. To ensure the performance of the column and the entire chromatographic conditions, resolution solution containing USP urea RS at a concentration of 3 mg/mL and USP biuret RS at 3 μg/mL was analyzed once at the beginning of the sequence to evaluate the system suitability. The proportion of biuret in the resolution solution is constituted to 0.1% (w/w), the maximum amount allowed in USP grade urea raw material. The chromatographic analysis was conducted to establish the purity and identity of urea as the raw material used in manufacturing using actual samples and the method meets all analytical performance characteristic described in Table I. Quantification of urea and biuret was achieved by resuspending the same quantity weight by weight of USP RS working standard and test samples in water. The working standard was analyzed six times (Table S-II) and the average of the analysis results was used to determine the quantity of urea and biuret in the test samples (Eq 2). The identity of urea was determined by comparing the average of the retention times of six urea working standard injections to the retention times corresponding to urea in the test sample (Identity test). Impurities in the urea were calculated by considering a response factor of 1 for urea and a response factor of 28.934 for biuret determined from the slopes of their calibration curves. To evaluate the capability of the assay, several lots of samples of different sources and vendors were analyzed. One lot typically used in manufacturing is shown in Table II and Table III. The method was able to analyze all urea raw materials lots from different vendors with the required precision and accuracy.
Results and Discussion
Initial Method Development
In an initial method development, a combination of 17 different column types and sizes were evaluated in an HPLC-UV analysis using suitable chromatographic conditions such as mobile phase composition, pH, and flow rates (Table S-I). These columns were explored for optimum separation and quantification of urea and its known major organic impurities. Figure 1 shows some chromatograms obtained using Zic-HILIC (Panel A), Supelcosil LC-SCX (Panel B), Dionex IonPac CS14 (Panel C) and Phenomenex C18 (Panel D) columns. Separation of urea and impurities achieved by WAX-1 column is shown in Figure 2. It is evident that Zic-HILIC is unable to sufficiently resolve biuret from cyanuric acid. Urea is not retained on the Supelcosil column, further urea and cyanuric acid are coeluting when using Dionex IonPac CS 14. Finally, both urea and biuret are not retained when applying Phenomenex C18 column. Ultimately, the Acclaim Mixed-Mode WAX-1 column gave the best separation and quantification of urea and this column was able to resolve all major known organic impurities and was therefore selected for further consideration.
Comparison of Method Performance
The performance of the Acclaim WAX-1 column was compared to that of the Ascentis Express OH5 column used in the USP Monograph Urea. Figure 2 shows a comparison of the chromatography for samples run employing the new method using the Acclaim WAX-1 column and USP monograph Urea. Panel A of Figure 2 shows the chromatograms of urea (5 mg/mL), biuret (0.05 mg/mL), cyanuric acid (0.005 mg/mL), and triuret (0.05 mg/mL) resuspended in the mobile phase and analyzed on an Acclaim Mixed-Mode WAX-1 column. Panel B of Figure 2 is the chromatograms of urea, cyanuric acid, and triuret separated by the USP method using an Ascentis Express OH5 column. The side by side comparison shows that the new method demonstrated better resolution capacity. The USP method was not able to separate biuret from cyanuric acid, nor was this method able to sufficiently separate triuret and urea. Moreover, the USP method showed higher background noise as the gradient commences.
Assay Performance Characteristics
The analytical procedure characteristics were evaluated (Table I) based on the requirement of ICH guideline Q2(R1) (22). The analytical characteristics evaluated were linearity and range, accuracy, specificity, precision (repeatability and intermediate precision), and robustness for both urea and biuret. The limit of quantification (LOQ) was also assessed for biuret. The assay acceptance targets and validation results for each of these performance characteristics are shown in Table I.
The linear regression equation obtained for the concentrations of urea and biuret demonstrated good agreement of response and concentration over the range covered (Figure 3). The accuracy results for both the urea and biuret assays are shown in Table II; both met the respective acceptance criterion. The assay showed excellent specificity as demonstrated by chromatographic overlay of urea peak and blank injections (Figure 4). Comparison of the retention times of the USP urea working standard (N = 6) and the urea test samples to evaluate identity demonstrates very good agreement (Table III).
All system suitability parameters assessed met acceptance targets (Table IV). The lowest observed resolution factor between the urea and biuret peaks that yielded acceptable urea assay was 2.30, and this low-resolution factor value was only obtained when using a very aged column. Evaluation of intraday precision (Table II) and inter-day precision (Table V) for urea and biuret assays demonstrate very high repeatability and reproducibility that met predefined targets.
Robustness evaluation for urea and biuret assays demonstrated that there was no loss in chromatographic resolution between urea and biuret peaks and that the smallest average resolution observed was 4.05. The peak tailing factors were least affected by the changes in robustness conditions. All the peak areas obtained under the robustness conditions for both urea and biuret were within three standard deviations of their respective mean.
Conclusion
In this manuscript we described the development of HPLC-UV urea purity and identity assays and the chromatographic analysis of its major known organic impurities: biuret, cyanuric acid, and triuret. The method resolves all the required compounds well and accurately quantifies urea and biuret. The method is intended to quantify urea and biuret and evaluate the presence or absence of cyanuric acid and triuret. The assay met all predefined performance characteristics: linearity, accuracy, spike recovery, specificity, intermediate precision, and robustness. This method has been successfully validated in the QC facility and implemented in the GMP environment as a purity method for quantification of urea and biuret.
Conflict of Interest
The authors declare that they have no competing interests.
Acknowledgements
The authors thank Anna Ip for helping with the Fusion Software used in the DOE experiment; Matthew Blake, Michael Ronk, Yasser Nashed-Samuel, Zhongqi Zhang, Jason Richardson, Kathleen Burke, and Katie Parks for useful discussion and input during method validation.
Appendix
Calculation of the Concentration of Urea in All Sample Types
Working Standard (CS), Check Standard (Cc), and Test Samples (Ct) using the following equation:
Where:
COA = Certificate of Analysis.
Determination of Urea Assay
Where:
Rx = Peak area from the Test Sample, Check Standard, or Bracket Standard.
Rs = Average of the Peak Area for the Working Standard (N = 6).
Cs = Urea concentration in the urea USP RS (Working Standard, mg/mL).
Cx = Urea concentration in the Test Sample, Check Standard, or Bracket Standard (mg/mL).
Identity Test
Calculate the percentage difference (%DRT) between the average retention time of urea from the Working Standard (N = 6) and the retention time of urea from the Test Sample using the following equation: Where:
RT1 = Average Retention Time of the Working Standard (N = 6).
RT2 = Retention Time of the Test Sample.
Impurities calculation
Use the following equation to calculate % impurities in a Test Sample: Where:
Ai = Peak Area of an individual impurity in the sample injection.
∑Aj = Sum of the (Aj/RF) Peak Area divided by RF and take sum for all impurities (consider all peaks, t = 3.4 to t = 10 min).
RF = Response Factor. For biuret, cyanuric acid, and triuret RF = 28.934; for urea RF = 1.
- © PDA, Inc. 2020
References
PDA members receive access to all articles published in the current year and previous volume year. Institutional subscribers received access to all content. Log in below to receive access to this article if you are either of these.
If you are neither or you are a PDA member trying to access an article outside of your membership license, then you must purchase access to this article (below). If you do not have a username or password for JPST, you will be required to create an account prior to purchasing.
Full issue PDFs are for PDA members only.
Note to pda.org users
The PDA and PDA bookstore websites (www.pda.org and www.pda.org/bookstore) are separate websites from the PDA JPST website. When you first join PDA, your initial UserID and Password are sent to HighWirePress to create your PDA JPST account. Subsequent UserrID and Password changes required at the PDA websites will not pass on to PDA JPST and vice versa. If you forget your PDA JPST UserID and/or Password, you can request help to retrieve UserID and reset Password below.