A IMS OF S TUDY
The aims of the present studies were: x Analysis of Biotechnological Synthesis
- To develop a strategy for rapid identification of the product in a fermentation process where reference sample is not available (Pa- per I).
- To develop and validate a method for analytical quantification of both the substrate and the product in a fermentation process (Paper
- To give general guidelines for development and validation of methods aimed for quantification of drugs produced through fer- mentations(Paper II). x Determination of Adsorption Isotherm Parameters
- To characterize a new CSP (Kromasil CHI-TBB) aimed for prepa- rative chromatograpy through the determination of its validated iso- therm parameters for a chiral model compound (Paper III).
- To develop a new injection technique for the perturbation peak method in order to determine adsorption isotherm parameters di- rectly from binary (Paper IV-V) or quaternary mixtures (Paper
- To validate the newly developed injection techniques for the binary (Paper V) and quaternary cases (Paper VI).
This section provides a brief overview of the development and validation of analytical methods for both qualitative and quantitative analysis of small drugs (Mw < 1000) derived from process liquids like fermentation medium The principles discussed are applicable to any low-molecular weight compounds in biotechnological or organic synthesis contexts.
B IOTECHNOLOGY
Biotechnology focuses on the commercial production and isolation of specific molecules from plant, animal, or microbial sources, often utilizing fermentation for the synthesis of hard-to-produce substances Over millennia, various biotechnological processes, such as brewing, wine-making, and cheese-making, have been developed and refined, although their early practitioners were unaware of the enzymatic mechanisms involved It wasn't until the 1940s to 1960s that large-scale industrial processes relying on enzyme-catalyzed bio-transformations were established.
Extensive research in steroid chemistry has focused on synthesizing key components, particularly cortisones A crucial early step in these transformations involves site- and stereo-specific hydroxylation of the steroid core Traditional chemical methods face challenges in introducing hydroxyl groups due to the complexity of typical steroids, which contain 19 to 29 carbon atoms Since the early 1950s, this challenge has been addressed by leveraging the high specificity of enzymes found in microorganisms.
In the early 1950s, a significant breakthrough was achieved with the development of the 11D-hydroxylation process Today, various reactions, including hydroxylation at different positions on the steroid skeleton, can be effectively performed using microbial cultures The primary advantage of microbial hydroxylation lies in the regio- and stereo-specificity of the enzymes involved In contrast, traditional synthetic methods face challenges due to the presence of multiple easily hydroxylated carbons, which can lead to either D- or E- attachment of the OH-group, increasing the risk of introducing multiple hydroxyl groups to the steroid skeleton.
In the studies presented in Papers I and II, progesterone (PS) served as the substrate, while the enzyme responsible for the production of 9D-hydroxyprogesterone (9D-OH-PS) was steroid 9D-hydroxylase, as illustrated in Figure 1.
Figure 1 Structures of substrate and product Position 9 and 11 is marked in the figure.
B IOPROCESS M ONITORING
Traditional Bioprocess Monitoring
The rapid advancement of biotechnological research in recent decades has highlighted the necessity for effective analytical techniques in bioprocess monitoring Accurate identification and quantification of critical components in biotechnological processes are vital for enhancing process development and optimization Enhanced monitoring practices have been shown to lead to improved control and increased productivity Although many bioprocesses continue to rely on traditional parameters like CO2, pH, and O2 for monitoring, there is a growing demand to also track the concentrations of fermentation substrates and products for better overall management.
The control of biotechnological processes can be specified in three groups:
Off-line control with distance involves taking samples from the production process for analysis in a central laboratory This approach is commonly used when the high investment and operational costs of sophisticated equipment, such as HPLC, gas chromatography (GC), or mass spectrometry (MS), affect the technology Typically, the response time for results is around 24 hours.
48 hours, which give opportunity only for subsequent measures
Offline control using local equipment allows for quicker response times by analyzing samples extracted from the process, resulting in more reliable adjustments This method offers clear advantages over remote control, despite the potential complexity and cost of analysis techniques Simple methods like thin layer chromatography (TLC) can be employed, while more advanced chromatographic methods such as high-performance liquid chromatography (HPLC) and gas chromatography (GC) are frequently utilized for monitoring bioprocesses HPLC is further detailed in section 2.5.
Online control involves the automatic sampling of processes, with immediate measurement and calculation of results by a computer, which then makes necessary decisions When the measurement time aligns with the process rates, such as growth or product formation rates, it is referred to as real-time control—an ideal approach for process control However, this method faces challenges, including the need for sterilized equipment and limited recalibration options, making the automatic collection of sterile samples from processes sometimes unreliable.
In many industries, offline methods dominate process analysis; however, there are ongoing efforts to integrate techniques like HPLC, GC, MS, and biosensors online Flow Injection Analysis (FIA) is a widely adopted method for analyzing fermentations, allowing for the mixing of samples with reagents (such as enzymes) during flow FIA offers advantages like high sampling frequency, low reagent consumption, and cost-effectiveness, although these benefits depend on the specific reaction employed A significant limitation of FIA is its often insufficient selectivity and sensitivity for product detection, typically relying on the measurement of co-substrates or co-products, such as O2 consumption or H2O2 formation Additionally, enzyme-based reactions can face stability challenges, and the lack of selectivity means that FIA can generally analyze only one analyte at a time, in contrast to chromatography.
It is possible to circumvent the drawbacks caused by the sample treatment by using non-invasive method, such as near infrared spectroscopy (NIR)
[26] However, this and such methods are generating indirect information from the process with a lot of noises The evaluation of such methods is dif- ficult and seldom reliable.
New Approaches in Bioprocess Monitoring
Mass spectrometry (MS) is an advanced technique recently utilized for monitoring biotechnological processes, particularly for the online detection and quantification of gases However, it requires costly equipment and is less user-friendly compared to HPLC with UV detection Recently, diode-array detectors (DAD) have been increasingly employed to monitor the fermentation processes of wines and ethanol.
Field flow fraction (FFF) is an effective elution technique designed for molecules with a molecular weight greater than 1000, although this thesis does not focus on large molecules The literature on FFF is extensive, highlighting its applications in the separation and characterization of proteins and enzymes, as well as the separation of human and animal cells.
[33] and the molecular weight and particle size distribution of polymers [34,
35] FFF is a relatively new approach in biotechnology; therefore practical experiences are not yet abundant.
Product Identification
HPLC-UV diode-array detection (DAD) and HPLC-MS techniques utilize chromatography for separation, with DAD or MS serving for identification and quantification These methods efficiently deliver real-time UV and MS data for each peak in a chromatogram, allowing for direct peak identification through comparison with existing literature or standard compounds However, the absence of standard compounds complicates the preliminary identification process, particularly in bioprocess monitoring when a new biotechnological process is initiated and reference substances are unavailable.
A critical phase in developing a biotechnological production protocol involves creating reference material for optimizing and validating processes at pilot (100-300 L) and production (~m³) scales This reference material can be synthesized on a lab scale (~10 L), yielding enough for clear chemical identification via NMR However, having access to a rapid preliminary identification method for the product, along with initial quantitative monitoring, is essential Given that fermentation processes can take a week or longer, it's crucial to assess early whether to continue the process, as halting it promptly in case of undesirable outcomes can lead to significant economic benefits.
Rapid analysis of the products of a biotechnological process at an early stage is essential to clearly distinguish the product peak from other growing peaks
In biotechnological processes, the composition of the culture medium evolves due to microbiological metabolism during fermentation, resulting in a complex fermentation broth with various breakdown products and intermediates When multiple peaks appear during this process, it is crucial to identify the desired product among them to optimize time and resources and prevent unnecessary fraction collections At this point, there may be insufficient time to purify the substantial amounts of product needed for accurate NMR identification, which is essential for confirming the final product.
For preliminary product identification when reference samples are unavailable, HPLC combined with DAD or MS proves effective This study aimed to evaluate and compare the efficacy of HPLC-DAD and electrospray ionization (ESI) MS for the rapid identification of products at the initial stages of synthesis, even in the absence of reference materials.
Product Quantification
Rapid quantification of products and substrates in fermentation processes is crucial for optimizing development Many fermentation laboratories utilize HPLC systems paired with affordable diode-array detectors for effective quantitative monitoring HPLC enables simultaneous analysis of multiple components within a single sample, making it an essential tool for bioprocess monitoring While mass spectrometry (MS) is not standard in most fermentation labs, DAD often suffices for quantifying higher product concentrations, allowing MS to be reserved for other analytical needs.
In Paper II, the objective was to develop and validate an analytical method for quantifying both the substrate and the product, facilitating a comprehensive characterization of the process kinetics, including the assessment of substrate conversion and product formation rates.
S TEROIDS
Steriod Analysis
Significant efforts have been dedicated to detecting steroids in biological fluids, employing various methods such as simple TLC for qualitative analysis While immunoassays have been utilized for quantification, they face challenges like cross-reactions and interference from other substances To address these issues, several chromatographic techniques have been developed, predominantly using gas chromatography (GC), which offers excellent detection limits but necessitates prior derivatization of steroids for volatilization High-performance liquid chromatography (HPLC) has also been reported, utilizing UV detection or LC-MS, although earlier stationary phases like Sephadex LH-20 and Lipidex were limited by their inability to handle high pressure, resulting in slow and time-consuming separations Currently, HPLC is the favored method for steroid separation, employing both normal-phase and reversed-phase chromatography.
HPLC
A modern High-Performance Liquid Chromatography (HPLC) system comprises several key components, including a high-pressure solvent delivery system, a sample auto injector, a separation column, and a detector, typically a UV or DAD Additionally, many systems feature a temperature-controlled oven and a pre-column to safeguard the analytical column from impurities The separation process occurs within the column, which is filled with chemically modified silica particles ranging from 3.5 to 10 micrometers As the mobile phase is pumped through the column, the analytes in the injected sample are separated based on their interactions with the stationary phase Selecting the appropriate stationary and mobile phases is crucial for achieving optimal separation results.
Figure 2 Block diagram of a general LC-system
The possibility to extract information (both qualitative and quantitative) about several important compounds in one single analytical run, makes HPLC to a strong tool for bioprocess monitoring.
M ETHOD D EVELOPMENT
Analytical chemistry focuses on techniques for determining the chemical composition of samples, with multiple methods available for measuring a compound The selection of analytical methods depends on various factors, including the analyte's chemical properties and concentration, sample matrix, analysis speed and cost, measurement type (quantitative or qualitative), and the number of samples involved Qualitative methods identify the chemical species present in the sample, while quantitative methods provide numerical data on the relative amounts of these species Typically, qualitative information is essential prior to conducting quantitative analysis, and a separation step is often required in both types of analysis.
In the early stages of a project, method development should be minimal, allowing for a more labor-intensive workup procedure with a limited number of samples As the project progresses, it becomes beneficial to focus on automation and convenience in the analytical methods employed Each analytical method consists of several steps, represented in a simplified flow diagram for chromatographic procedures, which illustrates chemical analysis as a sequence of information and operations structured as input, process, and output.
Figure 3 Simplified block diagram of an analytical chromatographic method
Before starting method development, it's essential to review existing knowledge about the sample and clearly define the analysis goals Considerations should include the number of samples to be analyzed and the available HPLC equipment The sample's characteristics, such as hydrophilicity, hydrophobicity, and protolytic functions, play a crucial role in determining the optimal HPLC method Notably, the steroids examined in studies I and II are neutral compounds.
Inpaper I the desired product was not available as a standard, which made method development more difficult
Chemical analysis typically involves examining a small fraction of the material to determine its composition, necessitating that this sample accurately represents the bulk material for valid results The process of obtaining a representative sample is known as sampling Prior to High-Performance Liquid Chromatography (HPLC), some samples may require pre-treatment to eliminate interferences or concentrate analytes, which can sometimes be more intricate than the HPLC separation itself Clearly defining the separation goals at the outset of method development is crucial, whether the aim is to resolve all components or to separate analytes from impurities and degradation products without further distinction among the latter Selecting a detector sensitive to all relevant sample components is essential before injection; variable-wavelength ultraviolet detectors are commonly preferred for their versatility Ultimately, the finalized HPLC procedure must fulfill all initial objectives and undergo validation for quantitative analysis.
The objective of Paper I was to create a method for effectively separating the product, substrate, degradation products, and impurities from the fermentation medium to facilitate quick product identification Meanwhile, Paper II focused on developing a precise quantitative analysis method for both products and substrates Additionally, the strategies outlined in Paper I and the guidelines provided in Paper II were intended to be applicable for standard fermentation laboratories.
Q UALITATIVE A NALYSIS
Reference Substance Available
The simplest qualitative analysis involves a comparison of the retention times between a chromatographic peak containing an unknown compound and peaks obtained for reference samples using more than one stationary phase.
Without Reference Substance
Structural identification of unknown compounds can be challenging, especially when reference compounds are unavailable While HPLC with UV detection allows for on-line measurements and monitoring at multiple wavelengths, it often falls short for safe identification Over the past 15 years, the advancement of combined liquid chromatography-mass spectrometry (LC-MS) has emerged as a crucial tool for qualitative analysis In cases where reference compounds are lacking, techniques such as NMR may be necessary to definitively identify unknown substances, including isomers of target compounds.
Q UANTITATIVE A NALYSIS
In quantitative analysis, the primary objective is to accurately measure the concentration of analyte molecules within a sample Typically, two distinct analytes at the same concentration yield varying detector responses in chromatography, necessitating the measurement of detector responses for known concentrations of each analyte to ensure precise results.
A standard curve is a graphical representation that illustrates the relationship between detector response and analyte concentration in a sample For accurate quantification analysis, three prevalent calibration methods are utilized: external standard calibration, internal standard method, and standard addition method.
The external standard calibration method is a straightforward approach that is best suited for simple sample preparations and situations where there are minimal instrumental variations While it lacks precision, it is commonly applied in pharmaceutical product analysis due to the uncomplicated nature of the matrices involved and the ease of sample preparation However, this method is not recommended for complex matrices.
To create a standard curve for analyte concentration analysis, known concentration standard solutions are prepared and injected into the chromatographic column The areas or heights of the resulting peaks in the chromatogram are measured and plotted against the injected amounts Unknown samples are then prepared and analyzed using the same method, allowing their concentrations to be determined from the calibration plot The process is referred to as "external standard calibration," indicating that the standards are analyzed in separate chromatographic runs from the unknown samples.
The internal standard (I.S.) method enhances accuracy in quantitative analysis by compensating for instrumental and sample preparation errors, such as dilution and extraction losses Selecting an appropriate I.S that mimics these variations is crucial for improving both accuracy and precision The I.S should closely resemble the analyte but remain distinguishable during chromatographic separation Standard curves are created by spiking blank samples with known analyte concentrations and a constant I.S concentration, with both standard and unknown samples processed concurrently The calibration curve plots the ratios of analyte to I.S peak areas against analyte concentrations An effective I.S must be well-resolved from other compounds, absent from the sample, similar in retention time, of high purity, stable, and structurally comparable to the analyte, ensuring it mirrors the analyte throughout all sample preparation stages.
The internal standard method, commonly used in chromatography, has gained popularity in quantitative HPLC-MS techniques due to its effectiveness in selecting compounds with similar structures.
In the method development outlined in paper II, we aimed to quantify both the fermentation product and the substrate, evaluating both external and internal standard methods The internal standard method was ultimately selected due to its superior precision Initially, structurally related compounds were tested as internal standards in the quantitative analysis described in paper II, following the established internal standard protocol for bioanalytical approaches However, there are no specific recommendations for selecting internal standards in biotechnological analysis Therefore, paper II investigates the applicability of the bioanalytical internal standard protocol to biotechnological contexts.
The standard addition method is a valuable technique used when suitable blank matrices are unavailable, such as in the analysis of endogenous compounds in body fluids This method involves adding varying amounts of the analyte to an unknown sample that already contains an unknown concentration of the analyte After conducting chromatographic analysis, the peak areas or heights are plotted against the added concentrations By extrapolating this calibration plot, the original unknown concentration of the analyte can be determined accurately.
A standard addition method that possesses even greater accuracy and preci- sion is obtained if one incorporates an internal standard [48].
S AMPLE P REPARATION
Extraction
The primary techniques for sample preparation include liquid-liquid extraction (LLE) and solid-phase extraction (SPE) These methods focus on isolating the target analyte from the sample matrix, minimizing the presence of interfering substances during the subsequent analytical separation process.
Liquid-liquid extraction (LLE) relies on the partitioning of an analyte between an aqueous phase and an immiscible organic phase, influenced by factors such as pH, ionic strength, and the type of organic solvent used Adjusting the pH of the aqueous phase can enhance the recovery of proteolytic analytes into the organic phase, while the choice of organic solvent and the volume ratio between the phases also play critical roles in optimizing recovery Hansch and colleagues have developed a predictive method for estimating distribution constants, which has proven to be effective LLE is a well-established technique for bioseparations and has recently evolved into semi-automated and fully automated systems using 96-well LLE plates and robotic liquid-handling workstations, significantly improving sample preparation time and throughput.
Solid-phase extraction (SPE) serves as an effective alternative to liquid-liquid extraction (LLE), where analytes are partitioned between a solid and a liquid phase During the SPE process, interfering compounds are removed from the solid adsorbent, allowing for the subsequent desorption of analytes using an eluting solvent Various sorbents and formats are available for SPE, including normal-phase, reversed-phase, ion exchange, and restricted access, making the system highly automatable To isolate products from the fermentation matrix, both SPE and LLE were assessed in the studies presented in papers I and II.
S EPARATION
For effective detection, it is essential to separate analytes in a mixture, and chromatography has emerged as the leading analytical technique for this purpose This method relies on the differing rates of migration of substances through a column, influenced by their partitioning between a stationary phase and a mobile phase Chromatographic techniques are categorized based on the physical state of the mobile phase, including gas chromatography (GC), supercritical fluid chromatography (SFC), and liquid chromatography (LC) Additionally, the diverse characteristics of stationary phases play a crucial role in further classifying these methods.
Liquid chromatography (LC) is a crucial analytical tool used for examining a variety of substances across different matrices It can be classified based on how the solute interacts with the stationary phase, including methods such as adsorption chromatography, partition chromatography, ion-exchange chromatography (IEC), size exclusion chromatography (SEC), and affinity chromatography.
Liquid chromatography initially utilized highly polar stationary phases with nonpolar solvents as mobile phases, known as normal-phase liquid chromatography (NPLC), exemplified by chromatography on bare silica In contrast, reversed-phase liquid chromatography (RPLC) employs a nonpolar stationary phase, typically a hydrocarbon, and a relatively polar mobile phase, where the most polar component is eluted first due to its higher solubility in the mobile phase RPLC is currently the predominant mode of liquid chromatography, often using octadecyl-bonded silica (C18) or octyl-bonded silica (C8) as stationary phases The retention mechanism of solute molecules in liquid chromatography is complex and influenced by various intermolecular interactions Many researchers suggest that adsorption and partitioning are the primary mechanisms governing retention in C18 columns.
In chromatography, substances A and B are separated within a column filled with solid particles and a mobile phase When a sample mixture is injected at the top, the components distribute themselves between the mobile and stationary phases As the mobile phase flows continuously down the column, it carries the sample, allowing for further distribution If solute B interacts more strongly with the stationary phase than solute A, it will spend less time in the mobile phase and consequently move down the column at a slower rate than solute A.
After solute A has eluted from the column, solute B follows, indicating that a chromatographic separation of the mixture has taken place When a detector is positioned after the column to sense the analyte's presence, its signal can be plotted against time, resulting in a graphical representation known as a "chromatogram."
Figure 4 (a) Illustration of the chromatographic separation of a mixture of compo- nents A and B in a column (b) The chromatogram of the separation of the mixtures
Successful chromatography hinges on the optimal balance of intermolecular forces among the solute, mobile phase, and stationary phase Achieving the right polarities for these three components is crucial for effective separation within a reasonable timeframe.
Commercial HPLC columns vary significantly among suppliers, with differences in plate number, band symmetry, and retention In the first study, the optimal HPLC column for separating various steroids was identified, followed by the need to optimize the mobile phase The composition of the mobile phase can be easily modified in HPLC separations, allowing for adjustments in retention through changes in solvent strength The second study evaluated two different organic solvents as modifiers in the mobile phase for the effective separation of steroids on a C18 stationary phase.
D ETECTION
UV and Diode Array Detection
UV detectors are the most prevalent type of HPLC detectors due to their robustness, affordability, and ease of use, as many solutes absorb light in the UV range Traditional UV detectors measure absorbance at a single wavelength, requiring mechanical adjustments of the monochromator, which can lead to measurement inconsistencies In contrast, diode-array detectors (DAD) can simultaneously measure multiple wavelengths without moving parts, eliminating mechanical errors and drift over time This operational efficiency distinguishes DADs from conventional single-wavelength detectors.
A diode array detector (DAD) enhances chromatographic analysis by allowing operators to display chromatograms at any wavelength between 190 and 400 nm, along with the UV spectra of each eluting peak This capability provides more comprehensive information about sample composition compared to single wavelength detectors, which are primarily used for quantitative analysis DADs facilitate both quantitative and qualitative assessments, enabling peak identification and purity analysis through the collection of UV spectra for resolved peaks By comparing recorded signals, analysts can effectively identify peaks within the chromatogram.
The UV spectra of a standard can be compared with a sample peak by overlapping the two spectra, allowing for identification of components in a chromatogram even in the absence of a standard This method can indicate relationships between compounds, such as small chemical modifications from fermentation processes, degradation products, or metabolites The DAD instrument's software typically includes a matching factor tool to assess the degree of overlap between acquired spectra In paper I, HPLC-DAD was utilized for rapid preliminary qualitative analysis of the fermentation process.
Figure 5 Simplified diagram of: (A) a conventional single wavelength detector and (B) a multi wavelength detector
Modern Diode Array Detectors (DADs) are versatile tools widely utilized in various applications, particularly in drug production control where the swift screening and identification of unknown impurities are essential Their ability to compare unknown high-resolution UV spectra with established library spectra makes them invaluable for tasks such as identifying steroidal glycosides in seeds, peptide mapping, analyzing sulfamethazine in animal tissues, and detecting pesticides in human biological fluids Despite limited literature on their use in biotransformation monitoring until the late 1990s, recent years have seen an increase in HPLC-DAD applications for wine fermentation and ethanol production While DADs offer rapid access to product information, their primary limitation lies in the lack of detailed structural insights compared to mass or NMR spectra.
Mass Spectrometry
Mass spectrometry (MS) is a widely used detection technique that provides quantitative and qualitative information about the components in a mixture
In qualitative analysis, determining the molecular weight of an unknown compound is crucial, and mass spectrometry (MS) is an effective technique for this purpose Additionally, MS offers greater sensitivity for quantification compared to UV detection methods.
An MS detector comprises three key components: the ionization source, which generates ions; the mass analyzer, which separates ions based on their mass-to-charge ratio (m/z); and the electron multiplier, which detects the ions Various ion sources employ different ionization techniques to produce charged particles.
Three prominent ionization techniques include electrospray ionization (ESI), atmospheric pressure chemical ionization (APCI), and matrix-assisted laser desorption (MALDI) Among these, electrospray ionization is the most commonly employed method in liquid chromatography-mass spectrometry (LC-MS) due to its versatility in soft ionization of a wide range of analytes The principle of ESI is illustrated in Figure 6.
Fig 6 Illustration of the electrospray process in positive ion mode The picture was kindly provided by Andreas Pettersson, Depertment of Analytical Chemistry, Upp- sala University
In Electrospray Ionization (ESI), ions are generated by applying an electric field to the liquid phase, resulting in the creation of charged droplets As the solvent evaporates, the droplets decrease in size while maintaining their charge, leading to an increased charge-to-volume ratio that ultimately produces gas-phase ions Although the exact mechanism behind the formation of gas-phase ions from small droplets remains somewhat unclear, it has minimal impact on the application of ESI in liquid chromatography-mass spectrometry (LC-MS).
Commonly available mass analyzers include quadrupole, ion-trap, time-of-flight (TOF), and magnetic sector analyzers In this study, an ion-trap mass analyzer was utilized, showcasing its effectiveness in identification across chemistry, biochemistry, and biotechnology fields Ion trap mass spectrometry (MS) excels in advanced qualitative analysis, offering scans up to the MS 5 range, which allows for the generation of multiple daughter ions and the acquisition of MS n spectra This capability yields extensive structural information, enabling the comparison of first and second-generation product ion spectra of known reference compounds with those of unknown samples In this research, MS 5 spectra were analyzed for various steroids, although the product 9D-OH-PS was not available as a reference substance.
11Į-hydroxyprogesterone (11Į-OH-PS) was used as a qualitative reference whose MS spectra showed convincing similarities (but not identical) with the spectra for our fermentation-derived product.
NMR
Nuclear magnetic resonance (NMR) spectroscopy is an essential analytical tool for chemists, enabling the determination of compound structures through nondestructive spectroscopic analysis This technique reveals the number of atoms, primarily carbons and protons, and their connectivity, thereby elucidating molecular conformations By measuring and interpreting NMR spectra from liquids, chemists can gather critical structural information Proton NMR offers high sensitivity, requiring only 1-10 mg of sample, while carbon NMR demands larger quantities, typically around 20 mg, which may not always be available from laboratory-scale biotechnological syntheses.
V ALIDATION
Precision and Accuracy
The precision of an analytical method refers to the closeness of multiple measurements of an analyte obtained from repeated applications of the procedure to different aliquots of a uniform biological matrix This precision is quantified using the coefficient of variation (C.V.), also known as relative standard deviation (RSD) RSD can be categorized into three types: repeatability (intra-day precision), intermediate precision (inter-day precision), and reproducibility (precision between laboratories).
Repeatability in laboratory analysis is achieved when the same analyst uses the same equipment on the same day, with a minimum of five determinations at three different concentrations (low, medium, and high) as per FDA guidelines The ICH allows for repeatability to be measured by three determinations at varying concentrations or six determinations at a constant concentration, typically for pharmaceutical products Acceptance criteria for precision vary by analysis type, with pharmaceutical quality control requiring precision better than 2%, while bioanalytical applications should have precision values better than 15% at each concentration level, except for the lower limit of quantification (LLOQ), which should not exceed 20% Intermediate precision assesses variations due to different analysts and instruments over time, while reproducibility measures precision across different laboratories.
Accuracy in analytical methods refers to how closely the measured value of analytes aligns with their true value, often termed trueness This can be assessed by analyzing samples with known concentrations and comparing the results to the actual values The FDA recommends that accuracy in bioanalysis be established through at least five determinations across three concentrations (low, medium, and high) within the expected range The mean measured value should fall within 15% of the true value, except for the Lower Limit of Quantification (LLOQ), which should be within 20% Both precision and accuracy can be derived from the same set of analytical experiments.
The precision and accuracy of 9Į-OH-PS were evaluated through six replicates analyzed across three different days and at three varying concentrations, demonstrating both intra-day and inter-day reliability.
Selectivity/Specificity
In analytical chemistry, the terms selectivity and specificity are often confused Selectivity refers to a method's ability to respond to multiple substances while distinguishing the analyte's response from others, whereas specificity indicates a response to only one analyte In chromatography with UV detectors, methods typically exhibit selectivity rather than specificity To evaluate a method's selectivity, it is essential to test blank samples with and without analytes to identify potential interferences Ensuring high selectivity is crucial for accurate quantification of analytes.
In paper II, the selectivity of the chosen system was investigated by comparing chromatograms of pure matrix solution with matrix solution to which analytes had been added.
Limit of Detection and Quantification
It's essential to differentiate between the limit of detection (LOD) and the lower limit of quantification (LLOQ) in analytical chemistry The LOD refers to the minimum concentration of an analyte that produces a peak height three times the baseline noise in a chromatographic system, indicating that the analyte can be detected but not necessarily quantified In contrast, the LLOQ is the lowest concentration at which an analyte can be accurately and precisely quantified, typically requiring an accuracy of 80-120% and a precision within 20% Understanding these distinctions is crucial for accurate analytical measurements.
% [16] Both the LLOQ and LOD were investigated in paper II.
Linearity and Range
The linearity of an analytical method reflects its capacity to produce test results that are directly proportional to the analyte concentration within a specified range For accurate calibration, standards should be prepared in the same matrix as the samples being analyzed, ideally consisting of six to eight non-zero standard samples It's crucial to exclude the blank and zero samples from the calibration curve to avoid skewed results Preparing standard samples independently enhances error detection compared to using serial dilutions, which, while providing a better correlation coefficient, may lead to misleading regression results if errors occur in the highest standard A valid standard curve should feature a linear regression equation with an intercept close to zero, and the linear correlation coefficient (r) must be at least 0.95, as per FDA guidelines, although a threshold of 0.99 is often recommended by experts.
A high linear correlation coefficient does not guarantee a linear standard curve, as lower concentration standards may deviate from linearity despite a strong r value, according to Bildlingmeyer [85] To ensure accuracy, the linear coefficient should be supported by a graph plotting response against logarithmic sample concentrations, with deviations on the y-axis not exceeding 5% [85] The range of an analytical method is defined as the validated concentration interval that meets the criteria of accuracy, precision, and linearity.
In paper II the linearity was evaluated by both the r-values and with the logarithmic linearity plot approach.
Recovery
Effective sample preparation aims for high recovery of analytes from the matrix, making it a crucial aspect of the extraction process Absolute recovery is defined as the ratio of the response from a spiked sample treated through the complete analytical procedure to that of a non-biological sample spiked in an aqueous solution and directly injected into the chromatographic system In contrast, relative recovery compares the responses of extracted spiked samples in a matrix to those of extracted spiked pure samples in an aqueous solution Both absolute and relative recovery metrics are essential for determining whether sample losses during extraction result from matrix effects or inefficient extraction methods.
To ensure accurate recovery in analytical processes, a minimum of six determinations across three concentrations (low, medium, and high) is essential, with absolute recovery ideally exceeding 90% and relative recovery surpassing 95% While prioritizing selectivity may occasionally necessitate lower recovery rates, maintaining acceptable sensitivity, precision, and accuracy is crucial The internal standard must exhibit a recovery comparable to the analyte, remaining within 15% of the analyte's recovery Research detailed in paper II examined both absolute and relative recoveries for the product, substrate, and internal standard across various fermentation media and time intervals, also exploring the potential of ultrasonication to enhance recovery rates.
Stability
In method validation, it is crucial to investigate the stability of the analyte under different conditions, as degradation can occur Ensuring that the analyte remains stable from the time of sample collection to analysis is essential Stability experiments should mimic real-world scenarios that are likely to arise during sample handling and analysis.
[16] The following five stability conditions are advisable to investigate for the analytes and the I.S.: x Short-Term Temperature Stability
The stability of the analyte in the matrix at ambient temperature must be assessed, requiring the investigation of at least three aliquots for both low and high concentrations.
“benchtop stabilliy” x Post-Preparative Stability
The stability of the analyte in the final extract during the expected maximum analysis time, which for automatic injections can be up to
48 h, should be assessed x Freeze and Thaw Stability
The stability of the analyte after at least three thaw-and-freeze cycles should be determined x Long-Term Stability
To ensure accurate analysis, it is crucial to assess the stability of frozen samples for a duration longer than the interval between sample collection and the final analysis Additionally, the stability of stock solutions must also be evaluated to maintain the integrity of the results.
The stability of the stock solution of the standard should be evaluated at room temperature
Inpaper II the stability of the stock solutions of PS, 9D-OH-PS, 11D OH-PS and binaphtol were investigated; also the short-term temperature stability, post-preparative stability and long term stability were investigated for the product and the internal standard, binaphtol.
Robustness
Robustness tests are essential for evaluating how minor alterations in operational parameters impact analysis results, despite the lack of mention in FDA guidelines and most validation protocols For methods employed over extended periods and across various laboratories, assessing robustness becomes crucial It is important to examine the effects of small changes in experimental conditions, such as variations in buffer pH, organic modifier concentration in the mobile phase, ambient temperature, and detection wavelength.
Quality Control
Once the analytical method is validated for routine use, it is essential to regularly monitor its accuracy and precision to ensure consistent performance This involves analyzing a set of quality control (QC) samples, prepared from different weightings than those used for the standard curve, in each run Typically, these QC samples include duplicates at three concentration levels: low, medium, and high To meet quality standards, at least four out of six QC samples must fall within 20% of their respective nominal values, with at least one sample at each concentration level Additionally, a standard curve should be processed during every run to maintain reliability.
Validation of Biotechnological Synthesis
The suggested tolerances for the validation parameters in the FDA recom- mendations for bioanalytical methods [16] are rather wide, with C.V values
For pharmaceutical product analysis, the requirements for coefficient of variation (C.V.) values are stringent, typically less than 2% This high level of precision is achievable due to the simpler matrix involved and the ability to select analyte concentrations freely, allowing for the avoidance of extremely low or high concentrations.
The lack of detailed recommendations for analytical procedures in biotechnological drug production, unlike the established FDA guidelines for bioanalytical methods, is noteworthy Validating the quantification of both substrates and products at specific time intervals is crucial for accurately calculating process kinetics, including substrate conversion coefficients and production rates This study aims to explore the applicability of FDA's bioanalytical validation guidelines to biotechnological synthesis and identify necessary modifications if they are not directly applicable.
3 Determination of Adsorption Isotherm Parameters
In this section a short overview is given of preparative chromatography and the determination of adsorption isotherm parameters - single and competitive
- to be used for computer-assisted optimization of separations.
P REPARATIVE C HROMATOGRAPHY
Nonlinear Chromatography
In linear chromatography, the adsorption isotherm is characterized by a straight line that begins at the origin, indicating that the amount of solute adsorbed on the stationary phase at equilibrium is directly proportional to its concentration in the mobile phase This linear relationship ensures that retention time and band shapes remain consistent, regardless of the sample composition or quantity Consequently, linear conditions are commonly observed in analytical chromatography.
Nonlinear chromatography occurs when the concentration of a component in the stationary phase is not proportional to its concentration in the mobile phase, leading to adsorption isotherms that are influenced by all other components in the sample mixture This results in variations in retention time, peak height, and band shape, which are affected by both the composition and quantity of the sample Such complexities are typical in preparative applications, as the interdependence of individual band profiles arises from the varying amounts of each component adsorbed onto the stationary phase.
Preparative chromatography is typically performed under mass and volume overloaded conditions to enhance product throughput In volume overloading, sample concentrations remain within the linear region of the isotherm, and the volume is increased to optimize throughput However, this method often leads to underutilization of the column and low throughput Conversely, mass overloading involves raising sample concentrations beyond the linear adsorption region, resulting in asymmetric band profiles characterized by self-sharpening fronts and tailing rear boundaries, particularly in the common "Langmuirian" isotherm To achieve maximum throughput, a combination of both volume and mass overloading is generally employed Additionally, large-scale chromatography consumes significant amounts of costly solvents, highlighting the critical need for process optimization.
Simulated Moving Bed
Batch chromatography is typically used for purifying small amounts of pure substances, ranging from a few milligrams to about 100 grams For larger quantities, from several hundred grams to kilograms, the simulated moving bed (SMB) process is a more efficient alternative Continuous SMB systems are gaining popularity for the purification of pharmaceuticals and specialty chemicals, including enantiomers and proteins, due to their advantages such as reduced solvent consumption, increased productivity and purity, and lower investment costs compared to traditional batch elution chromatography However, a notable limitation of SMB chromatography is that it can only separate compounds into two fractions.
An SMB (Simulated Moving Bed) system typically comprises 8 to 12 identical columns arranged in series, with four individually operable valves positioned between each column A recycling pump circulates the mobile phase flow through all columns, while two additional pumps continuously inject the feed and fresh eluent Additionally, two pumps are responsible for withdrawing the raffinate and extract flows, ensuring efficient separation processes.
Optimizing SMB chromatography requires careful selection of various parameters, including column diameter, length, and fluid flow rates, making computer simulations essential for design and process optimization Nonlinear conditions complicate this optimization process, as empirical methods often fail under overloaded scenarios This challenge is particularly significant in chiral separations, where chiral stationary phases (CSPs) typically exhibit lower saturation capacities compared to non-chiral columns, highlighting the necessity of computer-assisted optimization in these cases.
Preparative Chromatography in Biotechnology
Biotechnologically produced therapeutic substances undergo rigorous downstream processing to achieve high purity levels Preparative chromatography has become the primary purification method in modern biotechnology for isolating valuable proteins from complex mixtures The advancement of specialized stationary phase materials with unique selectivity has greatly enhanced the capabilities of large-scale chromatographic bioseparations.
Biotechnology employs various preparative chromatographic separation techniques, with the most prevalent being ion-exchange chromatography (IEC), hydrophobic interaction chromatography (HIC), reversed-phase liquid chromatography (RPLC), size exclusion chromatography (SEC), immobilized metal-affinity chromatography (IMAC), lectin-affinity chromatography, biospecific-affinity chromatography, and dye-affinity chromatography.
Reversed-phase liquid chromatography (RPLC) is the predominant method for separating small organic molecules and is effective for preparative peptide separations However, its application in protein purification is limited due to the risk of protein denaturation Despite this challenge, there are notable instances of successful large-scale protein purification using RPLC In this thesis, the focus is on the small molecules (molecular weight