Application of Ion Chromatography with Electrochemical Detection in Optimization and Control of Fermentation and Cell Culture

May 2, 2011
Lokesh Bhattacharyva
Volume 2011 Supplement, Issue 4

An overview of applications in the analysis of nutrients and metabolites.

Drugs, biological therapeutics, and vaccines manufactured under cGMP must be controlled at all stages of the production process. The final release testing is not in itself sufficient to guarantee quality, safety, purity and efficacy of the final product (1). It is necessary to ensure that the manufacturing operations perform as intended within the established specifications. This implies that any given process is well characterized, adequately controlled, and reproducible. Fermentations are complex processes used in the manufacture of recombinant proteins and peptides, vaccines, and other therapeutic products. The performance of fermentation (and cell culture) processes are critical to the quality and consistency of production of the therapeutic products and are influenced by parameters, such as pH, temperature, dissolved oxygen, as well as nutrients and other components necessary for cell growth, and metabolites. The nutrients include carbohydrates, amino acids, and other organic compounds, which act as carbon sources essential for cell growth and product synthesis. Inorganic cations (metal ions) also play critical roles for optimum activities of enzymes and in cell growth. The metabolic products sometimes act to inhibit the product synthesis and affect the desired yield. Thus, measurements of nutrients and metabolites in fermentation broths (and culture media) are important for understanding the roles they play and microbial physiology, as well as for the development, optimization, characterization, and control of the process. Such measurements also permit improvement of the process performance by providing the ability to determine the need for supplementation, as well as to the development of defined media to replace complex media formulations. Most of the nutrients and metabolites, including inorganic ions, carbohydrates, alditols, alcohols, and aliphatic carboxylic and amino acids are ionic or polar in nature, and do not have chromophores necessary for analysis by absorption measurements. Ion chromatography (IC) with electrochemical detection (ED) presents a suitable technique for the determination of these components. The technique usually requires no sample preparation other than dilution with water. Other chromatographic methods often require complex pre- or postcolumn derivatization for the detection of analytes.

This article will provide a brief overview of IC and ED, as well as their applications in the analysis of nutrients and metabolites in fermentation broths and neomycin in selective culture media.

Ion chromatography

IC is a form of high-performance liquid chromatography (HPLC), which involves separation based on ionic interactions among ionic or polar analytes, ions present in the mobile phase (eluent), and ionic functional groups on chromatographic support (stationary phase). This can lead to two distinct mechanisms of separation based on the type of interactions—(a) ion exchange in which the analyte ions are adsorbed to and desorbed from the stationary phases repeatedly leading to separation, and (b) ion exclusion due to repulsion between similarly charged analyte ions and the ions on the chromatographic support. Separation based on ion exchange has been the predominant application of IC to-date.

In ion-exchange chromatography, ionic functional groups on the chromatographic support have charges opposite to those of the analyte ions. That is, a column for separation of cations, a cation-exchange column, contains negative ions. This technique is widely used in the analysis of cations, including metal ions, anions, mono- and oligosaccharides, alditols (sugar alcohols), aminoglycosides (antibiotics), amino acids, peptides, organic acids, amines, alcohols, glycols, thiols, and other polar molecules. Ion-exclusion chromatography uses strong cation- or anion-exchange columns to separate ionic, polar, weakly polar, and apolar analytes. The ionic analytes experience strong repulsion from similarly charged functional groups on the chromatographic support and are eluted first, with little or no separation. An apolar analyte experiences no repulsion and is, therefore, eluted late from the column. Polar analytes experience less repulsion than an ion but more than an apolar molecule and are separated in order of their polarity (more polar compound eluting earlier). This technique has been used in the analysis of organic acids, alcohols, glycols, and sugars.

Any suitable detector can be used for the detection and quantitation of analytes by IC. However, two forms of ED, suppressed conductivity detection (SCD) and pulsed amperometry detection (PAD), have distinct advantages over other detection methods. They provide the selectivity and sensitivity necessary for the measurement of critical components of fermentation broths without any pre- or post-column derivatization or complex sample preparation procedures.

When a constant voltage is applied across two electrodes between which the effluent from a column flows, a current is generated because the effluent contains ions and polar molecules. The current is proportional to the conductivity of the solution, which, in turn, is proportional to the concentration of ionic species in solution, leading to the detection and quantitation of polar analytes by a conductivity detector. The problem, however, is that the conductivity of the mobile phase is significantly higher than the conductivities of the analytes, simply because the concentrations of ions in the mobile phase is 104–105 times higher than that of the analytes. SCD permits detection and quantitation of analytes by reducing the conductivity of the mobile phase (baseline conductivity) to nearly zero. It also enhances the conductivity of the analytes, thereby enhancing sensitivity by about an order of magnitude (2). The peaks of apolar and weakly polar molecules, such as sugars (except the ionic sugars, e.g., sialic acids, sugar-phosphates), alditols, alcohols, and glycols are not detected or remain close to the noise level, providing selectivity of detection to the ionic and polar components of the media.

PAD is specific for compounds, such as, carbohydrates, amino acids, alditols, glycols, and alcohols, which can be oxidized at a selected potential (applied voltage), leaving most of the other compounds undetectable, and thus, providing selectivity of detection.

A detailed discussion on the mechanisms of ion-exchange and ion-exclusion chromatographies as well as those of SCD and PAD is beyond the scope of this article. Interested readers are encouraged to read books and publications that provide details of the principles and mechanisms of IC, SCD and PAD, including selection of columns and detectors for different types of applications (3–6).

Cations and anions

Fermentation broths typically have several cations and anions, including amino, carboxylic and other organic acids, as well as amines and other organic bases. However, concentrations of inorganic cations and anions are significantly higher than their organic counterparts. Thus, inorganic ions show significant peaks by IC when fermentation broths are diluted 1000-fold or more, while their organic counterparts are reduced to undetectable levels. This provides necessary selectivity for analysis.

The first report of the determination of cations and anions using IC to follow a fermentation process of methanotrophic bacteria was published in 1992 (7). Using OmniPac PAX-500 anion-exchange column and SCD, the authors followed several anions over the course of fermentation and found acetate to accumulate when natural gas was used as the substrate, which inhibits the growth of the methanotrophic bacteria. Mono- and divalent cations were monitored by SCD using IonPac CS10 cation-exchange column eluted isocratically with 40 mM HCl–4 mM DL-2,3-diaminopropionic acid (DAP) as the eluent. The results indicated that the concentration of Mg2 + must be above 60 ppb to maintain optimum growth of the bacteria.

Robinette et al. (8, 9) reported quantitation of eight mono- and divalent cations in fermentation broths for recombinant and pathogenic microorganisms, as well as in mammalian and insect cell cultures in complex and defined media. The authors employed IonPac CS10 and CS12 cation-exchange columns eluted isocratically with 20 mM HCl – 4 mM DAP and 20 mM methanesulfonic acid (MSA), respectively. The cations are detected by SCD. The sample preparation involves simple 1000–2500 fold dilution of filtered broth with water. The reports indicate good precision and accuracy, and a wide linear range for each cation, with detection limits ≤1 µg/mL.

Figure 1: A time course of analysis of cations in an insect cell culture using a complex medium with glucose and sucrose as the primary carbon sources. Peaks: 1=sodium, 2=ammonium, 3=potassium, 4=magnesium, and 5=calcium. Column: IonPac CS12; elution: 20 mM MSA, isocratic, 1.0 mL/min. (Reprinted from Ref. 9 with permission from Elsevier)

Figure 1 shows a time course of an insect cell culture using a complex medium. The chromatograms show that the levels of sodium, potassium, magnesium, and calcium remained virtually unchanged throughout the fermentation, while ammonium concentration increased approximately 2-fold. Similar investigations were reported for S. cerevisiae (eukaryotic) grown in a chemically defined medium, and E.coli (prokaryotic) and a mammalian cell culture in complex media containing glucose as the primary carbon source.

Figure 2: Resolution of carbohydrates, alditols, alcohols, and glycols commonly found in fermentation broths by (A) CarboPac MA1 column and (B) CarboPac PA1 column. (Reprinted from Ref. 12 with permission from Dionex)

Aliphatic carboxylates (e.g., acetate, lactate, pyruvate) are metabolites that often reduce fermentation yields. In an Application Note (10), Dionex reported the analysis of aliphatic carboxylates and inorganic anions in the fermentation broths of S. cerevisiae in YPD medium and E.coli in LB medium using IonPac AS11-HC column eluted with 1-60 mM NaOH and SCD. The results show well resolved peaks with good precision and accuracy, wide linear range and low detection limit for each anion.

Figure 3: A time course of analysis of carbohydrates in H. influenzae cultivation using complex medium with glucose as the major carbon source. Peaks: 1=glucose, 2=fructose, 3=ribose; 4=maltose. H=hours. Elution: 150 mM NaOH, isocratic, 1.0 mL/min. (Reprinted from Ref. 11 with permission from Elsevier)

Carbohydrates and alcohols

Carbohydrates are carbon sources essential for cell growth and product synthesis, while alditols and alcohols are metabolites. High Performance Anion-Exchange Chromatography (HPAEC) coupled with PAD (HPAEC–PAD) or Integrated PAD (IPAD), which is an improved PAD application, is the most widely used chromatographic technique for the analysis of carbohydrates and alditols (3–6). The CarboPac MA1 column has the ability to analyze mono- and disaccharides, alditols, and alcohols present in fermentation broths (see Figure 2A) in a single run. However, the chromatography takes as long as 60 min for each run and uses a high concentration of NaOH as the mobile phase. The CarboPac PA series of columns (e.g., PA1, PA10, PA20) take significantly less time and use low NaOH concentration. But alcohols and alditols are eluted early and are poorly resolved (Figure 2B). Therefore, the PA series columns are better suited for mono- and disaccharide analysis. Thus, the choice of column depends on the analytes to be monitored in the media. Herber and his colleagues (8, 11) used a CarboPac PA1 column to monitor microbial fermentations using chemically defined and complex media. The samples were diluted 50 fold with water and eluted from the column with NaOH (isocratic and gradient) to monitor ethanol, glycerol, galactose, glucose, mannose, fructose, raffinose, ribose, and lactose. Carboxylic acids and inorganic ions are transparent to PAD. Most of the other media components were below the detection limit at this dilution. Indeed, only proline, arginine and lysine exhibited any noticeable detector response (see later for the detection of amino acids by PAD). Figure 3 shows a time-course study of carbohydrates in a H. influenzae fermentation in a complex medium with glucose as the major carbon source. Figure 4 shows the chromatograms of the analysis of S. cerevisiae fermentation broth using CarboPac MA1 (12). Glucose is the most predominant component of the media at 0 h. At the end of fermentation, no glucose was detected and the predominant components are the metabolites.

Figure 4: Chromatograms of the analysis of S. cerevisiae fermentation broth using CarboPac MA1; (A) 0 h and (B) 24 h. (Reprinted from Ref. 12 with permission from Dionex)

Amino Acids

The amino acid compositions of cell culture and fermentation broths can be determined by HPAEC-PAD without sample derivatization (13–15). However, the presence of carbohydrates and alditols in the media may interfere. By optimizing the elution method, Genzel et al. (13) reported that the amino acids were well resolved and also resolved from the predominant carbohydrate, glucose, in GMEM media used in the production of inactivated influenza vaccines (see Figure 5). The authors reported good precision, accuracy, linear range, and low quantitation limit for each amino acid. Similar results were also reported for the supernatants of YPD broth, LB broth, Minimum Essential Eagle's Medium and a serum-free, protein-free hybridoma medium (14, 15).

Figure 5: Amino acid analysis of (1) a standard amino acid mixture containing 277.5 µM glucose;(2) GMEM medium; (3) GMEM medium + 1% (v/v) peptone; (4) GMEM medium+10% (v/v) FCS + 1% (v/v) peptone. Norleucine was added as the internal standard. Column: AminoPac PA10; elution: as per the procedure described in Ref. 16. (Reprinted from Ref. 13 with permission from Elsevier)

Antibiotics

Selective media are used for the growth of select microorganisms. For example, if a microorganism is resistant to a certain antibiotic, then that antibiotic can be added to the medium in order to prevent other cells, which do not possess the resistance, from growing. Selective growth media for eukaryotic cells commonly contain neomycin to select cells that have been successfully transfected with a plasmid carrying the neomycin resistance gene as a marker. Generally, neomycin is assayed following the procedure described in USPNF General Chapter <81>, Antibiotics—Microbial Assay (17). However, the procedure is tedious and labor intensive, has poor precision, and requires days to complete. Neomycin is an aminoglycoside containing a carbohydrate moiety and, like other carbohydrates, can be assayed using HPAEC–PAD. Figure 6 shows the HPAEC–PAD chromatograms of two selective media containing neomycin. The media also contain glucose, sucrose, and amino acids at concentrations 100-1000 fold higher than that of neomycin. These components also show peaks in HPAEC–PAD. Thus, the resolution of neomycin was a challenge, which was accomplished using a CarboPac PA1 column and a linear elution gradient from 1.0 mM to 3.0 mM NaOH over 25 min.

Figure 6: HPAEC-PAD chromatograms of a neomycin standard and two selective culture media containing neomycin. Column: CarboPac PA1; flow rate: 0.5 mL/min. The effluent from the column was mixed with 100 mM NaOH before entering into the detector for baseline stabilization.

Conclusion

Monitoring nutrients and metabolites in cell culture media and fermentation broth provides critical pieces of information on their roles in fermentation, and in optimizing, characterizing, and controlling the process to produce safe and efficacious therapeutic products in a reproducible manner. The results presented in this review show that the unique selectivity and sensitivity of IC with ED permit determination of fermentation broth components. This is particularly important for those components which do not have suitable chromophores for detection by absorption measurement. The sample preparation involves simple filtration and dilution of fermentation broth with water. No pre- or postcolumn derivatization is necessary for detection. In addition, the technique provides the ability to analyze the desired components selectively in the presence of others, e.g., amino acids in the presence of large excess of carbohydrates or inorganic cations and anions in the presence of amino acids. These features make IC with ED an effective technique in analysis of fermentation broths in a simple and cost effective manner.

References

1. 21 CFR Parts 210 and 211, Parts 600–610 (Government Printing Office, Washington, DC).

2. Application Note 157, "Comparison of Suppressed to Nonsuppressed Conductivity Detection for the Determination of Common Inorganic Cations" (Dionex Corp., Sunnyvale, CA).

3. W.R. LaCourse, Pulsed Electrochemical Detection in High Performance Liquid Chromatography (Wiley, New York, NY, 1997).

4. J. Weiss, Ed., Handbook of Ion Chromatography, 3rd ed. (VCH Verlag, Weinheim, Germany, 2004).

5. S.J. Fritz and D.T. Gjerde, Ion Chromatography, 4th ed., (Wiley-VCH, Weinheim, Germany, 2009).

6. L. Bhattacharyya and J.S. Rohrer, Eds., Applications of Ion Chromatography for Pharmaceutical and Biological Products (Wiley, New York, NY), in press.

7. L. Joergensen, A. Weimann, and H.F. Botte, J. Chromatogr.602, 179–188 (1992).).

8. R.S.R. Robinett and W.K. Herber, J. Chromatogr. A 671, 315–322 (1994).

9. R.S.R. Robinett, H.A. George, and W.K. Herber, J. Chromatogr. A 718, 319–327 (1995).

10. Application Note 123, "The Determination of Inorganic Anions and Organic Acids in Fermentation Broths" (Dionex Corp., Sunnyvale, CA 2006).

11. W.K. Herber and R.S.R. Robinett, J. Chromatogr. A 676, 287–295 (1994).

12. Application Note 122, "The Determination of Carbohydrates, Alcohols, and Glycols in Fermentation Broths" (Dionex Corp., Sunnyvale, CA).

13. Y. Genzel, S. Königa, and U. Reichl, Anal Biochem. 335, 119–125 (2004).

14. V.P. Hanko and J.S. Rohrer, Anal. Biochem. 324, 29–38 (2004).

15. V.P. Hanko, A. Heckenberg, and J.S. Rohrer, J. Biomol. Tech. 15, 317–324 (2004).

16. P. Jandik et al., J. Chromatogr. B 732, 193–201 (1999).

17. USP–NF General Chapter <81>, "Antibiotics—Microbial Assay (USP, Rockville, MD).

*The findings and conclusions in this article have not been formally disseminated by the Food and Drug Administration and should not be construed to represent any Agency determination or policy.