Sticking In or Standing Out? Dichotomy in Vaccine Purification By Chromatography

View PDF


A general vaccine purification strategy can be divided into three stages, with one or more steps for each stage. The first stage is to concentrate and isolate the target molecule quickly to remove it from conditions that could lead to its inactivation or loss. Intermediate purification seeks to remove remaining contaminants, typically using an orthogonal approach. That is followed by a polishing step in which trace impurities are removed through high-efficiency steps because those impurities usually are similar to the target molecules.

Chromatography is a strong tool and a scalable method for achieving maximum purity in vaccine downstream processing and contaminants/impurities removal (1). Other common purification methods include precipitation, ultracentrifugation, tangential-flow filtration (TFF), and enzymatic digestion (2, 3). Chromatography benefits manufacturing through step reduction (4), higher yield, and faster processing (5). Charged depth and membrane filters (6) as well as chemical methods such as salting out and PEGylation also have been used for impurities reduction, but they should be complemented with chromatography techniques.

Here we review current challenges in vaccine purification and chromatography technology trends and demonstrates best practices for vaccine purification using chromatography.

Challenges in Vaccine Purification
At the start of purification process development, purity requirements and acceptable levels of impurities should be defined. Regulators ask that an impurities profile identify and quantify impurities that might be product- and process-related (7). In one example, the limit of DNA impurity was decreased to 10 ng/dose for continuous cell-line substrates (10 µg/dose of a plasmid containing 0.1% residual gDNA) (2). Regulators recommend that “products should be purified to be free of adventitious agents
and residual cells and should have low levels of cell-substrate DNA” (8). DNA content also contributes to production challenges including increase in fluid viscosity, fouling of separation media, and coprecipitation (3).

Product-related impurities also can result from aggregation, degradation, or incorrect composition (9). Several reported challenges include limited resolution and selectivity due to physical similarities of viruses and nucleic acids; loss of infectivity and/or potency due to physical and/or chemical degradation; and reduced downstream purification efficiency due to binding between nucleic acid and product of interest (3).

Alternative Chromatography Media
The large sizes of vaccine products (30 nm to >300 nm) limit intraparticle diffusion into the pores of most commonly available adsorbent resins (10), leading to decreases in binding capacity and throughput (11).

The use of membrane- and monolith-based media has surfaced in recent years (12). High operational flow rates and low pressure drops affect the short diffusion times common in membrane chromatography steps (13). A recent review reported recoveries of 60–80% for viral vector processes using membranebased separation media in bind and elute (B–E) mode (10), possibly due to limited binding capacity compared with traditional membrane chromatography with ligands bound to the membrane surface. Membrane chromatography also allows single-use application (11). However, membranes have larger void volumes than monoliths, which by comparison reduces their resolution (12).

Channels with diameters up to 4,000 nm in monoliths enable mass transport, mainly from convection, but mechanical instability and lack of homogeneity limit their scalability (11). A 2016 review reported that clogging occurred when high levels of host-cell DNA were present. Further work is needed to enable more successful applications (10).

A new class of three-dimensional hydrogel containing immobilized functional groups within a porous polymeric-based chromatographic support also has been reported. Tightly packed functional groups improve binding capacity while large pore size allows fast mass transfer. Productivity also can be improved through process miniaturization and rapid cycling capability (14).

Case Studies on Vaccine Purification 
Viral Vaccines: Early viral vaccines used inactivated whole-cell suspensions. For vaccine development, purification steps have been introduced to reduce allergic reactions. Use of chromatography has enabled complete removal of low molecular-weight, nonstructural proteins in the manufacture of a foot-and-mouth disease (FMD) marker vaccine — thus emphasizing the importance of chromatography (1517).

The large size of virus particles again limits mass transport, so those particles tend to bind to the bead surfaces. Beads with diameters optimized to allow for high viral recovery in flow-through (FT) mode achieved 80% virus recovery with 80–85% total protein clearance (18).

An example of rabies viral vaccine purification reportedly used a single cation-exchange chromatography (CEX) step in B/E mode. CEX of clarified harvest material achieved 60% reduction in host-cell protein, 99.7% DNA removal, and a step yield of 89% when compared with use of anion-exchange chromatography (AEX). The combination of CEX with an enzymatic nucleic acid digestion step further achieved at least a four-log scale reduction of DNA. Other CEX resins were evaluated at pilot scale. The ligand structure (sulfoisobutyl) and matrix (polymethacrylate) correlated well with the purification results. The high virus yield and DNA removal achieved in a single step also could be scaled up (19).

Polysaccharide Conjugate Vaccines (PCVs): PCVs typically are formed by linking a polysaccharide immunogen chemically to a carrier protein. A PVC process typically has two purification steps. The first step generally uses organic solvent precipitation or chromatography to separate polysaccharides from nucleic acids and proteins before conjugation.

The second purification step is intended to separate conjugated polysaccharides from unconjugated polysaccharides and carrier proteins — typically by chromatography and/or TFF. TFF often results in low yield because conjugation takes place nonstoichiometrically, resulting in similar molecular sizes among conjugated and unconjugated polysaccharides.

Advancements in membrane science have allowed for customization of membrane molecular weight cut-offs (MWCOs). When sufficient data are available, a balance between yield and impurity removal can be achieved (20). Manufacturers seek to keep excess free polysaccharides low (10 to ≤25%) because such impurities reportedly inhibit humoral immunity (21, 22). Ultrafiltration, hydrophobic-interaction chromatography (HIC), and selective precipitation (22) can reduce free polysaccharide content. A meningitis virus process using HIC and diafiltration (100kD TFF) achieved <2% free polysaccharides of the C, W-135, and Y serogroups (23). Increasing ionic strength and reducing pH to lessen intramolecular electrostatic interactions and polysaccharide compaction also shortened retention time in the size-exclusion chromatography (SEC) step (24).

Using organic solvents and detergents for the first purification step requires long process hours, introduces operational challenges, and requires special waste disposal (25, 26). In a reported pneumococcal polysaccharide vaccine process, the use of active carbon filtration, mixed-mode chromatography, and AEX resulted in a 75% shorter purification time and eliminated the need for hazardous material disposal (25). Manufacture of a next-generation vaccine with reduced use of organic solvents and detergents will require more work (27).

Poor recovery also is a common challenge in typical PCV purification processes. For example, the purification step yield of a Neisseria meningitidis X capsular polysaccharide was reported to be 60–65% (28). Poor conjugation rates of activated polysaccharide of under 20% (21) and up to 50% (29) also have been observed. Purification is often a multistep process because contaminants differ in molecular size, ionic charges, and hydrophobicity (30). A one-step, strong AEX process was successfully scaled up for a pneumococcal polysaccharide purification process (26). Similar purification performances at pilot and commercial scales were demonstrated for serotype 19-F. Other effective techniques for PCV purification have been reported (31).

Plasmid  DNA (pDNA) Vaccines: Similarities in structure between pDNA and other nucleic acid impurities complicate charge-based separation. The FDA requires >80% plasmid in supercoiled conformation for bulk release (32). Similar size-based difficulties using conventional chromatography media were observed in pDNA purification, as previously mentioned. In a comparison of various classical and novel types of AEX matrices, high binding capacities (4.5 mg/ mL) were reported. The capacity of purification matrices for DNA resuspended from fractions precipitated with polyethylene glycol (PEG) is double that for clarified lysate, possibly because contaminants were removed during the precipitation process or because DNA was compacted (33). Proper buffer selection reportedly minimized RNA binding and increased capacity from 3 to 5 g/L. In particular, use of a high–ionic-strength (>0.5 M) AEX as capture achieved 90% w/w plasmid purity (32, 34).

A pDNA process used HIC-AEX-SEC and scaled up to 200-L to 4,000-L fermentors (35). The capture step achieved a yield >80% and reduced the salt needed for binding pDNA to HIC. In the AEX step, the largest commercial monolithic column size limited production to 15 g of purified pDNA, by contrast with 10–15 g using weak AEX. The step yields of AEX and SEC exceeded 90% and 95%, respectively (35).

Virus-Like Particle (VLP)–Based Vaccines: A number of expression systems have been used for virus-like particle (VLP)-based vaccine production (36).

Enveloped baculovirus coproduced in insect-cell systems typically requires adsorptive chromatography in B–E mode because of its similar size and overall charge to VLPs. Other chromatographic matrices such as membrane layers or monoliths have been popular because of the smaller diffusion coefficients of VLPs (37).

Eluted VLPs can be unstable. In a hepatitis B process, VLP-like structures with irregular sizes and shapes were reported after binding to colloidal silica. That was likely to have resulted from unfavorable buffer conditions, resulting in surface-exposed hydrophobic patches (38). Similar structural conformation issues also were demonstrated for a rabbit hemorrhagic disease virus process (39).

To separate the helper virus in a hepatitis C process, several AEX resins were screened at different sodium chloride concentrations and linear velocities (40). The chromatography steps in low-through mode achieved 60% yields with ~2 log reduction values of baculovirus. In addition, the authors reported that higher flow rates or high load conductivity affected higher VLP recovery and lower baculovirus removal (6).

Impurities trapped within the VLP could complicate purification. Enzymatic treatment of the feedstock was reported to increase purity (41). Use of an anionic detergent and/or suitable buffer was recommended for DNA trapped by virus particles during aggregation (42).

In a human papillomavirus vaccine process, a serial AEX process was used to improve vaccine clinical manufacturability and product yields (43).

Viral Vectored Vaccines: Purification strategies for viral vectors are diverse because of differences in biophysiochemical properties and characteristics of each unique type of viral vector (44). Step-by-step optimization of the purification process is recommended to maximize yield and quality (45).

With an adenoassociated virus (AAV), the main challenges are to separate empty capsids from the filled ones, as well to separate plasmid DNA from RNA. The ratio of empty to filled particles in a recombinant AAV (rAAV) process (HEK293 cells were used) can range from 10:1 to 4:1. Gel filtration was reported to reduce empty capsids in an rAAV stock to <5% (46). A screen of 18 AEX resins for the polishing step also found that four resins enabled 100% RNA removal, and ≤99.5% plasmid content was recovered with one particular AEX resin (47). Furthermore, a dynamic capacity of 2.1 to 5.3 mg/mL was observed, and AEX achieved selectivity at high capacity within a wider range of loading salt concentration (47).

In an HEK293 system for an AAV process at 20-L scale, post–nuclease-treated cell lysate was filtered and fed at a titer of 1.1 × 1010 IVP/mL into a weak AEX with 80% recovery. The eluent then was concentrated 10× before being introduced into SEC (48).

Successful scaling of AAV to 200-L cultures also was reported using immunoaffinity chromatography for AAV serotypes 1, 2, 6, and 8 in a B/E mode. The sample then was concentrated and diafiltered before a gel filtration step that achieved quantitative recovery (49).

Chromatography Trends
Economic improvements to chromatography processes are imminent, especially with the large, potential demands for gene therapy and DNA vaccines (33). Strong biochemical engineering foundations are especially critical for future bioprocessing in that field (33).

High-throughput screening (HTS) technology could speed up process development by roughly 20%. In one example, evaluation of purification conditions for two chromatography steps for a recombinant protein-based vaccine took eight days rather than the usual two months through the use of HTS (50).

Single-use technology (SUT) eliminates cleaning and validation and improves process flexibility (5, 51). For example, similar product yield and purity to that of an open sanitized system was achieved in a 200-L single-use clinical batch of live viral vaccine using membrane chromatography (50).

Process intensification approaches also are in the limelight to reduce total processing costs and increase product quality. Technologies should complement each other and be adaptable for future bioprocessing advancements (4). Luttmann et al. reported a successful five-stage quasicontinuous bioprocess for a potential malaria vaccine using the Pichia pastoris platform (52).

In an influenza virus process, use of a three-zone simulating moving bed (SMB) heightened productivity when the system was adapted to semicountercurrent (SCC) mode (53). A process yield of 86% for adenovirus purification using a twocolumn SCC, with 90% DNA clearance and 89% HCP clearance, also was reported (10). For greater adoption of continuous SCC chromatography, the authors recommended product-oriented innovation combined with complex process design and validation considerations (10).

Talk has surfaced of process analytical technologies (PAT) and quality by design (QbD), including online high-performance liquid chromatography (HLPC) systems for process monitoring and unit operations control (54). Prediction of elution profiles for rota-VLPs on a DEAE membrane absorber using hydrodynamics and a steric mass-action (SMA) model was reported. To improve chromatographic recoveries and capacities, the influences of matrices, pore sizes, ligands, ligand densities, and stabilizers must be better understood (51). Continuous processing then can incorporate advances in continuous quality verification (CQV), continuous process verification (CPV), and real-time release (RTR)-enabling developments (54).

Prospects and Perspective
The chromatographic processing of vaccines must continue to focus on reducing the number of purification steps and improving yield (52). The main drawbacks of today’s vaccine purification processes result from the multistep purification process to which a product is subjected. The DiViNe project is a European consortium seeking to develop an affinity chromatography resin for multiple vaccine targets using Nanofitin-based technology (55). Several international collaborations also are in place, including a group working on the production of a polysaccharide vaccine (56, 57).

Merck KGaA. Biopharmaceutical Application Guide;

2 Carnes AE, Williams JA. Plasmid DNA Manufacturing Technology. Recent Pat. Biotechnol. 1(2) 2007: 151–166.

3 Gousseinov E, Kools W, Pattnaik P. Nucleic Acid Impurity Reduction in Viral Vaccine Manufacturing. BioProcess Int. 12(2) 2014: 59–64.

4 D’Souza RN, et al. Emerging Technologies for the Integration and Intensification of Downstream Bioprocesses. Pharm. Bioprocessing. 1(5) 2013: 423–440.

5 Ray S. Challenges and Trends in Vaccine Manufacturing. BioPharm Intl. 7(supplement) 2011.

6 Peixoto Christina, et al. Production and Purification of Virus-Like Particle (VLP) Based Hepatitis C Vaccine Candidate (Poster). Merck KGaA, 2015;

7 CBER. Guidance for Industry: Content and Format of Chemistry, Manufacturing and Controls Information and Establishment Description Information for a Vaccine or Related Product. US Food and Drug Administration: Rockville, MD, 1999;

8 CBER. Characterization and Qualification of Cell Substrates and Other Biological Materials Used in the Production of Viral Vaccines for Infectious Disease Indications. US Food and Drug Administration: Rockville, MD, 2017;

9 Wright FJ. Product-Related Impurities in Clinical-Grade Recombinant AAV Vectors: Characterization and Risk Assessment. Biomedicines 2(1) 2014: 80–97.

10 Nestola P, et al. Improved Virus Purification Processes for Vaccines and Gene Therapy. Biotechnol. Bioeng. 112(5) 2015: 843–857.

11 Teepakorn C, Fiaty K, Charcosset C. Comparison of Membrane Chromatography and Monolith Chromatography for Lactoferrin and Bovine Serum Albumin Separation. Processes 4(3) 2016: 31;

12 Kramberger P, Urbas L, Štrancar A. Downstream Processing and Chromatography Based Analytical Methods for Production of Vaccines, Gene Therapy Vectors, and Bacteriophages. Hum. Vaccin. Immunother. 11(4) 2015: 1010–1021; doi:10.1080/21645515.2015.1009817.

13 Vogel JH, et al. A New Large-Scale Manufacturing Platform for Complex Biopharmaceuticals. Biotechnol. Bioeng. 109(12) 2012: 3049–3058; doi:10.1002/bit.24578.

14 Zhao M, et al. Affinity Chromatography for Vaccines Manufacturing: Finally Ready for Prime Time? Vaccine, 2018 (in press).

15 Merck KGaA. Generic Manufacturing Process of Foot and Mouth Disease; Vaccines. 2013.

16 Paton DJ, Sumption KJ, Charleston B. Options for Control of Foot-and-Mouth Disease: Knowledge, Capability and Policy. Philos. Trans. R. Soc. Lond. B. 364(1520) 2009: 2657–2667.

17 Foot-and-Mouth Disease: Scientific Problems and Recent Progress 1st Annual Report (2003) Prepared for DEFRA, Science Directorate. Institute for Animal Health, Pirbright Laboratory;

18 Iyer G, et al. Reduced Surface Area Chromatography for Flow-Through Purification of Viruses and Virus Like Particles. J. Chromatogr. A. 1218(26) 2011: 3973–3981.

19 Fabre V, et al. Method for Purifying the Rabies Virus. US20100260798 A1 2010 (2009).

20 Beckett P. A New Paradigm for Vaccine Process Development: Custom TFF Membranes (Webinar, 2017);

21 Kapre SV, Chhikara MK, Soni DJ. A Novel Quantification Method for Vaccines. WO2013046226A2 (2013).

22 Berti F, et al. Separation of Conjugated and Unconjugated Components. US20090176311A1 (2009).

23 D’ambra AJ, et al. Novel Meningitis Conjugate Vaccine. 20030035806A1 (2003).

24 Hadidi M, Buckley JJ, Zydney AL. Effects of Solution Conditions on Characteristics and Size Exclusion Chromatography of Pneumococcal Polysaccharides and Conjugate Vaccines. Carbohydr. Polym. 152 2016: 12–18; doi:10.1016/j.carbpol.2016.06.095.

25 Arnold FJosef, Soike M. Alcohol-Free Pneumococcal Polysaccharide Purification Process. US005714354A (1998).

26 Kapre SV, Jana SK, Tushar D. A Novel Process for Preparation of Polysaccharides. WO2012127485A1 (2012).

27 MacDonald A. Next Generation Vaccine Manufacturing. Technology Networks 22 July 2016);

28 Shankar PS, Reddy CS, Reddy PS. Production of High Yields of Bacterial Polysaccharides. WO2014080423A2 (2014).

29 Frasch CE. Preparation of Bacterial Polysaccharide–Protein Conjugates: Analytical and Manufacturing Challenges. Vaccine 27(46) 2009: 6468–6470; doi:10.1016/j.vaccine.2009.06.013.

30 Damotharan V, Nettem SK, Maila R. Purification of Polysaccharide Protein Conjugates. WO2014188313A1 (2014).

31 McMaster, RP. Purification of Polysaccharide-Protein Conjugate Vaccines By Ultrafiltration with Ammonium Sulfate Solutions. EP2292261A1 (2011).

32 Xenopoulos A, Priyabrata P. Production and Purification of Plasmid DNA Vaccines: Is There Scope for Further Innovation? Expert Rev. Vaccines. 13 (12) 2014: 1537–1551.

33 Levy SM, et al. Biochemical Engineering Approaches to the Challenges of Producing Pure Plasmid DNA. Trends Biotechnol. 18(7) 2000: 296–305.

34 Ferreira GN, et al. Downstream Processing of Plasmid DNA for Gene Therapy and DNA Vaccine Applications. Trends Biotechnol. 18(9) 2000: 380–388.

35 Urthaler J, Buchinger W, Necina R. Improved Downstream Process for the Production of Plasmid DNA for Gene Therapy. Acta Biochim. Pol. 52(3) 2005: 703–711.

36 Roldão A, et al. Virus-Like Particles in Vaccine Development. Expert Rev. Vaccines. 9(10) 2010: 1149–1176;  doi:10.1586/erv.10.115.

37 Vicente T, et al. Large-Scale Production and Purification of VLP-Based Vaccines. J. Invertebr. Pathol. 107 (supplement) 2011: S42–S48; doi:10.1016/j.jip.2011.05.004.

38 Zahid M, Lunsdorf H, Rinas U. Assessing Stability and Assembly of the Hepatitis B Surface Antigen into Virus-Like Particles During Down-Stream Processing. Vaccine. 33(31) 2015: 3739–3745; doi:10.1016/j.vaccine.2015.05.066.

39 Fernández E, et al. Conformational and Thermal Stability Improvements for the Large-Scale Production of Yeast-Derived Rabbit Hemorrhagic Disease Virus-Like Particles As Multipurpose Vaccine. PLoS ONE 8(2) 2013: e56417; doi:10.1371/journal.pone.0056417.

40 Vicente V, et al. Anion-Exchange Membrane Chromatography for Purification of Rotavirus-Like Particles. J. Membr. Sci. 311(1–2) 2008: 270–283;

41 Steppert P, et al. Purification of HIV-1 Gag Virus-Like Particles and Separation of Other Extracellular Particles. J. Chromatogr. A. 1455, 2016: 93–101; doi:10.1016/j. chroma.2016.05.053.

42 Norman C, et al. Removal of Residual Cell Culture Impurities. WO2015071177A1 (2015).

43 Shelly DA, Van Cleave V. Parvovirus B19 VLP Vaccine Manufacturing. GEN. 29(16) 2009. 2?page=1.

44 Brindley DA, et al. Emerging Platform Bioprocesses for Viral Vectors and Gene Therapies. BioProcess Intl. 14(4s) 2016: 8–17;

45 Kallel H, Kamen AA. Large-Scale Adenovirus and Poxvirus-Vectored Vaccine Manufacturing to Enable Clinical Trials. Biotechnol. J. 10(5) 2015: 741–747; doi:10.1002/biot.201400390.

46 Urabe M, et al. Removal of Empty Capsids from Type 1 Adeno-Associated Virus Vector Stocks By Anion-Exchange Chromatography Potentiates Transgene Expression. Mol. Ther. 13(4) 2006: 823–828.

47 Eon-Duval Alex, Burke G. Purification of Pharmaceutical-Grade Plasmid DNA By Anion-Exchange Chromatography in an RNase-Free Process. J. Chromatogr B Analyt. Technol. Biomed. Life Sci. 804(2) 2004: 327–335.

48 Kamen A, Henry O. Development and Optimization of an Adenovirus Production Process. J. Gene Med. 6(1) 2004: s184–s192.

49 Cecchini S, Virag T, Kotin RM. Reproducible High Yields of Recombinant Adeno-Associated Virus Produced Using Invertebrate Cells in 0.02- to 200-Liter Cultures. Hum. Gene Ther. 22(8) 2011: 1021–1030; doi:10.1089/hum.2010.250.

50 Yang, Y-P. Advances in Purification Technologies Accelerate Vaccine Development. Am. Pharm. Rev. 30 July 2016;

51 Effio CL, Hubbuch J. Next Generation Vaccines and Vectors: Designing Downstream Processes for Recombinant Protein-Based Virus-Like Particles. Biotechnol. J. 10(5) 2015: 715– 727;  doi:10.1002/biot.201400392.

52 Luttmann R, et al. Sequential/Parallel Production of Potential Malaria Vaccines: A Direct Way from Single Batch to Quasi-Continuous Integrated Production. J. Biotechnol. 213, 2015: 83–96; doi:10.1016/j.jbiotec.2015.02.022.

53 Frensing T, et al. Options for Continuous Production of Cell Culture-Derived Viral Vaccines. Engineering Conferences International (ECI ), October 2013;

54 Whitford WG. Supporting Continuous Processing with Advanced Single-Use Technologies. BioProcess Intl. 11(4s) 2013: 46–52;

55 Pattnaik P. Sustainable Downstream Processing of Vaccines: The DiViNe Consortium. EPM Magazine, 2017;

56 Merck KGaA, Darmstadt, Germany. Merck to Develop Next-Generation Purification Processes with International Vaccine Institute. 2016;

57 International Vaccine Institute. International Vaccine Institute and MilliporeSigma to Develop Next-Generation Purification Processes, 2016;

Corresponding author Li-Jun Sim is technology manager – Singapore and Malaysia, at MilliporeSigma; +65-81138914; Takao Ito is associate director, manufacturing science and technology, Japan and Korea, at MilliporeSigma. Priyabrata Pattnaik is head of biologics operations – Asia Pacific, at MilliporeSigma.

Leave a Reply