Ad Capture By Anion Exchange Membrane Chromatography

A rapid, simple, and scalable process was developed in our laboratory with a minimum number of sample handling steps for chromatographic capture

FIGURE 20.3 Flow diagram for capture of Ad from lysate.

of intact, infectious Ad viral particles using Mustang®* Q membranes (Figure 20.3). To measure the dynamic binding capacity of anion exchange membranes for Ad, CsCl gradient-purified Ad (1.24 x 1012 virus particles total) was loaded at various flow rates onto a Mustang Q Acrodisc+ with a membrane volume (MV) of 0.03 ml. Fractions were analyzed at 280 nm. Figure 20.4 shows a breakthrough curve by plotting the absorbance at 280 nm vs. time. Since the sample used was purified Ad with contaminating host cell proteins or nucleic acids below detection levels at this wavelength, an anion exchange HPLC assay provided a good correlation between the absorbance at 280 nm and the number of virus particles. Recently, Sweeney and Hennessey [65] reported a more accurate and robust spectrophotometric method for Ad particle quant-itation. The breakthrough curve suggests that the dynamic binding capacity for Ad at 10% breakthrough at a flow rate of 3.0 ml/min or 100 MV/min was 1.9 x 1014 virus particles per ml of membrane. Up to 70% of the bound virus could be eluted following addition of 25 mM HEPES pH 7.4 buffer containing 1.0 M NaCl.

To measure the capacity of Mustang Q membranes for crude virus, DNase/RNase treated supernatant from a freeze-thaw lysate of Ad-infected 911 cells equivalent to ten 150 cm dishes were loaded onto a Mustang Q unit

* Mustang and Acrodisc are registered trademarks of Pall Corporation.

Time (min)

FIGURE 20.4 Dynamic binding capacity from breakthrough curve for the capture of CsCl-purified Ad on Mustang Q membranes. Purified Ad (1.24 x 1012 VP purified by CsCl centrifugation) were loaded onto a Mustang Q unit with a membrane volume of 0.03 ml using a buffer containing 25 mM HEPES, pH 7.4. The sample was loaded on the membranes using AKTA Purifier 100 system with Unicorn 3.2.1 software (Amersham Biosciences, Piscataway, NJ, USA). A flow rate of 3 ml per min was used. Virus breakthrough was monitored at 280 nm.

Time (min)

FIGURE 20.4 Dynamic binding capacity from breakthrough curve for the capture of CsCl-purified Ad on Mustang Q membranes. Purified Ad (1.24 x 1012 VP purified by CsCl centrifugation) were loaded onto a Mustang Q unit with a membrane volume of 0.03 ml using a buffer containing 25 mM HEPES, pH 7.4. The sample was loaded on the membranes using AKTA Purifier 100 system with Unicorn 3.2.1 software (Amersham Biosciences, Piscataway, NJ, USA). A flow rate of 3 ml per min was used. Virus breakthrough was monitored at 280 nm.

with a 0.03 ml MV in 25 mM HEPES pH 7.4 buffer containing of 0.2 M NaCl. Breakthrough was determined in the flow through fractions by real time PCR analysis. The results presented in Figure 20.5 show that the dynamic capacity for capture of Ad from crude cell lysates at 10% breakthrough was 4.9 x 1013 virus particles (VP) per ml of membrane at a flow rate of 100 MV/min. This has a profound influence on the process economics during scale-up. Evidently, one would require a much smaller anion exchange membrane chromatography device in order to capture Ad, compared to a conventional beaded chroma-tography media column at the manufacturing scale. For example, as titers of 1011 Ad particles per ml of cell culture media can now be routinely produced, a 1000 l batch that produces 1017viral particles could be captured by a 2 l membrane chromatography unit in under an hour at a flow rate of 20 l/min. In contrast, based on the Ad capacities for anion exchange chromatography columns determined by Huyghe and coworkers [17], it would require a 100 l column to process that amount of Ad or one would have to perform 10 cycles on a 10 l column.

The results presented in Figure 20.6 show analytical size exclusion chro-matograms of Ad purified from crude cell lysates by Mustang Q membrane chromatography or CsCl gradient centrifugation and following buffer exchange. The elution profiles were similar for both kinds of Ad preparations indicating

A Membrane Chromatography Application 1.00E + 11 1

FIGURE 20.5 Dynamic binding capacity from breakthrough curve for the capture of Ad from lysate using real time PCR. Infected 911 cells equivalent to ten 15 cm dishes were subjected to 3 freeze-thaw cycles and the lysate centrifuged for 15 min at 5000g (4°C). The supernatant was incubated at room temperature for 30 min with 100 units DNase and 50 units RNase per ml of cell lysate followed by a filtration step using a PALL SuporCap™-50 Capsule (0.2 /m) to ensure that the lysate was free of particulate matter. The suspension was adjusted to a final concentration of 0.3 M NaCl. This crude lysate sample was applied directly onto a Mustang Q membrane equilibrated with 0.3 M NaCl in 25 mM HEPES, pH 7.4 to determine Ad breakthrough. Virus particles were detected by real-time PCR.

that Ad purified by strong anion exchange membranes was as clean as virus purified by standard CsCl centrifugation.

The attractive feature of the strong anion exchange membrane units is portrayed by the small amount of membrane volume needed for reliable capture of sizeable amounts of Ad from cell lysates. Also, purification of Ad vectors using membrane-based anion exchange chromatography is significantly faster and more cost-effective than the traditional CsCl protocol where an ultracentrifuge is needed as opposed to a syringe adaptable membrane chro-matography unit. Also, protocols involving membrane-based anion exchange chromatography can easily be scaled up as in the plasmid DNA primary capture step [43].

One of the advantages of purification methods based on ion exchange chro-matography compared to methods based on CsCl gradient centrifugation is the high ratio of the number of infectious viral particles compared to the total number of virus particles. Huyghe and coworkers [17] have reported ratios of 1:80 using DEAE anion exchange columns. Membrane anion exchange chromato-graphy involving Mustang Q Acrodiscs on the other hand provided a ratio of

1.00E+09

Volume (ml)

FIGURE 20.6 Analytical size exclusion chromatography of Ad purified by (A) CsCl gradient centrifugation and (B) Mustang Q anion exchange membrane chromatography. Ad particles purified by CsCl centrifugation (2.3 x 1012 VP total) or Mustang Q anion exchange chromatography (1.8 x 1011 VP total) were loaded onto a Amersham XK-16 column packed with Sepharose CL-4B (bed volume 2.0 ml) using a buffer containing 25 mM HEPES, pH 7.4. Virus elution was monitored at 280 nm.

FIGURE 20.6 Analytical size exclusion chromatography of Ad purified by (A) CsCl gradient centrifugation and (B) Mustang Q anion exchange membrane chromatography. Ad particles purified by CsCl centrifugation (2.3 x 1012 VP total) or Mustang Q anion exchange chromatography (1.8 x 1011 VP total) were loaded onto a Amersham XK-16 column packed with Sepharose CL-4B (bed volume 2.0 ml) using a buffer containing 25 mM HEPES, pH 7.4. Virus elution was monitored at 280 nm.

1:9, (Table 20.1) indicating that this procedure was gentler on the virus than the procedures based on DEAE anion exchange columns.

A recent example showed rapid and efficient capture of Ad35 vector from a Benzonase and Triton X-100 treated 20 l cell culture supernatant containing 4 x 1015 VP on a 260 ml membrane volume Mustang Q capsule with 10fold reduction in host cell proteins and 60 to 70% Ad35 recovery in one hour processing time [66]. Aggregation of Ad through association with host cell DNA during the purification process is of major concern as it impacts meeting regulatory guidelines for DNA levels in Ad dosage form. Konz et al. [67] have developed an Ad purification process that involves addition of polysorbate-80 throughout the process as well as spiking with 1 M sodium chloride at two intermediate steps in order to dissociate the DNA/Ad complex.

0 0

Post a comment