Defining Hydrolysates: Generation of a Chemically Defined Alternative

Zachary W. Deeds, Ashley D. Smith, Benjamin J. Cutak, C. Steven Updike, Mindy S. Wilson, Daiva Dailide, Dennis W. Cooley, Barry R. Drew, and Matthew V. Caple
SAFC Biosciences, 2909 Laclede Avenue, Saint Louis, MO 63103 USA Correspondence to: Zachary.Deeds@sial.com

Abstract Results
Protein hydrolysates are commonly utilized in cell culture processes either as a component of a complete medium formulation Overlaid Traces

062804.001, Chan A

or as part of a fed-batch bioreactor process. It is well documented that hydrolysates can have a substantial positive impact on Screen Hydrolysates 062904.002, Chan A

cell growth and/or protein production. Given the undefined nature and the lot-to-lot variability associated with hydrolysates,
there exists a need to mitigate these risks with a chemically defined (CD) alternative that can maintain the desired performance. 750 750

RP-HPLC Fractionate
In order to create a CD alternative to protein hydrolysates, a two-pronged approach was taken to capture the associated & Lyophilize
effects. The first experimental path utilized standard analytical techniques to identify the nutritive components supplied by
hydrolysates (i.e. amino acids, vitamins). A “basal” supplement was designed to supply these nutrients, as this is a significant
portion of the functionality of hydrolysates. The second path was based upon Reverse Phase HPLC fractionation of multiple Screen Fractions (+ Base) 500 500

with CHO-IgG1
different hydrolysate types. Cell culture screening with Chinese Hamster Ovary (CHO) cells led to the identification of m
V
o
m
V
o
l

“bioactive” fractions. Subsequent identification of the components contained within the fractions led a greater understanding
l
t t
s s

of the effects of hydrolysates. The learnings from both approaches were utilized to generate a chemically defined supplement, ID Fraction Components
& Source CD Versions
EX-CELLTM CD Hydrolysate Fusion, which is capable of replacing the functions of hydrolysates in many CHO cell systems. 250 250

Screen Components with

Multiple CHO Lines
Materials and Methods
0 0

Sigma-Aldrich Corporation (St. Louis, MO, USA) supplied all chemicals, media, and solutions unless otherwise stated. Combine and Optimize
with Base Supplement for 0 50 100 150 200 250 300 350

Cell Lines Final Product

Minutes

Cell lines CHO-IgG1 and CHO-IgG2 are proprietary Chinese Hamster Ovary (CHO) clones expressing recombinant antibodies.
Figure 1. Process Flow Chart Figure 2. RP-HPLC Chromatogram.
Culture media Two 280nm traces (red & green) for a wheat
gluten hydrolysate fractionation are shown.
The media used in this study are all proprietary SAFC Biosciences formulations that are chemically defined or prepared without
hydrolysates. SAFC Biosciences’ imMEDIAte ADVANTAGE™ Small Volume Media Program prepared all media.
Cell Growth and Recombinant Protein Production Assays 9.00E+06 100.0%
The cells were routinely cultured in suspension in shaker flasks and were used to seed experiments conducted in duplicate 8.00E+06
50mL (30mL working volume) disposable TPP Bioreactor tubes (Techno Plastic Products AG, Switzerland). Initial cell density 80.0%
7.00E+06
was 200,000 viable cells/mL. The cells were cultured in a Multitron incubator (Infors HT, Switzerland) at 37 °C, 5% CO2 and

% Change from Base Control

No Feed
6.00E+06 60.0%

Viable Cells/mL
200rpm shaker speed. Base Supplement
5.00E+06 Yeast Extract 2g/L
Assays were counted using a Vi-CELLTM XR (Beckman Coulter, CA, USA). Spent medium samples were collected for the Base + Fraction A 40.0%
4.00E+06 4
analysis of nutrients/metabolites and IgG concentration. The secreted IgG was measured by an Octet QK (ForteBio, Inc., Menlo Base + Fraction B
Base + Fraction C 7
Park, CA, USA) using Protein A biosensors. 3.00E+06
Base + Fraction D
20.0%

Product quality was measured by Cation Exchange HPLC. MAb charge variants were separated by HPLC by injecting 50ug of 2.00E+06
0.0%
purified MAb (1mg/mL) into a 4.6mm × 250mm ProPac SCX-10 strong cation-exchange column (Dionex, CA, USA) connected 1.00E+06

Fraction A

Fraction C

Fraction D
Supplement

Fraction B
Extract 2g/L
to an Alliance HPLC with PDA detection (Waters, MA, USA). Mobile phase A consisted of 10mM sodium phosphate (pH

Yeast
Feed

Base
No Feed
0.00E+00 -20.0%
6.0±0.2). Mobile phase B consisted of 10mM sodium phosphate and 500mM sodium chloride (pH 7.5±0.2). Separation and 0 1 2 3 4 5 6 7 8

detection were performed at 30 °C and 214nm using the following operating parameters: Day -40.0%

Time (min) %A %B Flow Rate (mL/min) Figure 3. Growth Curve for Fraction Screening. Two fractions (of 66) Figure 4. Normalized IgG productivity from Fraction Screening. Four
showed a positive response for growth (Fractions A & D) when fed in fractions (of 66) showed a positive response for IgG production when fed
0 95 5 1.2 addition to the nutritive “Base” supplement with the CHO-IgG1 cell line. in addition to the nutritive “Base” supplement with the CHO-IgG1 cell line.
2 95 5 1.2 Data is normalized to the “Base” supplement.
10 60 40 1.2
10.5 0 100 1.2
1.00E+07
15.1 0 100 1.2 70.0%
9.00E+06
15.3 95 5 1.2

% Change from Base Control

60.0%
8.00E+06
7.00E+06 50.0%
Viable Cells/mL

For fed-batch assays the TPP tubes were fed with a one-time bolus on/around Day 2, depending on the cell density (0.8 to 6.00E+06
Hydrolysate 2g/L 40.0% 5
1.2e6 viable cells/mL). The feed included Glucose (6g/L) to prevent carbon source limitation and a “basal” supplement to 5.00E+06 Base Supplement
30.0% 7
provide much of the nutritive function of hydrolysates. It was designed to contain only compounds seen in the hydrolysates 4.00E+06 Base + Compound

or in a standard medium formulation (amino acids, trace elements, and vitamins). The basal supplement is added to help 3.00E+06 20.0%

elucidate the unknown positive effectors by supplying known nutritive compounds. This supplement continually evolved over 2.00E+06
10.0%
the course of the project as more was learned about the action of hydrolysates. As new compounds were identified they were 1.00E+06
Feed
0.00E+00 0.0%
titrated into the “Base” supplement. This process continued until the final product goals were achieved. Hydrolysate Base Base +
0 1 2 3 4 5 6 7 8
2g/L Supplement Compound
Day

Hydrolysate Fractionation
The Reverse Phase HPLC system consisted of a LC-6A separation module (Shimadzu, Japan) equipped with a binary gradient Figure 5. Growth Curve for Compound Screening. In this example, a Figure 6. Normalized IgG productivity from Compound Screening.
with high pressure mixing and a Shimadzu SPD-6AV UV detector. A preparative C18 column (2.5 cm x 22.5 cm) was used for single compound identified from a hydrolysate fraction showed a positive In this example, a single compound identified from a hydrolysate
the separation of the hydrolysates. Data collection and processing were performed using the Class-VP Chromatography Data response for cell growth when fed with the nutritive “Base” supplement fraction showed a positive response for IgG production when fed with
with the CHO-IgG1 cell line. the nutritive “Base” supplement with the CHO-IgG1 cell line. Data is
System. normalized to the “Base” supplement.
The primary chromatography method (TFA Separation) consisted of mobile phase A (0.1% trifluoroacetic acid in HPLC grade
water) and mobile phase B (0.1% TFA in acetonitrile). Separation of the hydrolysate components was performed using a 300
minute linear gradient from 0 to 30% B with a 30 minute hold at 30% B. The flow rate was 5mL/min and the column temperature 8.0E+06 500
was maintained at 25 °C. The injection volume was 5ml of a 200g/L solution of the representative hydrolysate (soy, wheat No Hydrolysate
7.0E+06
gluten, yeast extract, and animal tissue). Hydrolysate Control
400 CD Hydrol. Fusion 1x
6.0E+06
The hydrolysate fractions were collected at one-minute intervals with a fraction collector and then pooled in five-minute intervals No Hydrolysate
Viable Cells/mL

to simplify testing (66 fractions compared to 330). The samples were frozen in a -70 °C freezer and then lyophilized to remove 5.0E+06 Hydrolysate
300
mg/L IgG

Control
the solvents. After lyophilization the fractions were resuspended in 5mL water so they would be compatible with cell culture 4.0E+06
testing and mass spectrometry. CD Hydrol. Fusion 1x
3.0E+06
200
The fractions were cell culture tested as outlined above (feeding on Day 2). They were fed at an amount that would be
2.0E+06
equivalent to the representative quantity in 2g/L of the standard hydrolysate.
1.0E+06 100

0.0E+00
0 1 2 3 4 5 6 7 8 9 10 11
Discussion Day
0
Max IgG

The process developed for this project is outlined in Figure 1. Four different hydrolysate types were selected for analysis
(soy, wheat gluten, yeast extract, and meat). The goal was to identify the unknown active components contained within these Figure 7. Growth Curve for Batch Medium Use. In this example with Figure 8. Maximum IgG Production for Batch Medium Use. In this
the CHO-IgG1 cell line, an medium that was dependent on hydrolysates example with the CHO-IgG1 cell line, an medium that was dependent
hydrolysates, and Reverse Phase HPLC fractionation was used to aid with the separation of these compounds. In Figure 2,
was made into a CD formulation with the addition of the EX-CELLTM CD on hydrolysates was made into a CD formulation with the addition of the
two overlaid chromatograms (at 280nm) demonstrate the reproducibility of the separation. Hydrolysate Fusion. Growth performance was slightly improved with the EX-CELLTM CD Hydrolysate Fusion. IgG productivity performance was
Fractions with improved growth or productivity responses were identified from each hydrolysate. Examples of four fractions use of this product. virtually identical to the undefined control.
from a Yeast Extract are shown in Figures 3 & 4. Significant stimulatory effects are seen with addition on these fractions.
Fractions that gave a positive growth and/or productivity response were subjected to further analytical analysis to determine
the components contained within. Orthogonal separation techniques and mass spectrometry methods were employed for this 5.0E+06 1400
1300
analysis (details and data not shown). Chemically defined versions of the identified compounds were subsequently screened 4.5E+06 1200
for activity using the same CHO cell culture assay outlined above. An example of a compound that was identified to have a 4.0E+06 1100
Average Viable Cells/mL

1000
positive effect is shown in Figures 5 & 6. 3.5E+06 Glucose Only 900
mg/L IgG

Soy UF 2g/L 800

All identified active compounds and those contained in the basal supplement were further optimized with multiple CHO lines to 3.0E+06
Soy UF 4g/L 700
develop the EX-CELLTM CD Hydrolysate Fusion product. The design of this product allows it to be used in all CHO applications 2.5E+06 Yeast Extract 2g/L 600
Yeast Extract 4g/L 500
that currently utilize undefined hydrolysates. Example results for batch medium use are shown in Figures 7 & 8. Figures 9 & 2.0E+06
CD Hydrol. Fusion 1x 400
10 demonstrate how the EX-CELLTM CD Hydrolysate Fusion can be used in a fed-batch process. Lastly, in Figure 11 a CEX 1.5E+06 300
200
profile shows that only very minor product quality differences are seen when switching from hydrolysates 1.0E+06 100
to a chemically defined process with EX-CELLTM CD Hydrolysate Fusion. 5.0E+05 0
Glucose

Extract 2g/L

Extract 4g/L

Hydrolysate
Soy UF 2g/L

Soy UF 4g/L

Fusion 1x

Feed
Only

0.0E+00
Yeast

Yeast

CD

0 1 2 3 4 5 6 7 8 9 10 11
Day
Conclusions
• By analyzing the amino acid, vitamin, and metal composition of several protein hydrolysates and with subsequent Figure 9. Growth Curve for Fed-Batch Use. A fed-batch study was Figure 10. Maximum IgG Production for Fed-Batch Use. A fed-batch
performed with CHO-IgG2 in EX-CELLTM CD CHO Fusion (catalog study was performed with CHO-IgG2 in EX-CELLTM CD CHO Fusion
optimization, a hydrolysate “Base” supplement was designed to encompass much of the nutritive functions of hydrolysates. #14365C). The maximum cell density for the CD Hydrolysate Fusion is (catalog #14365C). The maximum IgG productivity levels for the CD
The use of this supplement helped to elucidate the unknown positive effectors contained within hydrolysates. very similar to the hydrolysate controls. Hydrolysate Fusion conditions are higher than the 2g/L hydrolysate
• Reverse Phase HPLC fractionation of four hydrolysate types and subsequent screening with CHO cells led to the controls and within 10% of the 4g/L hydrolysate controls.
identification of active fractions. Further analytical analysis of these fractions led to the discovery of important factors that
contribute to the effect seen with hydrolysates. CEX Product Quality Analysis
• A new supplement was created based on the work above, EX-CELLTM CD Hydrolysate Fusion. It is supplied as a 20x liquid 50
Hydrolysate Process
(product # 14700C) or powder (product # 24700C). It can be used as an alternative to hydrolysates in any part of a CHO cell 45
culture process. It also has application with other cell lines, such as NS0 and SP2/0 (data not shown). 40 CD Hydrolysate Fusion
Process
• EX-CELLTM CD Hydrolysate Fusion is designed to perform equivalently as compared to undefined hydrolysates. In most 35
% of Peak Area

instances this goal is met or exceeded. 30

25

20

15

10

0
Acidic Peak 1 Peak 2 Peak 3 Basic

Figure 11. Product Quality Analysis of IgG produced by CHO-IgG1.

A CEX-HPLC method was used to look at product quality difference
between IgG produced by a process using hydrolysates and the CD
Hydrolysate Fusion. Overall, the profiles are very similar.

71818-25465
0059
www.safcbiosciences.com