ProteoCool Pills#20: Micro PEG solubility screening, a simple Tool for Biologics Design and Formulation Development.

Adequate protein solubility is an important prerequisite for development, manufacture, and administration of biotherapeutic drug candidates, especially for high-concentration protein formulations. 

For example, in monoclonal antibody discovery, early identification of monoclonal antibody candidates whose development, as high concentration (≥100 mg/mL) drug products, could prove challenging, due to self-interaction that may induce high viscosity, can help define strategies for candidate engineering and selection.

If in theory, Rheology measurements are an effective means for characterizing therapeutic protein/antibody solutions, practically, the conventional measurements are hindered by the limited amount of material, especially during early development, when it is necessary to screen and compare several different molecules over a wide range of conditions (e.g different pH, additives, concentration)

Therefore alternative techniques able to provide hints about aggregation propensity and solubility using a smaller sample volume are essential to compare and select the best candidates in the early development and reduce the risk of move forward an candidate with high developability risk. 

Dynamic light Scattering (DLS) is probably the most used technique for this purpose since it allows to:

1) Characterize the sample intermolecular interactions (attractive or repulsive?) comparing how the diffusion coefficient (Dt) is affected by concentration since: 

In an ideal dilute solution, the diffusion coefficient (Dt) measured by DLS is not dependent on solute concentration. As concentration increases, the solution becomes less ideal. 

           Dt=D0(1+kD*C)

Attractive interactions (kD < 0) cause an apparent decrease in Dt and an apparent increase in Rh, while repulsive interactions (kD > 0) cause an apparent increase in Dt and an apparent decrease in Rh 

Therefore, decrease of Dt in function of the concentration, indicating the presence of repulsive intermolecular interactions while increase of Dt in function of the concentration, indicate presence of attractive interactions (sample more prone to aggregation)

2) Perform viscosity assessment by Microrheology: Using polystyrene beads with known values of R allows for the determination of the viscosity of the protein solution that the beads are suspended in. The size of the beads is larger than that of the protein molecules, and thus the DLS signals can easily be separated.

Bilayer interferometry (BLI) was also recently proposed as an alternative to DLS to assess protein self-interaction. (Sun et.al mabs 2013;  Domnowsky et. al International Journal of Pharmaceutics 2020) 

All those methods are fast and require a small amount of material but they require specific and expensive instrumentations those are not present in all the laboratories.

In this post, I would like to introduce you a simple method that could be done in every laboratory (since it requires the presence of a centrifuge for plates and nanodrop UV spectrophotometer or similar) based on PEG precipitation, a previously established method for determining the relative apparent solubility of adapted for screening in small scale and which is reported to correlate with the Kd values obtained by DLS (Scannell et.al, Pharm Res 2021) 

This method, adapted in 96 well plate allow to compare monoclonal antibody (mAb) candidates also if only limited quantities (eg. 1 mg) are available. 

Protocol 

(for A280nm reading with Nanodrop or similar)

(adapted from Toprani et.al J Pharm Sci. 2016) 

Day1 (afternoon)

1) 25ul of monoclonal antibody at 1mg/ml mixed with 25ul of PEG10K solutions at different concentrations (from 32% to 8%) in a 96well V-bottom plate

                           Example of a plate assembled to test 6 different mabs in duplicate: 

2) The plate was covered by aluminum foil and incubated O/N at 4°C

Day2

3) The plate was centrifuged 1h at 3200g at 4°C; 

4)10ul of surnatant were carefully transferred in a 96 well PCR plate (using a multichannel pipette). V bottom plates are preferable since the form of the well reduce the probability to resuspend the precipitate during the surnatant pick-up;

5) Amount of mab present in the surnatant was quantified by measuring the A(280nm) by NanoDrop Spectrophotometer;

6) Relative soluble fraction is calculated and plotted as function of PEG concentration;

Example of results

Example 1

 Comparision of  PEG solubility for 3 different mabs  (human igG1-CLk) in PBS buffer

Example 2

Comparision of PEG solubility for 2 different mabs in 2 different buffers (different pH)

The main limit of this protocoll is represented by the throughput, since sample reading by nanodrop allow to scale down the protocol and reduce a lot the protein amount but it is not very fast. 

6 mabs in duplicate --> 96 well --> more than 1h at nanodrop

For high number of samples, you can run a modified verision of the protocol, based on A280 deterination using 96well half volume UV clear plates (Greiner cod. 675801) in a multiplate reader. 

Since, in multiwell reader, the A280nm value change in fuction of both, concentration and optical path (that is function of the sample volume)  in this case to obtain a good sensitivitty the reaction volume (step1) were doubled  (50ul of mab 1mg/l + 50ul of PEG solution) and after centrifugation 70ul of the surnatant were tranfered to the  UV-clear 96 plate for the A280 determination with the multiplate reader. 

6 mabs in duplicate --> 96 well --> 1-2 minutes

Example 3

Comparision of PEG solubility for a wild tipe mab vs some mutant in PBS 

A(280nm) measure with Biotek- cytation5 multiplate reader

Microplate reader vs nanodrop:

Pros:

 Throughput (A plate could be acquired in few minutes)

Cons: 

Double amout of material required: (0,4 mg vs 0,2mg of mab each coloum)

Cost of the plates (about 8 euro/plates

Materials: 

- PEG 10K (Alfa Aesar cod. B21955)

- 96 well V-bottom plates (Costar cod. 3897)

- 96 UV-clear half volume (Greiner cod. 675801 cost ~ 8 euro/plate)

Other references: 

https://www.americanlaboratory.com/media/20/Document/DLS-Microrheology.pdf

ProteoCool Pills#6: Rapid protein Expression and solubility check using a 24 tip-horn sonicator

 

The use of high cell density media (e.g Enpresso reported in the ProteoCool n°7 or YGR reported in the ProteoCool n°32 ) for protein expression using E.coli allows to scale down culture volumes and easily express many clones in parallel.

Protein expression and solubility check can become a limiting step and the set-up of a simple and reliable protocol is essential.

Automation by using robotic platform is a possible solution for a real high-throughput approach, (e.g a screening projects that require to process  thousands samples for week) but in the most cases just the implementation of a optimized and rational protocol based on the use of multi-channel pipettes and 96-deep well plates can allow to manipulate in parallel up to hundreds samples/week with low cost and high protocol flexibility.

The lysis method used in the small scale test need to be similar to the ones that will be applied for the purification of the protein from rest of biomass in case of positive results, and therefore Sonication or is preferable respect than freeze-thaw cycles (work well in small scale but is not scalable) and chemical/enzymatic lysis ( expensive in scale-up)

Protocool

1) About 8OD of bacterial cells (for Enpresso correspond to 800ul not induced or 500ul induced 8h or 300 ul induced 24h) were transferred from culture into the 96 Deepwell plate following the scheme reported below (to fit the 24 tip sonicator horm)

2) Plate were centrifuged for 20' at 3700 rpm and 4°C

3) After centrifugation the surnatants were discharged 

4)    The pellets were resuspended adding 960ul of cell Lysis/binding buffer ( eg: Tris 20mM, NaCl 300mM, imidazole 10mM pH=8 for His-tagged protein that will be purified by IMAC) and mixed 3-5 times with the multichannel to allow complete pellet re-suspension. 

2)    Cell Lysis by sonication using the 24-tip horn:

Sonication cicle: 30’’ ON/45’’ OFF , 20 cycles  50% power – Qsonica Q700 sonicator.

The plate need be placed in ICE during the sonication to avoid sample heating.

3)    From each well 60ul of sonicated solution (Total fraction) e transferred in PCR plate no skirted containing 20ul of Loading sample buffer 4X ( eg. Thermo scientific cod. NP0008 ) with reducing agent.

4)    The 96-deepwell plate were centrifuged at 20’ a 4000rpm 4°C;

5)    From each well 60ul of surnatant (Souble fraction) e transferred in PCR plate no skirted containing 20ul of Loading sample buffer 4X ( eg. Thermo scientific cod. NP0008 ) with reducing agent.

6)    The no skirted PCR plate is closed by aluminum foil and incubated 5’ a 95°C in a PCR machine.

7)   12ul of each samples were loaded in a 26well 4-12% SDS-page gel.

Example of results: