Does loading method influence my peptide recovery after purification?

In peptide purification, sample loading onto the column is rarely considered.  Most, if not all, HPLC instruments come equipped with a sample injection loop which demands a liquid injection of the sample for purification.  If you decide to use flash chromatography to purify your peptides though, liquid injection is no longer the exclusive method for sample introduction to the column.  Alternatively, dry loading crude material is a strategy often used in small molecule purifications, particularly when sample solubility concerns arise.

The first question I asked myself when considering a new sample loading strategy is whether or not the purification efficiency will be maintained.  A close second though is whether or not the loading method will cause significant differences in peptide recovery.

In today’s post, I’ll compare recovery efficiencies for peptides purified using reversed phase flash chromatography but loaded onto the cartridge using either direct liquid injection or dry loaded onto reversed phase material.

Prior to HPLC purification, crude peptide samples are often dissolved in a solution of aqueous acetonitrile.  Sometimes, due to solubility limitations, this volume can be very large (10s of mL’s for just a few mg of peptide).  For purification with flash chromatography though, this large volume can hamper the purification efficiency so organic solvents like DMF or DMSO are often substituted for concentrated sample injections.  I started to ask myself though, can dry loading crude peptides also be a viable method for sample loading onto the cartridge.

To evaluate this question, I synthesized the Jung-Rudman sequence, a particularly nasty 10 amino acid peptide using my Biotage® Initiator+ Alstra™ peptide synthesizer with DIC/oxyma for my coupling reagents and default methods, Figure 1.

Figure 1: Crude analytical HPLC of the JR 10-mer used for the purification study.

The crude peptide is about 33% pure, which normally is very disappointing.  Because this experiment is designed to evaluate how loading impacts purification, I’m pleased that the synthesis turned out ugly.  Who really wants to compare recovery efficiencies when the peptide is nearly pure from the start?

For my first purification I dissolved about 100 mg of crude JR 10-mer in 300 µL of DMSO and directly injected the sample onto an equilibrated 10 gram Biotage® SNAP Bio C18 cartridge for purification with my Biotage® Isolera Dalton 2000™, Figure 2.

Figure 2: Purification of JR 10-mer peptide dissolved in DMSO for loading on the SNAP Bio cartridge. Fractionation was triggered by UV, but extracted ions were monitored for the (m+2)/2 (purple line) and m+1 (green line) throughout the purification. Fraction 7 (black arrow) was concentrated for recovery calculations.  Inset provides a zoomed in view of the peptide elution region.

Although I use this volume of DMSO to load peptides frequently, for this peptide a large degree of streaking is observed, decreasing the UV detector sensitivity and flattening the overall chromatogram.  This makes identifying the desired fractions to preserve and combine extremely difficult.  I was also monitoring the purification with mass spectrometry.  Using the 1213 (m+1) and 606 ((m+2)/2) masses as guidance, I concentrated fraction 7 for analytical HPLC and recovery analysis, Figure 3.

Figure 3: Analytical HPLC for the JR 10-mer purified using a direct liquid injection. 102 mg was dissolved in 300 uL DMSO and the purified peptide is 56% pure. Ultimately, 3 mg of the desired peptide was recovered from the purification.

The rapid purification cleaned up the JR peptide from 33% crude purity to about 56% purity and I recovered about 3 mg of the expected 33 mg (9% recovery, recovery quantity corrected for post purification purity).  While I would have liked better results, these results highlight the use of flash chromatography as a clean up technique prior to HPLC polishing.  The sample is now enriched in the desired product which will reduce the effort necessary to achieve the desired purity with HPLC.

For the second purification, I performed a true dry load of the peptide.  To do this, I fully dissolved about 100 mg the peptide in a solution of water, acetonitrile and methanol (the crude peptide isn’t very soluble in much) and added the solution to about 1 gram of silica functionalized with C18 alkyl chains.  Using my Biotage® V-10 Evaporation system I was able to fully evaporate the solvent in minutes, leaving behind a free flowing powder of stationary phase with crude peptide adsorbed to the surface.  I transferred the material to an empty Samplet® cartridge, equilibrated the cartridge, and loaded the Samplet onto the cartridge for purification with my Isolera Dalton 2000™, Figure 4.

Figure 4:  Purification of JR 10-mer peptide was evaporated onto C18-functionalized silica for loading on the SNAP Bio cartridge. Fractionation was triggered by UV, but extracted ions were monitored for the (m+2)/2 (purple line) and m+1 (green line) throughout the purification. Fractions 8 and 9 (black arrow) were concentrated for recovery calculations.  Inset is a zoomed in view of the peptide elution region.

The first difference to note is the significantly reduced size of the injection peak.  An advantage of this is that the peak signals from the peptide and impurities are now much sharper and clearer for identification.  As above I used the mass chromatogram as a guide and combined and concentrated fractions 8 and 9 for analytical HPLC and recovery analysis, Figure 5.

Figure 5:  Analytical HPLC for the JR 10-mer purified using dry loading. 104.2 mg was dried onto 1 gram of C18-functionalized silica and the purified peptide is 50% pure. Ultimately, 7.5 mg of the desired peptide was recovered from the purification.

Dry loading the crude peptide for purification resulted in a purity of 50% and I recovered about 7.5 mg of the expected 34.4 mg (21% recovery, corrected for purity).  Although the crude purity is slightly lower than the liquid load purification, the recovery is much better.  This is likely due to two factors; 1) crude sample is not analytically transferred from the syringe to the cartridge during a direct liquid injection and 2) the streaking caused by the DMSO injection solvent caused a greater degree of co-elution.

It is important to note that for both of these purifications, I allowed the Isolera to collect large fractions.  Purity can certainly be increased by decreasing the fraction volume, albeit with some cost in recovery.  Based on these results, I will probably continue to use dry loading a my primary method for sample introduction moving forward.  Drying the peptide onto the stationary phase was quick and easy with the V-10 evaporation system and the purification, although slightly less pure, yielded higher recovery and simplified interpretation of the chromatogram.

Stay tuned for an investigation of alternative materials for dry loading peptides.  Have you ever tried dry loading your crude peptides?

Leave a Reply

Your email address will not be published. Required fields are marked *