Note · 10 November 2026

Reconstituting Peptides: Solvents, Mixing and What Gets Lost

Supplier handbooks pick a solvent for a freeze-dried peptide by its charge: water first, dilute acid for basic sequences, dilute base for acidic ones, organic co-solvent for hydrophobic ones. Measured studies support the charge logic and add what handbooks omit: agitated proteins were damaged at air interfaces rather than by shear, peptides stuck to walls, and powder mass was not peptide mass.

The answer: what the measured literature supports

The measured literature supports choosing a solvent by charge. Solubility was lowest near a molecule's isoelectric point, the pH at which its charges balance, and amyloid beta gave more soluble peptide when dissolved at a pH well above that point. Dimethyl sulfoxide (DMSO), the co-solvent handbooks name for hydrophobic sequences, oxidised cysteine in measured systems, and methionine and tryptophan under some strongly acidic conditions but not others. When proteins were agitated or foamed, the measured damage came from air-liquid interfaces rather than from shear, and varied by molecule. Dilute peptides adsorbed to container walls, some far more than others, and the concentration reached depended on how much of the powder was peptide.

Supplier handbooks give the charge rule without data and disagree on acid strength and on DMSO. Two claims found on reconstitution pages, that shaking or a stream of liquid shears peptides and that slow dissolution signals degradation, traced to no study and are logged as untraced.

Three handbooks, one rule, different strengths

Bachem, GenScript and LifeTein each start from the sequence. Bachem's peptide guide and GenScript's guideline both count basic residues and the N-terminus against acidic residues and the C-terminus; GenScript scores histidine as basic and holds that peptides under six residues generally dissolve in water. None of the three cites a study, and each is recorded here as guidance.

For basic peptides they name acid at different strengths. GenScript's flowchart moves from water to 10 to 30 per cent acetic acid. Bachem names a small amount of acetic or trifluoroacetic acid, then dilution, and calls phosphate-buffered saline at pH 7.0 to 7.4 the safest diluent where 1 mg/mL or less suffices. LifeTein names diluted acetic acid after water. Ready-made diluents can sit well below GenScript's range: a listing for 0.6 per cent acetic acid describes it as an acidic diluent for laboratory work. None of the documents gives data on which strength dissolved which sequence.

For acidic peptides Bachem names 0.1 per cent aqueous ammonia, GenScript ammonium hydroxide except for cysteine-containing sequences, and LifeTein phosphate-buffered saline or ammonium bicarbonate. For hydrophobic sequences all three name a small volume of organic solvent, such as DMSO, dimethylformamide, acetonitrile or methanol, then dilution. They part on DMSO: GenScript lists it as the last step for every charge class, with no caveat on it beyond storing methionine, cysteine and tryptophan peptides oxygen-free; LifeTein advises against it where cysteine or methionine stability is a concern; Bachem lists it while requiring oxygen-free solvents or reducing agents for those residues.

Charge, the isoelectric point and the liquid

The charge rule rests on a measured relationship. Shaw and colleagues reported in 2001 on ribonuclease Sa, an acidic protein with an isoelectric point of 3.5, and two lysine-substituted variants with isoelectric points of 6.4 and 10.2: across the three, the pH of minimum solubility moved with the isoelectric point. Isoelectric points estimated from model-compound pK values could be in error by more than a full point of pH; the handbook count uses whole charges and no pK values at all. This site's note on isoelectric point and peptide solubility computed values from 10.40 to 11.43 for one sequence, depending on the pK set and on whether its termini were treated as blocked or free.

Fezoui and colleagues reported in 2000 that amyloid beta, whose precipitation and aggregation peak near its isoelectric point of about 5.5, gave significantly higher yields of monomer and dimer when first dissolved in sodium hydroxide and freeze-dried than as a trifluoroacetate or hydrochloride salt. Taken up in neutral buffer, the material moved from about pH 10.5 to neutral without crossing that point.

The liquid also changed the outcome. Zhang and colleagues freeze-dried keratinocyte growth factor in 1995 under conditions chosen to make it aggregate on reconstitution with water; sulfated polysaccharides, surfactants, polyphosphates or amino acids in the reconstitution medium, or a change in its ionic strength, cut aggregation and raised recovery of soluble monomer. This site's note on how long bacteriostatic water lasts recorded the reverse, a freeze-dried protein that aggregated more in 0.9 per cent benzyl alcohol than in water, and traced the 28-day figure to a regulatory default for opened, preserved containers.

What DMSO did to methionine and cysteine

The handbook concern about DMSO traces to measured chemistry, under narrower conditions than the warnings imply. Savige and Fontana reported in 1980 that DMSO with concentrated hydrochloric acid, in acetic acid, oxidised methionine to its sulfoxide and cysteine to cystine quantitatively, and converted the tryptophan of four peptides, among them somatostatin, to oxindolylalanine derivatives isolated in over 90 per cent yield. Shechter reported in 1986 that aqueous DMSO and hydrochloric acid oxidised methionine in peptides and proteins to the sulfoxide; the reaction was strongly pH-dependent, and thiols followed after a 2-hour lag. Tamamura and colleagues, forming disulfides in peptides in 1995 with silver triflate followed by DMSO in aqueous hydrochloric acid, reported no significant side reactions at methionine, tyrosine or tryptophan under their conditions, so the measured outcome depended on conditions rather than on DMSO alone.

Cysteine also reacted in an assay solvent. Akimoto and colleagues found in 2020 that oxidation of a cysteine-derived reagent and of a cysteine peptide rose with DMSO concentration in the DMSO and acetonitrile solvent used to dissolve test chemicals, and that 5 per cent DMSO lowered the reagent's concentration by about 0.4 micromolar, a shift they judged left their assay's results unaffected. A PubMed search on 10 November 2026 pairing DMSO with methionine oxidation in peptides returned 13 records; the two measuring the reaction both used hydrochloric acid, and none tested a neutral DMSO stock.

Shaking, foam and how long a cake takes

Of four reconstitution pages examined for this note, three state that shaking, or a stream of liquid striking the powder, damages, shears or denatures peptides or proteins; none cites a study for it. None of the three handbooks mentions shaking or foam. A PubMed search on 10 November 2026 pairing shear, shaking, agitation or vortexing with peptide-bond cleavage returned 12 records, all on von Willebrand factor and the enzyme that cleaves it, or unrelated. The claim that shaking shears or breaks peptide molecules is logged as untraced.

Measured damage lay elsewhere. Maa and Hsu reported in 1997 that shear alone did not significantly aggregate recombinant human growth hormone or deoxyribonuclease. High shear with air present had no major effect on the deoxyribonuclease but aggregated the growth hormone irreversibly, more so at higher protein concentration and air-liquid interfacial area, because shear kept creating new interface. Foaming by nitrogen bubbling, at low shear, was very detrimental to the growth hormone, and surfactant delayed or prevented aggregation.

Clarkson and colleagues reported in two 1999 papers that damage in protein foams came mainly from unfolding at the gas-liquid interface: about 10 per cent of adsorbed bovine serum albumin and 75 per cent of adsorbed pepsin stayed denatured after desorbing, oxidation was minimal, and bulk shear as bubbles burst caused no denaturation. Foam changed the tertiary structure of all six proteins tested and aggregated some, with no fragmentation.

One study compared reconstitution methods. Gerring and colleagues reported in 2013 that three batches of a freeze-dried ovine antibody-fragment product took a mean of 25 minutes 51 seconds to dissolve in 10 mL of sterile water with gentle swirling, and 3 minutes 7 seconds in 18 mL of 0.9 per cent saline with manual inversion, with no change in degradation, aggregation, purity, structure or mass. Volume, liquid and motion changed together, so the time saved belongs to none of them alone. None of these studies used a short research peptide.

How long a cake takes depends on the cake. Kulkarni and colleagues reported in 2021, for high-concentration protein formulations, that partially crystalline cakes with larger pores wetted, broke into small pieces and dissolved faster, while some amorphous cakes admitted no fluid at all and floated amid foam or bubbles, dissolving by gradual surface erosion, with reconstitution time highly correlated with the viscosity near the dissolving surface. Bachem's guide states that reconstitution occasionally takes several hours and that bath sonication may help. One page examined for this note states that particles left after 15 minutes of swirling may mean the peptide had degraded, citing no source for it; PubMed searches pairing reconstitution or dissolution time with degradation, potency or purity returned five records, none testing that link, and the claim is logged as untraced.

What reaches solution: walls and net content

Not all of what dissolves stays free. Livesey and Nicholls measured in 2012 how much human serum albumin it took to halve the adsorption of iodine-labelled peptides and proteins to polystyrene or polypropylene tubes: the figure varied more than ten-thousand-fold between angiotensin II, the least sticky, and tyr0-urocortin II, the most, with no significant difference between the plastics. The authors concluded that some peptides are sticky enough to risk adsorptive loss during sampling and analysis. None of the three handbooks addresses losses to container walls.

The concentration reached also depends on the powder. This site's note on what HPLC purity measures recorded peptide fractions of 530 to 896 mg/g in the freeze-dried materials it covered. Concentration is mass over volume, the arithmetic the reconstitution calculator performs from a stated mass, so a label figure gives a peptide concentration only if it states peptide rather than powder: 10 mg of powder at 530 to 896 mg/g, in 2 mL, carries 2.65 to 4.48 mg/mL of peptide rather than 5. Where a lot certificate reports net content, that is the mass the arithmetic needs.

What traces to a source, and what does not

The charge rule traces to measured solubility minima near the isoelectric point, though the handbooks' whole-charge count is cruder than pK-based estimates that could miss by more than a full point of pH. The DMSO warning is supported for cysteine and, under some strongly acidic conditions, for methionine and tryptophan; for neutral stocks it is recorded as guidance, as are the handbooks' acid and base strengths, DMSO volumes and sonication. The claims about shaking, streams of liquid and slow dissolution are logged as untraced. In the measured record, charge set solubility, interfaces set agitation damage in the proteins tested, and the composition of the powder and the walls of the container set how much peptide ended up in solution.

Disclosure. Vialog is published by the team behind Disguised Alpha, which sells the product linked in this note. The note follows the same editorial policy as every other page on this site.

References

  1. Bachem. Handling and Storage Guidelines for Peptides. The Online Peptide Guide, chapter 08 of 10; no author shown; page metadata: published 29 January 2025, modified 15 October 2026. Section on solubilisation read: residue counting including the N-terminal amine and C-terminal carboxyl; basic peptides in a small amount of acetic or trifluoroacetic acid, then diluted, with phosphate-buffered saline at pH 7.0 to 7.4 named the safest diluent where 1 mg/mL or less suffices, and Arg- and Lys-rich trifluoroacetate salts said to tend to dissolve at neutral pH; acidic peptides in 0.1 per cent aqueous ammonia; neutral or hydrophobic peptides in a small amount of DMSO, DMF, acetic acid, acetonitrile, methanol, propanol or isopropanol, then diluted; reconstitution occasionally taking up to several hours; several minutes of water-bath sonication, avoiding excessive warming; oxygen-free solvents or reducing agents for Trp, Met and Cys peptides. States that no single procedure can be given for every peptide. No study, dataset or reference cited. Guidance, not a study. No PMID. Retrieved 10 November 2026. View on bachem.com
  2. GenScript. Peptide Solubility Guidelines and Reconstitution Instructions (PDF, version 2023-V1.0); no author shown. Peptides under six residues generally dissolve in water; charge count of +1 for Arg, Lys, His and the N-terminus and -1 for Asp, Glu and the C-terminus; net negative: water, then ammonium hydroxide under 50 microlitres (not for Cys-containing peptides), then DMSO; net positive: water, then 10 to 30 per cent acetic acid, then DMSO; net neutral: acetonitrile or methanol, then DMSO for very hydrophobic peptides; flowchart gives 50 to 100 microlitres of DMSO per mg with sonication; stock solutions of about 1 to 2 mg/mL; Met, Cys and Trp peptides stored oxygen-free. States that solubility depends on impurities and salts in the powder as well as on sequence. No study, dataset or reference cited. Guidance, not a study. No PMID. Retrieved 10 November 2026. View on genscript.com
  3. LifeTein. Peptide Not Dissolving? Troubleshooting Guide. Supplier guide; no author or date shown. Basic peptides in water, then diluted acetic acid; acidic peptides in PBS or a mild basic buffer such as ammonium bicarbonate; hydrophobic peptides first in a small amount of DMSO, DMF, acetonitrile, methanol or isopropanol, then diluted slowly; cysteine peptides in degassed acidic solvent below neutral pH; DMSO avoided where cysteine or methionine stability is a concern; gentle sonication. Attributes most failures to hydrophobicity, aggregation or a mismatch between charge and solvent. No study, dataset or reference cited. Guidance, not a study. No PMID. Retrieved 10 November 2026. View on lifetein.com
  4. Shaw KL, Grimsley GR, Yakovlev GI, Makarov AA, Pace CN. The effect of net charge on the solubility, activity, and stability of ribonuclease Sa. Protein Sci. 2001;10(6):1206-15. doi:10.1110/ps.440101. PMID 11369859 View on pubmed.ncbi.nlm.nih.gov
  5. Fezoui Y, Hartley DM, Harper JD, Khurana R, Walsh DM, Condron MM, Selkoe DJ, Lansbury PT, Fink AL, Teplow DB. An improved method of preparing the amyloid beta-protein for fibrillogenesis and neurotoxicity experiments. Amyloid. 2000;7(3):166-78. doi:10.3109/13506120009146831. PMID 11019857 View on pubmed.ncbi.nlm.nih.gov
  6. Zhang MZ, Wen J, Arakawa T, Prestrelski SJ. A new strategy for enhancing the stability of lyophilized protein: the effect of the reconstitution medium on keratinocyte growth factor. Pharm Res. 1995;12(10):1447-52. doi:10.1023/a:1016219000963. PMID 8584478 View on pubmed.ncbi.nlm.nih.gov
  7. Savige WE, Fontana A. Oxidation of tryptophan to oxindolylalanine by dimethyl sulfoxide-hydrochloric acid. Selective modification of tryptophan containing peptides. Int J Pept Protein Res. 1980;15(3):285-97. doi:10.1111/j.1399-3011.1980.tb02579.x. PMID 6155358 View on pubmed.ncbi.nlm.nih.gov
  8. Shechter Y. Selective oxidation and reduction of methionine residues in peptides and proteins by oxygen exchange between sulfoxide and sulfide. J Biol Chem. 1986;261(1):66-70. PMID 3001062 View on pubmed.ncbi.nlm.nih.gov
  9. Tamamura H, Otaka A, Nakamura J, Okubo K, Koide T, Ikeda K, Ibuka T, Fujii N. Disulfide bond-forming reaction using a dimethyl sulfoxide/aqueous HCl system and its application to regioselective two disulfide bond formation. Int J Pept Protein Res. 1995;45(4):312-9. doi:10.1111/j.1399-3011.1995.tb01043.x. PMID 7601603 View on pubmed.ncbi.nlm.nih.gov
  10. Akimoto M, Yamamoto Y, Watanabe S, Yamaga H, Yoshida K, Wakabayashi K, Tahara Y, Horie N, Fujimoto K, Kusakari K, Kamiya K, Kojima K, Kawakami T, Kojima H, Ono A, Kasahara T, Fujita M. Oxidation of a cysteine-derived nucleophilic reagent by dimethyl sulfoxide in the amino acid derivative reactivity assay. J Appl Toxicol. 2020;40(6):843-854. doi:10.1002/jat.3948. PMID 32052484 View on pubmed.ncbi.nlm.nih.gov
  11. Maa YF, Hsu CC. Protein denaturation by combined effect of shear and air-liquid interface. Biotechnol Bioeng. 1997;54(6):503-12. PMID 18636406 View on pubmed.ncbi.nlm.nih.gov
  12. Clarkson JR, Cui ZF, Darton RC. Protein denaturation in foam [the first of two consecutive papers under this title; mechanism, bovine serum albumin and pepsin]. J Colloid Interface Sci. 1999;215(2):323-332. doi:10.1006/jcis.1999.6255. PMID 10419667 View on pubmed.ncbi.nlm.nih.gov
  13. Clarkson JR, Cui ZF, Darton RC. Protein denaturation in foam [the second of two consecutive papers under this title; six proteins foamed in a bubble column]. J Colloid Interface Sci. 1999;215(2):333-338. doi:10.1006/jcis.1999.6256. PMID 10419668 View on pubmed.ncbi.nlm.nih.gov
  14. Gerring D, King TR, Branton R. Validating a faster method for reconstitution of Crotalidae Polyvalent Immune Fab (ovine). Toxicon. 2013;69:42-9. doi:10.1016/j.toxicon.2012.12.005. PMID 23305624 View on pubmed.ncbi.nlm.nih.gov
  15. Kulkarni SS, Patel SM, Suryanarayanan R, Rinella JV, Bogner RH. Key factors governing the reconstitution time of high concentration lyophilized protein formulations. Eur J Pharm Biopharm. 2021;165:361-373. doi:10.1016/j.ejpb.2021.05.005. PMID 33974974 View on pubmed.ncbi.nlm.nih.gov
  16. Livesey JH, Nicholls MG. Stickiness to hydrophobic surfaces varies widely among peptides and proteins. J Immunoassay Immunochem. 2012;33(3):302-13. doi:10.1080/15321819.2011.647188. PMID 22738653 View on pubmed.ncbi.nlm.nih.gov