Backbone Protection for Aggregation-Prone Peptide Synthesis

#1 Pseudoproline (ΨPro) Dipeptides as First-Line Tool

**No proper head-to-head comparison exists.** Nobody has compared ΨPro vs Hmb vs Dmb vs unprotected on the same sequence with quantitative HPLC purity data. Sampson 1999 compared only ΨPro vs Hmb. ΨPro and Dmb dipeptides have never been tested against each other on the same difficult sequence. They are treated as complementary (Ser/Thr vs Gly junctions), but nobody has actually proved that framing. **Dmb removal is unpredictable.** Removal rates range from 48% to 78% across sequences and there is no model to predict which end you will land on. The newer Thp group (Paravizzini et al., 2025) is much more acid-labile, but it needs validation on diverse peptides before it can replace Dmb. **Heat helps aggregation but hurts elsewhere.** Elevated temperature (50–90°C) disrupts aggregation but accelerates aspartimide formation and His/Cys racemisation. One study showed 86°C eliminated all resin-dependent differences, which if broadly true would weaken the case for expensive PEG resins. But it was tested on only 3 peptides (9, 15, 24-mer). **CSY does not work with Met.** NCS deprotection oxidises Met and can chlorinate Cys/Trp. Many cosmetic peptides contain Met, so norleucine substitution or alternative approaches are needed. **Hmsb/Hmnb are academic only.** These faster Hmb derivatives produce excellent results in published studies but have not been used in manufacturing-scale cGMP synthesis. CDMOs stick with ΨPro + Dmb.

#2 Dmb Dipeptides for Gly Junctions and Asp-Gly Aspartimide

**No proper head-to-head comparison exists.** Nobody has compared ΨPro vs Hmb vs Dmb vs unprotected on the same sequence with quantitative HPLC purity data. Sampson 1999 compared only ΨPro vs Hmb. ΨPro and Dmb dipeptides have never been tested against each other on the same difficult sequence. They are treated as complementary (Ser/Thr vs Gly junctions), but nobody has actually proved that framing. **Dmb removal is unpredictable.** Removal rates range from 48% to 78% across sequences and there is no model to predict which end you will land on. The newer Thp group (Paravizzini et al., 2025) is much more acid-labile, but it needs validation on diverse peptides before it can replace Dmb. **Heat helps aggregation but hurts elsewhere.** Elevated temperature (50–90°C) disrupts aggregation but accelerates aspartimide formation and His/Cys racemisation. One study showed 86°C eliminated all resin-dependent differences, which if broadly true would weaken the case for expensive PEG resins. But it was tested on only 3 peptides (9, 15, 24-mer). **Loading x backbone interaction is unknown.** No study has varied loading density across a proper panel (e.g., 0.05 to 0.5 mmol/g) while also testing backbone protection. Everyone assumes these interact, but nobody has measured it. **CSY does not work with Met.** NCS deprotection oxidises Met and can chlorinate Cys/Trp. Many cosmetic peptides contain Met, so norleucine substitution or alternative approaches are needed. **Hmsb/Hmnb are academic only.** These faster Hmb derivatives produce excellent results in published studies but have not been used in manufacturing-scale cGMP synthesis. CDMOs stick with ΨPro + Dmb.

#3 Avoid Monomeric Hmb at Manufacturing Scale

**No proper head-to-head comparison exists.** Nobody has compared ΨPro vs Hmb vs Dmb vs unprotected on the same sequence with quantitative HPLC purity data. Sampson 1999 compared only ΨPro vs Hmb. ΨPro and Dmb dipeptides have never been tested against each other on the same difficult sequence. They are treated as complementary (Ser/Thr vs Gly junctions), but nobody has actually proved that framing. **Dmb removal is unpredictable.** Removal rates range from 48% to 78% across sequences and there is no model to predict which end you will land on. The newer Thp group (Paravizzini et al., 2025) is much more acid-labile, but it needs validation on diverse peptides before it can replace Dmb. **Heat helps aggregation but hurts elsewhere.** Elevated temperature (50–90°C) disrupts aggregation but accelerates aspartimide formation and His/Cys racemisation. One study showed 86°C eliminated all resin-dependent differences, which if broadly true would weaken the case for expensive PEG resins. But it was tested on only 3 peptides (9, 15, 24-mer). **CSY does not work with Met.** NCS deprotection oxidises Met and can chlorinate Cys/Trp. Many cosmetic peptides contain Met, so norleucine substitution or alternative approaches are needed. **Hmsb/Hmnb are academic only.** These faster Hmb derivatives produce excellent results in published studies but have not been used in manufacturing-scale cGMP synthesis. CDMOs stick with ΨPro + Dmb.

#4 Hmsb/Hmnb for Positions Lacking Ser/Thr/Gly

**No proper head-to-head comparison exists.** Nobody has compared ΨPro vs Hmb vs Dmb vs unprotected on the same sequence with quantitative HPLC purity data. Sampson 1999 compared only ΨPro vs Hmb. ΨPro and Dmb dipeptides have never been tested against each other on the same difficult sequence. They are treated as complementary (Ser/Thr vs Gly junctions), but nobody has actually proved that framing. **Dmb removal is unpredictable.** Removal rates range from 48% to 78% across sequences and there is no model to predict which end you will land on. The newer Thp group (Paravizzini et al., 2025) is much more acid-labile, but it needs validation on diverse peptides before it can replace Dmb. **Heat helps aggregation but hurts elsewhere.** Elevated temperature (50–90°C) disrupts aggregation but accelerates aspartimide formation and His/Cys racemisation. One study showed 86°C eliminated all resin-dependent differences, which if broadly true would weaken the case for expensive PEG resins. But it was tested on only 3 peptides (9, 15, 24-mer). **CSY does not work with Met.** NCS deprotection oxidises Met and can chlorinate Cys/Trp. Many cosmetic peptides contain Met, so norleucine substitution or alternative approaches are needed. **Hmsb/Hmnb are academic only.** These faster Hmb derivatives produce excellent results in published studies but have not been used in manufacturing-scale cGMP synthesis. CDMOs stick with ΨPro + Dmb.

#5 Combine ΨPro + Resin + Heat for Peptides >30 AA

**No proper head-to-head comparison exists.** Nobody has compared ΨPro vs Hmb vs Dmb vs unprotected on the same sequence with quantitative HPLC purity data. Sampson 1999 compared only ΨPro vs Hmb. ΨPro and Dmb dipeptides have never been tested against each other on the same difficult sequence. They are treated as complementary (Ser/Thr vs Gly junctions), but nobody has actually proved that framing. **Dmb removal is unpredictable.** Removal rates range from 48% to 78% across sequences and there is no model to predict which end you will land on. The newer Thp group (Paravizzini et al., 2025) is much more acid-labile, but it needs validation on diverse peptides before it can replace Dmb. **Heat helps aggregation but hurts elsewhere.** Elevated temperature (50–90°C) disrupts aggregation but accelerates aspartimide formation and His/Cys racemisation. One study showed 86°C eliminated all resin-dependent differences, which if broadly true would weaken the case for expensive PEG resins. But it was tested on only 3 peptides (9, 15, 24-mer). **CSY does not work with Met.** NCS deprotection oxidises Met and can chlorinate Cys/Trp. Many cosmetic peptides contain Met, so norleucine substitution or alternative approaches are needed. **Hmsb/Hmnb are academic only.** These faster Hmb derivatives produce excellent results in published studies but have not been used in manufacturing-scale cGMP synthesis. CDMOs stick with ΨPro + Dmb.

#6 Never Use ΨPro Adjacent to Asp Under Flow >60°C

**No proper head-to-head comparison exists.** Nobody has compared ΨPro vs Hmb vs Dmb vs unprotected on the same sequence with quantitative HPLC purity data. Sampson 1999 compared only ΨPro vs Hmb. ΨPro and Dmb dipeptides have never been tested against each other on the same difficult sequence. They are treated as complementary (Ser/Thr vs Gly junctions), but nobody has actually proved that framing. **Dmb removal is unpredictable.** Removal rates range from 48% to 78% across sequences and there is no model to predict which end you will land on. The newer Thp group (Paravizzini et al., 2025) is much more acid-labile, but it needs validation on diverse peptides before it can replace Dmb. **Heat helps aggregation but hurts elsewhere.** Elevated temperature (50–90°C) disrupts aggregation but accelerates aspartimide formation and His/Cys racemisation. One study showed 86°C eliminated all resin-dependent differences, which if broadly true would weaken the case for expensive PEG resins. But it was tested on only 3 peptides (9, 15, 24-mer). **CSY does not work with Met.** NCS deprotection oxidises Met and can chlorinate Cys/Trp. Many cosmetic peptides contain Met, so norleucine substitution or alternative approaches are needed. **Hmsb/Hmnb are academic only.** These faster Hmb derivatives produce excellent results in published studies but have not been used in manufacturing-scale cGMP synthesis. CDMOs stick with ΨPro + Dmb.

#7 CSY Masking and Depsipeptides for Extreme Cases

**No proper head-to-head comparison exists.** Nobody has compared ΨPro vs Hmb vs Dmb vs unprotected on the same sequence with quantitative HPLC purity data. Sampson 1999 compared only ΨPro vs Hmb. ΨPro and Dmb dipeptides have never been tested against each other on the same difficult sequence. They are treated as complementary (Ser/Thr vs Gly junctions), but nobody has actually proved that framing. **Dmb removal is unpredictable.** Removal rates range from 48% to 78% across sequences and there is no model to predict which end you will land on. The newer Thp group (Paravizzini et al., 2025) is much more acid-labile, but it needs validation on diverse peptides before it can replace Dmb. **Heat helps aggregation but hurts elsewhere.** Elevated temperature (50–90°C) disrupts aggregation but accelerates aspartimide formation and His/Cys racemisation. One study showed 86°C eliminated all resin-dependent differences, which if broadly true would weaken the case for expensive PEG resins. But it was tested on only 3 peptides (9, 15, 24-mer). **CSY does not work with Met.** NCS deprotection oxidises Met and can chlorinate Cys/Trp. Many cosmetic peptides contain Met, so norleucine substitution or alternative approaches are needed. **Hmsb/Hmnb are academic only.** These faster Hmb derivatives produce excellent results in published studies but have not been used in manufacturing-scale cGMP synthesis. CDMOs stick with ΨPro + Dmb.