Morpholine, a strong contender for Fmoc removal in solid‐phase peptide synthesis

Morpholine, which scores 7.5 in terms of greenness and is not a regulated substance, could be considered a strong contender for Fmoc removal in solid‐phase peptide synthesis (SPPS). Morpholine in dimethylformamide (DMF) (50%–60%) efficiently removes Fmoc in SPPS, minimizes the formation of diketopiperazine, and almost avoids the aspartimide formation. As a proof of concept, somatostatin has been synthesized using 50% morpholine in DMF with the same purity as when using 20% piperidine–DMF.


| INTRODUCTION
The increasing number of peptide-based therapeutics approved by the Food and Drug Administration (FDA) each year has brought about a boom in peptide pharmaceuticals and the related manufacturing sector. 1,2Like the manufacturing process of any other pharmaceutical, that of these peptidyl drugs requires optimization, good manufacturing practices, and overall sustainability to meet the requirements of regulatory authorities. 3,46][7] Generally, solution-phase synthesis, solid-phase synthesis, and/or a combination of the two have been adopted for peptide production. 8,9 this context, 9-fluorenylmethoxycarbonyl (Fmoc)-based solid-phase peptide synthesis (SPPS) is popular among chemists, mostly because of the mild conditions used. 10,11Successful peptide synthesis using Fmoc-SPPS protocols involves the efficient execution of two repetitive steps, namely, the incorporation of Fmoc-protected amino acids onto a growing peptide chain attached to a solid support and the removal of Fmoc using base after each incorporation. 12oc is a temporary protecting group for N α -amino acids, and it can be rapidly removed by treatments with a variety of primary or secondary amines. 10,13Generally, excess base in a relatively polar solvent can remove Fmoc efficiently. 10In this regard, cyclic secondary amines are the bases of choice in Fmoc-SPPS because of the advantage of the scavenging dibenzofulvene generated during Fmoc removal.[16] The use of these secondary amines renders successful Fmoc-SPPS protocols, although a plethora of side reactions occur. 17These reactions include the formation of diketopiperazine (DKP), epimerization, and formation of aspartimides and related side products, to name just a few.For piperidine (PIP), the most popular base used to remove Fmoc, the extent of these side reactions often implies a greater challenge in downstream processing, including a loss in the yield of the desired peptide.PIP is a controlled substance in terms of narcotic drug production and as such its use is restricted.Moreover, and most importantly, it is a hazardous chemical with a greenness score of 6.9 on a scale of 0-10 in GSK's reagent selection guide. 180][21] In this regard, we recently reported the use of 30% 4MP in 0.5 M OxymaPure-DMF for Fmoc removal. 22In this context, other secondary amines, such as pyrrolidine and piperazine (PPZ), have also been exploited on several occasions.For example, the use of PPZ supplemented with DBU in N-methylpyrrolidone (NMP) has been described to enhance the suppression of DKP formation in some sequences. 23However, several factors, including limited solubility in DMF, reduced base strength and thus necessitating the addition of another stronger base, and the requirement of additives like hydroxybenzotriazole (HOBt) 24 or formic acid 25 to suppress side reactions, circumscribe the general use of PPZ in Fmoc-SPPS.Likewise, the performance of pyrrolidine was not promising due to its higher base strength, similar to that of PIP, thus influencing side reactions and sometimes causing the degradation of the desired peptide. 26,27In contrast, in the case of morpholine (Morph), Fmoc removal is generally accompanied by fewer side reactions.9][30] Owing to its lower base strength, to date, the use of Morph has been limited to a couple of Fmoc removal cycles for peptide sequences prone to side reactions during SPPS. 31For instance, 20% Morph-NMP was used for Fmoc removal from the Asp-X (X = His or Ser) motif in sequences to suppress aspartimide formation. 32,33[36][37] Here we studied the performance of Morph as a general alternative to PIP for Fmoc removal in SPPS and evaluated its capacity to

| Swelling
Of AM-resin, 200 mg was placed in 5-mL syringes fitted with a filter, and a solvent was then added up to a volume of 5 mL.The mixture was allowed to stand for 30 min to swell the resin to its maximum capacity.Using the piston of a syringe, the swollen resin was compressed to totally drain the solvent.Thereafter, the piston was carefully pulled up until the resin attained its maximum volume in the syringe.The volume of the resin was read before and after swelling.

| Peptide synthesis
All peptides were prepared following the standard Fmoc/tert-butyl (tBu)-based SPPS protocol.Fmoc Rink Amide AM resin (0.64 mmol/g) F I G U R E 1 Cyclic secondary amines used for Fmoc group removal and their corresponding pK a values.was used as solid support for the peptides.Initially, the resin was washed using dichloromethane (DCM) (2 Â 1 min) and swelled in DCM (1 Â 10 min) and DMF (1 Â 10 min).Deprotection of the Fmoc group was achieved by standard treatment of the resin with 20% PIP-DMF (1 + 7 min), followed by washing with DMF.The protected Fmoc-amino acids (3 equiv.)were incorporated using DIC (3 equiv.)and OxymaPure (3 equiv.) in DMF, as coupling reagents, for 30-60 min at rt.This was repeated until the final sequence was achieved.
Fmoc from the last coupled amino acid was removed as explained above.
When using HO-Wang resin, the first amino acid was incorporated by the addition of Fmoc-AA-OH (4 equiv.),N,N-dimethylaminopyridine (DMAP) (0.2 equiv.)and DIC (2 equiv.) in DMF (1 ml/0.1 g of resin) onto the resin and the reaction mixture was stirred for 1 h.The suspension was filtrated, and the unreacted alcohol groups on the resin were capped using Ac 2 O (5 equiv.)and DMAP (0.5 equiv.)for 15 min and then washed with DMF.
For synthesis on 2-CT resin, the resin was washed with DCM (3 Â 1 min) followed by the first incorporation of Fmoc-amino acid (2 equiv.) in the presence of N,N-diisopropylethylamine (DIEA) (6 equiv.)for 1 h at rt.After that, the unreacted chlorotrityl groups were capped by adding MeOH to the loaded resin for 30 min at rt.Then, the suspension was washed (DCM, DMF) and Fmoc was removed.
Peptidyl resin was cleaved using TFA/TIS/H 2 O (95:2.5:2.5) for 45 min at rt.For linear somatostatin, TFA/anisole/TIS/H 2 O (90:5:2.5:2.5) was used as a cleavage mixture.After cleavage, the mixture was precipitated with chilled Et 2 O and then centrifuged.The pellet was then resuspended in H 2 O for analysis by HPLC and liquid chromatography-mass spectrometry (LCMS).

| RESULT AND DISCUSSION
In our previous study, the synthesis of linear somatostatin using 30 min (10 + 20 min) treatments with 20% Morph-DMF for Fmoc removal yielded unsatisfactory results, with the prevalence of several deletion sequences throughout the synthesis. 22This result indicated that the lower base strength of Morph impeded complete Fmoc removal during synthesis, even when it was used in large excess.
Given this consideration, we tested a higher proportion of Morph-DMF, starting from 50% Morph-DMF up to 100% Morph, over a range of Fmoc removal timeframes to determine the optimum condition.

| Swelling of the resin in Morph-DMF mixtures
As we intended to use a highly concentrated solution of Morph in DMF, we first studied the swelling of AM-resin (aminomethyl polystyrene [PS] resin crosslinked with 1% divinylbenzene) using 50% Morph-DMF and 100% Morph, as well as other standard removal solutions such as 20% PIP-DMF, 20% 4MP-DMF, and 10% PPZ (DMF-ethanol, 9:1).The results are outlined in Table 1 The results in Table 1 demonstrate that Morph is an excellent solvent to swell PS resin.In this regard, 100% Morph showed the best swelling properties, and 50% Morph-DMF gave the same swelling as that achieved with the standard 20% PIP-DMF.standard in research, were selected.These two combinations of time intervals were applied with increasing amounts of Morph (60%, 75%, 90%) in DMF, and even neat Morph (100%), to the same Fmoc-Lys (Boc)-Pro-NH-Rink-amide-AM resin.This was then followed by the incorporation of Fmoc-Phe-OH, cleavage of the peptides, and chromatographic analysis (Figure 3).

| Optimization of Fmoc removal condition using Morph mixtures
The results indicated that increasing from 20% to 50% of Morph has a strong positive effect on the removal of Fmoc, being the 50% Morph possibly optimal or very close to the optimal condition.Unexpectedly, a higher content of Morph (90% or 100%) was detrimental to the Fmoc removal process.This finding reinforces the notion that swelling capacity is not the only property of a solvent/reagent for it to be adequate for SPPS.Indeed, neat Morph swelled the AM resin even better than any of the mixtures.However, we should keep in mind the role of the polarity of the solvent in the Fmoc removal.By using 90%-100% Morph, it also is acting as a solvent for the reaction and its polarity is low compared with DMF.It has been demonstrated that solvents or mixtures of solvents with higher polarity are the best for Fmoc removal. 38ter initial optimization, we validated the Fmoc removal conditions with [65][66][67][68][69][70][71][72][73][74] ACP peptide, an aggregation-prone sequence. 39Previous experience indicated that Fmoc removal from Fmoc-Ile 69 -Asp (OtBu)-Tyr(tBu)-Ile-Asn(Trt)-Gly-NH-Rink-amide-AM resin is difficult during the assembly of 65-74 ACP peptide. 40The strategy followed is shown in Scheme 2.  2).

| DKP formation study
In SPPS, the formation of DKP is a side reaction that appears to be more pronounced in the synthesis of C-terminal peptide acids that have Gly or N-alkyl amino acids such as Pro residues in the terminal dipeptide.This effect is explained by the fact that the presence of these residues favors a cis-conformation for cyclization.The presence of one L and one D amino acid in the C-terminal dipeptide also predisposes to DKP formation as this arrangement confers greater stability to the resulting DKP with the side chains of the amino acid residues on opposite sides of the ring.Although this side reaction can take place by acid catalysis, it is much more prevalent in Fmoc chemistry during the removal of the Fmoc group of the second residue. 41This reaction is very severe when Wang resin is used.However, the steric hindrance around CT resin can minimize but not suppress it completely. 42,43DKP formation can also occur during the preparation of depsipeptides, where a peptide bond is substituted by a (thio)ester moiety, 44 and even in sequences where the leaving group is an amide from a peptide amide bond. 45In this study, we investigated the effect of Morph-DMF on DKP formation following the strategy shown in

| Study of aspartimide formation
One of the well-known side reactions in Fmoc-based SPPS is the formation of aspartimide (a five-membered ring structure) and byproducts derived thereof. 17,46Although this reaction can take place during acidolytic global deprotection, it occurs mainly during the removal of Fmoc groups by base (usually PIP) in each cycle of amino acid incorporation.The effect of this side reaction during peptide elongation is accumulative, so the greater the number of synthetic cycles, the more side reactions.Furthermore, the generation of aspartimide leads to a plethora of side products because aspartimide is not stable, and it can hydrolyze during the work-up, rendering the target peptide (α peptide) and also the β peptide.As aspartimide is prone to epimerization, the L and D α/β peptides are formed.Furthermore, aspartimide can be opened by the base (PIP) to give L/D α/β-piperidide peptides.In short, aspartimide and all related side products jeopardize the synthesis, leading to high impurities profile, which makes the purification harsh and as a result leads to poor yields.
Many strategies have been devised to minimize the formation of aspartimide during SPPS. 47These include the development of different side chain protections for Asp residues [48][49][50][51][52] and the use of backbone protections 51,[53][54][55] and different bases [56][57][58] and solvents for Fmoc removal. 59,60To study the extent of aspartimide formation driven by the use of 50% Morph-DMF compared with 20% PIP-DMF, we prepared a model α peptide (H-Ala-Lys(Boc)-Asp(OtBu)-Gly-Tyr(tBu)-Ile-NH-Rink-amide-AM resin).The sequence, derived from fragments 1-6 of toxin II from scorpion Androctonus australis Hector, is a widely known model for studying aspartimide and related side products. 61We ran two parallel syntheses using standard Fmoc SPPS conditions.In the first, Fmoc was removed with 20% PIP-DMF treatment with 50% Morph-DMF emerged as a suitable alternative to the 1 + 7 min treatment with 20% PIP-DMF for Fmoc removal.Furthermore, the β peptide had a greater tendency to give aspartimide than the α peptide.However, it remained inconclusive whether 50% Morph-DMF could minimize aspartimide formation when compared to the typical 20% PIP-DMF treatment.
Henceforth, we examined aspartimide formation under stress conditions.To this end, six equal portions of resin-bound α peptide were subjected to 4 h treatments at rt and 45 C using 20% PIP-DMF, 50% Morph-DMF, and 100% Morph.
After the treatments, the peptidyl resins were cleaved, and the resultant crude products were analyzed by HPLC and LCMS.
The results are collectively presented in Figure 6.Overall, the performance of 50% Morph-DMF and 100% Morph in minimizing aspartimide formation far outweighed that of 20% PIP.For the 20% PIP-DMF treatment, the conversion of α peptide to aspartimide and corresponding piperidides was 9.2% at rt whereas it increased to >70% at 45 C. In the case of 50% Morph-DMF and 100% Morph, the extent of α peptide conversion to morpholides was far less pronounced.For 50% Morph-DMF, the amount of such undesired sideproducts was 1.2% and 4.3% for rt and 45 C, respectively.Interestingly, the 100% Morph treatment produced merely 0.61% and 2.2% of morpholides at rt and 45 C, respectively.Finally, it can be concluded that Fmoc removal with a 10 + 20 min treatment of 50% Morph-DMF and 100% Morph is preferred for Fmoc-SPPS to minimize aspartimide formation.-Phe-Thr(tBu)-Ser(tBu)-Cys(Trt)-O-2-CT resin] following standard Fmoc-SPPS protocols.For both batches, the Fmoc removal cycles were performed with 10 + 20 min treatments using 20% PIP-DMF and 10 + 20 min treatments using 50% Morph-DMF, respectively.

| Synthesis of linear somatostatin
After cleavage with TFA, the crude peptides were evaluated by HPLC and LCMS (Figure 7).The two syntheses showed HPLC purity profiles of 77% and 79% for batches of peptide treated with 20% PIP-DMF and 50% Morph-DMF, respectively (peaks between 11 and 12 min are due to tert-butylation of Cys during the global deprotection).These results demonstrate that 20% PIP-DMF can be successfully substituted by 50% Morph-DMF with the bonus of fewer potential side reactions.

| CONCLUSION
During Fmoc-SPPS of a vulnerable peptide sequence, the Fmoc removal steps have always triggered a plethora of side reactions.
These are attributed mainly to the use of base, mostly PIP.To avoid or suppress side reactions during synthesis and also to comply with regulatory requirements, there is now a need to find a suitable alter- Fmoc-SPPS is a multi-factorial process that entails a simple yet complex network of interdependent reaction steps.The use of particular reagents during these reactions is often crucial in the sense that if they are not correctly optimized, the whole synthetic process is impaired.In this respect, Morph, which scores 7.5 in terms of greenness 18 and is not a regulated substance, could be considered a strong contender for Fmoc removal in SPPS.
Finally, the use of 60% minimize base-catalyzed side reactions, namely, DKP and aspartimide formation, and epimerization and β elimination when Cys is the C-terminal amino acid in the synthesis of peptide acids.To this end, we first tested the capacity of Morph to remove Fmoc by varying the proportion of Morph-DMF over different time intervals.With the best condition obtained through optimization, several experiments were then carried out to study the severity of prominent baseinfluenced side reactions in SPPS.The optimized Morph-DMF deprotection condition was tested for both DKP and aspartimide formation.Finally, the use of Morph-DMF as an Fmoc removal reagent was validated during the synthesis of linear somatostatin as a model peptide.
To determine the optimum condition for Fmoc removal, Fmoc-Lys-Pro-NH 2 was synthesized on Rink-AM resin following standard Fmoc-SPPS protocols.Next, 50% Morph in DMF over distinct deprotection timeframes was used to study Fmoc removal efficiency.Thus, samples of pre-swelled Fmoc-Lys(Boc)-Pro-NH-Rink-AM resin were first treated with 50% of Morph in DMF over combinations of time intervals (1 + 7 min, 2 + 10 min, 5 + 15 min, 10 + 2 min, 10 + 20 min, 10 min, 20 min, and 30 min).Fmoc-Phe-OH was then incorporated, the peptide was cleaved from the resin, and the crude product was analyzed by HPLC (UV detection at 299 nm) and LCMS.The sole presence of Fmoc-Phe-Lys-Pro-NH 2 in the crude product was expected to indicate efficient Fmoc removal.In contrast, the detection of Fmoc-Lys-Pro-NH 2 along with the desired Fmoctripeptide would be indicative of incomplete Fmoc removal (Scheme 1).The results are summarized in Figure 2. The results from the 50% Morph-DMF treatments indicated that the degree of Fmoc cleavage was proportional to increasing removal times, and 30 min of deprotection (in just one treatment or as two intervals (10 + 20 min)) emerged as the optimal time.For the next round of experiments, 10 + 20 min and 1 + 7 min, the latter the T A B L E 1 Swelling of the AM resin in different bases at room temperature.

10 +
HPLC chromatograms and percentage of the dipeptide Fmoc-Lys-Pro-NH 2 remaining after Fmoc removal with 50% Morph-DMF over different treatment times.F I G U R E 3 Fmoc removal study on the dipeptide Fmoc-Lys-pro-NH 2 using 10 + 20 min and 1 + 7 min treatments with increasing amounts of Morph (50%, 60%, 75%, and 90%) in DMF and with 100% Morph.S C H E M E 1 Study of Fmoc removal on the dipeptide Fmoc-Lys-Pro-NH 2 .At first, H-Asp(OtBu)-Tyr(tBu)-Ile-Asn(Trt)-Gly-NH-Rinkamide-AM resin was elongated following standard Fmoc SPPS protocols.The quality of the synthesis was verified by performing a minicleavage using TFA.Fmoc-Ile-OH was then incorporated, and Fmoc removal was performed with X% Morph-DMF (X = 50, 60, 75, 90, and 100) over two combinations of time intervals, 8 min (1 + 7 min) and 30 min (10 + 20 min).In addition to these, standard 20% PIP-DMF treatments for 8 min (1 + 7 min) and 30 min (20 min) were also performed to be used as references.Afterward, Fmoc-Gly-OH was incorporated into the free amino groups, followed by Fmoc removal (20% PIP-DMF; 1 + 7 min) and TFA cleavage.The crude peptides obtained were analyzed by HPLC and LCMS.A greater amount of Gly peptide should indicate the effectiveness of the Fmoc removal condition (Table

Scheme 3 .
Scheme 3. First, we looked for a suitable model dipeptide that would allow us to determine the impact of Morph instead of PIP on DKP formation.We chose Wang resin because the side reaction is more pronounced.Fmoc dipeptides containing D-Val-Pro, D-Val-Ala, Val-Pro, Lys-Pro, and Pro-Lys were synthesized on Wang resin using the conventional Fmoc/tBu-based SPPS strategy.The Fmoc group of the second residue was removed using 50% Morph-DMF or 100% Morph for 10 + 20 min or 20% PIP-DMF for 1 + 7 min as control.The formation of DKP releases free hydroxy groups with a double drawback: low yield and formation of deletion peptides where C-terminal amino acids are missing.The latter form during further elongation of the peptide chain even when using normal coupling reagents such as DIC and Oxyma.In each step, some Fmoc amino acid acylates the Wang resin released during DKP formation.The formation of these deletion peptides does not take place when CT resin is used due to the inability of the trityl alcohol to undergo esterification.In the case of the CT resin, the DKP formation is detrimental only to the yield.

S C H E M E 3 1 +
Study of DKP formation on the dipeptide Fmoc-Xaa 1 -Xaa 2 -O-Wang resin.After the elimination of the Fmoc group, Fmoc-Phe-OH was incorporated using DIC-OxymaPure to form mainly the Fmoc-tripeptide, although a small amount was incorporated into the Wang resin.Next, more Fmoc-Phe-OH is incorporated using DIC and DMAP for the acylation of the remaining hydroxy groups of the Wang resin.Cleavage of the peptide from the resin gave the tripeptide and Fmoc-Phe-OH.The latter is a measure of the DKP formed.HPLC and LCMS analysis of the crude product after cleavage revealed the extent of DKP formation (Figure4, SI-5 to SI-11 in SI).First, we studied DKP formation during the Fmoc removal on the sequence Fmoc-D-Val-Pro-O-Wang, a sequence exclusively prone to DKP formation.As expected, 20% PIP-DMF (1 + 7 min) showed 100% of Fmoc-Phe-OH, which in turn indicated 100% DKP formation.Next, the treatment with 50% Morph-DMF (10 + 20 min) resulted in a 98% DKP formation.Although this result is not acceptable for SPPS, it is the first indication that Morph tends to decrease DKP formation.Interestingly, the 100% Morph treatment (10 + 20 min) resulted in 78% DKP, and also, some traces of Fmoc-D-Val-Pro-OH were present in the crude product, thereby revealing the inefficiency of 100% Morph treatments to remove Fmoc, as observed earlier.When D-Val-Ala, Val-Pro, and Pro-Lys(Boc) were present, the formation of DKP was insignificant.Finally, we did the same study on Fmoc-Lys(Boc)-Pro-O-Wang resin.20% PIP-DMF treatments (7 min) yielded in 100% DKP formation, whereas 50% Morph-DMF (10 + 20 min) gave 87% DKP.The formation of DKP increased in the case of 100% Morph, reaching 98%.To conclude, our experiments reaffirm that DKP formation is highly sequence-dependent and more prominent when Pro is C-terminal, as Lys(Boc)-Pro, but not Pro-Lys (Boc), resulted in DKP formation.Importantly, the use of 50% Morph-DMF (10 + 20 min) reduced DKP formation when compared with 20% PIP-DMF treatments (1 + 7 min).Thus, the Morph mixture is recommended when there is a risk of DKP formation.
native to PIP.Our results show that the 10 + 20 min treatments with 50%-60% Morph-DMF efficiently remove Fmoc in SPPS.For the aggregation-prone sequence (Fmoc-IDYING-NH 2 ) used in our study to evaluate Fmoc removal, a 10 + 20 min reaction with 50%-60% Morph-DMF outperformed a typical 1 + 7 min treatment with 20% PIP-DMF.In terms of DKP and aspartimide formation, the two most prominent base-induced side reactions, the performance of 50%-60% Morph-DMF was superior to 20% PIP-DMF.These new Fmoc removal conditions could be implemented in many cases in which DKP formation can occur.45With respect to the suppression of aspartimide formation, Morph outperformed PIP at both rt and under stress conditions.Finally, all these positive outcomes culminated in the successful synthesis of two batches of linear somatostatin utilizing 50% Morph-DMF and 20% PIP-DMF.The two batches had similar chromatographic purity profiles.More importantly, here we demonstrate for the first time that 50% Morph-DMF can be used to remove Fmoc during the synthesis of a peptide sequence.On the basis of our results, 50%-60% Morph-DMF emerges as a viable alternative for Fmoc removal steps in SPPS.As Morph does not cause any major problems, Fmoc removal time for long sequences could be prolonged to 40 min or even longer to assure a complete Fmoc removal.A prolonged cycle time is actually a (probably acceptable) disadvantage.

Table 2
, the treatments with 50% and 60% Morph-DMF showed the best performance with both removal timeframes (1 + 7 min and 10 + 20 min).Interestingly, solutions of 50% or 60% Morph in DMF for 10 + 20 min gave similar results to the standard 20% PIP-DMF.Again, a larger amount of Morph resulted in poor performance, neat Morph being the worst scenario.T A B L E 2 Fmoc removal study on Fmoc-Ile-DYING-NH-Rink-AM resin.
62rph-DMF instead of 20% PIP-DMF has little economic significance.PIP is approximately double more expensive than Morph, which means that the Fmoc removal withMorph is approximately 50% more expensive, but considering that the cost of the base used to remove the Fmoc group accounts for less than 0.5% of the total cost including reagents and solvents,62the cost increase is negligible.On the other hand, as Morph is approximately four times cheaper than 4MP, the use of Morph in comparison with 4MP is even advantageous.