Position-dependent impact of hexafluoroleucine and trifluoroisoleucine on protease digestion

Article abstract of DOI:10.3762/bjoc.13.279

Rapid digestion by proteases limits the application of peptides as therapeutics. One strategy to increase the proteolytic stability of peptides is the modification with fluorinated amino acids. This study presents a systematic investigation of the effects of fluorinated leucine and isoleucine derivatives on the proteolytic stability of a peptide that was designed to comprise substrate specificities of different proteases. Therefore, leucine, isoleucine, and their side-chain fluorinated variants were site-specifically incorporated at different positions of this peptide resulting in a library of 13 distinct peptides. The stability of these peptides towards proteolysis by α-chymotrypsin, pepsin, proteinase K, and elastase was studied, and this process was followed by an FL-RP-HPLC assay in combination with mass spectrometry. In a few cases, we observed an exceptional increase in proteolytic stability upon introduction of the fluorine substituents. The opposite phenomenon was observed in other cases, and this may be explained by specific interactions of fluorinated residues with the respective enzyme binding sites. Noteworthy is that 5, 5, 5-trifluoroisoleucine is able to significantly protect peptides from proteolysis by all enzymes included in this study when positioned N-terminal to the cleavage site. These results provide valuable information for the application of fluorinated amino acids in the design of proteolytically stable peptide-based pharmaceuticals.

Full text of DOI:10.3762/bjoc.13.279

Position-dependent impact of hexafluoroleucine and
trifluoroisoleucine on protease digestion
Susanne Huhmann, Anne-Katrin Stegemann, Kristin Folmert, Damian Klemczak,
Johann Moschner, Michelle Kube and Beate Koksch*
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2017, 13, 2869–2882.
Department of Chemistry and Biochemistry, Freie Universität Berlin,
Takustraße 3, 14195 Berlin, Germany
doi:10.3762/bjoc.13.279
Received: 29 September 2017
Accepted: 13 December 2017
Published: 22 December 2017
Email:
Beate Koksch* - beate.koksch@fu-berlin.de
* Corresponding author
This article is part of the Thematic Series "Organo-fluorine chemistry IV".
Guest Editor: D. O'Hagan
Keywords:
fluorinated amino acids; hexafluoroleucine; peptide drugs; protease
stability; trifluoroisoleucine
© 2017 Huhmann et al.; licensee Beilstein-Institut.
License and terms: see end of document.
Abstract
Rapid digestion by proteases limits the application of peptides as therapeutics. One strategy to increase the proteolytic stability of
peptides is the modification with fluorinated amino acids. This study presents a systematic investigation of the effects of fluori-
nated leucine and isoleucine derivatives on the proteolytic stability of a peptide that was designed to comprise substrate specifici-
ties of different proteases. Therefore, leucine, isoleucine, and their side-chain fluorinated variants were site-specifically incorporat-
ed at different positions of this peptide resulting in a library of 13 distinct peptides. The stability of these peptides towards proteol-
ysis by α-chymotrypsin, pepsin, proteinase K, and elastase was studied, and this process was followed by an FL-RP-HPLC assay in
combination with mass spectrometry. In a few cases, we observed an exceptional increase in proteolytic stability upon introduction
of the fluorine substituents. The opposite phenomenon was observed in other cases, and this may be explained by specific interac-
tions of fluorinated residues with the respective enzyme binding sites. Noteworthy is that 5,5,5-trifluoroisoleucine is able to signifi-
cantly protect peptides from proteolysis by all enzymes included in this study when positioned N-terminal to the cleavage site.
These results provide valuable information for the application of fluorinated amino acids in the design of proteolytically stable
peptide-based pharmaceuticals.
Introduction
Peptide-based drugs are promising pharmaceuticals since they However, their application as drugs is often limited due to low
offer several advantages including high selectivity, specificity, oral bioavailability and a short half-life attributable in part to
and efficacy for recognizing and binding to their targets [1-6]. proteases of the digestive system and blood plasma [1-8]. Effi-
2869

Beilstein J. Org. Chem. 2017, 13, 2869–2882.
cient approaches to overcome these limitations have been de- phenylalanine variants into the histone acetyltransferase protein
veloped including the incorporation of non-natural amino acids, tGN5 resulted in destabilization in a chymotrypsin digestion
such as D-amino acids, backbone-extended or chemically modi- assay [30]. Substitution of tryptophan, tyrosine, and phenylala-
fied amino acids [1]. In this regard, the incorporation of fluo- nine residues in a glycosylation-deficient mutant of Candida
rine into amino acids has become a promising strategy. Fluo- antarctica lipase B, CalB N74D, by their monofluorinated ana-
rine’s unique properties, namely low polarizability, a strong logues, left the resistance to proteolytic degradation by
inductive effect, and high electronegativity, as well as its small proteinase K unchanged [31]. Incorporation of α-fluoroalkyl-
size, result in strong, short C–F bonds and perturb the acidity substituted amino acids can also lead to proteolytically stable
and basicity of adjacent functional groups. Moreover, these peptides, and proteases can even be used to synthesize α-fluo-
changes may strongly influence hydrogen bonding and electro- roalkyl-substituted peptides [32-38].
static interactions that are crucial for binding to receptors or, in
context of protease stability, enzymes. Thus, when introduced These results indicate that the influence of fluorinated amino
in the form of fluorinated amino acids, this unique element can acids on the proteolytic stability of peptides and proteins
alter the biophysical, chemical and pharmaceutical properties of remains difficult to predict. In an attempt to systematically
proteins and peptides including their interaction with proteases study the influence of fluorinated amino acids on the proteo-
[9,10].
lytic stability of peptides, a 10-amino acid peptide (FA) was
previously designed in our group, comprising the substrate
Several laboratories have focused on introducing highly fluori- specificities of the proteases α-chymotrypsin and pepsin
nated analogues of hydrophobic amino acids and have studied [39,40]. 2-Aminobutanoic acid (Abu) and its fluorinated ana-
the effects on stability of the resulting proteins towards thermal logues 2-amino-4,4-difluorobutanoic acid (DfeGly) and
and chemical denaturation [9,11-22]. These studies prompted 2-amino-4,4,4-trifluorobutanoic acid (TfeGly) were individu-
further investigation into the extent to which fluorinated amino ally incorporated at either the P2, the P1’ or the P2’ position
acids stabilize peptides and proteins against proteolytic degra- [41] to give nine different analogues of FA. In prior studies, we
dation in particular. Meng and Kumar reported that the incorpo- observed that the introduction of fluorine atoms into the Abu
ration of 5,5,5,5’,5’,5’-hexafluoroleucine (HfLeu) into the anti- side chain can significantly improve or dramatically reduce
microbial peptides magainin 2 amide and buforin enhanced resistance to hydrolysis by different enzymes and human blood
their resistance towards proteolytic degradation by trypsin [23]. plasma, depending upon the fluorine content of the side chain,
They also introduced HfLeu into the glucagon-like-peptide-1 the position of the substitution relative to the cleavage site and
(GLP-1), which is an attractive lead compound for the treat- the particular protease [39,40].
ment of diabetes mellitus type 2. Unfortunately, the clinical use
of the wild-type peptide is severely hampered due to rapid Here, we extend these studies to include highly fluorinated,
digestion (≈2 min) by the serine protease dipeptidyl peptidase sterically demanding HfLeu, and 5,5,5-trifluoroisoleucine
[24-26]. Satisfyingly, the fluorinated GLP-1 analogues (TfIle) and to investigate their effects on proteolytic stability
displayed higher proteolytic stability against this enzyme [27]. towards the serine proteases α-chymotrypsin, elastase, and
proteinase K, and the aspartate protease pepsin.
Usually, the enhanced proteolytic stability of fluorinated
Results and Discussion
peptides is explained by their greater hydrophobicity and altered
secondary structure compared to the parent, non-fluorinated Peptide design and structure
peptide. A further reason is the increased steric bulk of the fluo- To elucidate the impact of fluorination on proteolytic stability
rinated amino acid, meaning protection from protease degrada- we previously designed the peptide FA (Figure 1b) that
tion is a result of the steric occlusion of the peptide from the comprises the substrate specificities of α-chymotrypsin and
active site [23,28]. In contrast, the Marsh lab found that the pepsin [39,40]. Consequently, the P1 position is occupied by a
introduction of HfLeu into the antimicrobial peptide MSI-78 phenylalanine residue. Lysine residues were introduced at both
only renders it more stable towards proteolysis by trypsin and ends of the peptide sequence to enhance solubility, and
chymotrypsin in the presence of liposomes [29]. In the absence o-aminobenzoic acid (Abz) at the N-terminus serves as a fluo-
of liposomes, the fluorinated variants were as rapidly degraded rescence label. Alanine residues in positions P3, P3’, and P4’
as the non-fluorinated control, suggesting that the incorporation act as spacers as the peptide binds in an extended conformation
of HfLeu is not the only factor that prevents the peptide from to the enzyme’s active site [42]. The positions P2, P1’ and P2’
being digested by proteases [29]. Fluorinated aromatic amino at or adjacent to the cleavage site [41] carry the key residues for
acids were also investigated regarding their impact on peptide the recognition of the substrate by the protease and serve as
proteolysis. For instance, incorporation of monofluorinated substitution sites.
2870

Beilstein J. Org. Chem. 2017, 13, 2869–2882.
Figure 1: (a) Structures of isoleucine (1), leucine (2), and their fluorinated analogues 5,5,5-trifluoroisoleucine (3, TfIle) and 5,5,5,5’,5’,5’-hexafluoro-
leucine (4, HfLeu). The van der Waals volumes given in parentheses correspond to the side chains (starting at α-carbon), and were calculated accord-
ing to Zhao et al. [43]; (b) Names and amino acid sequences of the studied peptides; the substitution positions are marked as Xaa; Xaa = Leu, HfLeu,
Ile or TfIle. Positions are named according to Schechter and Berger nomenclature [41].
The alanines at P2, P1’ or P2’ positions were substituted indi- The course of proteolytic digestion was characterized by an an-
vidually with either TfIle or HfLeu (Figure 1a). Ile and Leu alytical HPLC-assay with fluorescence detection [39,40]. For
variants were also included in this study as non-fluorinated quantification of substrate degradation, integration of the corre-
controls. This led to a library of 12 FA variants (Figure 1b). sponding HPLC peak was conducted. Cleavage products were
identified by ESI–ToF mass spectrometry (see Supporting
Leu and Ile are larger and more hydrophobic than Ala. The Information File 1, Tables S4–S7). Figure S2 (Supporting Infor-
fluorinated amino acids are even larger and more hydrophobic mation File 1) shows the time course of all of the digestion ex-
than their hydrocarbon analogues [44,45]. Furthermore, fluo- periments. A detailed description of the results for the indi-
rine substitution has been shown to polarize neighboring C–H vidual enzymes is given in the following sections.
bonds (here the γ-hydrogens) that could affect noncovalent
Proteolytic stability towards α-chymotrypsin
interactions [9,11]. Since the amino acids used here (Figure 1a)
differ in their degree of fluorination, spatial demand and hydro- (EC 3.4.21.1)
phobicity, it is expected that they will have different impacts on α-Chymotrypsin is a serine endopeptidase with broad speci-
the enzyme’s binding pocket, reflected by different behavior in ficity. It preferably cleaves peptide bonds C-terminal to large
the proteolysis assay.
hydrophobic residues such as phenylalanine, tyrosine, trypto-
phan, and leucine in the P1 position. Secondary hydrolysis also
occurs at the carbonyl end of isoleucine, methionine, serine,
Determination of proteolytic stability
All peptides were incubated with the four different proteases threonine, valine, histidine, glycine, and alanine [47,58-60].
and their proteolytic degradation was followed over a period of
24 h. Both, α-chymotrypsin [46-49] and pepsin [50-54] are well The S2 subsite of α-chymotrypsin generally prefers to accom-
characterized digestive proteases. They are, together with modate hydrophobic residues [59,61]. We observed that the
trypsin, the main enzymes of the human digestive system. Elas- fluorinated P2 variants show a smaller amount of digestion after
tase possesses a wide substrate specificity for non-aromatic, 120 min compared to their hydrocarbon analogues, while all
neutral side chains [55,56] and is found in the human pancreas variants are more stable than the control FA (Figure 2). After
and in blood serum. Proteinase K, an enzyme widely used for 24 h, all P2 peptides except for P2-LeuFA are still more stable
inactivation and degradation studies of proteins, was included than FA. Incorporation of Leu into P2 leads to complete prote-
here since it shows a broad substrate specificity and high activi- olysis compared to FA, in which Ala occupies this position.
ty and, thus, is able to digest numerous native proteins, even in However, incorporation of six fluorine substituents into Leu (re-
the presence of detergents [57]. These four enzymes have dif- sulting in HfLeu) results in an almost 100% gain in proteolytic
ferent preferences at their subsites, thus providing a broad scope stability. Ile is not as highly preferred in P2 as Leu, but also
for our investigations.
here the introduction of three fluorine substituents leads to a
2871

Beilstein J. Org. Chem. 2017, 13, 2869–2882.
Figure 2: Percentage of substrate remaining after incubation for 120 min (left) and 24 h (right) with α-chymotrypsin in 10 mM phosphate buffer,
pH 7.4, at 30 °C. The data shown represent the average of three independent measurements. Errors are derived from the standard deviation.
50% gain in stability. P2-HfLeuFA and P2-TfIleFA are not ESI–ToF mass analysis confirms that the position P1 bearing
digested at all, suggesting that HfLeu and TfIle are not favored Phe is the main cleavage site for α-chymotrypsin (Figure 3).
within the S2 pocket of α-chymotrypsin.
Cleavage C-terminal to Leu and HfLeu in P1’-LeuFA and P1’-
HfLeuFA is also observed (Figure 3, see Supporting Informa-
P1’-substituted peptides are all more stable towards digestion tion File 1, Table S4), which means that the cleavage site was
than the control peptide FA, while Leu seems to provide the shifted towards the C-terminus by one residue. This is likely a
best protection from proteolysis. Here, introduction of fluorine consequence of α-chymotrypsin’s preference for not only aro-
makes the peptide prone to degradation. The opposite is true for matic residues but also bulky hydrophobic residues in the S1
Ile as TfIle leads to less efficient degradation. The S1’ subsite pocket, thus, HfLeu is accepted by the P1 binding pocket of
of α-chymotrypsin usually accommodates basic residues with α-chymotrypsin.
long side chains [59,62,63]. Ile, as a branched amino acid, is
obviously not well accommodated in this position for steric
reasons. A further increase in side chain volume with TfIle
exacerbates this effect. In the case of HfLeu, however, fluorine
substituents seem to engage in favorable interactions with
amino acid residues of the binding site, thus making P1’-
HfLeuFA a better substrate than the non-fluorinated Leu
peptide. Several such interactions are possible, as described in
our previous work [39,40,64].
The S2’ subsite of α-chymotrypsin exhibits a hydrophobic char-
acter and thus prefers to accommodate hydrophobic residues
[59,65]. However, the more hydrophobic peptides P2’-LeuFA,
P2’-HfLeuFA, P2’-IleFA, P2’-TfIleFA are more stable against
digestion by α-chymotrypsin compared to FA after 120 min of
incubation. After 24 h, only the fluorinated analogues are less
degraded than the control FA. Full length P2’-HfLeuFA and
Figure 3: Cleavage positions observed in the digestion of library
peptides with α-chymotrypsin.
P2’-TfIleFA are present at percentages up to 44% and 24%, re-
spectively, while substitution by Leu and Ile in P2’ position
leads to accelerated proteolysis compared to FA. Thus, both
HfLeu and TfIle have a protective effect towards proteolysis in In summary, the introduction of fluorinated Leu and Ile ana-
this position.
logues into a α-chymotrypsin specific peptide sequence can
2872

Beilstein J. Org. Chem. 2017, 13, 2869–2882.
improve proteolytic stability mainly at the P2 and P2’ positions, as well as Ile are well accommodated in the S2 subsite of
with the strongest effects observed for the P2 position.
pepsin. In contrast, TfIle, although smaller than HfLeu [44],
doesn’t appear to fit well into the S2 pocket of pepsin.
Proteolytic stability towards pepsin
(EC 3.4.23.1)
To compare the P1’ substituted peptides, only P1’-HfLeuFA
Pepsin is an aspartic endopeptidase and one of the main diges- shows the same persistence after 120 min as the control peptide
tive enzymes in humans. It exhibits specificity for hydrophobic, FA, while all other sequences are digested faster. Here, the
especially aromatic residues like Phe, Trp, and Tyr at the P1 introduction of fluorine into Leu seems to stabilize the peptide
and P1’ positions [50-54]. It has an extended active site that can by about 20%. Interestingly, after 24 h this trend is reversed,
bind at least seven residues [66,67], and peptide bond cleavage and the P1’-HfLeuFA peptide is destabilized to an amount of
occurs N-terminal to the residue at position P1. The cleavage 17% compared to the hydrocarbon analogue, but both peptides
efficiency heavily depends upon the identity of this amino acid, are somewhat more stable than the control FA. The incorpora-
with Phe and Leu being the most favored residues. At the P1’ tion of TfIle into this position doesn’t show a significant
position aromatic amino acid residues are preferred, however, impact. Although the fluorine substituents slow down the diges-
the influence of the P1’ position on proteolytic cleavage is not tion process (see Supporting Information File 1, Figure S2b),
as significant [68]. Pepsin typically does not cleave at Val, Ala, the TfIle containing peptide as well as its hydrocarbon ana-
or Gly linkages [60].
logue are fully digested after 24 h. The S1’ subsite has hydro-
phobic character and thus prefers to accommodate hydrophobic
The S2 subsite of pepsin preferentially accommodates hydro- or aromatic residues [71]. Ile and TfIle are obviously well
phobic residues such as Leu, Ala or norleucine as well as the accommodated in this position, while Leu and HfLeu are not.
β-branched species Ile and Val, but can also bind charged
residues [69,70]. Except for P2-TfIleFA, we observed that the The S2’ subsite of pepsin favors hydrophobic amino acids, but
P2-modified peptides are degraded more rapidly than the also accepts charged polar amino acids like Glu and Thr
control peptide FA and that these peptides are almost or com- [52,72]. After 120 min peptides P2’-LeuFA, P2’-IleFA and P2’-
pletely degraded after 120 min (Figure 4). For example, where- TfIleFA are degraded faster than the control FA. Instead, P2’-
as after 24 h FA is also almost completely degraded, HfLeuFA is only digested up to 5%. This effect is even more
P2-TfIleFA is still detected at a level of 100%. Incorporation of pronounced after 24 h. While all other P1’ substituted peptides,
Leu or Ile leads to complete proteolysis. Remarkably, the intro- along with the control peptide FA, are almost or completely
duction of six fluorine atoms into Leu doesn’t change this be- digested, P2’-HfLeuFA is still present to about 76%. In this
havior. In sharp contrast yet equally remarkable, the incorpora- case fluorination leads to protection against proteolysis by
tion of three fluorine substituents into Ile results in a 100% gain pepsin. Hfleu is obviously not well accommodated in this posi-
in proteolytic stability. These results indicate that Leu, HfLeu, tion. As already observed for position P1’, the introduction of
Figure 4: Percentage of substrate remaining after incubation for 120 min (left) and 24 h (right) with pepsin in 10 mM acetate buffer, pH 4.0, at 30 °C.
The data shown represent the average of three independent measurements. Errors are derived from the standard deviation.
2873

Beilstein J. Org. Chem. 2017, 13, 2869–2882.
three fluorine atoms into Ile slows down proteolysis, although tional cleavage site at which the peptide bonds LeuP1’-AlaP2’,
both peptides are completely digested after 24 h.
and HfLeuP1’-AlaP2’ are hydrolyzed, respectively, which means
that the cleavage site was shifted towards the C-terminus by one
For almost all peptides of our library, we observed the expected residue. This cleavage pattern was also detected for α-chymo-
cleavage pattern with Phe in the P1 position (Figure 5, see Sup- trypsin before, and indicates that HfLeu is well accepted by
porting Information File 1, Table S5). Only P2’-HfLeuFA is not pepsin in its S1 binding site. Furthermore, we identified a
hydrolyzed at the designed cleavage site, instead cleavage second cleavage site for P2-HfLeuFA at which the peptide bond
occurs exclusively N-terminal to the HfLeu residue, thus N-terminal to Phe is proteolytically cleaved as well. This means
demonstrating that HfLeu occupies the P1’ position. In the case that the cleavage site is shifted such that HfLeu occupies the P1
of P2’-TfIle we found two further peptide bonds that are and Phe the P1’ position. However, this perfectly matches the
cleaved by pepsin, namely N-terminal cleavage to TfIle and to specificity of pepsin that prefers bulky hydrophobic and aromat-
Phe. These findings indicate that the S1’ subsite accommodates ic amino acids both up- and downstream of the scissile bond.
bulky hydrophobic residues more readily than does the S2’ site
of pepsin. For P1’-LeuFA and P1’-HfLeuFA we found an addi- In summary, the introduction of fluorinated Leu into a pepsin
specific peptide sequence can improve the proteolytic stability
at the P2’ position, whereas the incorporation of a fluorinated
Ile into the P2 position shows the strongest effect in protection
from proteolysis.
Proteolytic stability towards elastase
(EC 3.4.21.36)
Elastase is a serine endopeptidase, and has a wide specificity for
non-aromatic uncharged side chains. It preferentially cleaves
peptide bonds C-terminal to small uncharged non-aromatic
amino acid residues such as glycine, alanine and serine, but also
valine, leucine, isoleucine [56,73]. Its binding site extends over
eight subsites (S5 to S1, and S1’ to S3’) [74].
The fact that in this study larger and more hydrophobic amino
acids [44,45] were introduced may explain why degradation of
Figure 5: Cleavage positions for digestion of the different peptides
with pepsin.
most of the variants during the first 120 min of incubation with
elastase is hardly observed (Figure 6). Only P2’-LeuFA, P2’-
Figure 6: Percentage of substrate remaining after incubation for 120 min (left) and 24 h (right) with elastase in 100 mM Tris/HCl buffer, pH 8.4, at
37 °C. The data shown represent the mean of three independent measurements. Errors are derived from the standard deviation.
2874

Beilstein J. Org. Chem. 2017, 13, 2869–2882.
IleFA, and P2’-TfIleFA were somewhat digested during this
time, however, all of the modified peptides are more stable than
the control FA.
After 24 h all P2 peptide variants are more stable than the
control FA (Figure 6), while TfIle provides the best protection
from elastase digestion. Leu and Ile are not quite as preferred in
P2 position as is Ala. Fluorination of Leu leads to an increase in
stability of around 19%. With 35% this effect is even higher
when three fluorine atoms are introduced into Ile.
Modification of the P1’ postion renders P1’-HfLeuFA and P1’-
IleFA more stable than the control peptide FA, while P1’-
TfIleFA is comparably stable. Incorporation of Leu into P1’
leads to complete digestion. However, introducing six fluorine
atoms into Leu results in an 85% gain in stability. The opposite
is observed for Ile, where TfIle accelerates enzymatic degrada-
tion.
Figure 7: Cleavage positions for digestion of the different peptides
with elastase.
The S1’ subsite usually prefers Lys residues, and to a lesser
Except for P2’-IleFA, all P2’ modified variants are more stable extent Ala or Glu [74,76]. Indeed, we found a fragment cleaved
compared to the control peptide FA after 24 h. Leu is not as off corresponding to a Lys in P1’, but primarily detected frag-
preferred in this position as Ala. Introduction of fluorine ments with Ala in P1’ and also Phe that was even more favored
strengthens this effect and effectively doubles the stability. In than Lys. We observed that Ile as well as its fluorinated ana-
contrast, introduction of Ile leads to almost complete proteol- logue TfIle are not accommodated in this subsite, probably due
ysis. However, substitution by TfIle slows down the degrada- to their β-branched topology.
tion rate and results in a stabilization of around 37%. In P2’,
fluorination shows in both cases a protective effect towards The S3’ pocket in elastase is known to have a high aromatic
hydrolysis by elastase.
specificity [74]. Interestingly, in our cases Phe in P3’ was less
favored. Instead, mainly Lys occupied this position.
Elastase preferably hydrolyses peptide bonds C-terminal to
uncharged non-aromatic amino acids and mainly between Ala is favored in P2. Its carboxyl group can form a hydrogen
Ala–Ala and Ala–Gly bonds [56,73]. Since Ala is the main bond with the amide nitrogen of Gly193 in the S2 pocket, and
residue present in the peptides studied here, we observed Ala’s methyl group faces the solvent [76].
various cleavage products in the ESI–ToF analysis (Figure 7,
see Supporting Information File 1, Table S6). For none of the Occupation of the S4 subsite is important for efficient catalysis
peptides were fragments with Phe in the P1 position observed. [76,77]. Thus elastase might not easily split the first three bonds
Since elastase has a constricted S1 pocket, the binding of aro- at the amino terminus of a peptide chain, since interactions of a
matic amino acids at P1 is deleterious [75]. Here, we also ob- residue with S4 is necessary [77]. Indeed, we only observed a
served that Leu appears to never occupy the P1 position, al- low amount of cleavage proximal to the N-terminus, while most
though it is known to occupy this position in other substrates of the hydrolysis occurred at the C-terminal end of the peptides.
[73]. Interestingly, the larger fluorinated variant was found in The S3 subsite seems to accommodate bulkier hydrophobic
the P1 position in one case, while Ile and its fluorinated ana- amino acids well, as we observed cleavage products containing
logue occupy this position in two of the three peptide ana- Ile and Leu in P3 position for all the peptides modified with
logues.
these residues, as well as their fluorinated analogues for two of
the three substituted peptides each.
The S2’ subsite of elastase has a marked specificity for Ala, and
can accommodate bulkier residues only with some difficulty In summary, introduction of HfLeu in different positions of a
[74]. Thus, we did not find the fluorinated amino acids HfLeu peptide can enhance the proteolytic stability up to 85% com-
and TfIle binding to the S2’ subsite of the enzyme as expected, pared to the corresponding Leu analogues. Replacing Ile with
whereas for the Leu and Ile variants this was observed only in TfIle can increase the stability against elastase as well, al-
one case each.
though not as efficiently as HfLeu.
2875

Beilstein J. Org. Chem. 2017, 13, 2869–2882.
Proteolytic stability towards proteinase K
(EC 3.4.21.64)
this position is not as preferred as is Leu and this enhances the
stability to a small extent compared to FA. Introducing three
Proteinase K is a non-specific serine endopeptidase and the fluorine atoms at the Ile side chains strengthens the stability
main proteolytic enzyme produced by the fungus Tritirachium against proteinase K even further.
album Limber [78]. It has a broad substrate specificity, cleaving
peptide bonds C-terminal to a number of amino acids, however, As already observed for the other two Leu containing peptides,
prefers aromatic or aliphatic hydrophobic amino acids in posi- also substitution of Ala at position P2’ with Leu does not
tion P1 [57,78]. Furthermore, Ala is favored in position P2 and change the resistance against proteinase K significantly. Inter-
enhances cleavage efficiency [79,80]. Proteinase K possesses a estingly, when six fluorine atoms are introduced, the digestion
very high proteolytic activity [79]. Its active center contains an process is faster and P2’-HfLeuFA is almost completely
extended binding region consisting of several subsites, at least degraded after 120 min. The opposite is observed for the fluori-
four or five subsites on the N-terminal side of the scissile bond nation of Ile. While P2’-IleFA is as stable as FA, P2’-TfIleFA
(S1 to S4/S5) and three subsites C-terminal to the scissile bond shows a small gain in stability of around 12%.
(S1’ to S3’) [81-83]. The “bottom” of substrate recognition site
is predominantly hydrophobic and there is evidence that not the Thus, in the case of P1’ as well as P2’ peptide variants, only
sequence of the substrate is of importance in the recognition but fluorination of Ile leads to a slower digestion by proteinase K,
only the volume of the side chains [84].
while introducing even more fluorine atoms into the Leu side
chain leads to more rapid hydrolysis compared to the non-fluo-
Substitution of Ala in position P2 with Ile and Leu leads to a rinated analogues. Ile seems in all investigated cases not as
greater or comparable amount of degradation after 120 min. preferred as Leu, since less efficient digestion is observed.
Introducing fluorine atoms in both cases slows down the diges- Introduction of three fluorine atoms even enhances this protec-
tion process, most pronounced for P2-TfIleFA with a gain of tive effect.
60% in stability compared to its non-fluorinated analogue. Ile is
not preferred to the extent that Leu is, and the introduction of Based on the wide substrate specificity of proteinase K and its
fluorine enhances this effect. While all other peptides are almost preference for alanine, and since our studied peptides have a
completely or entirely degraded after 24 h, P2-TfIleFA is the high number of alanine residues present, there are multiple
only peptide that is still left after 24 h of incubation (Figure 8). cleavage sites possible in addition to the designed site between
PheP1 and XaaP1’. Indeed, multiple cleavage patterns are ob-
Introduction of Leu at the P1’ position leads to an amount of served, especially cleavage C-terminal to Ala (Figure 9, see
digestion comparable to FA after 120 min. Fluorination of the Supporting Information File 1, Table S7). Thus, Ala mainly
Leu side chain leads to a small acceleration in digestion. Ile at occupies the S1 subsite, but is also found to bind to the S2 site
Figure 8: Percentage of substrate remaining after incubation for 120 min (left) and 24 h (right) with proteinase K in 50 mM Tris/HCl buffer, containing
10 mM CaCl2 pH 7.5, at 37 °C. The data shown represent the mean of three independent measurements. Errors are derived from the standard devia-
tion.
2876

Beilstein J. Org. Chem. 2017, 13, 2869–2882.
defined as a “cleft” or “pocket” [79]. The P3 residue of the
peptide substrate lies on the protein surface and the side chain
of P3 should be directed toward the solvent [79]. This arrange-
ment might explain the broad specificity of S3 [79]. We ob-
served that all the residues used in this study can occupy the P3
position, mainly Phe and Lys. Leu, Ile and TfIle are also found
to a great extent in P3.
In summary, fluorination of an Ile residue N-terminal to the
cleavage site can help to protect a peptide against proteolysis by
proteinase K. Due to its broad specificity and high activity,
proteinase K typically digests peptides quickly [57]. This was
also observed in this work in experiments in which all peptides,
except for P2-TfIleFA, a remarkably stable species, were com-
pletely degraded after an incubation time of 24 h.
Figure 9: Cleavage positions found with proteinase K.
Conclusion
The bulky side-chain fluorinated amino acids HfLeu and TfIle
to a greater extent. Ala is most effective in P2 [80] as the S2 have the power to significantly stabilize peptides against proteo-
subsite is a small and narrow cleft, which limits the possibili- lytic degradation. The impact of their incorporation on the
ties for effective side chain substitutions [79]. However, Ile and proteolytic stability of peptides does not follow a general trend
TfIle are well accepted here. A negative or positive charge at S2 but rather depends on a combination of factors including the
is not preferred and hampers the formation of the enzyme sub- nature of the fluorinated amino acid, the substitution position
strate complex [82]. Thus, it can be concluded that Lys is relative to the cleavage site and the studied protease. Also, in
poorly accepted at this position. HfLeu, the most sterically contrast to proteolytic studies published before [23,27,28], the
bulky amino acid investigated here, is not observed to occupy expectation of a general increase in proteolytic stability as a
the S2 subsite. Instead, HfLeu is mainly found to bind to the S1 result of steric occlusion of the peptide from the active site upon
pocket. Leu is also found to occupy the S1 site of proteinase K, incorporation of sterically demanding fluorinated amino acids
which is large and has mainly hydrophobic character could not be verified based on the results of our current study.
[82,83,85]. It does not impose too strong steric limitations on We found a significant stabilization towards proteolysis in 13 of
the amino acid side chain but prefers hydrophobic and aromatic a total of 24 peptides of the library studied here upon introduc-
residues, with a specificity for Ala [78,81,83,86]. Charged side tion of either HfLeu or TfIle (Figure 10). However, we ob-
chains of Glu and Lys are very poorly accepted, as are served that even these sterically demanding fluorinated amino
β-branched functional groups, because the entrance to the S1 acids show in some cases favorable interactions with the en-
subsite is too narrow to allow their passage [79]. Thus, Lys is zymes binding sites resulting in a more rapid digestion as the
not observed to occupy the S1 subsite. Neither Ile nor TfIle can non-fluorinated control.
be accommodated by the S1 pocket due to their β-branching.
Phe is found in S1 in only two cases in our study, and mainly The introduction of fluorinated Leu and Ile analogues into P2
occupies the S3 and S1’ pockets. The S3 pocket has a wide and P2’ position improved the proteolytic stability towards
specificity due to its location at the protein surface, but exhibits α-chymotrypsin. When introduced in the P1’ position a stabi-
preference for aromatic side chains in P3 (Trp, Phe) [79]. S1’ lization was still observed for TfIle, while incorporation of
shows a slight preference for smaller residues like Ala and Gly, HfLeu made the peptide more prone to proteolytic digestion
but also bulkier residues such as Phe and Leu are hydrolyzed to compared to the non-fluorinated control. Incorporation of
a significant extent [81]. In this study Leu apparently does not HfLeu had a significantly stabilizing effect towards hydrolysis
bind to the S1’ site at all, and this is also true of TfIle. Addition- by pepsin only in P2’ position, while TfIle develops a protec-
ally, Lys is not well accommodated here. Phe is also found to tive effect only when incorporated in P2 position.
occupy S4 to a great extent, and this subsite is known to have an
affinity for aromatic groups, especially a marked preference for As both, elastase and proteinase K possess a broad specificity,
Phe [79]. S4–P4 interactions are primarily hydrophobic in preferring C-terminal cleavage to Ala, we observed here a
nature [79], which might explain why we observed that Lys is rather unspecific cleavage pattern for both enzymes with
only poorly accepted in this position. The S3 subsite cannot be multiple cleavage products, in which the intended designed
2877

Beilstein J. Org. Chem. 2017, 13, 2869–2882.
Figure 10: Dimension of stabilization or destabilization upon TfIle or HfLeu incorporation compared to the non-fluorinated analogues containing Ile or
Leu, respectively, for all four different enzymes studied here and measured after 24 h of incubation.
cleavage site with Phe in P1 position wasn’t affected. However, Furthermore, our investigations show that fluorine’s impact on
we observed that the introduction of HfLeu has a general proteolytic stability needs to be investigated always case-by-
protective effect against degradation by elastase, whereas the case as there is no general trend to be concluded. Nevertheless,
effect of TfIle depends on the substitution position. Although the results of this current study provide valuable knowledge on
the introduction of fluorine substituents generally affected the how bulky fluorinated amino acids can help to increase the
rate of hydrolysis by proteinase K, only fluorination of an Ile proteolytic stability of peptides, and show that upon smart
residue N-terminal to the cleavage site effectively protected the design, these fluorinated amino acids can be used to engineer
peptide from digestion. Particularly noteworthy is the effect of peptide drug candidates.
fluorination of the Ile side chain in P2 position. The P2-TfIleFA
Experimental
peptide was the most resistant substrate towards proteolysis by
all four proteases applied in this study. In contrast, destabiliza- Materials
tion due to fluorination was only observed when TfIle and Fmoc-L-amino acids were purchased from ORPEGEN
HfLeu were incorporated into the P1’ position.
Peptide Chemicals GmbH (Heidelberg, Germany). Fmoc-
Lys(Boc)Wang resin was from Novabiochem (Merck Chemi-
In future studies, we will focus on a more precise characteriza- cals GmbH, Darmstadt, Germany). All solvents were used from
tion of the interaction of fluorinated substrates with proteolytic VWR (Darmstadt, Germany) without further purification. All
enzymes to which multiple factors contribute. The steric other chemicals were bought from Acros (Geel, Belgium), abcr
demand or conformation of the side chain, hydrophobicity, fluo- GmbH (Karlsruhe, Germany), fluorochem (Hadfield, United
rine induced polarity and significant pKa-value changes of Kingdom), VWR (Darmstadt, Germany) or Merck (Darmstadt,
neighboring groups [9,10] can lead to fluorine-specific interac- Germany) at highest commercially available purity and used as
tions between substrate and enzyme binding sites as well as to such. A detailed synthetic strategy for Fmoc-TfIle-OH is de-
an exclusion of the cleavage-relevant peptide bonds from the scribed in literature [44]. For the synthesis of Fmoc-HfLeu-OH
active site.
see Supporting Information File 1.
2878

Beilstein J. Org. Chem. 2017, 13, 2869–2882.
Peptide synthesis, purification and
characterization
powder. The purity of the peptides was controlled by analytical
HPLC (LUNATM C8 (2) column, 5 μm, 250 × 4.6 mm,
Peptides were synthesized manually in a 0.05 mmol scale Phenomenex®, Torrance, CA, USA), and the products were
on a solid support by means of an Fmoc/tert-butyl protecting identified by high-resolution ESI–ToF–MS (see Supporting
group strategy on a preloaded Fmoc-Lys(Boc)Wang resin Information File 1).
(0.57 mmol/g loading) using 10 mL polypropylene reactors.
HfLeu containing peptides were synthesized with an Activo- Protease digestion assay
P11 automated peptide synthesizer (Activotec, Cambridge, All peptides employed in the degradation studies were used as
United Kingdom). Couplings of non-fluorinated amino acids the TFA salts obtained after lyophilization. Stock solutions of
were performed in dimethylformamide (DMF) with the Fmoc- α-chymotrypsin (from bovine pancreas, EC 3.4.21.1,
L-amino acid, 1-hydroxybenzotriazole (HOBt) and N,N’-diiso- ≥40.0 units/mg of protein, Sigma Aldrich, St. Louis, MO,
carbodiimide (DIC) in an eight-fold excess with respect to the USA), and pepsin (from porcine stomach mucosa, EC 3.4.23.1,
resin amount. In order to ensure completion of the reaction the ≥250 units/mg of protein, Sigma Aldrich, St. Louis, MO, USA)
couplings were performed twice for 1 h each. The fluorinated were prepared at concentrations of 1 mg/mL in phosphate
amino acids and coupling reagents 1-hydroxy-7-azabenzotri- buffer (10 mM, pH 7.4), or in acetate buffer (10 mM, pH 4.0),
azole (HOAt)/DIC were used in 1.2-fold excess, and the cou- respectively. For proteinase K (from tritirachium album, EC
pling was carried out manually one time overnight. In the case 3.4.21.64, ≥30 units/mg of protein, Sigma Aldrich, St. Louis,
of an insufficient coupling, the coupling was repeated for 3 h MO, USA) and elastase (from porcine pancreas, EC 3.4.21.36,
with 0.5 equivalents. Prior to the Fmoc deprotection of the fluo- 6.2 units/mg of protein, Sigma Aldrich, St. Louis, MO, USA)
rinated amino acids, free N-termini were capped by adding a stock solutions were prepared also at concentrations of
mixture of acetic anhydride (Ac2O, 10% (v/v)) and N,N-diiso- 1 mg/mL in tris/HCl (50 mM) + CaCl2 (10 mM) buffer (pH
propylethylamine (DIPEA, 10% (v/v)) in DMF (3 × 10 min). 7.5), or in tris/HCl buffer (100 mM, pH 8.4), respectively.
Fmoc deprotection was achieved by treatment with 20% (v/v) Peptides (0.002 mmol) were prepared as stocks in DMSO
piperidine in DMF (3 × 10 min). All peptides were N-termi- (100 µL) and incubated with the respective enzyme at 30 °C
nally labeled with o-aminobenzoic acid (Abz) to enable photo- (for α-chymotrypsin and pepsin) or 37 °C (for proteinase K and
metric detection. The resin was washed between each step with elastase) with shaking at 300 rpm in a thermomixer over a
DMF and dicholoromethane (DCM, 3 × 2 mL each). After the period of 24 h. The reaction mixture consisted of DMSO
synthesis, the peptides were cleaved from the resin by treat- (15 µL), corresponding buffer (25 µL), peptide solution (5 µL)
ment with a solution (2 mL) containing triisopropylsilane (TIS, and the corresponding enzyme solution (5 µL). The concentra-
10% (v/v)), water (1% (v/v)), and trifluoroacetic acid (TFA) tion of enzyme was optimized so that the hydrolysis of the
(89% (v/v)) for 3 h. The resin was washed twice with TFA control peptide FA was about 40% after 120 min. Aliquots of
(1 mL) and DCM (1 mL) and excess solvent was removed by 5 µL were taken at fixed time points (0, 15, 30, 60, 90, 120 min
evaporation. The crude peptide was precipitated with ice-cold as well as 3 h and 24 h) and either quenched with ACN contain-
diethyl ether (80 mL), and after centrifugation dried by ing 0.1% (v/v) TFA (95 µL), in the case of α-chymotrypsin,
lyophilization. Purification of the synthesized peptides was per- proteinase K and elastase, or 2% aqueous ammonia (95 µL), in
formed on a LaPrepΣ low-pressure HPLC system (VWR, the case of pepsin. All samples were subjected to analytical
Darmstadt, Germany) using a Kinetex RP-C18 endcapped HPLC on a LaChrom-ELITE-HPLC-System equipped with a
HPLC column (5 µM, 100 Å, 250 × 21.2 mm, Phenomenex®, fluorescence detector (VWR International Hitachi, Darmstadt,
USA). A Security GuardTM PREP Cartridge Holder Kit Germany). A monolithic reversed-phase C8 Chromolith® Per-
(21.20 mm, ID, Phenomenex®, USA) served as pre-column. As formance HPLC column (100 × 4.6 mm, Merck KGaA, Darm-
eluents deionized water (Milli-Q Advantage® A10 Ultrapure stadt, Germany) was used to resolve and quantify the products
Water Purification System, Millipore®, Billerica, MA, USA) of digestion. The used system and gradients are described in
and acetonitrile (ACN), both containing 0.1% (v/v) TFA were detail in Supporting Information File 1. Detection based on the
used. HPLC runs were performed starting with an isocratic Abz label was carried out using a fluorescence detector with
gradient of 5% ACN over 5 min, flow rate: 10 mL/min, contin- λex = 320 nm and λem = 420 nm. In all cases, the peaks corre-
uing with a linear gradient of 5–70% ACN over 25 min, flow sponding to the starting materials (full-length peptides) or the
rate: 20.0 mL/min. UV-detection occurred at 220 nm. Data N-terminal fragments (products) were integrated and used to de-
analysis was performed with an EZChrom Elite-Software termine the velocity of the reaction (see Supporting Informa-
(Version 3.3.2 SP2, Agilent Technologies, Santa Clara, CA, tion File 1). The FA peptide was used as a reference. Each frag-
USA). The fractions containing pure peptide were combined, ment cleaved from the full-length peptide was identified by
reduced in vacuo and lyophilized to give the peptides as a white ESI–ToF mass analysis on an Agilent 6220 ESI–ToF–MS spec-
2879

Beilstein J. Org. Chem. 2017, 13, 2869–2882.
14.Vukelić, S.; Moschner, J.; Huhmann, S.; Fernandes, R.; Berger, A. A.;
Koksch, B. Synthesis of Side Chain Fluorinated Amino Acids and Their
Effects on the Properties of Peptides and Proteins. In Modern
Synthesis Processes and Reactivity of Fluorinated Compounds;
Leroux, F. R.; Tressaud, A., Eds.; Elsevier, 2017; pp 427–464.
doi:10.1016/B978-0-12-803740-9.00015-9
trometer (Agilent Technologies, Santa Clara, CA, USA, see
Supporting Information File 1). All experiments were per-
formed in triplicate.
Supporting Information
15.Buer, B. C.; Meagher, J. L.; Stuckey, J. A.; Marsh, E. N. G. Protein Sci.
2012, 21, 1705–1715. doi:10.1002/pro.2150
Supporting Information File 1
16.Buer, B. C.; Meagher, J. L.; Stuckey, J. A.; Marsh, E. N. G.
Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 4810–4815.
doi:10.1073/pnas.1120112109
Characterization and identification of synthesized peptides,
characterization of the enzymatic digestion reactions, and
identification of proteolytic cleavage products, HPLC
methods, and synthesis protocol for Fmoc-HfLeu-OH.
[http://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-13-279-S1.pdf]
17.Jäckel, C.; Salwiczek, M.; Koksch, B. Angew. Chem., Int. Ed. 2006, 45,
4198–4203. doi:10.1002/anie.200504387
18.Tang, Y.; Tirrell, D. A. J. Am. Chem. Soc. 2001, 123, 11089–11090.
doi:10.1021/ja016652k
19.Tang, Y.; Ghirlanda, G.; Petka, W. A.; Nakajima, T.; DeGrado, W. F.;
Tirrell, D. A. Angew. Chem., Int. Ed. 2001, 40, 1494–1496.
doi:10.1002/1521-3773(20010417)40:8<1494::AID-ANIE1494>3.0.CO;
2-X
Acknowledgements
20.Tang, Y.; Ghirlanda, G.; Vaidehi, N.; Kua, J.; Mainz, D. T.;
Goddard, W. A., III; DeGrado, W. F.; Tirrell, D. A. Biochemistry 2001,
40, 2790–2796. doi:10.1021/bi0022588
This work has been generously supported by the DFG in the
context of the Research Training Group 1582 “Fluorine as Key
Element”. Special thanks go to Dr. Holger Erdbrink and Prof.
Dr. Constantin Czekelius for providing the fluorinated amino
acid Boc-TfIle-OH. The authors thank Dr. Allison Ann Berger
for carefully proofreading the manuscript.
21.Yoder, N. C.; Kumar, K. Chem. Soc. Rev. 2002, 31, 335–341.
doi:10.1039/b201097f
22.Bilgiçer, B.; Xing, X.; Kumar, K. J. Am. Chem. Soc. 2001, 123,
11815–11816. doi:10.1021/ja016767o
23.Meng, H.; Kumar, K. J. Am. Chem. Soc. 2007, 129, 15615–15622.
doi:10.1021/ja075373f
References
24.Deacon, C. F.; Nauck, M. A.; Toft-Nielsen, M.; Pridal, L.; Willms, B.;
Holst, J. J. Diabetes 1995, 44, 1126–1131. doi:10.2337/diab.44.9.1126
25.Holst, J. J.; Deacon, C. F. Curr. Opin. Pharmacol. 2004, 4, 589–596.
doi:10.1016/j.coph.2004.08.005
1. Vlieghe, P.; Lisowski, V.; Martinez, J.; Khrestchatisky, M.
Drug Discovery Today 2010, 15, 40–56.
doi:10.1016/j.drudis.2009.10.009
2. Sato, A. K.; Viswanathan, M.; Kent, R. B.; Wood, C. R.
Curr. Opin. Biotechnol. 2006, 17, 638–642.
26.Drucker, D. J. Curr. Pharm. Des. 2001, 7, 1399–1412.
doi:10.2174/1381612013397401
doi:10.1016/j.copbio.2006.10.002
27.Meng, H.; Krishnaji, S. T.; Beinborn, M.; Kumar, K. J. Med. Chem.
2008, 51, 7303–7307. doi:10.1021/jm8008579
3. Albericio, F.; Kruger, H. G. Future Med. Chem. 2012, 4, 1527–1531.
doi:10.4155/fmc.12.94
28.Akçay, G.; Kumar, K. J. Fluorine Chem. 2009, 130, 1178–1182.
doi:10.1016/j.jfluchem.2009.09.002
4. Uhlig, T.; Kyprianou, T.; Martinelli, F. G.; Oppici, C. A.; Heiligers, D.;
Hills, D.; Calvo, X. R.; Verhaert, P. EuPa Open Proteomics 2014, 4,
58–69. doi:10.1016/j.euprot.2014.05.003
29.Gottler, L. M.; Lee, H.-Y.; Shelburne, C. E.; Ramamoorthy, A.;
Marsh, E. N. G. ChemBioChem 2008, 9, 370–373.
doi:10.1002/cbic.200700643
5. Fosgerau, K.; Hoffmann, T. Drug Discovery Today 2015, 20, 122–128.
doi:10.1016/j.drudis.2014.10.003
30.Voloshchuk, N.; Zhu, A. Y.; Snydacker, D.; Montclare, J. K.
Bioorg. Med. Chem. Lett. 2009, 19, 5449–5451.
doi:10.1016/j.bmcl.2009.07.093
6. Santos, G. B.; Ganesan, A.; Emery, F. S. ChemMedChem 2016, 11,
2245–2251. doi:10.1002/cmdc.201600288
7. Jäckel, C.; Koksch, B. Eur. J. Org. Chem. 2005, 4483–4503.
doi:10.1002/ejoc.200500205
31.Budisa, N.; Wenger, W.; Wiltschi, B. Mol. BioSyst. 2010, 6, 1630–1639.
doi:10.1039/c002256j
8. Latham, P. W. Nat. Biotechnol. 1999, 17, 755–757. doi:10.1038/11686
9. Salwiczek, M.; Nyakatura, E. K.; Gerling, U. I. M.; Ye, S.; Koksch, B.
Chem. Soc. Rev. 2012, 41, 2135–2171. doi:10.1039/C1CS15241F
10.Berger, A. A.; Völler, J.-S.; Budisa, N.; Koksch, B. Acc. Chem. Res.
2017, 50, 2093–2103. doi:10.1021/acs.accounts.7b00226
11.Salwiczek, M.; Samsonov, S.; Vagt, T.; Nyakatura, E.; Fleige, E.;
Numata, J.; Cölfen, H.; Pisabarro, M. T.; Koksch, B. Chem. – Eur. J.
2009, 15, 7628–7636. doi:10.1002/chem.200802136
12.Huhmann, S.; Nyakatura, E. K.; Erdbrink, H.; Gerling, U. I. M.;
Czekelius, C.; Koksch, B. J. Fluorine Chem. 2015, 175, 32–35.
doi:10.1016/j.jfluchem.2015.03.003
32.Bordusa, F.; Dahl, C.; Jakubke, H.-D.; Burger, K.; Koksch, B.
Tetrahedron: Asymmetry 1999, 10, 307–313.
doi:10.1016/S0957-4166(98)00508-4
33.Koksch, B.; Sewald, N.; Burger, K.; Jakubke, H.-D. Amino Acids 1996,
11, 425–434. doi:10.1007/BF00807946
34.Koksch, B.; Sewald, N.; Hofmann, H.-J.; Burger, K.; Jakubke, H.-D.
J. Pept. Sci. 1997, 3, 157–167.
doi:10.1002/(SICI)1099-1387(199705)3:3<157::AID-PSC94>3.0.CO;2-
W
35.Smits, R.; Koksch, B. Curr. Top. Med. Chem. 2006, 6, 1483–1498.
doi:10.2174/156802606777951055
13.Marsh, E. N. G. Designing Fluorinated Proteins. In Methods in
Enzymology; Vincent, L. P., Ed.; Academic Press, 2016; Vol. 580,
pp 251–278. doi:10.1016/bs.mie.2016.05.006
36.Thust, S.; Koksch, B. J. Org. Chem. 2003, 68, 2290–2296.
doi:10.1021/jo020613p
2880

Beilstein J. Org. Chem. 2017, 13, 2869–2882.
37.Thust, S.; Koksch, B. Tetrahedron Lett. 2004, 45, 1163–1165.
doi:10.1016/j.tetlet.2003.12.007
61.Brady, K.; Abeles, R. H. Biochemistry 1990, 29, 7608–7617.
doi:10.1021/bi00485a010
38.Sewald, N.; Hollweck, W.; Mütze, K.; Schierlinger, C.; Seymour, L. C.;
Gaa, K.; Burger, K.; Koksch, B.; Jakubke, H. D. Amino Acids 1995, 8,
187–194. doi:10.1007/BF00806491
62.Schellenberger, V.; Jakubke, H.-D.
Biochim. Biophys. Acta, Protein Struct. Mol. Enzymol. 1986, 869,
54–60. doi:10.1016/0167-4838(86)90309-2
39.Asante, V.; Mortier, J.; Schlüter, H.; Koksch, B. Bioorg. Med. Chem.
2013, 21, 3542–3546. doi:10.1016/j.bmc.2013.03.051
40.Asante, V.; Mortier, J.; Wolber, G.; Koksch, B. Amino Acids 2014, 46,
2733–2744. doi:10.1007/s00726-014-1819-7
63.Schellenberger, V.; Turck, C. W.; Rutter, W. J. Biochemistry 1994, 33,
4251–4257. doi:10.1021/bi00180a020
64.Ye, S.; Loll, B.; Berger, A. A.; Mülow, U.; Alings, C.; Wahl, M. C.;
Koksch, B. Chem. Sci. 2015, 6, 5246–5254. doi:10.1039/C4SC03227F
65.Antal, J.; Pál, G.; Asbóth, B.; Buzás, Z.; Patthy, A.; Gráf, L.
Anal. Biochem. 2001, 288, 156–167. doi:10.1006/abio.2000.4886
66.Sampath-Kumar, P. S.; Fruton, J. S. Proc. Natl. Acad. Sci. U. S. A.
1974, 71, 1070–1072. doi:10.1073/pnas.71.4.1070
67.Fruton, J. S. Acc. Chem. Res. 1974, 7, 241–246.
doi:10.1021/ar50080a001
41.Schechter, I.; Berger, A. Biochem. Biophys. Res. Commun. 1967, 27,
157–162. doi:10.1016/S0006-291X(67)80055-X
42.Davies, D. R. Annu. Rev. Biophys. Biophys. Chem. 1990, 19, 189–215.
doi:10.1146/annurev.bb.19.060190.001201
43.Zhao, Y. H.; Abraham, M. H.; Zissimos, A. M. J. Org. Chem. 2003, 68,
7368–7373. doi:10.1021/jo034808o
44.Erdbrink, H.; Nyakatura, E. K.; Huhmann, S.; Gerling, U. I. M.;
Lentz, D.; Koksch, B.; Czekelius, C. Beilstein J. Org. Chem. 2013, 9,
2009–2014. doi:10.3762/bjoc.9.236
68.Hamuro, Y.; Coales, S. J.; Molnar, K. S.; Tuske, S. J.; Morrow, J. A.
Rapid Commun. Mass Spectrom. 2008, 22, 1041–1046.
doi:10.1002/rcm.3467
45.Samsonov, S. A.; Salwiczek, M.; Anders, G.; Koksch, B.;
Pisabarro, M. T. J. Phys. Chem. B 2009, 113, 16400–16408.
doi:10.1021/jp906402b
69.Rao, C.; Dunn, B. M. Evidence for Electrostatic Interactions in the S2
Subsite of Porcine Pepsin. In Aspartic Proteinases: Structure, Function,
Biology, and Biomedical Implications; Takahashi, K., Ed.; Springer US:
Boston, MA, 1995; pp 91–94. doi:10.1007/978-1-4615-1871-6_10
70.Fujinaga, M.; Chernaia, M. M.; Mosimann, S. C.; James, M. N. G.;
Tarasova, N. I. Protein Sci. 1995, 4, 960–972.
46.Blow, D. M.; Birktoft, J. J.; Hartley, B. S. Nature 1969, 221, 337–340.
doi:10.1038/221337a0
47.Blow, D. M. The Structure of Chymotrypsin. In The Enzymes;
Paul, D. B., Ed.; Academic Press, 1971; Vol. 3, pp 185–212.
doi:10.1016/S1874-6047(08)60397-2
doi:10.1002/pro.5560040516
71.Kageyama, T. Cell. Mol. Life Sci. 2002, 59, 288–306.
doi:10.1007/s00018-002-8423-9
48.Derewenda, Z. S.; Derewenda, U.; Kobos, P. M. J. Mol. Biol. 1994,
241, 83–93. doi:10.1006/jmbi.1994.1475
72.Dunn, B. M.; Hung, S.-H.
49.Polgár, L. Cell. Mol. Life Sci. 2005, 62, 2161–2172.
doi:10.1007/s00018-005-5160-x
Biochim. Biophys. Acta, Protein Struct. Mol. Enzymol. 2000, 1477,
231–240. doi:10.1016/S0167-4838(99)00275-7
50.Fruton, J. S. The Specificity and Mechanism of Pepsin Action. In
Advances in Enzymology and Related Areas of Molecular Biology;
Nord, F. F., Ed.; John Wiley & Sons, Inc., 1970; Vol. 33, pp 401–443.
doi:10.1002/9780470122785.ch9
73.Naughton, M. A.; Sanger, F. Biochem. J. 1961, 78, 156–163.
doi:10.1042/bj0780156
74.Renaud, A.; Lestienne, P.; Hughes, D. L.; Bieth, J. G.; Dimicoli, J. L.
J. Biol. Chem. 1983, 258, 8312–8316.
51.Fruton, J. S. Pepsin. In The Enzymes; Paul, D. B., Ed.; Academic
Press, 1971; Vol. 3, pp 119–164. doi:10.1016/S1874-6047(08)60395-9
52.Dunn, B. M. Chem. Rev. 2002, 102, 4431–4458.
doi:10.1021/cr010167q
75.Qasim, M. A. Protein Pept. Lett. 2014, 21, 164–170.
doi:10.2174/09298665113206660093
76.Atlas, D. J. Mol. Biol. 1975, 93, 39–48.
doi:10.1016/0022-2836(75)90358-7
53.Antonov, V. K. New Data on Pepsin Mechanism and Specificity. In Acid
Proteases:Structure, Function, and Biology; Tang, J., Ed.; Springer US:
Boston, MA, 1977; pp 179–198. doi:10.1007/978-1-4757-0719-9_11
54.Powers, J. C.; Harley, A. D.; Myers, D. V. Subsite Specificity of Porcine
Pepsin. In Acid Proteases:Structure, Function, and Biology; Tang, J.,
Ed.; Springer US: Boston, MA, 1977; Vol. 92, pp 141–157.
doi:10.1007/978-1-4757-0719-9_9
77.Atlas, D.; Levit, S.; Schechter, I.; Berger, A. FEBS Lett. 1970, 11,
281–283. doi:10.1016/0014-5793(70)80548-8
78.Ebeling, W.; Hennrich, N.; Klockow, M.; Metz, H.; Orth, H. D.; Lang, H.
Eur. J. Biochem. 1974, 47, 91–97.
doi:10.1111/j.1432-1033.1974.tb03671.x
79.Georgieva, D.; Genov, N.; Voelter, W.; Betzel, C.
Z. Naturforsch., C: J. Biosci. 2006, 61, 445–452.
55.Largman, C.; Brodrick, J. W.; Geokas, M. C. Biochemistry 1976, 15,
2491–2500. doi:10.1021/bi00656a036
doi:10.1515/znc-2006-5-623
80.Morihara, K.; Tsuzuki, H. Agric. Biol. Chem. 1975, 39, 1489–1492.
doi:10.1080/00021369.1975.10861790
56.Shotton, D. M. Elastase. Methods in Enzymology; Academic Press,
1970; Vol. 19, pp 113–140. doi:10.1016/0076-6879(70)19009-4
57.Sweeney, P. J.; Walker, J. M. Proteinase K (EC 3.4.21.14). In
Enzymes of Molecular Biology; Burrell, M. M., Ed.; Humana Press:
Totowa, NJ, 1993; pp 305–311. doi:10.1385/0-89603-234-5:305
58.Czapinska, H.; Otlewski, J. Eur. J. Biochem. 1999, 260, 571–595.
doi:10.1046/j.1432-1327.1999.00160.x
81.Brömme, D.; Peters, K.; Fink, S.; Fittkau, S. Arch. Biochem. Biophys.
1986, 244, 439–446. doi:10.1016/0003-9861(86)90611-9
82.Kraus, E.; Femfert, U. Hoppe-Seyler's Z. Physiol. Chem. 1976, 357,
937–947. doi:10.1515/bchm2.1976.357.2.937
83.Betzel, C.; Singh, T. P.; Visanji, M.; Peters, K.; Fittkau, S.;
Saenger, W.; Wilson, K. S. J. Biol. Chem. 1993, 268, 15854–15858.
84.Wolf, W. M.; Bajorath, J.; Müller, A.; Raghunathan, S.; Singh, T. P.;
Hinrichs, W.; Saenger, W. J. Biol. Chem. 1991, 266, 17695–17699.
85.Betzel, C.; Bellemann, M.; Pal, G. P.; Bajorath, J.; Saenger, W.;
Wilson, K. S. Proteins: Struct., Funct., Genet. 1988, 4, 157–164.
doi:10.1002/prot.340040302
59.Hedstrom, L. Chem. Rev. 2002, 102, 4501–4524.
doi:10.1021/cr000033x
60.Sweeney, P. J.; Walker, J. M. Proteolytic Enzymes for Peptide
Production. In Enzymes of Molecular Biology; Burrell, M. M., Ed.;
Humana Press: Totowa, NJ, 1993; pp 277–303.
doi:10.1385/0-89603-234-5:277
2881

Beilstein J. Org. Chem. 2017, 13, 2869–2882.
86.Kraus, E.; Kiltz, H.-H.; Femfert, U. F. Hoppe-Seyler's Z. Physiol. Chem.
1976, 357, 233–237. doi:10.1515/bchm2.1976.357.1.233
License and Terms
This is an Open Access article under the terms of the
Creative Commons Attribution License
(http://creativecommons.org/licenses/by/4.0), which
permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
(http://www.beilstein-journals.org/bjoc)
The definitive version of this article is the electronic one
which can be found at:
doi:10.3762/bjoc.13.279
2882