Synthesis of C-Mannosylated Glycopeptides Enabled by Ni-Catalyzed Photoreductive Cross-Coupling Reactions

Article abstract of DOI:10.1021/jacs.1c05567

The biological functions of tryptophan C-mannosylation are poorly understood, in part, due to a dearth of methods for preparing pure glycopeptides and glycoproteins with this modification. To address this issue, efficient and scalable methods are required for installing this protein modification. Here, we describe unique Ni-catalyzed cross-coupling conditions that utilize photocatalysis or a Hantzsch ester photoreductant to couple glycosyl halides with (hetero)aryl bromides, thereby enabling the α-C-mannosylation of 2-bromo-tryptophan, peptides thereof, and (hetero)aryl bromides more generally. We also report that 2-(α-d-mannopyranosyl)-L-tryptophan undergoes facile anomerization in the presence of acid: something that must be considered when preparing and handling peptides with this modification. These developments enabled the first automated solid-phase peptide syntheses of C-mannosylated glycopeptides, which we used to map the epitope of an antibody, as well as providing the first verified synthesis of Carmo-HrTH-I, a C-mannosylated insect hormone. To complement this approach, we also performed late-stage tryptophan C-mannosylation on a diverse array of peptides, demonstrating the broad scope and utility of this methodology for preparing glycopeptides.

Full text of DOI:10.1021/jacs.1c05567

pubs.acs.org/JACS
Article
Synthesis of C‑Mannosylated Glycopeptides Enabled by Ni-
Catalyzed Photoreductive Cross-Coupling Reactions
Runyu Mao, Shiyi Xi, Sayali Shah, Michael J. Roy, Alan John, James P. Lingford, Gerd Gäde,
Nichollas E. Scott, and Ethan D. Goddard-Borger*
Cite This: https://doi.org/10.1021/jacs.1c05567
Read Online
ACCESS
Metrics & More
Article Recommendations
*
sı Supporting Information
ABSTRACT: The biological functions of tryptophan C-manno-
sylation are poorly understood, in part, due to a dearth of methods
for preparing pure glycopeptides and glycoproteins with this
modification. To address this issue, efficient and scalable methods
are required for installing this protein modification. Here, we
describe unique Ni-catalyzed cross-coupling conditions that utilize
photocatalysis or a Hantzsch ester photoreductant to couple
glycosyl halides with (hetero)aryl bromides, thereby enabling the
α-C-mannosylation of 2-bromo-tryptophan, peptides thereof, and
(
hetero)aryl bromides more generally. We also report that 2-(α-D-
mannopyranosyl)-L-tryptophan undergoes facile anomerization in
the presence of acid: something that must be considered when preparing and handling peptides with this modification. These
developments enabled the first automated solid-phase peptide syntheses of C-mannosylated glycopeptides, which we used to map
the epitope of an antibody, as well as providing the first verified synthesis of Carmo-HrTH-I, a C-mannosylated insect hormone. To
complement this approach, we also performed late-stage tryptophan C-mannosylation on a diverse array of peptides, demonstrating
the broad scope and utility of this methodology for preparing glycopeptides.
INTRODUCTION
A number of approaches to the synthesis of Trp(Man) 1
■
15−19
have been described (Scheme 1A).
Isobe and co-workers
Tryptophan C-mannosylation is the only known form of
protein C-glycosylation. It involves the formation of a C−C
bond between C-2 of L-tryptophan and the anomeric position
of D-mannopyranose to give 2-(α-D-mannopyranosyl)-L-
performed a Larock synthesis to obtain α-D-mannosyl indole 2
and a regio- and stereoselective alkylation with aziridine 3 to
15,16
access 1 (Scheme 1A-i),
while Ito and co-workers accessed
1
,2
1
via a stereocontrolled ring opening of epoxide 4 with
tryptophan or Trp(Man) 1. This C-mannoside adopts the
17,18
1
4
unusual C conformation rather than the C conformer
lithiated tryptophanol derivative 5 (Scheme 1A-ii).
These
4
1
3
−6
preferred by most other mannosides.
found on a diverse array of proteins, including those with a
The modification is
strategies are too long and low-yielding to provide the gram
quantities of 1 required for automated SPPS. A more recent
approach by Chen and co-workers used an isoquinoline-1-
carboxamide auxiliary on tryptophan 6 to direct a Pd-mediated
C−H activation and glycosylation at C-2 using glycosyl
7
thrombospondin repeat (TSR) domain, type-I cytokine
8
1,2
9
receptors, RNases, lipoprotein lyase, glycoside hydrolases,
5
10
11
glycosyltransferases, siglecs, and viral glycoproteins, yet its
19
biological functions remain poorly characterized.
Nonetheless, recent studies have demonstrated that Trp-
Man) appears to play a role in stabilizing protein tertiary
chloride 7 (Scheme 1A-iii). The limitation of this approach
is that the isoquinoline directing group requires strong acid for
removal and, as reported here for the first time, this results in
anomerization of 1 to the β-anomer 1β (Scheme 1A). The
positioning of this auxiliary group also precludes this strategy
from being useful for the late-stage modification of peptides.
(
12
structure, as demonstrated for a TSR domain, a type-I
5
5
cytokine receptor and RNase2. However, difficulties in
obtaining homogeneous glycoforms precludes a quantitative
examination of the impact of Trp(Man) on the thermody-
namics of protein and peptide folding. Progress in this field
13,14
Received: May 30, 2021
requires synthetic methods to access pure glycoforms.
The
main barriers to realizing this goal are the production of
Trp(Man) in sufficient quantities to facilitate automated solid
phase peptide synthesis (SPPS) and the absence of methods
for installing the modification at a late stage in peptide
synthesis.
©
XXXX American Chemical Society
https://doi.org/10.1021/jacs.1c05567
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
A

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Scheme 1. Synthetic Routes to Trp(Man) 1 and C-
Glycosides
RESULTS AND DISCUSSION
■
Our initial efforts to develop a scalable route to 1 were focused
on finding a dual Ni/photoredox catalytic reductive coupling
condition using a tertiary amine as both base and terminal
30,31
reductant.
This approach made use of the commercially
available mannosyl bromide 8 and 2-bromo-L-tryptophan 9,
which is accessible in one step from commercially available
materials. Initial experiments were not promising (Table 1,
Table 1. Optimization of Cross-Coupling Reaction
Conditions
e
a
entry
PC
reductant
base
LED
yield
b,c
1
2
3
4
5
6
7
8
9
[Ir]
[Ir]
[Ir]
[Ir]
[Ir]
−
−
−
−
HE
HE
HE
−
HE
HE
Et N
blue
blue
blue
blue
blue
blue
blue
−
0
3
b,
c
DIPEA
DIPEA
DIPEA
DIPEA
DIPEA
DIPEA
DIPEA
DIPEA
trace
c
21%
55%
65%
42%
c
,
d
d
d
d
−
−
−
0
d
0
violet
78%
b
a
0
.15 mmol scale, HPLC yield with an internal standard. without
MgCl2. 3 eq base. 48 h reaction time. [Ir]: 1.5 mol %
c
d
e
[
Ir{dF(CF )ppy} (dtbbpy)]PF .
3
2
6
entry 1, 2); however, the addition of stoichiometric amounts of
MgCl , which has been shown to enhance cross-coupling
2
31−34
reactions involving glycosyl halides,
afforded 10 in 21%
yield (Table 1, entry 3). Experiments utilizing other terminal
reductants (Table S1) identified Hantzsch ester (HE) as the
best additive (Table 1, entry 4) and with further optimization
the yield was increased to 65% (Table 1, entry 5).
In recent times, Ni/photoredox dual catalysis has shown
promise for the assembly of structurally diverse C-glycosides,
though the availability of suitable radical precursors has limited
20−23
the utility of these techniques (Scheme 1B).
However, α-
At this point in the optimization process, it became apparent
that the control reaction lacking PC also provided product 10,
albeit with a lower yield of 42% (Table 1, entry 6), while
controls without HE or irradiation (Table 1, entries 7, 8)
provided no product. We surmised from this that photoexcited
HE was sufficiently reducing to generate the low-valence nickel
species and glycosyl radical, directly or indirectly, needed to
configured mannopyranosyl radicals, which are relatively stable
and can be trapped to provide α-C-mannosides with excellent
2
4−26
stereoselectivity,
have been generated in other contexts
from simple glycosyl halides using transition metal-based
27−29
photocatalysts (PCs).
Thus, we reasoned that a Ni/
photoredox dual catalysis strategy could be used to reductively
couple mannosyl halides and haloarenes to access Trp(Man) 1,
polypeptides thereof, and other α-C-mannosides.
35−38
initiate the catalytic cross-coupling cycle (Scheme S2).
Since HE has much stronger absorption in the near-UV region
39
In pursuit of such conditions, we developed a PC-free, Ni-
catalyzed reductive cross-coupling reaction that utilizes
Hantzsch ester (HE) as a photoreductant (Scheme 1C).
This methodology provides gram quantities of 1, can be used
for the late-stage modification of simple peptides bearing a 2-
bromo-tryptophan residue, and enables the synthesis of diverse
α-C-mannosides. During our work with Trp(Man) 1 and its
peptides, we discovered that this C-glycoside undergoes facile
anomerization in the presence of strong acids. Accounting for
this issue in the design of SPPS protecting group strategies
enabled the automated synthesis of biologically relevant
glycopeptides using Fmoc-based methods. To complement
this approach, we also established a Ni/photoredox dual
catalysis methodology that operates in polar aprotic solvents
and tolerates water to facilitate the late-stage C-mannosylation
of larger and more complex peptides.
(350−400 nm) than blue-light region, we repeated the
reaction without PC using violet LED irradiation (peak
emission at 390 nm) to obtain product 10 in 78% yield
(Table 1, entry 9) using a shorter reaction time than for
previous examples. Other organic bases, solvents, nickel
precatalysts, ligands, and HE derivatives failed to improve on
this result (Table S1).
This condition is reminiscent of a recent method reported
by Molander and co-workers, whereby the Ni-catalyzed cross-
coupling of aryl halides and hydroxyphthalimide esters was
40
facilitated by a HE photoreductant. It is also congruent with
recent reports of 1,4-dihydropyridines mediating various
photochemical reactions and of 1,4-dihydropyridine derivatives
serving as photoactivatable substrates for Ni-catalyzed cross-
37−39,41−43
coupling reactions.
The inexpensive nature of HE,
the ability to dispense with costly PCs, and independence from
B
https://doi.org/10.1021/jacs.1c05567
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
preferred conformation from 4C1 to C , though the
1
auxiliary directing groups makes this an attractive methodology
4
2
3
for the construction of α-C-mannosides and C(sp )-C(sp )
bonds more generally. Furthermore, assembling Trp(Man) 1
in this way might plausibly be extended to performing the late-
stage C-glycosylation of peptides.
mechanism underlying this change is not immediately obvious.
To complete the synthesis of Trp(Man) 1, we saponified the
protected cross-coupling product 10. Following acidification
with hydrochloric acid, we observed by HPLC that the product
was now a mixture of two isomers. Repeating this experiment
with more careful neutralization and purification under neutral
conditions provided Trp(Man) 1, suggesting that the C-
glycoside may be acid sensitive. Trp(Man) 1 was readily
converted to the Fmoc-protected building block 13 for SPPS,
however, before proceeding we sought to better understand
the acid sensitivity of these compounds, since SPPS commonly
employs strong acids, such as trifluoroacetic acid (TFA), for
the removal of protecting groups.
To explore these ideas further, we tried these reaction
conditions on a suite of halogenated (hetero)arenes 11a−f and
peptides 11g−i to generate the α-C-mannosides 12a−i
(Scheme 2). Both electron-rich and -deficient systems were
Scheme 2. α-C-Mannosylation of Diverse Aryl Bromides
We started by treating the protected Trp(Man) derivatives
0 and 13 with various acids and solvents commonly used in
1
SPPS (Table 2). For both compounds, TFA in dichloro-
Table 2. Acid Sensitivity of Trp(Man) Derivatives
entry
substrate
% acid (v/v)
1% TFA
2% TFA
5% TFA
10% TFA
20% TFA
30% TFA
α:β
1
2
3
4
5
6
7
8
9
10
10
10
10
10
10
10
10
13
13
13
13
99:1
19:1
17:3
3:2
3:7
1:4
1:0
19:1
1:0
1:0
1:0
1% HCO H
2
5% HCO H
2
50% HFIP
100% HFIP
100% AcOH
20% AcOH + 30% HFIP
1
1
1
0
1
2
effective substrates in the cross-coupling reaction, with indoles,
pyridines, quinolines, and arene substrates providing product
in good to moderate yields. The only exception we
encountered was for thiazole 11f, which decomposed under
the reaction conditions and provided only a trace amount of
product 12f. As might be anticipated, the reaction was tolerant
of ester, amide, alcohol, nitrile, benzylic and aryl fluoride
functional groups.
The conversion of peptides 11g−i to glycopeptides 12g−i is
particularly noteworthy, as it demonstrated that this approach
can facilitate the late-stage C-glycosylation of peptides in
moderate to good yields. The ability to perform C-
glycosylation directly on peptides distinguishes this approach
from those previously reported (Scheme 1A). However, it
must be noted that the solubility of a given peptide substrate in
acetonitrile limited the scope of this approach: a problem we
would later solve by reoptimization of our cross-coupling
conditions to operate in polar aprotic solvents and tolerate the
addition of water.
1:0
methane was sufficient to produce significant amounts of the
undesired isomer (Table 2, entries 1−6), which was isolated
and identified as beta anomers 10β and 13β, respectively.
While 1% formic acid did not result in detectable levels of
anomerization, 5% formic acid did (Table 2, entries 7, 8).
Hexafluoroisopropanol (HFIP) and acetic acid, alone or in
combination, did not promote anomerization of 10 and 13
(Table 2, entries 9−12). Interestingly, the protecting groups
on the pyranose or amino acid made no significant difference
to the extent of anomerization. Further experimentation
revealed that solvents with basic oxygens, such as methanol
and water, were effective at suppressing anomerization,
however, these also suppress acid-mediated deprotection of
peptides (Table S2). Collectively, these experiments provided
a framework for understanding the resilience of Trp(Man) and
what limitations this places on SPPS strategies.
The conformation of the various α-C-mannosides prepared
The acid-catalyzed anomerization of 10 was performed with
1
in Scheme 2 is also worthy of comment. The C-mannosides
20% CF CO D in CDCl and the reaction monitored by H
3
2
3
4
1
2a−f all adopted the typical C conformation, while the
NMR (Figure S3). Integration of unique resonances enabled
quantitation of the relative abundance of 10 and 10β and
demonstration that the reaction proceeded with pseudo first-
order reaction kinetics (Figure 1A,B). No deuterium was
1
1
pyranose of the glycopeptides 12g−i all preferred the C
4
conformation. This provides further evidence that tryptophan
plays a defining role in driving a shift in the pyranose’s
C
https://doi.org/10.1021/jacs.1c05567
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
To understand how problematic this acid sensitivity was for
SPPS, we dissolved the protected C-mannosylated tripeptide
1
2i in 1:1 TFA/dichloromethane and monitored the sample by
HPLC over time. Within 2 h, the sample had completely
converted to an isomer with an identical mass but different
retention time: presumably the β-anomer (Figure 1D). To the
best of our knowledge, these data constitute the first evidence
of acid-mediated Trp(Man) anomerization. This finding is
relevant both to the synthesis of these peptides and the analysis
of biological samples: prolonged TFA exposure will result in
the anomerization of Trp(Man).
Having established the limits of Trp(Man) tolerance toward
acid, we explored the incorporation of this glycoamino acid
into peptides using automated SPPS. Our initial targets were
simple di- and tripeptides 14a−m (Figure 2), which would be
used to probe the epitope preferences of the 5G12 monoclonal
5
antibody (mAb). This mAb tool was recently reported as
being useful for the enrichment of C-mannosylated glycopep-
tides from biological samples to enable studies of the C-
5
glycome. Structural data for this mAb have been reported
(
PDB ID: 6PLH) and demonstrate that most of the mAb-
glycopeptide interactions occur through Trp(Man), with the
exception of an adjacent Glu: the side chain of this residue
makes intimate hydrogen bonds with the mAb (Figure 2A).
Understanding the limitations and biases of this antibody will
provide important context to the glycoproteomic data
generated using this first-generation tool.
When assembling C-mannosylated peptides 14a−m to
probe the binding preferences of this mAb (Figure 2B, Scheme
S3), we avoided the use of side chain protecting groups
wherever possible to obviate TFA deprotection steps that
would induce anomerization of the C-mannoside. This was not
possible for peptides containing Glu or Asp, and so benzyl
ester side chain protection was used instead, with hydro-
genolysis used to effect deprotection. All peptides were
assembled using automated SPPS on Sieber amide resin
(Scheme S3) and the N-terminal Fmoc protecting group left in
place to increase the molecular weight of the peptide and thus
enhance the sensitivity of subsequent surface plasmon
resonance (SPR) experiments.
Figure 1. Acid sensitivity of Trp(Man) and its derivatives. (a) The
relative abundance of 10 and 10β over time in the presence of 20%
(
v/v) CF CO D in CDCl . (b) Pseudo-first order reaction kinetics
3 2 3
are observed for this reaction. (c) Mechanistic rationalization of the
anomerization reaction. (d) Stacked HPLC traces for the TFA-
mediated anomerization of peptide 12i.
incorporated into the analyte during the course of this reaction,
1
as indicated by H NMR and LC-MS, suggesting that
anomerization most likely occurs as a result of protonation
of the endocyclic pyranose oxygen and formation of a
stabilized acyclic benzylic cation, rather than acid-catalyzed
formation of an indolidene intermediate (Figure 1C).
The affinity of each peptide for 5G12 was determined by
SPR under steady-state conditions (Figure 2B and S4) and,
Figure 2. Mapping the epitope of 5G12, an antibody developed to explore the C-glycome. (a) The structure of 5G12 (PDB ID: 6PLH) in complex
with a C-mannosylated peptide antigen. (b) The affinity of Trp(Man) peptides prepared by SPPS for the 5G12 mAb, as determined by SPR.
Graduated highlights are indicative of relative affinity, with tight binders in red and weak binders in blue. (c) A representative sensorgram for
binding and dissociation of the Fmoc-Glu-Trp(Man)-Ser-NH glycopeptide 14b to immobilized 5G12.
2
D
https://doi.org/10.1021/jacs.1c05567
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 3. Total synthesis of Carmo-HrTH-I 14n. (a) The strategy for automated SPPS of Carmo-HrTH-I 14n. Conditions: (i) 4 eq Fmoc-Xxx−
OH, 1.25 eq DIPEA, 5 eq DIC, 5 eq Oxyma, microwave 50 °C, 10 min; (ii) pyrrolidine/DMF (1:4), r.t. Five min; (iii) 4 eq Fmoc-Asn-OH, 4 eq
DIPEA, 4 eq HATU, 20 eq HOBt, microwave 50 °C, 1 h; (iv) TFA/TIPS/DMB/CH Cl (1:4:5:90), r.t. ten min. (b) The retention time of
2
2
44
synthetic 14n matches that of authentic material extracted from C. morosus corpora cardiaca. (c) Tandem MS data for both synthetic and
authentic 14n confirm the sequence and site of modification in both peptide samples.
where possible, kinetic fitting of sensorgrams (Figure 2C and
S5). These data confirmed that the 5G12 mAb has a strong
preference for peptides with a Glu-Trp(Man) motif but also a
similar affinity for the Gln-Trp(Man) and Leu-Trp(Man)
motifs, which bind with affinities only 5- and 6-fold lower,
respectively. This lies in stark contrast to peptides with a Ala/
Ser/Asp/Asn-Trp(Man) motif, which have 50- to 200-fold
worse affinity for the mAb. Substitution of the residue C-
terminal to Trp(Man) had minimal impact on binding,
commensurate with structural data showing that it makes no
direct interactions with the mAb (Figure 2A). In addition to
demonstrating that the residue N-terminal to Trp(Man) plays
an important role in binding to this antibody, these data also
reveal that the 5G12 mAb has very low affinity for Trp(Man)
alone and have confirmed that Trp(Man) is essential for
peptide binding. While this exercise was intended as a proving
ground for automated SPPS with Trp(Man), it has provided
valuable insights into the specificity and limitations of the only
tool available for mapping the C-glycome. It has also provided
a valuable set of peptides for the generation and evaluation of
next-generation mAb tools for enriching C-glycosylated
peptides in a less-biased fashion.
amidation. The latter modification could easily be installed
by performing SPPS on Sieber amide resin, while the size and
sequence of the peptide suggested we could proceed without
the need for side-chain protecting groups (Figure 3A).
Thus, Fmoc-Thr-OH, Fmoc-Gly-OH and Fmoc-Trp-
(Man)−OH 13 were coupled to Sieber amide resin in a
stepwise fashion using standard DIC/oxyma coupling con-
ditions. The subsequent Fmoc-Asn-OH was introduced using
45,46
HATU/HOBt to avoid dehydration side-products,
before
returning to DIC/oxyma coupling conditions to complete
assembly of the peptide. The final coupling involving pGlu
used a prolonged reaction time to ensure efficient addition of
this relatively unreactive residue. The final product was
liberated from the resin by brief exposure to a cleavage
cocktail containing 1% TFA followed quickly by careful
neutralization and purification by preparative HPLC to provide
20 mg of Carmo-HrTH-I 14n in 30% isolated yield. LC-MS
confirmed that the synthetic product had an identical retention
44
time to authentic material extracted from C. morosus (Figure
3B) and tandem mass spectrometry confirmed the sequence of
both synthetic and authentic peptides, as well as the location of
the modification (Figure 3C).
Our next goal was more ambitious: the automated SPPS of
Prolonged incubation of the peptide with TFA resulted in a
shift in retention time but not mass, which is consistent with
acid-catalyzed anomerization of the synthetic α-C-mannoside
to the β-anomer (Figure S6). Collectively, the synthesis of
peptides 14a−n demonstrates that automated Fmoc-based
SPPS strategies are viable for the preparation of glycopeptides
Carmo-HrTH-I 14n, a hypertrehalosaemic peptide hormone
44
from the stick insect Carausius morosus. This decapeptide,
originally isolated from the corpora cardiaca of 2000 stick
6
insects and characterized by NMR, possesses an N-terminal
pyroglutamate, Trp(Man) at position 8, and C-terminal
E
https://doi.org/10.1021/jacs.1c05567
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
a
Scheme 3. Late-Stage α-C-Mannosylation of Peptides Possessing 2-Bromo-L-Tryptophan Residues
a
Yields without parentheses were determined by HLPC, while isolated yields are presented in parentheses.
possessing Trp(Man) but that the acid lability of the
modification may present some challenges when selecting
side-chain protecting groups.
To provide an alternative to SPPS with Trp(Man), we
returned to our cross-coupling reaction to reoptimize it for use
in more polar solvents so that larger peptides might be
subjected to late-stage C-glycosyaltion. For this work, the
dipeptide 11h served as our model substrate. With HE as
photoreductant, very little product (<5%) was obtained and
hydrodehalogenation was the major byproduct (Table S3).
After some experimentation, we found that transitioning to
NMP as solvent, removal of the HE photoreductant and
reintroduction of a PC (4CzIPN), shifting back to blue light
irradiation and replacing mannosyl bromide 8 with mannosyl
F
https://doi.org/10.1021/jacs.1c05567
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
chloride 15 provided product in 45% yield. Use of the electron
rich diOMebpy ligand for the Ni catalyst further improved the
yield to 64%. The addition of water to the system then
dramatically improved the rate of the reaction and the yield to
CONCLUSION
■
In the process of developing an efficient, inexpensive, and
versatile route to Trp(Man) 1, we have demonstrated the
utility of HE as a photoreductant to drive the Ni-catalyzed
cross-coupling of glycosyl and aryl bromides. We have
repeatedly executed the synthesis of Fmoc-protected Trp-
7
3% (64% isolated): a change that was made possible by
shifting from the hydrolytically labile mannosyl bromide 8 to
the more resilient mannosyl chloride 15.
(
Man) 13 on a gram-scale and consistently obtained this
Having optimized this condition on the relatively small
dipeptide 11h, we next applied it to the hexapeptide 16a to
determine if larger substrates were efficiently modified.
Gratifyingly, the late-stage C-mannosylation of 16a proceeded
smoothly to give 17a in 40% isolated yield (Scheme 3). The
most significant byproduct observed in this and subsequent
reactions arose from hydrodehalogenation.
Though these photoredox conditions were not anticipated
to be compatible with unprotected amines and carboxylic
acids, we sought to demonstrate this empirically and offer
solutions to this limitation. Accordingly, peptides 16b and 16c,
which had a free α- and ϵ-amino group, respectively, were
prepared and subjected to the cross-coupling conditions,
providing only trace amounts of product and a complex
mixture of degradation products. This shortcoming was easily
overcome by protecting α- and ϵ-amino groups as trifluor-
oacetamides: as demonstrated by the efficient C-mannosyla-
tion of 16d to give 17d in 35% isolated yield.
important building block in 65% overall yield in four steps
from commercially available starting materials. This has
enabled the first automated SPPS of α-C-mannosylated
glycopeptides and the discovery of their facile acid-mediated
anomerization. We have also developed a photocatalytic
variation of this cross-coupling chemistry that operates in
polar aprotic solvent systems, is tolerant of water, and
facilitates the late-stage C-mannosylation of partially protected
peptides bearing a 2-bromo-L-tryptophan residue. The
advantage of the former method is that it does not require
expensive photocatalysts, though it is limited by substrate
solubility in acetonitrile. The latter cross-coupling condition
does require photocatalyst but can be applied to larger and
more complex peptide substrates, owing to its compatibility
with polar aprotic solvents and water. Collectively, these
methods empower efforts to delineate the biological roles of
tryptophan C-mannosylation by providing access to homoge-
neous glycopeptides and glycoproteins.
Unprotected carboxylates fared little better than the amines.
Peptide 16e, with a C-terminal carboxylate, provided little
product, while peptide 16f, with a free Glu side chain, provided
enough product to isolate. As expected, protecting these
carboxylates as simple methyl esters dramatically improved
yields, with 17g obtained in 34% isolated yield. Both the
trifluoroacetamide protection of amines and methyl ester
protection of acids can be removed at the same time as the
acetyl protecting groups on the mannose by saponification.
We continued to explore the compatibility of other
functional groups with these cross-coupling conditions. The
guanidinyl group of Arg-containing peptide 16h and phenolic
side chain of Tyr-containing peptide 16i were well-tolerated,
providing 17h and 17i in 33% and 55% yield, respectively. The
same was true for the thioethers in peptide 16j, which
possesses a Met and acetamidomethyl (Acm)-protected Cys, as
well as peptide 16k, which possesses a biotin moiety. Both 17j
and 17k were obtained in over 50% isolated yield.
Furthermore, the terminal alkyne of the propargylglycine
ASSOCIATED CONTENT
sı Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/jacs.1c05567.
■
*
Experimental procedures and analytical data (PDF)
AUTHOR INFORMATION
■
Corresponding Author
Ethan D. Goddard-Borger − The Walter and Eliza Hall
Institute of Medical Research, Parkville, Victoria 3052,
Australia; Department of Medical Biology, University of
Melbourne, Parkville, Victoria 3010, Australia; orcid.org/
0000-0002-8181-9733; Email: goddard-borger.e@
wehi.edu.au
Authors
Runyu Mao − The Walter and Eliza Hall Institute of Medical
Research, Parkville, Victoria 3052, Australia; Department of
Medical Biology, University of Melbourne, Parkville, Victoria
(
Ppg) residue in 16l was sufficiently well-tolerated to provide
the C-mannosylated product 17l in 21% yield.
We next sought to demonstrate how this late-stage
modification approach can be used to rapidly synthesize larger
glycopeptides. As an alternative route to Carmo-HrTH-I 14n,
the corresponding brominated peptide 16m was assembled
using SPPS and then subjected to the coupling conditions to
provide 17m in an impressive 37% isolated yield.
3010, Australia
Shiyi Xi − The Walter and Eliza Hall Institute of Medical
Research, Parkville, Victoria 3052, Australia; Department of
Medical Biology, University of Melbourne, Parkville, Victoria
3010, Australia
Sayali Shah − The Walter and Eliza Hall Institute of Medical
Research, Parkville, Victoria 3052, Australia; Department of
Medical Biology, University of Melbourne, Parkville, Victoria
As a final demonstration of the power of this methodology,
and the ease with which trifluoroacetamides and methyl esters
can be deprotected following C-mannosylation, we synthesized
peptide 18, which mimics the C-mannosylated repeats of the
holdfast protein pvfp-1 from Perna viridis (Asian green
3010, Australia
Michael J. Roy − The Walter and Eliza Hall Institute of
Medical Research, Parkville, Victoria 3052, Australia;
Department of Medical Biology, University of Melbourne,
Parkville, Victoria 3010, Australia; orcid.org/0000-0003-
0198-9108
4
7
mussel). The precursor peptide substrate 16n was easily
prepared using standard Fmoc-based SPPS chemistry and then
smoothly converted to the protected C-mannosylated product
1
7n in 30% isolated yield using our cross-coupling conditions.
Alan John − The Walter and Eliza Hall Institute of Medical
Research, Parkville, Victoria 3052, Australia; Department of
Saponification of 17n provided 18 in excellent yield.
G
https://doi.org/10.1021/jacs.1c05567
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Medical Biology, University of Melbourne, Parkville, Victoria
selective labeling approach. Angew. Chem., Int. Ed. 2020, 59 (46),
0659−20665.
5) John, A.; Järva,
2
(
3
010, Australia
̊
M. A.; Shah, S.; Mao, R.; Chappaz, S.;
James P. Lingford − The Walter and Eliza Hall Institute of
Medical Research, Parkville, Victoria 3052, Australia;
Department of Medical Biology, University of Melbourne,
Parkville, Victoria 3010, Australia
Gerd Gäde − Department of Biological Sciences, University of
Cape Town, RSA-7700 Rondebosch, South Africa
Nichollas E. Scott − Department of Microbiology and
Immunology, University of Melbourne at the Peter Doherty
Institute for Infection and Immunity, Parkville, Victoria 3010,
Australia; orcid.org/0000-0003-2556-8316
Birkinshaw, R. W.; Czabotar, P. E.; Lo, A. W.; Scott, N. E.; Goddard-
Borger, E. D. Yeast- and antibody-based tools for studying tryptophan
C-mannosylation. Nat. Chem. Biol. 2021, 17 (4), 428−437.
(6) Munte, C. E.; Gäde, G.; Domogalla, B.; Kremer, W.; Kellner, R.;
Kalbitzer, H. R. C-mannosylation in the hypertrehalosaemic hormone
from the stick insect Carausius morosus. FEBS J. 2008, 275 (6),
1
(
163−73.
7) Hofsteenge, J.; Huwiler, K. G.; Macek, B.; Hess, D.; Lawler, J.;
Mosher, D. F.; Peter-Katalinic, J. C-mannosylation and O-fucosylation
of the thrombospondin type 1 module. J. Biol. Chem. 2001, 276 (9),
6
(
485−98.
Complete contact information is available at:
https://pubs.acs.org/10.1021/jacs.1c05567
8) Furmanek, A.; Hess, D.; Rogniaux, H.; Hofsteenge, J. The
WSAWS motif is C-hexosylated in a soluble form of the
erythropoietin receptor. Biochemistry 2003, 42 (28), 8452−8.
Notes
(9) Okamoto, S.; Murano, T.; Suzuki, T.; Uematsu, S.; Niwa, Y.;
The authors declare no competing financial interest.
Sasazawa, Y.; Dohmae, N.; Bujo, H.; Simizu, S. Regulation of
secretion and enzymatic activity of lipoprotein lipase by C-
mannosylation. Biochem. Biophys. Res. Commun. 2017, 486 (2),
ACKNOWLEDGMENTS
■
5
(
58−563.
We would like to thank Dr. Martin Brzozowski for the
generous gift of 4CzIPN and the Melbourne Mass Spectrom-
etry and Proteomics Facility of The Bio21 Molecular Science
and Biotechnology Institute at The University of Melbourne
for assistance with mass spectrometry analyses. We would like
to acknowledge support from The Walter and Eliza Hall
Institute of Medical Research; National Health and Medical
Research Council of Australia (NHMRC) project grants
GNT1139546, GNT1139549 and GNT2000517; ARC Dis-
covery Project Grant DP210100362, the Australian Cancer
Research Fund and a Victorian State Government Operational
Infrastructure support grant. N.E.S is supported by an ARC
Future Fellowship and E.D.G.B. is supported by the Brian M.
Davis Charitable Foundation Centenary Fellowship.
10) Pronker, M. F.; Lemstra, S.; Snijder, J.; Heck, A. J.; Thies-
Weesie, D. M.; Pasterkamp, R. J.; Janssen, B. J. Structural basis of
myelin-associated glycoprotein adhesion and signalling. Nat. Commun.
2
(
016, 7, 13584.
11) Falzarano, D.; Krokhin, O.; Van Domselaar, G.; Wolf, K.;
Seebach, J.; Schnittler, H. J.; Feldmann, H. Ebola sGP–the first viral
glycoprotein shown to be C-mannosylated. Virology 2007, 368 (1),
8
3−90.
(12) Shcherbakova, A.; Preller, M.; Taft, M. H.; Pujols, J.; Ventura,
S.; Tiemann, B.; Buettner, F. F.; Bakker, H. C-mannosylation supports
folding and enhances stability of thrombospondin repeats. eLife 2019,
8, 52978 DOI: 10.7554/eLife.52978.
(13) Kulkarni, S. S.; Sayers, J.; Premdjee, B.; Payne, R. J. Rapid and
efficient protein synthesis through expansion of the native chemical
ligation concept. Nat. Rev. Chem. 2018, 2 (4), 0122.
(
14) Conibear, A. C.; Watson, E. E.; Payne, R. J.; Becker, C. F. W.
ABBREVIATIONS
■
Native chemical ligation in protein synthesis and semi-synthesis.
Chem. Soc. Rev. 2018, 47 (24), 9046−9068.
4
CzIPN, 1,2,3,5-tetrakis(carbazol-9-yl)-4,6-dicyanobenzene;
Acm, acetamidomethyl; DIC, N,N′-diisopropylcarbodiimide;
diOMebpy, 4,4′-dimethoxy-2,2′-bipyridine; DIPEA, N,N-dii-
sopropylethylamine; HATU, (1-[Bis(dimethylamino)-
methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium 3-oxide hexa-
fluorophosphate; HE, Hantzsch ester; HFIP, 1,1,1,3,3,3-
hexafluoro-2-propanol; HOBt, hydroxybenzotriazole; mAb,
monoclonal antibody; Oxyma, ethyl cyano(hydroxyimino)-
acetate; PC, photocatalyst; SPPS, solid phase peptide syn-
thesis; SPR, surface plasmon resonance; TFA, trifluoroacetic
acid or trifluoroacetyl; Trp(Man), 2-(α-D-mannopyranosyl)-L-
tryptophan; TSR, thrombospondin repeat
(15) Nishikawa, T.; Koide, Y.; Kajii, S.; Wada, K.; Ishikawa, M.;
Isobe, M. Stereocontrolled syntheses of α-C-mannosyltryptophan and
its analogues. Org. Biomol. Chem. 2005, 3 (4), 687−700.
(16) Nishikawa, T.; Ishikawa, M.; Wada, K.; Isobe, M. Total
Synthesis of α-C-Mannosyltryptophan, a Naturally Occurring C-
Glycosyl Amino Acid. Synlett 2001, 2001, 0945.
(
17) Manabe, S.; Marui, Y.; Ito, Y. Total synthesis of mannosyl
tryptophan and its derivatives. Chem. - Eur. J. 2003, 9 (6), 1435−47.
18) Manabe, S.; Ito, Y. Total synthesis of novel subclass of glyco-
(
2
amino acid structure motif: C -α- -C-mannosylpyranosyl- -trypto-
D
L
phan. J. Am. Chem. Soc. 1999, 121 (41), 9754−9755.
(19) Wang, Q.; Fu, Y.; Zhu, W.; An, S.; Zhou, Q.; Zhu, S.-F.; He, G.;
Liu, P.; Chen, G. Total synthesis of C-α-mannosyl tryptophan via
palladium-catalyzed C−H glycosylation. CCS Chem. 2021, 3, 1729−
REFERENCES
■
1736.
(
1) de Beer, T.; Vliegenthart, J. F.; Loffler, A.; Hofsteenge, J. The
(20) Cong, F.; Lv, X. Y.; Day, C. S.; Martin, R. Dual catalytic
hexopyranosyl residue that is C-glycosidically linked to the side chain
of tryptophan-7 in human RNase Us is alpha-mannopyranose.
Biochemistry 1995, 34 (37), 11785−9.
strategy for forging sp(2)-sp(3) and sp(3)-sp(3) architectures via β-
scission of aliphatic alcohol derivatives. J. Am. Chem. Soc. 2020, 142
(
49), 20594−20599.
(
2) Hofsteenge, J.; Muller, D. R.; de Beer, T.; Loffler, A.; Richter, W.
(
21) Ma, Y.; Liu, S.; Xi, Y.; Li, H.; Yang, K.; Cheng, Z.; Wang, W.;
J.; Vliegenthart, J. F. New type of linkage between a carbohydrate and
a protein: C-glycosylation of a specific tryptophan residue in human
RNase Us. Biochemistry 1994, 33 (46), 13524−30.
Zhang, Y. Highly stereoselective synthesis of aryl/heteroaryl-C-
nucleosides via the merger of photoredox and nickel catalysis.
Chem. Commun. 2019, 55 (97), 14657−14660.
(22) Wei, Y.; Ben-Zvi, B.; Diao, T. Diastereoselective synthesis of
aryl C-glycosides from glycosyl esters via C-O bond homolysis. Angew.
Chem., Int. Ed. 2021, 60, 9433.
(
3) Frank, M.; Beccati, D.; Leeflang, B. R.; Vliegenthart, J. F. G. C-
Mannosylation Enhances the Structural Stability of Human RNase 2.
iScience 2020, 23 (8), 101371.
4) Jonker, H. R. A.; Saxena, K.; Shcherbakova, A.; Tiemann, B.;
(
Bakker, H.; Schwalbe, H. NMR Spectroscopic characterization of the
C-mannose conformation in a thrombospondin repeat using a
(23) Dumoulin, A.; Matsui, J. K.; Gutierrez-Bonet, A.; Molander, G.
A. Synthesis of non-classical arylated C-saccharides through nickel/
H
https://doi.org/10.1021/jacs.1c05567
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
photoredox dual catalysis. Angew. Chem., Int. Ed. 2018, 57 (22),
(44) Gäde, G.; Kellner, R.; Rinehart, K. L.; Proefke, M. L. A
tryptophan-substituted member of the AKH/RPCH family isolated
from a stick insect corpus cardiacum. Biochem. Biophys. Res. Commun.
1992, 189 (3), 1303−9.
6
(
614−6618.
24) Giese, B.; Dupuis, J. Diastereoselective syntheses of C-
glycopyranosides. Angew. Chem., Int. Ed. Engl. 1983, 22 (8), 622−623.
25) Korth, H.-G.; Sustmann, R.; Dupuis, J.; Giese, B. Electron spin
(
(45) Mojsov, S.; Mitchell, A. R.; Merrifield, R. B. A quantitative
evaluation of methods for coupling asparagine. J. Org. Chem. 1980, 45
(4), 555−560.
resonance spectroscopic investigation of carbohydrate radicals. Part 2.
Conformation and configuration in pyranos-1-yl radicals. J. Chem.
Soc., Perkin Trans. 2 1986, 9, 1453−1459.
(46) Gausepohl, H.; Kraft, M.; Frank, R. W. Asparagine coupling in
Fmoc solid phase peptide synthesis. Int. J. Pept. Protein Res. 1989, 34
4), 287−94.
47) Hwang, D. S.; Zeng, H.; Lu, Q.; Israelachvili, J.; Waite, J. H.
(
26) Beckwith, A. L. J.; Duggan, P. J. The quasi-homo-anomeric
(
interaction in substituted tetrahydropyranyl radicals: Diastereoselec-
(
tivity. Tetrahedron 1998, 54 (24), 6919−6928.
Adhesion mechanism in a DOPA-deficient foot protein from green
mussels. Soft Matter 2012, 8 (20), 5640−5648.
(
27) Andrews, R. S.; Becker, J. J.; Gagne, M. R. Intermolecular
addition of glycosyl halides to alkenes mediated by visible light.
Angew. Chem., Int. Ed. 2010, 49 (40), 7274−6.
(
28) Andrews, R. S.; Becker, J. J.; Gagne, M. R. A photoflow reactor
for the continuous photoredox-mediated synthesis of C-glycoamino
acids and C-glycolipids. Angew. Chem., Int. Ed. 2012, 51 (17), 4140−
3
.
(
29) Andrews, R. S.; Becker, J. J.; Gagne, M. R. Investigating the rate
of photoreductive glucosyl radical generation. Org. Lett. 2011, 13 (9),
406−9.
30) Steiman, T. J.; Liu, J.; Mengiste, A.; Doyle, A. G. Synthesis of
2
(
beta-phenethylamines via Ni/photoredox cross-electrophile coupling
of aliphatic aziridines and aryl iodides. J. Am. Chem. Soc. 2020, 142
(
16), 7598−7605.
(
31) Duan, Z.; Li, W.; Lei, A. Nickel-catalyzed reductive cross-
coupling of aryl bromides with alkyl bromides: Et3N as the terminal
reductant. Org. Lett. 2016, 18 (16), 4012−5.
(
32) Wang, S.; Qian, Q.; Gong, H. Nickel-catalyzed reductive
coupling of aryl halides with secondary alkyl bromides and allylic
acetate. Org. Lett. 2012, 14 (13), 3352−5.
(
33) Zhao, C.; Jia, X.; Wang, X.; Gong, H. Ni-catalyzed reductive
coupling of alkyl acids with unactivated tertiary alkyl and glycosyl
halides. J. Am. Chem. Soc. 2014, 136 (50), 17645−51.
(
34) Wang, X.; Wang, S.; Xue, W.; Gong, H. Nickel-catalyzed
reductive coupling of aryl bromides with tertiary alkyl halides. J. Am.
Chem. Soc. 2015, 137 (36), 11562−5.
(
35) Biswas, S.; Weix, D. J. Mechanism and selectivity in nickel-
catalyzed cross-electrophile coupling of aryl halides with alkyl halides.
J. Am. Chem. Soc. 2013, 135 (43), 16192−7.
(
36) Diccianni, J. B.; Diao, T. Mechanisms of nickel-catalyzed cross-
coupling reactions. Trends in Chemistry 2019, 1 (9), 830−844.
(
37) Buzzetti, L.; Prieto, A.; Roy, S. R.; Melchiorre, P. Radical-based
C-C bond-forming processes enabled by the photoexcitation of 4-
alkyl-1,4-dihydropyridines. Angew. Chem., Int. Ed. 2017, 56 (47),
1
(
5039−15043.
38) van Leeuwen, T.; Buzzetti, L.; Perego, L. A.; Melchiorre, P. A
redox-active nickel complex that acts as an electron mediator in
photochemical Giese reactions. Angew. Chem., Int. Ed. 2019, 58 (15),
4
(
953−4957.
39) Jung, J.; Kim, J.; Park, G.; You, Y.; Cho, E. J. Selective
debromination and α-hydroxylation of α-bromo ketones using
Hantzsch esters as photoreductants. Adv. Synth. Catal. 2016, 358
(
1), 74−80.
(
40) Kammer, L. M.; Badir, S. O.; Hu, R.-M.; Molander, G. A.
Photoactive electron donor-acceptor complex platform for Ni-
mediated C(sp3)−C(sp2) bond formation. Chem. Sci. 2021, 12
(
15), 5450−5457.
(
41) Emmanuel, M. A.; Greenberg, N. R.; Oblinsky, D. G.; Hyster,
T. K. Accessing non-natural reactivity by irradiating nicotinamide-
dependent enzymes with light. Nature 2016, 540 (7633), 414−417.
(
42) Chen, X.; Ye, F.; Luo, X.; Liu, X.; Zhao, J.; Wang, S.; Zhou, Q.;
Chen, G.; Wang, P. Histidine-specific peptide modification via visible-
light-promoted C-H alkylation. J. Am. Chem. Soc. 2019, 141 (45),
1
(
8230−18237.
43) Wang, P. Z.; Chen, J. R.; Xiao, W. J. Hantzsch esters: an
emerging versatile class of reagents in photoredox catalyzed organic
synthesis. Org. Biomol. Chem. 2019, 17 (29), 6936−6951.
I
https://doi.org/10.1021/jacs.1c05567
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX