Solvent-Controlled Chemoselectivity in the Photolytic Release of Hydroxamic Acids and Carboxamides from Solid Support

Article abstract of DOI:10.1021/acs.orglett.7b01386

The synthetic utility and theoretical basis of a photolabile hydroxylamine-linker are presented. The developed protocols enable the efficient synthesis and chemoselective photolytic release of either hydroxamates or carboxamides from solid support. The bi

Full text of DOI:10.1021/acs.orglett.7b01386

Letter
pubs.acs.org/OrgLett
Solvent-Controlled Chemoselectivity in the Photolytic Release of
Hydroxamic Acids and Carboxamides from Solid Support
Katrine Qvortrup,*,† Rico G. Petersen,† Asmus O. Dohn,† Klaus B. Møller,† and Thomas E. Nielsen†,‡
†Department of Chemistry, Technical University of Denmark, DK-2800 Kgs. Lyngby, Denmark
‡Singapore Centre for Environmental Life Sciences Engineering, Nanyang Technological University, Singapore 637551, Singapore
S
* Supporting Information
ABSTRACT: The synthetic utility and theoretical basis of a
photolabile hydroxylamine-linker are presented. The devel-
oped protocols enable the efficient synthesis and chemo-
selective photolytic release of either hydroxamates or
carboxamides from solid support. The bidetachable mode of
the linker unit is uniquely dependent on the solvent.
Hydroxamic acids are obtained by performing photolysis in protic solvents, whereas photolysis in aprotic solvents enables the
selective release of carboxamides.
ydroxamic acids have been the source of much
1
Hbiochemical interest in recent years. Therefore, the use
of solid-phase2 combinatorial chemistry3 for high-throughput
generation of structurally diverse hydroxamic acids is highly
relevant. Although hydroxamic acids may be obtained by direct
cleavage of resin-bound esters with hydroxylamine derivatives,4
this strategy requires an excess of hydroxylamine and/or
addition of base which complicates postcleavage workup.
Several approaches involving resin-bound hydroxylamine link-
ers have been reported.5 However, these hydroxamate linkages
suffer from only being cleavable under acidic conditions, which
limits the range of chemical transformations applicable to the
solid-phase synthesis of structurally diverse hydroxamic acids.
Therefore, other cleavage principles are necessary in order to
provide complex molecules assembled through a diverse range
of chemical reactivity. A linker system that can be cleaved under
photolytic conditions may be considered truly orthogonal in
Figure 1. 4,5-Dialkoxy-2-nitrobenzyl moiety in photolabile linkers for
this context.6 Futhermore, photolytic cleavage offers a mild
solid-phase synthesis.
method of cleavage which is particularly attractive for the direct
release of screening compounds into biological screens without
contamination by cleavage reagents.
hydroxylamine-functionalized photolabile support. Using stand-
ard TBTU-mediated peptide coupling reactions, derivative 5a
was synthesized as a simple and easily monitorable model
system. Photolytic cleavage was carried out on resin suspended
in H2O/MeOH (4:1) by irradiating for 30 min at rt with 365
nm light using an LED UV-lamp. Analysis of the released
material via RP-UPLC, however, showed release of two
products: the hydroxamate 7a and the carboxamide 8a resulting
from C−O and N−O cleavage, respectively, in a 3:4 ratio.
The nature of the solvent9 and the acidity of the solution10,11
have been demonstrated to have pronounced effect on the
kinetics and equilibrium position of aci-nitro compounds
(Figure 2). We first explored the solvent effects in the
We now wish to report a complete study on a photolabile
linker based on the o-nitroveratryl group7 capable of releasing
hydroxamates upon UV irradiation. Uniquely, this linker unit
may function as a “bidetachable” system. By simply varying the
reaction solvent, the photolysis can be controlled to provide
either C−O or C−N bond cleavage, which allows for controlled
release of the hydroxamate or carboxamide, respectively (Figure
1). This strategy may introduce further diversity into target
molecules and compound libraries. Linker 4 was readily
prepared in a few high-yielding steps (Scheme 1)8 before
being explored as a hydroxamate-releasing linker. A N-[(1H-
benzotriazol-1-yl)(dimethylamino)-methylene]-N-methyl-
methanaminium tetrafluoroborate N-oxide (TBTU)-mediated
coupling of 4 to a Rink linker attached to the commercially
available amino-functionalized support (PEGA800) afforded the
Received: May 8, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b01386
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 1. (A) Synthesis of Fmoc-Protected Hydroxylamine-
Functionalized Carboxylic Acid Linker (4); (B) Application
of Linker 4 in Solid-Phase Peptide Synthesis (SPPS) for the
Photolytic Release of Peptide Hydroxamic Acids and
Carboxamides
Table 1. Relative Product Yields for Photolysis of 5a at 360
nm in Various Solvents
a
b
entry
solvent
mesitylene
product 7a:8a
A
B
C
D
E
0:100
67:33
60:40
98:2
MeOH/H2O (1:4)
H2O
HFIP
mesitylene/HFIP (1:1)
98:2
a
Photolytic cleavage was carried out for 0.5 h with an LED UV-lamp
b
(360 nm). Product distribution was determined by RP-HPLC.
The effect of Lewis acid catalysis on the photoreaction of 5a
has also been investigated (SI). The qualitative studies showed
that a wide range of Lewis acids favor the formation of the
hydroxamate product. The most efficacious Lewis acid was
found to be BF3, giving high selectivity toward formation of the
hydroxamate product 8a.
It is well-known that o-nitroveratryl compounds upon
irradiation undergo a Norrish Type II β-hydrogen abstraction
to give the biradical intermediate 11,12 which after photo-
isomerization forms the E,Z-12 and Z,Z-12 isomers (Figure 2).
From there it can again undergo isomerization to the E-aci-
nitro forms 14EE and 14ZE. Measurements of aci-nitro transients
have confirmed the presence of 13− as an intermediate between
12 and 14,11 and direct isomerization of 12 to 14 by rotation
about the CN bond has been excluded.13,14 Also, the
conversion of 12 to 14 via direct proton shift between the two
oxygen atoms of the aci-nitro group (without participation of a
water molecule) seems unlikely.15,11 The activation barrier
computed by our density functional theory calculations yielded
a barrier of 123 kJ mol−1 for this direct proton shift in the
species 12 derived from 9, where R1 = Ph, R2 = Me.
Furthermore, inclusion of a single water molecule to mediate
the shift of this proton was computed to lower the activation
barrier by at least 81 kJ mol−1 (see SI for computational
details). The presence of additional water molecules in bulk
solution should lower the activation barrier for water-mediated
proton exchange even further.15 Taking this solvation effect
into account, water-mediated proton exchange (or proton
transfer mediated by other protic solvents) via the anion 13 can
be assumed to be the most likely path between the nitronic acid
isomers 12 and 14, with an activation barrier of only a few kJ
mol−1. Based on this discussion, we assume that the equilibrium
between the two possible protonation sites on the nitronic acid
photolysis of 6a on the level of final product formation. The
photoreaction was studied by photolyzing aliquots of the resin
6a in various solvents and determining the product distribution
via HPLC analysis. Because the solvent also influences the
swelling and solvation properties of the support, the obtained
results are merely qualitative. While this technique did not
allow us to quantify the amount of products formed, it did
provide an expedient method to determine the relative
photoproduct formation. Selected product yield profiles are
listed in Table 1 (for a comprehensive list consult the
Supporting Information (SI)). It is evident that the solvent
has a strong influence on the product ratio of the reaction and
some general conclusions may be drawn. Polar solvents favor
formation of the hydroxamic acid product 7a, while apolar
solvents mainly give the carboxamide product 8a. In particular,
the polar fluorinated alcohol, hexafluoroisopropanol (HFIP),
with a high hydrogen-bond-donating ability led to hydroxamic
acid product 7a with high selectivity. Apolar solvents favor
formation of the carboxamide product 8a over hydroxamate
product 7a. Notably, when using mesitylene, carboxamide
product 8a was formed exclusively.
Figure 2. Proposed mechanism for the photolytic degradation of hydroxamate-functionalized o-nitroveratryl derivatives. For simplicity, only the E-
isomers with regard to the C−OR group are shown.
B
DOI: 10.1021/acs.orglett.7b01386
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
function is established on the ns time scale in polar protic
solutions.
Table 2. Synthesis and Photolytic Release of Hydroxamates
7a−h and Carboxamides 8a−h
It is generally assumed that the decay of the aci-nitro forms is
the rate-limiting step in the photoisomerization of o-nitrobenzyl
derivatives and that cyclization to form intermediate 15
proceeds only from the neutral aci-tautomer 14.11−13 Under
conditions where interconversion between the two aci-nitro
forms is efficient, we expect the “normal” hydroxamic acid
product forming pathway (14 → 15 → 16) to be fast.
However, in aprotic solvent where ionization to 13− does not
occur, the Z-aci-nitro species 12 give rise to a N−O bond
fragmentation pathway, which generates the amide and
nitroketone products 19 and 20. The proposed mechanism is
depicted in Figure 2.
To further investigate the photolysis of hydroxamate-
functionalized o-nitroveratryl compounds, we synthesized 21
(SI) as a model compound and studied the photolysis in
solution (Figure 3). Hereby we were able to identify the nature
a
Photolytic cleavage was carried out for 2 h with an LED UV-lamp
b
(360 nm). Purity was determined by RP-HPLC.
Figure 3. Study of product distribution for the photolytic degradation
of 21 in HFIP and toluene, respectively.
mg of resin suspended in appropriate solvent by irradiating for
0.5−3 h at rt with 365 nm light using an LED UV-lamp. We
showed the possibility of selectively cleaving these compounds
to give the hydroxamate and the carboxamide products,
respectively. Selected examples of cleavage of a variety of
compounds are presented in Table 2 (for a more elaborate
study on cleavage of the full compound library, see SI). From
Table 2 it can be concluded that the developed solid-phase
methodology is very robust and applicable to a range of both
aromatic and aliphatic hydroxamates. The liberated products
were recovered in high purity (90−95%) and satisfactory yields
(35−63%).
While the linker 4 has been shown to be stable toward both
acidic and basic condition, we investigated the utility of the
linker for the synthesis of acid- and base-labile substrates. Both
hydroxamate functionalized amino acid derivatives containing
Boc- (7h) and Fmoc- (7g) protected α-amino groups, Trt-
protected amide (7g), and Pbf-protected guanidinium (7h) side
chain groups were successfully released, demonstrating the
extraordinary protecting group compatibility of this linker resin.
To further demonstrate the synthetic potential of the linker
for the generation of more complex structures, we investigated
the use of linker 4 for the synthesis of a derivative of a known
diketopiperazine (DKP) hydroxamic acid HDAC inhibitor.17
Massive efforts in solid-phase synthesis have strived for the
development of synthesis methodology, which systematically
generates natural product-like compounds of high spatial
complexity. In this context a current limitation is the difficulties
faced in the synthesis of acid and base sensitive scaffolds,
including racemization-prone structures. To demonstrate the
use of linker 4 for the generation of the hydroxamate-
functionalized DKP derivative 25, a serine-terminated oligo-
meric peptide sequence 24 was assembled on a hydroxylamine-
functionalized photolabile support by standard SPPS protocols.
Exposing the resin 24 to classical periodate oxidation
of byproducts formed in the photolysis of a hydroxamate-
functionalized o-nitroveratryl compound. Furthermore, solution
phase photolysis experiments provide the opportunity to study
the photolysis without influences from swelling and solvation
properties of the solid support. Photolysis of 21 was carried out
in a broad range of solvents (see SI). The low solubility of 21
did not allow an investigation of irradiation experiments in
mesitylene and saturated hydrocarbon solvents. In polar
solvents and in acidic solutions (CH3CN, HFIP, 1% TFA in
MeOH) the hydroxamic acid product formation is the only
observed pathway, while the apolar solvent toluene gave a
mixture of hydroxamic acid 7a and carboxamide 8a in a ratio of
1:1. Two examples of our results with 21 are shown in Figure 3.
Each peak in the chromatograms is characterized and identified
by UPLC-MS. In agreement with our proposed mechanism, we
observed from these experiments that the major byproduct
formed in polar solvents was o-nitrosobenzaldehyde 22, with
only minor impurities of 23, while o-nitrosobenzaldehyde 22
and o-nitrobenzaldehyde 23 were formed in a ratio of ∼1:1 in
toluene. The absence of other peaks in the chromatograms
indicates that no other side reactions had occurred. Confident
with the photolysis strategy, we employed the hydroxylamine
linker 4 for the parallel synthesis of a library of putatively
HDAC inhibitors16 (Table 2 and SI). A Rink linker was
positioned between the support and the photolinker unit to
optimize and verify attachment chemistry of linker 4 on the
solid support. After incubating the supports 5a−e with TFA/
CH2Cl2 (1:1) for 2 h, one major peak corresponding to
cleavage of the Rink linker was generally observed (6a−e),
indicating high efficiency of the attachment chemistry of 4 and
high stability of the photolabile unit toward TFA deprotection
conditions normally used in standard peptide synthesis
procedures. Photolytic cleavage was carried out on 30−100
C
DOI: 10.1021/acs.orglett.7b01386
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
K.; Komnatnyy, V. V.; Taveras, K. M.; Le Quement, S. T.; Frauenlob,
R.; Givskov, M.; Nielsen, T. E. Tetrahedron Lett. 2011, 52, 7121.
(5) (a) Floyd, C. D.; Lewis, C. N.; Patel, S. R.; Whittaker, M.
Tetrahedron Lett. 1996, 37, 8045. (b) Mellor, S. L.; McGuire, C.; Chan,
W. C. Tetrahedron Lett. 1997, 38, 3311. (c) Bauer, U.; Ho, W.-B.;
Koskinen, A. M. P. Tetrahedron Lett. 1997, 38, 7233. (d) Barlaam, B.;
Koza, P.; Berriot, J. Tetrahedron 1999, 55, 7221. (e) Ngu, K.; Patel, D.
V. J. Org. Chem. 1997, 62, 7088. (f) Gazal, S.; Masterson, L. R.; Barany,
G. J. Pept. Res. 2005, 66, 324.
conditions generated the corresponding aldehyde, and
subsequent TFA treatment mediated the N-acyliminium
cyclization. Rewardingly, photolytic release gave the hydrox-
amate-functionalized DKP-derivative 25 in high purity (Scheme
2).
Scheme 2. Synthesis of a Hydroxamate-Functionalized Fused
Natural Product-like DKP Derivative (25)
(6) (a) For a definition of an orthogonal protection scheme: Barany,
G.; Merrifield, R. B. J. Am. Chem. Soc. 1977, 99, 7363. (b) For a recent
review of linkers, see: James, I. W. Tetrahedron 1999, 55, 4855.
(c) Yoo, D. J.; Greenberg, M. M. J. Org. Chem. 1995, 60, 3358.
(7) (a) Holmes, C. P.; Jones, D. G. J. Org. Chem. 1995, 60, 2318. (b)
For an overview on photolabile linkers for solid-phase organic
synthesis, see: Scott, P. J. H. Linker Strategies in Solid-Phase Organic
Synthesis; John Wiley and Sons: 2009.
(8) Qvortrup, K.; Nielsen, T. E. Angew. Chem., Int. Ed. 2016, 55,
4472.
In summary, we have developed a photolabile hydroxylamine
linker for the synthesis of hydroxamic acids on solid support.
The synthesis strategy shows excellent compatibility with a
range of structurally diverse compounds. The linker is
compatible with most commonly used protecting groups for
SPPS and remains intact throughout the multistep synthesis.
Products are ultimately released from the solid support in high
purity using light. In addition, this linker unit may also function
in a bidetachable mode, enabling the release of the
corresponding carboxamides when photolysis is performed in
an aprotic solvent. Based on results from density functional
theory calculations, the present paper provides evidence of the
mechanism allowing for the control and selection between
these two competing reaction pathways. Finally, we have
demonstrated the use of the linker for the generation of a
pharmacologically relevant hydroxamate-functionalized natural
product-like DKP derivative in high purity.
(9) Il’ichev, Y. V.; Schworer, M. A.; Wirz, J. J. Am. Chem. Soc. 2004,
̈
126, 4581.
(10) (a) Walbert, S.; Pfleiderer, W.; Steiner, U. E. Helv. Chim. Acta
2001, 84, 1601. (b) Corrie, J. E. T.; Barth, A.; Ranjit, V.; Munasinghe,
N.; Trentham, D. R.; Hutter, M. C. J. Am. Chem. Soc. 2003, 125, 8546.
(11) Schmierer, T.; Bley, F.; Schaper, K.; Gilch, P. J. Photochem.
Photobiol., A 2011, 217, 363.
(12) Il’ichev, Y. V.; Wirz, J. J. Phys. Chem. A 2000, 104, 7856.
(13) Dunkin, I. R.; Gębicki, J.; Kiszka, M.; Sanín-Leira, D. J. Chem.
Soc., Perkin Trans. 2 2001, 1414.
(14) Lill, M. A.; Hutter, M. C.; Helms, V. J. Phys. Chem. A 2000, 104,
8283.
(15) Schaper, K.; Dommaschke, D.; Globisch, S.; Madani-
Mobarekeh, S. A. J. Inf. Rec. 2000, 25, 339.
(16) Paris, M.; Porcelloni, M.; Binaschi, M.; Fattori, D. J. Med. Chem.
2008, 51, 1505.
(17) Petersen, R.; Le Quement, S. T.; Nielsen, T. E. Angew. Chem.,
Int. Ed. 2014, 53, 11778.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.7b01386.
1
Experimental details; RP-HPLC, RP-UPLC, MS, H and
13C NMR data; computational details (PDF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: kaqvo@kemi.dtu.dk.
ORCID
Katrine Qvortrup: 0000-0003-3828-2069
Thomas E. Nielsen: 0000-0001-8700-1951
Notes
The authors declare no competing financial interest.
REFERENCES
■
(1) Minucci, S.; Pelicci, P. G. Nat. Rev. Cancer 2006, 6, 38.
Rappoport, Z.; Liebman, J. F. In The Chemistry of Hydroxylamines,
Oximes and Hydroxamic Acids; Rappoport, Z., Ed.; Wiley-Interscience:
Chichester, 2009; Vols. 1−2.
(2) Merrifield, B. R. J. Am. Chem. Soc. 1963, 85, 2149.
(3) Moos, W. H.; Green, G. D.; Pavia, M. R. Annu. Rep. Med. Chem.
1993, 28, 315.
(4) (a) Thouin, E.; Lubell, W. D. Tetrahedron Lett. 2000, 41, 457.
(b) Ho, C. Y.; Strobel, E.; Ralbovsky, J., Jr.; Galemmo, R. A. J. Org.
Chem. 2005, 70, 4873. (c) Nandurkar, N. S.; Petersen, R.; Qvortrup,
D
DOI: 10.1021/acs.orglett.7b01386
Org. Lett. XXXX, XXX, XXX−XXX