Tunable pH-sensitive 2-carboxybenzyl phosphoramidate cleavable linkers

Article abstract of DOI:10.1016/j.tetlet.2020.151650

We previously described a pH-sensitive phosphoramidate linker scaffold that can be tuned to release amine-containing drugs at various pH values. In these previous studies it was determined that the tunability of this linker was dependent upon the proximity of an acidic group (e.g., carboxylic acid or pyridinium). In this study, we confirmed that the tunability of pH-triggered amine-release was also dependent upon the pK_a of the proximal acidic group. A series of 2-carboxybenzyl phosphoramidates was prepared in which the pK_a of the proximal benzoic acid was predictably attenuated by substituents on the benzoate ring consistent with their σ-values.

Full text of DOI:10.1016/j.tetlet.2020.151650

Tetrahedron Letters 61 (2020) 151650
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Tunable pH-sensitive 2-carboxybenzyl phosphoramidate cleavable
linkers
⇑
Brian S. Backer, Cindy J. Choy, Austen L. Davis, Zachery S. Browne, Clifford E. Berkman
Department of Chemistry, Washington State University, Pullman, WA 99164-4630, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
We previously described a pH-sensitive phosphoramidate linker scaffold that can be tuned to release
amine-containing drugs at various pH values. In these previous studies it was determined that the tun-
ability of this linker was dependent upon the proximity of an acidic group (e.g., carboxylic acid or pyri-
dinium). In this study, we confirmed that the tunability of pH-triggered amine-release was also
dependent upon the pKa of the proximal acidic group. A series of 2-carboxybenzyl phosphoramidates
was prepared in which the pKa of the proximal benzoic acid was predictably attenuated by substituents
Received 16 December 2019
Revised 8 January 2020
Accepted 17 January 2020
Available online 25 January 2020
Keywords:
on the benzoate ring consistent with their
r-values.
Cleavable Linker
Phosphoramidate
Hammet effects
Hydrolysis
Ó 2020 Elsevier Ltd. All rights reserved.
Introduction
Cleavable linker technology has been actively pursued for the
past two decades [1–3]. Cleavable linkers have found use in the
context of solid phase synthesis of small-molecule libraries and
as strategy for asymmetric synthesis [4,5]. More recently, the fields
of chemical biology and biochemistry have employed cleavable
linkers in the development of antibody drug conjugates [6–10] or
triggered release of condition-specific turn-on dyes [11]. Despite
the diversity of applications for cleavable linkers, the common
requirement is highly predictable and controlled release of a
desired compound from the linker scaffold under specific
conditions.
We recently reported on two generations of phosphoramidate
scaffolds (Fig. 1) that can be triggered to release amine-bearing
drugs or self-immolating spacers at varying rates based upon the
proximity of a mildly acidic group (e.g., carboxylic acid or pyri-
dinium) [12,13]. Due to the tunable nature of these scaffolds, we
envisioned that these cleavable linker scaffolds can be tailored to
the specific pH requirements of controlled release applications.
Applications for such tunable pH-programmable cleavable scaf-
folds include antibody or small-molecule drug conjugates, drug-
eluting stents, prodrugs, turn-on dyes for intracellular trafficking,
or protecting groups in the chemical synthesis [1–3,14].
Fig. 1. 1st and 2nd generation phosphoramidate-based pH-sensitive cleavable
linkers.
carboxylic acid led to greater susceptibility of the PAN bond to
hydrolysis at mild pH conditions, proceeding through two possible
mechanisms [12,13]. We have hypothesized that in addition to
proximity, the pKa of the neighboring acidic group, and its ability
to protonate the leaving amine, could influence the hydrolytic sus-
ceptibility of the PAN bond. The focus of this study was aimed at
attenuating the hydrolytic stability of the phosphoramidate scaf-
fold through substituent effects [20,21] on a proximal benzoic acid
(Fig. 2). We hypothesized that under mildly acidic conditions
(pH = 5.5), in which the the carboxylic acid is more fully proto-
It is known that the hydrolysis of phosphoramidate PAN bonds
is pH dependent [15–18] and influenced by the presence of a prox-
imal carboxylate [19]. We demonstrated that the proximity of a
⇑
Corresponding author.
https://doi.org/10.1016/j.tetlet.2020.151650
0040-4039/Ó 2020 Elsevier Ltd. All rights reserved.

2
B.S. Backer et al. / Tetrahedron Letters 61 (2020) 151650
Based on the data in Table 1, electron withdrawing groups
(EWGs) led to increased stability of the phosphoramidate scaffold
toward hydrolysis of the PAN bond, while electron donating
groups (EDGs) led to a decrease in stability (Table 1). This trend
was inversely correlated to the
r
values at both pH 7.4 and pH
5.5, resulting in negative
q
values of À0.84 and À1.02 for pH 7.4
Fig. 2. The 2-carboxybenzyl phosphoramidate linker scaffold 1. X = EWGs or EDGs
at the meta or para positions of benzoic acid.
and pH 5.5, respectively (Fig. 4). These results suggest that the
pH-triggered release of amines from the phosphoramidate scaffold
is also tunable through attenuating the pKa of the neighboring
acidic group.
nated from electron-donating substituents (
r < 0) present, the
While the mechanism for the hydrolysis of our previous pH-
triggered phosphoramidate linker scaffolds [12,13] has remained
conjectural (Scheme 2), the presence of a transient intermediate
in the 31P spectra (Fig. 3) supports the 2-step intramolecular sub-
stitution mechanism as shown in Scheme 2b. This mechanism
would be consistent with the inverse Hammett correlation
observed for the phosphoramidates (Fig. 4) whereby electron
donating groups promote the nucleophilicity of the benzoyl car-
boxylate for the substitution at phosphorus, forming the acyl-
phosphate intermediate. Because 1c showed the greatest forma-
tion of this putative acyl-phosphate intermediate, it was used as
a representative example to further elucidate the mechanism of
hydrolysis.
rates for hydrolysis of the PAN bond would be greater.
The substituted 2-carboxybenzyl phosphoramidates were pre-
pared as outlined in Scheme 1. Phthalates 6a–6g were saponified
and converted to the corresponding esters 5a-5g. Fluorenlymethyl
ester-protected phosphites 3a–3g were generated from 4a–4g
with diphenyl phosphite. Routine Atherton-Todd conditions
[12,13] provided the protected 2-carboxybenzyl phosphorami-
dates 2a–2g. Global deprotection with LiOH yielded the desired
substituted 2-carboxybenzyl phosphoramidates 1a–1g.
Once prepared, the hydrolytic stability of 2-carboxybenzyl
phosphoramidates 1a–g was assessed at pH 5.5, and 7.4 by 31P
NMR [12,13]. To determine the pH-dependent rates of hydrolysis,
the peak areas for each parent phosphoramidate (1a–g), and its
respective hydrolytic product and intermediate, were determined
over 8 h with respect to an internal standard (triphenylphosphine
oxide, TPPO). The observed hydrolysis rates of phosphoramidates
1a–g followed first order kinetics as previously noted [12,13].
Half-lives and rate constants for 2-carboxybenzyl phosphorami-
dates 1a–1g, as well as control compounds 13, 16, and 19, at pH
5.5 and 7.4 are summarized in Table 1. Our interest in pH-triggered
cleavable linkers is their utility in tumor biomarker-targeted drug
conjugates. Therefore, the most relevant conditions for our applica-
tions are physiological (pH 7.4) and that of early endosomes (pH
5.5) following the internalization of the biomarker-targeted drug
conjugates. However, to more fully illustrate the pH-dependence
of the 2-carboxybenzyl phosphoramidate scaffold, the stability of
1c was monitored at pH 5.0, 5.5, 6.0. 6.5, and 7.4 as an example
(Fig. 3).
The hydrolysis of 1c was monitored by 31P and 1H NMR
(Fig. 5) in pD 6.0 buffer (acetic acid d4 : sodium acetate-d3),
and the rate of its hydrolysis in pD 6.0 buffer was found to be
similar to that at pH 6.5 [23]. The chemical shift of the transient
intermediate P’ (I) was consistent with those of previously
reported acyl-phosphates [24]. An expansion of the phenethy-
lamine region in the 1H NMR spectra (II and V) showed no sig-
nificant change for the distant b-protons (a) as expected,
however, the
bond, rapidly decayed to the
a
-protons (b), which were proximal to the PAN
-protons (b’) of released
a
phenethylamine. Expansion of the benzylic region in the 1H
NMR spectra (III and VI) showed that the benzylic protons (c)
of 1c rapidly decayed to the benzylic protons (c’) of an acyl-
phosphate intermediate before finally decaying to the benzylic
protons (c”) of the hydrolyzed product. Combined, the data from
Scheme 1. Synthesis of substituted 2-carboxybenzyl phosphoramidates.

B.S. Backer et al. / Tetrahedron Letters 61 (2020) 151650
3
Table 1
Sigma values, hydrolysis half-lives and rates of hydrolysis for substituted 2-carboxybenzyl phosphoramidates at pH 5.5 and pH 7.4.
Entry
Compound
r1
pH 5.5
t1/2 (h)
pH 5.5
pH 7.4
t1/2 (h)
pH 7.4
ka (sÀ1
)
ka (sÀ1
)
1a
0.71
1.09
2.67b
0.52b
0.22
0.16
176 Â 10-6
49.37
46.96
20.06
8.51
3.9 Â 10-6
1b
1c
1d
1e
0.66
0.23
0.00
À0.17
84 Â 10À6b
373 Â 10À6b
878 Â 10À6
1192 Â 10À6
4.1 Â 10-6
9.6 Â 10À6
22.6 Â 10À6
28.1 Â 10À6
6.85
1f
À0.27
0.26c
0.29b
28.74
676 Â 10À6c
657 Â 10À6b
7 Â 10À6
10.03c
14.05b
49.37
19.2 Â 10À6c
13.7 Â 10À6b
3.9 Â 10À6
1g
13
—
—
16
19
—
87.52
6.71
2 Â 10À6
481.35
7.29
0.4 Â 10-6
——
29 Â 10À6
26.4 Â 10À6
1
a
b
c
Substituent sigma values for benzoic acid [20].
Observed t1/2 at 37 °C reported in hours as determined by 31P NMR data normalized to 40 mM triphenylphosphine oxide standard.
Buffer containing 10% dimethyl sulfoxide for solubility.
Buffer containing 10% dimethylformamide for solubility.
the 31P and 1H NMR spectra for the hydrolysis of 1c support the
proposed 2-step intramolecular mechanism (Scheme 2B) over a
general acid catalyzed mechanism (Scheme 2A).
In summary, we have demonstrated that by attenuating the pKa
of the neighboring acidic group in the 2-carboxybenzyl phospho-
ramidate scaffold the controlled hydrolytic release of an amine
can be predictably tuned. These results are consistent with those
reported by Kirby and coworkers [25–28] who noted the effects
of neighboring ionizable groups on hydrolysis of phosphate mono,
-di, and -tri esters. Together with our prior findings, we conclude
that the rate of hydrolysis of such phosphoramidate scaffolds
depends upon three main factors: (1) the pKa of the departing
amine [12], (2) the proximity of the ionizable group [13], and (3)
the pKa of the proximal ionizable group. Combined, these aspects
allow for considerable control of this scaffold for the release of
amine payloads in the context of cleavable linkers, which can be
tuned to meet the specific demands of various controlled-release
applications.

4
B.S. Backer et al. / Tetrahedron Letters 61 (2020) 151650
Fig. 3. (A) Representative stacked 31P NMR spectra for the hydrolysis of 1c (9.9 ppm) at pH 6.0 revealing both the formation of an acyl-phosphate intermediate species (–
6.3 ppm), and the final hydrolytic product (2.1 ppm). (B) Compiled and fitted pH-dependent kinetic data for the area of 1c (blue), the intermediate (green), and the hydrolytic
product (red) normalized to a 40 mM triphenylphosphine oxide (TPPO) standard.
Fig. 4. Inverse Hammet correlation of 2-carboxybenzyl phosphoramidates at pH 7.4.

B.S. Backer et al. / Tetrahedron Letters 61 (2020) 151650
5
Scheme 2. Proposed mechanisms for the hydrolysis of 2-carboxybenzyl phosphoramidates [22].
Fig. 5. Hydrolysis of compound 1c in pD 6.0 buffer at 37 °C over 8 h. (I) 31P NMR: P represents the phosphorus in 1c. P’ represents the phosphorus of the intermediate. P”
represents the phosphorus in hydrolytic product. (II) Expansion of the 1H NMR spectra of the phenethylamine region: a represents the b-protons of 1c and the released
phenethylamine, b represents the
represents the 1H resonance for the benzylic protons of 1c, c’ represents the benzylic protons of the intermediate species, and c” represents the benzylic protons of the hydrolytic
product. (IV) Integrated areas of the 31P NMR spectra for 1c (blue), the acyl-phosphate intermediate (green), and the hydrolytic product (red). (V) Integrated areas of the 1
spectra for the -protons b of 1c (blue) and the released phenethylamine
-protons b’ (red). (VI) Integrated areas of the 1H spectra for the benzylic protons of 1c (c, blue), the
a-protons of 1c, and b’ represents a
-protons of the released phenethylamine. (III) Expansion of the 1H NMR spectra of the benzylic region: c
H
a
a
acyl-phosphate intermediate (c’, green), and the hydrolytic product (c”, red).
NMR Spectroscopy), Gerhard Munske (WSU Laboratory of Biotech-
nology and Bioanalysis) and Mr. Feyisola P. Olatunji.
Declaration of Competing Interest
The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared
to influence the work reported in this paper.
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.tetlet.2020.151650.
Acknowledgments
References
This work was supported in part by the National Institutes of
Health (Grant CA223121). The authors extend their gratitude for
technical assistance to G. Helms and W. Hiscox (WSU Center for
[1] M. Shadid, S. Bowlin, J. Bolleddula, Bioorganic, Med. Chem. 25 (2017) 2933–
2945.

6
B.S. Backer et al. / Tetrahedron Letters 61 (2020) 151650
[2] M. Pátek, M. Lebl, Biopolym. – Pept. Sci. Sect. 47 (1998) 353–363.
[15] J.D. Chanley, E. Feageson, J. Am. Chem. Soc. 85 (1963) 1181–1190.
[16] J.D. Chanley, E. Feageson, J. Am. Chem. Soc. 80 (1958) 2686–2691.
[17] M. Halmann, A. Lapidot, D. Samuel, J. Chem. Soc. (1963) 1299.
[18] E.J. Sampson, S.J. Benkovic, J. Am. Chem. Soc. 93 (1971) 4009–4016.
[19] E.P. Savel’ev, N.N. Preobrazhenskaya, Z.A. Shabarova, M.A. Prokof’ev, Chem.
Nat. Compd. 3 (1967) 101–104.
[3] P. Blaney, R. Grigg, V. Sridharan, Chem. Rev. 102 (2002) 2607–2624.
[4] K.S. Lam, M. Lebl, V. Krchnák, Chem. Rev. 97 (1997) 411–448.
[5] G. Guazzelli, M. Miller, D.J. Procter, Linker Units Cleaved by Radical Processes:
Cleavage of Carbon-Sulfur -Selenium, -Tellurium, -Oxygen, -Nitrogen and -
Carbon Linkers (2009).
[6] R.V.J. Chari, ChemInform (2008) 39.
[20] C. Hansch, A. Leo, R.W. Taft, Chem. Rev. 91 (1991) 165–195.
[21] C. Gardner Swain, E.C. Lupton, J. Am. Chem. Soc. 90 (1968) 4328–4337.
[22] S.J. Benkovic, P.A. Benkovic, J. Am. Chem. Soc. 89 (1967) 4714–4722.
[23] A.K. Covington, M. Paabo, R.A. Robinson, R.G. Bates, Anal. Chem. 40 (1968)
700–706.
[24] A. Okruszek, W.J. Stec, Tetrahedron Lett. 23 (1982) 5203–5206.
[25] A. Kirby, J. Gesser, F. Hollfelder, J. Priebe, F. Nome, Can. J. Chem. 83 (2005)
1629–1636.
[7] M.J.M. Hinrichs, R. Dixit, AAPS J. 17 (2015) 1055–1064.
[8] A. Kumar, T. Mastren, B. Wang, J.T. Hsieh, G. Hao, X. Sun, Bioconjug. Chem. 27
(2016) 1681–1689.
[9] S.A. Kularatne, C. Venkatesh, H.K.R. Santhapuram, K. Wang, B. Vaitilingam, W.
A. Henne, P.S. Low, J. Med. Chem. 53 (2010) 7767–7777.
[10] S. Jayaprakash, X. Wang, W.D. Heston, A.P. Kozikowski, ChemMedChem 1
(2006) 299–302.
[11] A.K. Steiger, S. Pardue, C.G. Kevil, M.D. Pluth, J. Am. Chem. Soc. 138 (2016)
7256–7259.
[26] A.J. Kirby, J.R. Mora, F. Nome, Biochim. Biophys. Acta - Proteins Proteomics
2013 (1834) 454–463.
[12] C.J. Choy, J.J. Geruntho, A.L. Davis, C.E. Berkman, Bioconjug. Chem. 27 (2016)
824–830.
[27] A.J. Kirby, D.W. Tondo, M. Medeiros, B.S. Souza, J.P. Priebe, M.F. Lima, F. Nome,
J. Am. Chem. Soc. 131 (2009) 2023–2028.
[13] C.J. Choy, C.R. Ley, A.L. Davis, B.S. Backer, J.J. Geruntho, B.H. Clowers, C.E.
Berkman, Bioconjug. Chem. 27 (2016) 2206–2213.
[28] A.J. Kirby, M.F. Lima, D. Da Silva, C.D. Roussev, F. Nome, J. Am. Chem. Soc. 128
(2006) 16944–16952.
[14] J. Zhou, H. Ma, Chem. Sci. 7 (2016) 6309–6315.