Enzyme-triggered gelation: Targeting proteases with internal cleavage sites

Article abstract of DOI:10.1039/c3cc48132h

A generalizable method for detecting protease activity via gelation is described. A recognition sequence is used to target the protease of interest while a second protease is used to remove the residual residues from the gelator scaffold. Using this appro

Full text of DOI:10.1039/c3cc48132h

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Enzyme-triggered gelation: targeting proteases
with internal cleavage sites†
Steven C. Bremmer,ab Anne J. McNeil*a and Matthew B. Soellner*b
Cite this: Chem. Commun., 2014,
50, 1691
Received 23rd October 2013,
Accepted 18th December 2013
DOI: 10.1039/c3cc48132h
www.rsc.org/chemcomm
A generalizable method for detecting protease activity via gelation
is described. A recognition sequence is used to target the protease
of interest while a second protease is used to remove the residual
residues from the gelator scaffold. Using this approach, selective
assays for both MMP-9 and PSA are demonstrated.
Because both overexpressed and overactive enzymes correlate with
disease, simple visual assays that respond to enzymatic activity are in
high demand.1 Colorimetric and fluorimetric assays dominate the
field,2 but the analysis can be challenging in the complex medium of
biological fluids. An alternative approach is to use an enzyme to
trigger a phase transition (e.g., solution-to-gel) by cleaving a covalent
bond.3 Such assays can be performed in media as complex as blood
or cells, and the results are unambiguously detected without any
Scheme 1 General design strategy for detecting proteases via gelation
using recognition sequences and aminopeptidase M.
instrumentation. In 2004, Xu and coworkers reported the first the gelator scaffold can disrupt the gelation process,7 and as
enzyme-mediated gelation of small molecules.4 Since then, several a consequence, a new gelator may need to be discovered for
different enzymes have been used to trigger gelation.5
each enzyme.
Though promising, only a handful of reported enzyme-
We reasoned that an aminopeptidase, which nondiscriminately
triggered gelations are selective for a single enzyme.6 Most of these cleaves amino acids off the N-terminus of a peptide,8 could be used
selective systems use peptide-based recognition sequences. Enzy- to remove the leftover prime-side recognition sequence residues
matic cleavage at the N-terminal side of the recognition sequence from the gelator scaffold (Scheme 1). As a consequence, the same
serves to completely separate this sequence from the gelator and gelator can be used for detecting any protease by simply varying the
trigger gel formation. In 2012 we reported a highly selective and recognition sequence. Using this approach, we describe herein
generalizable method for detecting protease activity based on this the development of selective assays for matrix metalloprotease-9
approach.6a While this method is suitable for most serine and (MMP-9) and prostate specific antigen (PSA) using a single gelator
cysteine proteases, it is not suitable for aspartyl proteases and scaffold and aminopeptidase M (AP-M). Aminopeptidase M (AP-M)
metalloproteinases because these enzymes require specific residues was selected because it will remove residues from nearly every
on both sides of the scissile bond. Targeting these enzymes is more sequence of natural amino acids.8
challenging because the leftover peptide residues attached to
The sensing scheme is comprised of three key components:
a solubility factor (SF), recognition sequence, and gelator. The
solubility factor9 increases the solubility and ensures that the
recognition sequence is accessible to the target enzyme in
solution. The solubility factor also prevents premature cleavage of
the substrate by AP-M because it contains a non-natural oligo-
ethylene glycol unit at the N-terminus that is not recognized by
AP-M. The recognition sequence contains a short peptide
segment that is easily interchangeable and optimized for each
target enzyme. Once this sequence is cleaved by the protease,
a Department of Chemistry and Macromolecular Science and Engineering Program,
University of Michigan, 930 North University Avenue, Ann Arbor, Michigan,
48109-1055, USA. E-mail: ajmcneil@umich.edu
b Departments of Chemistry and Medicinal Chemistry, University of Michigan,
930 North University Avenue, Ann Arbor, Michigan, 48109-1055, USA.
E-mail: soellner@med.umich.edu
† Electronic supplementary information (ESI) available: Synthetic procedures;
peptide characterization; rheology data; gelation data; LC-MS data; control
experiments. See DOI: 10.1039/c3cc48132h
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 1691--1693 | 1691

View Article Online
Communication
ChemComm
the AP-M will non-discriminately cleave the remaining resi- This suggests that our MMP-9 sensing gelator could readily
dues. To prevent the AP-M from cleaving the peptide-based discriminate between healthy patients and patients with
gelator, a non-natural p-aminobenzamide (PABA) unit was aggressive cancers.
included at the N-terminus. Gelator 16a was selected for these
To demonstrate the generality of this approach, we targeted
studies because it forms stable gels in buffer (50 mM HEPES, a different protease: prostate-specific antigen (PSA). PSA is a serine
200 mM NaCl, 10 mM CaCl2, 7.5% DMSO, pH = 7.5) with a low protease, however, it was selected because the prime side residues
critical gel concentration (cgc, 1.7 mM).
have a dramatic influence over the reactivity.16 In addition, PSA is an
important biomarker for prostate cancer.17 Although overexpression
of active PSA correlates with cancer prognosis, current screening
methods measure total PSA concentration, not the fraction that is
active.18 While total PSA levels as a prostate cancer diagnostic is
controversial,18a recent studies have demonstrated that active PSA
enzyme in the blood is highly correlative to aggressive prostate
cancer.19 Hence, an assay based on activity is highly desirable.
A substrate containing an optimized PSA recognition sequence
(HSAKFY–SG)16a was synthesized and treated with PSA and
AP-M in buffer. Within 24 h, a stable gel was formed (ESI,†
Fig. S14). A control experiment lacking PSA showed no gelation
over the same time (ESI,† Fig. S14). Similarly, if the prime side
residues were missing (HSAKFY) from the substrate, no reaction
was observed by LC-MS even after 24 h (ESI,† Fig. S13).
Because PSA activity is rate-limiting, we hypothesized that
the response time could be shortened by adding the cleaved
substrate (SG-1) to the reaction mixture. Although SG-1 was
largely insoluble, adding a lysine (i.e., SGK-1) significantly improved
its solubility. This substrate is rapidly converted to 1 in situ using
AP-M. Gratifyingly, including SGK-1 (1 mM) reduced the time-to-
gelation from 24 h to 7.5 h (Fig. 2). The detection limit of this PSA
sensor is 32 nM of active enzyme, which is above the physiological
levels of active PSA found in the blood of prostate cancer patients
(B0.1 nM active enzyme).20 However, high levels of active PSA are
found in the extracellular fluid of prostate cancers (1.8 mM).18b On
this basis, our PSA sensing gelator could be used as an aid in biopsy
diagnosis. Our detection limit must be lowered considerably to be
useful as a diagnostic for measuring active PSA levels in blood.20 In
principle, a lower detection limit can be achieved if a gelator with a
lower cgc is used. Alternatively, a different recognition sequence
with higher activity towards PSA would lead to lower detection
limits. Additionally, patients with prostate cancer have high levels
We first targeted MMP-910 because it has stringent requirements
about which residues are present on both sides of the cleavage site,11
and it has been implicated in several pathological conditions,
including cancer metastasis12 and arthritis.13 A scaffold containing
an optimized recognition sequence for MMP-9 (GPKG–LKGA)14 was
treated with both MMP-9 and AP-M in buffer. Gratifyingly, a stable
gel was observed within 7 h (Fig. 1A and Fig. S10, ESI†). The
importance of prime side residues was demonstrated when a
recognition sequence lacking these residues showed no con-
version over the same time (Fig. 1B and Fig. S10, ESI†).
Because, in principle, the assay depends on the concentration
and activity of both enzymes, identifying conditions wherein the
target enzyme (MMP-9) is rate-limiting is desired, ensuring a low
detection limit for the target enzyme. LC-MS studies were used to
monitor conversion of LKGA-1 to 1 mediated by AP-M over time.
These studies revealed that approximately 45% conversion is
reached within 160 min, at which point gelation occurs (ESI,†
Fig. S17). This rate is significantly faster than the 7 h needed for
conversion of parent scaffold: SF-GPKG-LKGA-1 (ESI,† Fig. S10).
Thus, the rate-limiting step must be the initial MMP-9 triggered
cleavage. Overall, the detection limit is 50 nM under these optimized
conditions, which is within the levels present in cancer patients with
elevated levels of circulating MMP-9 (71 Æ 60 nM for non-small
cell lung cancer patients, 36 Æ 13 nM for healthy subjects).15
Fig. 1 (A) An opaque gel was observed within 7 h of adding MMP-9
(50 nM) to a buffer solution containing AP-M (1.2 mM) and the gelator
scaffold with the optimized recognition sequence. (B) No gel was observed
when the prime side residues were not included under otherwise identical
conditions. Pictures were taken after 24 h.
Fig. 2 (A) An opaque gel was observed within 7.5 h when PSA (1.6 mM) was
added to a buffer solution containing AP-M (1.2 mM), SGK-1 (1 mM) and the
gelator scaffold with the optimized recognition sequence. No gel was
observed when either (B) AP-M or (C) PSA was excluded, under otherwise
identical conditions.
1692 | Chem. Commun., 2014, 50, 1691--1693
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
V. Glattauer, J. White, P. J. Pasic, A. B. Sorensen, L. Waddington,
K. M. McLean, P. D. Currie and P. G. Hartley, Biomaterials, 2011, 32,
5304–5310; (e) M. J. Webber, C. J. Newcomb, R. Bitton and S. I. Stupp,
Soft Matter, 2011, 7, 9665–9672.
of proPSA, an inactive precursor of PSA, that could be converted
into active PSA in situ (B30 nM proPSA, compared to less than
1 nM for healthy individuals)20 using an additional enzyme.21
In summary, a generalizable method for detecting protease
activity using gelation is described. We demonstrated that
aminopeptidase M, which nondiscriminately cleaves amino
acids off the N-terminus of a peptide, can be used to remove
the residual components of a recognition sequence from the gel
scaffold. As a consequence, the same gelator can be used to detect
different proteases by simply switching the recognition sequences.
Using this approach, we were able to detect both MMP-9 and PSA
at physiological concentrations. Given that these proteases are
validated biomarkers for cancer, we believe that with further devel-
opment this simple visual assay may be useful for early detection.
Our future efforts will focus on accelerating the time for gelation by
employing additives (i.e., growth promoters),22 designing alternative
gelators with lower critical gel concentrations, and improving the
activity of the recognition sequences. Overall, the system described
herein is significantly more versatile than other enzyme-triggered
gelations in the literature. As a result, we anticipate that this method
will be utilized for the detection and diagnosis of other disease-
relevant proteases.
6 Few systems are selective for a single enzyme. For reference, see:
(a) S. C. Bremmer, J. Chen, A. J. McNeil and M. B. Soellner, Chem.
Commun., 2012, 48, 5482–5484; (b) D. Koda, T. Maruyama,
N. Minakuchi, K. Nakashima and M. Goto, Chem. Commun., 2010,
46, 979–981; (c) Z. Yang, P.-L. Ho, G. Liang, K. H. Chow, Q. Wang,
Y. Cao, Z. Guo and B. Xu, J. Am. Chem. Soc., 2007, 129, 266–267.
7 Seemingly minor structural changes can convert peptide-based gelators
into nongelators. For recent examples, see: (a) K. A. Houton, K. L. Morris,
L. Chen, M. Schmidtmann, J. T. A. Jones, L. C. Serpell, G. O. Lloyd and
D. J. Adams, Langmuir, 2012, 28, 9797–9806; (b) M. L. Muro-Small, J. Chen
and A. J. McNeil, Langmuir, 2011, 27, 13248–13253; (c) H. Wang, C. Yang,
M. Tan, L. Wang, D. Kong and Z. Yang, Soft Matter, 2011, 7, 3897–3905;
(d) D. M. Ryan, T. M. Doran, S. B. Anderson and B. L. Nilsson, Langmuir,
2011, 27, 4029–4039.
8 M. Drag, M. Bogyo, J. A. Ellman and G. S. Salvesen, J. Biol. Chem.,
2009, 285, 3310–3318.
9 The solubility factor structure is as follows: PEG4-Arg-Arg-Ser-Pro-.
10 For a related example of MMP-9 triggered gelation, see: Z. Yang,
M. Ma and B. Xu, Soft Matter, 2009, 5, 2546–2548, The detection
limit of this gelation system for MMP-9 is 168 nM, which is above
the physiological levels of MMP-9.
11 S. J. Kridel, E. Chen, L. P. Kotra, E. W. Howard, S. Mobashery and
J. W. Smith, J. Biol. Chem., 2001, 276, 20572–20578.
12 (a) Q. Yu and I. Stamenkovic, Genes Dev., 1999, 13, 35–48; (b) E. I.
Deryugina and J. P. Quigley, Cancer Metastasis Rev., 2006, 25, 9–34.
We thank Maureen Corrielus for help with peptide synthesis of 13 B. L. Gruber, D. Sorbi, D. L. French, M. J. Marchese, G. J. Nuovo,
R. R. Kew and L. A. Arbeit, Clin. Immunol. Immunopathol., 1996, 78,
161–171.
14 A. Prudova, U. Keller, G. S. Butler and C. M. Overall, Mol. Cell.
PSA substrates. We thank the Arnold and Mabel Beckman Founda-
tion (Young Investigator Award to A.J.M.) and National Institutes of
Health (R00RR024366 to M.B.S.) for support of this work.
Proteomics, 2010, 9, 894–911.
15 T. Iizasa, T. Fujisawa, M. Suzuki, S.-I. Motohashi, K. Yasufuku,
T. Yasukawa, M. Baba and M. Shiba, Clin. Cancer Res., 1999, 5, 149–153.
16 (a) G. S. Coombs, R. C. Bergstrom, J. L. Pellequer, S. I. Baker,
M. Navre, M. M. Smith, J. A. Tainer, E. L. Madison and D. R. Corey,
Chem. Biol., 1998, 5, 475–488; (b) M. Debela, V. Magdolen,
N. Schechter, M. Valachova, F. Lottspeich, C. S. Craik, Y. Choe,
W. Bode and P. Goettig, J. Biol. Chem., 2006, 281, 25678–25688.
17 (a) S. P. Balk, Y. J. Ko and G. J. Bubey, J. Clin. Oncol., 2003, 21,
383–391; (b) A. M. LeBeau, M. Kostova, C. S. Craik and
S. R. Denmeade, Biol. Chem., 2010, 391, 333–343.
Notes and references
1 For recent reviews, see: (a) S. B. Kim, M. Hattori and T. Ozawa, Int.
J. Mol. Sci., 2012, 13, 16986–17005; (b) A. Razgulin, N. Ma and J. Rao,
Chem. Soc. Rev., 2011, 40, 4186–4216.
2 For recent reviews, see: (a) D. Tang, Y. Cui and G. Chen, Analyst,
2013, 138, 981–990; (b) X. Xie, W. Xu and X. Liu, Acc. Chem. Res.,
2012, 45, 1511–1520; (c) P. Tallury, A. Malhotra, L. M. Byrne and
S. Santra, Adv. Drug Delivery Rev., 2010, 62, 424–437.
18 For a recent review on PSA screening: (a) J. Lilja, D. Ulmert and
A. J. Vickers, Nat. Rev. Cancer, 2008, 8, 268–278; (b) S. R. Denmeade,
L. J. Sokoll, D. W. Chan, S. R. Khan and J. T. Isaacs, Prostate, 2001,
48, 1–6.
19 (a) P. Wu, L. Zhu, U. H. Stenman and J. Leinonen, Clin. Chem., 2004,
50, 125–129; (b) P. Wu, U. H. Stenman, M. Pakkala, A. Narvanen and
J. Leinonen, Prostate, 2004, 58, 345–353.
3 For recent reviews, see: (a) Y. Zhang, Y. Kuang, Y. Gao and B. Xu,
Langmuir, 2011, 27, 529–537; (b) R. J. Williams, R. J. Mart and
R. V. Ulijn, Biopolymers, 2010, 94, 107–117; (c) Y. Gao, Z. Yang,
Y. Kuang, M.-L. Ma, J. Li, F. Zhao and B. Xu, Biopolymers, 2010, 94,
19–31.
4 Z. Yang, H. Gu, D. Fu, P. Gao, J. K. Lam and B. Xu, Adv. Mater., 2004,
16, 1440–1444.
20 P. Niemela, J. Lovgren, M. Karp, H. Lilja and K. Pettersson, Clin.
Chem., 2002, 48, 1257–1264.
21 T. K. Takayama, B. A. McMullen, P. S. Nelson, M. Matsumura and
K. Fujikawa, Biochemistry, 2001, 40, 15341–15348.
22 Y. J. Adhia, T. H. Schloemer, M. T. Perez and A. J. McNeil, Soft
Matter, 2012, 8, 430–434.
5 For recent examples, see: (a) J. Li, Y. Gao, Y. Kuang, J. Shi, X. Du,
J. Zhou, H. Wang, Z. Yang and B. Xu, J. Am. Chem. Soc., 2013, 135,
9907–9914; (b) J.-B. Guilbaud, C. Rochas, A. F. Miller and A. Saiani,
Biomacromolecules, 2013, 14, 1403–1411; (c) M. Hughes, L. S.
Birchall, K. Zuberi, L. A. Aitken, S. Debnath, N. Javid and R. V. Ulijn,
Soft Matter, 2012, 8, 11565–11574; (d) R. J. Williams, T. E. Hall,
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 1691--1693 | 1693