Human plasma-mediated hypoxic activation of indolequinone-based naloxone pro-drugs

Article abstract of DOI:10.1016/j.bmcl.2009.07.061

Hypoxia is known to occur in tissues in response to narcotic analgesic therapy using as a result of respiratory depression. The aim of this study was to synthesize a narcotic antagonist pro-drug that can be activated by tissue hypoxia to prevent the damage associated with respiratory depression. We synthesized three different pro-drugs of the narcotic antagonist naloxone utilizing indolequinone as the hypoxia-sensitive moiety. The indolequinone structure in the pro-drugs was designed to have an open reactive point at the N-1 position offering the possibility of further conjugation with macromolecules to modify the bio-availability of these pro-drugs in vivo. A pro-drug (labeled 1) where naloxone and the indolequinone moiety were linked through a carbonate bond was rapidly hydrolyzed in phosphate buffered saline. However, two additional pro-drugs (labeled 2 and 3) having carbamate linkers were stable in phosphate buffered saline for 24 h. The reductive release of naloxone from the pro-drugs was achieved in the presence of the bio-reductive enzyme DT-Diaphorase, with about 80% release occurring from the two pro-drugs in 24 h. More than 99% of naloxone was released from pro-drug 2 in 30% human plasma, however the release only occurred under hypoxic conditions. This system provides a potential means for feedback control to counter critical respiratory depression induced by narcotic analgesics.

Full text of DOI:10.1016/j.bmcl.2009.07.061

Bioorganic & Medicinal Chemistry Letters 19 (2009) 5016–5020
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Human plasma-mediated hypoxic activation of indolequinone-based
naloxone pro-drugs
Baohua Huang, Shengzhuang Tang, Ankur Desai, Xue-min Cheng, Alina Kotlyar, Abraham Van Der Spek,
*
Thommey P. Thomas, James R. Baker Jr.
Michigan Nanotechnology Institute for Medicine and Biological Sciences, 9220 MSRB III, Box 0648, University of Michigan, Ann Arbor, MI 48109, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Hypoxia is known to occur in tissues in response to narcotic analgesic therapy using as a result of respi-
ratory depression. The aim of this study was to synthesize a narcotic antagonist pro-drug that can be acti-
vated by tissue hypoxia to prevent the damage associated with respiratory depression. We synthesized
three different pro-drugs of the narcotic antagonist naloxone utilizing indolequinone as the hypoxia-sen-
sitive moiety. The indolequinone structure in the pro-drugs was designed to have an open reactive point
at the N-1 position offering the possibility of further conjugation with macromolecules to modify the bio-
availability of these pro-drugs in vivo. A pro-drug (labeled 1) where naloxone and the indolequinone moi-
ety were linked through a carbonate bond was rapidly hydrolyzed in phosphate buffered saline. However,
two additional pro-drugs (labeled 2 and 3) having carbamate linkers were stable in phosphate buffered
saline for 24 h. The reductive release of naloxone from the pro-drugs was achieved in the presence of the
bio-reductive enzyme DT-Diaphorase, with about 80% release occurring from the two pro-drugs in 24 h.
More than 99% of naloxone was released from pro-drug 2 in 30% human plasma, however the release only
occurred under hypoxic conditions. This system provides a potential means for feedback control to coun-
ter critical respiratory depression induced by narcotic analgesics.
Received 14 May 2009
Revised 7 July 2009
Accepted 9 July 2009
Available online 12 July 2009
Keywords:
Hypoxia
Selective drug release
Indolequinone
Naloxone
Human plasma
Ó 2009 Elsevier Ltd. All rights reserved.
There has been continued interest in the exploitation of indol-
equinones as a method of hypoxia-targeted drug delivery.1,2 Hyp-
oxic cells and tissues have been documented to occur in
cardiovascular diseases,3 wound healing,4 and neurodegenerative,5
inflammatory6 and gastrointestinal disorders.7 Hypoxic cells are
generally characterized by low oxygen tension, low pH, low nutri-
ent levels, and over-expression of angiogenic factors leading to cell
death and permanent tissue damage. This has led to the develop-
ment of bioreductive indolequinone-based pro-drugs that can
potentially target hypoxic cells.
Narcotic analgesics are the most effective pain relief medica-
tions for acute traumatic injury. However, narcotics have serious
side effects, such as hypoxia due to respiratory depression and
hypotension due to vasodilation.8 These adverse effects may cause
cardio-respiratory collapse in the setting of traumatic injury,
potentially leading to death. One approach to counteract adverse
narcotic effects would be to use low-level narcotic antagonist feed-
back that prevents respiratory failure while maintaining analgesia.
Applying an agonist-antagonist based strategy would involve a
combination therapeutic containing both a narcotic agonist to
achieve analgesia and a narcotic antagonist whose release could
be activated in the presence of tissue hypoxia to reverse respira-
tory depression. Specifically, the feedback component would need
to rapidly release a short-acting narcotic antagonist when a low
oxygen pressure is present. Naloxone is an effective, short-acting
opioid antagonist that can reverse a range of morphine’s effects,
including analgesia and respiratory depression.9 Although mor-
phine’s analgesic and respiratory effects overlap, low-dose infu-
sions of naloxone can reverse morphine’s respiratory depression,
while its analgesic effect can be relatively unaffected allowing pain
relief to continue.10 We report here the synthesis and reductively-
triggered drug release studies of naloxone pro-drugs bearing an
indolequinone structure.
The structures of indolequinone bearing naloxone pro-drugs 1,
2, and 3 are shown in Scheme 3. The relationship between pro-
drug structures and drug release kinetics has been previously stud-
ied.1 Generally, compounds with electron-withdrawing groups at
the indole 3-position were among the best substrates, and that
groups larger than methyl at N-1 position are clearly tolerated by
DT Diaphorase enzyme for bioreduction. A study on the reductively
triggered release of a variety of oxygen-based and sulfur-based
exiting groups from their 4,7-dioxoindole-3-methyl derivatives
was carried out.11 It has also been demonstrated that upon substi-
tution with a potential leaving group at the 3-position of the indol-
equinone moiety is a viable means for bioreductive drug delivery.12
* Corresponding author. Tel.: +1 734 647 2777.
E-mail address: jbakerjr@med.umich.edu (J.R. Baker Jr.).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.07.061

B. Huang et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5016–5020
5017
N
N
O
O
OH
OH
O
N
O
N
OH
OH
N
O
N
O
O
O
MeO
MeO
O
O
O
H
H
O
O
N
N
2e or
R1
R1
O
2x1e
reduction
2, 3
H2O
OH
OH
OH
O
N
MeO
O
+
CO2
N
N
OH
+ Naloxone
HN
N
N
O
R1
O
H
O
Scheme 1. Proposed mechanism of reductively triggered release of naloxone from pro-drugs 2 and 3.
In the current study, we used the indolequinone structure with
2-methyl 5-methoxy groups that has shown good drug release
properties. A proposed chemical structure for reductively activated
release of naloxone is shown in Scheme 1. When the pro-drugs 2,
and 3 are reduced to dihydroxyindole, the electron density at the
indole nitrogen increases drastically, triggering expulsion of the
exiting group (in this case, naloxone), in a reverse-Michael-like
process, releasing the drug coupled to a small spacer.13,14
The synthesis of the indolequinone linker compounds 11 and 12
is demonstrated in Scheme 2. The methyl analogue of the indol-
equinone structure has been synthesized by Naylor et al.11 Rather
than putting a methyl group at the N-1 position, we used an eth-
oxycarbonyl methyl functional group at the N-1 position. This syn-
thetic design allows this position to serve as an open reaction site
to facilitate conjugation of long chain tethers or nanoparticles. The
temporary functional group at N-1 position must be stable
throughout the synthetic process, yet must be able to be finally
modified and coupled with tethers. The pro-drugs can then be con-
jugated through the tethers’ structure to other macromolecules,
such as a dendrimer nanoparticle, to enhance its retention time
in vivo that can aid in situations that need prolong drug release.
Commercially available 5-methoxy-2-methylindole was first trea-
ted with Vilsmeier reagents to give 3-formyl compound 4 with
excellent yield. Alkylation of position 1 was then carried out with
sodium hydride and ethyl bromo-acetate. Nitration at the desired
4-position could only be achieved at a low reaction concentration
and with a large excess of nitric acid. However, a 6-nitro isomer
was presented at a low ratio (about 15%). The mixture had poor
solubility in organic solvents, making it impossible to purify
through chromatography, and when reduced with Sn/HCl gave a
mixture of 4- and 6-amine products. The desired 4-amino com-
pound 7 was then purified with chromatography. Fremy’s salt
was used to covert the aromatic ring to a quinone structure with
quantitative yield. The crucial step is the reduction of the 3-carbox-
aldehyde to a hydroxyl group because there are several other sites
can also be reduced in the structure, especially the ethyl ester. The
NO2
CHO
CHO
Me
CHO
Me
MeO
MeO
MeO
MeO
MeO
(iii)
(i)
(ii)
Me
Me
N
H
N
H
N
N
CO2Et
CO2Et
5
4
6
O
O
O
OH
Me
O
CHO
Me
NH2
OH
CHO
MeO
(vi)
(v)
MeO
(iv)
MeO
(vii)
Me
N
Me
N
N
N
CO2H
O
CO2Et
O
CO2Et
CO2Et
10
7
8
9
O
OH
Me
MeO
(viii)
N
R1
O
O
11 R1=
12 R1=
NHBoc
O
N
H
2
O
N3
O
N
H
3
Scheme 2. Synthesis of indolequinone linker. Reagents and conditions: (i) DMF/POCl3; (ii) NaH/DMF/BrCH2CO2Et; (iii) AcOH/HNO3; (iv) Sn/HCl/EtOH, reflux; (v) Fremy’s salt/
acetone; (vi) NaBH4/MeOH/THF; (vii) LiOH/MeOH/THF; (viii) mono-Bocdiamine or azido amine, HATU, Et3N, DMF.

5018
B. Huang et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5016–5020
N
O
O
OH
O
N
O
O
O
NH
N
O
O
N
MeO
O
MeO
O
(iii)
Me
H
O
N
N
(i)
O
OH
Me
R1
R1
O
R1=
O
O
MeO
R1=
O
NHBoc
N3
O
2
R1=
R1=
N
13
NHBoc
N3
O
N
H
N
2
H
2
R1
O
O
O
14 R1=
N
H
O
(ii)
O
3
N
H
3
3
11
NHBoc
N3
O
N
H
N
OH
2
O
O
O
O
O
12 R1=
O
O
H
N
H
MeO
3
O
N
R1
O
O
NHBoc
R1=
O
1
N
H
2
Scheme 3. Synthesis of pro-drugs 1, 2 and 3. Reagents and conditions: (i) (a) 4-nitrophenyl chloroformate/Et3N; (b) excess N,N0-methylethylenediamine; (ii) 4-nitrophenyl
chloroformate activated naloxone, Et3N/CH2Cl2; (iii) (a) Triphosgene; (b) Naloxone, triethylamine/dichloromethane.
selective reduction of the aldehyde could only be achieved by using
5 equiv of NaBH4 at 0 °C over 8 min, and any elevation of temper-
ature or prolonging the reaction time decreased the yield drasti-
cally. The yield of the reduction reaction was around 60%. After
the hydrolysis of the ester with 1 equiv of LiOH, compound 10
was then coupled with a mono-Boc protected diamine to produce
the indolequinone linker 11. An azido amine tether was also
coupled to 10 to give indolequinone linker 12 with a tether amena-
ble to ‘click’ chemistry via an alkyne functional group through
1,3-Huisgen cyclic addition.
(250 ꢀ 4.6 mm, 300 Å) connected to
a
C5 guard column
(4 ꢀ 3 mm). The mobile phase used for elution was a linear gradi-
ent beginning with 90:10 (v/v) water/acetonitrile at a flow rate of
1 mL/min. Trifluoroacetic acid at a 0.14 wt % concentration in
water as well as in acetonitrile was used as a counter ion. The data
were collected at kmax = 280 nm. The HPLC profiles of the pro-drugs
and free naloxone are shown in Figure 1. The retention time of free
naloxone is at 8.5 min, and the retention times for pro-drug 2 and 3
are 17.4 and 16.6 min, respectively. To quantify the released drug,
a calibration curve for free naloxone was generated, and the detec-
The synthesis of all the pro-drugs is shown in Scheme 3.
Pro-drug 1 was first prepared from compound 11 which has a car-
bonate bond between the indolequinone structure and naloxone,
however, the carbonate structure is unstable in phosphate buffer
saline (PBS) which indicates that it would not be stable in blood.
This would lead to no-specific naloxone release. To enhance the
stability of the pro-drugs, a small spacer was incorporated between
the indolequinone structure and the drug through two carbamate
bonds. The carbamate bonds in pro-drugs 2 and 3 are much more
stable so undesired release of naloxone can be avoided. In the
mean time, the spacer can reduce the steric interference between
the drug and the linker and thus enhance enzyme access to the
cleavage sit and facilitate hypoxic release of the drug. To achieve
this, the activated 4-nitrophenyl carbonate of compounds 11 and
12 were condensed with an excess of N,N0-methylethylene diamine
to give the compounds 13 and 14. Despite only a 3 equiv of dia-
mine being used in the reactions there was no significant dimeriza-
tion. The secondary amine at compounds 13 and 14 was then
converted to a carbomyl chloride with triphosgene, which was
subsequently reacted with naloxone to form the naloxone pro-
drugs 2 and 3. All compounds are characterized with NMR and
MS spectra (see Supplementary data).
tion limit of free naloxone using this technique is 8 ng/lL. The pro-
drugs were dissolved in DMSO and then diluted with PBS buffer to
make the working stock solutions at a concentration of 100 mM for
the stability and release studies. Stability of the pro-drugs under
non-hypoxic conditions was tested by incubation in PBS buffer.
The HPLC results showed that pro-drug 1 was hydrolyzed within
2 h (data not shown). However, pro-drug 2 and 3 were stable in
PBS buffer with no free drug detected after 24 h at 37 °C.
0.39
0.29
Pro-drug 2
Pro-drug 3
0.19
Free naloxone
0.09
The hypoxia-induced drug release was characterized and mon-
itored using a Reverse Phase High Performance Liquid Chromatog-
raphy (RP-HPLC). Briefly, HPLC analysis was carried out on a
Waters Delta 600 HPLC system equipped with a Waters 2996 pho-
todiode array (PDA) detector, a Waters 717 Plus auto sampler, and
a Waters Fraction collector III. The instrument is controlled by
Empower 2 software. For analysis, aliquots were taken from the
incubated samples and run on a C5 silica-based RP-HPLC column
-0.01
0
5
10
15
20
25
30
Time (min.)
Figure 1. A HPLC chromatogram showing profiles of naloxone and prodrugs. Free
naloxone, pro-drug 2 and 3 eluted at 8.5, 17.4 and 16.6 min, respectively.

B. Huang et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5016–5020
5019
0.25
0.2
0.15
0.1
0.05
0
100%
80%
60%
40%
20%
0%
With
DT-Diaphorase
PBS buffer
Hypoxia
7
8
9
10
Normoxia
0
5
10
15
20
25
30
Time (hour)
0
5
10
15
Time (min.)
20
25
30
Figure 4. HPLC time course of naloxone release for pro-drug 2 under hypoxic and
normoxia conditions in 30% human plasma.
Figure 2. HPLC of naloxone release for pro-drug
2
in the presence of DT-
Diaphorase.
installed inside the chamber to determine oxygen pressure
levels. Human blood was collected in Citrate Dextrose Solution
anticoagulant. The blood was centrifuged at 4 °C (7000g for
5 min) to collect the plasma. Pro-drug 2 (final concentration,
The release of naloxone from the pro-drugs was first examined
in the presence of the enzyme DT-Diaphorase (DTD; NQO1) in PBS
buffer. DTD is an obligate 2-electron reductase that is characterized
125 lM) was mixed with PBS buffer and 30% human plasma.
by its ability to use either NADH or NADPH as a cofactor.15 A 10
lL
Hypoxic condition was established by degassing the sample with
argon slowly for 20 min. After degassing, the partial pressure of
oxygen (pO2) value was measured using a blood gas analyzer.
This protocol allowed us to achieve a pO2 value of 18 mm Hg.
After the initial setting of these conditions, the sample vials were
degassed and immediately closed and transferred to an air-tight
chamber that is previously set under hypoxic conditions. Hypox-
ia was maintained by passing argon gas through the chamber
4 mM pro-drug solution was diluted with 180
a 5 L 2 mM NADH coenzyme and a 5 L 40 mU/
in PBS buffer (pH 7.0) were added. The final concentration of pro-
drugs was 200 M and the mixture was incubated at 37 °C for 24 h
lL phosphate buffer,
l
l
lL NQO1 solution
l
and the result showed that 78% of the naloxone was released for
pro-drug 2 and 82% for pro-drug 3 (see Figs. 2 and 3). The identity
of the naloxone peak at 8.5 min was confirmed by their UV profiles
as well as mass spectra. The structure-bioreductive–activity rela-
tionship of a series of indolequinone compounds has been studied
by Moody16 Naylor,11 and Stratford.17 They synthesized a series of
2- and 3-substituted indolequinone compounds and evaluated
their release activities. The results demonstrated that substitution
at the 1-position of the indolequinone nucleus with bulky groups
can be tolerated by the enzyme while compounds with an elec-
tron-withdrawing substituent at C-3 were better substrates.13
Our results and observations were in agreement with those previ-
ous studies. Pro-drug 2 and 3 has a bulky group at position 1, and
the drug moiety is at position 3. Clearly indolequinone binds to the
active site in an orientation that allows space for a large substitute
in the vicinity of the N-1 position.
throughout the sample incubation time. Simultaneously,
control experiment was carried out under normal atmospheric
conditions without any degassing procedure. Samples (30
a
lL
each) were withdrawn from each mixture at certain time points,
namely 1, 2, 5, and 24 h, and were immediately frozen in dry ice.
Samples were thawed prior to HPLC analysis. The time course of
the naloxone release under hypoxic conditions is shown in Fig-
ure 4. It is noteworthy that 64% of the drug was released in
the first 5 h, and after 24 h, almost all the naloxone (99.2%)
was released form pro-drug 2. However, the drug release was
completely inhibited under normoxic conditions (Fig. 4). This
remarkable difference suggests that the drug release rates are
dependent on the oxygen level in the reaction media, which fur-
ther indicates that the release kinetics involve an intermediate
QÅꢁ radical. The radical is very reactive toward oxygen, thus
the half-life time of the species is extremely short and the inter-
mediate will be oxidized back to quinone under a higher oxygen
level, which will not release the drug.18 The results suggest that
under hypoxic conditions, reductive enzymes in blood plasma
can catalyze the release the naloxone from the pro-drug through
a bio-reductive mechanism.
We further tested the release of drug from pro-drug 2 in
fresh human plasma under conditions of low-pressure oxygen.
An air-sealed chamber was made to simulate a hypoxic environ-
ment to test the drug release kinetics. An oxygen gauge was
0.2
With
DT-Diaphorase
In conclusion, we have successfully synthesized naloxone pro-
drugs bearing an indolequinone structure that enzymatically
releases free naloxone. The results of this study suggest that pro-
drug 2 may be useful in a narcotic antagonist feedback system
releasing free naloxone, when hypoxic conditions occur, such as
in respiratory failure. Furthermore, a long chain spacer at indole
position 1 may serve as a reactive point for conjugation of these
pro-drugs to macromolecules to further modify the pharmacoki-
netics of these pro-drugs in vivo.
0.15
PBS buffer
7
8
9
10
0.1
0.05
0
0
5
10
15
Time (min.)
20
25
30
Acknowledgement
Financial support from Defense Advanced Research Projects
Agency (DARPA 53270-MS-DRP) is gratefully acknowledged.
Figure 3. HPLC of naloxone release for pro-drug
Diaphorase.
3
in the presence of DT-

5020
B. Huang et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5016–5020
10. Wilson, W. C.; Benumof, J. L. In Anesthesia; Miller, R. D., Ed., 6th ed.; Churchill
Supplementary data
Livingstone: New York, 2005.
11. Naylor, M. A.; Jaffar, M.; Nolan, J.; Stephens, M. A.; Butler, S.; Patel, K. B.; Evertt,
S. A.; Adams, G. E.; Stratford, I. J. J. Med. Chem. 1997, 40, 2335.
12. Naylor, M. A.; Swann, E.; Everett, S. A.; Jaffar, M.; Nolan, J.; Robertson, N.;
Lockyer, S. D.; Patel, K. B.; Dennis, M. F.; Stratford, M. R. L.; Wardman, P.;
Adams, G. E.; Moody, C. J.; Stratford, I. J. J. Med. Chem. 1998, 41, 2720.
13. Hernick, M.; Borch, R. F. J. Med. Chem. 2003, 46, 148.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2009.07.061.
References and notes
14. Ferrer, S.; Naughton, D. P.; Threadgill, M. D. Tetrahedron 2003, 59, 3445.
15. Ernster, L. Methods Enzymol. 1967, 10, 309.
16. (a) Beall, H. D.; Hudnott, A. R.; Winski, S.; Siegel, D.; Swann, E.; Ross, D.; Moody,
C. J. Bioorg. Med. Chem. Lett. 1998, 8, 545; (b) Swann, E.; Barraja, P.; Oberlander,
A. M.; Gardipee, W. T.; Hudnott, A. R.; Beall, H. D.; Moody, C. J. J. Med. Chem.
2001, 44, 3311.
17. Phillips, R. M.; Naylor, M. A.; Jaffar, M.; Doughty, S. W.; Evereet, S. A.; Breen, A.
G.; Choudry, G. A.; Stratford, I. J. J. Med. Chem. 1999, 42, 4071.
18. Everett, S. A.; Swann, E.; Naylor, M. A.; Stratford, M. R. L.; Patel, K. B.; Tian, N.;
Newman, R. G.; Vojnovic, B.; Moody, C. J.; Wardman, P. Biochem. Pharmacol.
2002, 63, 1629.
1. Stratford, I. J.; Workman, P. Anti-Cancer Drug Des. 1998, 13, 519.
2. Hernick, M.; Flader, C.; Borch, R. F. J. Med. Chem. 2002, 45, 3540.
3. Schmedtje, J. F.; Ji, Y. S. Trends Cardiovasc. Med. 1998, 8, 24.
4. Braddock, M.; Campbell, C. J.; Zuder, D. Int. J. Dermatol. 1999, 38, 808.
5. Ikonomidou, C.; Torski, L. Curr. Opin. Neurol. 1995, 8, 487.
6. Bodayali, T.; Stevens, C. R.; Billingham, M. E. J.; Blake, D. R. Ann. Rheum. Dis.
1998, 57, 703.
7. Bielefeldt, K.; Conklin, J. L. Dig. Dis. Sci. 1997, 42, 878.
8. Bloodworth, D. Phys. Med. Rehabil. Clin. N. Am. 2006, 17, 355.
9. Linder, C.; Fishman, J. J. Med. Chem. 1973, 16, 553.