George Mattheolabakis, Chi C Wong, Yu Sun, Carol A Amella...
Transcript of George Mattheolabakis, Chi C Wong, Yu Sun, Carol A Amella...
JPET #217208
1
Pegylation improves the pharmacokinetics and bioavailability of small-molecule drugs
hydrolysable by esterases: A study of phospho-ibuprofen
George Mattheolabakis, Chi C Wong, Yu Sun, Carol A Amella, Robert Richards, Panayiotis P.
Constantinides, Basil Rigas
Department of Medicine, Stony Brook University (G.M., C.C.W., Y.S., C.A.A., R.R., B.R.) and
Medicon Pharmaceuticals (P.C., B.R.), Stony Brook, NY, USA
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
JPET #217208
2
Running Title: Pegylation prevents phospho-ibuprofen hydrolysis
Correspondence
Basil Rigas
Department of Medicine
HSC, T17-080
Stony Brook University
Stony Brook, NY 11794-8173, USA
Tel: 631-444-9538; Fax: 631-444-9553
E-mail: [email protected]
Number of Text Pages: 21
Number of Tables: 1
Number of Figures: 4
Number of Words:
• Abstract: 243
• Introduction: 483
• Discussion: 776
Abbreviations
CES = carboxylesterases
DCM = dichloromethane
NSAIDs = Nonsteroidal anti-inflammatory drugs
PI = phospho-ibuprofen
PI-PEG = phospho-ibuprofen – polyethylene glycol
PK = pharmacokinetic
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
JPET #217208
3
ABSTRACT
Esterase hydrolysis of drugs can accelerate their elimination thereby limiting their efficacy. Poly-
ethylene glycol covalently attached on drugs (pegylation) is known to improve the efficiency of
many drugs. Using as a test agent the novel phospho-ibuprofen (PI), we examined whether
pegylation of PI could abrogate its hydrolytic degradation by esterases; PI, known to inhibit
colon cancer growth, has a carboxylic ester hydrolysable by carboxylesterases (CES). We
covalently attached PEG-2000 to PI (PI-PEG) and studied its stability by exposing it to cells
overexpressing CES and by administering it to mice. We also evaluated PI-PEG’s anticancer
efficacy in human colon cancer xenografts and in Apcmin/+ mice. PI-PEG was stable in the
presence of cells overexpressing CES1 or CES2, while PI was extensively hydrolyzed (90.2 ±
0.7 %, 14.3 ± 1.1 %, mean ± SEM). In mice, PI was nearly completely hydrolyzed. Intraveous
administration of PI-PEG resulted in significant levels in blood and in colon cancer xenografts
(xenograft values in parentheses): AUC0-24h = 2,351 (2,621) (nmole/g)xh; Cmax = 1,965 (886)
nmole/g; Tmax = 0.08 (2) h. The blood levels of ibuprofen, its main hydrolytic product, were
minimal. Compared to controls, PI-PEG inhibited the growth of the xenografts by 74.8%
(p<0.01), and reduced intestinal tumor multiplicity in Apcmin/+ mice by 73.1% (p<0.01)
prolonging their survival (100% vs. 55.1% of controls; p=0.013).Pegylation protects PI from
esterase hydrolysis and improves its pharmacokinetics. In preclinical models of colon cancer, PI-
PEG is a safe and efficacious agent that merits further evaluation.
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
JPET #217208
4
1. INTRODUCTION
Carboxylesterases (CES) are a multigene family of mammalian enzymes widely
distributed throughout the body that catalyze the hydrolysis of esters, amides, thioesters, and
carbamates. Humans have two carboxylesterases, hCE1 and hCE2. Both are expressed in the
liver, but the intestine expresses only hCE2. CES represent a first-line defense against
internalized molecules, a function crucial to the survival of an organism. However, the function
of CES may be detrimental to drug development, if preservation of the integrity of a drug is
required for its function. This has been the case with phospho-ibuprofen (PI, MDC-917), which
consists of ibuprofen covalently attached to a diethylphosphate group via a butyl spacer moiety
(Mattheolabakis et al., 2012a). The spacer and ibuprofen are bound through a carboxylic ester
bond, which we have shown to be subject to hydrolysis by CES 1 and 2. Since PI is not a pro-
drug, we sought to preserve the molecule intact in vivo by circumventing the ability of CES to
hydrolyze it. We explored whether pegylating PI would serve that purpose.
Pegylation, the covalent attachment of one or more molecules of polyethylene glycol
(PEG) to a target molecule, is an efficient method used to improve the pharmacokinetic (PK)
properties and therapeutic efficacy of both low- and high- molecular weight compounds.
Pegylation of drugs improves their hydrophilicity and other physiochemical properties, enhances
their bioavailability, stability and circulation half-life in vivo, and reduces their potential toxicity
(Greenwald et al., 2003; Hong et al., 2010). Pegylated compounds have shown promising
efficacy in humans. For example, PEG-camptothecin (PROTHECAN®), an ester-based drug, is
in phase 2 clinical trials (Greenwald et al., 2003; Scott et al., 2009); PEG-paclitaxel is currently
in phase 1 trials (Choi and Jo, 2004); and PEG-interferon is currently used for treatment of
hepatitis C (Hoofnagle and Seeff, 2006). We reasoned that in the case of PI, pegylation will have
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
JPET #217208
5
two desirable effects. First, it may enhance its solubility because of PEG’s hydrophilicity;
second, it may protect it from CES due to PEG’s sheer size making it a poor substrate for CES.
PI is a promising candidate anticancer agent. In preclinical models it has exhibited
significant efficacy against colon and breast cancer (Mattheolabakis et al., 2012a; Sun et al.,
2012). It is of interest that in both applications the efficacy of PI against cancer was enhanced by
formulating it in a nanocarrier that likely protected it from CES. Our recent PK study revealed
that the bioavailability of PI in mice is low (<5%) independently of the administration route, a
finding ascribed to limited aqueous solubility, hydrolysis by esterases, and rapid clearance from
the systemic circulation (Xie et al., 2011).
Here, we report the synthesis of pegylated PI (PI-PEG), consisting of mPEG2000
attached to PI at its phosphate moiety (Fig. 1); mPEG2000 is a neutral, biocompatible and
biodegradable methoxy-poly(ethylene oxide). Pegylation of PI rendered it resistant to esterase
hydrolysis, resulting in improved pharmacokinetic (PK) properties and enhanced biodistribution.
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
JPET #217208
6
2. MATERIALS AND METHODS
Ibuprofen was obtained from TCI America (Portland, OR). Methoxy-polyethylene glycol
MW=2,000 (mPEG2000), 1,4-butanodiol, and all other reagents were purchased from Sigma-
Aldrich (St. Louis, MO). Human colon cancer cell lines were purchased from American Type
Culture Collection (ATCC, Manassas, VA) and cultured following the recommendations of
ATCC.
2.1. PI-PEG synthesis
A modified H-phosphate reaction was used to synthesize PI-PEG (Tirosh et al., 1997).
Nuclear magnetic resonance (NMR) spectra were recorded in CDCl3 solution at 400 Mhz using a
Varian Instrument (Santa Clara, CA). The reaction proceeded in two stages as depicted in Fig.1.
2.1.1. Synthesis of 4-hydroxybutyl 2-(4-isobutylphenyl)acetate 3
Ibuprofen 1, (5 g, 0.024 mol) was dissolved with 4-dimethylaminopyridine (0.59 g, 0.004
mol), dicyclohexylazodicarboxylate (DCC, 6.5 g, 0.031 mol) and pyridine (1.91 g, 0.024 mol)
in 80 ml of dry dichloromethane (DCM). The reaction was allowed to proceed for 30 min at
room temperature. This mixture was slowly added to a suspension of 1,4-butanodiol 2 (13 g,
0.144 mol) in 140 ml of dry DCM and allowed to react overnight with continuous stirring. The
resulting solution was filtered, washed with 10% K2CO3, water and brine, dried over magnesium
sulfate and evaporated under reduced pressure. The residue was subjected to silica gel
chromatography and eluted with a gradient of hexane-ethyl acetate to obtain 3.
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
JPET #217208
7
2.1.2. Synthesis of 4-(hydroxy(mPEG)phosphoryloxy butyl 2-(4-isobutylphenyl) propanoate 5
1.6 ml of PCl3 was added to a solution of imidazole (3.89 g, 0.056 mol) in 180 ml of dry
DCM under stirring over an ice-water bath. This was followed by the addition of 8.2 ml of
triethylamine in 20 ml DCM. After 15 minutes, 20 ml of 3 dissolved in DCM was added
dropwise over 30 min. The reaction continued for an additional 40 min before being quenched by
adding 100 ml of water-pyridine (1:4 v/v). After 15 min, 300 ml of chloroform were added and
the organic layer was washed twice with 100 ml of water, dried over MgSO4 and evaporated
under reduced pressure. The residue was dissolved in 50 ml of DCM. Lyophilized mPEG2000
(3.5 g) and pivaloyl chloride (0.35 g, 0.0029 mol) were added to the reaction mixture. After 10
minutes of stirring, the organic solvent was evaporated to dryness under reduced pressure. A
solution of 0.8 g I2 in 15 ml of water-pyridine (1:1 v/v) was added in order to oxidize the H-
phosphate. Oxidation took place for 10 min and was stopped by the addition of 100 ml 5%
aqueous Na2S2O3 solution. PI-PEG 5 was extracted from the aqueous medium with 100 ml of
chloroform and 100 ml of DCM. The organic layers were combined, dried over MgSO4 and
evaporated under reduced pressure. PI-PEG 5 was purified by repeated cycles of acetone
precipitation at -80oC.
2.2. HPLC analysis
PI-PEG was quantified using an HPLC Waters Alliance 2695 equipped with a Waters
2998 photodiode array detector (220 nm, Waters, Milford, MA) and a Thermo BDS Hypersil
C18 column (150 X 4.6 mm, particle size 3 μm, Thermo Fisher Scientific, Waltham, MA). The
mobile phase consisted of a gradient between buffer A (H2O, acetonitrile, trifluoroacetic acid
94.9:5:0.1 v/v/v) and buffer B (acetonitrile). Using various concentrations of PI-PEG in
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
JPET #217208
8
acetonitrile we determined that the range of linear responses of our HPLC method was between
10 and 400 μM.
2.3. In situ hydrolysis by CES1 and CES2
HEK293 cells were seeded into poly-L-lysine coated 24-well plates at a density of 2.0 x
105 cells/per well. After 24 h, HEK293 cells were transfected with the carboxylesterases 1
(CES1) or carboxylesterases 2 (CES2) plasmids, or empty pCMV-XL6 vector with
Lipofectamine 2000. Briefly, the transfection complexes were formed in Opti-MEM (Invitrogen,
Carlsbad, CA) and then added to cells after incubation for 20 min. Over-expression of CES1 and
CES2 was confirmed by RT-PCR. Hydrolysis assays were performed 22-24 h after transfection.
The media were aspirated and replaced with complete RPMI media containing 100 µM PI-PEG
or PI. After 1 h incubation, the cells were washed once with complete media, and collected in
200 µl lysis solution (50% ethanol). Extraction was performed by sonication for 5 min followed
by addition of 400 µl of ethanol. The samples were centrifuged at 17,000 x g for 10 min and
analyzed by HPLC.
2.4. Acute toxicity studies
All animal studies were approved by the Institutional Animal Care and Use Committee at
Stony Brook University. PI-PEG dissolved in water was injected intraperitoneally (i.p.) with
escalating single doses of 400, 800, 1,600, 3,200 and 4,000 mg/kg that corresponds to the
equivalent dose of 66.7, 113.3, 266.7, 533,3 and 666,7 mg/kg of PI, respectively, (2 mice per
dose) into female CD-1 mice (Harlan, Indiana, IN) with 3 days intervals between injections for 3
subsequent injections. Mice were monitored daily for body weight changes, overall health
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
JPET #217208
9
condition, and signs of toxicity. At the end of the study, mice were euthanized and necropsies
were performed.
2.5. PK and drug distribution studies in mice bearing colon cancer xenografts
A single dose of an aqueous solution of PI-PEG, 2,400 mg/kg (equimolar to 400 mg/kg
of PI), was injected i.p. or intravenously (i.v.) into female athymic nude mice (Harlan, Indiana,
IN) bearing SW-480 human colon cancer xenografts. Groups of 2 mice were sacrificed at various
time points following drug administration. Blood was collected through heart puncture and drug
was extracted with two volumes of acetonitrile. Various organs and the xenografts from the
animals were also collected, homogenized and sonicated in PBS, and extracted with two volumes
of acetonitrile. After centrifugation for 10 min at 5,000 g, drug levels were measured in the
supernatants by HPLC, to determine the area under the curve (AUC0-24h), the maximum
concentration, the time of the maximum time and drug’s circulation half-time. As a control,
equimolar amounts of PI were administered i.p. or i.v. .
2.6. Efficacy studies in mice
Human colon cancer xenografts in mice: SW-480 human colon cancer cells (2×106 cells
suspended in 100 μl PBS) were implanted subcutaneously (s.c.) into both flanks of 5-6 week old
female athymic nude mice. When the average tumor volume reached ∼100 mm3, mice were
given i.p. vehicle (water) or PI-PEG 4,000 mg/kg daily x5d/week. Tumor dimensions were
measured using a caliper, and tumor volumes were calculated using the following formula: L ×
W × (L + W/2) × 0.56. On day 17, animals were euthanized and tumors were harvested. We
performed hematoxylin-and-eosin (H&E) staining on lung, liver and kidneys sections to evaluate
their histological characteristics and determine any abnormalities.
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
JPET #217208
10
Efficacy in Apcmin/+ mice: Six-week-old C57BL/6J Apcmin/+ mice (The Jackson
Laboratories, Bar Harbor, ME; n = 9/group) were treated daily by oral gavage as follows: vehicle
(water); PI 400 mg/kg; PI-PEG 2,400 mg/kg; and PEG 2,000 mg/kg (equimolar to PI-PEG). At
16 weeks of age, the animals were sacrificed, their small intestines and colons were removed,
opened longitudinally and tumors were counted using a magnifying lens. Kaplan-Meier survival
estimates were calculated for each group and compared using the log-rank test. Animal weights
were compared across time using linear mixed models with group included as a between subjects
effect and time as a within-subjects factor. The best-fitting variance-covariance structure was
chosen using information criteria.
3. RESULTS
3.1. Synthesis of PI-PEG
PI-PEG was synthesized as depicted in Fig. 1 and characterized by 1H NMR, 31P NMR
and HPLC. The 1H NMR (Supplemental Figure) shows an excess of PEG molecules (peak at 3.6
ppm) present in the final product compared to ibuprofen alone (benzyl peaks at 7.05 and 7.2
ppm). In all samples, unreacted PEG was <10% of the total on a molar basis. The HPLC
chromatogram showed a single peak at 4.8 min representing PI-PEG (Fig 2). Because the UV
absorption of the PI-PEG originates from the ibuprofen part of the molecule, we concluded that
our samples contained <0.5% of other ibuprofen by-products.
3.2. PI-PEG is resistant to carboxylesterase-mediated hydrolysis in vitro
We have previously shown that intact phospho-NSAIDs are much more potent than their
hydrolyzed products (Wong et al., 2012). However, phospho-NSAIDs are highly susceptible to
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
JPET #217208
11
degradation by carboxylesterases in vivo at the carboxyl ester bond. Thus, the stability of PI-PEG
is critical for optimal efficacy. We evaluated the metabolic stability of PI-PEG in CES1- and
CES2-overexpessing cells and compared it to PI, the non-pegylated counterpart. In CES1- and
CES2-overexpressing cells, PI undergoes rapid degradation with 90.2 ± 0.7 % (mean ± SEM for
this and all subsequent values) and 14.3 ± 1.1 % of the total drug hydrolyzed, respectively, after
1 h incubation. In contrast, PI-PEG is highly stable in the presence of carboxylesterase, as we
could not detect any apparent hydrolysis in either CES1- or CES2- expressing cells. Hence,
pegylation significantly protected the carboxyl-ester moiety in PI-PEG from carboxylesterase-
mediated hydrolysis, suggesting that this chemical modification can significantly improve the
metabolic stability in vivo.
3.3. PK and biodistribution of PI-PEG
We studied the PK properties of PI-PEG and PI after a single i.v. or i.p. administration to
mice; these compounds were given at equimolar doses (2400 mg/kg and 400 mg/kg,
respectively). In contrast to PI that was essentially completely hydrolyzed, PI-PEG remained
predominantly intact in vivo, regardless of its route of administration.
As shown in Fig. 3 and Table 1, i.v. administration of PI-PEG led to similar AUC0-24h
values in both blood and tumor xenografts (2,351 and 2,621 nmole/g*h, respectively). As
expected for this route of administration , the blood Cmax was higher (2.2 fold) than that of
tumors but the Tmax in blood (5 min) preceded considerably that of tumors (2 h), indicating an
active and prolonged drug uptake process by the tumors. Both the blood and tumor drug levels
showed a similar decay, leveling off to near zero after 4 h. Of note, ibuprofen, a hydrolysis
product of PI-PEG was detected in both blood and tumors. However, the levels of ibuprofen
were small indicating a low level PI-PEG hydrolysis. Based on AUC0-24h values, ibuprofen in
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
JPET #217208
12
blood was 4% of PI-PEG (99 vs. 2,351 nmole/g*h, respectively) and in tumor 11% (293 vs.
2,621 nmole/g*h, respectively), confirming the marked resistance of PI-PEG to in vivo
hydrolysis. In fact, given the similar Tmax of ibuprofen in blood and tumors (2 h), it is likely that
the low-level degradation of PI-PEG occurs in tumors.
In contrast to PI-PEG and in agreement with our previous report (Xie et al., 2011), when
administered i.v., PI was almost completely hydrolyzed. Even at its maximum tolerated dose
(100 mg/kg), PI exhibited low levels in the blood (AUC0-24h 2.6 nmole/g*h) and was
undetectable in the tumors, being rapidly hydrolyzed to release quantitatively its parent
compound, ibuprofen.
PI-PEG is encountered at distinctly high levels in kidneys that exceed those of all other
organs (Fig. 3C). The Cmax of PI-PEG in kidneys is 5-fold higher than that of stomach, the organ
with the next highest Cmax (4,884 vs. 1,003 μM). The higher levels of PI-PEG in the kidneys
(increased AUC0-24h, Suppl. Table) indicate that the kidneys are an important elimination
pathway for the hydrophilic PI-PEG.
When PI-PEG was given i.v. at doses as high as 4,000 mg/kg that corresponds to the
equivalent dose of 666.7 mg/kg of PI, there was no evidence of toxicity. Further increase in its
dose was precluded by the high viscosity of the i.v. solution. Thus we could not determine acute
or sub-chronic maximum tolerated dose of PI-PEG. In contrast, the MTD of non-pegylated PI
was 100 mg/kg thus, pegylation of PI results in at least 7-fold increase in its MTD in mice. These
findings indicate that pegylation of PI is associated with a significant increase in its safety.
The PK and biodistribution parameters of PI-PEG after its i.p. and i.v. administration are
summarized in Table 1, Fig. 3 and Supplementary Fig. 2. PI-PEG levels after i.p. administration
increased rapidly and blood Cmax=922 µM, Tmax=30 min, and AUC0-24h =2,044 nmole/g*h. In the
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
JPET #217208
13
tumors, the PI-PEG levels peaked at 1 h with Cmax=245 nmole/g and AUC0-24h =1,248 nmole/g*h
(Table 1). We also detected ibuprofen in the plasma and tumors, albeit at much lower levels
compared to PI (Fig. 3, middle panels).
Administration by i.v. injection resulted in a 3.6-fold higher peak drug level and 2.1-fold
higher total drug exposure in the tumors and resulted in reduced drug accumulation in other
organs (based on their respective Cmax and AUC0-24h values) compared to i.p. injection. Overall,
i.v. administration enhanced the delivery of PI-PEG to tumors compared to the i.p. route.
Compared to i.v. administration, when PI-PEG was given i.p. it generated higher AUC0-24h in the
lungs (9.7 fold), liver (3 fold), the spleen (12.8 fold), and the heart (5.3 fold). However, i.v.
administration in mice is limited by the viscosity of the PI-PEG solution, especially when high
drug concentrations are used (3,000 to 4,000 mg/kg). Thus, we continued our studies using i.p.
administration.
3.4. Animal toxicity studies
We evaluated the in vivo toxicity of PI-PEG in mice applying standard protocols for acute
toxicity studies. Mice treated with PI-PEG up to 4,000 mg/kg were healthy and without any body
weight loss. In particular, there were no abnormalities in the major organs upon gross and
histological examination. In addition, all mice treated with PI-PEG on a long-term basis for
efficacy studies showed no evidence of toxicity (Fig. 4).
3.5. PI-PEG is efficacious in the treatment and prevention of colon cancer
We evaluated the chemotherapeutic potential of PI-PEG in a subcutaneous xenograft
model of SW-480 human colon cancer cells. When the tumors reached ~100 mm3 volume, mice
were treated with PI-PEG (4,000 mg/kg/d, i.p., 5 times per wk) for 17 days. As shown in Fig. 4,
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
JPET #217208
14
PI-PEG suppressed tumor growth. This effect was statistically significant, beginning at 7 days
after the initiation of treatment and continuing until the end of the study (day 7 p<0.03; day 10 to
17, p<0.01). Compared to the vehicle control, PI-PEG reduced tumor volume by 74.8% (p<0.01,
Fig. 4).
Next, we evaluated the chemopreventive potential of PI-PEG in the Apcmin/+ mouse
model. Equimolar doses of PI (400 mg/kg/d), PEG (2,000 mg/kg/d) and PI-PEG (2,400 mg/kg/d)
were administered to Apcmin/+ mice by oral gavage once daily for 10 weeks. Both PI and PI-PEG
decreased the total number of tumors in the small intestine by 77.2% and 73.1%, respectively, as
compared to the control group (p<0.01 for both compared to control; no significance between PI
and PI-PEG groups). PI-PEG and PI reduced colon tumor multiplicity by 91.1% (p<0.01) and
64.3% (p<0.02, no significance for PI vs. PI-PEG), respectively. PEG (mPEG2000) reduced the
number of colon tumors by 40.5% (p>0.1 compared to control; Figure 4). Based on the log-rank
tests, survival among both PI and PI-PEG treated animals (100% for both) was significantly
greater (p=0.013) compared to survival for the control or PEG groups (55.6% and 71.4%
respectively, Fig. 4). In both studies, PI-PEG was well tolerated with no weight loss during
treatment.
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
JPET #217208
15
4. DISCUSSION
Pegylation has emerged as an efficacious approach to overcoming inherent limitations of
candidate drugs such as low aqueous solubility, rapid metabolism and elimination, low
bioavailability and toxicity. Indeed, several clinically used products are based on this approach
(Harris and Chess, 2003; Mattheolabakis et al., 2012b). Here, we demonstrate that pegylation of
a new class of small molecules can be used to circumvent the hydrolytic activity of the
ubiquitous carboxylesterases that inactivates otherwise efficacious compounds.
The test compound used to demonstrate the feasibility of this approach was PI, a member
of the promising phospho-NSAIDs (Xie et al., 2011). The pegylation of PI was achieved through
a three-step process starting with ibuprofen. The final product had a purity of over 90%; the low
levels of PEG in PI-PEG were of no biological significance, e.g., PEG even at doses equimolar
to PI-PEG failed to affect tumor multiplicity.
The pegylation of PI had a major impact on its physicochemical properties, interaction
with CESs and pharmacokinetic profile. The lipophilic PI (logP = 5.22) became readily soluble
in aqueous solutions, including blood and other body fluids. In addition, PI-PEG resisted CES
hydrolysis both in vitro and in vivo. In sharp contrast to PI that was hydrolyzed by both isoforms
of CES (nearly completely by CES1), PI-PEG remained essentially intact when exposed to cell
lines overexpressing CES1 and CES2. A similar pattern was observed in vivo as well. PI is
known to be completely hydrolyzed by mice due to the action of their carboxylesterases
independently of the administration route (Wong et al., 2012).
A likely explanation of the resistance of PI-PEG to CES hydrolysis is that the PEG
moiety through steric hindrance rendered the PI-PEG molecule inaccessible to the catalytic site
of CES. PEG is not only 5-fold larger than PI based on molecular weight, but can also behave as
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
JPET #217208
16
a random coil in an aqueous solvent. Thus it is conceivable that the PEG tail sterically prevents
the enzymatic hydrolysis of the new molecule, which, as a result, remains intact. Since we have
shown that all phospho-NSAIDs synthesized to date are subject to CES degradation (unpublished
data), their pegylation may offer a significant pharmacological advantage for this class of drugs.
The two critical changes in the properties of PI brought about by its pegylation
(hydrophilicity and resistance to enzymatic hydrolysis) had a direct effect on its PK and
biodistribution. The levels of PI-PEG in the circulation of mice were very high (Cmax >1,000
µM) in contrast to the essentially undetectable levels of PI (Xie et al., 2011). PI-PEG survived its
exposure to CES which led to high tumor drug levels, where it was present consistently and
independently of its administration route. It is of interest that while the i.p. route delivered
similar amounts of PI-PEG to the blood (practically identical AUC0-24h values), i.v.
administration led to almost double tumor drug levels. Since the i.v. route led to more than
double Cmax values of PI-PEG, it is conceivable that tumor drug uptake is dependent on the
blood-tumor drug gradient.
The biodistribution of PI-PEG likely reflects its enhanced hydrophilicity. The organ with
the highest levels of PI-PEG is the kidney (5 fold higher than the next highest). Thus is possible
that PI-PEG is predominantly eliminated through the kidney.
PI-PEG showed considerable efficacy against colon cancer in animal models and
inhibited tumorigenesis when employed as either a chemopreventive or a chemotherapeutic
agent. In fact, PI-PEG showed roughly the same efficacy in both applications (~75%).
Remarkably, the effect of PI-PEG in the colon in terms of tumor multiplicity was clearly superior
to that of PI (>90% reduction vs. 64%). Even though the difference between the two compounds
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
JPET #217208
17
failed to reach statistical significance (probably due to our modest sample size), the trend in
favor of PI-PEG is clear.
In addition to efficacy, the safety of anticancer agents is of extreme importance,
especially for chemoprevention, when they are administered for prolonged periods of time to
healthy subjects. NSAIDs have as strong a potential to serve as chemopreventive agents
(Johnson et al., 2010), but their toxicity, mostly gastrointestinal and renal, will be limiting in
that context, especially since toxic levels can become cumulative with prolonged use (Rainsford,
1999). PI-PEG appears to be a very safe agent, as evidenced by the absence of gastrointestinal
and other organ toxicity in mice as presented in this study.
In summary, our preclinical data show that pegylation is promising as an approach that
can enhance the pharmacological properties of compounds like PI and structurally similar
phospho-NSAIDs that are subject to enzymatic degradation. Furthermore, PI-PEG is a safe and
efficacious anticancer agent, exhibiting the essential pharmacologic and safety properties
required of a successful candidate anticancer agent.
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
JPET #217208
18
Authorship contributions
Participated in research design: Mattheolabakis, Rigas
Conducted experiments: Mattheolabakis, Wong, Sun
Performed data analysis: Mattheolabakis, Rigas
Wrote or contributed to the writing of the manuscript: Mattheolabakis, Wong, Amella, Richards,
Constantinides, Rigas
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
JPET #217208
19
REFERENCES
Choi JS and Jo BW (2004) Enhanced paclitaxel bioavailability after oral administration of pegylated
paclitaxel prodrug for oral delivery in rats. Int J Pharm 280:221-227.
Greenwald RB, Choe YH, McGuire J and Conover CD (2003) Effective drug delivery by PEGylated drug
conjugates. Advanced drug delivery reviews 55:217-250.
Harris JM and Chess RB (2003) Effect of pegylation on pharmaceuticals. Nature reviews Drug discovery
2:214-221.
Hong M, Zhu S, Jiang Y, Tang G, Sun C, Fang C, Shi B and Pei Y (2010) Novel anti-tumor strategy: PEG-
hydroxycamptothecin conjugate loaded transferrin-PEG-nanoparticles. Journal of controlled
release : official journal of the Controlled Release Society 141:22-29.
Hoofnagle JH and Seeff LB (2006) Peginterferon and ribavirin for chronic hepatitis C. The New England
journal of medicine 355:2444-2451.
Johnson CC, Hayes RB, Schoen RE, Gunter MJ and Huang WY (2010) Non-steroidal anti-inflammatory
drug use and colorectal polyps in the Prostate, Lung, Colorectal, And Ovarian Cancer Screening
Trial. The American journal of gastroenterology 105:2646-2655.
Mattheolabakis G, Nie T, Constantinides PP and Rigas B (2012a) Sterically stabilized liposomes
incorporating the novel anticancer agent phospho-ibuprofen (MDC-917): preparation,
characterization, and in vitro/in vivo evaluation. Pharmaceutical research 29:1435-1443.
Mattheolabakis G, Rigas B and Constantinides PP (2012b) Nanodelivery strategies in cancer
chemotherapy: biological rationale and pharmaceutical perspectives. Nanomedicine (Lond)
7:1577-1590.
Rainsford KD (1999) Profile and mechanisms of gastrointestinal and other side effects of nonsteroidal
anti-inflammatory drugs (NSAIDs). The American journal of medicine 107:27S-35S; discussion
35S-36S.
Scott LC, Yao JC, Benson AB, 3rd, Thomas AL, Falk S, Mena RR, Picus J, Wright J, Mulcahy MF, Ajani JA
and Evans TR (2009) A phase II study of pegylated-camptothecin (pegamotecan) in the
treatment of locally advanced and metastatic gastric and gastro-oesophageal junction
adenocarcinoma. Cancer chemotherapy and pharmacology 63:363-370.
Sun Y, Rowehl LM, Huang L, Mackenzie GG, Vrankova K, Komninou D and Rigas B (2012) Phospho-
ibuprofen (MDC-917) suppresses breast cancer growth: an effect controlled by the thioredoxin
system. Breast cancer research : BCR 14:R20.
Tirosh O, Kohen R, Katzhendler J, Gorodetsky R and Barenholz Y (1997) Novel synthetic phospholipid
protects lipid bilayers against oxidation damage: Role of hydration layer and bound water. J
Chem Soc Perk T 2:383-389.
Wong CC, Cheng KW, Xie G, Zhou D, Zhu CH, Constantinides PP and Rigas B (2012) Carboxylesterases 1
and 2 hydrolyze phospho-nonsteroidal anti-inflammatory drugs: relevance to their
pharmacological activity. The Journal of pharmacology and experimental therapeutics 340:422-
432.
Xie G, Sun Y, Nie T, Mackenzie GG, Huang L, Kopelovich L, Komninou D and Rigas B (2011) Phospho-
ibuprofen (MDC-917) is a novel agent against colon cancer: efficacy, metabolism, and
pharmacokinetics in mouse models. The Journal of pharmacology and experimental therapeutics
337:876-886.
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
JPET #217208
20
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
JPET #217208
21
Footnotes Section
Statement of conflicts of interest: Dr. Rigas has an equity position in Medicon
Pharmaceuticals, Inc. and Dr. Constantinides is a consultant for the company.
This work was supported in part by an assistance award (W81XWH-10-1-0873) from the US Army Medical Research Acquisition Activity
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
JPET #217208
22
FIGURE LEGENDS
Figure 1: Synthesis of PI-PEG. The synthesis was carried out using the H-phosphate strategy as
described in Methods. Briefly, Ibuprofen 1 reacted with excess 1,4-butanodiol 2 according to a
DCC coupling method to obtain 3, which reacted with PCl3 and mPEG2000 to produce the final
product PI-PEG 5.
Figure 2: HPLC chromatogram of PI-PEG. The prominent peak with retention time 4.8 min
originates from PI-PEG. The minor peaks at 6.1 min and 8.5 min originate from ibuprofen
byproducts, and represent < 0.5% of all detected compounds.
Figure 3: PK and biodistribution of PI-PEG. Blood and tumor levels of PI-PEG (A) and its
metabolite, ibuprofen (B), after i.v. or i.p. administration, determined by HPLC as in Methods.
The biodistribution of PI-PEG after i.v. administration (C), and of ibuprofen, its metabolite (D).
S.I., small intestine.
Figure 4: PI-PEG inhibits colon cancer in vivo. A. PI-PEG administered by i.p inhibits the
growth of SW-480 human colon cancer cell xenografts in nude mice. Statistically significant
inhibition started on day 11. B. PI-PEG administered orally to Apcmin/+ mice suppressed tumor
multiplicity in the entire gastrointestinal (GI) tract and also in the colon presented separately. It
also improved the survival rate compared to controls (p<0.05). C. Hematoxylin and eosin
staining of xenograft tissue sections from animals treated with PI-PEG and sacrificed on day 17
of treatment: 1- lung, 2- liver, and 3- kidney.
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
JPET #217208
23
TABLES
Table 1: PK parameters of PI-PEG after i.v. and i.p. injection
Delivery Route
Cmax
(nmole/g)
Tmax
(h)
AUC0-24h
(nmole/g x h)
t1/2
(h)
Intravenously
Blood 1,965 0.08 2,351 0.69
Tumor 886 2 2,621 2.44
Intraperitoneally
Blood 922 0.5 2,044 2.54
Tumor 245 1 1,248 2.87
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
OH
O
+HOOH DMAP, Pyridine
DCC, DCM
O
O
OH
1PCL3
CH2Cl2-TEA
(i) (ii)
H2O
O
O
O P
O
H
OH(i) PV-Cl, pyridine
(ii) CH3O-PEG-OH
(iii) I2, H20
O
O
O PO-
O
O-PEG-OCH3
2
3
45
Figure 1
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
Figure 2
0.4
0.3
0.2
0.1
0.0
-0.1
0 5 10
Retention time (min)
UV Absorption (AU)
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
Figure 3
0
1200
2400
0 8 16 24
Intraperitoneal
Intravenous
Blood (nmole/g)
0
1200
2400
0 8 16 24
Tumor (nmole/g)
Time (h)
0
60
120
0 8 16 24
Intraperitoneal
Intravenous
PI-PEG
Ibuprofen
0
60
120
0 8 16 24
Blood (nmole/g)
Tumor (nmole/g)
A
B
0
2500
5000
S.I. Lung Liver Stomach Spleen Kidney Heart Tumor
5 min 30 min
1h 2h
4h
0
100
200
PI-PEG (nmole/g)
Ibuprofen (nmole/g)
C
D
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from
Figure 4
0
40
80
Control PI-PEG PI PEG
G.I. tract
0
50
100
0 5 10
Survival (%)
Treatment (wks)
Control
PEG
PI
PI-PEG
0
400
800
0 5 10 15 20
Tumor volume (mm3)
Treatment (days)
Control
PI-PEG
* *
0
1
2
Control PI-PEG PI PEG
Colon * *
** *
Numberof tumors at sacrifice
A
B
A B C1 2 3
C
This article has not been copyedited and formatted. The final version may differ from this version.JPET Fast Forward. Published on July 21, 2014 as DOI: 10.1124/jpet.114.217208
at ASPE
T Journals on January 20, 2021
jpet.aspetjournals.orgD
ownloaded from