Dipeptidyl-quinolone derivatives inhibit hypoxia inducible factor-1α prolyl hydroxylases-1, -2, and -3 with altered selectivity

Article abstract of DOI:10.1021/cc100073a

Intracellular levels of the hypoxia-inducible transcription factor (HIF) are regulated under normoxic conditions by prolyl hydroxylases (PHD1, 2, and 3). Treatment of cells with PHD inhibitors stabilizes HIF-1α, eliciting an artificial hypoxic response that includes the transcription of genes involved in erythropoiesis, angiogenesis, and glycolysis. The different in vivo roles of the three PHD isoforms are not yet known, making a PHD-selective inhibitor useful as a biological tool. Although several chemical series of PHD inhibitors have been described, significant isoform selectivity has not been reported. Here we report the synthesis and activity of dipeptidyl analogues derived from a potent but non-selective quinolone scaffold. The compounds were prepared by Pd-catalyzed reductive carbonylation of the 6-iodoquinolone derivative to form the aldehyde directly, which was then attached to a solid support via reductive amination. Amino acids were coupled, and the resulting dipeptidyl-quinolone derivatives were screened, revealing retention of PHD inhibitory activity but an altered PHD1, 2, and 3 selectivity profile. The compounds were found to be ～10-fold more potent against PHD1 and PHD3 than against PHD2, whereas the specific parent compound had shown no appreciable selectivity among the different PHD isoforms.

Full text of DOI:10.1021/cc100073a

676
J. Comb. Chem. 2010, 12, 676–686
Dipeptidyl-Quinolone Derivatives Inhibit Hypoxia Inducible Factor-1r
Prolyl Hydroxylases-1, -2, and -3 with Altered Selectivity
Justin K. Murray,† Chenera Balan,† Alan M. Allgeier,§ Annie Kasparian,§
Vellarkad Viswanadhan,† Christopher Wilde,† Jennifer R. Allen,† Sean C. Yoder,‡
Gloria Biddlecome,‡ Randall W. Hungate,† and Les P. Miranda*,†
Chemistry Research & DiscoVery, Chemical Process Research and DeVelopment, and Hematology,
Amgen, Inc., One Amgen Center DriVe, Thousand Oaks, California 91320
ReceiVed April 30, 2010
Intracellular levels of the hypoxia-inducible transcription factor (HIF) are regulated under normoxic conditions
by prolyl hydroxylases (PHD1, 2, and 3). Treatment of cells with PHD inhibitors stabilizes HIF-1R, eliciting
an artificial hypoxic response that includes the transcription of genes involved in erythropoiesis, angiogenesis,
and glycolysis. The different in vivo roles of the three PHD isoforms are not yet known, making a PHD-
selective inhibitor useful as a biological tool. Although several chemical series of PHD inhibitors have been
described, significant isoform selectivity has not been reported. Here we report the synthesis and activity of
dipeptidyl analogues derived from a potent but non-selective quinolone scaffold. The compounds were
prepared by Pd-catalyzed reductive carbonylation of the 6-iodoquinolone derivative to form the aldehyde
directly, which was then attached to a solid support via reductive amination. Amino acids were coupled,
and the resulting dipeptidyl-quinolone derivatives were screened, revealing retention of PHD inhibitory
activity but an altered PHD1, 2, and 3 selectivity profile. The compounds were found to be ∼10-fold more
potent against PHD1 and PHD3 than against PHD2, whereas the specific parent compound had shown no
appreciable selectivity among the different PHD isoforms.
Introduction
inhibitors were non-specific iron coordinators and/or ana-
logues of 2OG.4 Further investigation led to the discovery
of a variety of iron-coordinating aromatic heterocycles whose
activity was enhanced by appending a carboxylate to mimic
2OG.5 Co-crystallization of PHD2 with an isoquinoline
inhibitor, (((4-hydroxy-8-iodoisoquinolin-3-yl)carbonyl)ami-
no)acetic acid,6 in the 2OG binding pocket has facilitated
the structure-based design and discovery of several new
chemical classes of potent PHD inhibitors,7 including the
newly identified quinolones (Figure 1).8 Most PHD inhibitors
reported to date are 2OG competitive, bind Fe2+ in a
bidentate manner, have a carboxylate that forms a salt bridge
with the guanadinium side chain of HIF-1R residue Arg383,
and make favorable van der Waals contacts with the
hydrophobic residues lining the PHD active site (e.g., Tyr310,
Met 299, and Trp389).6 The highly conserved 2OG binding
pocket (all 14 hydrophobic residues lining the enzyme active
site are completely conserved among the three PHD iso-
forms6) has hindered identification of a PHD-selective
inhibitor, which would be valuable for elucidation of the
The cellular response to changes in the level of dioxygen
is coordinated through the R/ꢀ heterodimeric hypoxia-
inducible transcription factor (HIF). During normoxia, the
presence of sufficient cellular pO2 is “sensed” through
increased activity of a small class of dioxygenases, the HIF
prolyl hydroxylases (PHD1, 2, and 3), in catalyzing the trans-
4-hydroxylation of HIF-1R residues Pro402 or Pro564 within
the oxygen-dependent degradation domain (ODD).1 This
post-translational modification allows HIF-1R to be bound
by the von Hippel Lindau protein (pVHL),2 thus targeting
HIF-1R for poly ubiquitination by an associated E3 ubiquitin
ligase and subsequent degradation by the proteosome.
However, under hypoxic conditions, HIF-1R is stabilized
through reduced PHD action, leading to HIF-mediated
transcription of an array of genes that contain the hypoxia-
responsive element (HRE), including those coding for
glycolytic enzymes, erythropoietin, and the vascular endot-
helial growth factor (VEGF).
Stabilization of HIF-1R through inhibition of PHD has
been examined as a potential treatment for ischemic diseases
including anemia, myocardial infarction, and stroke.3 PHD
enzymatic activity is dependent on Fe2+, ascorbate, and
2-oxoglutarate (2OG), and the first small molecule PHD
* To whom correspondence should be addressed. E-mail: lesm@
amgen.com.
† Chemistry Research & Discovery.
‡ Hematology.
Figure 1. (A) Quinolone PHD inhibitor8 and (B) numbering of
positions around the quinolone ring.
§ Chemical Process Research and Development.
10.1021/cc100073a  2010 American Chemical Society
Published on Web 07/28/2010

Dipeptidyl-Quinolones Display Altered PHD Selectivity
Journal of Combinatorial Chemistry, 2010 Vol. 12, No. 5 677
Scheme 1
Scheme 2
different roles of PHD1, 2, and 3 in vivo. To our knowledge,
compounds with significant selectivity between the three
PHD isoforms have yet to be reported in the literature.
We sought to maintain the potency and alter the selectivity
of the quinolone-based inhibitors by extending the molecule
out of the 2OG binding site into a putative hydrophobic
pocket within the HIF-1R (559-574) peptide binding site.9
By analogy to the isoquinoline series, we assumed the amide
carbonyl and phenolic oxygen of the quinolone coordinate
the Fe2+ and the carboxylate of the attached glycine interacts
with Arg383 to mimic the binding of 2OG in the bottom of
the pocket, while the substituent at the 6-position of the
heterocyclic ring is directed toward solvent at the mouth of
the pocket.6 We decided to derivatize the quinolone scaffold
at this position via attachment of amino acids to explore the
HIF-1R binding site on the surface of PHD. We designed
and prepared a set of 96 dipeptidyl-quinolone analogues and
screened for PHD inhibition. The compounds were initially
found to be 10- to 30-fold more potent against PHD1 and
PHD3 than against PHD2, whereas the parent compound had
shown no appreciable selectivity among the different PHD
isoforms. Several “hits” from the initial screen were resyn-
thesized and validated as having half maximal inhibitory
concentrations (IC50) < 45 nM against PHD2 and IC50 values
<10 nM for PHD1 and PHD3.
signed to allow incorporation of amino acids via 9-fluore-
nylmethoxycarbonyl solid-phase peptide synthesis (Fmoc-
SPPS). The initial steps of the synthesis involved the
conversion of an aryl halide into a form suitable for
immobilization on the solid-phase support (Scheme 1). Ethyl
4-hydroxy-6-iodo-1-methyl-quinolone-3-carboxylate (1), a
derivative prepared during exploration of the structure-activity
relationship of the quinolones,8 was refluxed in dioxane with
tert-butyl glycine hydrochloride to produce compound 2.
Using a slight modification of the conditions developed by
Beller, Almena, and co-workers for the Pd-catalyzed reduc-
tive carbonylation of aryl bromides using CO/H2 (synthesis
gas), compound 2 was converted directly to the desired
aldehyde, compound 3.10,11 Reductive amination of com-
pound 3 with polystyrene Rink amide resin and sodium
cyanoborohydride produced the resin-bound quinolone 4.
Acetylation of the quinolone-resin 4 was followed by
trifluoroacetic acid (TFA) mediated cleavage from the solid
support and removal of the tert-butyl protecting group to
provide compound 5.
Next, we prepared acylated amino acid-quinolone deriva-
tives by following the above strategy but by introducing
amino acids at the 6-position (Scheme 2). Our initial
objective was to determine if derivatization at this site would
significantly reduce PHD activity. Accordingly, we prepared
and tested a relatively small number of compounds. Amino
acids were chosen to maximize the diversity of size,
hydrophobicity, polarity, and chirality within a very small
set. Acylation of quinolone-resin 4 with NR-Fmoc amino acid
Results and Discussion
Initial Synthesis of Quinolone Derivatives. Our synthetic
approach to the dipeptidyl-quinolone derivatives was de-

678 Journal of Combinatorial Chemistry, 2010 Vol. 12, No. 5
Table 1. PHD2 Inhibition by Quinolone-Amino Acid Conjugates
Murray et al.
achieve >90% conversion. Fmoc removal with 20% piperi-
dine in N,N-dimethylformamide (DMF), followed by acety-
lation, and TFA cleavage with simultaneous global side chain
deprotection of all members of chemset 7 afforded chemset
8. Reversed phase (RP) HPLC purification produced com-
pounds 8{1-7} in sufficient quantity (1-2 mg) for screening
in a representative PHD enzymatic assay. From the assay
results (Table 1), we concluded that derivatization at position
6 of the quinolone did not abrogate PHD inhibition and that
incorporation of an amino acid (D-phenylalanine or aspartic
acid) could improve activity relative to acetylation (8{2} and
8{6}). Further exploration of the series was justified, and
we proceeded with preparation of additional analogues.
Synthesis of Dipeptidyl-Quinolone Derivatives. Dipep-
tidyl-quinolone derivatives were synthesized via Fmoc SPPS
from the previously prepared quinolone-amino acid resins
7{1-4} via the following reaction steps (Scheme 3).12 Fmoc
removal with 20% piperidine in DMF was followed by the
coupling of NR-Fmoc-amino acid reagent chemset 9 (Figure
3) with N-[(1H-benzotriazol-1-yl)(dimethylamino)methyl-
ene]N-methylmethanaminium hexafluorophosphate N-oxide
(HBTU) to produce chemset 10. Treatment with piperidine
removed the Fmoc protecting group from chemset 10
members before capping the N-terminus of the dipeptide
chain with reagent chemset 11 (Figure 4). In the final step,
acylation with anhydrides 11{1,2,4} was performed in
dichloromethane (DCM) with N,N-diisopropylethylamine
(DIEA), and coupling of carboxylic acid 11{3} was achieved
via preactivation with HBTU. TFA was added to the resin-
bound compounds to cleave product chemset 12 from the
solid support. The cleavage solution was collected, concen-
trated in a rotary evaporator, and dissolved in dimethylsul-
foxide (DMSO) for analysis and purification. To ascertain
the initial synthetic quality of the chemset 12 members, a
small portion of the DMSO solution was subjected to liquid
chromatography-mass spectrometry (LC-MS) analysis. The
analytical data showed that the compounds in chemset 12
had an average initial purity of 70%, with 25% of the
compounds being g80% pure (area percent of UV absor-
reagent chemset 6 (Scheme 2, Figure 2) to produce chemset
7 proved challenging. Repeated (3-4 ×) 24 h treatments of
the resin with a large excess (>10 equiv) of the preformed
amino acid symmetrical anhydrides were necessary to
Scheme 3

Dipeptidyl-Quinolones Display Altered PHD Selectivity
Journal of Combinatorial Chemistry, 2010 Vol. 12, No. 5 679
Figure 2. NR-Fmoc amino acid diversity reagents 6{1-7}.
Figure 3. NR-Fmoc amino acid diversity reagents 9{1-6}.
Figure 5. Histogram of initial compound purity from prepurification
analytical LC-MS of chemset 12.
Dipeptidyl-Quinolone Derivatives Have Altered PHD
Selectivity Profiles. The compounds in chemset 12 were
screened for inhibition of the PHD-mediated hydroxylation
of Pro564 in the hHIF-1R(559-574) peptide.13 All com-
pounds showed complete inhibition of the three PHD
isoforms at the highest concentration tested (see Supporting
Information). The initial screening results (Figure 8) were
analyzed for any potential structure/activity relationship
(SAR) among the inhibitors. We found that chemset 12
compounds that had been capped with acetic anhydride
(reagent 11{2}) tended to be more active than compounds
containing other chemset 11 reagents; products containing
N-ethylmorpholine (reagent 11{3}) showed the lowest activ-
ity. The incorporation of D-phenylalanine (reagent 6{2}) into
compounds in chemset 12 tended to produce more potent
inhibitors, in good agreement with the result of 8{2} being
the most active compound in chemset 8. No clear preferred
member from reagent chemset 9 was discernible; however,
compound 12{4,1,2}, resulting from the sequential incor-
poration of glycine, L-2-naphthylalanine, and acetic anhy-
dride, was identified as having a higher than expected
potency based upon its constituents. Additionally, we found
that compounds from chemset 12 tended to be selective
toward the PHD1 and PHD3 isoforms (Figure 9 and
Supporting Information), whereas the parent quinolone had
not shown a bias toward any of the three isoforms (Figure
1). Two compounds with D-phenylalanine at the first diversity
position and capped with a benzoyl group showed ap-
proximately 30-fold selectivity against PHD2 (see 12{2,3,1}
Figure 4. Diversity reagents 11{1-4}.
bance at 214 nm of peak with molecular ion corresponding
to target mass, see Figure 5).
Purification and Analysis. The 96 compounds generated
by combining chemset 6{1-4} (Figure 2), chemset 9{1-6}
(Figure 3), and chemset 11{1-4} (Figure 4) to yield chemset
12 (nomenclature example: compound 12[6{1},9{1},11{1}]
represented as 12{1,1,1}) were RP-HPLC purified and
characterized before submission to the PHD assays. Figure
6 shows the HPLC trace before and after purification for
compounds 12{1,5,2}, 12{1,5,1}, 12{1,5,4}, and 12{1,4,2},
and Figure 7 shows the purity profile of the 96-member
library after purification. More than 90% of the compounds
were 90% pure, but eight compounds had purities between
85% and 89% (see Supporting Information). One compound,
12{1,3,1}, which had an initial purity of only 11% from
prepurification analysis, had a final purity of 73% and was
not included in the screen. The samples were submitted for
screening in the PHD enzymatic assays. After the biochemi-
cal screening, the remaining sample material was character-
1
ized by H NMR spectroscopic analysis (see Supporting
Information).

680 Journal of Combinatorial Chemistry, 2010 Vol. 12, No. 5
Murray et al.
Figure 6. LC-MS chromatograms (UV absorbance at 214 nM) before (left) and after (right) purification of chemset compounds 12 (top to
bottom) {1,5,2}, 12{1,5,1}, 12{1,5,4}, and 12{1,4,2}. Peak with a molecular ion corresponding to the target mass indicated in green with
area percent.
(Table 3). We confirmed the activity of five potent PHD2
inhibitors with IC50 values <45 nM, including compounds
12{2,4,2}, 12{4,1,2}, and 12{2,5,2}, which had been the
most active “hits” from the initial screen. These same
compounds were also low nanomolar inhibitors of PHD1
and PHD3 with 3- to 8-fold selectivity over PHD2. The most
selective compound was 12{2,1,2}, which was 7- and 16-
fold selective for PHD1 and PHD3 over PHD2, respectively,
and a significant improvement over the parent quinolone
(Figure 1). The trends in potency and selectivity remained
consistent for the products from the first and second
syntheses; however, the activity values were somewhat
shifted (more potent but less selective), possibly because of
Figure 7. Final purity (area percent of UV absorbance at 214
nm of peak with molecular ion corresponding to target mass) of
day-to-day assay variability and higher accuracy of com-
pound quantification on the larger resynthesis scale. Modeling
suggested that an aminoacyl- or dipeptidyl-sized group may
be sufficient to present a hydrophobic side chain within close
proximity of the hydrophobic pocket created by PHD2
residues Trp258 and Met299 that is normally occupied by
HIF-1R (Figure 10).14 Structural characterization of this
additional interaction will be required to understand the
altered selectivity and/or potency of this series of compounds
as the residues surrounding the immediate opening of the
PHD active site are completely conserved among the three
compounds in chemset 12 from post-purification analytical LC-
MS. The compound with purity <80% was not submitted for
assay.
and 12{2,4,1} in Table 2). With further development, these
PHD-selective inhibitors could be used to understand which
isoform of the PHD enzyme is responsible for the different
aspects of the hypoxia response in vivo.
To confirm the screening results and further validate the
SAR, 12 compounds from the array (including the 3 most
active “hits”) were resynthesized, purified, and assayed

Dipeptidyl-Quinolones Display Altered PHD Selectivity
Journal of Combinatorial Chemistry, 2010 Vol. 12, No. 5 681
Figure 8. Screening of compounds in chemset 12 for inhibition of PHD2. * - Activity for compound 12{1,3,1} was not determined because
of its low purity.
Also, the dipeptidyl-quinolones showed increased potency
(IC50 values <10 nM) and selectivity (∼10-fold) toward
PHD1 and PHD3, which could potentially be developed into
useful compounds for the further understanding of HIF
biology.
Experimental Methods
Preparation of Compound 5. In a 100 mL round-bottom
flask were combined ethyl 4-hydroxy-6-iodo-1-methyl-2-oxo-
1,2-dihydroquinoline-3-carboxylate8 (1, 4.135 g, 11.52 mmol),
glycine tert-butyl ester hydrochloride (2.23 g, 13.3 mmol),
triethylamine (3.08 mL, 22.14 mmol), and 1,4-dioxane (22
mL). The reaction mixture was refluxed for 24 h. The solvent
was removed by rotary evaporation. The residue was
Figure 9. Selectivity of quinolone-dipeptides among PHD1, 2, and
3 from initial screening data.
dissolved in dichloromethane (50 mL) and was washed with
water (2 × 100 mL) and brine (1 × 100 mL). The organic
layer was dried over MgSO4, filtered, and concentrated under
isoforms (Asp254, Ile256, Trp258, Met299, Tyr310, His313,
vacuum. The product was purified by flash chromatography
Asp315, His 374, Trp389).6
(10-50% ethyl acetate in hexanes over 20 min). The
combined fractions were concentrated under vacuum to give
2 (3.7 g, 70% yield). LC-MS: 20-100% B (water and
Conclusions
Modulating HIF transcriptional activity via HIF-1R sta-
bilization by inhibiting PHD activity may have therapeutic
application for anemia and ischemic diseases. We have
investigated modification at the 6-position of the quinolone
series of PHD inhibitors with a dipeptide group. After
developing our synthetic approach with a small set of
compounds, we prepared and screened a number of dipep-
tidyl-quinolone derivatives in the PHD1, 2, and 3 enzymatic
assays, identifying several “hits.” Resynthesis, purification,
and testing of a subset of the original compounds confirmed
that five of the compounds had PHD2 IC50 values <45 nM.
acetonitrile with 0.1% TFA in each as solvents A and B,
respectively) over 2 min at 2.0 mL/min on a Phenomenex
SB-C18, 4.6 × 50 mm, 1.8u column at 45 °C, rt ) 1.79
1
min, m/z ) 403.2 (-tBu C13H11IN2O5 MW ) 402.14). H
NMR (400 MHz, chloroform-d) δ ppm 10.67 (br. s., 1 H)
8.48 (d, J ) 2.15 Hz, 1 H) 7.91 (dd, J ) 8.90, 2.05 Hz, 1
H) 7.10 (d, J ) 9.00 Hz, 1 H) 4.09-4.15 (m, 2 H) 3.65 (s,
3 H) 1.51 (s, 9 H).
To a 160 cc Parr bomb with a stir bar were added tert-
butyl 2-(4-hydroxy-6-iodo-methyl-2-oxo-1,2-dihydroquino-
line-3-carboxamido)acetate (2, 3.07 g, 6.70 mmol), CataXium

682 Journal of Combinatorial Chemistry, 2010 Vol. 12, No. 5
Murray et al.
Table 2. Quinolone-Dipeptides with Greater than 30-Fold Selectivity against PHD2
A (0.137 g, 0.382 mmol), Pd(OAc)2 (0.0292 g, 0.13 mmol),
tetramethylethylenediamine (0.74 mL, 4.91 mmol), and
toluene (50 mL). The reactor was assembled and purged with
Ar (5 × 40 psig) and CO (3 × 20 psig). The reactor was
pressurized with CO to 20 psig and sealed. The manifold
was purged with syngas (1:1 CO/H2). The reactor was
charged with syngas to a total pressure of 60 psig to give a
final CO/H2 ratio of 2:1. The vessel was heated to 100 °C
for 48 h. The reactor was cooled, vented, and purged with
Ar. The reactor contents were washed into a round-bottom
flask with ethyl acetate and concentrated under vacuum. The
product was purified by flash chromatography (10-50%
ethyl acetate in hexanes over 20 min) to give 3 (0.9 g, 37%
yield). LC-MS: 20-100% B over 2 min at 2.0 mL/min on
a Phenomenex SB-C18, 4.6 × 50 mm, 1.8u column at 45
°C, rt ) 1.47 min, m/z ) 305.2 (-tBu C14H12N2O6 MW )
dichloromethane, methanol, and dichloromethane to give 4.
A small amount of resin was treated with triisopropylsilane
(TIS, 0.05 mL), water (0.05 mL), dichloromethane (0.45
mL), and trifluoroacetic acid (TFA, 0.45 mL). The mixture
was shaken for 1 h. The solution was drained and concen-
trated under vacuum to afford (2-(6-(aminomethyl)-4-hy-
droxy-1-methyl-2-oxo-1,2-dihydroquinoline-3-carboxami-
do)acetic acid. LC-MS: 10-100% B water over 3 min at
0.8 mL/min on a Phenomenex Synergi MAX-RP, 2 × 50
mm, 4u column at 40 °C, rt ) 1.48 min, m/z ) 306.0
1
(C14H15N3O5 MW ) 305.29) H NMR (400 MHz, chloro-
form-d + methanol-d3) δ ppm 7.98 (s, 1 H) 7.80 (dd, J )
8.78, 1.76 Hz, 1 H) 7.55 (br. d, J ) 8.80 Hz, 1 H) 4.22 (s,
2 H) 4.20 (s, 2 H) 3.72 (s, 3 H) 1.05 (s, 9 H).
To the resin (4, 42 mg, 0.025 mmol) was added a solution
of acetic anhydride and triethylamine in dichloromethane (5:
1:14, 2 mL). The mixture was shaken for 15 min. The
solution was drained, and the resin was washed (5 × DCM).
To the resin was added triisopropylsilane (0.2 mL), water
(0.2 mL), dichloromethane (1.0 mL), and trifluoroacetic acid
(TFA, 1.0 mL). The mixture was shaken for 1 h. The solution
was drained and concentrated under vacuum. The residue
was dissolved in DMSO (1 mL) and purified by RP-HPLC
5-50%B over 20 min on a Phenomenex Jupiter 4u Proteo
90A, 100 × 7.80 mm column, 3.0 mL/min. The collected
fractions were pooled and concentrated under vacuum to
afford 5 (3.0 mg, 35% yield). LC-MS analysis: 20-100%
B over 3 min at 2.0 mL/min Phenomenex SB-C18, 4.6 ×
50 mm, 1.8u, 45 °C rt ) 2.28 min. m/z ) 348.1 (C16H17N3O6
1
304.25). H NMR (400 MHz, chloroform-d) δ ppm 10.07
(s, 1 H) 10.56 (br. s., 1 H) 8.69 (d, J ) 1.51 Hz, 1 H) 8.21
(dd, J ) 8.78, 1.76 Hz, 1 H) 7.49 (d, J ) 8.53 Hz, 1 H)
4.14 (d, J ) 5.52 Hz, 2 H) 3.74 (s, 3 H) 1.52 (s, 9 H).
To polystyrene Rink amide resin (3.33 g, 0.6 mmol/g,
Novabiochem) and tert-butyl 2-(6-formyl-4-hydroxy-1-meth-
yl-2-oxo-1,2-dihydroquinoline-3-carboxamido)acetate (3, 0.801
g, 2 mmol) in N-methylpyrrolidinone (50 mL) was added
trimethyl orthoformate (0.2 mL, 2 mmol) and glacial acetic
acid (0.2 mL). The reaction mixture was stirred at room
temperature for 1 h followed by the addition of sodium
cyanoborohydride (0.4 g, 7 mmol). The reaction mixture was
stirred at room temperature overnight, filtered, washed with

Dipeptidyl-Quinolones Display Altered PHD Selectivity
Journal of Combinatorial Chemistry, 2010 Vol. 12, No. 5 683
Table 3. PHD2 Inhibition by Selected Quinolone-Dipeptide Conjugates

684 Journal of Combinatorial Chemistry, 2010 Vol. 12, No. 5
Table 3. Continued
Murray et al.
1
MW ) 347.32). H NMR (400 MHz, DMSO-d6) δ ppm
diisopropylcarbodiimide (DIC, 0.125 mmol, 0.02 mL). The
mixture was stirred for 20 min, during which time a white
precipitate formed. The solvent was removed under vacuum
to afford the symmetrical anhydride as a waxy solid. The
solid was dissolved in DMF (2 mL) and added to the resin
(4, 42 mg, 0.025 mmol). The mixture was shaken for 24 h.
The solution was drained, and the resin was washed (3 ×
DCM). The acylation was repeated twice more to achieve
10.58 (t, J ) 5.52 Hz, 1 H) 7.97 (s, 1 H) 8.45 (t, J ) 5.77
Hz, 1 H) 7.70 (dd, J ) 8.53, 1.51 Hz, 1 H) 7.60 (d, J )
8.50 Hz, 1 H) 4.35 (d, J ) 6.02 Hz, 1 H) 4.14 (d, J ) 5.52
Hz, 1 H) 3.63 (s, 3 H)1.88 (s, 3 H).
Preparation of Chemset 8. Fmoc-amino acid (6{1-7},
0.25 mmol, Novabiochem) was dissolved in dichloromethane
(10 mL) in a round-bottom flask. To the solution was added
Figure 10. Predicted binding of quinolone and acetylaminoacyl-quinolone 8{1} to PHD2.

Dipeptidyl-Quinolones Display Altered PHD Selectivity
Journal of Combinatorial Chemistry, 2010 Vol. 12, No. 5 685
90% conversion to resin 7. Fmoc removal was accomplished
by addition of a solution of 20% piperidine in DMF (2 mL)
to the resin. The mixture was shaken for 5 min, the solution
was drained, and the process was repeated. After draining,
the resin was washed (5 × DMF, 5 × DCM). Acetylation,
cleavage from the solid support, and LC-MS purification
were accomplished as described for compound 5 to afford
8{1-7}. LC-MS: 20-100% B over 2 min at 2.0 mL/min
on a Phenomenex SB-C18, 4.6 × 50 mm, 1.8u column at
45 °C 8{1}: 1.0 mg, rt ) 1.30 min, m/z ) 495.2 (C25H26N4O7
plate and concentrated under vacuum. The residue in each
well was dissolved in DMSO (0.5 mL).
A portion (10 µL) of the DMSO solution of the crude
peptide product mixtures was transferred to a shallow-well
96-well plate and diluted with MeOH (30 µL). The LC-MS
analysis showed that the crude compounds were initially 70%
pure on average. The remainder of the DMSO stock solution
was purified by preparative mass-triggered RP-HPLC (250
µL injection on a Synergi 4u MAX-RP 80A, 150 × 21.20
mm column, 5-95%B over 10 min, 20 mL/min). The
collected fractions were concentrated under vacuum in tared
vials and subjected to final characterization by analytical LC-
1
MW ) 494.50), H NMR (400 MHz, DMSO-d6) δ ppm
10.58 (t, J ) 5.58 Hz, 1 H) 8.41 (t, J ) 5.77 Hz, 1 H) 8.33
(t, J ) 6.06 Hz, 1 H) 8.23 (d, J ) 7.83 Hz, 1 H) 7.99 (d, J
) 1.96 Hz, 1 H) 7.83-7.87 (m, 1 H) 7.80 (d, J ) 8.02 Hz,
2 H) 7.73 (s, 1 H) 7.59 (d, J ) 8.80 Hz, 1 H) 7.38-7.51 (m,
3 H) 7.70 (dd, J ) 8.80, 1.96 Hz, 1 H) 4.58 (ddd, J ) 9.29,
8.02, 4.79 Hz, 1 H) 4.37 (d, J ) 5.87 Hz, 2 H) 4.14 (d, J )
5.48 Hz, 2 H) 3.79 (dd, J ) 16.60, 6.10 Hz, 1 H) 3.70 (dd,
J ) 16.60, 5.90 Hz, 1 H) 3.63 (s, 3 H) 3.19 (dd, J ) 13.89,
4.89 Hz, 1 H) 2.93 (dd, J ) 13.79, 9.68 Hz, 1 H) 1.74 (s, 3
H); 8{2}: rt ) 1.30 min, 5.6 mg, m/z ) 495.2 (C25H26N4O7
MW ) 494.50); 8{3}: 2.0 mg, rt ) 1.27 min, m/z ) 461.2
(C22H28N4O7 MW ) 460.48); 8{4}: 1.6 mg, rt ) 1.03 min,
m/z ) 405.2 (C18H20N4O7 MW ) 404.38); 8{5}: 1.8 mg, rt
) 0.99 min, m/z ) 435.2 (C19H22N4O8 MW ) 434.40); 8{6}:
2.3 mg, rt ) 1.02 min, m/z ) 463.2 (C20H22N4O9 MW )
462.41); 8{7}: 2.1 mg, rt ) 0.98 min, m/z ) 476.2
(C22H29N5O7 MW ) 475.50).
1
MS and H NMR to provide Chemset 12 (see Supporting
Information). Compounds were resynthesized from resins
7{1-4} on a 0.05 mmol scale and purified to >95% purity
by LC-MS.
PHD Enzymatic Assay. PHD1, 2, and 3 activity was
measured utilizing homogeneous time-resolved fluorescence
energy transfer technology by detecting the trans-4-hydroxyl-
ation of HIF-1R residue Pro564 in Biotin-hHIF-1R(558-574)
(Biotin-DLEMLAPYIPMDDDFQL) peptide substrate result-
ing in recognition by the Europium-tagged Von Hippel-
Lindau, Elongin B, and Elongin C heterotrimeric complex
(VCB-Eu complex). Compound inhibitor potency was de-
termined using 1 nM PHD1, 2, or 3, 100 nM Biotin-hHIF-
1R(558-574), 0.25 µM 2-OG, 100 µM FeCl2, and 2 mM
ascorbic acid in reaction buffer (30 mM MES, pH 6, 10 mM
NaCl, 10 mM CaCl2, 0.25% Brij-35). The reaction was
terminated after 1 h with 50 mM succinic acid in detection
buffer (50 mM Tris-HCl, pH 8.0, 100 mM NaCl, 0.05%
Tween 20, 0.5% NaN3) containing a final concentration of
25 nM streptavidin-APC and 2.5 nM VCB-Eu. The POC
(percentage of control) was determined by comparing the
signal from hydroxylated peptide substrate, Biotin-
[Hyp564]hHIF-1R(558-574), in the enzyme reaction con-
taining inhibitor compound with that from PHD2 enzyme
with DMSO vehicle alone, and no enzyme. The data were
fit to the 4-parameter model using a Levenberg-Marquardt
non-linear regression algorithm.8
Preparation of Chemset 12. Resins 7{1-4} were pre-
pared on large scale from resin 4 (0.5 mmol, 830 mg each)
by repeated treatment with the preformed symmetrical
anhydrides of Fmoc-amino acids (6{1-4}, 4 × 5 mmol
each). Resins 7{1-4} were distributed into each well of a
96-well filter plate (1.0 mL of a 3:2 DCM/DMF resin slurry
was delivered to each well using a pipet, 24 wells per
derivative, 0.02 mmol per well). The resin was washed (8
× DMF). To each well was added 120 µL of 20% piperidine
in DMF and allowed to stand for 5 min. The solution was
drained, and the process was repeated. After draining, the
resin was washed (8 × DMF). In separate vials, Fmoc amino
acids 9{1-6} (0.2 mL of a 0.5 M solution in DMF, 0.1 mmol
per well) were activated with N-[(1H-benzotriazol-1-yl)(dim-
ethylamino)methylene]N-methylmethanaminium hexafluo-
rophosphate N-oxide (HBTU, 0.2 mL of a 0.5 M solution in
DMF) and N-methylmorpholine (NMM, 0.1 mL of 1.0 M
solution in DMF), added to the wells, and allowed to stand
for 20 min. The solution was drained, and this process was
repeated. After draining, the resin was washed (8 × DMF).
To each well was added 120 µL of 20% piperidine in DMF
and allowed to stand for 5 min. The solution was drained,
and the process was repeated. After draining, the resin was
washed (8 × DMF, 8 × DCM). To the resin was added 200
µL of acylation cocktail (1.4:0.1:0.5 DCM/TEA/acetic,
pivalic, benzoic anhydride) or preactivated 3-morpholino-
propanoic acid (11{1-4}) and allowed to stand for 15 min.
The solution was drained, and the process was repeated. The
resin was washed (5 × DCM). To the resin was added 1
mL of cleavage cocktail (90:5:5 TFA/TIS/water). The
cleavage solution was drained into a solid-bottom 96-well
Acknowledgment. The authors thank Ankita Patel, Breanne
Grady, Justin Provchy, Stephanie J. Mercede, Steven Notari,
and Zheng Hua for technical assistance.
1
Supporting Information Available. H NMR spectra,
LC-MS chromatograms before and after purification, and
PHD screening data for Chemset 12. This material is
available free of charge via the Internet at http://pubs.acs.org.
References and Notes
(1) (a) Ivan, M.; Kondo, K.; Yang, H.; Kim, W.; Valiando, J.;
Ohh, M.; Salic, A.; Asara, J. M.; Lane, W. S.; Kaelin, W. G.
Science 2001, 292, 464–468. (b) Jaakkola, P.; Mole, D. R.;
Tian, Y.-M.; Wilson, M. I.; Gielbert, J.; Gaskell, S. J.; von
Kriegsheim, A.; Hebestreit, H. F.; Mukherji, M.; Schofield,
C. J.; Maxwell, P. H.; Pugh, C. W.; Ratcliffe, P. J. Science
2001, 292, 468–472.
(2) (a) Ohh, M.; Park, C. W.; Ivan, M.; Hoffman, M. A.; Kim,
T.-Y.; Huang, L. E.; Pavletich, N.; Chau, V.; Kaelin, W. G.
Nat. Cell Biol. 2000, 2, 423–427. (b) Min, J.-H.; Yang, H.;
Ivan, M.; Gertier, F.; Kaelin, W. G. Science 2002, 296, 1886–
1889.

686 Journal of Combinatorial Chemistry, 2010 Vol. 12, No. 5
Murray et al.
(3) (a) William, C.; Masson, N.; Tian, Y.-M.; Mahmood, S. A.;
Wilson, M. I.; Bicknell, R.; Eckhardt, K.-U.; Maxwell, P. H.;
Ratcliffe, P. J.; Pugh, C. W. Proc. Natl. Acad. Sci. U.S.A.
2002, 99, 10423–10428. (b) Siddq, A.; Ayoub, I. A.; Chavez,
J. C.; Aminova, L.; Shah, S.; LaManna, J. C.; Patton, S. M.;
Connor, J. R.; Cherny, R. A.; Volitakis, I.; Bush, A. I.;
Langsetmo, I.; Seeley, T.; Gunzler, V.; Ratan, R. R. J. Biol.
Chem. 2005, 280, 41732–41743. (c) Safran, M.; Kim, W. Y.;
O-Connell, F.; Flippin, L.; Gu¨nzler, V.; Horner, J. W.;
DePinho, R. A.; Kaelin, W. G. Proc. Natl. Acad. Sci. U.S.A.
2006, 103, 105–110.
(4) (a) Schlemminger, I.; Mole, D. R.; McNeill, L. A.; Dhanda,
A.; Hewitson, K. S.; Tian, Y.-M.; Ratcliffe, P. J.; Pugh, C. W.;
Schofield, C. J. Bioorg. Med. Chem. Lett. 2003, 13, 1451–
1454. (b) Mole, D. R.; Schlemminger, I.; McNeill, L. A.;
Hewitson, K. S.; Pugh, C. W.; Ratcliffe, P. J.; Schofield, C. J.
Bioorg. Med. Chem. Lett. 2003, 13, 2677–2680.
Biswas, K.; Frohn, M. J.; Peterkin, T. A. N.; Chang, C.; Bu¨rli,
R. W.; Dao, J. H.; Veith, H.; Rogers, N.; Yoder, S. C.;
Biddlecome, G.; Tagari, P.; Allen, J. R.; Hungate, R. W.
Bioorg. Med. Chem. Lett. 2008, 18, 3925–3928. (f) Frohn,
M.; Viswanadhan, V.; Pickrell, A. J.; Golden, J. E.; Muller,
K. M.; Bu¨rli, R. W.; Biddlecome, G.; Yoder, S. C.; Rogers,
N.; Dao, J. H.; Hungate, R.; Allen, J. R. Bioorg. Med. Chem.
Lett. 2008, 18, 5023–5026. (g) Kawamoto, R. M. Prolyl
hydroxylases inhibitors and methods of use. WO Patent
2008002576, Jan 3, 2008. (h) Hocutt, F. M.; Leonard, B. E.;
Peltier, H. M.; Phuong, V. K.; Rabinowitz, M. H.; Rosen,
M. D.; Tarantino, K. T.; Venkatesan, H.; Zhao, L. X.
Benzoimidazoles as prolyl hydroxylases inhibitors. WO Patent
2009134750, Nov 5, 2009. (i) Rosen, M. D.; Rabinowitz,
M. H.; Zhao, L. X.; Hocutt, F. M. Benzoimidazole glycina-
mides as prolyl hydrodroxylase inhibitors. WO Patent
2009134754, Nov 5, 2009.
(5) (a) Franklin, T. J.; Morris, W. P.; Edwards, P. N.; Large, M. S.;
Stephenson, R. Biochem. J. 2001, 353, 333–338. (b) Ivan, M.;
Haberberger, T.; Gervasi, D. C.; Michelson, K. S.; Gu¨nzler,
V.; Kondo, K.; Yang, H.; Sorokina, I.; Conaway, R. C.;
Conaway, J. W.; Kaelin, W. G. Proc. Natl. Acad. Sci. U.S.A.
2002, 99, 13459–13464. (c) Warnecke, C.; Griethe, W.;
Weideimann, A.; Ju¨rgensen, J. S.; William, C.; Bachmann,
S.; Ivashchenko, Y.; Wagner, I.; Frei, U.; Weisner, M.;
Eckardt, K.-U. FASEB J. 2003, 17, 1186–1188. (d) Hirsila,
M.; Koivunen, P.; Gu¨nzler, V.; Kivirikko, K. I.; Myllyharju,
J. J. Biol. Chem. 2003, 278, 30772–30780. (e) Knowles, H. J.;
Tian, Y.-M.; Mole, D. R.; Harris, A. L. Circ. Res. 2004, 162–
169.
(6) McDonough, M. A.; Li, V.; Flashman, E.; Chowdhury, R.;
Mohr, C.; Lie´nard, B. M. R.; Zondlo, J.; Oldham, N. J.;
Clifton, I. J.; Lewis, J.; McNeill, L. A.; Kurzeja, R. J. M.;
Hewitson, K. S.; Yang, E.; Jordan, S.; Syed, R. S.; Schofield,
C. J. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 9814–9819.
(7) (a) Warshakoon, N. C.; Wu, S.; Boyer, A.; Kawamoto, R.;
Sheville, J.; Renock, S.; Xu, K.; Pokross, M.; Zhou, S.; Winter,
C.; Walter, R.; Mekel, M.; Evdokimov, A. G. Bioorg. Med.
Chem. Lett. 2006, 16, 5517–5522. (b) Warshakoon, N. C.;
Wu, S.; Boyer, A.; Kawamoto, R.; Sheville, J.; Renock, S.;
Xu, K.; Pokross, M.; Evdokimov, A. G.; Walter, R.; Mekel,
M. Bioorg. Med. Chem. Lett. 2006, 16, 5598–5601. (c)
Warshakoon, N. C.; Wu, S.; Boyer, A.; Kawamoto, R.;
Sheville, J.; Bhatt, R. T.; Renock, S.; Xu, K.; Pokross, M.;
Zhou, S.; Walter, R.; Mekel, M.; Evdokimov, A. G.; East, S.
Bioorg. Med. Chem. Lett. 2006, 16, 5616–5620. (d) Warsha-
koon, N. C.; Wu, S.; Boyer, A.; Kawamoto, R.; Renock, S.;
Xu, K.; Pokross, M.; Evdokimov, A. G.; Zhou, S.; Winter,
C.; Walter, R.; Mekel, M. Bioorg. Med. Chem. Lett. 2006,
16, 5687–5690. (e) Tegley, C. M.; Viswanadhan, V. N.;
(8) Allen, J. R.; Tegley, C. M.; Biswas, K.; Burli, R.; Muller,
K. M.; Frohn, M. J.; Golden, J. E.; Mercede, S. J.; Neira,
S. C.; Peterkin, T. A. N.; Hungate, R. W.; Kurzeja, R.; Yu,
V.; Dao, J. Quinolone based compounds exhibiting prolyl
hydroxylase inhibitory activity, and compositions, and uses
thereof. WO Patent 2007070359, Apr 3, 2008.
(9) Hon, W.-C.; Wilson, M. I.; Harlos, K.; Claridge, T. D. W.;
Schofield, C. J.; Pugh, C. W.; Maxwell, P. H.; Ratcliffe, P. J.;
Stuart, D. I.; Jones, E. Y. Nature 2002, 417, 975–978.
(10) Klaus, S.; Neumann, H.; Zapf, A.; Stru¨bing, D.; Hu¨bner, S.;
Almena, J.; Reirmeier, T.; Grob, P.; Sarich, M.; Krahnert, W.-
R.; Rossen, K.; Beller, M. Angew. Chem., Int. Ed. 2006, 45,
154–158. The CO/H2 ratio was increased from 1:1 to 2:1 to
limit reductive de-iodination.
(11) Initial synthetic attempts using formic acetic anhydride as the
CO source produced compound 3, albeit in much lower yield,
with predominant formation of the des-iodoquinolone. (a)
Cacchi, S.; Fabrizi, G.; Goggiamani, A. J. Comb. Chem. 2004,
6, 692–694. (b) Morimoto, T.; Kakiuchi, K. Angew. Chem.,
Int. Ed. 2004, 43, 5580–5588.
(12) Because of a bias against compounds containing two car-
boxylic acid functional groups, further investigation of the
scaffold represented by 8{6} was not pursued in spite of its
improved PHD inhibitory activity.
(13) Dao, J. H.; Kurzeja, R. J. M.; Morachis, J. M.; Veith, H.;
Lewis, J.; Yu, V.; Tegley, C. M.; Tagari, P. Anal. Biochem.
2009, 384, 213–223.
(14) Molecular modeling studies were performed as described in
ref 7f.
CC100073A