Discovery of 5-substituted pyrrolo[2,3- d ]pyrimidine antifolates as dual-acting inhibitors of glycinamide ribonucleotide formyltransferase and 5-aminoimidazole-4-carboxamide ribonucleotide formyltransferase in de novo purine nucleotide biosynthesis: Implications of inhibiting 5-aminoimidazole-4- carboxamide ribonucleotide formyltransferase to AMPK activation and antitumor activity

Article

pubs.acs.org/jmc

Discovery of 5‑Substituted Pyrrolo[2,3‑d]pyrimidine Antifolates as

Dual-Acting Inhibitors of Glycinamide Ribonucleotide

Formyltransferase and 5‑Aminoimidazole-4-carboxamide

Ribonucleotide Formyltransferase in De Novo Purine Nucleotide

Biosynthesis: Implications of Inhibiting 5‑Aminoimidazole-4-

carboxamide Ribonucleotide Formyltransferase to AMPK Activation

and Antitumor Activity

Shermaine Mitchell-Ryan,^†,⊥Yiqiang Wang,^‡,⊥,#Sudhir Raghavan,^‡Manasa Punaha Ravindra,^‡

Eric Hales,^§,†Steven Orr,^§,†Christina Cherian,^§,†Zhanjun Hou,^§,†Larry H. Matherly,*^{,†,§,∥,⊥}

and Aleem Gangjee*^,‡,⊥

^†Department of Oncology, Wayne State University School of Medicine, Detroit, Michigan 48201, United States

^‡Division of Medicinal Chemistry, Graduate School of Pharmaceutical Sciences, Duquesne University, 600 Forbes Avenue, Pittsburgh,

Pennsylvania 15282, United States

^§Molecular Therapeutics Program, Barbara Ann Karmanos Cancer Institute, 110 East Warren Ave, Detroit, Michigan 48201, United

States

^∥Department of Pharmacology, Wayne State University School of Medicine, Detroit, Michigan 48201, United States

S

* Supporting Information

ABSTRACT: We synthesized 5-substituted pyrrolo[2,3-d]pyrimidine

antifolates (compounds 5−10) with one-to-six bridge carbons and a

benozyl ring in the side chain as antitumor agents. Compound 8 with a

4-carbon bridge was the most active analogue and potently inhibited

proliferation of folate receptor (FR) α-expressing Chinese hamster

ovary and KB human tumor cells. Growth inhibition was reversed

completely or in part by excess folic acid, indicating that FRα is

involved in cellular uptake, and resulted in S-phase accumulation and

apoptosis. Antiproliferative eﬀects of compound 8 toward KB cells

were protected by excess adenosine but not thymidine, establishing de

novo purine nucleotide biosynthesis as the targeted pathway. However,

5-aminoimidazole-4-carboxamide (AICA) protection was incomplete,

suggesting inhibition of both AICA ribonucleotide formyltransferase

(AICARFTase) and glycinamide ribonucleotide formyltransferase (GARFTase). Inhibition of GARFTase and AICARFTase by

compound 8 was conﬁrmed by cellular metabolic assays and resulted in ATP pool depletion. To our knowledge, this is the ﬁrst

example of an antifolate that acts as a dual inhibitor of GARFTase and AICARFTase as its principal mechanism of action.

Antifolates targeting de novo purine nucleotide biosynthesis

were also described and include lometrexol ((6R)5,10-

dideazatetrahydrofolate, LMTX) (Figure 1), (2S)-2-((5-(2-

((6R)-2-amino-4-oxo-5,6,7,8-tetrahydro-1H-pyrido[2,3-d]-

pyrimidin-6-yl)ethyl)thiophene-2-carbonyl)amino)-

pentanedioic acid (LY309887),^3,4and (2S)-2-((5-(2-((6S)-2-

amino-4-oxo-1,6,7,8-tetrahydropyrimido[5,4-bate][1,4]thiazin-

6-yl)ethyl)thiophene-2-carbonyl)amino)pentanedioic acid

(AG2034).⁵All of these drugs inhibit the ﬁrst folate-dependent

step in purine nucleotide biosynthesis, catalyzed by β-

INTRODUCTION

■

The antifolates are a class of antiproliferatives widely

recognized for their inhibition of folate metabolism.^1,2Major

antifolate enzyme targets include thymidylate synthase (TS)

and dihydrofolate reductase (DHFR). Inhibition of these

enzymes suppresses de novo nucleotide biosynthesis, resulting

in an imbalance of purine and pyrimidine precursors and

rendering cells incapable of undergoing accurate DNA

replication, ultimately resulting in cell death. Clinically relevant

TS (e.g., pemetrexed, PMX) and DHFR (e.g., methotrexate

(MTX) and pralatrexate) inhibitors (Figure 1) continue to play

important roles in treating hematologic malignancies and solid

tumors.^1,2

Received: August 28, 2013

Published: November 20, 2013

10016

dx.doi.org/10.1021/jm401328u | J. Med. Chem. 2013, 56, 10016−10032

Journal of Medicinal Chemistry

Article

Figure 1. Structures of classical antifolates including methotrexate (MTX), pemetrexed (PMX), raltitrexed (RTX), lometrexol (LMTX), and

pralatrexate.

Figure 2. Design of 5-substituted pyrrolo[2,3-d]pyrimidines 5 and 7−10 (n = 1 or 3−6) based on structures of 6-substituted analogues 1−4 and

pemetrexed.

glycinamide ribonucleotide (GAR) formyltranferase (GARF-

Tase), and have progressed to clinical trials.^4,6,7However, their

toxicities were dose-limiting, most likely a consequence of their

cellular uptake and metabolism to polyglutamates in normal

tissues.

mechanism is enhanced.^10,11For FRα, the apical spatial

orientation in normal epithelial tissues is often disrupted in

tumors such that its basolateral membrane expression in tumors

results in exposure to the circulation.⁹These features provide a

compelling rationale for developing folate-based therapeutics

that target PCFT and FRα for cancer therapy. Examples of

FRα-targeted therapies include a monoclonal antibody,

Farletuzumab (Morphotech),¹²a cytotoxic folate conjugate,

Vintafolide (EC145; Endocyte),¹³and, ONX0801, a classical

antifolate that is selectively transported into cells by FRs over

RFC and inhibits de novo thymidylate biosynthesis.¹⁴

PMX is a 5-substituted pyrrolo[2,3-d]pyrimidine antifolate

with a 2-carbon bridge linked to a p-aminobenzoyl glutamate.

PMX is a good transport substrate for RFC and is among the

best PCFT substrates,^8,10,11yet it is a modest substrate for

FRs.¹⁵⁻¹⁹We previously described 6-substituted pyrrolo[2,3-

d]pyrimidine benzoyl antifolates related to PMX (compounds

1−4) (Figure 2).^15,20Whereas the 6-regioisomer of PMX was

pharmacologically inert, the 3- and 4-bridge carbon analogues

The reduced folate carrier (RFC) is one of three principal

mechanisms of (anti)folate uptake into mammalian cells.⁸

Cellular requirements for folate cofactors for DNA replication

provide a plausible explanation for the high levels of RFC in

most tumors. However, demands for folates, and consequently

RFC, are also shared by normal tissues such that RFC may not

be the optimal mechanism for tumor-selective uptake of

cytotoxic folate analogues. Other cellular uptake mechanisms,

notably, the proton-coupled folate transporter (PCFT) and

folate receptors (FRs) α and β, are also expressed in tumors

and show more restricted expression in normal tissues.⁹

Furthermore, PCFT is a proton symporter such that in the

acidic microenvironment generated by glycolytic tumors,

membrane transport and selective tumor targeting by this

10017

dx.doi.org/10.1021/jm401328u | J. Med. Chem. 2013, 56, 10016−10032

Journal of Medicinal Chemistry

Article

Scheme 1

of this series (compounds 1 and 2, respectively) were

selectively transported by FRα and PCFT over RFC and

were potently inhibitory toward FRα- and PCFT-expressing

tumor cells.^15,20

might inhibit AICARFTase, analogous to PMX, albeit without

TS inhibition. Furthermore, if this was accompanied by FRα

and/or PCFT transport selectivity over RFC, then this would

be especially appealing for targeting tumors that express high

levels of these systems.

PMX was originally described as a multitargeted antifolate

with a primary inhibition at TS and secondarily at GARFTase

and DHFR.²¹However, recent studies suggested that although

TS was the primary cellular target of PMX, 5-aminoimidazole-

4-carboxamide (AICA) ribonucleotide (ZMP) formyltransfer-

ase (AICARFTase) could be a pharmacologically important

target in the presence of excess thymidine (circumvents TS) at

suﬃciently high PMX concentrations.^22,23Hence, in the

presence of thymidine, inhibition of proliferation of CCRF-

CEM leukemia cells and solid tumor cell lines was circum-

vented by hypoxanthine but only partially by AICA.^22,23

Furthermore, PMX resulted in marked accumulations of

ZMP, the AICARFTase substrate. By contrast, the 6-substituted

pyrrolo[2,3-d]pyrimidine analogues 1 and 2 are exclusively

GARFTase inhibitors whose inhibitory eﬀects are completely

circumvented by AICA.^15,20

The accumulation of ZMP in PMX-treated cells is intriguing

because ZMP is an AMP mimetic that activates AMP-activated

protein kinase (AMPK).²⁴AMPK negatively regulates mTOR,

a critical prosurvival pathway that is activated in many tumor

cells along with PI3K/AKT, secondary to loss or mutation of

PTEN.²⁵⁻²⁷This may provide a possible explanation for the

growth inhibitory eﬀects of PMX in the presence of thymidine,

as purine nucleotides are not depleted.^22,23However, this has

not been directly tested because no AICARFTase-targeted

drugs without TS inhibition have been described.

We describe herein the impact of these structural

modiﬁcations on (i) transporter speciﬁcity and (ii) the targeted

pathway (TS versus de novo purine nucleotide biosynthesis),

including (iii) the extent of cellular GARFTase and

AICARFTase inhibition. Our results document an emerging

structure−activity relationship (SAR) for the pyrrolo[2,3-

d]pyrimidine antifolates, accompanying translocation of the 6-

pyrrolo[2,3-d]pyrimidine substituent to the regio 5-position,

including sustained FRα cellular uptake, accompanying

restoration of RFC and loss of PCFT transport activities. The

4-carbon bridge analogue of this series (compound 8) is unique

in that it shows dual inhibition of both AICARFTase and

GARFTase exclusive of TS in tumor cells, resulting in depletion

of purine nucleotides. Finally, our results directly examine the

possibility of whether AMPK activation secondary to

AICARFTase inhibition contributes to decreased proliferation

of KB human tumor cells.

CHEMISTRY

■

Compounds 5−10 (n = 1−6) were obtained from a modiﬁed

synthesis of PMX^28,29using an α-bromo aldehyde condensation

with 2,6-diamino-4-oxo-pyrimidine 15 as the key step (Scheme

1). Thus, a Heck coupling reaction³⁰of aryl iodide 11 with

commercially available unsaturated alcohols 12a−f in DMF at

70 °C aﬀords the unsaturated, coupled alcohols that rearrange

to the vinyl alcohols and tautomerise to aﬀord aldehydes 13a−f

(yields: 83−92%), as reported by Larock et al.³⁰α-

Bromonation of 13a−f with 5,5-dibromo-2,2-dimethyl-4,6-

dioxo-1,3-dioxane³¹at room temperature aﬀorded correspond-

ing α-bromo aldehydes 14a−f (yields: 74−78%). 5-Substituted

pyrrolo[2,3-d]pyrimidines 16a−f were synthesized by con-

In this article, we synthesized and systematically charac-

terized the in vitro antiproliferative activities and cellular

mechanisms of novel 5-substituted pyrrolo[2,3-d]pyrimidine

antifolates (including PMX) with a benzoyl ring in the side

chain and one-to-six bridge carbons (compounds 5−10) closely

related to 6-substituted pyrrolo[2,3-d]pyrimidine analogues 1−

4^15,20(Figure 2). We reasoned that the 5-substituted analogues

10018

dx.doi.org/10.1021/jm401328u | J. Med. Chem. 2013, 56, 10016−10032

Journal of Medicinal Chemistry

Article

Table 1. IC₅₀Values (nM) for 5- and 6-Substituted Pyrrrolo[2,3-d]pyrimidine Antifolates and Classical Antifolates in RFC-,

a

PCFT-, and FR-Expressing Cell Lines

hRFC

hFRα

hPCFT

hRFC/FRα/hPCFT

KB

compound

structure

PC43-10

R2

RT16

RT16 (+FA) R2/hPCFT4 R2(VC)

KB

KB (+FA)

(+Thd/Ade/AICA)

1

6-pyrrole/3C

6-pyrrole/4C

6-pyrrole/5C

6-pyrrole/6C

5-pyrrole/1C

5-pyrrole/2C

5-pyrrole/3C

5-pyrrole/4C

5-pyrrole/5C

5-pyrrole/6C

648.6(38.1)

>1000

216(8.7)

>1000

4.1(1.6)

6.3(1.6)

54(21)

>1000

461(62)

>1000

188(41)

23.0(3.3)

213 (28)

>1000

1.7(0.4)

1.9(0.7)

13(7.2)

23(12)

>1000

Ade/AICA

2

>1000

3

>1000

4

>1000

162(18)

>1000

5

>1000

6 (PMX)

30.6 (6.2)

68.8(21.2)

56.6(5.8)

196.4 (55.0)

>1000

18.2(3.8)

72.0(27.1)

8.6(2.1)

33.5(2.5)

>1000

22.3(8.6)

329(61)

840(90)

>1000

9.94(3.11)

49.5(13.2)

12.7(5.4)

17.3(8.9)

>1000

690 (310)

533(233)

700 (300)

898(102)

>1000

Thd/Ade

Ade

7

8

Ade

9

Ade/AICA

10

>1000

MTX

RTX

LMTX

12(1.1)

114(31)

15(5)

120.5(16.8)

99.5(11.4)

38.0(5.3)

6.0(0.6)

5.9(2.2)

1.2(0.6)

20(2.4)

22(5)

Thd/Ade

Thd

6.3(1.3)

12(2.3)

12(8)

31(7)

Ade/AICA

a

Growth inhibition assays were performed for CHO sublines engineered to express human RFC (PC43-10), FRα (RT16), or PCFT (R2/PCFT4)

for comparison with transporter-null (R2 and R2(VC)) CHO cells and for the KB human tumor sublines (expressing RFC, FRα, and PCFT). For

the FRα experiments, growth inhibition assays were performed in the presence and absence of 200 nM folic acid (FA). The data shown are mean

values from 3 to 10 experiments ( standard errors in parentheses). The results are presented as IC₅₀values corresponding to the concentrations

that inhibit growth by 50% relative to cells incubated without drug. Data for MTX, RTX, LMTX, and compounds 1, 2, 3, and 4 were previously

published¹⁵(structures for compounds 1−10 are in Figure 2). For KB cells, data are shown for the protective eﬀects of nucleoside additions,

including adenosine (60 μM) or thymidine (10 μM), or 5-aminoimidazole-4-carboxamide (320 μM). Examples of these data are in Figure 5. For 5-

and 6-pyrrolo-2,3-d]pyrimidine compounds 1−10, the structural motifs are noted (i.e., the pyrrole substitution/bridge length). Experimental details

are provided in the Experimental Section. Abbreviations: Ade, adenosine; AICA, 5-aminoimidazole-4-carboxamide; and Thd, thymidine.

densation of 14a−f with 15 at 45 °C in the presence of sodium

acetate (yields: 52−70%). Subsequent hydrolysis with 3 N

NaOH followed by coupling with diethyl ^L-glutamate using N-

methyl morpholine and 2,4-dimethoxy-6-chlorotriazine as the

activating agents aﬀorded diesters 17a−f (yields: 51−78% over

two steps). Final saponiﬁcation of the diesters with 1 N NaOH

provided target compounds 5−10 (yields: 88−97%).

The antiproliferative eﬀects of compounds 1−4 toward the

FRα-expressing cell lines were abolished in the presence of

excess (200 nM) folic acid, establishing that the cellular uptake

of these analogues is mediated primarily via FRα¹⁵(Table 1).

5-Substituted pyrrolo[2,3-d]pyrimidine analogues with a

side-chain benzoyl group (analogues of compounds 1−4) and

bridge lengths from one-to-six carbons (compounds 5−10)

were synthesized (Scheme 1) and tested for their antiprolifer-

ative activities toward the CHO and human tumor cell line

models (Table 1). The 5-pyrrolo[2,3-d]pyrimidine analogue

with a 2-carbon bridge (compound 6) is PMX. 5-Pyrrole

substitution restored RFC-directed antiproliferative activity, as

reﬂected in the dramatically increased antiproliferative eﬀects of

this series toward PC43-10 cells compared to RFC-, PCFT-,

and FR-null R2 cells in the following order of potency

(numbers of bridge carbons are noted in parentheses): 6 (2C)

> 8 (4C) ∼7 (3C) > 9 (5C) ≫ 5 (1C) ∼10 (6C). For PCFT-

expressing R2/PCFT4 CHO cells and with the exception of

compound 6, antiproliferative eﬀects for the 5-substituted

analogues were attenuated compared to the 6-substituted

analogues (e.g., for direct comparisons between paired 5- and 6-

substituted analogues, compare IC₅₀’s for compounds 7−10 to

1−4, respectively, in Table 1).

With the FRα-expressing sublines, potencies were also

reduced for the 5-substituted compounds compared to the 6-

substituted analogues. These diﬀerences ranged from slight

(compare 2 to 8 and 3 to 9) to substantial (compare 1 to 7)

(Table 1). For both series, the 6-carbon bridge analogues

(compounds 4 and 10) were inactive. In contrast to the results

for PCFT- and RFC-expressing cells, the impacts of bridge

length were distinctly diﬀerent, with antiproliferative activities

toward FRα-expressing RT16 CHO cells in rank order 8 (4C)

> 6 (2C) > 9 (5C) > 7 (3C) ≫ 5 (1C) ∼ 10 (6C). An

analogous pattern of sensitivities was obtained with KB human

tumor cells, and for both the KB and RT16 cell lines, inhibitory

BIOLOGICAL EVALUATION

■

Transporter Selectivity for Inhibition of Cell Prolifer-

ation by 5- and 6-Substituted Pyrrolo[2,3-d]pyrimidine

Regioisomers. Our initial goal was to deﬁne better the SARs

for membrane transport of pyrrolo[2,3-d]pyrimidine antifolates

relating to the impact of 5- versus 6-substitutions on the pyrrole

ring, which is based on the established antitumor activities of 6-

substituted pyrrolo[2,3-d]pyrimidine-substituted compounds 1

and 2 with 3- and 4-bridge carbons, respectively,^15,20and the

preclinical and clinical eﬃcacy of PMX,²¹a 5-substituted

pyrrolo[2,3-d]pyrimidine with a 2-carbon bridge (Figure 2). In

our previous studies,^15,20compounds 1 and 2 were eﬃciently

internalized by tumor cells via FRα and PCFT but not by RFC,

whereas PMX was a substrate by all three systems.¹⁵⁻²⁰

Transporter speciﬁcities are exempliﬁed by patterns of

inhibition of cell proliferation toward isogenic Chinese hamster

ovary (CHO) cell models engineered to express human RFC

(PC43-10), PCFT (R2/PCFT4), or FRα (RT16)¹⁵⁻¹⁹treated

with compounds 1−4 (Table 1). The results are compared to

those for RFC-, PCFT-, and FR-null MTXRIIOua^R2-4 (R2)

CHO cells. As previously reported,^15,20compounds 1−4

inhibited proliferation of FRα-expressing RT16 and R2/

PCFT4 cells in the following order of potency: 1 > 2 > 3 >

4. However, compounds 1−4 were largely inactive toward

RFC-expressing PC43-10 and R2 cells. Analogous results were

obtained with KB nasopharyngeal carcinoma cells that express

elevated FRα along with modest levels of RFC and PCFT.¹⁶

10019

dx.doi.org/10.1021/jm401328u | J. Med. Chem. 2013, 56, 10016−10032

Journal of Medicinal Chemistry

Article

Figure 3. Inhibition of RFC-mediated transport and relative FRα-binding aﬃnities by 6- (compound 2) and 5-substituted (compound 8)

pyrrolo[2,3-d]pyrimidine antifolates. (A) To determine if novel antifolates were able to bind human RFC and inhibit transport of [³H]MTX,

reﬂecting transport of the novel analogues by this mechanism, we used PC43-10 CHO cells ectopically expressing human RFC but not FRs or

PCFT. Uptake of 0.5 μM [³H]MTX was measured in the absence or presence of 10 μM nonradioactive inhibitors, including compound (Cpd) 2 or

8 or the classical agents PMX and MTX. Transport results are compared to those for RFC-null R2 cells. (B) To assess the relative binding aﬃnities

of the novel antifolates to FRα, we performed direct competitive binding assays using [³H]folic acid (FA) in the FRα-engineered CHO cell line,

RT16, in the presence or absence of various concentrations of unlabeled pyrrolo[2,3-d]pyrimidine compounds 2 or 8 or folic acid, PMX, or MTX.¹⁵

eﬀects were substantially reversed by 200 nM folic acid,

conﬁrming FRα as an important uptake mechanism for these

compounds.

Leads to a Cytotoxic Response via Purine Nucleotide

Depletion. De novo purine nucleotide biosynthesis includes

10 reactions by which phosphoribosyl pyrophosphate (PRPP)

is converted into inosine monophosphate (IMP), the precursor

of AMP and GMP (Figure 4). There are two folate-dependent

enzymes in the pathway that are possible targets for folate-

based therapies including GARFTase (catalyzes steps 2, 3, and

5) and AICARFTase (catalyzes steps 9 and 10). Previous

studies established that GARFTase was the intracellular enzyme

target for LMTX^3,32as well as for compounds 1 and 2.¹⁵For

PMX, TS is the primary intracellular target, although modest

inhibitions of GARFTase and DHFR were also reported.²¹

Most recently, AICARFTase was implicated as a potentially

important secondary enzyme target for PMX (in the presence

of excess thymidine to circumvent TS inhibition) by nucleo-

side/AICA protection experiments and metabolic assays.^22,23

To identify the targeted pathway for 6-substituted com-

pounds 1 and 2, we previously used nucleoside protection

experiments with adenosine (60 μM) and thymidine (10 μM)

to distinguish de novo purine nucleotide from thymidylate

biosynthesis, respectively.^15−20,32To identify further the likely

folate-metabolizing-enzyme targets in purine nucleotide biosyn-

thesis (GARFTase versus AICARFTase), cells were treated

with the antifolates in the presence of AICA (320 μM), which is

metabolized to AICA ribonucleotide (ZMP), the substrate for

AICARFTase, thus bypassing the step catalyzed by GARF-

Tase^15−20,32(Figure 4).

To complement the results showing antiproliferative eﬀects

of the most potent 5-substituted analogue 8 toward PC43-10

CHO cells expressing human RFC, we measured competitive

inhibitions of [³H]MTX (0.5 μM) transport in this model

(Figure 3A). MTX is a well-characterized surrogate substrate

for RFC, and the extent of [³H]MTX transport inhibition by

competitive unlabeled (anti)folate ligands correlates with their

membrane transport by RFC.⁸For our study, we compared

inhibitory eﬀects of 10 μM compound 8 to those for

corresponding 6-substituted compound 2 as well as to PMX

and MTX. Transporter-null R2 cells were used as a measure of

the baseline [³H]MTX uptake. Whereas 2 was a poor inhibitor

of [³H]MTX uptake (<25% inhibition), 10 μM compound 8

was a reasonably good inhibitor (>90% inhibition), with an

inhibitory potency approximating that for PMX and a net

transport only slightly greater than the very low levels recorded

for R2 cells. Clearly, unlike the 6-substituted pyrrolo[2,3-

d]pyrimidine analogues, the 5-substituted compounds lack

speciﬁcity for FRα and/or PCFT over RFC.

Binding aﬃnities for FRα are expressed relative to that for

folic acid. For both the transport and binding assays,

experimental details are provided in the Experimental Section,

and the results are presented as mean values standard errors

from three experiments.

Compound 8 showed antiproliferative eﬀects toward FRα-

expressing RT16 CHO cells and KB human tumor cells (Table

1). To demonstrate drug binding to FRα directly, a prerequisite

of cellular uptake by this endocytotic mechanism, we used

competitive binding assays of [³H]folic acid with RT16 cells

treated with unlabeled ligands that compete for FRα bind-

ing¹⁵⁻¹⁹(Figure 3B). Relative FRα binding aﬃnities for the 5-

(compound 8) versus 6-substituted (compound 2) analogues

paralleled their antiproliferative activities. However, this

correlation did not quite extend to PMX (compound 6) or

to MTX, indicating that multiple factors in addition to FRα

binding aﬃnities contribute to net growth inhibition of these

FRα-expressing cells for the antifolates.

We used this approach for KB cells treated with compounds

7−9, with results compared to those for compound 2 and PMX

(compound 6) (Table 1; Figure 5 shows the nucleoside/AICA

protection results for PMX and for compound 2 compared to

those for compound 8). With compound 2, both adenosine and

AICA were completely protective, establishing de novo purine

biosynthesis and GARFTase as the principal cellular

targets.^15,20With PMX, thymidine, adenosine, and AICA

were all partially protective, albeit to diﬀerent extents.

Combined thymidine and adenosine completely protected KB

cells from the growth inhibitory eﬀects of PMX (not shown;

Table 1). The growth inhibitory eﬀects of the 5-substituted

compounds 7−9 with KB cells were unaﬀected by excess

thymidine but were completely reversed in the presence of

adenosine alone, indicating that de novo purine nucleotide

Inhibition of De Novo Purine Nucleotide Biosynthesis

by Both 5- and 6-Pyrrolo[2,3-d]pyrimidine Compounds

10020

dx.doi.org/10.1021/jm401328u | J. Med. Chem. 2013, 56, 10016−10032

Journal of Medicinal Chemistry

Article

d]pyrimidine ring system results in an altered inhibition of the

targeted enzymes.

Decreased purine nucleotide biosynthesis resulting from

inhibition of folate-dependent enzymes can be monitored by

following changes in ATP pools, a surrogate measure of total

purine nucleotides (Figure 6). KB tumor cells were treated for

24 h with novel 5-substituted analogue 8 or corresponding 6-

subsituted regioisomer compound 2. An additional incubation

was performed with PMX. ATP pools were extracted and

measured by an HPLC ion-pairing method.^20,33In KB cells

following a 24 h incubation, compound 8, like 2, potently

depleted ATP pools (>90%), whereas PMX was less eﬀective

(∼60% inhibition).

Inhibition of De Novo Purine Nucleotide Biosynthesis

by Compound 8 Results in Cytotoxicity and Apoptosis.

To conﬁrm that depletion of purine nucleotides by compound

8 results in cytotoxicity toward KB cells, we performed colony-

forming assays. In these experiments, KB tumor cells were

treated with various concentrations (0.5−10 μM) for 48 h, then

washed and plated without drug. Colonies were enumerated

after 10 days (Figure 7A). Following drug exposure, the

inhibitory eﬀects of compound 8 were irreversible with

complete loss of colony formation at the highest drug

concentrations, thereby establishing compound 8 as cytotoxic.

Programmed cell death type I (mitochondrial pathway)

requires ATP to form the death complex with cytochrome C

and caspase 9 and to carry out other energy-requiring processes

for enzymatic degradation and modiﬁcation of cellular

components.³⁴Previous studies suggested that 6-substituted

pyrrolo[2,3-d]pyrimidine antipurine antifolate 1 was capable of

inducing apoptosis, albeit to a lower extent than for

daunorubicin over 24 h.¹⁵Additional experiments were

performed with KB cells treated with compound 8 to assess

its eﬀects on cell cycle progression and induction of apoptosis,

with comparisons to PMX and compound 2. Cells were

exposed to 1 μM drugs for 48 or 96 h. Cultures were then

divided, with one fraction ﬁxed and stained with propidium

iodide (PI) for ﬂow cytometry analysis of cell cycle distribution

(Figure 7B). An additional fraction was assayed for apoptosis

by ﬂow cytometry with annexin V-ﬂuorescein isothiocyanate

(FITC) /7-amino actinomycin D (AAD) staining (Figure 7C).

At 48 h, minimal eﬀects on the cell cycle were observed

except for PMX, which induced a G₁/G_o-phase arrest relative to

the untreated (DMSO) control (Supporting Information,

Figure 1S). At 96 h, all compounds tested increased the

percentages of cells in S-phase (albeit to diﬀerent extents) and

reduced those in the G₁/G_o- and G₂/M-phases of the cell cycle

(Figure 7B). A prominent increase in the sub-G₁population at

96 h was also seen (Figure 7B; most notable for compounds 2

and 8), indicating that apoptosis was induced. Annexin V-

FITC/7-AAD staining was used to monitor the induction of

apoptosis further. For compounds 2 and 8 as well as PMX,

positive annexin V-FITC staining increased over time (results

for 96 h are shown in Figure 7C; additional data at 48 h of drug

exposure are shown in the Supporting Information, Figure 1S),

with distinct increases in both early (annexin-V-FITC^high/7-

AAD^low) and late (annexin-V-FITC^high/7-AAD^high) apoptotic

fractions compared those for the untreated control (i.e., the

fractions of cells in early apoptosis were 33.8 and 0.49% for

compound 8-treated and control cells, respectively, whereas the

fractions in late apoptosis at 96 h were 26.7 and 2.1%,

respectively). Nearly identical results were obtained with

compound 2, although for PMX the distribution between the

Figure 4. De novo purine nucleotide biosynthesis and relationship to

AMPK. The de novo purine nucleotide biosynthetic pathway including

the steps from phosphoribosyl pyrophosphate (PRPP) to IMP is

shown. Reactions 1, 4, and 8 are catalyzed by glutamine

phosphoribosylpyrophosphate amidotransferase, formylglycinamide

ribonucleotide synthase, and adenylosuccinate lyase, respectively.

Reactions 2, 3, and 5 are catalyzed by the trifunctional glycinamide

ribonucleotide (GAR) formyltransferase (GARFTase), which contains

GAR synthase (reaction 2), GAR formyltransferase (GARFTase;

reaction 3), and 5-aminoimidazole ribonucleotide synthase (reaction

5) activities. Reactions 6 and 7 are catalyzed by the bifunctional

phosphoribosylaminoimidazole carboxylase/phosphoribosylaminoimi-

dazole succinocarboxamide synthetase enzyme, which contains

carboxyaminoimidazole ribonucleotide synthase (reaction 6) and 5-

aminoimidazole-4-(N-succinylocarboxamide) ribonucleotide synthase

(reaction 7) activities. Reactions 9 and 10 are catalyzed by 5-

aminoimidazole-4-carboxamide ribonucleotide formyltransferase (AI-

CARFTase)/IMP cyclohydrolase that catalyzes the last two steps in

the pathway for de novo synthesis of IMP. Folate-dependent reactions

(reactions 3 and 9), in which 10-formyl tetrahydrofolate (THF) serves

as the one-carbon donor, are catalyzed by GARFTase and

AICARFTase. 5-Aminoimidazole-4-carboxamide (AICA) and AICA

ribonucleoside (AICAR) can be metabolized to 5-aminoimidazole-4-

carboxamide ribonucleotide monophosphate (ZMP) by either adenine

phosphoribosyl transferase (APRT) or adenosine kinase (AK), thus

circumventing the reaction catalyzed by GARFTase. The activation of

AMPK that results in inhibition of mTOR is also depicted.

Abbreviations: AICAR, 5-aminoimidazole-4-carboxamide ribonucleo-

side; FGAR, formyl glycinamide ribonucleotide; FAICAR, formyl 5-

aminoimidazole-4-carboxamide ribonucleotide; and MP, monophos-

phate.

synthesis was being targeted exclusively (rather than combined

thymidylate and purine nucleotide synthesis as for PMX).

Whereas AICA alone completely protected KB cells from the

inhibitory eﬀects of compound 9, AICA alone was only partially

protective with compounds 7 and 8. Thus, inhibition of purine

nucleotide biosynthesis by compounds 7 and 8 is likely a

composite eﬀect of dual inhibitions involving both GARFTase

and AICARFTase. These results strongly imply that shifting the

substituent from the 6- to the 5-position on the pyrrolo[2,3-

10021

dx.doi.org/10.1021/jm401328u | J. Med. Chem. 2013, 56, 10016−10032

Journal of Medicinal Chemistry

Article

Figure 5. Cell proliferation assays with protection by nucleosides including thymidine and adenosine and by 5-aminoimidazole-4-carboxamide

(AICA) to identify intracellular targets of PMX and compounds 2 and 8. To identify the targeted pathways and the folate-dependent intracellular

enzymes in KB cells treated with compound 8 (1−1000 nM), cell proliferation assays were performed in the presence of 10 μM thymidine, 60 μM

adenosine, or 320 μM AICA. The results were normalized to those for untreated cells (no drug) and for compound 8, the results are compared to

those for PMX and compound 2. Details are provided in the Experimental Section. The results shown are representative of triplicate experiments.

Analogous protection experiments were performed for compounds 7 and 9, and the results are summarized in Table 1 along with those for

compounds 2 and 8.

Cells were treated with trichloroacetic acid (TCA), and,

following deproteinization, acid-soluble metabolites were ether-

extracted and fractionated by ion-exchange chromatography,

permitting isolation and quantitation of [¹⁴C]formyl GAR. Both

5- and 6-substituted analogues showed a dose-dependent

decrease in the formation of [¹⁴C]formyl GAR. However,

compound 2 was more potent than either PMX or compound

8, with an IC₅₀for this compound exceeding those for the 5-

substituted compounds by 2- and 5-fold, respectively (Figure

8).

The antiproliferative eﬀects of PMX independent of its

impact on TS were previously attributed to its inhibition of

AICARFTase in addition to GARFTase, as reﬂected in the

incomplete protection by AICA in the presence of thymidine

Figure 6. Compounds 2 and 8 deplete ATP pools in KB cells. KB cells

and the accumulation of the AICARFTase substrate ZMP in

were treated with 1 μM of novel 5- and 6-substituted pyrrolo[2,3-

tumor cell lines treated with PMX, as measured by anion-

d]pyrimidine antifolates 2 and 8 or with PMX for 24 h. Nucleotides

were extracted and analyzed by ion-pair HPLC. Details are in the

exchange HPLC analysis.^22,23ZMP accumulation showed both

dose and time dependence for PMX and resulted in activation

of AMPK (via its role as an AMP mimetic) and suppression of

mTOR.

Experimental Section. The results are presented as mean values

standard errors.

To validate the ﬁndings from our AICA protection

experiments with compound 8 (Figure 6) and to explore

whether AICARFTase is a bona ﬁde target for this novel 5-

substituted pyrrolo[2,3-d]pyrimidine, analogous experiments

were performed to measure accumulation of ZMP in KB cells

treated with 1 μM compound 8 as a measure of AICARFTase

inhibition. Parallel assays in KB cells treated with PMX were

performed. ZMP accumulated in the presence of PMX and

compound 8 in a linear fashion for up to 48 h (not shown),

with PMX showing approximately 2-fold increased levels over

compound 8 (Figure 9A).

early and late apoptotic fractions was somewhat diﬀerent. Thus,

5- and 6-substituted pyrrolo[2,3-d]pyrimidines 2 and 8 can

both induce apoptosis, resulting in cell death, in spite of

depleted ATP pools.

Dual Inhibition of GARFTase and AICARFTase by 5-

Substituted Pyrrolo[2,3-d]pyrimidine Antifolates and

Downstream Eﬀects of Accumulated ZMP on AMPK.

Experiments were performed to measure cellular GARFTase

activity in KB tumor cells treated with the 6- (compound 2)

and 5-substituted (compound 8 and PMX) pyrrolo[2,3-

d]pyrimidines. For this purpose, we used an in situ activity

assay for GARFTase that measures incorporation of [¹⁴C]-

glycine into [¹⁴C]formyl GAR in the presence of azaserine over

16 h.^15−20,32

To verify the impact of ZMP accumulation on AMPK

phosphorylation in KB cells treated with compound 8 or with

PMX, western blot analysis was performed for total and

phosphorylated AMPK in KB cells treated with these antifolates

10022

dx.doi.org/10.1021/jm401328u | J. Med. Chem. 2013, 56, 10016−10032

Journal of Medicinal Chemistry

Article

Figure 7. Inhibition of de novo purine biosynthesis by compound 8 results in cytotoxicity and apoptosis. (A) KB cells (1000 cells) were treated with

compound 8 (0.5−10 μM) for 48 h, after which time the drug was removed and colonies were allowed to grow in drug-free medium over 10 days.

Colonies were enumerated as described in the Experimental Section. The results are presented as mean values

standard errors from three

experiments. (B) Cell cycle analysis. KB cells were treated with 1 μM antifolate (PMX, compound 2, or compound 8) for 96 h, washed, ﬁxed, and

stained with PI. Cell cycle distributions were analyzed by ﬂow cytometry. The sub-G₁fraction (white ﬁll) increased from 3.5% in the absence of

drugs upon treatment with PMX (17.3%) or compounds 2 (45.1%) or 8 (36.7%). For the nonsub-G₁population, the cell cycle distributions (colors

are green for G₁/G₀, yellow for S, and blue for G₂/M) are as follows: no additions, 67.4% G₁/G₀, 15.9% S, 13.2% G₂/M; PMX, 47.8% G₁/G₀, 27.5%

S, 7.5% G₂/M; compound 2, 22.3% G₁/G₀, 31.4% S, 1.2% G₂/M; and compound 8, 12.9% G₁/G₀, 48.2% S, 2.3% G₂/M. (C) Analysis of apoptosis by

annexin V-FITC/7-AAD staining and ﬂow cytometry is shown for KB cells treated with PMX or with compounds 2 or 8 for 96 h. The results are

compared to those for controls treated with DMSO in lieu of drug. The percentages of cells in each quadrant (viable cells (annexin V-FITC^low/7-

AAD^low; Q4), early apoptosis (annexin V-FITC^high/7-AAD^low; Q3), and late apoptosis/necrosis (annexin V-FITC^high/7-AAD^high; Q2)) are

summarized.

for 48 h compared to levels of phospho-AMPK resulting from

treatments with established AMPK activators including AICA,

AICA ribonucleoside (AICAR), and metformin²⁴(Figure 9B).

Although KB cells are LKB1-null (the principal upstream kinase

responsible for phosphorylating AMPK), AMPK is still

activated upon treatment with known AMPK activators (Figure

9B). Although AMPK is activated by PMX, compound 8 was

not signiﬁcantly activating despite the accumulation of ZMP.

Interestingly, neither AICA (not shown) nor metformin

(Supporting Information, Figure 2S) at concentrations found

to activate AMPK in Figure 9B (1 and 10 mM, respectively)

were particularly growth inhibitory toward KB cells (20 and

32% inhibition, respectively).

Molecular Modeling: Docking Studies of Compound 8

with GARFTase and AICARFTase. On the basis of the

cellular metabolic data that compound 8 inhibits both

GARFTase and AICARFTase in KB cells (Figures 8 and 9,

respectively), molecular modeling studies were performed.

Molecular modeling of compound 8 used LeadIT 1.3.0 and was

visualized using MOE 2011.10. Redocking the native crystal

structure ligands 10-CF₃CO-DDACTHF³⁵(Figure 10) for

human GARFTase (PDB ID 1NJS) and BW2315, 18,³⁶(Figure

10) for AICARFTase (PDB ID 1PL0) into their respective

crystal structures aﬀorded docked poses with a rmsd ∼1 Å.

Figure 11 shows the docked pose of compound 8 in the

human GARFTase active site. The cofactor binding pocket of

GARFTase is located at the interface between the N-terminal

mononucleotide binding domain and the C-terminal half of the

structure. The binding site for the folate cofactor moiety

consists of three parts: the pteridine binding cleft, the

benzoylglutamate region, and the formyl transfer region. The

docked pose shows the pyrrolo[2,3-d]pyrimidine scaﬀold of 8

to be buried deep in the active site and to occupy the same

location as the diaminopyrimidone ring in the native crystal

structure ligand (10-CF₃CO-DDACTHF). This orientation of

the scaﬀold permits the 2-amino moiety to form hydrogen

bonds to the backbone of Glu141 and Ser93. The N1 nitrogen

interacts with the backbone of Leu92 to form a hydrogen bond.

The 4-oxo moiety forms a hydrogen bond with Asp144 and

forms water-mediated hydrogen bonds with Asp142 and

Ala140 (not shown).

The molecule is oriented in a manner that aids the N7

nitrogen to form a hydrogen bond with Arg90. The

pyrrolo[2,3-d]pyrimidine scaﬀold resides in a hydrophobic

10023

dx.doi.org/10.1021/jm401328u | J. Med. Chem. 2013, 56, 10016−10032

Journal of Medicinal Chemistry

Article

compound 8 into the benzoylglutamate region of the protein.

The amide NH of the glutamate side chain forms a hydrogen

bond with Met89. The α-carboxylic acid of the glutamate side

chain interacts with Arg64 and additionally interacts with the

backbone of Ile91. The γ-carboxylic acid can form a water-

mediated hydrogen bond with Arg90. The interaction of the

ﬂexible glutamate side chain is very similar to the interaction

network observed for the glutamate side chain of 10-CF₃CO-

DDACTHF.

Figure 12 shows the overlay of the docked poses of 8 (blue)

with the crystal structure ligand 18 (red) (a potent inhibitor) in

human AICARFTase (PDB ID 1PL0). The pyrrolo[2,3-

d]pyrimidine scaﬀold of 8 occupies the same location as the

dihydroquinazoline scaﬀold of 18. Analogous to 18, the 2-NH₂

and N3 nitrogens of 8 interact with Asp546, whereas the 4-oxo

moiety forms a hydrogen bond with the side chain of Asn547.

The N7 nitrogen of 8 forms a hydrogen bond with the

backbone of Met312. The pyrrolo[2,3-d]pyrimidine scaﬀold of

8 forms hydrophobic interactions with Met312, Phe315, Ile452,

Pro543, and Phe544. The aryl glutamate section of 8 is oriented

similarly to the phenyl glutamate side chain of 18.

These docking results predict that 8 should bind and inhibit

the two folate-dependent purine biosynthetic enzymes

(GARFTase and AICARFTase) and are entirely consistent

with the results of our in situ metabolic assays (Figures 8 and

9).

Figure 8. In situ GARFTase assay. GARFTase activity and inhibition

by 5- and 6-substituted pyrrolo[2,3-d]pyrimidine analogues 2, 8, and

PMX were evaluated in situ with KB cells. Results are presented as

mean values

standard errors from three experiments. IC₅₀’s are

summarized in the table. Additional details are described in the

Experimental Section.

DISCUSSION

■

The development of rationally designed cytotoxic antifolates

remains an area of signiﬁcant interest to anticancer drug

development, as exempliﬁed by the relatively recent approval of

PMX³⁷and pralatrexate³⁸for cancer therapy more than 50

years since the introduction of MTX.¹Unfortunately, the

ubiquitous presence of RFC may contribute to untoward

toxicities of these classical antifolates toward normal tissues. On

the basis of this, growing attention has focused on developing

novel folate-based compounds with tumor targeting premised

on transport selectivity for FRs or PCFT over RFC,^11,13−20

a

pursuit that until recently^39,40continued to be challenged by

the lack of structural information on the major folate transport

systems.

The antitumor eﬃcacy for 5-substituted 2-carbon bridge

pyrrolo[2,3-d]pyrimidine PMX reﬂects its cellular uptake by all

of the major folate transport systems.¹⁶⁻²⁰Although the

corresponding 6-substituted analogue of PMX was pharmaco-

logically inert, with increasing bridge lengths the 6-substituted

pyrrolo[2,3-d]pyrimidines were surprisingly potent inhibitors of

cell proliferation, but membrane transport by RFC was lost.^15,20

The 6-pyrrolo[2,3-d]pyrimidine benzoyl side-chain analogues

with 3- and 4-bridge carbons (compounds 1 and 2,

respectively) were the most active of this series, reﬂecting

their FR and PCFT cellular uptake with little to no transport by

RFC as well as their potent inhibition of GARFTase, the ﬁrst

folate-dependent step in de novo purine nucleotide biosyn-

thesis leading to the synthesis of IMP.^15,20Inhibition of

GARFTase resulted in decreased de novo synthesis of purine

nucleotides and ATP depletion, leading to cytotoxicity and

apoptosis.^15,20

Figure 9. Accumulation of ZMP by 5-substituted pyrrolo[2,3-

d]pyrimidines PMX and compound 8. (A) ZMP (AICA ribonucleo-

tide) accumulation was measured in KB cells treated for 48 h with the

5-substituted pyrrolo[2,3-d]pyrimidine antifolates 8 and PMX.

Methods for extracting nucleotide metabolites and for measuring

ZMP levels are detailed in the Experimental Section. The results are

shown as mean values standard errors for three experiments. (B) KB

cells were treated as indicated for 48 h followed by Western blot

analysis of phosphorylated AMPK. Blots were probed with antibodies

to phospho- and total AMPK as well as β-actin. Details are included in

the Experimental Section. Abbreviations: AICA, 5-aminoimidazole-4-

carboxamide; AICAR, 5-aminoimidazole-4-carboxamide ribonucleo-

side; and NA, no additions.

In this article, our goal was to establish further a

comprehensive SAR for the 5- versus 6-substituted pyrrolo-

[2,3-d]pyrimidine ring system in relation to folate transporter

speciﬁcity and inhibition of intracellular targets based in part on

ﬁndings of transporter promiscuity for PMX¹⁶⁻²⁰and reports

pocket formed by Leu85, Ile91 (not shown), Leu92, Val97, and

the folate binding loop, residues 141−146. The ﬂexible four-

atom side chain helps to orient the benzoyl moiety of

10024

dx.doi.org/10.1021/jm401328u | J. Med. Chem. 2013, 56, 10016−10032

Journal of Medicinal Chemistry

Article

Figure 10. Structures of cocrystallized ligands 10-CF₃CO-DDACTHF (human GARFTase, PDB ID 1NJS) and BW2315 18 (human AICARFTase,

PDB ID 1PL0).

Figure 11. Stereoview overlay of the docked pose of compound 8 (blue) with 10-CF₃CO-DDACTHF (red) in human GARFTase (PDB ID 1NJS).

Figure 12. Stereoview overlay of the docked pose of compound 8 (blue) with 18 (red) in human AICARFTase (PDB ID 1PL0).

that PMX may in part derive its antitumor eﬀects by inhibiting

AICARTase, resulting in accumulation of ZMP and activation

of AMPK in addition to inhibiting TS.^22,23Compared to the 6-

substituted analogues, we found that the 5-substituted pyrrolo-

[2,3-d]pyrimidine compounds in general (i) were better

substrates for RFC, with the highest antiproliferative activities

for compounds with two-to-four bridge carbons independent of

the nature of the side-chain ring system; (ii) were poorer

substrates for PCFT with the exception of 2-carbon bridge

compound PMX (6); and (iii) preserved FRα cellular uptake,

with 4-carbon bridge analogue 8 showing the greatest FRα

potency in this expanded 5-substituted series of analogues

without TS inhibitory activity.

Compound 8 was cytotoxic, as reﬂected in the loss of

clonogenicity in KB cells, and eﬀected S-phase accumulation

and apoptosis over 96 h. Like its 6-substituted counterpart

10025

dx.doi.org/10.1021/jm401328u | J. Med. Chem. 2013, 56, 10016−10032

Journal of Medicinal Chemistry

Article

compound 1,²⁰compound 8 resulted in ATP depletion,

although this appeared to be only partly due to inhibition of

GARFTase. Rather, a unique feature of 5-substituted analogue

8 relates to its dual inhibition of both the GARFTase and

AICARFTase reactions in de novo purine nucleotide biosyn-

thesis. Although PMX also inhibited GARFTase, the ﬁnding

that the AICARFTase substrate ZMP accumulates in the

presence of this drug suggests that GARFTase inhibition must

be incomplete and that AICARFTase must also be a target as

previously described,^22,23whereas for PMX, total purine

nucleotides, as reﬂected in ATP pools, were signiﬁcantly

preserved. Compound 8 also inhibited AICARFTase, as

reﬂected in the incomplete protection in cell proliferation

assays by AICA, and demonstrated accumulation of the

AICARFTase substrate ZMP, albeit to a lesser extent than for

a comparable concentration of PMX. ATP pools were

signiﬁcantly depleted (∼90%) with compound 8.

Previous studies suggested that AMPK activation secondary

to ZMP accumulation and AICARFTase inhibition may result

in inhibition of mTOR and contribute to the antitumor eﬀects

of high levels of PMX in the presence of thymidine.^22,23In the

present study, we used KB tumor cells treated with PMX or

with compound 8. In this LKB1-deﬁcient tumor cell line

treated with PMX or compound 8, ZMP accumulation was

accompanied by AMPK activation in the presence of PMX but

not compound 8. AMPK activation was also seen upon

treatment with AICA, AICAR, or metformin. However, ZMP

accumulation and AMPK activation did not appear to be likely

to contribute to the antiproliferative eﬀects of PMX in KB cells,

as neither AICA nor metformin (both of which activated

AMPK more than PMX) potently inhibited cell proliferation.

The antiproliferative and cytotoxic eﬀects of compound 8

toward KB cells must derive from a depletion of ATP pools

secondary to inhibition of both GARFTase and AICARFTase.

In summary, a series of classical 5-substituted pyrrolo[2,3-

d]pyrimidine antifolates, 5−10, was designed and synthesized

as a hybrid of PMX and 6-substituted pyrrolo[2,3-d]-

pyrimidines 1−4. Compound 8, the lead compound of this

series, has an IC₅₀of 12.7 nM against KB tumor cells in culture.

Rather than only targeting GARFTase, as with 1 and 2, both

GARFTase and AICARFTase were identiﬁed as the intra-

cellular targets of 8. Furthermore, on the basis of the patterns of

inhibition of cell proliferation toward isogenic CHO cells

expressing FRα or RFC, compound 8 appears to be transported

by both FRα and RFC. The unique dual GARFTase and

AICARFTase inhibition of compound 8 makes it an excellent

lead compound for further study, including the design of

additional antitumor analogues to optimize cellular folate

uptake selectivity by FRs and inhibition of folate metabolizing

enzymes, leading to a new generation of potent tumor-targeted

agents. Furthermore, the dual enzyme inhibition results in

potent inhibition of purine nucleotide biosynthesis and could

preserve antitumor activity in the event that tumors become

resistant as a result of alterations in one or the other targeted

enzyme. Such dual-targeted inhibitors would seem to have

substantial advantages over currently used agents that

principally target only one intracellular enzyme target.

electronic thermometer and are uncorrected. Nuclear magnetic

resonance spectra for proton (¹H NMR) were recorded on either a

Bruker WH-400 (400 MHz) spectrometer or a Bruker WH-500 (500

MHz) spectrometer. The chemical-shift values are expressed in ppm

(parts per million) relative to tetramethylsilane as an internal standard:

s, singlet; d, doublet; t, triplet; q, quartet; m, multiplet; and br, broad

singlet. Mass spectra were recorded on a VG-7070 double-focusing

mass spectrometer or in a LKB-9000 instrument in the electron

ionization (EI) mode. Chemical names follow IUPAC nomenclature.

Thin-layer chromatography (TLC) was performed on Whatman Sil

G/UV254 silica gel plates with a ﬂuorescent indicator, and the spots

were visualized under 254 and 365 nm illumination. All analytical

samples were homogeneous on TLC in three diﬀerent solvent systems.

The proportions of solvents used for TLC are by volume. Column

chromatography was performed on a 230−400 mesh silica gel (Fisher,

Somerville, NJ) column. Elemental analyses were performed by

Atlantic Microlab, Inc., Norcross, GA. Element compositions are

within 0.4% of the calculated values. Fractional moles of water

frequently found in the analytical sample of antifolates could not be

prevented in spite of 24−48 h of drying in vacuo and were conﬁrmed,

where possible, by their presence in the ¹H NMR spectra. All solvents

and chemicals were purchased from Aldrich Chemical Co. or Fisher

Scientiﬁc and were used as received. For all of the compounds

submitted for biological evaluation, elemental analysis (C,H, N) was

performed to conﬁrm >95% purity (Supporting Information, Table

1S).

Ethyl 4-(3-Oxopropyl)benzoate (13a). To a solution of ethyl 4-

iodobenzoate (11) (5 mmol, 1.38 g) in 20 mL of anhydrous DMF

were added prop-2-en-1-ol 12a (6 mmol, 348 mg), LiCl (5 mmol, 210

mg), LiOAc (12.5 mmol, 850 mg), Bu₄NCl (2.5 mmol, 840 mg), and

Pd(OAc)₂(0.3 mmol, 60 mg), and the mixture was stirred at 70 °C for

3 h. TLC (hexane/EtOAc, 3:1) showed the disappearance of the

starting material (R_f= 0.70) and formation of one major spot at R_f=

0.60. To the reaction mixture cooled to room temperature was added

ethyl acetate (30 mL). The resulting solution was extracted with H₂O

(10 mL × 3) and dried over Na₂SO₄. After evaporation of solvent, the

residue was loaded on a silica gel column (4 × 20 cm) and ﬂash-

chromatographed with hexane/EtOAc (2:1), and the desired fractions

were pooled. After evaporation of the solvent, the residue was dried in

vacuo using P₂O₅to aﬀord 13a (0.83 g, yield 83%) as a colorless

liquid. R_f= 0.60 (hexane/EtOAc, 3:1). ¹H NMR (500 MHz, CDCl₃) δ

1.38 (t, 3H, CH₃, J = 7.0 Hz), 2.81 (t, 2H, CH₂, J = 7.5 Hz), 3.01 (t,

2H, CH₂, J = 7.5 Hz), 4.36 (q, 2H, CH₂, J = 7.0 Hz), 7.26 (d, 2H, 2

CH, J = 4.0 Hz), 7.97 (d, 2H, 2 CH, J = 4.3 Hz), 9.82 (t, 1H, CHO, J

= 1.5 Hz).

Ethyl 4-(4-Oxobutyl)benzoate (13b). Compound 13b was

synthesized as described for 13a (yield 88%) as a colorless liquid. R_f

1

= 0.62 (hexane/EtOAc, 3:1). H NMR (500 MHz, CDCl₃) δ 1.39 (t,

3H, CH₃, J = 7.0 Hz), 1.95−2.01 (m, 2H, CH₂), 2.46 (t, 2H, CH₂, J =

7.5 Hz), 2.71 (t, 2H, CH₂, J = 7.5 Hz), 4.37 (q, 2H, CH₂, J = 7.0 Hz),

7.24 (d, 2H, 2 CH, J = 4.3 Hz), 7.97 (d, 2H, 2 CH, J = 4.3 Hz), 9.76 (t,

1H, CHO, J = 1.5 Hz).

Ethyl 4-(5-Oxopentyl)benzoate (13c). Compound 13c was

synthesized as described for 13a: (yield 84%) as a colorless liquid. R_f=

1

0.62 (hexane/EtOAc, 3:1). H NMR (500 MHz, CDCl₃) δ 1.38 (t,

3H, CH₃, J = 7.0 Hz), 1.66−1.70 (m, 4H, 2 CH₂), 2.46 (t, 2H, CH₂, J

= 7.0 Hz), 2.69 (t, 2H, CH₂, J = 7.0 Hz), 4.36 (q, 2H, CH₂, J = 7.0

Hz), 7.23 (d, 2H, 2 CH, J = 4.0 Hz), 7.96 (d, 2H, 2 CH, J = 4.0 Hz),

9.75 (t, 1H, CHO, J = 1.5 Hz).

Ethyl 4-(6-Oxohexyl)benzoate (13d). Compound 13d was

synthesized as described for 13a (yield 86%) as a colorless liquid. R_f

1

= 0.62 (hexane/EtOAc, 3:1). H NMR (500 MHz, CDCl₃) δ 1.25−

1.27 (m, 2H, CH₂), 1.38 (t, 3H, CH₃, J = 7.0 Hz), 1.64−1.68 (m, 4H,

2 CH₂), 2.42 (t, 2H, CH₂, J = 7.5 Hz), 2.66 (t, 2H, CH₂, J = 7.5 Hz),

4.36 (q, 2H, CH₂, J = 7.0 Hz), 7.23 (d, 2H, 2 CH, J = 4.3 Hz), 7.96 (d,

2H, 2 CH, J = 4.3 Hz), 9.75 (t, 1H, CHO, J = 2.0 Hz).

EXPERIMENTAL SECTION

■

Ethyl 4-(7-Oxoheptyl)benzoate (13e). Compound 13e was

synthesized as described for 13a (yield 92%) as a colorless liquid. R_f=

0.63 (hexane/EtOAc, 3:1). ¹H NMR (500 MHz, CDCl₃) δ 1.34−1.36

(m, 4H, 2 CH₂), 1.38 (t, 3H, CH₃, J = 7.0 Hz), 1.58−1.65 (m, 4H, 2

All evaporations were carried out in vacuo with a rotary evaporator.

Analytical samples were dried in vacuo (0.2 mmHg) in a CHEM-DRY

drying apparatus over P₂O₅at 80 °C. Melting points were determined

on a MEL-TEMP II melting point apparatus with a FLUKE 51 K/J

10026

dx.doi.org/10.1021/jm401328u | J. Med. Chem. 2013, 56, 10016−10032

Journal of Medicinal Chemistry

Article

CH₂), 2.42 (t, 2H, CH₂, J = 7.5 Hz), 2.66 (t, 2H, CH₂, J = 7.5 Hz),

4.36 (q, 2H, CH₂, J = 7.0 Hz), 7.23 (d, 2H, 2 CH, J = 4.3 Hz), 7.95 (d,

2H, 2 CH, J = 4.3 Hz), 9.75 (t, 1H, CHO, J = 2.0 Hz).

starting materials and the formation of one major spot at R_f= 0.38

(CHCl₃/MeOH, 5:1). After evaporation of the solvent, CH₃OH (10

mL) was added followed by silica gel (3 g). Evaporation of the solvent

aﬀorded a plug, which was loaded onto a silica gel column (3.5 cm ×

15 cm) and eluted initially with CHCl₃followed by 10% MeOH in

CHCl₃and then 15% MeOH in CHCl₃. Fractions showing R_f= 0.38

were pooled and evaporated to aﬀord 16b (230 mg, yield 70%) as a

Ethyl 4-(8-Oxooctyl)benzoate (13f). Compound 13f was

synthesized as described for 13a (yield 87%) as a colorless liquid. R_f

1

= 0.63 (hexane/EtOAc, 3:1). H NMR (500 MHz, CDCl₃) δ 1.28−

1.34 (m, 6H, 3 CH₂), 1.38 (t, 3H, CH₃, J = 7.0 Hz), 1.60−1.63 (m,

4H, 2 CH₂), 2.42 (t, 2H, CH₂, J = 7.5 Hz), 2.64 (t, 2H, CH₂, J = 7.5

Hz), 4.36 (q, 2H, CH₂, J = 7.0 Hz), 7.23 (d, 2H, 2 CH, J = 4.3 Hz),

7.95 (d, 2H, 2 CH, J = 4.3 Hz), 9.75 (t, 1H, CHO, J = 2.0 Hz).

Ethyl 4-(3-Bromo-4-oxobutyl)benzoate (14b). To a solution of

aldehyde 13b (1 mmol, 220 mg) in 5 mL of anhydrous Et₂O were

added 5,5-dibromo-2,2-dimethyl-4,6-dioxo-1,3-dioxane (0.5 mmol, 150

mg) and 2 N HCl in Et₂O solution (0.1 mmol, 50 μL), and the

mixture was stirred at room temperature for 24 h. TLC (hexane/

EtOAc, 3:1) showed the disappearance of the starting material (R_f=

0.62) and formation of one major spot at R_f= 0.50. The reaction

solution was washed with 5% NaHCO₃solution, extracted with H₂O

(10 mL × 3), and dried over Na₂SO₄. After evaporation of the solvent,

the residue was loaded on a silica gel column (4 × 20 cm) and ﬂash-

chromatographed with hexane/EtOAc (2:1), and the desired fractions

were pooled. After evaporation of the solvent, the residue was dried in

vacuo using P₂O₅to aﬀord 14b (235 mg, yield 78%) as a colorless oil.

1

pink solid. mp > 250 °C decomposed. H NMR (500 MHz, DMSO-

d₆) δ 1.30 (t, 3H, CH₃, J = 7.5 Hz), 2.84 (t, 2H, CH₂, J = 7.0 Hz), 2.99

(t, 2H, CH₂, J = 7.0 Hz), 4.29 (q, 2H, CH₂, J = 7.5 Hz), 6.03 (s, 2H, 2-

NH₂), 6.29 (d, 1H, CH, J = 1.0 Hz), 7.33 (d, 2H, CH, J = 4.0 Hz),

7.85 (d, 2H, CH, J = 4.0 Hz), 10.21 (s, 1H, 3-NH), 10.60 (s, 1H, 7-

NH).

Ethyl 4-((2-Amino-4-oxo-4,7-dihydro-3H-pyrrolo[2,3-d]-

pyrimidin-5-yl)methyl)benzoate (16a). Compound 16a was

synthesized as described for 16b (yield 35%) over two steps as a

yellow solid. mp > 232 °C decomposed. R_f= 0.42 (CHCl₃/MeOH,

5:1). ¹H NMR (500 MHz, DMSO-d₆) δ 1.29 (t, 3H, CH₃, J = 7.5 Hz),

4.00 (s, 2H, CH₂), 4.28 (q, 2H, CH₂, J = 7.5 Hz), 6.00 (s, 2H, 2-NH₂),

6.34 (d, 1H, CH, J = 1.0 Hz), 7.41 (d, 2H, CH, J = 4.0 Hz), 7.83 (d,

2H, CH, J = 4.0 Hz), 10.12 (s, 1H, 3-NH), 10.74 (s, 1H, 7-NH).

Ethyl 4-(3-(2-Amino-4-oxo-4,7-dihydro-3H-pyrrolo[2,3-d]-

pyrimidin-5-yl)propyl)benzoate (16c). Compound 16c was

synthesized as described for 16b (yield 68%) as a purple solid. mp

1

R_f= 0.50 (hexane/EtOAc, 3:1). H NMR (500 MHz, CDCl₃) δ 1.39

1

> 264 °C decomposed. R_f= 0.42 (CHCl₃/MeOH, 5:1). H NMR

(t, 3H, CH₃, J = 7.0 Hz), 2.19−2.27 (m, 1H, CH₂), 2.34−2.42 (m, 1H,

CH₂), 2.79−2.85 (m, 1H, CH₂), 2.91−2.97 (m, 1H, CH₂), 4.16−4.18

(m, 1H, CHBr), 4.37 (q, 2H, CH₂, J = 7.0 Hz), 7.28 (d, 2H, 2 CH, J =

4.3 Hz), 7.99 (d, 2H, 2 CH, J = 4.3 Hz), 9.46 (d, 1H, CHO, J = 1.0

Hz).

Ethyl 4-(2-Bromo-3-oxopropyl)benzoate (14a). Compound

14a was synthesized as described for 14b. 14a is not stable and was

used directly in the next step without puriﬁcation.

(500 MHz, DMSO-d₆) δ 1.30 (t, 3H, CH₃, J = 7.5 Hz), 1.89−1.95 (m,

2H, CH₂), 2.57 (t, 2H, CH₂, J = 7.5 Hz), 2.66 (t, 2H, CH₂, J = 7.5

Hz), 4.29 (q, 2H, CH₂, J = 7.5 Hz), 5.96 (s, 2H, 2-NH₂), 6.35 (d, 1H,

CH, J = 1.0 Hz), 7.34 (d, 2H, CH, J = 4.0 Hz), 7.86 (d, 2H, CH, J =

4.0 Hz), 10.08 (s, 1H, 3-NH), 10.63 (s, 1H, 7-NH).

Ethyl 4-(4-(2-Amino-4-oxo-4,7-dihydro-3H-pyrrolo[2,3-d]-

pyrimidin-5-yl)butyl)benzoate (16d). Compound 16d was synthe-

sized as described for 16b (yield 53%) as a blue solid. mp > 256 °C

Ethyl 4-(4-Bromo-5-oxopentyl)benzoate (14c). Compound

1

decomposed. R_f= 0.44 (CHCl₃/MeOH, 5:1). H NMR (500 MHz,

14c was synthesized as described for 14b (yield 77%) as a colorless

1

DMSO-d₆) δ 1.30 (t, 3H, CH₃, J = 7.5 Hz), 1.56−1.62 (m, 4H, 2

CH₂), 2.58 (t, 2H, CH₂, J = 7.5 Hz), 2.65 (t, 2H, CH₂, J = 7.5 Hz),

4.29 (q, 2H, CH₂, J = 7.5 Hz), 5.94 (s, 2H, 2-NH₂), 6.31 (d, 1H, CH, J

= 1.0 Hz), 7.33 (d, 2H, CH, J = 4.0 Hz), 7.85 (d, 2H, CH, J = 4.0 Hz),

10.08 (s, 1H, 3-NH), 10.59 (s, 1H, 7-NH).

oil. R_f= 0.52 (hexane/EtOAc, 3:1). H NMR (500 MHz, CDCl₃) δ

1.39 (t, 3H, CH₃, J = 7.0 Hz), 1.73−1.82 (m, 2H, CH₂), 1.86−1.96

(m, 1H, CH₂), 2.03−2.10 (m, 1H, CH₂), 2.71−2.75 (m, 2H, CH₂),

4.21−4.25 (m, 1H, CHBr), 4.37 (q, 2H, CH₂, J = 7.0 Hz), 7.24 (d, 2H,

2 CH, J = 4.3 Hz), 7.97 (d, 2H, 2 CH, J = 4.3 Hz), 9.42 (d, 1H, CHO,

J = 1.3 Hz).

Ethyl 4-(5-(2-Amino-4-oxo-4,7-dihydro-3H-pyrrolo[2,3-d]-

pyrimidin-5-yl)pentyl)benzoate (16e). Compound 16e was

synthesized as described for 16b (yield 60%) as a pink solid. mp >

Ethyl 4-(5-Bromo-6-oxohexyl)benzoate (14d). Compound 14d

was synthesized as described for 14b (yield 75%) as a colorless oil. R_f=

1

212 °C decomposed. R_f= 0.45 (CHCl₃/MeOH, 5:1). H NMR (500

0.52 (hexane/EtOAc, 3:1). H NMR (400M, CDCl₃) δ 1.39 (t, 3H,

MHz, DMSO-d₆) δ 1.29−1.32 (m, 5H, CH₃, CH₂), 1.58−1.62 (m,

4H, 2 CH₂), 2.53 (t, 2H, CH₂, J = 7.5 Hz), 2.63 (t, 2H, CH₂, J = 7.5

Hz), 4.29 (q, 2H, CH₂, J = 7.5 Hz), 5.94 (s, 2H, 2-NH₂), 6.30 (d, 1H,

CH, J = 1.0 Hz), 7.33 (d, 2H, CH, J = 4.0 Hz), 7.85 (d, 2H, CH, J =

4.0 Hz), 10.08 (s, 1H, 3-NH), 10.58 (s, 1H, 7-NH).

CH₃, J = 7.0 Hz), 1.63−1.75 (m, 4H, 2 CH₂), 1.87−1.99 (m, 1H,

CH₂), 2.03−2.12 (m, 1H, CH₂), 2.69 (t, 2H, CH₂, J = 7.0 Hz), 4.19−

4.22 (m, 1H, CHBr), 4.36 (q, 2H, CH₂, J = 7.0 Hz), 7.23 (d, 2H, 2

CH, J = 4.0 Hz), 7.96 (d, 2H, 2 CH, J = 4.0 Hz), 9.42 (d, 1H, CHO, J

= 1.2 Hz).

Ethyl 4-(6-(2-Amino-4-oxo-4,7-dihydro-3H-pyrrolo[2,3-d]-

pyrimidin-5-yl)hexyl)benzoate (16f). Compound 16f was synthe-

sized as described for 16b (yield 52%) as a pink solid. mp > 208 °C

Ethyl 4-(6-Bromo-7-oxoheptyl)benzoate (14e). Compound

14e was synthesized as described for 14b (yield 77%) as a colorless

1

oil. R_f= 0.54 (hexane/EtOAc, 3:1). H NMR (500 MHz, CDCl₃) δ

1

decomposed. R_f= 0.45 (CHCl₃/MeOH, 5:1). H NMR (500 MHz,

1.30−1.40 (m, 2H, CH₂), 1.38 (t, 3H, CH₃, J = 7.0 Hz), 1.62−1.70

(m, 4H, 2 CH₂), 1.86−1.94 (m, 1H, CH₂), 2.00−2.07 (m, 1H, CH₂),

2.66 (t, 2H, CH₂, J = 7.5 Hz), 4.18−4.22 (m, 1H, CHBr), 4.36 (q, 2H,

CH₂, J = 7.0 Hz), 7.23 (d, 2H, 2 CH, J = 4.3 Hz), 7.95 (d, 2H, 2 CH, J

= 4.0 Hz), 9.42 (d, 1H, CHO, J = 1.5 Hz).

Ethyl 4-(7-Bromo-8-oxooctyl)benzoate (14f). Compound 14f

was synthesized as described for 14b (yield 74%) as a colorless oil. R_f=

0.56 (hexane/EtOAc, 3:1). ¹H NMR (500 MHz, CDCl₃) δ 1.30−1.40

(m, 4H, 2 CH₂), 1.38 (t, 3H, CH₃, J = 7.0 Hz), 1.62−1.64 (m, 4H, 2

CH₂), 1.86−1.94 (m, 1H, CH₂), 2.00−2.07 (m, 1H, CH₂), 2.65 (t,

2H, CH₂, J = 7.5 Hz), 4.19−4.22 (m, 1H, CHBr), 4.36 (q, 2H, CH₂, J

= 7.0 Hz), 7.23 (d, 2H, 2 CH, J = 4.0 Hz), 7.95 (d, 2H, 2 CH, J = 4.0

Hz), 9.42 (d, 1H, CHO, J = 1.5 Hz).

Ethyl 4-(2-(2-Amino-4-oxo-4,7-dihydro-3H-pyrrolo[2,3-d]-

pyrimidin-5-yl)ethyl)benzoate (16b). To a solution of 2,6-

diamino-64-oxopyrimidine 15 (151 mg, 1.2 mmol) and sodium

acetate (180 mg, 2.2 mmol) in water (3 mL) and methanol (3 mL)

was added α-bromo aldehyde 14b (330 mg, 1.1 mmol) The reaction

mixture was stirred at 45 °C for 3 h. TLC showed the disappearance of

DMSO-d₆) δ 1.27−1.33 (m, 7H, CH₃, 2 CH₂), 1.53−1.60 (m, 4H, 2

CH₂), 2.53 (t, 2H, CH₂, J = 7.5 Hz), 2.62 (t, 2H, CH₂, J = 7.5 Hz),

4.29 (q, 2H, CH₂, J = 7.5 Hz), 5.94 (s, 2H, 2-NH₂), 6.30 (d, 1H, CH, J

= 1.0 Hz), 7.33 (d, 2H, CH, J = 4.0 Hz), 7.86 (d, 2H, CH, J = 4.0 Hz),

10.06 (s, 1H, 3-NH), 10.57 (s, 1H, 7-NH).

(S)-Diethyl 2-(4-((2-Amino-4-oxo-4,7-dihydro-3H-pyrrolo-

[2,3-d]pyrimidin-5-yl)methyl)benzamido)pentanedioate (17a).

To a suspension of 16a (100 mg, 0.35 mmol) in 10 mL of CH₃OH

was added 3 N NaOH (10 mL). The resulting mixture was stirred

under N₂at 40−50 °C for 24 h. TLC indicated the disappearance of

starting material and the formation of one major spot at the origin.

The resulting solution was passed through Celite and washed with a

minimum amount of CH₃OH. The combined ﬁltrate was evaporated

under reduced pressure to dryness. To this residue was added distilled

water (10 mL). The solution was cooled in an ice bath, and the pH

was adjusted 3 to 4 using 3 N HCl. The resulting suspension was

chilled in a dry ice/acetone bath and thawed to 4 °C overnight in a

refrigerator. The precipitate was ﬁltered, washed with cold water, and

10027

dx.doi.org/10.1021/jm401328u | J. Med. Chem. 2013, 56, 10016−10032

Journal of Medicinal Chemistry

Article

dried in a desiccator under reduced pressure using P₂O₅to a brown

powder, which was used directly for the next step. To a solution of this

brown powder in anhydrous DMF (10 mL) were added 6-chloro-2,4-

dimethoxy-1,3,5-triazine (72 mg, 0.42 mmol) and N-methylmorpho-

line (43 mg, 0.42 mmol). After the mixture was stirred at rt for 2 h, N-

methylmorpholine (43 mg, 0.42 mmol) and dimethyl ^L-glutamate

hydrochloride (126 mg, 0.53 mmol) were added all at once. The

mixture was stirred at rt for 4 h. TLC showed the formation of one

major spot at R_f= 0.38 (CHCl₃/MeOH, 5:1). The reaction mixture

was evaporated to dryness under reduced pressure. The residue was

dissolved in a minimum amount of CHCl₃/MeOH, 5:1, and

chromatographed on a silica gel column (2 cm × 15 cm) with 4%

MeOH in CHCl₃as the eluent. Fractions that showed the desired

single spot at R_f= 0.38 were pooled and evaporated to dryness to

MHz, DMSO-d₆) δ 1.15−1.20 (m, 6H, 2 CH₃), 1.27−1.34 (m, 4H, 2

CH₂), 1.53−1.62 (m, 4H, 2 CH₂), 1.97−2.03 (m, 1H, CH₂), 2.06−

2.13 (m, 1H, CH₂), 2.42−2.45 (m, 2H, CH₂), 2.51−2.53 (m, 2H,

CH₂), 2.61 (t, 2H, CH₂, J = 7.5 Hz), 4.02−4.12 (m, 4H, 2 CH₂),

4.41−4.44 (m, 1H, CH), 5.93 (s, 2H, 2-NH₂), 6.30 (d, 1H, CH, J = 1.0

Hz), 7.28 (d, 2H, C₆H₄, J = 4.0 Hz), 7.78 (d, 2H, C₆H₄, J = 4.0 Hz),

8.62 (d, 1H, CONH, J = 4.0 Hz), 10.06 (s, 1H, 3-NH), 10.57 (s, 1H,

7-NH).

(S)-2-(4-((2-Amino-4-oxo-4,7-dihydro-3H-pyrrolo[2,3-d]-

pyrimidin-5-yl)methyl)benzamido)pentanedioic Acid (5). To

17a (75 mg, 0.26 mmol) was added 1 N NaOH (3 mL). The resulting

mixture was stirred at rt for 2 h. TLC indicated the disappearance of

starting material and the formation of one major spot at the origin.

The resulting solution was passed through Celite. The combined

ﬁltrate was evaporated under reduced pressure to dryness. To this

residue was added distilled water (10 mL). The solution was cooled in

an ice bath, and the pH was adjusted 3 to 4 using 1 N HCl. The

resulting suspension was chilled in a dry ice/acetone bath and thawed

to 4 °C overnight in a refrigerator. The precipitate was ﬁltered, washed

with cold water, and dried in a desiccator under reduced pressure using

P₂O₅(yield 94%) as a yellow powder. mp > 264 °C decomposed. R_f=

1

aﬀord 17a (100 mg, yield 61%) as a yellow syrup. H NMR (400M,

DMSO-d₆) δ 1.14−1.19 (m, 6H, 2 CH₃), 1.88−2.15 (m, 2H, CH₂),

2.42−2.45 (m, 2H, CH₂), 3.98 (s, 2H, CH₂), 4.03−4.12 (m, 4H, 2

CH₂), 4.36−4.46 (m, 1H, CH), 6.00 (s, 2H, 2-NH₂), 6.32 (d, 1H, CH,

J = 1.0 Hz), 7.37 (d, 2H, C₆H₄, J = 4.0 Hz), 7.74 (d, 2H, C₆H₄, J = 4.0

Hz), 8.60 (d, 1H, CONH, J = 4.0 Hz), 10.13 (s, 1H, 3-NH), 10.74 (s,

1H, 7-NH).

1

0.05 (CHCl₃/MeOH, 5:1). H NMR (400 MHz, DMSO-d₆) δ 1.88−

(S)-Diethyl 2-(4-(2-(2-Amino-4-oxo-4,7-dihydro-3H-pyrrolo-

[2,3-d]pyrimidin-5-yl)ethyl)benzamido)pentanedioate (17b).

Compound 17b was synthesized as described for 17a (yield 51%) as

2.15 (m, 2H, CH₂), 2.42−2.45 (m, 2H, CH₂), 3.98 (s, 2H, CH₂),

4.37−4.39 (m, 1H, CH), 6.00 (s, 2H, 2-NH₂), 6.32 (s, 1H, CH), 7.37

(d, 2H, C₆H₄, J = 4.0 Hz), 7.75 (d, 2H, C₆H₄, J = 4.0 Hz), 8.47 (d, 1H,

CONH, J = 4.0 Hz), 10.13 (s, 1H, 3-NH), 10.73 (s, 1H, 7-NH). Anal.

(C₁₉H₁₉N₅O₆·1.5H₂O).

1

a colorless syrup. R_f= 0.40 (CHCl₃/MeOH, 5:1). H NMR (500

MHz, DMSO-d₆) δ 1.15−1.20 (m, 6H, 2 CH₃), 1.97−2.03 (m, 1H,

CH₂), 2.06−2.13 (m, 1H, CH₂), 2.42−2.45 (m, 2H, CH₂), 2.85 (t,

2H, CH₂, J = 7.0 Hz), 2.97 (t, 2H, CH₂, J = 7.0 Hz), 4.02−4.13 (m,

4H, 2 CH₂), 4.40−4.44 (m, 1H, CH), 5.99 (s, 2H, 2-NH₂), 6.30 (d,

1H, CH, J = 1.0 Hz), 7.29 (d, 2H, C₆H₄, J = 4.0 Hz), 7.77 (d, 2H,

C₆H₄, J = 4.0 Hz), 8.61 (d, 1H, CONH, J = 4.0 Hz), 10.13 (s, 1H, 3-

NH), 10.60 (s, 1H, 7-NH).

(S)-2-(4-(2-(2-Amino-4-oxo-4,7-dihydro-3H-pyrrolo[2,3-d]-

pyrimidin-5-yl)ethyl)benzamido)pentanedioic Acid (6). Com-

pound 6 was synthesized as described for 5 (yield 96%) as a pink

powder. mp > 250 °C decomposed. Identical in all respects (NMR,

mp) to that reported by the method of Taylor et al.⁴¹R_f= 0.07

1

(CHCl₃/MeOH, 5:1). H NMR (400 MHz , DMSO-d₆) δ 1.97−2.03

(S)-Diethyl 2-(4-(3-(2-Amino-4-oxo-4,7-dihydro-3H-pyrrolo-

[2,3-d]pyrimidin-5-yl)propyl)benzamido)pentanedioate (17c).

Compound 17c was synthesized as described for 17a (yield 65%) as

(m, 1H, CH₂), 2.06−2.13 (m, 1H, CH₂), 2.42−2.45 (m, 2H, CH₂),

2.85 (t, 2H, CH₂, J = 7.0 Hz), 2.97 (t, 2H, CH₂, J = 7.0 Hz), 4.36−4.40

(m, 1H, CH), 6.00 (s, 2H, 2-NH₂), 6.30 (s, 1H, CH), 7.28 (d, 2H,

C₆H₄, J = 4.0 Hz), 7.78 (d, 2H, C₆H₄, J = 4.0 Hz), 8.51 (d, 1H,

CONH, J = 4.0 Hz), 10.15 (s, 1H, 3-NH), 10.61 (s, 1H, 7-NH). Anal.

(C₂₀H₂₁N₅O₆·1.2H₂O).

1

a colorless syrup. R_f= 0.40 (CHCl₃/MeOH, 5:1). H NMR (DMSO-

d₆) δ 1.15−1.20 (m, 6H, 2 CH₃), 1.91−1.94 (m, 2H, CH₂), 1.97−2.03

(m, 1H, CH₂), 2.06−2.13 (m, 1H, CH₂), 2.42−2.45 (m, 2H, CH₂),

2.58 (t, 2H, CH₂, J = 7.0 Hz), 2.64 (t, 2H, CH₂, J = 7.0 Hz), 4.02−4.13

(m, 4H, 2 CH₂), 4.42−4.45 (m, 1H, CH), 5.95 (s, 2H, 2-NH₂), 6.35

(d, 1H, CH, J = 1.0 Hz), 7.29 (d, 2H, C₆H₄, J = 4.0 Hz), 7.79 (d, 2H,

C₆H₄, J = 4.0 Hz), 8.63 (d, 1H, CONH, J = 4.0 Hz), 10.08 (s, 1H, 3-

NH), 10.62 (s, 1H, 7-NH).

(S)-2-(4-(3-(2-Amino-4-oxo-4,7-dihydro-3H-pyrrolo[2,3-d]-

pyrimidin-5-yl)propyl)benzamido)pentanedioic Acid (7). Com-

pound 7 was synthesized as described for 5 (yield 97%) as a pale-blue

powder. mp > 189 °C decomposed. R_f= 0.07 (CHCl₃/MeOH, 5:1).

¹H NMR (500 MHz, DMSO-d₆) δ 1.90−1.94 (m, 2H, CH₂), 1.94−

2.00 (m, 1H, CH₂), 2.06−2.13 (m, 1H, CH₂), 2.33−2.36 (m, 2H,

CH₂), 2.58 (t, 2H, CH₂, J = 7.0 Hz), 2.64 (t, 2H, CH₂, J = 7.0 Hz),

4.36−4.40 (m, 1H, CH), 5.96 (s, 2H, 2-NH₂), 6.35 (d, 1H, CH, J = 1.0

Hz), 7.29 (d, 2H, C₆H₄, J = 4.0 Hz), 7.79 (d, 2H, C₆H₄, J = 4.0 Hz),

8.51 (d, 1H, CONH, J = 4.0 Hz), 10.08 (s, 1H, 3-NH), 10.62 (s, 1H,

7-NH). Anal. (C₂₁H₂₃N₅O₆·0.70H₂O).

(S)-2-(4-(4-(2-Amino-4-oxo-4,7-dihydro-3H-pyrrolo[2,3-d]-

pyrimidin-5-yl)butyl)benzamido)pentanedioic Acid (8). Com-

pound 8 was synthesized as described for 5 (yield 88%) as a blue

powder. mp > 153 °C decomposed. R_f= 0.07 (CHCl₃/MeOH, 5:1).

¹H NMR (500 MHz, DMSO-d₆) δ 1.55−1.63 (m, 4H, 2 CH₂), 1.92−

2.00 (m, 1H, CH₂), 2.03−2.13 (m, 1H, CH₂), 2.34 (t, 2H, CH₂, J =

7.5 Hz), 2.58 (t, 2H, CH₂, J = 7.5 Hz), 2.63 (t, 2H, CH₂, J = 7.5 Hz),

4.37−4.39 (m, 1H, CH), 5.95 (s, 2H, 2-NH₂), 6.31 (d, 1H, CH, J = 1.0

Hz), 7.28 (d, 2H, C₆H₄, J = 4.0 Hz), 7.78 (d, 2H, C₆H₄, J = 4.0 Hz),

8.49 (d, 1H, CONH, J = 4.0 Hz), 10.09 (s, 1H, 3-NH), 10.58 (s, 1H,

7-NH). Anal. (C₂₂H₂₅N₅O₆·1.2H₂O).

(S)-Diethyl 2-(4-(4-(2-Amino-4-oxo-4,7-dihydro-3H-pyrrolo-

[2,3-d]pyrimidin-5-yl)butyl)benzamido)pentanedioate (17d).

Compound 17d was synthesized as described for 17a (yield 70%) as

1

a light blue syrup. R_f= 0.40 (CHCl₃/MeOH, 5:1). H NMR (500

MHz, DMSO-d₆) δ 1.15−1.20 (m, 6H, 2 CH₃), 1.55−1.64 (m, 4H, 2

CH₂), 1.97−2.03 (m, 1H, CH₂), 2.06−2.13 (m, 1H, CH₂), 2.42−2.45

(m, 2H, CH₂), 2.58 (t, 2H, CH₂, J = 7.5 Hz), 2.65 (t, 2H, CH₂, J = 7.5

Hz), 4.02−4.13 (m, 4H, 2 CH₂), 4.42−4.45 (m, 1H, CH), 5.94 (s, 2H,

2-NH₂), 6.31 (d, 1H, CH, J = 1.0 Hz), 7.28 (d, 2H, C₆H₄, J = 4.0 Hz),

7.77 (d, 2H, C₆H₄, J = 4.0 Hz), 8.61 (d, 1H, CONH, J = 4.0 Hz),

10.08 (s, 1H, 3-NH), 10.59 (s, 1H, 7-NH).

(S)-Diethyl 2-(4-(5-(2-Amino-4-oxo-4,7-dihydro-3H-pyrrolo-

[2,3-d]pyrimidin-5-yl)pentyl)benzamido)pentanedioate (17e).

Compound 17e was synthesized as described for 17a (yield 67%) as

1

a light blue syrup. R_f= 0.40 (CHCl₃/MeOH, 5:1). H NMR (500

MHz, DMSO-d₆) δ 1.15−1.20 (m, 6H, 2 CH₃), 1.29−1.33 (m, 2H,

CH₂), 1.55−1.62 (m, 4H, 2 CH₂), 1.97−2.03 (m, 1H, CH₂), 2.06−

2.13 (m, 1H, CH₂), 2.42−2.45 (m, 2H, CH₂), 2.51−2.53 (m, 2H,

CH₂), 2.62 (t, 2H, CH₂, J = 7.5 Hz), 4.02−4.12 (m, 4H, 2 CH₂),

4.41−4.45 (m, 1H, CH), 5.94 (s, 2H, 2-NH₂), 6.31 (d, 1H, CH, J = 1.0

Hz), 7.28 (d, 2H, C₆H₄, J = 4.0 Hz), 7.78 (d, 2H, C₆H₄, J = 4.0 Hz),

8.62 (d, 1H, CONH, J = 4.0 Hz), 10.07 (s, 1H, 3-NH), 10.58 (s, 1H,

7-NH).

(S)-2-(4-(5-(2-Amino-4-oxo-4,7-dihydro-3H-pyrrolo[2,3-d]-

pyrimidin-5-yl)pentyl)benzamido)pentanedioic Acid (9). Com-

pound 9 was synthesized as described for 5 (yield 93%) as a blue

powder. mp > 167 °C decomposed. R_f= 0.08 (CHCl₃/MeOH, 5:1).

¹H NMR (500 MHz, DMSO-d₆) δ 1.29−1.32 (m, 2H, CH₂), 1.58−

1.62 (m, 4H, 2 CH₂), 1.92−2.00 (m, 1H, CH₂), 2.03−2.13 (m, 1H,

CH₂), 2.34 (t, 2H, CH₂, J = 7.5 Hz), 2.53−2.55 (m, 2H, CH₂), 2.62 (t,

2H, CH₂, J = 7.5 Hz), 4.37−4.40 (m, 1H, CH), 5.94 (s, 2H, 2-NH₂),

6.31 (d, 1H, CH, J = 1.0 Hz), 7.28 (d, 2H, C₆H₄, J = 4.0 Hz), 7.78 (d,

(S)-Diethyl 2-(4-(6-(2-Amino-4-oxo-4,7-dihydro-3H-pyrrolo-

[2,3-d]pyrimidin-5-yl)hexyl)benzamido)pentanedioate (17f).

Compound 17f was synthesized as described for 17a (yield 78%) as

1

a colorless syrup. R_f= 0.42 (CHCl₃/MeOH, 5:1). H NMR (500

10028

dx.doi.org/10.1021/jm401328u | J. Med. Chem. 2013, 56, 10016−10032

Journal of Medicinal Chemistry

Article

2H, C₆H₄, J = 4.0 Hz), 8.50 (d, 1H, CONH, J = 4.0 Hz), 10.08 (s, 1H,

3-NH), 10.58 (s, 1H, 7-NH). Anal. (C₂₃H₂₇N₅O₆·1.0H₂O).

(S)-2-(4-(6-(2-Amino-4-oxo-4,7-dihydro-3H-pyrrolo[2,3-d]-

pyrimidin-5-yl)hexyl)benzamido)pentanedioic Acid (10). Com-

pound 10 was synthesized as described for 5 (yield 93%) as a blue

powder. mp > 178 °C decomposed. R_f= 0.10 (CHCl₃/MeOH, 5:1).

¹H NMR (500 MHz, DMSO-d₆) δ 1.27−1.32 (m, 4H, 2 CH₂), 1.53−

1.62 (m, 4H, 2 CH₂), 1.92−2.00 (m, 1H, CH₂), 2.03−2.13 (m, 1H,

CH₂), 2.34 (t, 2H, CH₂, J = 7.5 Hz), 2.52−2.54 (m, 2H, CH₂), 2.60 (t,

2H, CH₂, J = 7.5 Hz), 4.36−4.41 (m, 1H, CH), 5.94 (s, 2H, 2-NH₂),

6.31 (d, 1H, CH, J = 1.0 Hz), 7.28 (d, 2H, C₆H₄, J = 4.0 Hz), 7.78 (d,

2H, C₆H₄, J = 4.0 Hz), 8.50 (d, 1H, CONH, J = 4.0 Hz), 10.07 (s, 1H,

3-NH), 10.57 (s, 1H, 7-NH). Anal. (C₂₄H₂₉N₅O₆·1.77CH₃OH·

0.1HCl).

Molecular Modeling and Computational Studies. The X-ray

crystal structures of human GARFTase at 1.98 Å resolution (PDB ID

1NJS)³⁵and human AICARFTase at 2.55 Å resolution (PDB ID

1P4R)³⁶were obtained from the Protein Data Bank. The GARFTase

crystal structure contains human GARFTase complexed with the

hydrolyzed form of 10-triﬂuoroacetyl-5,10-dideaza-acyclic-5,6,7,8-

tetrahydrofolic acid (10-CF₃CO-DDACTHF), whereas the AICARF-

Tase crystal structure contains human AICARFTase complexed with

the sulfamido-bridged 5,8-dideazafolate analogue BW1540.

and RT16 (expresses FRα).^15−20,45The CHO sublines were all

maintained in α-minimal essential medium (α-MEM) supplemented

with a 5% penicillin−streptomycin solution, 2 mM ^L-glutamate, and

10% heat-treated bovine calf serum (Invitrogen). Transfected cell lines

were cultured in the presence of 1.5 mg/mL of G418. CHO cell lines

were cultured prior to cell viability experiments in folate-free RPMI

(FF-RPMI) (Invitrogen) with dialyzed fetal bovine serum (DFBS)

(Invitrogen) supplemented with 5% penicillin-streptomycin and ^L-

glutamate. FRα-expressing human KB nasopharyngeal carcinoma cells

were continuously maintained in complete FF-RPMI with 10% fetal

bovine serum, 5% penicillin−streptomycin, and ^L-glutamate.

Cell proliferation assays were performed exactly as previously

described.¹⁵⁻¹⁹Cell lines were treated with a range of concentrations

of standard or novel antifolates (0−1 μM), metformin (0−10 mM), or

AICA (0−2 mM) in a 96-well plate in FF-RPMI complete with DFBS,

2 (RT-16, KB, IGROV1) or 25 nM (R2/PCFT4) LCV, penicillin−

streptomycin, and ^L-glutamate over a 96 h incubation period at 37 °C

with 5% CO₂. To conﬁrm FRα-mediated drug uptake and growth

inhibition, 200 nM folic acid was added to the incubations with drug.

For proliferation assays with PC43-10 cells, cells were cultured in

standard RPMI1640 with DFBS, penicillin−streptomycin, and ^L-

glutamate for 96 h. To quantitate viable cells, media was removed and

Cell Titer-blue ﬂuorescent reagent (Promega) was added. Relative cell

numbers were determined by relative ﬂuorescence units measured with

a ﬂuorescence plate reader (Molecular Devices) at 590 nm emission

and 560 nm excitation with Softmax Pro plate reader software. IC₅₀

values, corresponding to the drug concentrations that resulted in 50%

loss of cell growth, were calculated from dose−response curves using

SigmaPlot (v.12) software.

To identify the targeted pathways/folate-dependent enzymes by the

classic and novel pyrrolo[2,3-d]pyrimidine antifolates, proliferation

assays were performed in the presence of adenosine (60 μM),

thymidine (10 μM), or AICA hydrochloride (320 μM), and the results

were compared to incubations performed in parallel without

nucleoside/AICA or drug additions.

Colony-forming assays were performed with compound 8 to verify a

cytotoxic (as opposed to cytostatic) response. KB cells (∼500 cells)

were plated in 60 mm dishes in FF-RPMI/DFBS supplemented with 2

nM LCV for a 24 h incubation period at 37 °C in the presence of 5%

CO₂. Concentrations from 0 to 10 μM of compound 8 were added to

media for an additional 24 h, after which the media was aspirated and

replaced with complete FF-RPMI supplemented with 2 nM LCV for

8−10 days. To visualize colonies, the cells were washed (2×) with

room temperature Dulbecco’s phosphate-buﬀered saline (PBS)

(Sigma) followed by a wash of 5% trichloroactetic acid (TCA). The

TCA was removed, and 10 mM borate buﬀer (pH 8.8) was added

followed by a 30 min incubation with 1% methylene blue in borate

buﬀer at room temperature. The cells were washed three times with

borate buﬀer, and the stained colonies were counted.

Competitive MTX Transport Assays. To determine if novel

antifolates bound to RFC, reﬂecting membrane transport by this

mechanism, competition with [³H]MTX for cellular uptake was

measured in RFC-expressing PC43-10 CHO cells.^15,18PC43-10 cells

(∼1.5 × 10⁶cells) were seeded in 60 mm dishes. After 48 h, cells were

washed with PBS and uptake of [³H]MTX (0.5 μM) was measured in

Hank’s balanced salts solution for 2 min at 37 °C. Ice-cold PBS was

used to quench the reactions. Cells were washed with ice-cold PBS

(3×) and then solubilized with 0.5 N NaOH. The cellular

homogenates were assayed for radioactivity with a liquid scintillation

counter, and protein concentrations were measured with Folin-phenol

reagent.⁴⁶Levels of drug uptake were expressed as picomoles per

milligram of cell protein.

FRα Binding Assay. To determine the binding aﬃnity of novel

antifolates to FRα, FRα-expressing RT16 CHO cells (∼1.5 x10⁶cells)

were seeded in a monolayer with complete α-MEM media. Cells were

washed with ice-cold PBS followed by ice-cold acetate buﬀer (10 mM

sodium acetate, 150 mM NaCl, pH 3.5) to release FRα-bound folates

and then neutralized with ice-cold HEPES-buﬀered saline (HBS) (20

mM HEPES, 140 mM NaCl, 5 mM KCl, 2 mM MgCl₂, 5 mM glucose,

pH 7.4). Cells were then incubated at 0 °C for 15 min in a cocktail

Docking studies were performed using LeadIT 1.3.⁴²The

protonation state of the proteins and the ligands were calculated

using the default settings. Water molecules in the active site were

permitted to rotate freely. The active site was deﬁned by a sphere of

6.5 Å from the native ligand in the crystal structure. Ligands for

docking were prepared using MOE 2010.10⁴³and energy-minimized

using the MMF94X forceﬁeld to a constant of 0.05 kcal/mol. Triangle

matching was used as the placement method, and the docked poses

were scored using default settings. The docked poses were exported

and visualized in MOE.

Molecules used for the docking experiments were constructed in

MOE 2010.11 and minimized using the MMFF94x forceﬁeld to a

constant of 0.05 kcal/mol. To validate the docking software for

docking the proposed compounds, the native ligands, 10-CF₃CO-

DDACTHF for GARFTase and BW1540 for AICARFTase, were built

using the molecule builder function in MOE, energy minimized, and

docked, as described above. rmsd of the docked poses were calculated

using an SVL code obtained from the MOE web site (www.

chemcomp.com) and compared to the conformation of the crystal

structure ligands.

The best docked pose for 10-CF₃CO-DDACTHF in the human

GARFTase crystal structure had an rmsd of 1.0368 Å, and the best

docked pose for BW1540 had an rmsd of 1.0995 Å in the

AICARFTase crystal structure. Thus, LeadIT 1.3. was validated for

our docking purposes in GARFTase and AICARFTase.

Reagents for Biological Studies. [3′,5′,7-³H]MTX (20 Ci/

mmol), [3′,5′,7,9-³H] folic acid (25 Ci/mmol), and [¹⁴C(U)]-glycine

(87mCi/mmol) were purchased from Moravek Biochemicals (Brea,

CA). Unlabeled folic acid was purchased from the Sigma Chemical Co.

(St. Louis, MO). LCV ((6R,S)-5-formyl tetrahydrofolate) was

provided by the Drug Development Branch, National Cancer Institute,

Bethesda, MD. The sources of the classical antifolate drugs were as

follows: MTX, Drug Development Branch, National Cancer Institute

(Bethesda, MD); RTX (N-(5-(N-(3,4-dihydro-2-methyl-4-oxyquina-

zolin-6-ylmethyl)-N-methyl-amino]-2-thienoyl)-^L-glutamic acid), As-

traZeneca Pharmaceuticals (Maccesﬁeld, Cheshire, England); LMTX

(5,10-dideaza-5,6,7,8-tetrahydrofolate), Eli Lilly and Co. (Indianapolis,

IN); and PMX (N-{4-(2-(2-amino-3,4-dihydro-4-oxo-7H-pyrrolo[2,3-

d]pyrimidin-5-yl)ethyl)benzoyl}-^L-glutamic acid) (Alimta) (LC Labo-

ratories, Woburn, MA). Other chemicals were obtained from

commercial sources in the highest available purity.

Cell Culture. MTXRIIOua^R2-4 (referred to as R2) is a RFC-,

PCFT-, and FRα-null Chinese hamster ovary (CHO) cell line that was

a gift from Dr. Wayne Flintoﬀ (University of Western Ontario).⁴⁴

From this parental cell line, RFC, PCFT, and FRα were transfected to

give rise to isogenic CHO cell lines designated PC43-10 (expresses

human RFC but not PCFT or FRα), R2/PCFT4 (expresses PCFT),

10029

dx.doi.org/10.1021/jm401328u | J. Med. Chem. 2013, 56, 10016−10032

Journal of Medicinal Chemistry

Article

containing [3′,5′,7′,9-³H]folic acid in the absence or presence of

unlabled folic acid or novel antifolate compounds over a range of

concentrations. Cells were washed with ice-cold HBS and solubilized

with 0.5 N NaOH. The cell homogenates were collected and analyzed

for radioactivity using a liquid scintillation counter. Protein

concentrations were measured with Folin-phenol reagent.⁴⁶Bound

[³H]folic acid was expressed as picomoles per milligram of protein.

Using SigmaPlot (v. 12) software, dose−response curves were created,

and IC₅₀’s were calculated. Relative binding aﬃnities were calculated as

the inverse molar ratios of unlabeled ligands required to inhibit

[³H]folic acid by 50%, as determined graphically with GraphPad (v.

6.0) software. By deﬁnition, the relative aﬃnity of folic acid is 1.

HPLC Detection of ATP. Intracellular ATP concentrations were

determined using nucleotide extractions and an HPLC method exactly

as previously described.^20,33Brieﬂy, KB cells were seeded in a T75

ﬂask with FF-RPMI with 10% DFBS, 5% penicillin−streptomycin, 2

mM ^L-glutamine, and 2 nM LCV. Cells were treated with 1 μM of

PMX or compounds 2 or 8 for 24 h. After the incubations, cells were

trypsinized and washed with cold PBS. Nucleotide pools were

fractionated by reversed-phase HPLC on a C18 column for

determination of ATP pools.²⁰

Apoptosis and Cell Cycle Analyses. Apoptosis and cell cycle

analyses were performed through the Microscopy, Imaging, and

Cytometry Resources Core of the Karmanos Cancer Institute. KB cells

were treated with 1 μM of PMX or compounds 2 or 8 for 48 and 96 h

in FF-RPMI media complete with 10% DFBS, 5% penicillin−

streptomycin, and 2 nM LCV. Cells were harvested by trypsinization

and washed with cold PBS. Cells were equally divided into two pools

and either assayed for cell cycle distribution or for apoptosis by ﬂow

cytometry using the BD FACSCanto II ﬂow cytometer operated with

BD FACSDiva software (v6.0) (BD Biosciences). For cell cycle

analysis, cells were ﬁxed with ice-cold ethanol (added dropwise) and

then stained with 50 μg/mL of PI and 100 μg/mL of RNase (type I-A)

(Sigma) in PBS for at least 20 min at room temperature in the dark

prior to analysis by ﬂow cytometry, with a minimum of 10 000 events

(cells) analyzed. Apoptosis was measured using the annexin V-FITC/

7-AAD kit, as speciﬁed by the manufacturer (Beckman Coulter), with

a minimum of 20 000 events (cells) analyzed for early apoptosis

(annexin-V-FITC^high/7-AAD^low; Q3) and late apoptosis/necrosis

(annexin-V-FITC^high/7-AAD^high; Q2)]; viable cells are shown in Q4

(annexin-V-FITC^low/7-AAD^low). All data were analyzed with FlowJo

(v7.6.1) software (Tree Star, Inc.). The Watson Pragmatic model was

used to analyze cell cycle distributions of G₁/G_o-, S-, and G₂/M-phase

cells and to estimate the sub-G₁(late apoptotic cell) fraction.

nonspeciﬁc [¹⁴C]fractions. The results were calculated as pmol

[¹⁴C]formyl GAR/mg protein. Data were graphed using SigmaPlot

(v.12), and the IC₅₀’s were calculated for the antifolate-treated samples

compared to the untreated controls.

HPLC Detection of ZMP Pools. Methods for detecting ZMP in

KB cells treated with the antifolate drugs were based on those

described by Rascanelli et al.^22,23Brieﬂy, KB cells were treated with

PMX or compound 8 for 48 h in FF-RPMI with 10% DFBS, 5%

penicillin−streptomycin, 2 mM ^L-glutamine, and 2 nM LCV. Cells

were washed with PBS and treated with ice-cold 5% TCA, and the

precipitated proteins were collected by centrifugation (1600 rpm, 10

min). The soluble fractions were extracted with ether (2×) and

fractionated by ion-exchange HPLC on a SAX 250 × 2 mm column

(Phenomenex) with a linear gradient from 5 mM NH₄H₂PO₄(pH

2.8) to 750 mM NH₄H₂PO₄(pH 3.9) over 25 min at a ﬂow rate of 0.2

mL/min. Absorbance was monitored at 280 nm, and the intracellular

ZMP concentrations were established by ﬁtting the peak absorbances

to a standard curve with commercial ZMP. AraCMP was used as an

internal control.

Western Blot Analysis. KB cells were treated for 48 h with PMX

or with compound 8 in the presence of thymidine (10 μM). Controls

include vehicle control (water or DMS0 as appropriate; results were

identical) or known activators of AMPK, including metformin (10

mM), AICA (1 mM), or AICA ribonucleoside (AICAR) (1 mM).

Cells were washed, suspended in 10 mM Tris-HCl (pH 7.0) plus

proteolytic and phosphatase inhibitors (Roche Applied Science), and

disrupted by sonication. Samples were centrifuged (14 000 rpm, 10

min), and 90 μg of protein was fractionated by SDS polyacrylamide gel

electrophoresis on Laemmli gradient gels (4−20%).⁴⁷Proteins were

electrotransferred to polyvinylidene diﬂuoride membranes (Pierce).⁴⁸

Detection of immunoreactive AMPK was with rabbit total and

phospho-AMPK antibodies (Cell Signaling) and an IRDye800-

conjugated secondary anti-rabbit antibody (LI-COR, Lincoln, NE)

using an Odyssey infrared imaging system (LI-COR). β-actin was

detected with an anti-β-actin mouse antibody (Sigma) and an

IRDye800-conjugated secondary anti-mouse antibody (LI-COR).

ASSOCIATED CONTENT

* Supporting Information

■

S

Cell cycle distributions and apoptosis in KB cells treated with

pemetrexed (PMX) or compounds 2 or 8 for 48 h, inhibition of

KB cell proliferation with the AMPK activator metformin, and

elemental analyses. This material is available free of charge via

the Internet at http://pubs.acs.org.

In Situ GARFTase Inhibition Assays. To measure the extent of

intracellular GARFTase inhibition in the presence of antifolate

inhibitors, we examined the metabolic incorporation of [¹⁴C(U)]-

glycine into [¹⁴C]formyl GAR in the presence of azaserine.^15−20,32KB

cells were seeded in complete media (above) and allowed to adhere to

the substratum for 24 h. Cells were then washed with FF-RPMI, ^L-

glutamine-free, with 10% DFBS, 5% penicillin−streptomycin, and 2

nM LCV. Cells were incubated at 37 °C with 5% CO₂for 1 h with

folate- and ^L-glutamine-depleted media containing 2 nM LCV with

PMX or with compounds 2 or 8 over a range of concentrations.

Control cells were treated with DMSO of equal volume. Azaserine (4

μM ﬁnal concentration) was added to the cells and allowed to incubate

for 30 min. This was followed by the addition of ^L-glutamine (2 mM)

and [¹⁴C(U)]glycine (ﬁnal speciﬁc activity 0.1 mCi/L). Cells were

incubated 16 h. Cells were washed with ice-cold PBS, trypsinized, and

collected. Cells were treated with 5% TCA at 0 °C. Samples were

centrifuged (4 °C, 14 000 rpm), and the protein precipitants were

solubilized with 0.5 N NaOH for quantitation of protein

concentrations with the Folin-phenol protein method.⁴⁶The TCA

supernatants were extracted with cold ether. After evaporation of

ether, the remaining aqueous layer was gravity ﬁltered through a 1 cm

chromatography column (Biorad) of AG1 × 8 (chloride form). The

columns were washed with 10 mL of 0.5 N formic acid followed by 4

N formic acid (10 mL) and ﬁnally eluted with 8 mL of 1 N HCL

collected in eight fractions. The eluate was measured for radioactivity,

and the percentages of radioactivity in the [¹⁴C]formyl GAR and

AUTHOR INFORMATION

Corresponding Authors

*Phone: 313-578-4280; Fax: 313-578-4287; E-mail: matherly@

karmanos.org (L.H.M.).

*Phone: 412-396-6070; Fax: 412-396-5593; E-mail: gangjee@

duq.edu (A.G.).

■

Present Address

^#Center for Drug Discovery and Translational Research, Beth

Israel Deaconess Medical Center, Harvard Medical School,

Boston, Massachusetts 02215, United States.

Author Contributions

^⊥These authors contributed equally to this work

Notes

The authors declare no competing ﬁnancial interest.

ACKNOWLEDGMENTS

■

This work was supported in part by grants from the National

Institutes of Health, National Cancer Institute R01 CA152316

(L.H.M. and A.G.), R01 CA166711 (A.G. and L.H.M.), and

R01 CA53535 (L.H.M.), and the Duquesne University Adrian

10030

dx.doi.org/10.1021/jm401328u | J. Med. Chem. 2013, 56, 10016−10032

Journal of Medicinal Chemistry

Article

Biological Ramifications and Impact on the Activity of Pemetrexed.

Cancer Metastasis Rev. 2007, 26, 129−139.

Van Kaam Chair in Scholarly Excellence (A.G.). Ms. Mitchell-

Ryan was supported by T32 CA009531 (L.H.M.) and F31

CA165853 (S.M.R.). We thank Dr. Mark Stout for his

assistance with the HPLC analyses of cellular ZMP. The

Microscopy, Imaging and Cytometry Resources Core is

supported, in part, by NIH Center Grant P30 CA22453 (to

the Barbara Ann Karmanos Cancer Institute) and the

Perinatology Research Branch of the NICHD, NIH, Wayne

State University.

(11) Kugel Desmoulin, S.; Hou, Z.; Gangjee, A.; Matherly, L. H. The

Human Proton-Coupled Folate Transporter: Biology and Therapeutic

Applications to Cancer. Cancer Biol. Ther. 2012, 13, 1−19.

(12) Kamen, B. A.; Smith, A. K. Farletuzumab, an Anti-Folate

Receptor α Antibody, Does Not Block Binding of Folate or Anti-

Folates to Receptor nor Does It Alter the Potency of Anti-Folates In

Vitro. Cancer Chemother. Pharmacol. 2012, 70, 113−120.

(13) Xia, W.; Low, P. S. Folate-Targeted Therapies for Cancer. J.

Med. Chem. 2010, 53, 6811−6824.

(14) Gibbs, D. D.; Theti, D. S.; Wood, N.; Green, M.; Raynaud, F.;

Valenti, M.; Forster, M. D.; Mitchell, F.; Bavetsias, V.; Henderson, E.;

Jackman, A. L. BGC 945, a Novel Tumor-selective Thymidylate

Synthase Inhibitor Targeted to Alpha-Folate Receptor-Overexpressing

Tumors. Cancer Res. 2005, 65, 11721−11728.

(15) Deng, Y.; Wang, Y.; Cherian, C.; Hou, Z.; Buck, S. A.; Matherly,

L. H.; Gangjee, A. Synthesis and Discovery of High Affinity Folate

Receptor-Specific Glycinamide Ribonucleotide Formyltransferase

Inhibitors with Antitumor Activity. J. Med. Chem. 2008, 51, 5052−

5063.

(16) Deng, Y.; Zhou, X.; Desmoulin, S. K.; Matherly, L. H.; Gangjee,

A. Synthesis and Biological Activity of a Novel Series of 6-Substituted

Thieno[2,3-d]pyrimidine Antifolate Inhibitors of Purine Biosynthesis

with Selectivity for High Affinity Folate Receptors over the Reduced

Folate Carrier and Proton-Coupled Folate Transporter for Cellular

Entry. J. Med. Chem. 2009, 52, 2940−2951.

(17) Wang, L.; Cherian, C.; Desmoulin, S. K.; Polin, L.; Deng, Y.;

Wu, J.; Hou, Z.; White, K.; Kushner, J.; Matherly, L. H.; Gangjee, A.

Synthesis and Antitumor Activity of a Novel Series of 6-Substituted

Pyrrolo[2,3-d]pyrimidine Thienoyl Antifolate Inhibitors of Purine

Biosynthesis with Selectivity for High Affinity Folate Receptors and

the Proton-Coupled Folate Transporter over the Reduced Folate

Carrier for Cellular Entry. J. Med. Chem. 2010, 53, 1301−1318.

(18) Wang, L.; Desmoulin, S. K.; Matherly, L. H.; Gangjee, A.

Synthesis, Biological, and Antitumor Activity of a Highly Potent 6-

Substituted Pyrrolo[2,3-d] pyrimidine Thienoyl Antifolate Inhibitor

with Proton-Coupled Folate Transporter and Folate Receptor

Selectivity over the Reduced Folate Carrier that Inhibits β-

Glycinamide Ribonucleotide Formyltransferase. J. Med. Chem. 2011,

54, 7150−7164.

ABBREVIATIONS USED

■

IC₅₀, 50% inhibitory concentration; AAD, amino actinomycin

D; α-MEM, α-minimal essential medium; AICA, 5-amino-

imidazole-4-carboxamide; AICAR, 5-aminoimidazole-4-carbox-

amide ribonucleoside; AICARFTase, 5-aminoimidazole-4-car-

boxamide ribonucleotide formyltransferase; AMPK, AMP-

activated protein kinase; CHO, Chinese hamster ovary;

DFBS, dialyzed fetal bovine serum; DHFR, dihydrofolate

reductase; FITC, ﬂuorescein isothiocyanate; FF-RPMI, folate-

free RPMI1640; FR, folate receptor; GAR, glycinamide

ribonucleotide; GARFTase, glycinamide ribonucleotide formyl-

transferase; HBS, HEPES-buﬀered saline; LCV, leucovorin;

LMTX, lometrexol; MTX, methotrexate; PMX, pemetrexed;

PBS, phosphate-buﬀered saline; PCFT, proton-coupled folate

transporter; PI, propidium iodine; RTX, raltiterxed; RFC,

reduced folate carrier; SAR, structure−activity relationship;

TCA, trichloroacetic acid; TS, thymidylate synthase; ZMP, 5-

aminoimidazole-4-carboxamide ribonucleotide

REFERENCES

■

(1) Visentin, M.; Zhao, R.; Goldman, I. D. The Antifolates. Hematol.

Oncol. Clin. North Am. 2012, 26, 629−648.

(2) Gonen, N.; Assaraf, Y. G. Antifolates in Cancer Therapy:

Structure, Activity and Mechanisms of Drug Resistance. Drug Resist.

Updates 2012, 15, 183−210.

(3) Mendelsohn, L. G.; Worzalla, J. F.; Walling, J. M. Preclinical and

Clinical Evaluation of the Glycinamide Ribonucleotide Formyltransfer-

ase Inhibitors Lometrexol and LY309887. In Anticancer Drug

Development Guide: Antifolate Drugs in Cancer Therapy; Jackman, A.

L., Ed.; Humana Press, Inc.: Totowa, NJ, 1999; pp 261−280.

(4) Budman, D. R.; Johnson, R.; Barile, B.; Bowsher, R. R.;

Vinciguerra, V.; Allen, S. L.; Kolitz, J.; Ernest, C. S., II; Kreis, W.;

Zervos, P.; Walling, J. Phase I and Pharmacokinetic Study of

LY309887: A Specific Inhibitor of Purine Biosynthesis. Cancer

Chemother. Pharmacol. 2001, 47, 525−553.

(19) Wang, L.; Cherian, C.; Desmoulin, S. K.; Matherly, L. H.;

Gangjee, A. Synthesis and Biological Activity of 6-Substituted

Pyrrolo[2,3-d]pyrimidine Thienoyl Regioisomers as Inhibitors of De

Novo Purine Biosynthesis with Selectivity for Cellular Uptake by High

Affinity Folate Receptors and the Proton-Coupled Folate Transporter

over the Reduced Folate Carrier. J. Med. Chem. 2012, 55, 1758−1770.

(20) Desmoulin, S. K.; Wang, Y.; Gangjee, A.; Matherly, L. H.

Targeting the Proton-Coupled Folate Transporter for Selective

Delivery of 6-Substituted Pyrrolo[2,3-d]pyrimidine Antifolate Inhib-

itors of De Novo Purine Biosynthesis in the Chemotherapy of Solid

Tumors. Mol. Pharmacol. 2010, 78, 577−587.

(5) Boritzki, T. J.; Zhang, C.; Bartlett, C. A.; Jackson, R. C. AG2034 a

GARFTASE Inhibitor with Selective Cytotoxicity to Cells That Lack a

G1 Checkpoint. In Anticancer Drug Development Guide: Antifolate

Drugs in Cancer Therapy; Jackman, A. L., Ed.; Humana Press Inc.:

Totowa, NJ, 1999; pp 281− 292.

(21) Chattopadhyay, S.; Moran, R. G.; Goldman, I. D. Pemetrexed:

Biochemical and Cellular Pharmacology, Mechanisms, and Clinical

Applications. Mol. Cancer Ther. 2007, 6, 404−417.

(6) Ray, M. S.; Muggia, F. M.; Leichman, C. G.; Grunberg, S. M.;

Nelson, R. L.; Dyke, R. W.; Moran, R. G. Phase I Study of (6R)-5,10-

Dideazatetrahydrofolate: A Folate Antimetabolite Inhibitory to De

Novo Purine Synthesis. J. Natl. Cancer Inst. 1993, 85, 1154−1159.

(7) Bissett, D.; McLeod, H. L.; Sheedy, B.; Collier, M.; Pithavala, Y.;

Paradiso, L.; Pitsiladis, M.; Cassidy, J. Phase I Dose-Escalation and

Pharmacokinetic Study of a Novel Folate Analogue AG2034. Br. J.

Cancer 2001, 84, 308−312.

(22) Racanelli, A. C.; Rothbart, S. B.; Heyer, C. L.; Moran, R. G.

Therapeutics by Cytotoxic Metabolite Accumulation: Pemetrexed

Causes ZMP Accumulation, AMPK Activation, and Mammalian

Target of Rapamycin Inhibition. Cancer Res. 2009, 69, 5467−5474.

(23) Rothbart, S. B.; Racanelli, A. C.; Moran, R. G. Pemetrexed

Indirectly Activates the Metabolic Kinase AMPK in Human

Carcinomas. Cancer Res. 2010, 70, 10299−10309.

(24) Mihaylova, M. M.; Shaw, R. J. The AMPK Signalling Pathway

Coordinates Cell Growth, Autophagy and Metabolism. Nat. Cell Biol.

2011, 13, 1016−1023.

(25) Gwinn, D. M.; Shackelford, D. B.; Egan, D. F.; Mihaylova, M.

M.; Mery, A.; Vasquez, D. S.; Turk, B. E.; Shaw, R. J. AMPK

Phosphorylation of Raptor Mediates a Metabolic Checkpoint. Mol.

Cell 2008, 30, 214−226.

(8) Matherly, L. H.; Hou, Z.; Deng, Y. Human Reduced Folate

Carrier: Translation of Basic Biology to Cancer Etiology and Therapy.

Cancer Metastasis Rev. 2007, 26, 111−128.

(9) Elnakat, H.; Ratnam, M. Distribution, Functionality and Gene

Regulation of Folate Receptor Isoforms: Implications in Targeted

Therapy. Adv. Drug Delivery Rev. 2004, 56, 1067−1084.

(10) Zhao, R.; Goldman, I. D. The Molecular Identity and

Characterization of a Proton-Coupled Folate Transporter−PCFT;

10031

dx.doi.org/10.1021/jm401328u | J. Med. Chem. 2013, 56, 10016−10032

Journal of Medicinal Chemistry

Article

(26) Inoki, K.; Zhu, T.; Guan, K. L. TSC2 Mediates Cellular Energy

Response to Control Cell Growth and Survival. Cell 2003, 115, 577−

590.

(27) Ma, X. M.; Blenis, J. Molecular Mechanisms of mTOR-

Mediated Translational Control. Nat. Rev. Mol. Cell Biol. 2009, 10,

307−318.

(28) Barnett, C. J.; Wilson, T. M.; Kobierski, M. E. A Practical

Synthesis of Multitargeted Antifolate LY231514. Org. Process Res. Dev.

1999, 3, 184−188.

(29) Aso, K.; Imai, Y.; Yukishige, K.; Ootsu, K.; Akimoto, H.

Pyrrolo[2,3-d]pyrimidine Thymidylate Synthase Inhibitors: Design

and Synthesis of One-Carbon Bridge Derivatives. Chem. Pharm. Bull.

2001, 49, 1280−1287.

(30) Larock, R. C.; Leung, W. Y.; Stolzdunn, S. Synthesis of Aryl-

Substituted Aldehydes and Ketones via Palladium-Catalyzed Coupling

of Aryl Halides and Non-Allylic Unsaturated Alcohols. Tetrahedron

Lett. 1989, 30, 6629−6632.

(31) Bloch, R. 5,5-Dibromo-2,2-dimethyl-4,6-dioxo-1,3-dioxane −

New Brominating Agent for Saturated and Alpha,Beta-Unsaturated

Carbonyl-Compounds. Synthesis 1978, 140−142.

from Chinese Hamster Ovary Cells. Somatic Cell Genet. 1976, 2, 245−

261.

(45) Wong, S. C.; Proefke, S. A.; Bhushan, A.; Matherly, L. H.

Isolation of Human cDNAs That Restore Methotrexate Sensitivity and

Reduced Folate Carrier Activity in Methotrexate Transport-Defective

Chinese Hamster Ovary Cells. J. Biol. Chem. 1995, 270, 17468−17475.

(46) Lowry, O. H.; Rosebrough, N. J.; Farr, A. L.; Randall, R. J.

Protein Measurement with the Folin Phenol Reagent. J. Biol. Chem.

1951, 193, 265−275.

(47) Laemmli, U. K. Cleavage of Structural Proteins During the

Assembly of the Head of Bacteriophage T4. Nature 1970, 227, 680−

685.

(48) Matsudaira, P. Sequence from Picomole Quantities of Proteins

Electroblotted onto Polyvinylidene Difluoride Membranes. J. Biol.

Chem. 1989, 262, 10035−10038.

(32) Beardsley, G. P.; Moroson, B. A.; Taylor, E. C.; Moran, R. G. A

New Antimetabolite, 5,10-Dideaza-5,6,7,8-tetrahydrofolate is a Potent

Inhibitor of De Novo Purine Synthesis. J. Biol. Chem. 1989, 264, 328−

333.

(33) Huang, D.; Zhang, Y.; Chen, X. Analysis of Intracellular

Nucleoside Triphosphate Levels in Normal and Tumor Cell Lines by

High-Performance Liquid Chromatography. J. Chromatogr., B: Anal.

Technol. Biomed. Life Sci. 2003, 784, 101−109.

(34) Brown, J. M.; Attardi, L. D. The Role of Apoptosis in Cancer

Development and Treatment Response. Nat. Rev. Cancer 2005, 5, 31−

37.

(35) Cheong, C.-G.; Wolan, D. W.; Greasley, S. E.; Horton, P. A.;

Beardsley, G. P.; Wilson, I. A. Crystal Structures of Human

Bifunctional Enzyme Aminoimidazole-4-carboxamide Ribonucleotide

Transformylase/IMP Cyclohydrolase in Complex with Potent

Sulfonyl-Containing Antifolates. J. Biol. Chem. 2004, 279, 18034−

18045.

(36) Zhang, Y.; Desharnais, J.; Marsilje, T. H.; Li, C.; Hedrick, M. P.;

Gooljarsingh, L. T.; Tavassoli, A.; Benkovic, S. J.; Olson, A. J.; Boger,

D. L.; Wilson, I. A. Rational Design, Synthesis, Evaluation, and Crystal

Structure of a Potent Inhibitor of Human GAR Tfase: 10-

(Trifluoroacetyl)-5,10-dideazaacyclic-5,6,7,8-tetrahydrofolic Acid. Bio-

chemistry. 2003, 42, 6043−6056.

(37) Hazarika, M.; White, R. M.; Johnson, J. R.; Pazdur, R. FDA

Drug Approval Summaries: Pemetrexed (Alimta). Oncologist 2004, 9,

482−488.

(38) Thompson, C. A. FDA Approves Pralatrexate for Treatment of

Rare Lymphoma. Am. J. Health-Syst. Pharm. 2009, 66, 1890.

(39) Chen, C.; Ke, J.; Zhou, X. E.; Yi, W.; Brunzelle, J. S.; Li, J.; Yong,

E. L.; Xu, H. E.; Melcher, K. Structural Basis for Molecular

Recognition of Folic Acid by Folate Receptors. Nature 2013, 500,

486−489.

(40) Wibowo, A. S.; Singh, M.; Reeder, K. M.; Carter, J. J.; Kovach,

A. R.; Meng, W.; Ratnam, M.; Zhang, F.; Dann, C. E., 3rd. Structures

of Human Folate Receptors Reveal Biological Trafficking States and

Diversity in Folate and Antifolate Recognition. Proc. Natl. Acad. Sci.

U.S.A. 2013, 110, 15180−15188.

(41) Taylor, E. C.; Kuhnt, D.; Shih, C.; Rinzel, S. M.; Grindey, G. B.;

Barredo, J.; Jannatipour, M.; Moran, R. G. A Dideazatetrahydrofolate

Analog Lacking a Chiral Center at C-6: N-[4-[2-(2-Amino-3,4-

dihydro-4-oxo-7H-pyrrolo[2,3-d]pyrimidin-5yl)ethyl[benzoyl]-^L-gluta-

mic Acid Is an Inhibitor of Thymidylate Synthase. J. Med. Chem. 1992,

35, 4450−4454.

(42) LeadIT, version 1.3; BioSolveIT GmbH: Sankt Augustin,

Germany, 2011.

(43) MOE, 2010.10; Chemical Computing Group: Montreal,

Quebec, Canada.

(44) Flintoff, W. F.; Davidson, S. V.; Siminovitch, L. Isolation and

Partial Characterization of Three Methotrexate-Resistant Phenotypes

10032

dx.doi.org/10.1021/jm401328u | J. Med. Chem. 2013, 56, 10016−10032

Article Doi

DOI: 10.1021/jm401328u

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jm401328u

Full text of DOI:10.1021/jm401328u

Products guided by the article

R&D Labs maybe for 1150670-22-6

Relevant to this article

Hot Product