Discovery and characterization of nonpeptidyl agonists of the tissue-protective erythropoietin receptors

Article abstract of DOI:10.1124/mol.115.098400

Erythropoietin (EPO) and its receptor are expressed in a wide variety of tissues, including the central nervous system. Local expression of both EPO and its receptor is upregulated upon injury or stress and plays a role in tissue homeostasis and cytoprotection. High-dose systemic administration or local injection of recombinant human EPO has demonstrated encouraging results in several models of tissue protection and organ injury, while poor tissue availability of the protein limits its efficacy. Here, we describe the discovery and characterization of the nonpeptidyl compound STS-E412 (2-[2-(4-chlorophenoxy) ethoxy]-5,7-dimethyl-[1,2,4]triazolo[1,5-a]pyrimidine), which selectively activates the tissue-protective EPO receptor, comprising an EPO receptor subunit (EPOR) and the common β-chain (CD131). STS-E412 triggered EPO receptor phosphorylation in human neuronal cells. STS-E412 also increased phosphorylation of EPOR, CD131, and the EPO-associated signaling molecules JAK2 and AKT in HEK293 transfectants expressing EPOR and CD131. At low nanomolar concentrations, STS-E412 provided EPO-like cytoprotective effects in primary neuronal cells and renal proximal tubular epithelial cells. The receptor selectivity of STS-E412 was confirmed by a lack of phosphorylation of the EPOR/EPOR homodimer, lack of activity in off-target selectivity screening, and lack of functional effects in erythroleukemia cell line TF-1 and CD34⁺ progenitor cells. Permeability through artificial membranes and Caco-2 cell monolayers in vitro and penetrance across the blood-brain barrier in vivo suggest potential for central nervous system availability of the compound. To our knowledge, STS-E412 is the first nonpeptidyl, selective activator of the tissue-protective EPOR/CD131 receptor. Further evaluation of the potential of STS-E412 in central nervous system diseases and organ protection is warranted.

Full text of DOI:10.1124/mol.115.098400

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
Discovery and characterization of nonpeptidyl agonists of the tissue-protective
erythropoietin receptor
James L. Miller, Timothy J. Church, Dmitri Leonoudakis, Karen Lariosa-Willingham, Normand
L. Frigon, Connie S. Tettenborn, Jeffrey R. Spencer, and Juha Punnonen
STATegics, Inc., 1455 Adams Dr., Menlo Park, CA 94025.

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
Running title:
Nonpeptidyl agonists of the tissue-protective EPO receptor
Corresponding author:
James L. Miller, PhD
STATegics, Inc.
1455 Adams Dr.
Menlo Park, CA 94025
Telephone: (650) 248-7944
Fax:
E-mail: jim.miller@stategics.com
Number of text pages: 39
Number of tables: 0
Number of figures: 5
Number of references: 72
Number of words in Abstract: 247
Number of words in Introduction: 746
Number of words in Discussion: 1488
List of nonstandard abbreviations:
AKT, v-akt murine thymoma viral oncogene homolog 1; BBB, blood-brain-barrier; BCA, bicinchoninic
acid; BSA, bovine serum albumin; CEPO, carbamoylated erythropoietin; CD131, colony stimulating
factor 2 receptor, beta, low-affinity (CSF2RB); CNS, central nervous system; DMEM, Dulbecco’s
modified Eagles medium; DMSO, dimethyl sulfoxide; DNase I, deoxyribonuclease I; DPBS, Dulbecco’s
phosphate-buffered saline; ELISA, enzyme-linked immunosorbent assay; EPO, erythropoietin; EPOR,
erythropoietin receptor; EPOR/EPOR, homodimeric erythropoietin receptor; EPOR/CD131, tissue-

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
protective heteromeric receptor; FBS, fetal bovine serum; G-CSF, granulocyte colony-stimulating factor;
HBSS, Hank’s balanced salt solution; HPLC, high performance liquid chromatography; JAK2, Janus
kinase 2; KIU, Kallikrein inhibitory units; LC-MS/MS, Liquid chromatography-tandem mass
spectrometry; LDH, lactate dehydrogenase; MEM Minimum essential medium; PAMPA, Parallel
artificial membrane permeability assay; rhEPO, recombinant human erythropoietin; RLU, relative
luminescence units; RPMI, Roswell Park Memorial Institute medium; RPTEC, human renal proximal
tubule epithelial cells; SAR, structure-activity relationship; TEER, trans-epithelial electrical resistance;
TPO, thrombopoietin; TPOR, thrombopoietin receptor.

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
ABSTRACT
Erythropoietin (EPO) and its receptor are expressed in a wide variety of tissues, including the central
nervous system. Local expression of both EPO and its receptor is upregulated upon injury or stress and
plays a role in tissue homeostasis and cytoprotection. High-dose systemic administration or local
injection of recombinant human EPO has demonstrated encouraging results in several models of tissue
protection and organ injury, while poor tissue availability of the protein limits its efficacy. Here, we
describe the discovery and characterization of the nonpeptidyl compound STS-E412, which selectively
activates the tissue-protective EPO receptor, comprising an EPO receptor subunit (EPOR) and the
common beta-chain (CD131). STS-E412 triggered EPO receptor phosphorylation in human neuronal
cells. STS-E412 also increased phosphorylation of EPOR, CD131 and the EPO-associated signaling
molecules JAK2 and AKT in HEK-293 transfectants expressing EPOR and CD131. At low nanomolar
concentrations, STS-E412 provided EPO-like cytoprotective effects in primary neuronal cells and renal
proximal tubular epithelial cells. The receptor selectivity of STS-E412 was confirmed by a lack of
phosphorylation of the EPOR/EPOR homodimer, lack of activity in off-target selectivity screening, and
lack of functional effects in erythroleukemia cell line TF-1 and CD34+ progenitor cells. Permeability
through artificial membranes and Caco-2 cell monolayers in vitro and penetrance across the blood-brain
barrier in vivo suggest potential for CNS availability of the compound. To our knowledge, STS-E412 is
the first nonpeptidyl, selective activator of the tissue-protective EPOR/CD131 receptor. Further
evaluation of the potential of STS-E412 in CNS diseases and organ protection is warranted.

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
Introduction
Erythropoietin (EPO) is a pleiotropic cytokine primarily known as the key regulator of erythroid
progenitor cell differentiation (Jacobson et al., 1957). EPO is synthesized in the renal cortex in response
to erythropoietic signaling, and the distribution of EPO thus produced is generally restricted to the
vasculature (Eckardt and Kurtz, 2005). EPO also exhibits broad cytoprotective and neuroprotective
effects in vitro and in vivo. Biosynthesis of EPO mediating cytoprotection occurs locally in tissues, and
EPO and EPO receptor (EPOR) expression increase following injury (Juul et al., 1998, Grasso et al.,
2005). EPOR can be detected in neurons, astrocytes, and microglia in the central nervous system (CNS)
and in multiple peripheral organs, including the heart, kidney, liver, lung, pancreas, and skeletal muscle
(Liu et al., 1997, Juul et al., 1998, David et al., 2002, Calvillo et al., 2003, Assaraf et al., 2007, Medana et
al., 2009, Joshi et al., 2014).
The receptors mediating erythropoiesis and neuroprotection by EPO are distinct. A homodimeric
receptor comprising two EPOR mediates erythropoiesis, while the cytoprotective effects of EPO are
mediated by a heteromeric receptor composed of EPOR and the GM-CSF/IL-3/IL-5 receptor common
beta-chain (CD131, CSF2RB) (Jubinsky et al., 1997, Brines et al., 2004, Leist et al., 2004, Chateauvieux
et al., 2011). CD131 physically associates with the EPOR (Jubinsky et al., 1997, Brines et al., 2004,
Bennis et al., 2012). Recombinant human EPO (rhEPO) induces phosphorylation of CD131 in UT-7 cells
(Hanazono et al., 1995) and transfection of CD131 into EPOR-expressing Ba/F3 cells enhances the
proliferative effects induced by rhEPO (Jubinsky et al., 1997). No protective effects of EPOR agonists
are observed in CD131-deficient mice (Brines et al., 2004), and CD131 is crucial for rhEPO’s anti-
apoptotic effects in endothelial cells (Su et al., 2011, Bennis et al., 2012). Like EPOR, the expression of
CD131 is upregulated following tissue injury; demonstrated by co-localization and upregulation of human
EPOR and CD131 in muscle fiber sarcolemma during critical limb ischemia (Joshi et al., 2014).
RhEPO demonstrates beneficial effects in preclinical and early clinical studies in several CNS
diseases, such as Friedreich’s ataxia, ischemic stroke, multiple sclerosis, Parkinson’s disease, spinal cord
injury, and traumatic brain injury (Konishi et al., 1993, Sakanaka et al., 1998, Brines et al., 2000,

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
Erbayraktar et al., 2003, Puskovic et al., 2006, Savino et al., 2006, Mammis et al., 2009, Ehrenreich et al.,
2011, Nachbauer et al., 2011, Jang et al., 2014). Achieving therapeutically meaningful levels of rhEPO in
the CNS with an intact blood-brain-barrier (BBB) requires direct intracranial injection or high-dose
systemic administration. High-dose rhEPO, however, is associated with a risk of side effects, including
thrombotic events (Drüeke et al., 2006, Singh et al., 2006); therefore, a selective EPOR/CD131 agonist
may have a therapeutic advantage over rhEPO for CNS indications.
Selective activation of the EPOR/CD131 receptor was reported using recombinant EPO variants
and EPO mimetic peptides. For example, carbamoylated EPO (CEPO), desialylated derivatives of EPO,
and several EPO-derived peptides signal via the tissue-protective EPOR/CD131 without any measurable
erythropoietic activity (Erbayraktar et al., 2003, Brines et al., 2004, Leist et al., 2004, Coleman et al.,
2006, Brines and Cerami, 2008, Brines et al., 2008, Ahmet et al., 2011, Chen et al., 2012). CEPO
provided potent neuroprotection with efficacy comparable to high-dose rhEPO in vivo (Leist et al., 2004).
However, CEPO exhibits pharmacokinetic properties similar to rhEPO (Leist et al., 2004), likely limiting
the therapeutic applicability of this variant for indications where tissue-availability is critical.
Small molecules have potential advantages over proteins, particularly when tissue permeability or
CNS availability is desired. Agonists of several cytokine receptors, including the receptors for
granulocyte colony-stimulating factor (G-CSF) (Tian et al., 1998) and thrombopoietin (TPO) (Erickson-
Miller et al., 2005, Desjardins et al., 2006, Dziewanowska et al., 2007, Kim et al., 2007, Kalota and
Gewirtz, 2010, Abe et al., 2011) have been described, and the TPO receptor (TPOR) agonist eltrombopag
is FDA-approved for the treatment of thrombocytopenia (Kuter, 2013). The feasibility of discovering
small molecule agonists of receptors for EPO was also supported by the small allosteric peptide agonists
that activated the EPOR/EPOR homodimer and triggered EPO-like biological activity (Naranda et al.,
1999, Naranda et al., 2002).
We identified a series of nonpeptidyl compounds that selectively activate the tissue-protective
EPOR/CD131 heteromeric receptor. The compound designated STS-E412 activates the EPOR/CD131
receptor at low nanomolar concentrations but does not activate the EPOR/EPOR homodimer. STS-E412

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
exhibits potent, EPO-like cytoprotective effects in primary human neuronal cells, rat hippocampal
neurons, and human renal proximal tubule epithelial cells (RPTEC) following a range of cytotoxic
challenges.

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
Materials and Methods
Materials. Unless otherwise indicated, the reagents for these studies were obtained from the
following sources: Roche Pharma AG (rhEPO, NeoRecormon®); Peprotech, [recombinant human IL-3
(rhIL-3)]; HyClone/Thermo Fisher, [Dulbecco's modified Eagles medium (DMEM) and minimal essential
medium (MEM) cell culture media, fetal bovine serum (FBS), sodium pyruvate, L-glutamine,
hygromycin B, Dulbecco’s phosphate-buffered saline (DPBS), Gentamycin, trypsin/EDTA, Hanks’
balanced salt solution (HBSS)]; Sigma, [DNase1, leupeptin, puromycin, L-glutamate, polyethyleneimine
(PEI), dextrose]; Invitrogen/Life Technologies, [amphotericin B (Fungizone®), B27 supplement,
Neurobasal medium, Glutamax®]; Fisher, [dimethyl sulfoxide (DMSO)].
STS-E412 (2-[2-(4-
chlorophenoxy)ethoxy]-5,7-dimethyl-[1,2,4]triazolo[1,5-a]pyrimidine) was synthesized in our laboratory
as described in the Supplemental Methods.
Preparation of human fetal brain tissue. Primary human brain tissue was obtained from
Advanced Bioscience Resources, Inc., Alameda, CA, as dissected 13- to 16-week-old human fetal brain
tissue. The gender of the fetuses was unknown. Processing and maintenance of human cortical tissue
was performed essentially according to Wogulis et al. (Wogulis et al., 2005) with modifications as
described below. The initial tissue processing was performed in HBSS supplemented with 250 ng/ml
amphotericin B. Brain tissue was cut into ~2-mm3 pieces with a sterile razor blade. The tissue fragments
were washed twice in amphotericin B-supplemented HBSS followed each time by centrifugation. The
tissue pieces were then dispersed by trituration in amphotericin B-supplemented HBSS containing DNase
I (1 mg/ml). Individual cells were separated from the tissue debris by passing the suspension through a
nylon mesh cell strainer (100 µm), collected by centrifugation, and digested in 0.05% trypsin/EDTA at
37°C for 20 minutes. The digestion was terminated by adding MEM supplemented with 10% FBS and
DNase I (1 mg/ml), and the cells were again dispersed by gentle trituration. The isolated cells were
collected by centrifugation and washed twice in culture medium containing MEM supplemented with L-
glutamine (1 mM), dextrose (1% w/v), sodium pyruvate (1 mM), B27 and Gentamycin (10 µg/ml).
Finally, the cells were plated in PEI-coated 24-well plates (750,000 cells per well). The medium was

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
refreshed twice weekly, and assays were performed between 7 and 21 days in culture. The cortical cell
cultures were > 90% neurons, based on positive staining for neurofilament and microtubule-associated
protein (Wogulis et al., 2005), and are referred to herein as human neuronal cells.
Generation of EPOR- and EPOR/CD131-overexpressing cell lines. The human embryonic
kidney cell line HEK-293 (ATCC-1573) was propagated in DMEM supplemented with FBS (10%).
HEK-293 cells over-expressing the EPOR alone were produced by transfection with a commercial
mammalian expression vector containing the human EPOR gene and a puromycin selection marker
(Gencopoeia, Rockville MD; EX-A0440-M68) using the FuGENE HD transfection reagent (Promega,
Madison WI) followed by selection in 5 µg/ml puromycin. HEK-293 cells over-expressing CD131 alone
were produced by transfection with a commercial mammalian expression vector containing the human
CSF2RB gene and a hygromycin B selection marker (Gencopoeia EX-Z5874-M67) followed by selection
in 150 µg/ml hygromycin B. HEK-293 cells over-expressing both the EPOR and CD131 were generated
by transfecting the EPOR expression vector (EX-A0440-M68) into the HEK293/CD131 cell line
followed by selection in puromycin and hygromycin B (5 µg/ml and 150 µg/ml, respectively).
Preparation of hippocampal neuronal cultures. Embryonic day 18 rat hippocampi (BrainBits,
Springfield, IL) were dissociated with papain (Worthington Biochemical Corporation) and plated in 96-
well plates at a density of 50,000 cells per well in Neurobasal/B27 medium as follows. Brain tissue was
received packaged in a proprietary preservation “Hibernate”. The Hibernate medium overlaying the
brain tissue was removed and replaced with100 ug/ml papain in DPBS (1 ml/brain); the digestion mixture
was then transferred to a 30°C water bath for 30 minutes. After incubation, the digestion solution was
removed and replaced with 3 ml of DMEM supplemented with 10% FBS and 2 U/ml DNase I. Using a
fire-polished 9-inch Pasteur pipette, the digested tissue was transferred to a 15-ml tube and triturated (5-
10 cycles) to disperse the individual cells. The dispersed cells were separated from larger tissue pieces by
allowing the latter to settle to the bottom of the centrifuge tube and transferring the supernatant to a new
15-ml centrifuge tube. The cells were collected by centrifugation (200g, 5 min). The cell pellet was
dispersed with a fire-polished 9-inch Pasteur pipette into 2 ml of growth medium containing Neurobasal

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
medium, B27 supplement, and Glutamax® (5 mM). The cells were plated at a density of 50,000 cells per
well in 96-well multiwell plates and maintained in a humidified incubator at 37°C with 5% CO2.
Primary human renal epithelial cells. Normal human RPTEC (Lonza/Clonetics) were received
at passage 2 and were seeded and grown according to the manufacturer’s protocol. The RPTEC were
maintained in basal medium (REBM; Lonza/Clonetics) supplemented with recombinant human epidermal
growth factor, recombinant human insulin, hydrocortisone, epinephrine, transferrin, triiodothyronine,
gentamycin, amphotericin B, and 0.5% FBS (Lonza/Clonetics).
Signaling assays. For assays employing HEK293 and derived cell lines, the cells were plated in
6- or 24-well plates 48 hour prior to the assay. For assays employing human neuronal cells, assays were
performed 7 to 21 days after plating. One hour prior to the assay, the media were removed and replaced
with fresh media. For the signaling assays measuring phosphorylation of AKT (HEK293 and derived
cells lines only), cells were detached using DPBS without Ca2+ or Mg2+, collected by centrifugation,
suspended in DMEM supplemented with 0.1% FBS, plated in 6-well tissue-culture plates, and assayed
between 30-45 minutes after plating. Compound stock solutions were prepared in DMSO at 1,000x final
concentration, and final dilutions were prepared immediately before the assay in cell culture medium.
Compound, rhEPO, and control cultures contained identical final concentrations of DMSO (0.1% by
volume). Cell plates and assay solutions were equilibrated to room temperature prior to the assay. The
reaction was initiated by the addition of test articles (5x final concentration) and mixing by gentle
trituration. Following 11 minutes at room temperature, the media overlaying the cells were removed, and
the cells were washed by the addition of 1 ml of DPBS. The DPBS overlaying the cells was then
removed and replaced with ice-cold lysis buffer (see Supplemental Data for composition of ELISA
reagents). . The lysate was frozen in place, thawed and clarified by centrifugation prior to analysis.
Determination of phospho-EPOR content in cell lysates. A standard sandwich enzyme-linked
immunosorbent assay (ELISA) was performed to assess the level of phospho-EPOR (pEPOR) according
to the manufacturer’s protocol (R&D Systems).
The results were normalized to protein contents

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
determined using the bicinchoninic acid (BCA) method (Pierce). See Supplemental Data, Supplemental
Figures 3-4 and Supplemental Table 1 for assay validation.
Determination of phospho-CD131 content in cell lysates. A standard sandwich ELISA was
performed to assess the level of phospho-CD131 (pCD131). Briefly, total CD131 was captured with an
immobilized anti-CD-131 antibody (BD Biosciences, # 554534) and detected with a horseradish
peroxidase-conjugated secondary antibody directed against phosphotyrosine (R&D Systems). The
pCD131 content was normalized to the protein content determined using the BCA method (Pierce).
Positive control membrane lysates were generated using TF-1 cells that were starved of IL-3 overnight
and treated with IL-3 (10 ng/ml) for 12 minutes. See Supplemental Data and Supplemental Figures 5-7
for assay validation.
Determination of phospho-JAK2 and phospho-AKT content in cell lysates. Commercial
ELISA kits for the detection of pY1007pY1008-JAK2 (pJAK2) and pS473-AKT (pAKT) were obtained from
Life Technologies and Cell Signaling Technology, respectively. Both assays were performed according
to the manufacturers’ instructions. The pJAK2 content was quantified according to a standard provided
with the kit in which 1 Unit of pYpY-JAK2 is equivalent to 3.3 µg of HEL cell lysate treated with 100
µM sodium vanadate for 30 minutes. The results were normalized to protein contents determined using
the BCA method (Pierce) and are expressed as units per 100 µg protein.
Primary rat cortical neuronal cultures. Cells were isolated from micro-surgically dissected
embryonic day 18 rat cortices that were obtained from Neuromics (Edina, MN), and cultures were
prepared according to the supplier’s protocol.
TF-1 cell proliferation assay. TF-1 cells (ATCC CRL-2003) were cultured in DMEM
supplemented with FBS (10%) and rhIL-3 (1 ng/ml). The cells were starved of IL-3 for 24 hours prior to
the assay and then plated at 5,000 cells per well in 96-well multiwell plates in the presence of test articles.
Cell proliferation was measured 72 hours later using the CellTiter-Glo® reagent (Promega) according to
the manufacturer’s instructions.

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
CD34+ cell proliferation assay. Frozen primary CD34+ cells from human bone marrow (Stem
Cell Technologies) were seeded into 96-well multiwell plates at 3,000 cells/well in StemSpan® SFEM
medium (Stem Cell Technologies) containing bovine serum albumin (1%), recombinant human insulin
(10 µg/ml), human transferrin (iron saturated, 200 µg/ml), 2-mercaptoethanol (20 µM) and L-glutamine
(2 mM) in Iscove’s MDM (Stem Cell Technologies). Cell proliferation was measured after 10 days in
culture using the CellTiter-Glo® reagent (Promega) according to the manufacturer’s instructions.
Protection of primary human neuronal cells from staurosporine challenge. Primary human
neuronal cells were treated with STS-E412 or rhEPO for 24 hours (37°C, 5% CO2) prior to initiating a
staurosporine insult (500 nM, final) and incubating for an additional 24 hours. Test articles were present
at the indicated concentrations for the duration of the experiment. Cytotoxicity was measured using a
commercially available lactate dehydrogenase (LDH) assay kit (Clontech) according to the
manufacturer’s instructions.
Excitotoxicity assay of rat hippocampal cultures. After 12 days in culture, the culture medium
was changed to Neurobasal medium supplemented with 0.1% BSA but without B27 supplement, and the
cultures were returned to the incubator (37°C, 5% CO2). Test articles were added in the same medium
yielding a final concentration of 0.2% DMSO. The following day, glutamate was added to a final
concentration of 100 µM, and the cell-culture plate was returned to the incubator for 24 hours. Test
articles were present at the indicated concentrations for the duration of the experiment. Cytotoxicity was
measured using a colorimetric LDH assay kit (Clontech) according to the manufacturer’s instructions.
The total LDH activity was determined by measuring the LDH activity released by medium supplemented
with 1% Triton X-100.
Hydrogen peroxide challenge. Twenty-four hours prior to the assay, RPTEC were harvested,
seeded into 96-well multiwell plates and returned to the incubator (37°C, 5% CO2). Components
contained in the RPTEC culture medium reduce hydrogen peroxide (H2O2); therefore, the peroxide
challenge was performed in RPMI-1640 medium supplemented with 0.5% FBS. The RPTEC medium
was replaced with RPMI/0.5% FBS by removing the media overlaying the cells and rinsing the tightly

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
adhering cells with RPMI/0.5% FBS. Fresh working stocks of H2O2 were prepared in HBSS. Fresh
working stocks of control solutions and compounds were prepared in RPMI/0.5% FBS. The challenge
was initiated by sequential additions of 10x control or compound working stocks followed by 10x H2O2
working stock (150 µM final) and mixing via gentle trituration. The culture plates were returned to the
incubator for 24 hours. Cytotoxicity was quantified using CellTiter-Glo® and LDH release assays
according to manufacturer’s instructions.
Parallel artificial membrane permeability assays. Parallel artificial membrane permeability
assays (PAMPAs) were performed using the pre-coated PAMPA Plate System (BD Biosciences,
GenTestTM, Part # 353015) according to the manufacturer’s protocol. DMSO stock solutions of test
articles and reference compounds (dopamine, amitriptyline, propranolol, carbamazepine, sulfasalazine)
were diluted in DPBS to a final concentration of 50 µM and 0.1% DMSO (v/v) and added to the donor
well in a volume of 300 µl. The acceptor wells contained 50 µM Lucifer yellow in DPBS in a volume of
200 µl. Leakage of Lucifer yellow into the donor well was interpreted as a failure in the PAMPA
membrane integrity. The incubation times were 2, 4, and 6 hours at room temperature. At the conclusion
of the incubation period, aliquots were removed and analyzed by reversed-phase high-performance liquid
chromatography (HPLC) [Agilent 1100, detection at 215/254 nm, linear gradient from 0.1%
trifluoroacetic acid in water (80%)/acetonitrile (20%) to 0.1% trifluoroacetic acid in water
(20%)/acetonitrile (80%)]. Permeability (Pe), efflux ratio (ER) and mass balance were calculated
according to the manufacturer’s methodology (SPC-353015-G rev 1.0). Permeability (cm/s): Pe = {-ln[1-
CA(t)/Ceq]}/[A*(1/VD+1/VA)*t], where A = filter area (0.3 cm2), VD = donor well volume (0.3 ml), VA =
acceptor well volume (0.2 ml), t = incubation time (s), CA(t) = compound concentration in acceptor well
at time t, CD(t)
=
compound concentration in donor well at time t, and Ceq
=
[CD(t)*VD+CA(t)*VA]/(VD+VA).
Caco-2 permeability assay. A bidirectional permeability assessment in Caco-2 cell monolayers
was performed at BD Biosciences. Caco-2 cells (human adenocarcinoma colonic cell line Caco-2, ATCC
Cat. No. HTB-37) were used at passage 44 and cultured for 24 days in DMEM supplemented with 10%

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
FBS, 5 mM HEPES, 0.4% glucose, non-essential amino acids, and 50 g/ml gentamycin. The monolayer
integrity was evaluated via pre-experimental trans-epithelial electrical resistance (TEER) measurements
and post-experimental Lucifer yellow A-to-B flux determinations for each cell monolayer. Bidirectional
transport (A to B and B to A) of compounds at 10 µM (with final 0.1% DMSO v/v) was determined in
duplicates at 0 and 90 minutes via liquid chromatography-mass spectrometry (LC-MS/MS) detection in
donor and acceptor wells. Apparent Permeability (Papp) was calculated according to the equation Papp
(cm/s) = (Flux × Vd)/(t × A) = dQ/dt × 1/(A × Cd), where Flux is the fraction of the donated amount
recovered in the receiver chamber, Vd is the volume in the donor chamber (cm3), Cd is the initial
concentration in the donor solution (mM), A is the surface area of the insert filter membrane (cm2), t is
the incubation time (s), and dQ/dt is the amount of drug transported within a given time period (pmol/s).
The ER was calculated as ER = Papp B to A / Papp A to B. The Mass balance was calculated as follows: Mass
balance = 100 × ((Vr × Crfinal) + (Vd × Cdfinal))/(Vd × C0), where C0 is the initial concentration in the donor
compartment at t = 0 min (nmol/cm3), Vr is the volume of the receiver compartment (cm3), Vd is the
volume of the donor compartment (cm3), Crfinal is the receiver concentration at the end of the incubation
(nmol/cm3), and Cdfinal is the donor concentration at the end of the incubation (nmol/cm3).
Off-target profiling assays. Enzyme activity and radioligand binding assays against 68
pharmacologically relevant molecular targets (enzymes, receptors, ion channels) were performed at
Ricerca, Concord OH (Part # 1160393). Test articles were evaluated in duplicate at a concentration of 10
µM. All assays included a reference positive control. Significant responses were defined as 50%
inhibition or stimulation for the biochemical assays or
radioligand binding assays. The tested molecular targets included Protein Serine/Threonine Kinase,
GSK3B, Adenosine A1, Adenosine A2A, Adenosine A3, Adrenergic 1A, Adrenergic 1B, Adrenergic
50% inhibition of ligand binding in the
,
1D
Adrenergic 2A, Adrenergic 1, Adrenergic 2, Androgen (Testosterone) AR, Bradykinin B1, Bradykinin
B2, Calcium Channel L-Type (Benzothiazepine), Calcium Channel L-Type (Dihydropyridine), Calcium
Channel N-Type, Cannabinoid CB1, Dopamine D1, Dopamine D2S, Dopamine D3, Dopamine D4.2,
Endothelin ETA, Endothelin ETB, Epidermal Growth Factor (EGF), Estrogen ER , GABAA

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
(Flunitrazepam, Central), GABAA (Muscimol, Central), GABAB1A, Glucocorticoid, Glutamate (Kainate),
Glutamate (NMDA, Agonism), Glutamate (NMDA, Glycine), Glutamate (NMDA, Phencyclidine),
Histamine H1, Histamine H2, Histamine H3, Imidazoline I2, IL-1, Leukotriene (Cysteinyl CysLT1),
Melatonin MT1, Muscarinic M1, Muscarinic M2, Muscarinic M3, Neuropeptide Y Y1, Neuropeptide Y Y2,
Nicotinic Acetylcholine, Nicotinic Acetylcholine 1 (Bungarotoxin), Opiate 1 (OP1, DOP), Opiate
(OP2, KOP), Opiate (OP3, MOP), Phorbol Ester, Platelet Activating Factor, Potassium Channel [KATP],
Potassium Channel hERG, Prostanoid EP4, Purinergic P2X, Purinergic P2Y, Rolipram, Serotonin (5-
Hydroxytryptamine)
5-HT1A,
Serotonin
(5-Hydroxytryptamine)
5-HT2B,
Serotonin
(5-
Hydroxytryptamine) 5-HT3, Sigma 1, Tachykinin NK1, Thyroid Hormone, Transporter, Dopamine
(DAT), Transporter (GABA), Transporter (Norepinephrine NET), Transporter (Serotonin, 5-
Hydroxytryptamine).

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
Results
Cytoprotective EPO receptor agonist compounds were identified from a small-molecule library of
putative EPO receptor agonist tool compounds (Punnonen et al., Abstract #335AAN11D1; 63rd Annual
Meeting of the American Academy of Neurology, 2011). The design of the initial library was facilitated
by the structure-activity relationship (SAR) information available from small molecule (Qureshi et al.,
1999, Goldberg et al., 2002) and allosteric peptide (Naranda et al., 1999, Naranda et al., 2002) agonists of
the EPO receptor, and allosteric small molecule agonists of the structurally related TPOR (Erickson-
Miller et al., 2005, Desjardins et al., 2006, Dziewanowska et al., 2007, Kim et al., 2007, Kalota and
Gewirtz, 2010, Abe et al., 2011), as well as structural data on the receptors (Livnah et al., 1996, Livnah et
al., 1999, Carr et al., 2001). An initial hit was identified based on receptor interaction and EPO-like
functional activity, and subsequent compound designs for developing an SAR for in vitro neuroprotection
focused on 1,2,4-triazolo[1,5-a]pyrimidines. The neuroprotection SAR optimization included calculated
BBB permeability [logP – (N+O) values and multi-parameter optimization] (Hitchcock, 2008, Wager et
al., 2010), cytoprotective activity utilizing serum-deprived PC12 cells (Kunioku et al., 2001), aqueous
solubility, in vitro metabolic stability, permeability in PAMPAs and Caco-2 assays, cytoprotection assays
in primary neuronal and hematopoietic cells, and receptor activation in transfected cell lines. Compounds
with EPO-like protective effects were identified, and STS-E412 (2-[2-(4-chlorophenoxy)ethoxy]-5,7-
dimethyl-[1,2,4]triazolo[1,5-a]pyrimidine) was selected for more detailed characterization based on
significant receptor activation, neuroprotection, and a lack of erythropoietic activity as shown below. The
structure of STS-E412 is shown as an insert in Figure 1B, and its synthesis is described under
Supplemental Data. STS-E412 has drug-like physical properties (MW = 318.8, cLogP = 3.3, topological
polar surface area = 62 Å2) and a water solubility of 0.0032 mg/ml (pH 7, PBS), which was determined by
HPLC-UV analysis of the dissolved compound in a saturated solution after 90 minutes of shaking.
To evaluate EPOR activation in primary non-erythroid cells, we established cultures of primary
human neuronal cells derived from human fetal cortical brain tissue. rhEPO produced a dose-dependent
increase in the amount of phosphorylated EPOR in these primary human neuronal cell cultures (Figure

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
1A). Basal levels of the pEPOR averaged 0.33 ± 0.1 pg/µg protein (mean ± S.D., range: 0.23-0.91
pg/µg). The apparent EC50 for rhEPO regarding EPOR phosphorylation was near 1 IU/ml (Figure 1A),
and increasing the rhEPO concentration to greater than ~10 IU/ml did not provide an additional increase
in EPOR phosphorylation. In the same primary cultures, STS-E412 also increased phosphorylation of the
EPOR (Figure 1B). The maximal effect of STS-E412 was at least comparable to or higher than that
induced by rhEPO (Figure 1B). The STS-E412-induced increase in EPOR phosphorylation was dose-
dependent and significant at concentrations as low as 60 nM. STS-E412 increased pEPOR levels
approximately 4-fold over background in three independent cultures of primary human neuronal cells.
The increase in pEPOR in HCC produced by 10 IU/ml rhEPO (n=6) was not statistically different from
that produced by 5 μM STS-E412 (n=8) based on student t-test (p=0.479).
To further examine signaling events induced by STS-E412, we generated HEK293 cell lines
over-expressing the EPOR and CD131 (Figure 2, A-D) or the EPOR alone (Figure 2E). In HEK293 cells
over-expressing the EPOR and CD131, both rhEPO and STS-E412 produced increases in EPOR and
CD131 phosphorylation (Figure 2A and 2B, respectively). The basal level of pEPOR in untransfected
HEK-293 cells was 0.33 ± 0.3 pg/µg protein (mean ± S.D., range: 0.01-0.86, n = 5), while cells over-
expressing the EPOR and CD131 had a basal pEPOR level of 0.66 ± 0.25 pg/µg protein (mean ± S.D.,
range: 0.25-0.86 pg/µg protein, n = 4). In these transfected cells, STS-E412 produced a dose-dependent
increase in pEPOR that reached a maximum near 100 nM with a bell-shaped dose-response curve (Figure
2A). The magnitude of the maximal increase in pEPOR induced by STS-E412 was approximately 50%
of that induced by rhEPO (Figure 2A). Increases in pEPOR produced by 10 IU/ml rhEPO were greater
than those produced by STS-E412 (tested at concentrations from 1-30 nM; p<0.014) while increases in
pEPOR produced by 10 IU/ml rhEPO vs. 100 nM STS-E412 were not significantly different (p=0.083).
Both STS-E412 and rhEPO also increased the phosphorylation of CD131 in these cells (Figure 2B). In
this case, no significant difference in the maximal response was observed between STS-E412 and rhEPO
(Figure 2B). Increases in pCD131 produced by 10 IU/ml rhEPO vs. those produced by 50 nM STS-E412
were not significantly different (p=0.21).

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
Both rhEPO and STS-E412 triggered phosphorylation of JAK2 (Figure 2C) and AKT (Figure 2D)
in HEK293 cells over-expressing the EPOR and CD131. In four independent experiments, STS-E412
increased pJAK2 by 1.52 ± 0.23 fold over basal levels with apparent EC50 values of approximately 1-10
nM (Figure 2C and data not shown). By comparison, 10 IU/ml rhEPO increased pJAK2 by 2.2 ± 0.2 fold
over basal levels (Figure 2C). Increases in pJAK2 produced by 10 IU/ml rhEPO vs. that produced by 1-
3.7 nM STS-E412 were statistically different by student t-test (p<0.035) while increases in pJAK2
produced by 10 IU/ml rhEPO vs. 11-100 nM STS-E412 were not significantly different (rhEPO vs. 11
nM STS-E412, p=0.072; rhEPO vs. 33 nM STS-E412, p=0.151; rhEPO vs. 100 nM STS-E412, p=0.109).
HEK293 cells and HEK293-derived cell lines contained relatively high basal levels of pAKT, and
increases in pAKT above basal levels, with either rhEPO or STS-E412 treatment, were very small (data
not shown). To circumvent this limitation, we found that transferring cells to media containing 0.1% FBS
produced a transient decrease in the level of pAKT (to approximately 30-50% of basal levels) that was
persistent for approximately 1 hour (data not shown). Under these conditions, STS-E412 also
significantly increased the level of pAKT in HEK293 cells over-expressing the EPOR and CD131 (Figure
2D). In four independent experiments, STS-E412 increased pAKT levels by 1.82 ± 0.4 fold over basal
levels with apparent EC50 values of 1-10 nM (Figure 2D and data not shown). In six independent
experiments, 10 IU/ml rhEPO increased pAKT by 2.45 ± 0.91 fold over basal levels (Figure 2D and data
not shown). Increases in pAKT produced by 10 IU/ml rhEPO vs. those produced by STS-E412 were
significant at all tested concentrations (p<5.6x10-8).
Using HEK293 cells over-expressing the EPOR alone, rhEPO strongly increased the
phosphorylation of the EPOR (Figure 2E), JAK2 and AKT (data not shown) as expected. The average
fold-increases in the levels of pEPOR and pJAK2 produced by rhEPO were 2.7 ± 0.7 fold (mean ± S.D.,
range: 1.9-3.6, n = 5) and 7.9 ± 4.3 (mean ± S.E.M., range: 2.7-20.8, n = 4), respectively. The fold-
increases in pAKT levels (1.45-fold over the basal level), in the presence of 0.1% FBS, in response to
rhEPO were smaller than the increases in pEPOR and pJAK2. In untransfected HEK293 cells, rhEPO
produced a small, statistically insignificant increase in pEPOR (p = 0.062, triplicate measurements;

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
Figure 2E), while pEPOR levels in the presence of STS-E412 were indistinguishable from those in the
controls. In EPOR-over-expressing HEK293 cells, rhEPO produced an increase in pEPOR in a dose-
dependent manner. In contrast, STS-E412 at concentrations ranging from 1 nM to 60 µM did not increase
pEPOR levels in the HEK293 cells over-expressing the EPOR alone (Figure 2E and data not shown).
Increases in pEPOR produced by 10 IU/ml rhEPO vs. those produced by STS-E412 (tested at
concentrations from 2 μM – 60 μM) were significant at all tested concentrations (p<0.00011).Consistent
with our finding in HEK293 cells, STS-E412 did not produce increases in pEPOR in the erythroid cell
line TF-1 over a similar concentration range, whereas 10 IU/ml rhEPO resulted in increases in pEPOR of
approximately 2 fold (1.94 ± 0.16 fold, mean ± S.D., range: 1.83-2.12, n = 3; data not shown).
Proliferation assays with the erythroleukemia cell line TF-1 (Figure 3A, 3B) and primary human
bone marrow-derived CD34+ cells (Figure 3C, 3D) support the view that STS-E412 does not stimulate the
erythropoietic EPOR/EPOR homodimer. RhEPO induced dose-dependent increases in both TF-1 and
CD34+ cell numbers (Figure 3A and Figure 3C, respectively). In contrast, STS-E412 at concentrations up
to 1,000 nM had no effect on TF-1 or CD34+ cell proliferation (Figure 3B and Figure 3D, respectively).
At concentrations above 1,000 nM, STS-E412 decreased the proliferation of TF-1 cells (Figure 3B), while
the proliferation of CD34+ cells was increased (Figure 3D). Student t-tests comparing of TF-1
proliferation in the presence of 10 IU/ml rhEPO vs. STS-E412 (Figure 3-B) indicated these differences
were significant (p<8.03x10-5). In contrast, student t-tests comparing CD34+ cell proliferation in the
presence of 10 IU/ml rhEPO vs. 30 μM STS-E412 were not significant (p=0.11) while differences
between 10 IU/ml rhEPO and STS-E412 at 3 μM and lower concentrations were significant (p<0.024).
STS-E412 protected primary human neuronal cells from a staurosporine challenge (Figure 4A),
protected primary rat hippocampal neurons from an excitotoxic glutamate challenge (Figure 4B) and
protected primary human RPTEC from a H2O2 challenge (Figure 5A, 5B). All the protection assays
followed a paradigm consisting of a 24-hour treatment with the test articles followed by a 24-hour
challenge, and rhEPO was included in the experiments for comparison.

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
In human neuronal cells (Figure 4A), the basal extracellular LDH was 19.3 ± 1.7% of the total
LDH (mean ± S.D.). Staurosporine treatment alone (500 nM) increased the extracellular LDH to 36.8 ±
7.2% of the total LDH (mean ± S.D., 1.9-fold increase). Pre-treatment with rhEPO attenuated the
staurosporine-dependent increase in the extracellular LDH by approximately 40% (to 29.9 ± 2.9% of the
total LDH). STS-E412 also protected the human neuronal cells from a staurosporine insult in a dose-
dependent manner with an apparent EC50 value in the 1-10 nM range (Figure 4A). STS-E412 attenuated
the increase in extracellular LDH by approximately 50% at concentrations above 10 nM (e.g., to 27.3 ±
1.6% of the total LDH in the presence of 1,000 nM STS-E412). Based on student t-tests, difference in
protection from staurosporine insult of human neuronal cells produced by 10 IU/ml rhEPO vs. those
produced by STS-E412 at concentrations > 1 nM were not significant (rhEPO vs. STS-E412 at 1 nM,
p=0.097; rhEPO vs. STS-E412 10 nM, p=0.697; rhEPO vs. STS-E412 at 100 nM, p=0.361; rhEPO vs.
STS-E412 at 1000 nM, p=0.547).
In primary rat hippocampal neuronal cultures, pre-treatment with rhEPO attenuated the
glutamate-dependent increase in extracellular LDH by approximately 70%. Similarly, STS-E412
protected primary rat hippocampal neurons from glutamate challenge with a maximal protection at
concentrations above 10 nM (Figure 4B). Concentrations of STS-E412 above 10 nM attenuated the
increase in extracellular LDH by up to 46 ± 16%. Based on student t-tests, 10 IU/ml rhEPO produced
significantly greater protection from glutamate insult in rat hippocampal neurons than STS-E412 (rhEPO
vs. STS-E412 at 1 nM, p=0.0071; rhEPO vs. STS-E412 10 nM, p=0.0128; rhEPO vs. STS-E412 at 100
nM, p=0.0133).
In primary cultures of human RPTEC, pre-treatment with rhEPO attenuated the H2O2-dependent
decrease in intracellular ATP in a dose-dependent manner with complete or near-complete protection in
the 1-10 IU/ml concentration range (102 ± 0.75% of control at 10 IU/ml rhEPO; Figure 5A). This
protection was also reflected in a decrease in the level of extracellular LDH at similar rhEPO
concentrations; rhEPO attenuated the increase in extracellular LDH by greater than 79% (97.5 ± 25% for
2 IU/ml rhEPO and 79.4 ± 0.1% for 10 IU/ml rhEPO). Pre-treatment with STS-E412 also attenuated the

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
H2O2-dependent decrease in intracellular ATP in a dose-dependent manner with nearly complete
protection at low nanomolar concentrations and complete protection at concentrations near 1 nM (103.3 ±
3.8% at 1 nM STS-E412; Figure 5A). This protection was also reflected in a decrease in extracellular
LDH at similar STS-E412 concentrations, and complete protection was observed at 0.1 nM STS-E412
(102 ± 13% compared with the unchallenged control; Figure 5B). Based on student t-test, 10 IU/ml
rhEPO did not produce significantly more protection form H2O2 challenge than the optimal concentration
of STS-E412 (LDH read-out: rhEPO vs. 0.1 nM STS-E412, p=0.303; Cell Titer Glo read-out: rhEPO vs.
1 nM STS-E412, p=0.933). Protection from H2O2 challenge was reduced at the highest concentrations of
rhEPO and STS-E412 (Figure 5A, 5B).
A preliminary assessment of the suitability of STS-E412 for in vivo studies was performed using
permeability and off-target activity assays. STS-E412 demonstrated high in vitro permeability in a
PAMPA with a Pe value of 15 × 10-6 cm/s. The bidirectional transport of STS-E412 at 10 µM across a
Caco-2 cell monolayer was determined via LC-MS/MS detection. STS-E412 demonstrated high
transcellular permeability in both directions, with Papp values > 17 × 10-6 cm/s. The calculated ER was <
1.0, suggesting that the compound is not a substrate for P-glycoprotein, a BBB efflux transporter.
Importantly, permeability of STS-E412 across the BBB was observed in vivo. Thirty minutes after 10
mg/kg IP dose of STS-E412 in Sprague-Dawley rats (n=3), plasma and brain concentrations of the
compound were 4073 121 ng/ml and 1589 25 ng/ml, respectively (compound calculations for brain
homogenate are based on a density of 1 ml per gram tissue; see supplemental data for experimental
details). The calculated brain/plasma concentration ratio was 0.39. Next, potential off-target effects were
evaluated in an enzyme- and radio-ligand-binding panel of 68 molecular targets. The assay package was
designed for prioritizing drug candidates utilizing a broad selection of targets to detect the drug’s
potential for adverse events and included several CNS targets to evaluate the potential for drug
dependence (the complete list of targets is presented in the Materials and Methods). Significant responses
were defined as > 50% inhibition or stimulation at 10 µM. All the assay values for STS-E412 were <

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
50%, suggesting no significant off-target effects among those evaluated in the panel (Supplemental Data,
Table 1).

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
Discussion
EPO has emerged as a pleiotropic cytokine that is important not only in the regulation of
erythroid cell differentiation and survival but also as a potent cytoprotective agent in multiple organs.
The characterization of the tissue-protective effects of EPO has led to the identification of two distinct
receptors for EPO: an EPOR/EPOR homodimer that mediates erythropoiesis and an EPOR/CD131
heteromeric receptor that mediates tissue-protective effects. Here we describe the discovery and
characterization of the nonpeptidyl compound STS-E412, which selectively activates the tissue-protective
EPOR/CD131 receptor.
The first reports of nonpeptidyl small molecule EPO receptor agonists described compounds
activating the EPOR/EPOR homodimer. These compounds demonstrated EPO-like activities on
hematopoietic precursor cells in vitro, while the potencies were relatively low (EC50 > 1 µM), and the
compounds had molecular weights > 1,000 Da (Qureshi et al., 1999, Goldberg et al., 2002). A dimeric
20-amino-acid peptide lacking sequence homology with EPO also activated the EPOR/EPOR homodimer
in preclinical and clinical studies (Wrighton et al., 1996, Johnson et al., 1998, Besarab et al., 2012). The
crystal structure of the dimeric EPO mimetic peptide bound to the EPOR confirmed that the binding is
orthosteric (Livnah et al., 1996). While none of these mimetics are likely to be orally bioavailable, these
studies demonstrated the feasibility of developing small molecules that compete with EPO to bind the
EPOR/EPOR homodimer.
Given the similarities in the mechanisms of activation and structures of the receptors for TPO and
EPO (Vigon et al., 1992, Boulay et al., 2003), the success in discovering and developing small molecule
TPOR agonists further suggested the feasibility of discovering small molecule mimetics of EPO. All the
small molecule TPOR agonists reported to date activate the receptor by binding to allosteric sites outside
of the native ligand-binding site and proximal to residues near the plasma membrane (Desjardins et al.,
2006, Dziewanowska et al., 2007, Kim et al., 2007, Erickson-Miller et al., 2009, Kalota and Gewirtz,
2010, Abe et al., 2011), suggesting that allosteric compounds may have more promise as therapeutic
agents targeting the EPO receptor when compared to compounds binding at the orthosteric sites. The

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
allosteric TPOR agonists described to date are also smaller than the previously described orthosteric
EPOR agonists, supporting the potential to design compounds that cross the BBB. Allosteric small
peptide agonists of the EPO receptor have been described, while no tissue-protective effects were
reported (Naranda et al., 1999, Naranda et al., 2002). Small molecule compounds selective for the tissue-
protective EPOR/CD131 heterodimer have the potential to overcome the limitations of EPO regarding
tissue availability and may improve safety by avoiding the hematopoietic side effects.
A fundamental mechanism of the EPO-induced protective effect is the inhibition of apoptosis via
the JAK2/AKT pathway and subsequent transcriptional regulation of anti-apoptotic proteins, including
Bcl-xL and Bax (Digicaylioglu and Lipton, 2001, Wen et al., 2002, Um and Lodish, 2006, Wu et al.,
2007a). EPO also protects cells against oxidative damage in vitro and in vivo (Solaroglu et al., 2003, Wu
et al., 2007b). STS-E412 activates the tissue protective EPOR/CD131 receptor at nanomolar
concentrations in primary human neuronal cells and cell lines stably expressing the EPOR and CD131.
The magnitude of CD131 phosphorylation in response to STS-E412 was comparable to that induced by
rhEPO in EPOR/CD131 transfectants. STS-E412 also increased phosphorylation of EPOR in these
transfectants, but the magnitude of the effect was substantially lower than that induced by rhEPO. STS-
E412 triggered JAK2 and AKT phosphorylation in the EPOR/CD131 transfectants with a maximal
increase approximately 50% of that induced by rhEPO. No significant activation of the EPOR/EPOR
homodimer by STS-E412 was observed, illustrating selective activation of the EPOR/CD131 receptor by
STS-E412. In contrast to transfected cell lines, the maximal effects of STS-E412 in primary neuronal
cells were comparable to rhEPO regarding both EPOR phosphorylation and cytoprotection. The lack of
EPOR phosphorylation in transfectants expressing the EPOR alone and the absence of EPO-like
proliferative effects in TF-1 and primary human CD34+ hematopoietic progenitor cells further support the
view that STS-E412 does not effectively activate EPOR/EPOR homodimeric receptors. The difference in
the relative effects of rhEPO and STS-E412 on the EPOR/CD131 transfectants and primary cells is likely
due to the presence of both EPOR/EPOR and EPOR/CD131 receptors in the heterologous expression
system, while primary human neuronal cells express primarily the EPOR/CD131 receptor (Brines et al.,

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
2004). Previous studies demonstrating comparable cytoprotective effects of CEPO and EPO in non-
hematopoietic tissues, including the CNS and kidneys (Leist et al., 2004, Imamura et al., 2007, Wang et
al., 2007), also support the conclusion that the primary receptor for EPO on neurons and RPTEC is
EPOR/CD131.
Proliferation assays in TF-1 and CD34+ progenitor cells provide a more complex view of
interactions between STS-E412 and cell surface receptors. TF-1 cells express IL-3, IL-5 and GM-CSF
receptors (heteromeric receptors composed of distinct alpha subunits and CD131), and these cytokines
stimulate TF-1 cell proliferation. TF-1 cells also express thrombopoietin receptor (TPOR) and
thrombopoietin stimulates TF-1 cell proliferation (Drexler and Quentmeier, 1996). That STS-E412 does
not stimulate proliferation in TF-1 cells suggests that STS-E412 does not effectively interact with these
receptors at concentrations below 3-10 µM. The observed decrease in TF-1 proliferation at higher
concentrations may suggest STS-E412-mediated sequestration of the CD131 subunit, which is required
for growth of the IL-3- or GM-CSF-dependent TF-1 cells. In contrast to TF-1 cells, the increase in CD34+
progenitor cell proliferation at STS-E412 concentrations above 3-10 µM may suggest a low affinity
interaction with the EPOR/EPOR homodimer or TPOR, although this was not detectable in the receptor
phosphorylation assays. The increase in CD34+ cell proliferation also suggests that the decrease in TF-1
proliferation at high concentrations is not due to a non-specific cytotoxic effect of STS-E412. Regardless
of the mechanism underlying these contrasting results with different cell types, these effects occurred at
STS-E412 concentrations several orders-of-magnitude above those producing cytoprotective effects and
EPOR/CD131 receptor activation.
STS-E412 exhibited a bell-shaped dose-response curve in the EPOR/CD131 transfectants and
RPTEC cells, but not in primary neuronal cells. These observations are not surprising in light of previous
studies on the small molecule TPOR agonist eltrombopag, which resulted in bell-shaped dose-response
curves in receptor-expressing cell lines (Xie et al., 2012). In addition, eltrombopag triggered TPO-like
STAT5 activation in CD34+CD41- progenitor cells, but in contrast to rhTPO, eltrombopag did not activate
STAT3 or AKT in CD34+CD41+ cells (Sun et al., 2012). Bell-shaped dose-response curves were also

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
observed in previous in vitro studies on nonpeptidyl compounds activating the receptors for G-CSF and
EPO (Tian et al., 1998, Qureshi et al., 1999). The formation of 1:1 receptor/ligand complex at high ligand
concentrations was also shown to inhibit the activation of receptors for growth hormone and EPO (Fuh et
al., 1992, Schneider et al., 1997). The crystal structure of the GM-CSF receptor in complex with CD131
revealed a 2:2:2 hexamer consisting of two GM-CSFR alpha chains, two CD131 chains, and two GM-
CSF molecules, and also demonstrated the existence of a functional higher-order dodecamer, a hexamer
dimer (Hansen et al., 2008). The higher-order structures of the EPOR/CD131 complex are not known,
while the GM-CSFRα/CD131 data suggest that similar higher-order structures form when CD131
associates with the EPOR. The bell-shaped dose-response curve triggered by STS-E412 suggests that,
similar to the EPOR/EPOR, growth hormone, G-CSF and GM-CSF receptors, the formation of 1:1
receptor/ligand complex at high STS-E412 concentrations prevents the optimal activation of the
EPOR/CD131 receptor. However, the exact mechanism of receptor activation by STS-E412 remains to
be elucidated.
In the primary neuronal and kidney cell cytoprotection assays, STS-E412 exhibited significant
cytoprotective effects following a range of cytotoxic challenges. These data are consistent with previous
reports demonstrating broad cytoprotective effects of rhEPO, EPO variants selective for EPOR/CD131,
and mimetic peptides (Konishi et al., 1993, Sakanaka et al., 1998, Erbayraktar et al., 2003, Brines et al.,
2004, Leist et al., 2004, Coleman et al., 2006, Brines et al., 2008, Ahmet et al., 2011, Chen et al., 2012).
The maximal protective effects in response to STS-E412 were comparable to those induced by rhEPO,
and optimal responses were observed at low nanomolar concentrations in both neurons and RPTEC. It is
worth noting that staurosporine inhibits many intracellular kinases and may inhibit activation of the
tissue-protective EPOR/CD131 receptor and signaling molecules involved in protection. While our
experimental paradigm used a 24 hours pre-treatment (in the absence of staurosporine) and staurosporine
concentrations used in our experiments were relatively low, we cannot exclude the possibility that kinase
inhibition affected the observed cytoprotective effects of rhEPO and STS-E412. This concern, however,
does not associate with cytoprotective effects observed following glutamate and H2O2 challenge.

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
In summary, we report the discovery of STS-E412, which selectively activates the tissue-
protective EPOR/CD131 heteromeric receptor. The small size, permeability across PAMPA and Caco-2
cellular monolayers, and observed CNS availability in vivo suggest improved tissue permeability
compared to previously described nonpeptidyl EPO receptor agonists. Combined with the EPO-like
protective effects in primary neurons and RPTEC, these compound characteristics support further
evaluation of STS-E412 in neurodegenerative diseases and organ protection.

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
Author Contributions
Participated in research design: J.L.Miller, T.J.Church, D.Leonoudakis, K.Lariosa-Willingham,
N.L.Frigon, C.S.Tettenborn, J.R.Spencer, J.Punnonen
Conducted experiments: J.L.Miller, T.J.Church, D.Leonoudakis, K.Lariosa-Willingham,
N.L.Frigon, C.S.Tettenborn, J.R.Spencer, J.Punnonen
Contributed new reagents or analytic tools: J.L.Miller, T.J.Church, C.S.Tettenborn, J.R.Spencer,
J.Punnonen
Wrote or contributed to the writing of the manuscript: J.L.Miller, T.J.Church, D.Leonoudakis,
K.Lariosa-Willingham, N.L.Frigon, C.S.Tettenborn, J.R.Spencer, J.Punnonen

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
References
Abe M, Suzuki K, Sakata C, Sugasawa K, Hirayama F, Koga Y, Kawasaki T, Naganuma S, Itoh H (2011)
Pharmacological profile of AS1670542, a novel orally-active human thrombopoietin receptor
agonist. Eur J Pharmacol 650:58-63.
Ahmet I, Tae HJ, Juhaszova M, Riordon DR, Boheler KR, Sollott SJ, Brines M, Cerami A, Lakatta EG,
Talan MI (2011) A small nonerythropoietic helix B surface peptide based upon erythropoietin
structure is cardioprotective against ischemic myocardial damage. Mol Med 17:194-200.
Assaraf MI, Diaz Z, Liberman A, Miller WHJ, Arvanitakis Z, Li Y, Bennett DA, Schipper HM (2007)
Brain erythropoietin receptor expression in Alzheimer disease and mild cognitive impairment. J
Neuropathol Exp Neurol 66:389-398.
Bennis Y, Sarlon-Bartoli G, Guillet B, Lucas L, Pellegrini L, Velly L, Blot-Chabaud M, Dignat-Georges
F, Sabatier F, Pisano P (2012) Priming of late endothelial progenitor cells with erythropoietin
before transplantation requires the CD131 receptor subunit and enhances their angiogenic
potential. J Thromb Haemost 10:1914-1928.
Besarab A, Zeig SN, Martin ER, Pergola PE, Whittier FC, Zabaneh RI, Schiller B, Mayo M, Francisco
CA, Polu KR, Duliege AM (2012) An open-label, sequential, dose-finding study of peginesatide
for the maintenance treatment of anemia in chronic hemodialysis patients. BMC Nephrol 13:95.
Boulay JL, O'Shea JJ, Paul WE (2003) Molecular phylogeny within type I cytokines and their cognate
receptors. Immunity 19:159-163.
Brines M, Cerami A (2008) Erythropoietin-mediated tissue protection: reducing collateral damage from
the primary injury response. J Intern Med 264:405-432.
Brines M, Grasso G, Fiordaliso F, Sfacteria A, Ghezzi P, Fratelli M, Latini R, Xie QW, Smart J, Su-Rick
CJ, Pobre E, Diaz D, Gomez D, Hand C, Coleman T, Cerami A (2004) Erythropoietin mediates

Molecular Pharmacology Fast Forward. Published on May 27, 2015 as DOI: 10.1124/mol.115.098400
This article has not been copyedited and formatted. The final version may differ from this version.
tissue protection through an erythropoietin and common beta-subunit heteroreceptor. Proc Natl
Acad Sci U S A 101:14907-14912.
Brines M, Patel N, Villa P, Brines C, Mennini T, De Paola M, Erbayraktar Z, Erbayraktar S, Sepodes B,
Thiemermann C, Ghezzi P, Yamin M, Hand C, Xie Q, Coleman T, Cerami A (2008)
Nonerythropoietic, tissue-protective peptides derived from the tertiary structure of erythropoietin.
Proc Natl Acad Sci U S A 105:10925-10930.
Brines ML, Ghezzi P, Keenan S, Agnello D, de Lanerolle NC, Cerami C, Itri LM, Cerami A (2000)
Erythropoietin crosses the blood-brain barrier to protect against experimental brain injury. Proc
Natl Acad Sci U S A 97:10526-10531.
Calvillo L, Latini R, Kajstura J, Leri A, Anversa P, Ghezzi P, Salio M, Cerami A, Brines M (2003)
Recombinant human erythropoietin protects the myocardium from ischemia-reperfusion injury
and promotes beneficial remodeling. Proc Natl Acad Sci U S A 100:4802-4806.
Carr PD, Gustin SE, Church AP, Murphy JM, Ford SC, Mann DA, Woltring DM, Walker I, Ollis DL,
Young IG (2001) Structure of the complete extracellular domain of the common beta subunit of
the human GM-CSF, IL-3, and IL-5 receptors reveals a novel dimer configuration. Cell 104:291-
300.
Chateauvieux S, Grigorakaki C, Morceau F, Dicato M, Diederich M (2011) Erythropoietin, erythropoiesis
and beyond. Biochem Pharmacol 82:1291-1303.
Chen H, Spagnoli F, Burris M, Rolland WB, Fajilan A, Dou H, Tang J, Zhang JH (2012)
Nanoerythropoietin is 10-times more effective than regular erythropoietin in neuroprotection in a
neonatal rat model of hypoxia and ischemia. Stroke 43:884-887.
Coleman TR, Westenfelder C, F.E. T, Yang Y, Hu Z, Swenson L, Leuvenink HG, Ploeg RJ, d'Uscio LV,
Katusic ZS, Ghezzi P, Zanetti A, Kaushansky K, Fox NE, Cerami A, Brines M (2006)
Cytoprotective doses of erythropoietin or carbamylated erythropoietin have markedly different
procoagulant and vasoactive activities. Proc Natl Acad Sci USA 103:5965-5970.