Discovery of Novel Small Molecules that Activate ... - ACS Publications

Subscriber access provided by CMU Libraries - http://library.cmich.edu

Article

Discovery of Novel Small Molecules that Activate Satellite Cell Proliferation and Enhance Repair of Damaged Muscle Andrew N. Billin, Marcus Bantscheff, Gerard Drewes, Sonja Ghidelle-Disse, Jason A. Holt, Henning F Kramer, Alan J. McDougal, Terry L. Smalley, Carrow W. Wells, William J. Zuercher, and Brad R. Henke ACS Chem. Biol., Just Accepted Manuscript • DOI: 10.1021/acschembio.5b00772 • Publication Date (Web): 22 Dec 2015 Downloaded from http://pubs.acs.org on December 26, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Chemical Biology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Discovery of Novel Small Molecules that Activate Satellite Cell Proliferation and Enhance Repair of Damaged Muscle Andrew N. Billin†, Marcus Bantscheff‡, Gerard Drewes‡, Sonja Ghidelli-Disse‡, Jason A. Holt†, Henning F. Kramer†, Alan J. McDougal†, Terry L. Smalley†, Carrow W. Wells¥, William J. Zuercher¥, and Brad R. Henke*,† †

Muscle Metabolism Discovery Performance Unit, GlaxoSmithKline, Research Triangle Park, North Carolina, 27709, USA

‡

Cellzome AG, Meyerhofstrasse 1, 69117 Heidelberg, Germany

¥

Department of Chemical Biology, GlaxoSmithKline, Research Triangle Park, North Carolina, 27709, USA

ABSTRACT:

Skeletal muscle progenitor stem cells (referred to as satellite cells) represent the primary

pool of stem cells in adult skeletal muscle responsible for the generation of new skeletal muscle in response to injury. Satellite cells derived from aged muscle display a significant reduction in regenerative capacity to form functional muscle. This decrease in functional recovery has been attributed to a decrease in proliferative capacity of satellite cells. Hence, agents that enhance the proliferative abilities of satellite cells may hold promise as therapies for a variety of pathological settings, including repair of injured muscle and age- or disease-associated muscle wasting. Through phenotypic screening of isolated murine satellite cells we identified a series of 2,4-diaminopyrimidines (e.g. 2), that increased satellite cell proliferation. Importantly, compound 2 was effective in accelerating repair of damaged skeletal muscle in an in vivo mouse model of skeletal muscle injury. While these compounds were originally prepared as c-Jun N-terminal kinase 1 (JNK-1) inhibitors, structure-activity analyses indicated JNK-1 inhibition does not correlate with satellite cell activity. Screening against a broad panel of kinases did not result in identification of an obvious molecular target, so we conducted cell-based proteomics experiments in an attempt to identify the molecular target(s) responsible for the potentiation of the satellite cell proliferation. This data provides the foundation for future efforts to design improved small molecules as potential therapeutics for muscle repair and regeneration.

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Skeletal muscle, which accounts for approximately 40% of human body weight, is crucial for both locomotion and energy metabolism. Age-associated muscle atrophy, sarcopenia, often leads to progressive disability and loss of independence in the elderly and is an independent risk factor for falls, duration of hospitalization, and mortality.1 Cachexia, the loss of muscle mass (with or without loss of fat mass) due to underlying diseases such as cancer, chronic heart failure, chronic obstructive pulmonary disease (COPD), end-stage renal disease, and HIV infection, is also an independent risk factor for mortality and morbidity. Genetic loss of function in genes required for skeletal muscle structural integrity leads to muscular dystrophies that exhibit cycles of muscle damage and repair2, where the ability to adequately repair damaged skeletal muscle degrades over time. Clearly, interventions that improve the retention of skeletal muscle mass, regenerative capacity, and function would find wide medical application. Skeletal muscle is comprised of multinucleated contractile muscle cells called myofibers, which are formed by fusion of progenitor cells called myoblasts. During neonatal / juvenile development muscle mass increases mainly by proliferation of myoblasts, which involves the fusion of muscle stem cells also called satellite cells.3 In adults, the contribution of cell proliferation decreases, and hypertrophic growth and remodeling of pre-existing myofibers dominates. Adult skeletal muscle is relatively stable under normal conditions, with only sporadic fusion of satellite cells to compensate for muscle turnover caused by daily activity. However, skeletal muscle possesses a remarkable ability to regenerate after being damaged. This regenerative process involves a dynamic interplay between satellite cells and their local environment (i.e., stem cell niche). Like all tissues, skeletal muscle undergoes a progressive decline in both function and regenerative capacity as a result of the aging process. Much of this decline has been attributed to loss of satellite cells and alteration of the stem cell niche.4 The long lifespan of satellite cells makes them particularly susceptible to the accumulation of cellular damage, and ultimately leads to defects in selfrenewal ability. This results in a depletion of the satellite cell pool with defects in activation/proliferation potential, resulting in impaired generation of new myoblasts.5 The practical consequence of the diminished regenerative capacity is delayed or incomplete muscle repair and restoration of function, which can be particularly debilitating after severe injury, or upon reloading of an immobilized limb. Recent advances in molecular biology, cell biology, and genetics has greatly improved the understanding of the role satellite cells play in skeletal muscle regeneration.6,7 Importantly, heterochronic parabiosis experiments, in which the circulatory systems of old and young mice are joined, have shown that the age-related decline in satellite cell activity can be restored by systemic factors.8 In addition, rejuvenation of muscle stem cell niches has been achieved through heterografting of muscle between young and old rats and mice.4 Small molecules that regulate resident hematopoetic stem cell populations are now marketed, and regulators of other tissue-specific progenitor cells are advancing through clinical development9,10, suggesting that modulation of satellite cells through oral


Page 2 of 23

Page 3 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


dosing of small molecules is a viable therapeutic modality. In work similar to the studies we describe below, Nierobisz et. al.11 have reported the development of a high-content imaging assay using satellite cells obtained from obese subjects. This assay was utilized to screen a 1600-member compound library, resulting in the identification of small molecules with proliferative activity on satellite cells. As part of our group’s ongoing effort to discover and develop novel therapeutic agents for skeletal muscle diseases, herein we report the discovery of a series of 2,4-diaminopyrimidines that increase proliferation of murine satellite cells in vitro. These compounds were identified via phenotypic screening of a population of murine adult satellite / precursor cells.12 Subsequent initial refinement by standard methods of medicinal chemistry and drug design afforded additional compounds with modestly improved cellular activity. One of these compounds, compound 2 (GSK 2226649A) was effective in accelerating repair of damaged skeletal muscle in an in vivo mouse model of skeletal muscle injury. In addition, we describe our initial efforts to determine the molecular target(s) by which these compounds potentiate satellite cell proliferation.

RESULTS AND DISCUSSION It is well established that satellite cells from aged animals have a decrease in proliferative capacity and therefore are inadequate at regenerating functional muscle to replace damaged or atrophied skeletal muscle.4 Thus we set out to identify small molecule inhibitors capable of causing an increase in the proliferative capacity of satellite cells isolated from aged (>24 months) mice. We isolated satellite cells via Fluorescent Activated Cell Sorting (FACS). Using satellite cell specific positive and negative selection markers described in Conboy, et. al.,4 we isolated a pure population of skeletal muscle stem cells with no other cell type present. Satellite cells were then plated in culture and treated with test compounds at 10 µM for 5 days in satellite cell proliferation media. After 5 days of treatment the number of cells per well were manually counted by two independent researchers to determine if there was an increase in the number of cells compared to vehicle control. If there was a 1.3-fold increase in the number of satellite cells, we then performed a three-point dose response assay at compound concentrations of 0.1, 1, and 10 µM to gauge the potency of compounds on proliferation. As a positive control, we used GW849825X, an activin receptor-like 4/5 (Alk 4/5) kinase inhibitor which has previously been shown to increase the proliferation of satellite cells.13,14 This compound was used at 0.5 µM concentration and consistently provided >2-fold increase in satellite cell proliferation vs. vehicle control. The cost and technical challenges with isolating large numbers of stem cells from aged mice made a ‘non-directed’ screening strategy using large numbers of compounds unfeasible. We therefore chose to initially screen a set of compounds (the “Stem Cell Toolbox”) that was designed as a focused set of small molecules used to explore stem cell determination pathways. This compound set was handcurated by GSK medicinal chemists; we did not use any computational algorithms or pathway analysis as selection criteria for inclusion into the toolbox. The Stem Cell Toolbox consisted of approximately 800 compounds, which had demonstrated an ability to affect stem cell populations. Included within the set



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

were diverse kinase inhibitors, compounds that modulate pathways (e.g. Wnt, Notch, β-catenin, Hedgehog, etc.) or molecular targets known to influence stem cell pathways, and compounds that affected epigenetic targets (e.g. bromodomain and HDAC modulators). Compounds of unknown mechanism observed to induce stem cell maintenance or determination formed an additional component of the Toolbox. Approximately 15% of the compounds in the set were from the published literature, with the remainder being proprietary compounds from various GSK preclinical discovery programs. Importantly, all compounds in this set were both well-annotated and had demonstrated activity in some type of cellular screen. Screening this ca. 800-compound set in our aged mouse stem cell assay at 10 µM single concentration in duplicate resulted in 27 compounds that displayed >1.3-fold increase in the number of satellite cells vs. vehicle control (3.3% hit rate). Of those 27 hits from the primary screen, 8 were confirmed as active via the three-point dose response protocol described above. It should be noted that due to the low yield of satellite cells from aged mice, we decided to screen with a format that did not allow us to run an adequate number of controls to properly calculate Z prime. However, the screen was adequately powered to differentiate a positive hit from an inactive compound. Six of the eight confirmed hits were weakly active at the highest concentration and thus were not pursued. One of the two remaining hits turned out to be a ‘flagpole’ compound, with close structural analogs having no activity in the satellite cell assay (data not shown). The 2,4diaminopyrimidine compound 1 (Figure 1), which showed robust proliferative effects (2.4-fold increase vs. controls) on mouse stem cells at a 10 µM concentration, was the most active hit from the screen. This compound had been previously prepared as part of an internal drug discovery program focused on c-Jun N-terminal kinase 1 (JNK-1) inhibitors for Type 2 diabetes. A search of the GSK database for similar compounds allowed us to screen a few close structural analogues of compound 1. Through this brief screening exercise we identified compound 2 (GSK2226649A) (Figure 1) as a more potent modulator of stem cell proliferation versus the original hit compound.

b


Page 4 of 23

Page 5 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


0.1% DMSO

Compound 2 @ 10 µM

Figure 1. (a) Chemical structures of original screening hit 1 and in vivo tool molecule 2. (b) Proliferative effects of compound 2 on murine muscle stems cells when dosed at 10 µM concentration. Before investing additional resource into screening and chemistry, we wished to determine whether the proliferative effects observed in our stem cell assay could be translated into a functional benefit in an in vivo setting. While the original screening hit 1 did not have pharmacokinetic (PK) properties suitable for an in vivo tool, the PK parameters of compound 2 were sufficiently good for use in an in vivo setting. Table 1 summarizes the PK properties of compound 2. We anticipated dosing via subcutaneous (SC) injection in our animal model, and the exposure and half-life of compound 2 via SC dosing in normal mice suggested sufficient drug concentration could be achieved in muscle with sufficient duration to enable proliferative effects on muscle stem cells in vivo. Table 1. Pharmacokinetic summary of compound 2. Dose (mg)

Route

Cmax (ng/mL)

Tmax (hrs)

T1/2 (hrs)

AUC0-inf (ng*hr/mL)

30 Oral 1384 0.5 1.6 1720 30 SC 1783 1.0 6.1 12077 Compound was dosed as a solution in 0.5% HPMC: 0.1% Tween 80; data are an average of n=2 mice. There are a paucity of in vivo models that have been developed to examine repair of damaged muscle. We utilized a modified version of a mouse model of eccentric (lengthening) contraction-induced muscle damage.15,16 This model uses a repetitive eccentric muscle contraction protocol to induce muscle injury in which the hind limb plantarflexor muscles are forcibly lengthened using an automated footplate while being concomitantly stimulated to contract. This results in damage to the hindlimb myofibers and is intended to reasonably mimic the physiological damage that would occur from a bout of heavy resistance training in an untrained individual. The resulting muscle damage reproducibly manifests as a reduction in maximal hindlimb tetanic force production of approximately 30-40%, and generally takes approximately 30 days to recover to baseline. As a control, a group of mice were subjected to non-injuring isometric contractions in order to establish a baseline for force production in the particular batch of mice used for the experiment. Either the compound or vehicle was dosed by subcutaneous injection 10 minutes prior to initiating the injury protocol and once a day thereafter. At the days indicated after the first dose and injury the mice were



tested for limb force production and the recovery of force production was followed for 20 days. The results of the experiment are shown in Figure 2. Mice treated with vehicle exhibited a 30% force production deficit 24 hours after the initial injury. As expected, mice subjected to isometric contractions exhibited no such deficit. Mice dosed with compound 2 at 10 mg/kg exhibited a force deficit similar to that seen in vehicle treated mice. Mice dosed with 30 mg/kg compound 2 were less affected by the injury and exhibited a 20% force reduction. Between days 1 and 3 after injury no improvement in force production was observed in any of the injured groups. Between day 6 and 10 after injury the vehicle treated group still exhibited ~30% force deficit while the drug-treated groups demonstrated robust recovery of force production (10–20% force deficit). By 20 days post-injury the drug-treated groups had fully recovered force production (0% deficit compared to isometric controls) and the vehicle treated group still exhibited a ~15% force deficit. Thus compound 2 promotes functional recovery from a muscle injury consistent with its activity as a promoter of muscle precursor cell out-growth. Interestingly at the 30 mg/kg dose compound 2 also appears to preserve force production or prevent loss of force production in the first three days after the injury, suggesting that the compound many also affect the extent of the initial injury. Since muscle satellite cells and their progeny do not begin to differentiate into myocytes capable of force production until four or five days after injury it is unlikely that the effects observed during the first three days postinjury are mediated by stem cell activity. The exact mechanisms regulating the observed effects remain to be determined. Fraction of Maximal Undamaged Limb Force

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1.1 1 0.9 Veh Iso 0.8

Veh Ecc 649A 10mg Ecc

0.7

649A 30mg Ecc 0.6 0.5 0

2

4

6

8 10 12 14 16 18 20 22 Day

Figure 2. Muscle regenerative effects of compound 2 in an in vivo mouse model of muscle injury. After induction of muscle injury to one hindlimb via repetitive eccentric contractions (Ecc), mice were dosed subcutaneously with either vehicle (Veh) or compound 2 (649A) for three weeks. The contralateral limb (Iso) in each mouse served as a control.


Page 6 of 23

Page 7 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Having established that compound 2 demonstrated efficacy in an in vivo model of muscle damage, we initiated simultaneous efforts to both improve the cellular potency of our compounds and to determine the molecular target(s) of our compounds. The most obvious possible molecular target was JNK-1, since compound 1 was a potent inhibitor of the isolated JNK-1 enzyme in a biochemical assay and showed submicromolar effects in a JNK-1 cellular assay measuring inhibition of c-Jun phosphorylation. Therefore we screened a small set of structural analogues of compound 1 from our defunct JNK-1 program that had a range of potencies against JNK-1 in both enzyme and cell-based JNK-1 assays in our satellite cell proliferation assay. The chemical structures of these compounds can be found either in Table 3 (vide infra) or in Supplementary Figure S1. The data from this screen are summarized in Table 2. A number of compounds in this set (e.g. 3, 41, 42, 46) profiled similarly to compound 1, showing both potent JNK-1 inhibition and proliferative activity in our muscle stem cell assay. However, compounds such as 43 and 49 displayed potent enzyme and cellular JNK-1 inhibition profiles yet were inactive in the muscle stem cell assay. In addition, compounds such as 2, 27, 50, and 51 had proliferative effects in the muscle stem cell assay at concentrations that were below what one might expect based on their JNK-1 potencies if the effects were mediated via JNK-1 inhibition. We concluded that these results showed enough divergence in JNK-1 activity and muscle stem cell proliferation to indicate that inhibition of JNK-1 is not a key mediator of the muscle stem cell effects exhibited by compound 1. Table 2. Activities of JNK-1 series compounds in JNK-1 biochemical and cellular assays and in the muscle stem cell proliferation assay. Compound1

JNK-1 JNK-1 Cell Stem cell Stem cell Stem cell Enzyme IC50 Assay IC50 activity3 activity3 activity3 2 2 (0.1 µM) (1 µM) (10 µM) (µM) (µM) 1 0.002 0.316 + + ++ 41 0.006 0.158 + +++ ++ 42 0.016 0.316 + +++ 43 0.020 0.126 4 44 0.020 NT 45 0.020 2.0 + + 46 0.025 0.50 ++ ++ 47 0.04 6.31 + 48 0.04 4.00 + 49 0.04 0.40 3 0.316 NT + ++ +++ 50 0.40 10.0 + + 2 1.0 NT + +++ +++ 51 10.0 NT + + + 52 0.063 0.79 ++ 29 0.40 NT + +++ +++ 27 0.25 31.6 + +++ 1 2 Compound structures are in Table 3 or in Supplementary Figure S1. See Methods section for details. 3 Activity in muscle stem assay is defined as: 2.5x increase in proliferation vs. vehicle = “+++”.4NT = not tested.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

In an attempt to improve cellular potency and obtain further structure-activity data, we prepared additional 2,4-diaminopyrimidine analogues following the general procedure depicted in Scheme 1. Initial amine substitution occurred exclusively at the 4-chloro position by heating 2,4,5trichloropyrimidine with an amine of choice in isopropanol at 90 °C.17 The products were routinely isolated by cooling the reaction mixture and filtering the product. Diversification of the 2-position was accomplished by heating the 2,5-dichloropyrimidine with amines of interest in a number of different solvents at slightly higher temperatures (100–120 °C), with the choice of solvent being dependent on the solubility of the amine monomers. Purification of the target compounds was achieved by silica gel chromatography or HPLC.

Scheme 1: (a) R1-NH2, (i-Pr)2NEt, i-PrOH, 90°C; (b) R2-NH2, (i-Pr)2NEt, solvent, 100–120°C. Initially we chose to explore different heterocyclic rings at the C2 position, with the goal of replacing the thiazole moiety. Our rationale to initially target the C2 position was based on two factors. First, this position was more accessible to analog synthesis given it is the last step in the synthetic sequence. Second, some unsubstituted thiazoles have been reported to be metabolized to hepatotoxic compounds18, so we wished to find a suitable replacement for this group early in our hit optimization work. Compounds 3–28, where the thiazole was replaced with various aromatic and saturated heterocyclic rings, were synthesized and shown to affect satellite cell proliferation. In particular, compound 3 showed very good activity at both 1 µM and 10 µM, similar to 2. Interestingly, compounds 5, 7, 8, 20 and 28 showed cell proliferation effects at 1 µM but the activity was reduced at 10 µM. We reasoned that the reduction in activity at 10 µM was due to unknown cell toxicity of the compounds at the higher concentration. Weak proliferation effects were observed for 6, 10, 11, 14, 16, 19, 21 and 25 with activity only being observed at the 10 µM concentration. No effect was observed for compounds 22–24. Table 3. Proliferative activity in the muscle stem cell assay for compounds 2–40.


Page 8 of 23

Page 9 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Compound


R1

Stem cell activity1 Stem cell activity1 (0.1 µM) (1 µM)

R2

Stem cell activity1 (10 µM)

2 -

+

+++

-

+++

+++

+

+++

+

-

+++

++

-

-

+

-

+

-

N

-

+++

++

N N H

-

-

++

-

-

++

-

-

+

-

-

+++

-

-

++

-

-

+++

-

+

-

-

-

++

-

-

+

-

+

++

-

-

++

3 O N

4 5 6 7 8

N

9 10 11 12 13 14 15 N

16 N

17 18 19



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

20

Page 10 of 23

O N

+

+++

+

-

-

+++

-

-

-

-

-

-

-

-

-

-

-

+++

+

+++

+++

+

+

+++

+

+++

+++

-

+++

-

-

-

+

-

-

-

-

++

+++

-

+

+++

-

-

+

-

-

-

-

+

-

21

22 23 24 O

25

N H

26 27 28 29 30

31

32

H2N

33 S

O

34 35

36


Page 11 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


37 -

-

-

+

+

-

+

+++

+++

-

++

++

38

39

40

1

Activity in muscle stem assay is defined as: 2.3x increase in proliferation vs. vehicle = “+++“. See Methods section for experimental details. Values are from n=1 experiment.

The results of our initial compound set showed several interesting trends. Subtle effects in regards to the substitution point were observed, for example, in the N-methylimidazole analogues 6–8. The 4-substituted analogue 7 was weakly active at 1 and 10 µM, whereas the 2-substituted analogue 6 showed an effect only at 10 µM. The 5-substituted analogue 8 displayed a ‘bell-shaped’ concentrationresponse, with excellent proliferative efficacy at 1 µM but less efficacy at 10 µM, suggestive of cytotoxicity at the higher concentration. A similar ‘bell-shaped response was observed with Nmethylpyrazole analogue 5 which was highly efficacious at 1 µM but showed lower proliferative capacity at 10 µM. Replacement of the C-2 aromatic substituents with saturated heterocycles was generally tolerated, as shown by compounds 13–29. In general these compounds were active only at the highest concentration, although compounds 26 and 27 showed excellent efficacy at the 1 µM concentration. The N-methylpyrrolidin-2-one (20) and the dioxane (29) analogues had profiles that suggested cytotoxicity at the higher concentration. Interestingly, removal of the methyl group from the pyrrolidine analogues 16 and 17 produced compounds with loss of activity (23–24). In summary, thiazole substitution with several different groups could be achieved at the C2 position while maintaining cellular activity, with the isoaxazole (3), tetrahydrofuran (26–27), and tetrahydropyran (28) analogues showing very good activity at both the 1 and 10 µM concentrations. With several interesting heterocyclic replacements available at C2, we turned our attention on making different analogues to replace the C4 fluoroamide portion of 2. The fluoroamide benzene ring is unique in that the conformation appears highly rigid, with an intramolecular hydrogen bonding network. Several different analogues were designed with the goal to challenge and/or confirm the importance of such a flat, rigid structure to the activity. Analogues such as 33 and 36 were chosen as simple benzene ring replacements. Compound 32 was chosen to mimic the putative F-H hydrogen bond, which is believed to be one factor in restricting the conformation, while retaining an amide N-H. Analogues such



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

as 34–35 were chosen to mimic the putative N-H carbonyl hydrogen bond, believed to be a factor in conformationally constricting the ring, yet removes the N-H bond. Of these analogues, 32 and 33 showed excellent activity at 1 and 10 µM. The N-methylpyrazole analogue 36 showed weak activity at 1 µM but toxicity at 10 µM. The methyl ether analogue 34 showed activity only at 10 µM, and the saturated analogue 37 and oxadiazole analogue 35 showed no activity in the assay. These results seemed to show that heterocyclic replacements of the benzene ring are possible, along with bicyclic modification. The results also suggest that the amide N-H may be important for activity, although additional work is needed to confirm the observed trend. Based on the activity profiles generated, a few analogues combining some of the better C2 and C4 pieces were made (38–40). Surprisingly, while compound 38 showed activity at 0.1 µM, the fold induction was weak and did not improve with increasing compound concentration. Compound 39 was active at all 3 concentrations and very efficacious at the 1 and 10 µM concentrations, while compound 40 showed good proliferative activity at the 1 and 10 µM concentrations. In summary, this small set of analogues provided several important pieces of information. First, a number of active analogues were made and there is clear SAR within the series, suggesting that the early hit compounds in the series are not ‘flagpoles’. Second, the early SAR is relatively flat thus far, as we were unable to significantly increase cellular potency over the original hit, albeit with only a small number of compounds prepared to date. Third, conducting this exercise reinforced the challenges of utilizing this cellular assay for compound optimization purposes. The low compound throughput, coupled with narrow dynamic range of proliferation and apparent cellular toxicity issues at higher compound concentrations led to slow progress and difficulty in drawing firm SAR conclusions from the resulting assay data. This underscored the importance of identifying the molecular target(s) responsible for the stem cell proliferation in order for the program to rapidly advance. Though we were able to rule out JNK-1 as the mediator of stem cell activity, the structure of our lead series contains a classic kinase ATP-site binding motif. Therefore we conducted a broad panel kinase screen to investigate whether another kinase was responsible for the cellular effects. Compounds 1, 2, and 45, all active at both 1 and 10 µM in the stem cell rejuvenation assay, were selected for testing against a panel of 294 kinases. The compounds were tested in duplicate experiments at a single concentration of 1 µM in the presence of 10 µM ATP and compared to DMSO controls. A detailed summary of these screening results is found in Supplementary Table S1. We defined less than 50% enzyme activity relative to DMSO controls as ‘active’ and therefore of potential further interest. Only 2 kinases, casein kinase 1 gamma (CK1g) and NIMA-related kinase 1 (NEK1), profiled as showing 10 >10 NA3 + 15 >10 NA NA ++ 16 >10 >10 >10 ++ 24 >10 >10 >10 + 1 Compounds were tested in 10-dose IC50 mode with 3-fold serial dilution starting at 10 μM. The following ATP concentrations were used: CK1g1, CK1g3 = 20 µM ATP; CK1g2 = 30 µM ATP. 2Activity in muscle stem assay is defined as: 2.3x increase in



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

proliferation vs. vehicle = “+++“. 3NA = no inhibition or compound activity that could not be fit to an IC50 curve. At this point we decided to apply a chemoproteomics strategy in an attempt to elucidate the molecular target(s) responsible for the effects of our compounds. In these experiments, sepharose beads are derivatized with small molecule ligands designed to bind to the targets of interest.19 The beads are then incubated with cell extracts in the presence of either vehicle (DMSO) or various concentrations of the bioactive molecules of interest, or with inactive analogues. The presence of an active molecule in the cell extract will lead to selective dose-dependent reduction of binding of the target proteins to the beads. The proteins captured on the beads are then identified and quantified by LC-MS/MS analysis, following tryptic digestion and isobaric peptide tagging with tandem-mass-tag (TMT) reagents. We used the TMT6 reagents which enables the direct quantitative comparison of six different samples in a single LC-MS/MS run, which is ideal to generate dose-response binding (IC50) curves.

Figure 3. Structure of chemical probes used in chemoproteomics experiments.


Page 14 of 23

Page 15 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


In order to maximize the chances of capturing the targets of our compounds, we used two different types of beads. The first set of experiments was performed with beads derivatized with tagged analogues of the active molecule GSK2226649A modified in three different positions (Figure 3). These positions were chosen primarily based on synthetic ease, since we had minimal structure-activity information from our stem cell assay. The synthesis of these chemical probes are found in Supplementary material. For synthetic reasons, the des-fluoro analogue of compound 2 was utilized in the preparation of probe N. Importantly, the intermediates of all three probes containing the linkers prior to attachment to the beads (Figure 3) all showed activity in the muscle stem cell assay, which suggested that attachment of the linkers in these regions did not negate interactions with the molecular target(s) responsible for the observed proliferative response. Since satellite cells were not available in sufficient quantities to conduct these experiments, we initially used HeLa cells. Two of the immobilization positions (probes N and W, Figure 3) appeared to be permissive for kinase binding whereas the third position (probe E) retained little specific protein binding as indicated by few proteins affected by competing with excess compound 2. The data obtained with probes N and W indicated that only a few kinases, including TP53RK and ASK2/3, were more efficiently competed from the beads by the active compared to compound 43, which served as our inactive control (Supplementary Table S2). In order to extend our analysis to kinases expressed in a muscle-derived cell type, we performed a similar pulldown experiment in A204 rhabdomyosarcoma cells using probe N (Supplementary Table S3). This experiment confirmed the preference of compound 2, as compared to the inactive analogue 43, for TP53RK, and suggested TAOK3 as an additional target candidate. We repeated this experiment, and profiled compound 2 over a range of concentrations, rather than a single concentration, to obtain IC50 values to rank the potential targets according to their affinity (Supplementary Table S4). The experiment yielded IC50 values of 0.39 µM for TAOK3 and 0.78 µM for TP53RK, respectively. For TAOK3, there was a clear window compared to the inactive analogue 43 (IC50 = 6.6 µM), but not for TP53RK (IC50 = 0.42 µM). However, TP53RK is identified in our experiments as a complex with OSGEP and TPRKB (TP53RK binding protein). For these proteins we also obtained IC50 values, which showed a more pronounced binding to the active over the inactive compound. A second set of experiments was performed with “kinobeads”, which are derivatized with a combination of different promiscuous kinase inhibitors.20 The kinobeads protocol enables the profiling of a compound against a large fraction of the kinases expressed in the cell line under study.21 We profiled the active compound 2 and the inactive analogue 43 in this system for approximately 120 kinases expressed in A204 cells. The data confirm and extend the data from biochemical profiling and were also in line with the data obtained with the bespoke tagged analogues (Supplementary Table S5). Both compounds are potent inhibitors of the p90 S6 kinases (RPS6KA1 and RPS6KA3), whereas the inactive compounds is 50-100 times more potent for the JNKs (MAPK8, MAPK9). Notable targets affected by the active but not the inactive compound included PKCd, TAOK3, CK1a, and CK1d, with IC50 values in the low micromolar range. Taking all the chemoproteomics data together, the best target candidates, based on the assumption that the target should exhibit a window between compound 2 and the inactive analogue 43, are TAOK3 and PKCd, followed by the TP53RK complex.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Of the three potential targets obtained from our proteomics studies, TAOK3 is arguably the best validated target based on the scientific literature. TAOK3 is known to act as an upstream activator of the p38 MAP kinase pathway in response to stress such as DNA damage.22 Activation of this pathway can lead to cell cycle arrest and cellular senescence. It is well known that senescence is a key negative regulator of muscle stem cell activity during the aging process23 and it was recently demonstrated that inhibiting the p38 MAPK pathway in satellite cells reverses senescence and restores proliferative capacity to these stem cells.24,25 Thus the known biology of TAOK3 as a positive regulator of p38 MAP kinase activity is consistent with the possibility that it is a key protein target of our molecules. PKCd is not known to be connected to the biology of stem cell aging or senescence and thus the relevance of this protein to the activity of compound 2 in our satellite cell assay is not clear. On the other hand, TP53RK is a kinase that phosphorylates and may activate the tumor suppressor p53, a molecule with clear links to aging and senescence. The activation of p53 activity is known to inhibit myogenesis.26 Thus, as a positive regulator of p53, TP53RK may potentiate p53 activity and inhibit myogenesis. Inhibition of TP53RK activity by our compounds would therefore produce a proliferative effect in our satellite cell assay, and hence there is some rationale for TP53RK as a target of our compounds. Further interrogation of the function of TAOK3 and TP53RK with additional small molecule inhibitors or via siRNA may serve to clarify the identity of the targets(s) of compound 2. In the present report, we describe the identification of a series of compounds that stimulate the proliferation of murine muscle stem cells (satellite cells) in vitro. By using previously established cell selection markers and FACS techniques we isolated a pure population of murine satellite cells from aged mice and established a robust, albeit low-throughput, phenotypic screen. Using a pre-designed, wellcurated screening set of 800 compounds that had previously demonstrated effects on stem cell populations, we screened for compounds that could potentiate satellite cell proliferation. Compound 1, a proprietary compound from a previous JNK-1 inhibitor program at GSK, was identified as the most interesting of the screening hits. Through the preparation and screening of a small set of structural analogues, we have established that multiple compounds within this series show proliferative activity, and we have seen modest increases in cellular potency and efficacy. One of the compounds from this series, compound 2, was utilized to establish that the effects observed on satellite cell proliferation in vitro could translate into a functional response in vivo. Compound 2 demonstrated a dose-dependent ability to accelerate repair of damaged skeletal muscle in a mouse model of skeletal muscle injury. We have also attempted to elucidate the molecular target(s) through which these compounds enhance satellite cell proliferation. Screening a larger set of known JNK-1 inhibitors led us to conclude that JNK-1 was not responsible for the observed effects in the satellite cells. Screening selected active compounds against a broad panel of kinases did not lead to an obvious molecular target, though due to the small dataset and lack of available structurally distinct inhibitors we cannot rule out CK1g as a possible protein target. We also applied a chemoproteomics strategy for target elucidation that has provided TAOK3, TP53RK, and PKCd as potential candidates, though further work is necessary to validate these proteins as bona fide targets of our compounds.


Page 16 of 23

Page 17 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Many questions and challenges remain besides the identification of the protein target(s) responsible for the observed cellular effects of these compounds. Though the compound numbers are small, the initial structure-activity profile of this series is relatively flat, with only modest gains in cellular potency seen to date. In addition, the presence of apparent cellular toxicity in some of the compounds, leading to a ‘bell-shaped’ dose-response, confounds interpretation of these results. Improvements in cellular potency and physicochemical properties of this series will be required in order to produce a high-quality clinical drug candidate. While we have initial evidence that the observed effects on satellite cell proliferation may translate in an in vivo setting, we have not yet fully established causality. Importantly, we have yet to determine the translatability of the observed effects to human satellite cells. Our pursuit of these challenges will be the subject of future reports.

METHODS Satellite Cell Isolation and Treatment Satellite cell isolation from aged mice (>24 months) was adapted from Conboy et. al.12 Briefly, skeletal muscle from the hindlimbs of aged mice were harvested. Muscles were digested in collagenase/dispase mixture to release satellite cells from myofibers. Samples were run on a FACSAria II using positive and negative selection markers. Once the pure population of cells were isolated they were plated in 384 well microclear plates pre-coated with collagen and laminin to ensure adherence of cells. 24 hours after plating a 2x volume of compound diluted in satellite cell proliferation media was added to each well. Cells were treated for 5 days in proliferation media. After 5 days of treatment cells were fixed and stained with Hoecht-33342. Cells were then imaged on BD Pathways imager. Number of cells per well were determined by counting number of nuclei. In Vivo Isometric Force Measurement Mice were placed on a warming plate (30-32°C) and anesthetized using 2%/L O2 isoflurane. Right hind limbs were first shaved and then pinned at the knee at 90° with foot strapped into a mouse “shoe” which doubles as a motor arm and force transducer (Aurora Scientific Instruments 305C, Aurora). Platinum sub-dermal electrodes (Grass Instruments; West Warwick, RI) were inserted dorsally and ventrally to the femur for field stimulation of the sciatic nerve. Because the mouse gastrocnemius, plantaris, and soleus muscles are larger and stronger as a complex than the ankle dorsiflexors (TA and EDL), hind limbs produced a net plantarflexion response to sciatic activation. Tetanic stimuli of 150Hz at 200µsec pulse width (with 2mA current and 15V) for 0.8 seconds were recorded as maximal isometric limb forces while single 200usec pulses were used to elicit maximal twitch force values. Eccentric Contraction-Induced Muscle Damage To induce muscle injury, mice were positioned as described above and initial baseline maximal twitch and tetanic force measurement were performed. After a 2 minute rest interval, mice underwent a unilateral eccentric muscle contraction protocol in which the hind limb plantarflexor muscles were forcibly lengthened using the automated footplate (30 degrees angular rotation, 1800 degrees/s) while being concomitantly stimulated to contract (150 Hz; 200 µsec pulse width; 2 mA and 15V). A bout of eccentric exercise consisted of 1 forced lengthening contraction every 10 seconds for a total of 60 repetitions. HTRF JNK-1 Enzyme assay



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Test compounds were diluted to appropriate concentration and added to a 384 well plate. 5 µl of 2x enzyme free solution (50 mM HEPES pH7.4, 0.2 mg/ml BSA, 10 mM MgCl2, 2 mM DTT) was added to control wells. 5 µl of 2x enzyme solution (Enzyme free solution + 2 nM JNK-1 enzyme) was added to test wells. Plates were covered to minimize evaporation. Plates were incubated for 30 minutes at RT. After 30 minute incubation, 5 µl 2x substrate solution (50 mM HEPES pH7.4, 0.2 mg/ml BSA, 10 mM MgCl2, 4 µM ATP, 200 nM GST-biotin-c-Jun) was added. Plates were covered and incubated for 45 minutes at 22 °C. Next,5 µl of stop solution (50 mM HEPES pH7.4, 50 mM EDTA pH8.0) was added to each well. 5 µl of HTRF solution (50 mM HEPES pH7.4, 0.2 mg/ml BSA, 12 nM Streptavidin SureLight APC, 6 nM anti phospho-c-Jun Ab) was then added to each well. Approximately 30 minutes after addition of HTRF solution, plates were read on a Viewlux or other appropriate plate reader. JNK-1 Cell based assay On day 1, 3T3 cells were plated at 20,000 cells per well in 96 well plates pre-coated with collagen. Cells were incubated at 37° C for 6 hours at 10% CO2. Cells were then transfected with JNK-1 and c-Jun. Cells were incubated overnight. On day 2, 30 nM Mifepristone (Invitrogen, H11-01) was added to each well followed by incubation at 37° C for 4 hours. After 4 hour incubation with Mifepristone, cells were treated with compound at test concentration. Cells were treated for 2 hours at 37° C. Separately on day 1, Anti-flag antibody was added to wells of a 96 well half area black EIA plate and the plate was incubated overnight at 4 °C. The EIA plate was washed 5 times with wash buffer. Blocking buffer (5% BSA containing PBST) was added to each well and the plate was incubated at RT for 2 hours, then the blocking buffer was removed from the EIA plate. The compound-treated 3T3 cells were then lysed and the cell lysate was transferred to the 96 well EIA plate. The EIA plate was incubated at 4° C overnight. The EIA plate was washed with buffer 5 times. Rabbit anti-phospho c-jun (Ser63) antibody (Cell Signalling Technology, 250552) was then added to each well and the cells were incubated at RT for 6 hours. The plate was washed 5 times, then goat anti-rabbit IgG HRP (Cell Signalling Technology, 246210) was added to each well and incubated at RT for 1 hour. The plate was again washed 5 times, then chemiluminescent substrate was added to each well and incubated 3 minutes at RT under mixing conditions. Chemiluminescence was then measured. Chemoproteomics Detailed procedures for chemoproteomics experiments can be found in references 18 and 19. General Procedure for Synthesis of 4-amino-2,5-dichloropyrimidines: 2-((2,5-Dichloropyrimidin-4yl)amino)-6-fluorobenzamide

2,4,5-Trichloropyrimidine (5.0371 g, 27.5 mmol) was dissolved in isopropanol (54 mL) in a 150 mL pressure tube. 2-Amino-6-fluorobenzamide (4.3285 g, 28.1 mmol) and N,N-diisopropylethylamine (24.0 mL, 137 mmol) were added, the tube was sealed and the mixture was heated to 90 °C for 21 hours. The mixture was cooled to RT and the precipitated solid was collected by filtration. The solid was washed with isopropanol and dried under vacuum to provide the product (3.7130 g, 45%) as a tan solid. 1H NMR


Page 18 of 23

Page 19 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(d6-DMSO): δ 10.86 (s, 1H), 8.51 (s, 1H), 8.22 (s, 1 H), 8.18–8.07 (m, 2H), 7.59 (td, J=8.4, 6.4 Hz, 1H), 7.13 (ddd, J=10.4, 8.4, 1.0 Hz, 1H). General Procedure for Synthesis of 2,4-diamino-5-dichloropyrimidines: 2-((5-chloro-2-((thiazol-2ylmethyl)amino)pyrimidin-4-yl)amino)-6-fluorobenzamide

2-[(2,5-Dichloro-4-pyrimidinyl)amino]-6-fluorobenzamide (3.0559 g, 10.2 mmol) and 2aminomethylthiazole dihydrochloride (2.2395 g, 12.0 mmol) were suspended in acetonitrile (35 mL) in a 150 mL pressure tube. N,N-Diisopropylethylamine (5.20 mL, 29.8 mmol) was added, the tube was sealed and the mixture was heated to 100 °C for 4 days. The mixture was cooled to RT and the solid was collected by filtration, washed with acetonitrile and dried to provide the desired product (2.2658 g, 59%) as a tan solid. 1H NMR (d6-DMSO): δ 10.33 (s, br, 1H), 8.25–7.98 (m, 4H), 7.88 (s, br, 1H), 7.75 (s, br, 1H), 7.56 (d, J=3.2 Hz, 1H), 7.22 (s, br, 1H), 6.95 (s, br 1H), 4.83–4.54 (m, 2H). LCMS calculated for C15H12ClFN6OS: m/z= 378; found: m/z= 379 (M+H+). HPLC purity= 99%.

ASSOCIATED CONTENT Supporting Information Detailed experimental procedures for compound syntheses, Supplemental Figure S1, and Supplemental Tables S1–S5. This material is available free of charge via the Internet at http://pubs.acs.org.

AUTHOR INFORMATION Corresponding Author *E-mail: [email protected]

Notes The authors declare no competing financial interest.

ACKNOWLEDGMENTS



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The authors would like to acknowledge M. Boesche, M. Jundt, and M. Klös-Hudak for expert technical assistance with the chemoproetomics experiments.


Page 20 of 23

Page 21 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


REFERENCES 1. Kraemer, R.R., Castracane, V.D. (2015) Novel insights regarding mechanisms for treatment of sarcopenia. Metab. Clin. Exp.64, 160–162. 2. Emery, A. E. H. (2002) The muscular dystrophies. Lancet 359, 687–695. 3. For an excellent review, see: Brack, A.S., Rando, T.A (2012) Tissue-Specific Stem Cells: Lessons from the Skeletal Muscle Satellite Cell. Cell Stem Cell 10, 504–514. 4. García-Prat, L., Sousa-Victor, P., Muñoz-Cánoves, P. (2013) Functional dysregulation of stem cells during aging: a focus on skeletal muscle stem cells. FEBS Journal 280, 4051–4062. 5. Oh, J., Lee, Y. D., Wagers, A. J. (2014) Nature Medicine 20, 870–880. 6. Yin, H., Price, F., Rudnicki, M. A. (2013) Satellite cells and the muscle stem cell niche. Physiol. Rev. 93, 23–67. 7. Braun, T., Gautel, M. (2011) Transcriptional mechanisms regulating skeletal muscle differentiation, growth and homeostasis. Nature Rev. Mol. Cell Biol. 12, 349–361. 8. Conboy, I. M., Conboy, M. J., Wagers, A. J., Girma, E. R., Weissman, I. L., Rando, T. A. (2005) Rejuvenation of aged progenitor cells by exposure to a young systemic environment. Nature 433, 760–764. 9. Längle, D., Halver, J., Rathmer, B., Willems, E., Schade, D. (2014) Small Molecules Targeting in Vivo Tissue Regeneration. ACS Chemical Biology 9, 57–71. 10. Billin, A. (2012) Hematopoietic, CNS and skeletal muscle stem cells as drug targets: opportunities, progress and challenges. Future Med. Chem. 4, 615–623. 11. Nierobisz, L.S., Cheatham, B., Buehrer, B.M., Sexton, J.Z. (2013) High-Content Screening of Human Primary Muscle Satellite Cells for New Therapies for Muscular Atrophy/Dystrophy. Curr. Chem. Genom. Transl. Med. 7, 21–29. 12. Conboy, M. J., Cerletti, M., Wagers, A. J., Conboy, I. M. (2010) Immuno-analysis and FACS sorting of adult muscle fiber-associated stem/precursor cells. Meth. Mol. Biol. 621, 165– 173. 13. Carlson, M. E., Conboy, M. J., Hsu M, Barchas, L., Jeong, J., Agrawal, A., Mikels, A. J., Agrawal, S., Schaffer, D. V., Conboy, I. M. (2009) Relative roles of TGF-beta1 and Wnt in the systemic regulation and aging of satellite cell responses. Aging Cell. 8, 676–89. 14. Fornaro, M., Hinken, A. C., Needle, S., Hu E., Trendelenburg, A. U., Mayer, A., Rosenstiel, A., Chang, C., Meier, V., Billin, A. N., Becherer, J. D., Brace, A. D., Evans, W. J., Glass, D. J., Russell, A. J. (2014) Mechano-growth factor peptide, the COOH terminus of unprocessed insulin-like growth factor 1, has no apparent effect on myoblasts or primary muscle stem cells. Am J Physiol Endocrinol Metab. 306, E150–E156. 15. Lovering, R. M., Roche, J. A., Goodall, M. H., Clark, B. B., McMillan, A. (2011) An in vivo rodent model of contraction-induced injury and non-invasive monitoring of recovery. J. Vis. Exp. 51, DOI: 10.3791/2782. 16. Kramer, H. F., McDougal, A. J., Clifton, L., Russell A. M. (2012) Development of a minimallyinvasive, longitudinal eccentric muscle damage model in the mdx mouse. New Directions in Biology and Disease of Skeletal Muscle, New Orleans, LA.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

17. Yoshida, K., Taguchi, M. (1992) Reaction of N-substituted cyclic amines with 2,4dichloroquinazoline, 2,4-dichloropyrimidine, and its 5-methyl derivative. J. Chem. Soc., Perkin Trans. 1,, 919–922. 18. Obach, R. S, Kalgutkar, A. S. Ryder, T. F.; Walker, G. S. (2008) In Vitro Metabolism and Covalent Binding of Enol-Carboxamide Derivatives and Anti-Inflammatory Agents Sudoxicam and Meloxicam: Insights into the Hepatotoxicity of Sudoxicam. Chem. Res. Toxicol. 21, 1890–1899. 19. Bantscheff, M., Drewes, G. (2012) Chemoproteomic approaches to drug target identification and drug profiling. Bioorg. Med. Chem. 20, 1973–1978. 20. Kruse, U., Pallasch, C. P., Bantscheff, M., Eberhard, D., Frenzel, L., Ghidelli, S., Maier, S. K., Werner, T., Wendtner, C. M., Drewes G. (2011) Chemoproteomics-based kinome profiling and target deconvolution of clinical multi-kinase inhibitors in primary chronic lymphocytic leukemia cells. Leukemia, 25, 89–100. 21. Bantscheff, M., Eberhard, D., Abraham, Y., Bastuck, S., Boesche, M., Hobson, S., Mathieson, T., Perrin, J., Raida, M., Rau, C.; Reader, V., Sweetman, G., Bauer, A., Bouwmeester, T., Hopf, C., Kruse, U., Neubauer, G., Ramsden, N., Rick, J., Kuster, B., Drewes, G. (2007) Quantitative chemical proteomics reveals mechanisms of action of clinical ABL kinase inhibitors. Nat. Biotechnol. 25, 1035–1044. 22. Raman, M., Earnest, S., Zhang, K., Zhao, Y., Cobb, M. H. (2007) TAO Kinases mediate activation of p38 in response to DNA damage. EMBO J. 26, 2005–2014. 23. Sousa-Victor, P., Gutarra, S., Garcia-Prat, L., Rodriguez_Ubreva, J., Ortet, L., Ruiz-Bonilla, V., JArdi, M., Ballestar, E., González, S., Serrano, A. L., Perdiguero, E., Muñoz-Cánoves, P. (2014) Geriatric muscle stem cells switch reversible quiescence into senescence. Nature 506, 316–321. 24. Bernet, J. D., Doles, J. D., Hall, J. K., Kelly Tanaka, K., Carter, T. A., Olwin, B. B. (2104) p38 MAPK signaling underlies a cell-autonomous loss of stem cell self-renewal in skeletal muscle of aged mice. Nat. Med. 20, 265–271. 25. Cosgrove, B. D., Gilbert, P. M., Porpiglia, E., Mourkioti, F., Lee, S. P., Corbel, S. Y., Llewellyn, M. E., Delp, S. L., Blau, H, M. (2104) Rejuvenation of the muscle stem cell population restores strength to injured aged muscles. Nat. Med. 20, 255–264. 26. Schwarzkopf, M., Coletti, D., Sassoon, D., Marazzi, G. (2006) Muscle cachexia is regulated by a p53-PW1/Peg3-dependent pathway. Genes Dev. 20, 3440–3452.


Page 22 of 23

Cell sorting

Page 23 of 23

Muscle Regeneration Satellite Cells

Aged phenotype

Phenotypic screen

GSK649A

Fraction of Maximal Undamaged Limb Force

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42


1.1 1 0.9 0.8 0.7 0.6 0.5 0

2

4

6

8 10 12 14 16 18 20 22 Day

satellite cell proliferation In vitro


Youth phenotype

Improved recovery from Injury In vivo

Discovery of Novel Small Molecules that Activate ... - ACS Publications

Recommend Documents