CoChat
Sign Up
Cited by 359Protein Degradation and InhibitorsUbiquitin and proteasome pathwaysChemical Synthesis and Analysis

Mapping the Degradable Kinome Provides a Resource for Expedited Degrader Development

Open access
Donovan, Katherine A., Ferguson, Fleur M., Bushman, Jonathan W., Eleuteri, Nicholas A., Bhunia, Debabrata, Ryu, SeongShick, Tan, Li, Shi, Kun +26 more·2020·Cell
DOI 10.1016/j.cell.2020.10.038PMC PMC10294644
Open this paper in CoChat — free account

Abstract

Targeted protein degradation (TPD) refers to the use of small molecules to induce ubiquitin-dependent degradation of proteins. TPD is of interest in drug development, as it can address previously inaccessible targets. However, degrader discovery and optimization remains an inefficient process due to a lack of understanding of the relative importance of the key molecular events required to induce target degradation. Here, we use chemo-proteomics to annotate the degradable kinome. Our expansive dataset provides chemical leads for ∼200 kinases and demonstrates that the current practice of starting from the highest potency binder is an ineffective method for discovering active compounds. We develop multitargeted degraders to answer fundamental questions about the ubiquitin proteasome system, uncovering that kinase degradation is p97 dependent. This work will not only fuel kinase degrader discovery, but also provides a blueprint for evaluating targeted degradation across entire gene families to accelerate understanding of TPD beyond the kinome.

AI snapshot

Written from the paper’s full text.

TL;DR

This study uses chemo-proteomics to map the degradable human kinome with a library of kinase-targeting degraders, identifying chemical leads for ~200 kinases and showing that starting from the highest-potency binder is often ineffective for degrader discovery. It also finds that most kinase degradation is p97 dependent and provides a public resource for expedited degrader development.

Key findings

  • Profiling 91 degraders across 7 cell lines quantified 411 protein kinases and identified 172 degraded protein kinases, corresponding to 33% of the human kinome and 42% of the detected kinome; including kinase-related proteins, 212 total targets were identified, versus 57 previously reported degradable kinases.
  • Degradability varied widely: 136 of 172 hits were downregulated in at least two independent treatments, and 9 kinases (CDK4, AURKA, FER, WEE1, BLK, LIMK2, CDK6, GAK, LIMK1) were each degraded in at least 40 of 155 independent treatments.
  • Cellular target engagement and stable ternary complex formation did not generally predict degradation efficiency, and target protein abundance showed a U-shaped relationship with degradation frequency rather than a simple linear dependence.
  • Switching between CRBN- and VHL-recruiting degraders expanded the accessible target space by 16 additional kinases, a 23% increase, and most kinase degradation was rescued by p97 inhibition, indicating p97 dependence across CRBN, VHL, and IAP-recruiting degraders.

Methods

  • The authors synthesized and selected 91 kinase-targeting degraders incorporating diverse kinase scaffolds, linkers, and CRBN- or VHL-recruiting ligands, then profiled them by global TMT LC-MS3 proteomics in HEK293T, MOLT-4, Mino, MM1.S, OVCAR-8, KATO III, and KELLY cells, using 5 h treatment with 1 μM degrader and hit thresholds of fold change < −1.25 and P < 0.01.
  • The degradable kinome database included 155 degrader-treated samples from 50 multiplexed global proteomics experiments; additional assays included KiNativ live-cell target engagement, FLAG-CRBN affinity purification mass spectrometry, RNA sequencing, immunoblots, and CellTiter-Glo viability assays.
  • Degradability scores were calculated from the frequency of degradation across unique compound treatments, and p97 dependence was assessed by co-treatment with the p97 inhibitor CB-5083.

Limitations

  • The paper states that its approach informs TPD only in the context of degraders developed from reported kinase binders and commonly employed linkers and E3-recruiting ligands, and that it is implausible to generate all possible variants; it also notes testing in immortalized cancer cell lines and describes the database as a living resource.
  • The paper states that some detected but not degraded kinases may be tractable under different experimental conditions, citing CK2 degraders active only at 24 h, and that the short 5 h screening time may miss slow degradation kinetics.
  • The paper states limitations of the CRBN AP-MS experiment, including potential loss of transient or weakly bound complexes during enrichment and washing, inherent noise relative to global proteomics, and the short time point precluding detection of slow degradation events.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1016/j.cell.2020.10.038

Cites

Statistics

Nothing to recompute.

Claims the paper makes

0 of 39 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    Deep proteome coverage permitted quantification of 411 protein kinases across 7 cell lines: HEK293T, MOLT-4, Mino, MM1.S, OVCAR-8, KATO III and KELLY cells.

    “Deep proteome coverage permitted quantification of 411 protein kinases across 7 cell lines: HEK293T, MOLT-4, Mino, MM1.S, OVCAR-8, KATO III and KELLY cells”
    Quote in the paperSubstantial claim
  • Findingown data

    Nine kinases (CDK4, AURKA, FER, WEE1, BLK, LIMK2, CDK6, GAK, LIMK1) were each degraded in at least 40 of the 155 independent treatments.

    “9 of these (CDK4, AURKA, FER, WEE1, BLK, LIMK2, CDK6, GAK, LIMK1) were each degraded in at least 40 of the 155 independent treatments”
    Quote in the paperSubstantial claim
  • Findingown data

    CDK17 is degraded by 15 different compounds.

    “For example, cyclin-dependent kinase 17 (CDK17) is degraded by 15 different compounds.”
    Quote in the paperNot a substantial claim
  • Findingown data

    Leads were identified for putative cancer targets PI3K-γ, PIP5K1A, PIP4K2B and PIP4K2C.

    “we examined their degradability and identified leads for putative cancer targets PI3K-γ, PIP5K1A, PIP4K2B and PIP4K2C”
    Quote in the paperSubstantial claim
  • Findingown data

    Fifty-two of the 57 kinases with at least one active degrader reported were also identified as degradable in this study (>90%).

    “Here, we found that 52 of the 57 kinases with at least one active degrader reported were also identified as degradable in this study (> 90%, Figure 1F).”
    Quote in the paperSubstantial claim
  • May combine two claims. 45
  • Classified as probably not a substantial claim. 19

  • May combine two claims.

    Findingown data

    The negative data contained within the dataset illuminates the kinases that are not yet 'degradable' and reveals the chemical structures that are not active towards a particular kinase target.

    “Equally important is the negative data contained within the dataset which illuminates the kinases that are not yet ‘degradable’ and reveals the chemical structures that are not active towards a particular kinase target”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Out of 42 pseudokinases quantified, 10 were degradable by at least one compound in the set, including IRAK3 and TRIB3.

    “Out of 42 pseudokinases quantified, 10 were degradable by at least one compound in our set, including well characterized pseudokinases IRAK3 and TRIB3”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    DD-03-156 induces potent and selective degradation of CDK17 and LIMK2.

    “Here, we identified at least one lead-like degrader molecule, DD-03–156, which induces potent and selective degradation of CDK17 and LIMK2”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Transcript levels were largely unchanged up until the 4 h time point in response to SK-3-91, indicating that the hits in the database are unlikely to be transcriptionally downregulated.

    “found that transcript levels were largely unchanged up until the 4 h time point (Figure S1B), indicating that the hits in our database are unlikely to be transcriptionally downregulated.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Despite an absence of prior reports that CSK is a degradable kinase, 15 compounds in the library were able to target CSK.

    “Despite an absence of prior reports that CSK is a degradable kinase, 15 compounds in our library were able to target CSK.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The dataset identifies kinases that may be refractory to degradation using currently available TPD technologies and characterizes their binary target engagement, ternary complex formation and expression profiles.

    “Additionally, our dataset identifies kinases that may be refractory to degradation using currently available TPD technologies, and characterizes their binary target engagement, ternary complex formation and expression profiles”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Forty proteins from the extended human kinome list were identified as degradable, validating them as pharmacologically related to protein kinases and tractable TPD targets.

    “We detected 173 of these proteins in at least one experiment, and identified degraders for 40 proteins from this list (Table S4; Figure S1A), validating them as pharmacologically related to protein kinases, and tractable TPD targets.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Using significance cutoffs of FC < −1.25 and P-value < 0.01, 172 degraded protein-kinases were identified, corresponding to 33% of the human kinome and 42% of the detected kinome.

    “Using significance cutoffs of FC < −1.25, P-value < 0.01, identified 172 degraded protein-kinases, corresponding to 33% of the human kinome, and 42% of the detected kinome”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The parental inhibitors corresponding to degraders profiled in this study are able to engage 370 of the 395 unique kinases present in the DiscovRX kinomeSCAN panel (93%), corresponding to at least 70% coverage of the human kinome.

    “Based on reported and in-house biochemical data, the parental inhibitors corresponding to degraders profiled in this study are able to engage 370 of the 395 unique kinases present in the DiscovRX kinomeSCAN panel (93%), corresponding to at least 70% coverage of the human kinome”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    GCK is bound and degraded by DB0646 (58 %I) and SB1-G-187 (94 %I), but not degraded by SK-3-91 (50 %I).

    “For example, GCK is bound and degraded by DB0646 (58 %I) and SB1-G-187 (94 %I), but not degraded by SK-3–91 (50 %I).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingcitation

    DB-3-291 incorporates dasatinib as the kinase binding ligand, which had a 1 nM in vitro binding affinity to CSK (KINOMEscan), but CSK was ranked 40th of over 100 kinases with sub μM binding affinity.

    “The DB-3-291 degrader incorporates the multitargeted inhibitor dasatinib as the kinase binding ligand, which was found to have a 1 nM in vitro binding affinity to CSK (KINOMEscan), however CSK was ranked 40th of over 100 kinases that had sub μM binding affinity (KD) (Davis et al., 2011).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Davis et al., 2011 source

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    Proteins identified as complexed with CRBN were enriched for kinases as well as their known binding partners such as Cyclin B (CDK1) and RASSF1 (STK4), consistent with the binding profiles of the assayed degraders.

    “The proteins identified as complexed with CRBN were enriched for kinases as well as their known binding partners such as Cyclin B (CDK1) and RASSF1 (STK4), consistent with the binding profiles of the assayed degraders.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Incorporation of alisertib as the target binder resulted in potent and selective degradation by dAURK-4, validated by immunoblotting.

    “Incorporation of an AURKA selective inhibitor, alisertib, as the target binder resulted in potent and selective degradation by dAURK-4, validated by immunoblotting”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    BUB1 was complexed but not degraded by all 4 degraders in HEK293T experiments, but was degraded in 24 independent treatments across the database in MOLT-4 and MM.1S cell lines, including by DB0646 and SK-3-91.

    “Finally, we observed that BUB1 was complexed but not degraded by all 4 degraders in HEK293T experiments, but was degraded in 24 independent treatments across the database in MOLT-4 and MM.1S cell lines, including by DB0646 and SK-3–91”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Transcriptional analysis showed no upregulation of CAMKK2, DNAPK, IKKe, and JAK2 in response to a 4 h treatment with SK-3-91, ruling out compensatory upregulation as a cause of the lack of observed degradation.

    “The resulting transcriptional analysis showed no upregulation of these kinases (Figure S2F), ruling out compensatory upregulation (Figure S2C; Figure S2B), as causes of a lack of observed degradation.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The number of degraded kinase targets per compound varies by cell line, with the largest number in MOLT-4, followed by KELLY and HEK293T.

    “We observed differences in the number of degraded kinase targets of each molecule dependent on the cell line. In all cases the largest number of protein kinases per compound were degraded in MOLT-4 cells, followed by KELLY and HEK293T (Figure S3C).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    IRAK1 and CDK17 were engaged by multiple degraders to a similar extent but were degraded by only one of these molecules.

    “We also observed examples of kinases that were engaged by multiple degraders to a similar extent, but were degraded by only one of these molecules, such as IRAK1 and CDK17”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    All kinases identified by AP-MS except CSNK1A1 were also detected in the global degradation profiling, allowing interrogation across 52 unique kinases.

    “All kinases identified by AP-MS except CSNK1A1 were also detected in the global degradation profiling, allowing us to interrogate trends across 52 unique kinases”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Compound DB-3-291 induced the strongest degradation of CSK and had the greatest selectivity.

    “Compound DB-3-291 was found to induce the strongest degradation, in addition to having the greatest selectivity”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The proportion of degraded kinases with detectable binding varied dramatically between compounds and was unrelated to cellular permeability.

    “Furthermore, the proportion of degraded kinases with detectable binding varied dramatically between compounds, and was unrelated to cellular permeability”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    For YES1, IRAK1 and LYN, complex formation and degradation are sometimes detected together, but complex formation does not predict degradation.

    “For YES1, IRAK1 and LYN, complex formation and degradation are sometimes detected together (YES1: DB0646, SB1-G-187. IRAK1, LYN: SB1-G-187), but complex formation does not predict degradation (YES1: SK-3–91, IRAK1: WH-10417–099).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    A subset of kinases had strong linker preferences, ranging from preference for a specific molecule (CSK, CDK9), preference for short linker (ABL2, CDK4, CDK5, CDK12 and LIMK2), or specific linker-attachment regioselectivity (CDK7, AAK1, BLK).

    “In our data, a subset of kinases had strong linker preferences, ranging from preference for a specific molecule (CSK, CDK9), preference for short linker (ABL2, CDK4, CDK5, CDK12 and LIMK2), or specific linker-attachment regioselectivity (CDK7, AAK1, BLK).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In the absence of degrader, p97 inhibition does not cause global blocking of kinase degradation; expression levels of top kinase hits were stable over 6 h.

    “Analysis of protein expression of the top kinase hits from each experiment revealed that expression levels were stable over 6 h, confirming that in absence of degrader, p97 inhibition does not cause global blocking of kinase degradation (Figure 7B).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    NEK9 is degradable by both CRBN and VHL ligases but clearly favors CRBN.

    “Such as NEK9, which is degradable by both ligases but clearly favors CRBN (Figure 5A; Figure S4B, C).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Many targets can be degraded with both CRBN and VHL targeting degraders, but a significant number show clear preferential compatibility with one over the other.

    “While many targets can be degraded with both CRBN and VHL targeting degraders, a significant number show clear preferential compatibility with one over the other.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    ZNF653 and IKZF1 clearly favor ortho-linked degraders, whereas RNF166 favors meta-linked degraders.

    “revealing that ZNF653 and IKZF1 clearly favor ortho-linked degraders, whereas RNF166 favors meta-linked degraders (Figure 6D, Table S10).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    A number of kinases showed the same ligase preference across more than one pair, and this preference held true across the entire degradable kinome database; for example, MAP3K11 and SRC were degraded exclusively by CRBN-recruiting degraders.

    “We found that a number of kinases showed the same ligase preference across more than one pair, and this preference held true across the entire degradable kinome database, for example MAP3K11 and SRC, which despite being targeted by many different compounds (8 and 14 respectively), were found to be degraded exclusively by CRBN-recruiting degraders (Table S9).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingnone given

    Differences in target profile of compounds were observed when tested in MOLT-4, KELLY, or HEK293T cells, and target expression levels were ruled out as the determining factors driving these differences.

    “We observed differences in the target profile of compounds when tested in MOLT-4, KELLY or HEK293T cells, and ruled out target expression levels as the determining factors driving these differences.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Side-by-side comparison of matched pairs of CRBN- and VHL-recruiting degraders revealed only minor differences in intracellular E3 ligase engagement, except for desmethoxy-TAE684-based degraders where the CRBN-based degrader was significantly more cell permeable.

    “Side-by-side comparison of each of the matched pairs of degrader molecules revealed only minor differences (Figure S4A-C), with the exception of the desmethoxy-TAE684 based degraders where the CRBN-based degrader was significantly more cell permeable (Figure S4C).”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    The degradable kinome database included 155 degrader treated samples.

    “We included 155 degrader treated samples in this database”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Active degrader molecules were found for more than 16 understudied kinases, including two potent and selective degraders for CDK17.

    “Strikingly, we found active degrader molecules for more than 16 understudied kinases including two potent and selective degraders for CDK17.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    All 34 degraders containing an aryl amine exit vector were able to induce degradation of at least one, but in most cases several, IMiD targets.

    “Consistent with previous reports, we found that all 34 degraders containing an aryl amine exit vector, were able to induce degradation of at least one, but in most cases several, IMiD targets.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Over 60 proteins of interest from diverse protein families and structural classes have been successfully degraded using a TPD strategy.

    “Over 60 POIs from diverse protein families and structural classes have been successfully degraded using a TPD strategy.”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites Lai et al., 2016 source

      Not checked in the free run.

    • cites Lu et al., 2015

      Not checked in the free run.

    • cites Wu et al., 2019

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    TPD may expand the ability to target the remaining 80% of the human proteome previously considered undruggable by utilizing phenotypically silent ligands as POI-recruiting handles.

    “Finally TPD may expand our ability to target the remaining 80% of the human proteome that has previously been considered undruggable (Russ and Lampel, 2005), due to the ability to utilize phenotypically silent ligands as POI-recruiting handles (Silva et al., 2019).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Russ and Lampel, 2005

      Not checked in the free run.

    • cites Silva et al., 2019 source

      Not checked in the free run.

  • May combine two claims.

    Interpretationown data

    The top degradable kinases from the corrected analysis mirror those from the frequency analysis (CDK4, AURKA, FER, WEE1, BLK), confirming that in sufficiently large datasets frequency of degradation is a good measure of general tractability.

    “Unsurprisingly, the top degradable kinases mirror those from the previous analysis (CDK4, AURKA, FER, WEE1, BLK), confirming that in sufficiently large datasets, even with over-representation of certain molecules, frequency of degradation is a good measure of general tractability.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The majority of degraders recruit the CRL4crbn or CRL2vhl E3 ligases to ubiquitinate a protein of interest.

    “The majority of degraders recruit the CUL4-RBX1-DDB1-CRBN (CRL4crbn) (Ito et al., 2010; Kronke et al., 2014; Winter et al., 2015), or CUL2-RBX1-ElonginB-ElonginC-VHL (CRL2vhl) (Buckley et al., 2012; Galdeano et al., 2014) E3 ligases to ubiquitinate a POI.”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites Ito et al., 2010 source

      Not checked in the free run.

    • cites Buckley et al., 2012 source

      Not checked in the free run.

    • cites Winter et al., 2015 source

      Not checked in the free run.

  • May combine two claims.

    Interpretationreasoning

    Outliers such as CK2 represent limitations of the study and indicate that some detected but not degraded kinases may indeed be tractable under different experimental conditions.

    “Outliers such as CK2 represent limitations of the study, and indicate that some detected but not degraded kinases may indeed be tractable under different experimental conditions.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    A single degrader molecule can degrade multiple protein molecules, a mechanism referred to as event-driven pharmacology, which may increase efficacy.

    “A key promise of TPD is the prospect of increased efficacy due to a single degrader molecule being able to degrade multiple protein molecules, this unique mechanism of action is referred to as event-driven pharmacology (Bondeson et al., 2015).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Bondeson et al., 2015 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    The human protein kinase super family consists of 514 protein kinases, making up 2.5% of the total human genome.

    “The human protein kinase super family consists of 514 protein kinases (Manning et al., 2002), which makes up 2.5% of the total human genome.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Manning et al., 2002 source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Premisecitation

    TPD uses small molecules (degraders) to hijack the cellular degradation machinery by recruiting E3 ubiquitin ligases to proteins of interest, which would not otherwise be recognized as substrates.

    “TPD uses small molecules (degraders) to hijack the cellular degradation machinery by recruiting E3 ubiquitin ligases to proteins of interest (POI), which would not otherwise be recognized as substrates (Sakamoto et al., 2001)”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Sakamoto et al., 2001 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Interpretationown data

    Altered degradation kinetics or other cell-type related variables could be drivers of the discrepancy for BUB1.

    “Here, altered degradation kinetics or other cell-type related variables could be drivers of this discrepancy.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationreasoning

    It is surprising that DB-3-291 does not degrade additional kinases.

    “Thus, it is surprising that this molecule does not degrade additional kinases.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    DB-3-291 provides an advanced lead for further development into a chemical probe and demonstrates the utility of this resource for providing lead molecules for indications beyond oncology.

    “DB-3-291 not only provides an advanced lead for further development into a chemical probe, but also demonstrates the utility of this resource for providing lead molecules for indications beyond oncology.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S1 (asserted).

    Premisecitation

    It is widely believed that degraders uncouple efficacy from target occupancy.

    “It is widely believed that degraders uncouple efficacy from target occupancy (Gadd et al., 2017; Nowak et al., 2018; Olson et al., 2018; Roy et al., 2019).”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites Gadd et al., 2017 source

      Not checked in the free run.

    • cites Nowak et al., 2018 source

      Not checked in the free run.

    • cites Olson et al., 2018 source

      Not checked in the free run.

  • May combine two claims.

    Interpretationown data

    Kinases have high compatibility for degrader induced binding with CRBN, but different complexes differ in their ability to efficiently catalyze degradation.

    “These data indicate that kinases have high compatibility for degrader induced binding with CRBN, but that different complexes differ in their ability to efficiently catalyze degradation.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductcitation

    Limitations of the AP-MS experiment include the potential loss of transient or weakly bound complexes, the inherent noise associated with AP-MS relative to global proteomics analysis, and the short time point of the experiment, which precludes detection of degradation events with slow kinetics.

    “Limitations of this experiment include the potential loss of transient or weakly bound complexes in the enrichment and subsequent wash steps, the inherent noise associated with AP-MS relative to global proteomics analysis (Dunham et al., 2012; Yugandhar et al., 2019) and the short time point of the experiment, which precludes detection of degradation events with slow kinetics.”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Dunham et al., 2012 source

      Not checked in the free run.

    • cites Yugandhar et al., 2019 source

      Not checked in the free run.

  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation

    Multiple studies have reported that the stability of the E3-degrader-POI ternary complex may influence degradation kinetics and selectivity, and may be a more reliable predictor for degradation than target engagement alone.

    “Multiple studies have reported that the stability of the E3-degrader-POI ternary complex may influence degradation kinetics and selectivity, and may be a more reliable predictor for degradation than target engagement alone (Bondeson et al., 2018; Roy et al., 2019).”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Bondeson et al., 2018 source

      Not checked in the free run.

    • cites Roy et al., 2019 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    Proteome Discoverer was used for controlling peptide and protein level false discovery rates.

    “Proteome Discoverer 2.1, 2.2 or 2.4 (Thermo Fisher Scientific) was used for .RAW file processing and controlling peptide and protein level false discovery rates, assembling proteins from peptides, and protein quantification from peptides.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    MS/MS spectra were searched against a Uniprot human database with both forward and reverse sequences as well as known contaminants.

    “MS/MS spectra were searched against a Uniprot human database (September 2016 or December 2019) with both the forward and reverse sequences as well as known contaminants such as human keratins.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductnone given

    A protein was considered a hit if it met a pre-determined threshold of P-value < 0.01 and fold change > 1.25.

    “A protein was considered a ‘hit’ if it met our pre-determined ‘hit’ threshold of P-value < 0.01 and fold change > 1.25.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductnone given

    Peptide spectral matches with poor quality MS3 spectra were excluded from quantitation using thresholds of summed signal-to-noise across channels < 100 for whole proteome or < 50 for affinity purification and precursor isolation specificity < 0.5, and proteins were required to have at least 2 unique peptides quantified.

    “Peptide spectral matches with poor quality MS3 spectra were excluded from quantitation (summed signal-to-noise across channels < 100 (Whole Proteome) or < 50 (Affinity Purification) and precursor isolation specificity < 0.5), and resulting data was filtered to only include proteins that had a minimum of 2 unique peptides quantified.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductreasoning

    The degradable kinome database includes only degrader treatments judged to best represent direct degrader-induced target degradation, excluding treatments longer than 8 h and treatments that induced transcriptional collapse.

    “In this database we only include degrader treatments that we believe best represents direct degrader-induced target degradation (excluded treatments of > 8 h and treatments that induced transcriptional collapse to reduce misinterpretation of cellular stress and secondary effects as direct targets).”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Treatments longer than 8 h and treatments that induce transcriptional collapse are not representative of direct degrader-induced target degradation and would cause misinterpretation of cellular stress and secondary effects as direct targets.

  • Classified as probably not a substantial claim.

    Premisecitation

    AAA+ ATPase p97 unfoldase activity is necessary for extracting a subset of proteins marked for degradation from multi-protein complexes, chromatin, or membrane bound complexes.

    “AAA+ ATPase p97 unfoldase activity has been demonstrated to be necessary for extracting a subset of proteins marked for degradation from multi-protein complexes, chromatin, or membrane bound complexes (Ramadan et al., 2007; Shcherbik and Haines, 2007; Verma et al., 2011).”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites Ramadan et al., 2007 source

      Not checked in the free run.

    • cites Shcherbik and Haines, 2007

      Not checked in the free run.

    • cites Verma et al., 2011 source

      Not checked in the free run.

  • May combine two claims.

    Interpretationown data

    The subset of kinases with comparable degradation across all 6 compounds is highly tolerant of linker alterations and may have the ability to adopt multiple productive complex conformations with CRBN.

    “We found that of the 26 kinases degraded, a subset showed comparable degradation across all 6 compounds suggesting that they are highly tolerant of linker alterations (Figure 6C; Table S3-4) and indicating that these specific kinases may have the ability to adopt multiple productive complex conformations with CRBN.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    Each independent multiplexed experiment contained three biological replicates of DMSO control treatments.

    “Each independent multiplexed experiment contained three biological replicates of DMSO control treatments and the number of biological replicates for each sample treatment is listed in Table S1.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    The majority of the degradable kinome is processed in a p97-dependent fashion, and this dependence occurs irrespective of the E3 ligase recruited (CRBN, VHL, and IAP).

    “Our observations suggest that the majority of the degradable kinome is processed in a p97-dependent fashion, and that this dependence occurs irrespective of the E3 ligase recruited (CRBN, VHL and IAP).”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Previous studies have shown that a nitrogen to oxygen modification on the thalidomide aryl attachment point can reduce or remove IMiD off-target effects in BTK and CDK4/6 targeted degraders.

    “Previous studies have shown that a nitrogen to oxygen modification on the thalidomide aryl attachment point can reduce or remove IMiD off-target effects in BTK and CDK4/6 targeted degraders (Dobrovolsky et al., 2019; Jiang et al., 2019).”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Dobrovolsky et al., 2019 source

      Not checked in the free run.

    • cites Jiang et al., 2019 source

      Not checked in the free run.

  • May combine two claims.

    Interpretationreasoning

    Cellular target engagement is not a good predictor of degrader efficacy, suggesting a catalytic mechanism uncoupled from primary affinities.

    “For example, while we can generally conclude that cellular target engagement is not a good predictor of degrader efficacy, suggesting a catalytic mechanism uncoupled from primary affinities, we also discovered kinases where an affinity threshold must be met for degradation, such as GCK.”
    Quote in the paperSubstantial claim
58 of 58

  • The paper relies on this without support.

    Degrader discovery and optimization remain an inefficient process due to a lack of understanding of the relative importance of the key molecular events required to induce target degradation.

    “However, degrader discovery and optimization remain an inefficient process due to a lack of understanding of the relative importance of the key molecular events required to induce target degradation.”

  • The paper relies on this without support.

    It remains difficult to predict which proteins are tractable and which may prove recalcitrant to TPD.

    “Despite the widespread exploration of TPD as a novel pharmacological modality, it remains difficult to predict which proteins are tractable and which may prove recalcitrant to this approach.”

  • The paper relies on this without support.

    The probability of identifying the same kinase as a false positive in multiple treatments is low.

    “We rationalized that the probability of identifying the same kinase as a false positive in multiple treatments is low”

  • The paper relies on this without support.

    Treatments longer than 8 h and treatments that induce transcriptional collapse are not representative of direct degrader-induced target degradation and would cause misinterpretation of cellular stress and secondary effects as direct targets.

    “In this database we only include degrader treatments that we believe best represents direct degrader-induced target degradation (excluded treatments of > 8 h and treatments that induced transcriptional collapse to reduce misinterpretation of cellular stress and secondary effects as direct targets).”
4 of 4

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Several mechanisms contribute to increased selectivity of PROTACs, including steric incapability of some targets to form a productive ternary complex, or lack of a suitably positioned lysine residue for ubiquitination.

    “Several mechanisms have been shown to contribute to this increase in selectivity. This includes steric incapability of some targets to form a productive ternary complex, or lack of a suitably positioned lysine residue for ubiquitination.62−69”

    From Advancing Targeted Protein Degradation via Multiomics Profiling and Artificial Intelligence (2023) · cites this paper as [69]

  • A recent massive kinase degradation study revealed the degradable kinome and a diversity of kinase-degrading profiles among E3 ligases.

    “a recent massive kinase degradation study using various kinase inhibitors as warheads has revealed the ‘degradable kinome’ and a diversity of kinase-degrading profiles among E3 ligases”

    From PROTAC targeted protein degraders: the past is prologue (2022) · cites this paper as [126]

  • PROTACs can comparatively achieve greater specificity with respect to off-target effects of some kinase inhibitors.

    “but given that PROTACs can comparatively achieve greater specificity with respect to off-target effects of some kinase inhibitors126,181”

    From PROTAC targeted protein degraders: the past is prologue (2022) · cites this paper as [126]

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Results describe a 12-hour RNA-sequencing time point absent from the methods

    Minor

    The results report transcriptional collapse at 8 and 12 hours, but the RNA-sequencing method lists treatments lasting only 1, 2, 4, or 8 hours.

  • RNA-sequencing design requires 32 flasks, not 24

    Minor

    Four time points, two treatments, and four replicates per treatment and time point require 32 flasks. The method says 24 were seeded.

    Reported 24 · recomputed 32

2 of 2

The citation lists haven’t been fetched yet

References 90 · Cited by 359

Authors
Donovan, K. A., Ferguson, F. M., Bushman, J. W., Eleuteri, N. A., Bhunia, D., Ryu, S., Tan, L., Shi, K., Yue, H., Liu, X., Dobrovolsky, D., Jiang, B., Wang, J., Hao, M., You, I., Teng, M., Liang, Y., Hatcher, J., Li, Z., Manz, T. D., Groendyke, B., Hu, W., Nam, Y., Sengupta, S., Cho, H., Shin, I., Agius, M. P., Ghobrial, I. M., Ma, M. W., Che, J., Buhrlage, S. J., Sim, T., Gray, N. S., & Fischer, E. S.
Affiliations
  • Harvard University (US)
  • Dana-Farber Cancer Institute (US)
  • Korea Institute of Science and Technology (KR)
  • Korea University (KR)
  • Chinese Academy of Sciences (CN)
  • Yonsei University (KR)
Corresponding author
  • Sim, Taebo
  • Gray, Nathanael S.
  • Fischer, Eric S.
ORCID
Venue
Cell
Publisher
Elsevier BV
Published
2020-12
Volume
183
Issue
6
Pages
1714-1731.e10
ISSN
0092-8674 (ISSN-L)
DOI
10.1016/j.cell.2020.10.038
PMC
PMC10294644
OpenAlex
W3107242996
Cited by
359
References
90
Citations by year
2026: 54 · 2025: 55 · 2024: 60 · 2023: 47 · 2022: 67 · 2021: 58 · 2020: 17
FWCI
15.82
Type
journal article
Language
en
Open access
Licence
Locations
  • Cell (publishedVersion)
  • PubMed (publishedVersion)
  • PubMed Central (acceptedVersion, other-oa)
  • YUHSpace (Yonsei University Medical Library) (submittedVersion, cc-by-nc-nd)
Funders
  • National Natural Science Foundation of China
  • Natural Science Foundation of Shanghai
  • Korea Institute of Science and Technology
  • Damon Runyon Cancer Research Foundation
  • National Research Foundation of Korea
  • National Cancer Institute
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Protein Degradation and Inhibitors
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Ubiquitin and proteasome pathways
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Chemical Synthesis and Analysis
Keywords
targeted protein degradation, ubiquitin-dependent protein degradation, VCP, ubiquitin-proteasome system
MeSH terms
Adult, Cell Line, Female, Humans, Male, Middle Aged, Protein Kinases, RNA, Messenger, Proteome, Databases, Protein, Proteomics, Ubiquitin-Protein Ligases, Proteasome Endopeptidase Complex, Young Adult, Proteolysis

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-10.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-10.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review