Elizabeth A. Komives
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Elizabeth A. Komives has a total of 331 co-authors. Here are the top co-authors by number of shared publications:
Steve Silletti
Department of Chemistry & Biochemistry
8 shared publications- Wei Chen
The Pennsylvania State University
8 shared publications Simon Bergqvist
Department of Chemistry and Biochemistry, University of California, San Diego, 9500 Gilman Drive, La Jolla, California 92092-0378, United States
7 shared publications- Cesar A. Ramirez-Sarmiento
Pontificia Universidad Católica de Chile
6 shared publications - Hannah E. R. Baughman
Department of Biochemistry and Molecular Biophysics, University of California, San Diego, 9500 Gilman Drive, La Jolla, CA 92092-0359
6 shared publications Carla F. Cervantes
Department of Chemistry and Biochemistry, University of California San Diego, 9500 Gilman Drive, La Jolla, CA 92037-0378, USA
5 shared publications
Latest Publications
- Open Access
Optimization of the NFκB-DNA ensemble leads to high affinity binding of partial consensus sites
Our ability to respond to a myriad of external stressors is dependent on the activity of specific transcription factors that precisely regulate gene expression. While some transcription factors have high specificity for their consensus DNA binding site, others such as NFκB (RelA/p50) readily bind to partial and nonconsensus sites. In fact, ChIP-Seq data shows that nearly half of the DNA-bound RelA occupies sequences that don’t match the consensus. Here we explore how NFκB (RelA/p50) binds to alternative sequences that differ subtly from the consensus. The two subunits of the NFκB heterodimer, RelA and p50, form an extensive pocket where DNA binding is mediated by both base-specific contacts and nonspecific interactions with the DNA backbone. RelA and p50 each form sequence-specific contacts with DNA at different 4-nucleotide sequences separated by two nucleotides, together forming a 10-base pair consensus binding sequence. Partial consensus sequences, known as “half-sites”, match the consensus sequence at only one of these two 4-nucleotide interaction sites. Here, using hydrogen deuterium exchange mass spectrometry (HDX-MS) and molecular dynamics (MD) simulations to investigate protein structural dynamics, we observe that rearranging nucleotides at the RelA-interacting end of a consensus DNA sequence to form a half-site induces dynamic structural changes through both subunits of RelA/p50. MD simulations reveal a shift in the position of DNA within the RelA/p50 binding pocket that allows optimal interaction with the half-site sequence. Our results demonstrate how RelA/p50 fine-tunes its interactions to achieve high-affinity binding with multiple different target DNA sequences.
Open AccessOptimization of the NFκB-DNA ensemble leads to high affinity binding of partial consensus sites
Our ability to respond to a myriad of external stressors is dependent on the activity of specific transcription factors that precisely regulate gene expression. While some transcription factors have high specificity for their consensus DNA binding site, others such as NFκB (RelA/p50) readily bind to partial and nonconsensus sites. In fact, ChIP-Seq data shows that nearly half of the DNA-bound RelA occupies sequences that don’t match the consensus. Here we explore how NFκB (RelA/p50) binds to alternative sequences that differ subtly from the consensus. The two subunits of the NFκB heterodimer, RelA and p50, form an extensive pocket where DNA binding is mediated by both base-specific contacts and nonspecific interactions with the DNA backbone. RelA and p50 each form sequence-specific contacts with DNA at different 4-nucleotide sequences separated by two nucleotides, together forming a 10-base pair consensus binding sequence. Partial consensus sequences, known as “half-sites”, match the consensus sequence at only one of these two 4-nucleotide interaction sites. Here, using hydrogen deuterium exchange mass spectrometry (HDX-MS) and molecular dynamics (MD) simulations to investigate protein structural dynamics, we observe that rearranging nucleotides at the RelA-interacting end of a consensus DNA sequence to form a half-site induces dynamic structural changes through both subunits of RelA/p50. MD simulations reveal a shift in the position of DNA within the RelA/p50 binding pocket that allows optimal interaction with the half-site sequence. Our results demonstrate how RelA/p50 fine-tunes its interactions to achieve high-affinity binding with multiple different target DNA sequences.
- Open Access
- 28 October 2026
- Published by Elsevier in Current Opinion in Structural Biology
- Vol. 102
The relationship between local frustration and protein function
- 28 October 2026
- Published by Elsevier in Current Opinion in Structural Biology
- Vol. 102
Open AccessThe relationship between local frustration and protein function
- Open Access
- 27 November 2025
- Published by Elsevier in Journal of Molecular Biology
- Vol. 438 (2) , 169566
Synthetic Type III-E CRISPR-Cas Effectors for Programmable RNA-targeting
- 27 November 2025
- Published by Elsevier in Journal of Molecular Biology
- Vol. 438 (2) , 169566
Open AccessSynthetic Type III-E CRISPR-Cas Effectors for Programmable RNA-targeting
- Open Access
- 20 November 2025
- Published by Elsevier in Molecular & Cellular Proteomics
- Vol. 25 (2) , 101471
The Mechanism of Histone Ubiquitylation by the ASB9-CUL5 Ubiquitin Ligase
- 20 November 2025
- Published by Elsevier in Molecular & Cellular Proteomics
- Vol. 25 (2) , 101471
Open AccessThe Mechanism of Histone Ubiquitylation by the ASB9-CUL5 Ubiquitin Ligase
- 28 August 2025
- Published by American Chemical Society (ACS) in Biochemistry
- Vol. 64 (18) , 3935-3955
A Structurally Divergent Class Ia Ribonucleotide Reductase from a Tick-Borne Pathogen
Ribonucleotide reductases (RNRs) generate 2'-deoxynucleotides for DNA biosynthesis, a reaction essential to all life. Class I RNRs have two subunits, α and β. α binds and reduces the substrate, whereas β oxidizes one of the cysteines in α to a C3'-H-bond-cleaving thiyl radical to begin the reaction. The α-Cys oxidant in β is variously a tyrosyl radical (Y•) generated by a diiron or dimanganese cluster, a high-valent dimetal cluster [Mn(IV)/Fe(III) or Mn2(IV/III)], or a dihydroxylphenylalanine (DOPA) radical that operates without need of a transition metal. The metal (in)dependence of the Cys oxidant in β correlates loosely with sequence-similarity groupings. We show here that Francisella hispaniensis (Fh) β, which lies within an uncharacterized sequence cluster that contains orthologs from multiple human pathogens, harbors a Fe2(III/III)/Y• cofactor, as in class Ia RNRs from eukaryotes and Escherichia coli. Fh β has several unusual structural features that may reflect adaptation to the bacterium's environment(s). In its apo form, an unwound helix everts a metal ligand toward solvent, and the radical-harboring Y points away from the diiron cluster. An additional aromatic residue (W194), conserved within the sequence cluster, is found close to the universally conserved W37, which is thought to mediate α-Cys oxidation in all class I enzymes. The Y• in resting β is remarkably resistant to reduction by hydroxyurea but becomes 8000 times more sensitive when β is engaged in turnover with α. These structural and functional distinctions could be counter measures against host redox defenses that would target the pathogen's RNR and its cofactor.
- 28 August 2025
- Published by American Chemical Society (ACS) in Biochemistry
- Vol. 64 (18) , 3935-3955
A Structurally Divergent Class Ia Ribonucleotide Reductase from a Tick-Borne Pathogen
Ribonucleotide reductases (RNRs) generate 2'-deoxynucleotides for DNA biosynthesis, a reaction essential to all life. Class I RNRs have two subunits, α and β. α binds and reduces the substrate, whereas β oxidizes one of the cysteines in α to a C3'-H-bond-cleaving thiyl radical to begin the reaction. The α-Cys oxidant in β is variously a tyrosyl radical (Y•) generated by a diiron or dimanganese cluster, a high-valent dimetal cluster [Mn(IV)/Fe(III) or Mn2(IV/III)], or a dihydroxylphenylalanine (DOPA) radical that operates without need of a transition metal. The metal (in)dependence of the Cys oxidant in β correlates loosely with sequence-similarity groupings. We show here that Francisella hispaniensis (Fh) β, which lies within an uncharacterized sequence cluster that contains orthologs from multiple human pathogens, harbors a Fe2(III/III)/Y• cofactor, as in class Ia RNRs from eukaryotes and Escherichia coli. Fh β has several unusual structural features that may reflect adaptation to the bacterium's environment(s). In its apo form, an unwound helix everts a metal ligand toward solvent, and the radical-harboring Y points away from the diiron cluster. An additional aromatic residue (W194), conserved within the sequence cluster, is found close to the universally conserved W37, which is thought to mediate α-Cys oxidation in all class I enzymes. The Y• in resting β is remarkably resistant to reduction by hydroxyurea but becomes 8000 times more sensitive when β is engaged in turnover with α. These structural and functional distinctions could be counter measures against host redox defenses that would target the pathogen's RNR and its cofactor.
- Open Access
An allosteric network governs Tom70 conformational dynamics to coordinate mitochondrial protein import
Tom70 mediates mitochondrial protein import by coordinating the transfer of cytosolic preproteins from Hsp70/Hsp90 to the translocase of the outer membrane (TOM) complex. In humans, the cytosolic domain of Tom70 ( Hs Tom70c) is entirely α -helical and comprises modular TPR motifs divided into an N-terminal chaperone-binding domain (NTD) and a C-terminal preprotein-binding domain (CTD). However, the mechanisms linking these functional regions remain poorly understood. Here, we present the 2.04 Å crystal structure of unliganded Hs Tom70c, revealing two distinct conformations – open and closed – within the asymmetric unit. These states are stabilized in part by interdomain crystal contacts and are supported in solution by hydrogen–deuterium exchange mass spectrometry (HDX-MS) and molecular dynamics (MD) simulations. Principal component and dynamical network analyses reveal a continuum of motion linking the NTD and CTD via key structural elements, notably residues in helices α 7, α 8, and α 25. Engagement of the CTD by the viral protein Orf9b interrupts this network, stabilizing a partially-closed intermediate conformation and dampening dynamics at distal NTD sites. Collectively, our findings lay the groundwork for understanding Tom70 allostery and provide a framework for dissecting its mechanistic roles in chaperone engagement, mitochondrial import, and viral subversion.
Open AccessAn allosteric network governs Tom70 conformational dynamics to coordinate mitochondrial protein import
Tom70 mediates mitochondrial protein import by coordinating the transfer of cytosolic preproteins from Hsp70/Hsp90 to the translocase of the outer membrane (TOM) complex. In humans, the cytosolic domain of Tom70 ( Hs Tom70c) is entirely α -helical and comprises modular TPR motifs divided into an N-terminal chaperone-binding domain (NTD) and a C-terminal preprotein-binding domain (CTD). However, the mechanisms linking these functional regions remain poorly understood. Here, we present the 2.04 Å crystal structure of unliganded Hs Tom70c, revealing two distinct conformations – open and closed – within the asymmetric unit. These states are stabilized in part by interdomain crystal contacts and are supported in solution by hydrogen–deuterium exchange mass spectrometry (HDX-MS) and molecular dynamics (MD) simulations. Principal component and dynamical network analyses reveal a continuum of motion linking the NTD and CTD via key structural elements, notably residues in helices α 7, α 8, and α 25. Engagement of the CTD by the viral protein Orf9b interrupts this network, stabilizing a partially-closed intermediate conformation and dampening dynamics at distal NTD sites. Collectively, our findings lay the groundwork for understanding Tom70 allostery and provide a framework for dissecting its mechanistic roles in chaperone engagement, mitochondrial import, and viral subversion.
- Open Access
The Mechanism of Histone Ubiquitylation by the ASB9-CUL5 Ubiquitin Ligase
The E3 ligase substrate receptor ankyrin and SOCS box protein 9 (ASB9) was shown to bind over 10 different proteins including metabolic enzymes such as creatine kinase, filament proteins such as vimentin, and histones. In previous work, we characterized the ASB9-Cullin 5 E3 ligase (ASB9-CRL) ubiquitylation of creatine kinase and showed that ubiquitylation required the ring-between-ring ligase, ARIH2. Here we characterize the ASB9-CRL ubiquitylation of histones and show that histones H3 and H4 are polyubiquitylated by the ASB9-CRL whereas histones H2A and H2B are not. Many, but not all lysines in the histones are ubiquitylated suggesting some substrate specificity. Binding experiments show that the ligase-histone interaction is highly electrostatic and the neddylated ASB9-CRL binds with highest affinity. Only free histones are ubiquitylated. When the histones are in nucleosomes or in complex with the chaperone Asf1, they are not ubiquitylated. Only K48 and K63 polyubiquitin chains were observed, suggesting that the ubiquitylation probably drives protein degradation. The presence of ASB9 in specific cell types correlates with situations in which free histones H3 and H4 need to be degraded. In this work, we demonstrate that the ASB9-CRL is the ligase that facilitates degradation of histones H3 and H4. In addition, this work represents the first example of Cullin 5 mediated ubiquitylation that does not require a ring-between-ring “helper” ligase.
Open AccessThe Mechanism of Histone Ubiquitylation by the ASB9-CUL5 Ubiquitin Ligase
The E3 ligase substrate receptor ankyrin and SOCS box protein 9 (ASB9) was shown to bind over 10 different proteins including metabolic enzymes such as creatine kinase, filament proteins such as vimentin, and histones. In previous work, we characterized the ASB9-Cullin 5 E3 ligase (ASB9-CRL) ubiquitylation of creatine kinase and showed that ubiquitylation required the ring-between-ring ligase, ARIH2. Here we characterize the ASB9-CRL ubiquitylation of histones and show that histones H3 and H4 are polyubiquitylated by the ASB9-CRL whereas histones H2A and H2B are not. Many, but not all lysines in the histones are ubiquitylated suggesting some substrate specificity. Binding experiments show that the ligase-histone interaction is highly electrostatic and the neddylated ASB9-CRL binds with highest affinity. Only free histones are ubiquitylated. When the histones are in nucleosomes or in complex with the chaperone Asf1, they are not ubiquitylated. Only K48 and K63 polyubiquitin chains were observed, suggesting that the ubiquitylation probably drives protein degradation. The presence of ASB9 in specific cell types correlates with situations in which free histones H3 and H4 need to be degraded. In this work, we demonstrate that the ASB9-CRL is the ligase that facilitates degradation of histones H3 and H4. In addition, this work represents the first example of Cullin 5 mediated ubiquitylation that does not require a ring-between-ring “helper” ligase.
- Open Access
- 10 April 2025
- Published by Oxford University Press (OUP) in Nucleic Acids Research
- Vol. 53 (7)
CRISPR RNA binding drives structural ordering that primes Cas7-11 for target cleavage
Type III-E CRISPR–Cas effectors, referred to as Cas7-11 or giant Repeat-Associated Mysterious Protein, are single proteins that cleave target RNAs (tgRNAs) without nonspecific collateral cleavage, opening new possibilities for RNA editing. Here, biochemical assays combined with amide hydrogen–deuterium exchange mass spectrometry (HDX–MS) experiments reveal the dynamics of apo Cas7-11. The HDX–MS results suggest a mechanism by which CRISPR RNA (crRNA) stabilizes the folded state of the protein and subsequent tgRNA binding remodels it to the active form. HDX–MS shows that the four Cas7 RNA recognition motif (RRM) folds are well-folded, but insertion sequences, including disordered catalytic loops and β-hairpins of the Cas7.2/Cas7.3 active sites, fold upon binding crRNA leading to stronger interactions at domain–domain interfaces, and folding of the Cas7.1 processing site. TgRNA binding causes conformational changes around the catalytic loops of Cas7.2 and Cas7.3. We show that Cas7-11 cannot independently process the CRISPR array and that binding of partially processed crRNA induces multiple states in Cas7-11 and reduces tgRNA cleavage. The insertion domain interacts most stably with mature crRNA. Finally, we show a crRNA-induced conformational change in one of the tetratricopeptide repeat fused with Cas/HEF1-associated signal transducer (TPR-CHAT) binding sites providing an explanation for why crRNA binding facilitates TPR-CHAT binding.
- 10 April 2025
- Published by Oxford University Press (OUP) in Nucleic Acids Research
- Vol. 53 (7)
Open AccessCRISPR RNA binding drives structural ordering that primes Cas7-11 for target cleavage
Type III-E CRISPR–Cas effectors, referred to as Cas7-11 or giant Repeat-Associated Mysterious Protein, are single proteins that cleave target RNAs (tgRNAs) without nonspecific collateral cleavage, opening new possibilities for RNA editing. Here, biochemical assays combined with amide hydrogen–deuterium exchange mass spectrometry (HDX–MS) experiments reveal the dynamics of apo Cas7-11. The HDX–MS results suggest a mechanism by which CRISPR RNA (crRNA) stabilizes the folded state of the protein and subsequent tgRNA binding remodels it to the active form. HDX–MS shows that the four Cas7 RNA recognition motif (RRM) folds are well-folded, but insertion sequences, including disordered catalytic loops and β-hairpins of the Cas7.2/Cas7.3 active sites, fold upon binding crRNA leading to stronger interactions at domain–domain interfaces, and folding of the Cas7.1 processing site. TgRNA binding causes conformational changes around the catalytic loops of Cas7.2 and Cas7.3. We show that Cas7-11 cannot independently process the CRISPR array and that binding of partially processed crRNA induces multiple states in Cas7-11 and reduces tgRNA cleavage. The insertion domain interacts most stably with mature crRNA. Finally, we show a crRNA-induced conformational change in one of the tetratricopeptide repeat fused with Cas/HEF1-associated signal transducer (TPR-CHAT) binding sites providing an explanation for why crRNA binding facilitates TPR-CHAT binding.
- Open Access
- 19 October 2024
- Published by Springer Nature in Nature Communications
- Vol. 15 (1) , 1-12
The homeodomain regulates stable DNA binding of prostate cancer target ONECUT2
The CUT and homeodomain are ubiquitous DNA binding elements often tandemly arranged in multiple transcription factor families. However, how the CUT and homeodomain work concertedly to bind DNA remains unknown. Using ONECUT2, a driver and therapeutic target of advanced prostate cancer, we show that while the CUT initiates DNA binding, the homeodomain thermodynamically stabilizes the ONECUT2-DNA complex through allosteric modulation of CUT. We identify an arginine pair in the ONECUT family homeodomain that can adapt to DNA sequence variations. Base interactions by this ONECUT family-specific arginine pair as well as the evolutionarily conserved residues are critical for optimal DNA binding and ONECUT2 transcriptional activity in a prostate cancer model. The evolutionarily conserved base interactions additionally determine the ONECUT2-DNA binding energetics. These findings provide insights into the cooperative DNA binding by CUT-homeodomain proteins.
- 19 October 2024
- Published by Springer Nature in Nature Communications
- Vol. 15 (1) , 1-12
Open AccessThe homeodomain regulates stable DNA binding of prostate cancer target ONECUT2
The CUT and homeodomain are ubiquitous DNA binding elements often tandemly arranged in multiple transcription factor families. However, how the CUT and homeodomain work concertedly to bind DNA remains unknown. Using ONECUT2, a driver and therapeutic target of advanced prostate cancer, we show that while the CUT initiates DNA binding, the homeodomain thermodynamically stabilizes the ONECUT2-DNA complex through allosteric modulation of CUT. We identify an arginine pair in the ONECUT family homeodomain that can adapt to DNA sequence variations. Base interactions by this ONECUT family-specific arginine pair as well as the evolutionarily conserved residues are critical for optimal DNA binding and ONECUT2 transcriptional activity in a prostate cancer model. The evolutionarily conserved base interactions additionally determine the ONECUT2-DNA binding energetics. These findings provide insights into the cooperative DNA binding by CUT-homeodomain proteins.