Oligomerization and Auto-methylation of the Human Lysine Methyltransferase SETD6
Abstract
The intricate mechanisms of cellular signaling are profoundly influenced by post-translational modifications, among which lysine methylation plays a pivotal and diverse role in biological processes and disease progression. This specific modification is orchestrated by a specialized family of enzymes known as protein lysine methyltransferases, or PKMTs. SETD6, or SET-domain-containing protein 6, stands as a notable member within this extensive PKMT family, identified as a mono-methyltransferase. Previous investigations have established SETD6′s critical involvement in the regulation of fundamental cellular pathways, including the crucial NF-κB signaling cascade, the WNT pathway, and mechanisms combating oxidative stress. Despite these significant functional insights, a comprehensive understanding of SETD6′s precise enzymatic mode of action at the biochemical level has remained largely elusive.
In this current study, we present compelling evidence elucidating novel aspects of SETD6′s biochemical properties, specifically demonstrating its capacity to form higher molecular weight structures. Our findings reveal that the monomeric, dimeric, and trimeric configurations of SETD6 are remarkably stabilized by the presence of S-adenosyl-L-methionine (SAM), the essential methyl donor molecule required for methylation reactions. Furthermore, we unequivocally establish that SETD6 possesses intrinsic auto-methylation activity, a process where the enzyme modifies itself. Through meticulous mapping, we identify Lysine 39 (K39) and Lysine 179 (K179) as the predominant auto-methylation sites, with a comparatively moderate auto-methylation activity also observed towards Lysine 372 (K372). Intriguingly, introducing a precise point mutation at K179, a residue situated within the catalytic SET domain of SETD6, significantly impaired the enzyme’s ability to assemble into its trimeric form. This observation, in contrast to mutations at K39 and K372 which did not produce the same effect, strongly suggests a profound and direct connection between SETD6′s auto-methylation status and its oligomerization state.
Proceeding further, through a series of rigorous radioactive in-vitro methylation experiments complemented by detailed biochemical kinetics analysis, we demonstrate a critical functional consequence of this auto-methylation. Specifically, we show that auto-methylation occurring at K39 and K179 leads to a substantial increase in the catalytic rates of SETD6. Collectively, the comprehensive data presented in this study robustly support a mechanistic model wherein SETD6′s auto-methylation and its intrinsic self-interaction dynamically and positively regulate its enzymatic activity under in-vitro conditions. This paradigm of self-regulation through auto-methylation and oligomerization may not be unique to SETD6, and our findings potentially suggest that similar intricate regulatory mechanisms could govern the activities of other protein lysine methyltransferases, opening new avenues for future research into this important enzyme family.
Introduction
Protein lysine methylation represents a widespread and remarkably versatile post-translational modification, serving as a fundamental regulatory mechanism across a vast array of cellular processes and signaling pathways. This modification, which involves the covalent addition of one or more methyl groups to a lysine residue within a protein, is precisely catalyzed by a specialized class of enzymes known as Protein Lysine Methyltransferases (PKMTs). The human genome encodes over 60 candidate members of this diverse enzyme family. A defining characteristic for the vast majority of PKMTs is the presence of a highly conserved catalytic SET domain, which is directly responsible for facilitating the transfer of methyl groups. These enzymes are capable of adding mono-, di-, or tri-methyl groups to both histone and non-histone proteins, thereby modulating their structure, function, and interactions. The essential methyl donor for these methyl-transferase reactions is S-adenosyl-L-methionine (SAM), often referred to as the universal methyl donor. The enzymatic transfer of a methyl group from SAM results in the formation of a methylated lysine residue on the target protein and S-adenosyl-L-homocysteine (SAH) as a by-product.
Enzymes frequently employ sophisticated strategies to fine-tune their activity and stability. Among these, self-modification and self-interaction are fundamental biochemical mechanisms observed across numerous enzyme families, including a significant number of kinases. These intrinsic properties are critical for maintaining enzyme stability, modulating their catalytic efficiency, and enabling them to effectively transduce downstream signals within complex cellular networks. While the phenomenon of homo-dimerization, where two identical protein subunits associate, has not been widely reported for PKMTs in the same manner as for some other enzyme classes, compelling evidence exists demonstrating a direct link between auto-methylation and the enzymatic activity of various PKMTs. For instance, the activity of G9a, an H3K9 PKMT, has been shown to be intricately regulated by auto-methylation at Lysine 239. This auto-methylation of G9a functions as an effective mediator for its interaction with HP1, mimicking the binding affinity typically observed with H3K9me3. Similarly, in yeast, the methyl-transferase Clr4 (a homolog of mammalian Suv39h) undergoes a conformational switch, mediated by auto-methylation at multiple lysine residues, specifically K455 and K464, which consequently enhances its H3K9 methylation activity. Conversely, recent research has illustrated scenarios where auto-methylation can have an inhibitory effect; for example, auto-methylation of SUV39H2 at K392 has been shown to reduce its enzymatic activity by impeding its interaction with target substrates. In another distinct case, auto-methylation of PRMT8 leads to a reduction in its binding affinity for SAM, its essential methyl donor. Given the increasing recognition of PKMTs as highly attractive therapeutic targets for a wide range of diseases, a comprehensive understanding of the precise regulatory mechanisms governing their activity, particularly how their auto-methylation influences their function, remains critically important and largely undefined.
The SET-domain-containing protein 6, or SETD6, is a distinguished member of the PKMT enzyme family. Its gene is precisely located on the long arm of chromosome 16, specifically at position 16q21. The gene itself encompasses 5049 base pairs and comprises eight coding exons. Human SETD6 exists in two primary splice variants: the longer “isoform a,” which consists of 473 amino acid residues, and the shorter “isoform b,” which is 449 residues long due to the absence of an in-frame segment spanning residues 40 to 63. SETD6 has been consistently implicated in the regulation of a diverse array of crucial cellular processes. For example, it is known to mono-methylate the NF-κB subunit RelA at Lysine 310, a modification that can either suppress the activation of NF-κB target genes in certain contexts or, paradoxically, promote RelA transcriptional activity in bladder cancer, highlighting its context-dependent roles. Beyond NF-κB signaling, SETD6 has also been demonstrated to play a significant role in the NRF2 oxidative stress response pathway through its interaction with the protein DJ1. Furthermore, it contributes to the WNT signaling pathway by methylating PAK4, participates in nuclear hormone receptor signaling, and is involved in embryonic stem cell differentiation via the methylation of histone H2AZ.
Despite the continuously emerging and undeniable importance of SETD6 in orchestrating a multitude of diverse cellular processes, there remains a notable paucity of information regarding the fundamental biochemical properties that govern and regulate its enzymatic activity. Addressing this critical knowledge gap, the current study delves into the intrinsic regulatory mechanisms of SETD6. Our investigations reveal that SETD6 dynamically forms multiple distinct oligomeric states, including monomers, dimers, and trimers, and crucially, that the stability of these higher-order structures is directly influenced and enhanced by the presence of SAM. We further report the significant discovery that SETD6 exhibits robust auto-methylation activity, a self-modifying process. Through meticulous site-mapping experiments, we precisely identify Lysine 39, Lysine 179, and Lysine 372 as key auto-methylation sites. A targeted point mutation at Lysine 179 was found to impair SETD6′s capacity to form a stable trimer, providing a strong implication for a direct functional link between the enzyme’s auto-methylation status and its precise oligomerization properties. Finally, through comprehensive biochemical kinetics analysis, we definitively demonstrate that auto-methylation occurring at Lysine 39 and Lysine 179 results in a significant increase in SETD6′s affinity for its methylated substrate, thereby enhancing its catalytic efficiency. This integrated approach sheds light on the sophisticated self-regulatory mechanisms governing SETD6′s function, offering critical insights into its diverse biological roles.
Results
SETD6 forms monomers, dimers and trimers as well as higher molecular weight structures.
To rigorously investigate whether SETD6 has the propensity to form molecular weight structures beyond its basic monomeric unit, we initiated a series of crosslinking experiments utilizing recombinant His-tagged SETD6. For these experiments, the protein was incubated in the presence of 68 µM bis-sulfosuccinimidyl-suberate (BS3), a well-established homobifunctional cross-linking reagent. BS3 functions by covalently linking lysine residues that are positioned approximately 1.14 nanometers apart, thereby stabilizing transient protein-protein interactions and allowing for their detection. Under conditions where no crosslinker was added, our analysis revealed the presence of a single band, corresponding precisely to the monomeric form of SETD6. However, upon initiating the incubation with BS3, we observed the progressive formation of distinct protein complexes. Within 30 minutes of incubation, complexes corresponding to the expected sizes of SETD6 dimers and trimers became readily detectable. The formation of these higher-order structures reached a saturation point after approximately 2 hours of incubation, indicating a stable population of oligomers. To ensure the specificity of the BS3 crosslinking assay, His-SUMO was employed as a negative control, and as expected, it did not exhibit similar crosslinking patterns, confirming that the observed oligomerization was specific to SETD6.
To further characterize the oligomerization state of SETD6 under native, non-denaturing conditions, recombinant His-SETD6 was subjected to size exclusion chromatography (SEC) using a Superdex200 Increase column. This technique separates proteins based on their hydrodynamic radius, allowing for the determination of their native molecular weight and oligomeric state. Consistent with the findings from our crosslinking experiments, the SEC profiles revealed a predominant population of SETD6 monomers, alongside detectable but lower amounts of dimer and trimer forms. The identity and presence of these various oligomeric states were robustly validated through subsequent Coomassie staining, which visualizes all proteins, and Western blot analysis using an anti-SETD6 antibody, ensuring specificity. To ascertain whether a dynamic equilibrium exists between these monomeric, dimeric, and trimeric forms of SETD6, we meticulously collected the eluted fractions corresponding to the trimeric and dimeric populations (specifically fractions 13 and 14 from the initial SEC run). These enriched fractions were then re-loaded onto the SEC column for a second separation. The results of this re-analysis showed a distribution of SETD6 native complexes strikingly similar to the initial SEC run, strongly supporting the existence of a constitutive and reversible equilibrium between the different oligomerization states of SETD6 in solution. While native SEC provided valuable insights, better separation and enrichment of the dimer-trimer populations were achieved when recombinant SETD6 was subjected to crosslinking prior to loading onto the SEC column, indicating that covalent stabilization of the oligomers improved their resolution. Finally, to extend our understanding to a more physiologically relevant context, we investigated the oligomerization state of SETD6 within living cells. Flag-tagged SETD6 was over-expressed in HEK293T cells, subsequently immunoprecipitated, and then eluted using Flag-peptides. The eluted protein was then separated on a native gel, followed by Western blot analysis using an anti-SETD6 antibody. The results clearly demonstrated that SETD6 forms dimeric and trimeric structures within cells, exhibiting a pattern and relative ratio remarkably similar to those observed with the recombinant His-SETD6 in vitro. Taken comprehensively, our findings from both in-vitro biochemical assays and in-cell analyses consistently demonstrate that SETD6 dynamically forms monomeric, dimeric, and trimeric structures, both in purified recombinant form and within cellular environments.
SAM stabilizes SETD6 monomeric, dimeric and trimeric structures.
Given that S-adenosyl-L-methionine (SAM) functions as the indispensable methyl donor for all methylation reactions catalyzed by SETD6, we formulated a hypothesis that SAM might not only participate in catalysis but also play a crucial role in stabilizing the distinct monomeric, dimeric, and trimeric complexes of SETD6. To test this hypothesis, recombinant SETD6 was incubated under standard methylation reaction conditions, both in the absence and presence of SAM. Subsequent analysis was performed using native gel electrophoresis, a technique that preserves protein complexes, followed by detection via Coomassie staining and Western blot analysis. These complementary methods allowed for a comprehensive assessment of changes in the abundance and distribution of SETD6′s monomeric, dimeric, trimeric, and any other higher molecular weight structures. Our observations revealed a striking effect: the addition of SAM led to a marked stabilization of the monomeric, dimeric, and trimeric forms of SETD6. Concurrently, there was a significant reduction in the presence of very high molecular weight bands, which are typically indicative of protein aggregation. Conversely, in the absence of SAM, we consistently detected numerous bands with significantly larger sizes than those corresponding to the expected monomers, dimers, and trimers, suggesting a tendency towards aggregation or formation of non-specific, unstable higher-order structures. Importantly, this phenomenon was demonstrated to be specifically dependent on SAM, as the high molecular weight bands were not diminished after the addition of an equivalent amount of S-adenosyl-L-homocysteine (SAH), the by-product of the methylation reaction. This distinction strongly indicates that the stabilization effect is directly mediated by SAM itself, rather than by a general presence of an adenosine-containing molecule.
To further strengthen the evidence for SAM’s critical role in promoting the enrichment of SETD6′s monomeric, dimeric, and trimeric forms, we employed a specific point mutation in SETD6, namely Y285A. This mutation targets Tyrosine 285, a residue known to be essential for SAM binding and which consequently leads to the catalytic inactivation of the enzyme. As observed through native gel electrophoresis, the addition of SAM resulted in the significant reduction of high molecular weight aggregates in wild-type SETD6. In stark contrast, for the SETD6 Y285A mutant, the presence of SAM had no such effect, and high molecular weight bands persisted, clearly demonstrating that an intact SAM binding site is absolutely requisite for the observed stabilization.
To provide additional quantitative support for SAM’s role in stabilizing these specific oligomeric structures of SETD6, we conducted dynamic light scattering (DLS) analysis. DLS measures the hydrodynamic diameter of particles in solution, providing insights into their size and aggregation state. We compared wild-type SETD6 to the SETD6 Y285A mutant, after both were incubated at 30°C for 2 hours with and without SAM. For wild-type SETD6 in the absence of SAM, the DLS profile revealed a heterogeneous population of aggregates with hydrodynamic diameters ranging broadly from 20 nm to 80 nm. This observation is in excellent agreement with the native gel electrophoresis results, which showed the existence of various molecular weights, not all of which are distinctly resolved by DLS. However, upon the addition of SAM to wild-type SETD6, the DLS profile dramatically shifted, yielding a single, much smaller population of structures with a narrow hydrodynamic diameter range of 4-10 nm. This size range correlates perfectly with the expected dimensions of SETD6 monomers, dimers, and trimers, confirming SAM’s role in promoting the formation of these defined, smaller oligomeric species. In stark contrast, for the SETD6 Y285A mutant, whether incubated with or without SAM, the DLS consistently indicated the presence of a single population of very large aggregates, with hydrodynamic diameters spanning 300-600 nm. This result definitively demonstrates that the addition of SAM had no discernible effect on the aggregation state of the SETD6 Y285A mutant, further underscoring the critical requirement of a functional SAM-binding site for the observed stabilization effect. Collectively, these multifaceted results from native gel electrophoresis, mutation analysis, and dynamic light scattering unequivocally suggest that SAM plays a fundamental role in stabilizing SETD6′s functional monomeric, dimeric, and trimeric structures, preventing the formation of larger, potentially inactive, aggregates.
K39 and K179 serve as the major auto-methylation sites of SETD6.
Prior research has consistently indicated that SETD6 possesses a robust auto-methylation activity, signifying its capacity to methylate itself. However, despite this established activity, the precise lysine residue(s) within SETD6 that undergo this auto-modification had not been previously identified or mapped. To begin the intricate process of roughly pinpointing the auto-methylation site within SETD6, we engineered and purified a truncated version of the protein that specifically lacked its C-terminus. This N-terminal truncation, which notably retained the catalytic SET domain, was then rigorously assessed for its in-vitro auto-methylation activity. Experiments were conducted both with the N-terminal truncation alone and in combination with the full-length wild-type SETD6 protein. Interestingly, in sharp contrast to the full-length protein, we were unable to detect any auto-methylation activity when the N-terminal truncation was incubated in isolation. However, a pivotal observation emerged when both the wild-type full-length SETD6 and the N-terminal truncation were incubated together in the same reaction mixture: both proteins were found to be methylated. These preliminary yet crucial results strongly suggested that the auto-methylation site, or at least a critical component necessary for auto-methylation, is indeed located within the N-terminal fragment of SETD6. Furthermore, these findings provided compelling evidence supporting a *trans*-auto-methylation mechanism, where one SETD6 molecule methylates another. Nevertheless, while *trans*-auto-methylation was strongly indicated, we could not definitively exclude the concurrent possibility of *cis*-auto-methylation, where an enzyme modifies itself.
To precisely and unambiguously map the auto-methylation site(s) within SETD6, recombinant His-tagged SETD6 was incubated under methylation reaction conditions, with and without SAM, and subsequently subjected to comprehensive mass spectrometry analysis. This highly sensitive analytical technique allowed for the detection and identification of modified lysine residues. The mass spectrometry analysis successfully identified two distinct lysine residues that underwent mono-methylation: Lysine 39 (K39) and Lysine 372 (K372). K39 is located strategically at the N-terminus of SETD6, immediately adjacent to the conserved catalytic SET domain. In contrast, K372 is positioned towards the C-terminus of the protein, residing within the Rubisco domain, which is a structural motif known for its importance in mediating protein-protein interactions. Building upon these initial mass spectrometry findings, and also considering our preliminary rough mapping experiments which localized the auto-methylation to the N-terminus of SETD6, we utilized structural bioinformatics tools, specifically the Swiss-PDB-Viewer (sPDBv) with the known SETD6 crystal structure (PDB ID: 3QXY). This structural analysis allowed us to identify exposed lysine residues that would be sterically accessible for enzymatic modification. Through this analysis, we identified two prominently exposed lysine residues: K39, which was already confirmed by mass spectrometry, and critically, Lysine 179 (K179), which is located directly within the catalytic SET domain itself. Based on this convergent evidence from rough mapping, mass spectrometry, and structural accessibility analysis, we formulated the specific hypothesis that K39, K179, and K372 collectively serve as the primary auto-methylation sites on SETD6.
To definitively test this hypothesis and quantify the contribution of each hypothesized site, purified SETD6 point mutants, where the lysine residues K39, K179, and K372 were individually mutated to arginine (K39R, K179R, K372R to prevent methylation), were subjected to a sensitive radioactive in-vitro methylation assay. This assay directly measures the incorporation of radiolabeled methyl groups onto the protein. Our quantitative analysis revealed nuanced differences in auto-methylation activity among the mutants. Specifically, the K372R mutant exhibited only a moderate decrease in auto-methylation activity compared to wild-type SETD6, suggesting a less critical role for this site. In stark contrast, the auto-methylation activity was dramatically reduced in the K39R mutant. Most strikingly, auto-methylation was almost completely abolished in the K179R mutant when compared to the wild-type enzyme. These compelling results strongly suggest that while K39, K179, and K372 all contribute to the auto-methylation activity of SETD6, Lysine 39 and Lysine 179 serve as the predominant and major auto-methylation sites, playing a critical role in the self-modification process of SETD6.
Auto-methylation at K179 is required for SETD6 trimeric state.
Having firmly established that SETD6 engages in self-dimerization and, crucially, undergoes auto-methylation primarily at Lysine 39 (K39) and Lysine 179 (K179), we then advanced a compelling hypothesis: that these two fundamental biochemical characteristics—oligomerization and auto-methylation—are intrinsically linked and interdependent. To rigorously investigate this proposed connection, we performed a series of native gel electrophoresis experiments. This technique is particularly valuable as it allows for the separation of proteins based on their native charge and size, thereby preserving and revealing their oligomeric states. We meticulously compared the oligomeric profiles of wild-type SETD6 with those of specific SETD6 mutants: K39R (where K39 is replaced by Arginine), K179R (where K179 is replaced by Arginine), and Y285A (a catalytically inactive mutant with a compromised SAM binding site). All experiments were conducted under standardized methylation reaction conditions, critically in the presence of S-adenosyl-L-methionine (SAM), the essential methyl donor.
The results from these native gel electrophoresis analyses revealed a significant and distinct shift in the oligomerization equilibrium specifically for the K179R mutant when compared to wild-type SETD6. Most notably, the K179R mutant displayed a discernible increase in protein aggregation, manifesting as higher molecular weight smudges or large insoluble complexes, concurrently with a significant reduction in its ability to form the discrete trimeric state. This striking alteration in the oligomeric profile of the K179R mutant strongly suggests a direct and essential link between the auto-methylation of SETD6 at K179 and its capacity to form stable trimers. The diminished trimer formation alongside increased aggregation in the K179R mutant indeed implies that the precise oligomeric state of SETD6 is intimately connected to its auto-methylation status.
Further substantiating this link, an in-vitro methylation assay performed under native conditions demonstrated a clear correlation. We observed a time-dependent increase in the auto-methylation signal specifically associated with the trimeric form of the wild-type enzyme. This indicates that the trimeric state is either particularly active in auto-methylation or that auto-methylation itself stabilizes this oligomeric configuration over time. Complementary results were obtained when His-tagged SETD6 was subjected to a radioactive methylation assay in the absence or presence of the chemical cross-linker bis-sulfosuccinimidyl-suberate (BS3). This experiment revealed a clear enrichment of the trimeric form, and to a lesser extent the dimeric form, in terms of auto-methylation signal when crosslinking was employed. It is crucial to acknowledge a technical nuance: while this assay demonstrated auto-methylation activity within the trimer and dimer bands, the absolute intensity of these bands, particularly when compared to the monomeric state in non-crosslinked samples, does not accurately reflect the true native equilibrium between the different oligomeric states. This limitation arises due to the inherent very low efficiency of the BS3 cross-linking reaction itself, meaning only a small fraction of the total protein is crosslinked into oligomers at any given time, preventing a direct quantitative comparison of the overall population distribution. Nevertheless, the findings consistently point towards a critical role for auto-methylation at K179 in the stable formation of SETD6′s trimeric state.
The auto-methylation at K39 and K179 increases SETD6 enzymatic activity.
The compelling evidence of a direct linkage between SETD6′s oligomeric state and its auto-methylation status, particularly at Lysine 39 (K39) and Lysine 179 (K179), led us to hypothesize a profound functional consequence: that SETD6′s auto-methylation directly influences and modulates its enzymatic activity. To rigorously test this critical hypothesis, we conducted a series of in-vitro methylation assays. Our experimental design involved comparing the catalytic efficiency of wild-type SETD6 against a carefully engineered double mutant of SETD6, specifically carrying arginine substitutions at both K39 and K179 (K39R/K179R), rendering it deficient in auto-methylation at these key sites. The substrates chosen for these assays were RelA and PAK4, both of which we had previously demonstrated to be bona fide methylation targets of SETD6.
As anticipated, wild-type SETD6 readily methylated both RelA and PAK4, confirming its known catalytic function. Strikingly, and providing strong support for our hypothesis, the methylation of both substrates was dramatically and significantly reduced when catalyzed by the K39R/K179R double mutant. This profound reduction in substrate methylation by the auto-methylation-deficient mutant strongly indicates that auto-methylation at K39 and K179 is indeed critical for optimal SETD6 enzymatic activity.
To obtain more quantitative insights into the observed differences in enzymatic activity, we further characterized the wild-type SETD6 and the K39R/K179R mutant using the MTase-Glo™ methyltransferase assay. This bioluminescent assay indirectly measures methyltransferase activity by quantifying the S-adenosyl-L-homocysteine (SAH) by-product formed during the methylation reaction, thereby allowing for the determination of Michaelis-Menten kinetic parameters. In these experiments, we employed constant amounts of the enzymes while systematically varying the concentrations of RelA, our chosen substrate. Reactions were performed at two distinct time points, 10 and 15 minutes, which were predetermined to be within the linear range of SAH production, ensuring that the measured rates accurately reflected initial velocities. A particularly significant difference was observed in the apparent Michaelis constant (KM) between wild-type SETD6 and the K39R/K179R mutant. A lower KM value typically indicates a higher affinity of the enzyme for its substrate. The observed disparity in KM values strongly suggests that the auto-methylation of SETD6 at positions K39 and K179 plays a crucial role in enhancing the enzyme’s affinity towards its substrate, RelA. Taken together, our comprehensive experimental data provide robust evidence that the oligomeric state of SETD6 and its auto-methylation at K39 and K179 are not merely linked phenomena but are intricately and positively coupled, jointly regulating and enhancing SETD6′s catalytic activity under in-vitro conditions. This establishes a novel and critical regulatory mechanism for this important protein lysine methyltransferase.
Discussion
Protein Lysine Methyltransferases (PKMTs) are indispensable enzymes that serve as pivotal regulators across a multitude of intricate cellular signaling pathways. They exert their profound influence by catalyzing the methylation of specific lysine residues on a diverse array of substrates, encompassing both canonical histone proteins and a wide range of non-histone proteins. Despite their recognized importance in cellular biology, the precise biochemical mechanisms governing PKMT activity, and more specifically, how their intrinsic auto-methylation properties contribute to and regulate their overall enzymatic function, have remained largely enigmatic and significantly less understood compared to other enzyme classes, such as kinases. In this study, we have systematically addressed this critical knowledge gap, providing compelling evidence that the oligomeric state of SETD6 and its inherent capacity for auto-methylation are not merely incidental features but are, in fact, absolutely critical determinants for its enzymatic activity under controlled in-vitro conditions.
Our comprehensive data consistently support a refined model wherein S-adenosyl-L-methionine (SAM), the universal methyl donor, plays a multifaceted role, extending beyond mere substrate provision. Specifically, we demonstrate that SAM actively stabilizes the monomeric, dimeric, and trimeric oligomeric states of SETD6. SAM is a molecule of immense biological significance, serving as the primary methyl donor for a vast array of methylation reactions involving lysine, arginine, and DNA. Its intracellular production is tightly coupled to cellular metabolism, formed through the condensation of the amino acid methionine and adenosine triphosphate (ATP). ATP, in turn, functions as the primary molecular currency of intracellular energy, generated through fundamental cellular metabolic processes such as glycolysis, the citric acid (Krebs) cycle, and oxidative phosphorylation. This intricate metabolic link suggests a profound potential correlation between the prevailing cellular metabolic state and the enzymatic activity of various methyltransferases, including the specific catalytic activity of SETD6. It is highly plausible that the intracellular concentration of SAM, along with SETD6′s accessibility to this vital co-factor, could directly dictate the dynamic oligomerization and auto-methylation states of SETD6. Consequently, these biochemically regulated states would profoundly influence SETD6′s cellular function, particularly its precise ability to interact with and methylate its diverse protein substrates. Nevertheless, it is an important caveat that all these biochemical properties were meticulously measured in a controlled in-vitro environment. Therefore, future research is imperative to thoroughly explore and confirm whether these intricate regulatory mechanisms indeed occur under the complex physiological settings and dynamic cellular contexts of living organisms.
While our investigations have definitively established that SETD6 exhibits the ability to form both dimeric and trimeric oligomers, a more precise mapping of the exact amino acid residues or structural interfaces that mediate these distinct oligomeric states has proven challenging in this current study. Although we have recently resolved the crystal structure of SETD6 in complex with a RelA peptide in the presence of SAM at a commendable resolution of 2.2 Å, the available structural information, unfortunately, is not yet sufficient to definitively pinpoint the specific interaction sites responsible for the observed oligomerization. To overcome this limitation and gain deeper structural insights, we anticipate that future structural biology approaches will be indispensable. Techniques such as high-resolution X-ray crystallography, Nuclear Magnetic Resonance (NMR) spectroscopy, and advanced electron microscopy, particularly for visualizing and modeling larger protein complexes, will be crucial. Applying these sophisticated methods to wild-type SETD6 and its various mutants, especially those affecting oligomerization and auto-methylation, will allow us to further dissect the structural consequences of SETD6′s different oligomeric and auto-methylation states. Such detailed structural information is essential to potentially uncouple these two intimately linked phenomena, enabling a more nuanced understanding of the individual contribution of each to SETD6′s overall activity and regulation. Furthermore, these advanced structural approaches may also provide the definitive structural evidence required to distinguish unequivocally between *cis*- and *trans*-auto-methylation mechanisms within SETD6, shedding light on how one SETD6 molecule self-modifies or modifies another.
Signaling pathways orchestrated by lysine methylation and the enzymes that catalyze them are increasingly recognized as highly attractive targets for therapeutic intervention across a spectrum of diseases. The auto-methylation and oligomerization properties of SETD6, which we have meticulously described in this study, represent novel regulatory mechanisms that profoundly influence SETD6′s enzymatic activity. This groundbreaking understanding opens up exciting possibilities for the rational design of highly specific inhibitors that could effectively block this newly discovered mode of enzyme activation. In a recent related publication, our group successfully designed a short peptide derived from the RelA sequence, which was strategically fused to a cell-penetrating peptide. We demonstrated that this chimeric peptide could directly interact with SETD6 and effectively inhibit its enzymatic activity, both in-vitro and within live cells. Building upon this success, further comprehensive investigation is now required to fully elucidate the complex biological and physiological roles of SETD6 auto-methylation and its precise oligomeric conditions in both normal physiological states and in the context of various disease pathologies. Such a profound understanding will be instrumental in enabling us to design bespoke, highly specific inhibitors that are precisely based on the newly elucidated biochemical properties of SETD6 presented here. These next-generation inhibitors hold significant potential to be considerably more selective compared to existing PKMT competitive inhibitors. This enhanced selectivity stems from the unique and functional linkage between SETD6′s specific oligomeric states and its auto-methylation at K39 and K179. In conclusion, our collective findings add an entirely new dimension to our current understanding of SETD6′s intricate enzymatic properties, thereby opening novel and promising avenues of research at both the fundamental biochemical level and the more complex functional cellular level. Moreover, these insights may strongly suggest that similar intricate self-regulatory mechanisms could govern the activities of other protein lysine methyltransferases, offering a broader paradigm for future investigations into this crucial enzyme family.
Materials and methods
Plasmids
The genetic constructs for human SETD6 isoform b, including both the wild-type sequence and the Y285A mutant, were carefully subcloned into the pET-Duet plasmid, which was engineered to incorporate a 6xHis-tag for subsequent purification. The pGEX-6P-1 plasmid, containing RelA fused with a GST-tag, has been previously described in relevant scientific literature. Similarly, the pETS-SUMO PAK4 plasmid, featuring a His-tag, was detailed in an earlier publication. To generate the specific SETD6 mutants, namely K39R, K179R, and the double mutant K39R/K179R, we employed site-directed mutagenesis techniques with precisely designed primers. Following successful mutagenesis, the DNA sequences were thoroughly verified through sequencing to confirm the desired modifications, and these mutant constructs were subsequently cloned into the pET-Duet plasmid, which also included the 6xHis-tag. The SETD6-N’ construct, an N-terminal truncation, was amplified and cloned into the pET-Duet plasmid, again incorporating a 6xHis-tag, as has been described in prior work. For cellular overexpression experiments in HEK-293T cells, the pcDNA Flag-SETD6 plasmid was utilized, having been previously established and described by Vershinin et al.
Recombinant Protein Expression and Purification
For the efficient production of recombinant proteins, the pET-DUET plasmid containing the SETD6 gene with its associated 6xHis-tag was transformed into Escherichia coli Rosetta strain. Bacterial cultures were subsequently grown in Lysogeny Broth (LB) medium and induced with Isopropyl β-D-1-thiogalactopyranoside (IPTG) to initiate protein expression, followed by an overnight incubation at 18°C to optimize protein folding and solubility. Bacterial cells were harvested from the growth medium through centrifugation, and the resulting cell pellet was meticulously lysed by sonication on ice. The sonication process involved a total duration of 1.5 minutes, delivered in cycles of 10 seconds on and 5 seconds off, at 25% amplitude, to ensure efficient cell disruption while minimizing protein denaturation. Following lysis, the cellular debris was removed by centrifugation at 18,000 revolutions per minute for 20 minutes at 4°C, and the supernatant containing the soluble protein was then filtered. The His-tagged proteins were subsequently purified to high homogeneity using an ÄKTA gel filtration system equipped with a HisTrap column (from GE Healthcare), leveraging the high affinity of the poly-histidine tag for immobilized metal ions. Proteins were eluted from the column using a gradient of imidazole, typically reaching a concentration of 0.5 M. The purified proteins were then subjected to dialysis against a buffer containing 10% glycerol in Phosphate-Buffered Saline (PBS) to ensure stability and proper storage conditions.
Size exclusion chromatography
To assess the native oligomeric state and molecular weight of the purified His-SETD6 protein, samples were loaded onto a SuperdexTM200 Increase 10/300 size exclusion column (from GE Healthcare). The chromatographic separation was performed at a flow rate of 0.5 milliliters per minute, maintained at a temperature of 4°C, utilizing an eluent composed of 250 mM Sodium Chloride in PBS buffer. This setup allows for the separation of proteins based on their hydrodynamic volume, providing insights into their native size and complex formation.
Crosslinking assay
Crosslinking experiments were meticulously conducted using the BS3 (bis(sulfosuccinimidyl)suberate) crosslinker, a product sourced from Thermo-Fisher. In these assays, 20 micrograms of the purified His-SETD6 protein were incubated in a 25 microliter reaction volume, in the precise presence of 68 micromolar BS3, at room temperature for various durations. The crosslinking reaction was then promptly terminated by the addition of protein sample buffer (comprising 250 mM Tris-HCl at pH 6.8, 10% Sodium Dodecyl Sulfate (SDS), 30% Glycerol, 5% β-mercaptoethanol, and Bromophenol blue). Samples were subsequently heated at 95°C for 5 minutes to fully denature the proteins and resolve the complexes, followed by separation via SDS-PAGE. The resolved proteins were then visualized either by Coomassie staining, utilizing Expedeon InstantBlueTM for general protein detection, or directly used for an in-vitro methylation assay, as detailed in a subsequent section, depending on the experimental objective.
Antibodies and western blot Analysis
For immunodetection in Western blot analyses, the following primary antibodies were employed: a rabbit polyclonal anti-SETD6 antibody, which has been previously described in reference [16], and a mouse monoclonal anti-His antibody, commercially available from Thermo Fisher (Cat. No. MA1-21315). For secondary detection, HRP-conjugated antibodies were used: goat anti-rabbit (Jackson ImmunoResearch, 111-035-144) and goat anti-mouse (Jackson ImmunoResearch, 115-035-062). All antibodies were appropriately diluted in a 10% skim milk solution prepared in PBST buffer, strictly adhering to the manufacturer’s recommended concentrations to ensure optimal signal and minimal background. For Western blot analysis, protein samples were denatured by heating at 95°C for 5 minutes in either denaturing or native sample buffer, depending on the desired outcome (e.g., assessing protein quantity under denaturing conditions or oligomeric state under native conditions). Subsequently, the proteins were separated by either SDS-PAGE for molecular weight determination or native gel electrophoresis for assessing native complex formation, prior to transfer and antibody detection.
Native gel electrophoresis
To assess the native oligomeric states of proteins, samples were meticulously prepared by mixing with native sample buffer, which comprised 250 mM Tris (HCl) at pH 6.8, 30% Glycerol, and Bromophenol blue. The prepared samples were then subjected to separation on a 6% polyacrylamide gel under native conditions. The electrophoresis was conducted at a constant voltage of 150 V for a duration of 60 minutes, ensuring that the process was performed on ice to maintain protein stability and preserve their native conformations.
Cell line, transfection and Immunoprecipitation
Human embryonic kidney cells, specifically the HEK-293T cell line, were meticulously maintained in Dulbecco’s modified Eagle’s medium (Sigma, D5671) supplemented with 10% fetal bovine serum (Gibco), 2 milligrams per milliliter of L-glutamine (Sigma G7513), a penicillin-streptomycin solution (Sigma, P0781) to prevent bacterial contamination, and non-essential amino acids (Sigma, M7145). These cells were cultured within a humidified incubator maintained at 37°C with a controlled atmosphere of 5% CO2. For experimental procedures, cells were plated onto 10-centimeter dishes and subsequently transfected using the Mirus transfection reagent, TransIT®-LT1, strictly adhering to the manufacturer’s comprehensive instructions to achieve efficient gene delivery. Following the appropriate incubation period, cells were harvested and lysed using RIPA lysis buffer, a robust solution consisting of 50 mM Tris–HCl at pH 8, 150 mM NaCl, 1% Nonidet P-40, 0.5% deoxycholate, and 0.1% SDS (volume/volume), along with 1 mM dithiothreitol (DTT) and a Sigma protease inhibitor cocktail (P8340, diluted 1:100) to prevent protein degradation. The resultant cell lysate was then incubated with FLAG-M2-affinity gel beads (A2220; Sigma) for 1 hour at 4°C, allowing for the specific capture of Flag-tagged proteins. After this incubation, the beads were thoroughly washed four times with lysis buffer to remove non-specifically bound proteins. Finally, the Flag-tagged proteins were eluted from the beads by incubating with 80 micrograms of Flag peptides (Sigma, F3290-4MG) for 1 hour at 4°C, which competitively displaces the bound proteins.
Dynamic light scattering (DLS)
Dynamic Light Scattering (DLS) measurements were systematically conducted utilizing a Zetasizer Nano-ZS instrument, manufactured by Malvern, UK. Recombinant SETD6 protein samples were carefully prepared and subjected to an overnight incubation, both in the presence and absence of S-adenosyl-methionine, within a specialized PKMT buffer (composed of 10 mM Tris-HCl at pH 8, 2% glycerol, 0.8 mM KCl, and 1 mM MgCl2). The incubation was performed at a controlled temperature of 30°C to simulate relevant biochemical conditions. Prior to data acquisition, each sample was precisely equilibrated for 1 minute at 4°C to ensure thermal stability and consistency. Correlograms, which represent the fluctuation of light scattering intensity over time due to particle movement, were consistently collected at a scattering angle of 173°. A minimum of 10 runs, each lasting 10 seconds, were acquired for every sample at 4°C. The raw correlograms were then expertly analyzed using the CONTIN procedure, a widely recognized algorithm for deconvolution of DLS data, which is integrated within the software provided with the instrument, allowing for the determination of particle size distribution based on hydrodynamic diameter.
Mass spectrometry
Samples of recombinant SETD6 were prepared for mass spectrometry analysis by incubation overnight, both with and without S-adenosyl-methionine, in PKMT buffer at a controlled temperature of 30°C. Following the incubation, the protein samples were subjected to enzymatic digestion with trypsin, a protease commonly used to generate peptides suitable for mass spectrometry. The resulting peptide mixtures were then analyzed using a Q-Exactive LC-MS/MS instrument operating in data-dependent acquisition (DDA) mode. The raw mass spectrometry data were initially processed and pre-analyzed using Preview software (from Protein Metrics Inc) to determine optimal search parameters. Subsequently, a comprehensive database search was performed against the human proteome database, which was appended with a list of common protein contaminants, to identify and quantify peptides. Peptide and protein identifications were rigorously filtered to ensure a false discovery rate (FDR) of less than 1%, ensuring high confidence in the results and the precise mapping of modified residues.
In-vitro methylation assay
The in-vitro methylation assay was carefully constituted within a reaction tube, containing the recombinant protein of interest, precisely 2 microcuries of 3H-labeled S-adenosyl-methionine (AdoMet), obtained from Perkin-Elmer, serving as the radiolabeled methyl donor, and the appropriate PKMT buffer Ademetionine, all combined in a total reaction volume of 25 microliters. Following an incubation period at 30°C for varying durations, tailored to the specific experimental design, Ademetionine the methylation reactions were terminated. The reaction products were then resolved by either SDS-PAGE for molecular weight separation under denaturing conditions or loaded onto a native gel electrophoresis system to preserve and analyze native protein complexes. Subsequent visualization and quantification of the incorporated radioactivity were performed through autoradiography, allowing for the detection of methylated proteins, and/or by Coomassie staining, utilizing Expedeon InstantBlueTM, for general protein visualization.
Bioluminescent methyltransferase assay
The measurement of methyltransferase enzyme activity was conducted using the Methyltransferase-Glo™ assay, a bioluminescent detection system. The assay was performed rigorously following the detailed protocols and specifications provided by the manufacturer, ensuring consistent and accurate quantification of enzyme kinetics.