Engineered depalmitoylases enable selective manipulation of protein localization and function

Jayaraman, Srinidhi; Kochiss, Audrey; Alcalay, Thy-Lan; del Rivero Morfin, Pedro J.; Ben-Johny, Manu

doi:10.1038/s41467-025-58908-x

Download PDF

Article
Open access
Published: 13 April 2025

Engineered depalmitoylases enable selective manipulation of protein localization and function

Nature Communications volume 16, Article number: 3514 (2025) Cite this article

5287 Accesses
1 Citations
3 Altmetric
Metrics details

Subjects

Abstract

S-Palmitoylation is a reversible post-translational modification that tunes the localization, stability, and function of an impressive array of proteins including ion channels, G-proteins, and synaptic proteins. Indeed, altered protein palmitoylation is linked to various human diseases including cancers, neurodevelopmental and neurodegenerative diseases. As such, strategies to selectively manipulate protein palmitoylation with enhanced temporal and subcellular precision are sought after to both delineate physiological functions and as potential therapeutics. Here, we develop chemogenetically and optogenetically inducible engineered depalmitoylases to manipulate the palmitoylation status of target proteins. We demonstrate that this strategy is programmable allowing selective depalmitoylation in specific organelles, triggered by cell-signaling events, and of individual protein complexes. Application of this methodology revealed bidirectional tuning of neuronal excitability by distinct depalmitoylases. Overall, this strategy represents a versatile and powerful method for manipulating protein palmitoylation in live cells, providing insights into their regulation in distinct physiological contexts.

Mechanisms and functions of protein S-acylation

Article 14 February 2024

Identification and prognostic biomarkers among ZDHHC4/12/18/24, and APT2 in lung adenocarcinoma

Article Open access 04 January 2024

Transcriptional and morphological profiling of parvalbumin interneuron subpopulations in the mouse hippocampus

Article Open access 04 January 2021

Introduction

S-palmitoylation or S-acylation is a biologically consequential post-translational modification essential for fine-tuning protein assembly, localization, and function^1,2,3. At the molecular level, palmitoylation involves the attachment of a 16-carbon palmitic acid to cysteine residues via a thioester bond catalyzed by multi-pass zinc-finger transmembrane palmitoyl acyl transferases known as zDHHC proteins^4,5,6,7,8. Palmitoylation is a reversible process⁹ with the removal of fatty acid groups catalyzed by diverse serine hydrolases such as: (1) acyl-protein thioesterases^10,11 (APT1-2 or Lypla1-2), (2) palmitoyl protein thioesterases¹², and (3) α/β hydrolase domain-containing protein family (Abhd17a-c^13,14,15). The dynamic acylation/deacylation cycle is crucial for maintaining their steady-state subcellular distributions and function¹⁶. Global proteomic analysis has revealed a “palmitoylome” consisting of over 5000 proteins^17,18, including ion channel subunits^{19,20,21,22,23}, G-proteins^24,25, scaffolding proteins²⁶, and kinases²⁷, that impact a range of biological processes. Indeed, altered protein palmitoylation and/or mutations in enzymes that mediate palmitoylation are linked to diverse human diseases including cancers²⁸, viral infections²⁹, diabetes³⁰, irritable bowel syndrome³¹, cardiac disease³², neurodevelopmental³³ and neurodegenerative³⁴ diseases. Consequently, selective manipulation of protein palmitoylation is highly desired both to delineate its complex physiology and as potential therapies³⁵. In this regard, although acyltransferases and depalmitoylases show some preference for modifying targets or target sites^22,36, multiple depalmitoylases often act on a protein¹⁴. The prevailing strategy to manipulate palmitoylation involves small molecule inhibitors or knockdown of specific palmitoyl acyl transferases or depalmitoylases. However, several of these compounds have varied selectivity and may lead to cytotoxicity, limiting their utility^{37,38,39,40,41}. Despite these important advances, several key challenges persist. First, to ensure proper subcellular trafficking of palmitoylated proteins, dynamic acylation, and deacylation are thought to be precisely orchestrated, with enzymes differentially localized to distinct subcellular locations¹⁶. Therefore, manipulating protein palmitoylation with subcellular precision would be advantageous¹¹. Second, selective manipulation of palmitoylation of specific proteins is essential to avoid off-target effects and cytotoxicity³⁵. However, given the relative promiscuity of zDHHCs³⁶ and depalmitoylases¹¹, engineering target specificity has remained challenging. These gaps hinder an in-depth understanding of complex mechanisms by which palmitoylation tunes protein function in physiology and for development of targeted therapeutic strategy.

To address these limitations, we here develop a toolkit of engineered depalmitoylases that may be inducibly activated, subcellularly regulated, and with substrate specificity. To do so, we modify the Abhd17 family of depalmitoylases to be chemically activated by the heterodimerization of FK506 binding protein (FKBP) and FKBP rapamycin binding domain (FRB)^42,43 or optically activated by heterodimerization of photoactivated enhanced magnets⁴⁴. We employ this approach to manipulate palmitoylation of a diverse family of proteins including (1) Ca_Vβ_2A, an obligatory subunit of Ca_V1/2 channels⁴⁵, (2) KChip2, a regulator of Kv4 channels, implicated as a seizure susceptibility gene⁴⁶ and in association with cardiac arrhythmias⁴⁷, (3) Lyn, a Src family tyrosine kinase implicated in various cellular processes⁴⁸, (4) NRas, a small G-protein and a commonly mutated oncogene in human cancer⁴⁹, and (5) PSD95, a synaptic scaffolding protein²⁶. We show that this approach can be leveraged for depalmitoylation of proteins from specific subcellular locations, targeted depalmitoylation of individual protein complexes, and inducible depalmitoylation in response to cell signaling. Altogether, this platform represents a powerful methodology to tune protein palmitoylation.

Results

Design of a chemogenetically-activated depalmitoylase

To rationally design a chemogenetic depalmitoylase, we leveraged the recently identified Abhd17 family of acyl-protein thioesterases. This class of enzymes contain two key domains: (1) a carboxy-terminal (CT) α/β hydrolase domain that forms the catalytic domain responsible for hydrolyzing palmitoyl groups from target proteins, and (2) an amino-terminal (NT) region that is itself palmitoylated and recruits the enzyme to relevant membrane domains^13,50 (Fig. 1a). Functionally, either mutations in the catalytic domain or deletions of the NT abolish Abhd17-mediated depalmitoylation^13,14. Given this molecular architecture, we hypothesized that the two functional domains may be split, rendering the enzyme basally inactive (Fig. 1b). The holoenzyme may then be reconstituted using a chemically inducible dimerization system (Fig. 1b), thereby activating protein depalmitoylation. We utilized the rapid and irreversible rapamycin-induced dimerization of FKBP and FRB, which has been used for various cell biological applications^51,52,53,54. We termed this overall strategy: FKBP/FRB-dimerization activated chemogenetic excision of palmitoylation by Abhd17c or FacePalm-17C. AlphaFold⁵⁵ predicted structures of Abhd17_A-Abhd17_C showed high structural similarity throughout the enzyme, with some divergence immediately following the amino-terminus, suggesting that this region may be well suited for splitting the holo-enzyme (Fig. 1c, Supplementary Fig. 1a-c). We attached FRB to an N-terminal fragment and cyan fluorophore yielding FacePalm-17C_NT and FKBP to the CT of Abhd17_C and mcherry yielding FacePalm-17C_CT (Fig. 1b, c). Importantly, previous studies have shown that membrane localization of Abhd17 catalytic subunit is insufficient to restore its ability to robustly depalmitoylate target proteins¹⁴. Here, FacePalm-17C instead relies on reconstituting the holoenzyme using the chemically induced dimerizer.

**Fig. 1: Design of functional validation of a chemogenetic depalmitoylase.**

To test the functionality of FacePalm, we considered the dually palmitoylated Ca_Vβ_2A subunit, which allows for its segregation to the inner leaflet of the plasma membrane. We chose this target given its physiological role in tuning Ca_V channel dynamics. We tagged Ca_Vβ_2A with a yellow fluorescent protein (YFP) to monitor its localization using confocal microscopy. When expressed alone, Ca_Vβ_2A is enriched at the plasma membrane consistent with high levels of basal palmitoylation (Fig. 1d; Supplementary Fig. 1d). Co-expression of wild-type Abhd17_C but not catalytically inactive Abdh17c S/A mutant results in redistribution of Ca_Vβ_2A throughout the cytoplasm (Fig. 1d; Supplementary Fig. 1e, f), consistent with its depalmitoylation and separation from the plasma membrane. In like manner, Abhd17_A and Abhd17_B also result in cytoplasmic localization of Ca_Vβ_2A (Supplementary Fig. 1e, f). These findings suggest that palmitoylation of Ca_Vβ_2A is tuned by the Abhd17 class of enzymes. We next considered changes in Ca_Vβ_2A localization with our engineered depalmitoylases, FacePalm-17C (Fig. 1e). At baseline, we found that Ca_Vβ_2A largely preserved its membrane localization even in the presence of FacePalm-17C (Fig. 1e), consistent with low basal activity of the split enzyme. Addition of rapamycin resulted in the rapid recruitment of the FacePalm17C_CT (red) to the FacePalm17C_NT (blue) fragment (Fig. 1e). The time course of change in FacePalm17C_CT localization was quantified by measuring Pearson’s correlation between the FacePalm17C_CT and the FacePalm17C_NT fragments (Fig. 1f). Subsequently, we observed a gradual cytoplasmic accumulation of Ca_Vβ_2A (Fig. 1e bottom row). The time course of change in Ca_Vβ_2A localization following activation of FacePalm-17C by rapamycin was quantified by measuring changes in cytosolic fluorescence normalized to initial cytosolic fluorescence (Fig. 1g). To further validate this metric, we monitored co-localization with a fluorescent membrane marker (CellMask) (Supplementary Fig. 2a-c) by quantifying Pearson’s correlation (r_mem). Indeed, rapamycin addition resulted in a decrease in r_mem demonstrating reduced membrane localization (Supplementary Fig. 2d). In addition, the time course for decrease in r_mem closely matched the increase in cytosolic fluorescence (ΔF/F₀) with no significant differences in the estimated time constants (Supplementary Fig. 2e-g). This change in cytosolic fluorescence of Ca_Vβ_2A was absent in (1) cells lacking FacePalm expression (Supplementary Fig. 2h), (2) when co-expressed with a catalytically inactive FacePalm-17C S/A mutant (Fig. 1g; Supplementary Fig. 2i), and (3) in the presence of Palmostatin B, a pharmacological antagonist of Abhd17 (Supplementary Fig. 2j, k). These results suggest that the catalytic activity of FacePalm-17C is required to elicit a dynamic change in Ca_Vβ_2A localization_. Furthermore, to exclude the possibility that the changes in Ca_Vβ_2A localization observed with FacePalm is due to a failure of newly synthesized proteins to be palmitoylated, we used cycloheximide to block protein synthesis. Even still, activation of FacePalm-17C elicited a robust increase in cytosolic fluorescence of Ca_Vβ_2A, suggesting that the engineered depalmitoylases can act on existing proteins (Supplementary Fig. 2l, m). To further corroborate whether changes in Ca_Vβ_2A localization correspond to altered palmitoylation, we used an Acyl-RAC assay to directly detect the palmitoylated fraction. In cells co-transfected with YFP-tagged Ca_Vβ_2A and FacePalm-17C, incubation with rapamycin yielded a ~60% reduction in palmitoylation (Fig. 1h, i). Taken together, these findings demonstrate that FacePalm serves as a chemogenetic approach to manipulate protein palmitoylation in live cells.

We probed whether FacePalm can evoke dynamic changes in the Ca²⁺ channel function through depalmitoylation of the Ca_Vβ_2A subunit. We considered the Ca_V2.2 channel which is functionally important for initiating presynaptic vesicle release in both the central and peripheral nervous system. We co-transfected HEK293 cells with Ca_V2.2 and auxiliary subunits α₂δ₁ and Ca_Vβ_2A subunits. Whole-cell voltage-clamp recordings showed rapid activation and minimal inactivation of Ca_V2.2 in the presence of Ca_Vβ_2A (Fig. 1j, top). Addition of rapamycin yielded no change in current dynamics (Fig. 1j, top). When FacePalm was co-transfected, the baseline currents remained unperturbed (Fig. 1j, bottom). However, addition of rapamycin yielded a marked increase in inactivation (Fig. 1j, bottom), consistent with previous studies where palmitoylation was disabled through mutagenesis of the Ca_Vβ_2A subunit⁵⁶. Population r₈₀₀ data measuring the fraction of peak current remaining following 800 ms of depolarization further confirmed this trend (Fig. 1k). In all, these findings suggest that FacePalm is a powerful approach to dynamically tune protein palmitoylation and to delineate molecular and physiological consequences.

FacePalm is a generalizable strategy for chemogenetic protein depalmitoylation

Given that both Abhd17a and Abhd17b share a similar architecture as Abhd17c, we sought to determine whether the FacePalm approach may be generalizable. We split Abhd17a and Abhd17b at analogous positions as Abhd17c and attached FRB and FKBP to the NT and CT regions, yielding FacePalm-17A and FacePalm-17B. Accordingly, we probed whether rapamycin activation of FacePalm-17A altered palmitoylation of Ca_Vβ_2A using an acyl-RAC assay (Fig. 2a, b). As with FacePalm-17C, we observed a strong reduction in the palmitoylated fraction of Ca_Vβ_2A (Fig. 2a, b). We next examined the effect of rapamycin-induced activation of FacePalm-17A on Ca_Vβ_2A localization. We found that with rapamycin activation, FacePalm-17A evoked a substantial increase in cytosolic fluorescence of Ca_Vβ_2A, albeit blunted compared to FacePalm-17C (Fig. 2c, d). Similar analysis confirmed functionality of FacePalm-17B in depalmitoylating (Fig. 2e, f) and altering localization of Ca_Vβ_2A (Fig. 2g, h). These findings illustrate the generality of this approach for inducible activation of Abhd17 enzymes.

**Fig. 2: FacePalm is a generalizable strategy for inducible protein depalmitoylation.**

More broadly, a staggering array of proteins with widely varying physiological functions are known to be palmitoylated. Here, we considered whether FacePalm might alter localization of three unrelated proteins whose palmitoylation is thought to be important for its function: (1) Kchip2, a Ca²⁺-binding protein that regulates K_V4 ion channels; (2) Lyn, a member of the Src family of tyrosine kinases; and (3) NRAS, a small G-protein involved in cell signal transduction and a proto-oncogene. Given that FacePalm-17C evoked the strongest change in Ca_Vβ_2A localization, we probed its effect on these palmitoylated proteins. For Kchip2 and Lyn, acyl-RAC assays revealed statistically significant reduction in palmitoylation (Kchip2, Fig. 2i-j; Lyn, Fig. 2m, n), and confocal imaging showed increased cytosolic localization of both Kchip2 (Fig. 2k, l) and Lyn (Fig. 2o, p) following activation of FacePalm-17C. In a similar manner, we found that activation of FacePalm-17C resulted in reduced palmitoylation of NRas (Fig. 2q, r) and increased internalization of NRas, likely reflecting its accumulation in the Golgi complex as expected following depalmitoylation. Changes in NRas dynamics were quantified by measuring the total area of intracellular puncta (Fig. 2s, t). Collectively, these results highlight the generality of FacePalm as a toolkit for chemogenetic manipulation of palmitoylation for various proteins.

Optogenetic and reversible manipulation of protein palmitoylation

A key limitation of the FacePalm approach for manipulating protein palmitoylation is poor reversibility, owing to the ultra-high affinity of rapamycin for FKBP and the slow clearance of rapamycin from cells⁵⁷. We, therefore, sought to devise an optogenetic approach, whereby the dimerization of the N- and C-termini of the Abhd17c depalmitoylase may be induced by light illumination. To do so, we utilized the recently engineered ‘enhanced Magnets,’ derived from the Vivid photoreceptor from Neurospora crassa⁴⁴. The enhanced Magnets are composed of protein pairs eMagA and eMagB which rapidly dimerize in the presence of blue-light and reverses in darkness. We attached eMagA to the Abhd17 NT fragment and eMagB to Abhd17 CT catalytic subunit and named the pair: opto-depalm (optogenetically activated depalmitoylases) (Fig. 3a). We monitored changes in localization of mCherry-tagged Ca_Vβ_2A in the presence and absence of opto-depalm. Without opto-depalm, we observed no changes in Ca_Vβ_2A localization following blue light illumination (Fig. 3b, c). With opto-depalm, Ca_Vβ_2A is initially membrane localized. However, with blue light (488 nm) illumination, we observed an increase in cytosolic localization of Ca_Vβ_2A (Fig. 3b, c, blue region), consistent with activation of the optogenetic depalmitoylase. In darkness, Ca_Vβ_2A once again accumulated near the plasma membrane (Fig. 3b, c), demonstrating reversibility. Disabling the catalytic activity of opto-depalm (S/A mutant) abolished light-dependent alterations in localization of Ca_Vβ_2A (Fig. 3b, c). To probe whether light-activation of optopalm changes Ca_Vβ_2A palmitoylation, we undertook Acyl-RAC assay. At baseline, a significant proportion of Ca_Vβ_2A is palmitoylated. Following 60-min of optopalm activation, we observed a transient reduction in Ca_Vβ_2A palmitoylation, which is then reversed in darkness (Fig. 3d, e). These findings illustrate the suitability of optogenetically-activated depalmitoylases for reversible manipulation of protein palmitoylation.

**Fig. 3: An optogenetic actuator allows reversible manipulation of protein palmitoylation.**

Organelle-specific depalmitoylation by engineering FacePalm

Subcellular localization has emerged as a key contributor for specificity and for determining the biological consequences of protein palmitoylation¹³. Strategies to preferentially manipulate protein palmitoylation in distinct intracellular compartments are sought after. As FacePalm is genetically encoded, this approach is amenable for targeted depalmitoylation in distinct subcellular localizations. We reasoned that the N-terminal localizing component of FacePalm-17C (FacePalm-17C_NT) may be redirected to various subcellular locations via well-established subcellular targeting sequences (Fig. 4a). The addition of rapamycin would then recruit FacePalm-17C_CT, i.e., the catalytic domain of Abhd17c, to the intended organelle, allowing depalmitoylation to proceed preferentially from this compartment. We generated a library of FacePalm-17C_NT variants containing targeting sequences^42,58 to the (1) ER, (2) Golgi, (3) PM, (4) early endosomes, and (5) late endosomes (Fig. 4a). Of note, with this strategy, the FacePalm-17C_CT fragment which contains the catalytic subunit remains the same. We systematically co-expressed various subcellularly-targeted FacePalm_NT with FacePalm-17C_CT and assessed changes in membrane localization of Ca_Vβ_2A.

**Fig. 4: Organelle-targeted depalmitoylation by FacePalm.**

Upon rapamycin activation, non-targeted FacePalm-17C_CT is broadly recruited to the PM and endosomal compartments. The degree of colocalization with each subcellular domain is quantified by determining Pearson’s correlation with a marker for respective subcellular locus (Supplementary Fig. 3a, d, g, j, m). These changes are largely consistent with the localization of Abhd17 holoenzyme as established by previous studies¹³. By contrast, with subcellularly targeted FacePalm, rapamycin activation results in strong recruitment of FacePalm-17C_CT to desired compartments (Supplementary Fig. 3b, e, h, k, n), as confirmed by a statistically significant increase in Pearson’s co rrelation with independent markers for each organelle (Supplementary Fig. 3c, f, i.l,o). Functionally, with ER-localized FacePalm, we found essentially no change in normalized cytosolic fluorescence of Ca_Vβ_2A (Fig. 4b, c, l)_, despite robust recruitment of the FacePalm-17C_CT to this domain. By comparison, Golgi-localized FacePalm evoked a modest increase in normalized cytosolic fluorescence of Ca_Vβ_2A (Fig. 4d, e, l). Interestingly, both PM-localized (Fig. 4f, g) and early endosome-localized (Fig. 4h, i) FacePalm evoked a still larger increase in cytosolic fluorescence (Fig. 4l). However, localizing FacePalm to late endosomes resulted in a modest reduction in cytosolic fluorescence compared to early-endosome localized FacePalm (Fig. 4j–l). The varying degree of increase in cytosolic fluorescence of Ca_Vβ_2A upon targeting FacePalm to distinct compartments points to heterogeneity in the efficacy of depalmitoylation from various intracellular organelles. In all, these results highlight the spatial and temporal precision with which FacePalm can manipulate protein palmitoylation subcellularly.

Cell-signaling programmable activation of depalmitoylation

From a pathophysiological perspective, manipulation of protein palmitoylation has emerged as an attractive therapeutic possibility³⁵. To this end, the ability to conditionally depalmitoylate proteins in response to a cell signaling event that may be upregulated in pathological settings is desirable, yet challenging. A prominent example involves activating mutations in NRas that are commonly observed in highly aggressive cancers associated with high mortality⁴⁹. Constitutive activation of NRas results in upregulation of MAPK and PI3K pathways that cause sustained proliferation and tumor progression⁴⁹. In this regard, palmitoylation of NRas is important for its localization and downstream signaling, and inhibitors of the NRas palmitoylation cycle have been proposed as potential therapies (Fig. 5a). Accordingly, we sought to engineer a synthetic feedback loop that activates a depalmitoylase in response to Erk, a downstream effector of NRas, and, in so doing, inhibit NRas in a context-dependent manner (Fig. 5a). To do so, we leveraged the design principle of FRET based ERK biosensors⁵⁹, and engineered the N-terminus of Abhd17_C with a WW phospho-binding domain and the C-terminus of Abhd17C with an Erk substrate from cdc24C along with an FQFP ERK-docking site (Fig. 5b). We termed this bipartite system: dePalmitoylation, Erk-activated (dePalm-Er). With low Erk activity, the two components will remain unbound. However, Erk activation will result in phosphorylation of the Erk substrate in the C-terminal component of dePalm-Er, which in turn will bind the WW phospho-binding domain and thereby reconstitute the holo-enzyme. To probe the functionality of this engineered depalmitoylase, we undertook confocal imaging to monitor the subcellular localization of both wild-type and Q61K mutant NRas that is prevalent in melanomas and adenocarcinomas⁴⁹. As depalmitoylated NRas accumulates in the Golgi, we monitored colocalization of YFP-tagged NRas with a red Golgi marker (mCherry-golgin). At baseline, both wild-type (Fig. 5c, top) and Q61K mutant NRas (Fig. 5c, bottom) are largely localized to the plasma membrane with minimal overlap with the Golgi compartment. (Fig. 5c, left top and bottom). Further quantification revealed low Pearson’s correlation between red and yellow fluorescence at baseline (Fig. 5d). Over-expression of Abhd17c results in internalization of both wild-type and mutant NRas (Fig. 5c, middle top), as evident from increased co-localization with the Golgi marker (Fig. 5d). Co-expression of dePalm-Er with wild-type NRas largely preserved the robust membrane localization of NRas as evident from both confocal images (Fig. 5c, top right), and minimal change in the Pearson’s correlation (Fig. 5d). In sharp contrast, co-expression of the activating Q61K mutant NRas with dePalm-Er resulted in increased Golgi localization (Fig. 5c, bottom right). Population data further confirmed this trend (Fig. 5d). These results suggest that dePalm-Er allows selective activation of the Abhd17_C depalmitoylase in response to elevated or constitutive Erk activity. These findings illustrate the versatility of the genetically encoded approach for programmable activation of protein depalmitoylation in response to a physiological stimulus.

**Fig. 5: Engineering a depalmitoylation feedback loop for NRas.**

Engineering FacePalm to target specific protein complexes

Depalmitoylating enzymes such as Abhd17 are often promiscuous and act on a wide range of cellular targets. Broad activation of FacePalm could contribute to global changes in the palmitoylome, potentially resulting in off-target effects and complex physiological outcomes. We considered whether FacePalm may be engineered to selectively target individual protein complexes. To do so, we reasoned that we could attach to FacePalm-17C_NT a selective nanobody that binds to a specific target. The addition of rapamycin will then recruit FacePalm-17C_CT to the target complex, resulting in its depalmitoylation (Fig. 6a). We termed this strategy FacePalm individually targeted or FacePalm-it.

**Fig. 6: Targeted depalmitoylation of specific protein complexes.**

To test whether this design could enable target specificity, we considered Kchip2-tagged to YFP and Ca_Vβ_2A tagged to a CFP. Both proteins are depalmitoylated by FacePalm, resulting in increased cytosolic localization (Figs. 1e–i, 2i-l). To selectively tune palmitoylation of Kchip2, we attached a previously published YFP-targeting nanobody⁶⁰ to FacePalm-17C_NT and co-expressed it with FacePalm-17C_CT. Rapamycin activation of FacePalm-it resulted in a time-dependent increase in cytosolic fluorescence of Kchip2, but not Ca_Vβ_2A (Fig. 6c, d), consistent with selective regulation of Kchip2. We next considered whether we can selectively depalmitoylate Ca_Vβ_2A while sparing Kchip2. To do so, we modified FacePalm-17C_NT with a selective high affinity nanobody for Ca_Vβ_2A (nb.F3)⁶¹ and co-expressed it with FacePalm-17C_CT (Fig. 6e, f). Here, Kchip2 was tagged with YFP, while Ca_Vβ_2A was tagged with an mCherry fluorescent protein for convenience. Application of rapamycin resulted in an increase in normalized cytosolic fluorescence of Ca_Vβ_2A without a corresponding increase in the cytosolic fluorescence of Kchip2 (Fig. 6f), consistent with targeted manipulation of Ca_Vβ_2A. Taken together, these findings illustrate the potential selectivity of FacePalm-it for modulating protein depalmitoylation.

While FacePalm-it allows inducible depalmitoylation of targeted proteins, in some cases, it may be advantageous to constitutively depalmitoylate proteins. We reasoned that directly attaching the catalytic subunit of Abhd17c to a target-selective nanobody may enable the constitutive depalmitoylation of the desired target protein (Fig. 6b). To test this possibility, we attached the YFP-binding nanobody to Abhd17c catalytic domain (Supplementary Fig. 4a), yielding Nb(YFP)-17c_cat. Co-expression of this fusion protein with YFP-tagged Ca_Vβ_2A, KChip2, Lyn, NRas, and Gαq resulted in decreased membrane localization of these proteins (Supplementary Fig. 4b-f). Furthermore, to probe whether this approach confers target selectivity we co-expressed YFP-tagged Kchip2 and mcherry-tagged Ca_Vβ_2A along with either (i) YFP-targeting nanobody tethered to Abhd17C_CT (Nb(YFP)-17C_CT) or (ii) Ca_Vβ_2A-targeting nanobody tethered to Abhd17C_CT (Nb(Ca_Vβ_2A)-17C_CT). At baseline, both Kchip2 and Ca_Vβ_2A exhibit strong membrane localization (Fig. 6g). However, with Nb(YFP)-17C_CT, Kchip2 is largely cytosolic while Ca_Vβ_2A remains membrane localized (Fig. 6h). By contrast, with Nb(Ca_Vβ_2A)-17C_CT, Kchip2 remains membrane localized while Ca_Vβ_2A is cytosolic (Fig. 6i). We quantified these changes by measuring the ratio of cytosolic to total fluorescence of either Ca_Vβ_2A or Kchip2 (F̅_cyt(Ca_Vβ_2A) and F̅_cyt(Kchip2)) and calculating the ratio ζ = F̅_cyt(Ca_Vβ_2A)/F̅_cyt(Kchip2) (Fig. 6j). Taken together, this overall approach may be generalized to either constitutively or inducibly depalmitoylate any target by developing selective nanobodies or other synthetic binding proteins.

FacePalm permits targeted tuning of neuronal function

Protein palmitoylation is thought to be critical for neuronal function and regulation. Indeed, at the molecular level, a wide range of neuronal ion channels and synaptic proteins undergo palmitoylation that tunes their localization and function¹⁹. We sought to determine whether the FacePalm approach may be utilized to tune neuronal function. Accordingly, we deployed all three FacePalm variants individually in cultured rat hippocampal neurons using viral transduction. As palmitoylation of PSD95 is well established to be important for its clustering and localization, we first probed the effect of FacePalm variants on PSD95 using immunocytochemistry and confocal microscopy. Without FacePalm, the addition of rapamycin yielded no appreciable change in spine density (Supplementary Fig. 5a, b). However, rapamycin-induced activation of all three FacePalm variants resulted in a strong reduction in PSD95 puncta density along neuronal process (Fig. 7a-f). To ensure that changes in PSD95 puncta density indeed stem from enzymatic depalmitoylation, we disabled the catalytic activity of FacePalm-17A through FacePalm-17C by introducing the S/A mutation in the catalytic triad. Reassuringly, the addition of rapamycin resulted in no apparent changes in PSD95 clustering in all three cases (Supplementary Fig 6a-f). These results confirm the functionality of FacePalm in neurons and highlight its potential utility to manipulate protein palmitoylation in a native setting.

**Fig. 7: Bidirectional modulation of neuronal function by FacePalm variants.**

Thus affirmed, we probed the effect of FacePalm variants on neuronal excitability using current clamp recordings. We measured the action potential (AP) firing rate evoked in response to a 1 s pulse to a family of current amplitudes (I_inj). In cells expressing FacePalm-17A, we found no change in AP firing rate with rapamycin activation (Fig. 7g, h), similar to cells lacking FacePalm (Supplementary Fig. 5c, d). When FacePalm-17B is overexpressed, however, we found a marked increase in AP firing rate in the presence of rapamycin (Fig. 7i, j), suggesting that FacePalm-17B can upregulate neuron AP firing. In sharp contrast, when FacePalm-17C is activated, we observed an overall decrease in AP firing rate (Fig. 7k, l). To ensure that these changes in neuronal AP firing properties are due to altered palmitoylation, we expressed both catalytically-inactive FacePalm-17B S/A mutant and FacePalm-17C S/A mutant. Reassuringly, we found no appreciable change in neuronal AP firing rates in both cases (Supplementary Fig. 6g–j). As neuronal action potentials are coordinated by multiple ion channel complexes, changes in neuronal firing properties here could reflect altered regulation of one or more ion channel complexes. Indeed, several ion channel pore-forming subunits as well as auxiliary proteins have been shown to be palmitoylated. As a result, identifying precise targets regulated by FacePalm variants is challenging.

Given its potentially broad impact a wide range of proteins, targeted depalmitoylation to tune specific aspects of neuronal function would be advantageous. Accordingly, we sought to determine whether inducibly localizing a depalmitoylase to PSD95 may reduce its clustering without impacting AP firing properties which presumably reflect the combined effect of various ion channels. To this end, PSD95 binding fibronectin intrabodies generated by mRNA display (FingR) have been developed for imaging and manipulating PSD95 and postsynaptic function in neurons⁶². Following our strategy with FacePalm-it, we attached PSD95-FingR to FacePalm_N and bicistronically expressed FacePalm_C. In addition, mcherry was used as an expression marker to identify cells expressing the two FacePalm components. At baseline, we observed robust clustering of PSD95 quantified as the puncta density along neuronal processes (Fig. 7m). In the presence of rapamycin, PSD95 clustering is significantly reduced, confirming the functionality of the targeted depalmitoylases (Fig. 7n). To probe changes in neuronal firing properties, we undertook current clamp recordings of hippocampal neurons expressing the targeted depalmitoylases. Reassuringly, we found minimal changes in AP firing properties following rapamycin activation of the targeted depalmitoylases (Fig. 7o, p).

Taken together, these findings highlight the utility of FacePalm to manipulate endogenously palmitoylated proteins and to delineate its physiological functions. Furthermore, targeted depalmitoylation may provide an alternate avenue to selectively tune specific aspects of neuronal function.

Discussion

Strategies to dynamically manipulate protein palmitoylation are desired for both identifying physiological regulatory mechanisms and as potential therapeutics. However, the relative promiscuity of (de)acylating enzymes has made this a challenging endeavor. We, here, developed a toolkit of chemically/optically activated depalmitoylases by engineering the Abhd17 family of enzymes. Specifically, we split Abhd17 enzyme into two fragments: (1) FacePalm_NT containing the Abhd17 N-terminal region and (2) FacePalm_CT containing the Abhd17 catalytic domain. We utilized both a chemically induced dimerization system (Fkbp/Frb activated by rapamycin) and an optogenetic heterodimerizer (‘enhanced magnets’) to inducibly reconstitute the holo-enzyme to increase its activity. This genetically encoded strategy is generalizable across the Abhd17 family and can be used to inducibly depalmitoylate various target proteins including ion channel subunits, kinases, and G-proteins. This approach is advantageous as it can be engineered to enable depalmitoylation in specific cell-types or subcellular locations with target specificity. Beyond this, we demonstrate that this strategy could be leveraged to develop depalmitoylation feedback circuits that alter protein localization in response to specific cell signaling events. These findings highlight the versatility of engineered depalmitoylases to modify protein function and provides an avenue to study dynamic regulation of protein function by palmitoylation.

Given the emerging role of palmitoylation in diverse human diseases, harnessing (de)palmitoylation as a therapeutic modality has garnered considerable interest^35,63. The APT1 inhibitor ML348 has been shown to reverse neuronal function and behavioral phenotype in Huntington disease models³⁴. Recently developed Abhd17 small molecule inhibitor ABD957 has been used to inhibit NRas signaling and growth of NRAS-mutant human acute myeloid leukemia cells⁴¹. Here, we developed FacePalm-it as an orthogonal avenue for targeted depalmitoylation of proteins by inducing proximity of a depalmitoylase using a high-affinity nanobody, allowing selective manipulation of the desired protein complex. Application of this approach in cultured neurons revealed targeted regulation of PSD95. This strategy parallels recent developments in using induced proximity of enzymes to manipulate various post-translational modifications of proteins including ubiquitination and phosphorylation. With the increased availability of selective nanobodies targeting various proteins and streamlined approaches to screen for synthetic nanobodies, it may be feasible to develop a toolkit of engineered depalmitoylases targeting a wide array of proteins. Moreover, it is also possible to constitutively promote depalmitoylation of a target by directly attaching a selective nanobody to the catalytic domain of Abhd17. This overall molecular design may form the basis for a distinct class of palmitoylation modulators with potential therapeutic utility.

Ras is the most frequently mutated gene family in cancers^49,64. However, developing allele-specific Ras inhibitors has been challenging. As Ras is active when localized to cell membrane, one approach has been to develop inhibitors of Ras lipid modifications such as farnesyl transferase inhibitors⁴⁹. Both NRas and HRas undergo dynamic palmitoylation/depalmitoylation suggesting that palmitoylation inhibitors may have potential utility²⁴. Here, we engineered dePalm-er which allows depalmitoylation of proteins in response to Erk phosphorylation. Mutant NRas upregulate the Ras-Mek-Erk pathway leading to increased cell proliferation and differentiation⁶⁵. Thus, dePalm-er could provide a strategy to disrupt the pathological overactivation of the Erk pathway. Of note, in our experiments, dePalm-er, although activated by Erk could still target a wide range of proteins. It may be feasible to engineer dePalm-er further by introducing selective nanobodies targeting NRas. More broadly, a similar strategy could be employed to engineer depalmitoylases activated by a wide range of cellular signals, such as for e.g., Ca²⁺ or ATP etc., which may be useful to tune protein function in response to neuronal/cardiac hyperactivity or altered metabolic state.

In neurons, we found that the expression of distinct FacePalm variants differentially tuned neuronal activity. Specifically, FacePalm-17B increases AP firing rate while FacePalm-17c evoked a decrease in AP firing rate, potentially as a result of depolarization block. Although our study does not identify an exact molecular mechanism underlying this change, altered AP properties likely reflect depalmitoylation of one or more targets as the catalytically inactivated FacePalm-17B and FacePalm-17C fail to appreciably perturb AP firing properties. One possibility is that although Abhd17 variants are homologous and have overlapping targets, they may also differentially impact disparate proteins leading to distinct outcomes. Consistent with this possibility, recent studies have shown that a BK channel splice variant is depalmitoylated by both Abhd17a and Abhd17c but is insensitive to Abhd17b^22,66. Even so, determining the exact mechanisms that contribute to differential regulation of neuronal activity is challenging as a wide range of ion channel complexes are regulated by palmitoylation^63,64. These include not only the pore-forming subunit but also auxiliary subunits and regulatory proteins. However, many of these processes are isoform-specific and further regulated by alternative splicing resulting in heterogeneity. Furthermore, for many proteins, the impact of palmitoylation on channel function is not established and could include both altered gating and/or trafficking. Future studies that utilize global proteomic analysis along with in depth electrophysiological studies are essential to dissect the various possibilities.

A few limitations merit further attention. First, engineered depalmitoylases critically modify Abhd17 enzymes by heterodimerization motifs. As these motifs are often large, this modification may itself impact the kinetics and selectivity of these proteins. Therefore, differences observed with FacePalm variants may not necessarily reflect those of the parent enzyme. Second, only a subset of palmitate modifications are thought to be dynamic¹⁷. Previous studies using ABD957, a selective Abhd17 antagonist, has shown that these enzymes target a restricted set of dynamically palmitoylated proteins. Future proteomic studies are necessary to determine the full extent of proteins targeted by FacePalm.

In all, we here developed a toolkit of engineered depalmitoylases as a versatile approach for inducible and selective manipulation of protein palmitoylation. These findings set the stage to unravel complex physiologically relevant regulatory mechanisms and to harness protein palmitoylation as a therapeutic modality for a wide range of diseases.

Methods

Ethical statement

This study uses E18 Sprague Dawley Rat hippocampal neurons isolated from tissue purchased from BrainBits LLC(TransnetYX). The tissues was extracted by BrainBits under animal protocol (#233-08-013) approved by the Southern Illinois University School of Medicine Laboratory Animal Care and Use Committee.

Molecular biology and virus construction

The Ca_Vβ_2a subunit (M80505.1) was tagged with YFP or mCherry at the C-Terminus with GSG linker in the N3 vector. NRas (NCBI accession #NM_002524.5) and Q61K mutant of NRas were synthesized from Twist Biosciences with cutting sites HindIII and XbaI and inserted into vector containing a N-Terminus-Venus. Kchip2 (XM_048505192.1) were synthesized from Twist Biosciences as DNA fragments with cutting sites NheI and SalI and C-terminal YFP. Lyn was a gift from Jesse Boehm & William Hahn & David Root (Addgene plasmid # 82215). G protein alpha-q-GFP was a gift from Catherine Berlot (Addgene plasmid # 66080). N-terminus of Abhd17 (Abhd17a: NM 145421; Abhd17b: NM146096, Abhd17c: NM 133722) (1-55 amino acid sequence) tagged with an FRB was cloned into a plasmid using cutting sites EcoRI and BamHI upstream of a Cerulean fluorophore. FKBP with C-Terminus of Abhd17 (residues 56-end containing the catalytic site) was cloned downstream of an mCherry fluorophore using HindIII and BamHI cutting sites. For organelle targeted Facepalm, we modified the N-terminal component of FacePalm-17C (which contains the N-terminus of Abhd17C, FRB, and cerulean) by attaching to the C-terminus various targeting sequences to intracellular domains: for ER-targeted: cytochrome b5 (residues 100-134); for golgi-targeted: giantin (residue 3131-3159); for PM-targeted: CaaX motif from KRas-4b (183-199); For early-endosome-targeted: Rab5; For late endosome-targeted: Rab7. These sequences were synthesized as gene fragments and cloned into pcDNA3 with cutting sites: EcoRI and BamHI. Similarly, N-terminus of Abhd17c tagged with an FRB and either GFP targeted nanobody or nb.F3 targeting Ca_Vβ2A was inserted into a pcDNA3.1 vector with HindIII and EcoRI. For PSD95-targered FacePalm-it, we used gene synthesis (Vector Builder) to generate a lentiviral vector that encodes Abhd17C_NT tagged with FRB and a PSD95-binding FingR⁶² as well as Abhd17C_CT tagged with FKBP, separated by a P2A sequence for bicistronic expression. In addition, mCherry was used as an expression marker with an EF1 promoter. The lentivirus was packaged by Vector Builder. For Erk-activated dePalmer, we used gene synthesis (Twist bioscience) to generate a fragment containing Abhd17C-NT tagged with a proline-directed WW phosphobinding domain⁵⁹ followed by a P2A sequence, Erk substrate with ERK docking domain⁵⁹ tagged to Abhd17C_CT flanked by HindIII and EcoRI. The fragment was cloned in a pcDNA expression vector. All viral vectors were generated by gene synthesis and packaged by VectorBuilder. For all of these: Abhd17 NT- tagged FRB and FKBP-tagged Abhd17 CT, separated by a P2A sequence for bicistronic expression, were designed and packaged into lentiviral (FacePalm-17A, FacePalm-17B) or adenoviral (FacePalm-17c) vectors containing an mCherry driven by an EF1 promoter. To disable catalytic activity of FacePalm variants, we introduced the following mutations in the catalytic triads based on sequences of Abhd17A-C: for FacePalm-17C, S211A based on Abhd17C; for FacePalm-17B, S170A based on Abhd17B; for FacePalm-17A, S190A based on Abhd17A. The mutant genes were synthesized and packaged by VectorBuilder. For generation of Opto-depalm and Opto-depalmS/A, we used gene synthesis (Twist Biosciences) to attach eMagA to the C-terminus of Abhd17C_NT, and eMagB to the N-terminus of Abhd17C_CT. The two components were bicistronically expressed with Abhd17C_NT-eMagA driven by a CMV promoter followed by an IRES sequence and eMagB-Abhd17C_CT.

Cell culture and transfections

Cultures were maintained at 37 ˚C, 5% CO₂, and 95% humidity. HEK293 cells (ATCC, CRL 1573) were maintained in Dulbecco’s Modified Eagle Medium containing 10% FBS, L-Glutamine (2 mM), 1% Penicillin- Streptomycin and Gentamicin (50 µg/ml). For electrophysiology, cells were plated on coverslips and transfected using calcium phosphate method⁶⁷; Ca_V2.2 (0.5–1 µg), Ca_Vβ_2a (0.5–1 µg) and α2δ (0.5–1 µg) and SV40 driven T-antigen (0.5–1 µg) depending on expression and co-expressed with FacePalm (0.4-0.7 µg) depending on expression. For AcylRAC/Western Blot assay, cells were plated on 60 mm dishes, and T-antigen (0.5–1 µg), C_aVβ_2a-YFP (2.5 µg), and either FacePalm-17A, FacePalm17b or Facepalm-17C (1 µg) were transfected using calcium phosphate method. Similarly, cells were plated on 60 mm dishes and T-antigen (0.5–1 µg), Facepalm-17C (1 µg), along with either (1) Venus-NRAS (1 µg), (2) Kchip2-YFP (1 µg), and (3) Lyn-YFP (1 µg). For confocal microscopy, cells were plated in glass-bottomed chambered coverslips. Cells were transfected using polyethyleneimine method of transfection as previously reported⁶⁷; T-antigen (0.2–0.3 µg), C_aVβ_2a (0.5–0.7 µg), Kchip2.2 (0.5–0.7 µg), N-Ras (0.1–0.5 µg), Q61K-NRas (0.1–0.5 µg) and Lyn (0.5–1 µg), co-transfected with Facepalm (0.4–0.7 µg). For the opto-depalm experiments, cells were plated on precoated glass-bottomed Mattek dishes and transfected with Mcherry tagged-C_aVβ_2a (0.5–0.7 µg) and opto-depalm (0.5–0.7 µg) or opto-depalm S/A (0.5–0.7 µg).

Acyl RAC assay/western blotting

Acyl Resin Assisted Capture assay kit was purchased from Badrilla (K010-311). The assay was performed per the manufacturer’s instructions. In brief, cells were collected 48 hrs post-transfection, and free cysteines were blocked for 4 h using a thiol-blocking agent in Buffer A, following which the proteins were precipitated using acetone. The proteins were resuspended in binding buffer and quantified using BCA assay. The same quantity of protein was loaded onto the beads for the comparison conditions. The resuspended protein was treated with the cleavage reagent in the presence of thiopropyl Sepharose beads. The palmitoylated fraction was eluted with the 2X SDS buffer provided, loaded onto 4–12% Bis-Tris gradient gels (NP0321BOX; Thermofisher), and western blotting was performed. The blots were probed with rabbit anti GFP (1:2000) (Genscript, catalog number: A01388-40) or rabbit anti-Mcherry (1:1500) (Cell signaling, catalog number: 43590) overnight. The blots were treated with secondary anti-rabbit HRP (ThermoFisher, catalog number 31460) and developed using chemiluminescence. Data was analyzed using Prism GraphPad 10; an unpaired t-test was used to obtain statistics between no rapamycin and rapamycin groups. Uncropped and unprocessed scans are provided as Source Data file.

Neuronal culture

E18 Sprague Dawley Rat primary hippocampal neuronal culture kit was purchased from BrainBits (Transnetyx, SKU KTSDEHP). Brain tissue was acquired from BrainBits under animal protocol (#233-08-013) approved by the Southern Illinois University School of Medicine Laboratory Animal Care and Use Committee. Cells were cultured as per instructions by BrainBits. Briefly, the hippocampi were digested in papain (2 mg/ml) in Hibernate (-Ca) for 10 min at 30 °C. Following this, the digested tissue was triturated and centrifuged at 200 × g for 1 min at room temperature. The cells were resuspended in Neurobasal A media with B27 and Glutamax with glutamate(25 µM). Cells were plated at 16,000 cells/cm² onto cell culture plates coated with poly-D-lysine. Cells were transferred to maintenance media (Neurobasal A media with B27 and Glutamax without glutamate) 4 days after plating. The media was then changed every 3 days. Cells were transduced on day 7 or day 10 and experiments were performed on day 13–16.

Immunocytochemistry

Neurons were fixed in 4% PFA containing 4% sucrose and 20 mM EGTA. Neurons were then permeabilized with 0.1% Triton X-100 and blocked with 10% FBS. Neurons were incubated with anti-mouse PSD95 (1:500; NeuroMab, SKU 75-028), and anti-rabbit mCherry (1:500, Cell Signaling, #43590) overnight and treated with secondary anti-rabbit Alexa Fluor 488 (Invitrogen, A-11008, 1:1000) and anti-rabbit Alexa Fluor 568 (Invitrogen, A11011, 1:1000) before mounting the coverslips onto glass slides with Anti-FADE mounting medium with DAPI. For organelle colocalization experiments, The HEK293 cells were probed with primary antibodies/stain as follows: for ER, Cytopainter (1:1000; AbCaM catalog number ab139482); for Golgi, RCAS (Cell Signaling, #12290.1:100); for PM, Cell Mask (ThermoFisher: C10046, 1:2500); for early endosome, EEA1 (Cell signaling, #3288, 1:100); for late endosome, Rab9A (Cell signaling, #5118, 1:50). Endoplasmic reticulum (Cytopainter, Abcam 139482, 1:1000) and Plasma Membrane (Cell Mask, ThermoFisher C10045 1:2500) staining using was done as per manufacturer instructions. For others, HEK293 cells were fixed, permeabilized, and blocked. Cells incubated overnight in primary antibody were rinsed and probed with secondary anti-rabbit Alexa Fluor 647 (Invitrogen, A27040, 1:1000) or anti-mouse Alexa Fluor 488 (Invitrogen. A10680, 1:1000) before confocal microscopy.

Confocal microscopy

Time-lapse live microscopy

24–48 h post transfection, HEK293 cells in chamber slides were rinsed with PBS (with Ca²⁺ and Mg²⁺) and Tyrode solution (125 mM NaCl, 2.5 mM KCl, 3 mM CaCl₂, 1 mM MgCl₂, 10 mM HEPES and 30 mM Glucose; pH7.4) was added. The chamber slides were placed on a stage top incubator and maintained at 37 °C and 5% CO₂. Cells were imaged using a 40× lens objective mounted on a Nikon Ti Eclipse inverted microscope equipped with a Yokogawa CSU-X1 confocal spinning disk and an Andor Zyla sCMOS camera. YFP, mCherry, and Cerulean were excited using 488, 568, and 407 lasers, and the gain and laser intensity were maintained across multiple fields, throughout the time-lapse measurements. Two to three fields of view were chosen for each sample and imaged before adding rapamycin (−10 min). Post-rapamycin, cells were imaged every 10 min up to 90–120 min.

Optogenetic depalmitoylation

Cells were moved to 30–32° 6–7 h post-transfection overnight. Cells were protected from light until imaging. Mattek dishes containing cells were mounted on the stage-top incubator. Opto-depalm was activated by a 488 laser every minute for 60 min. Depalmitoylation of Ca_Vβ_2a (Mcherry) was recorded using the 560 laser every 10 minutes. Following this, to probe the reversibility of the optogenetic actuator, the 488 laser was turned off and cells were imaged every 10 min using the 560 laser for 130 min. For acyl-RAC experiments, cells were exposed to blue LED for 60 min and turned off. Samples were collected at 0 min, 60 min, and 180 min and processed.

Fixed cell imaging

Slides containing mounted coverslips were imaged using 100×/60× lens objective, and PSD95, mCherry and DAPI were imaged using the 488, 568 and 407 lasers. Images were acquired as a Z-stack and maximum-intensity projections were generated. Images were analyzed using Fiji (Image J, Dendritic spine counter plugin). For live cell imaging, images were normalized to the background fluorescence and the fluorescence intensity was measured by drawing a region of interest across all the time points. NRAS undergoes a palmitoylation/depalmitoylation that shuttles between plasma membrane and golgi. To analyze NRAS depalmitoylation, the time scale images were converted to 8-bit and the threshold was adjusted to remove the cell outline and generate a mask. Analyze particles was used to calculate the area of the particles for each time scale measurement. For all colocalization analyses, Pearson’s correlation coefficient was calculated for Golgin (mCherry) and NRAS/Q61K (YFP) using the coloc 2 plugin. For neurons, PSD95 staining was quantified by using the dendritic spine counter plugin.

Whole-cell electrophysiology

Whole-cell voltage-clamp recordings for HEK293 were collected at room temperature using an Axopatch 200B amplifier (Axon Instruments). Borosilicate glass pipettes (2–4 Mohm) were pulled with a horizontal puller (P97; Sutter Instruments Co.) and fire-polished (Microforge, Narishige, Tokyo, Japan). Recordings were low-pass filtered at 2 kHz and 70% series resistance and capacitance compensation. For Ca_V2.2 experiments, Internal solutions contained 135 mM CsMeSO₃, 5 mM CsCl₂, 1 mM MgCl₂, 4 mM MgATP, 10 mM HEPES, 10 mM BAPTA, adjusted to 290–295 mOsm with CsMeSO₃ and pH 7.4 with CsOH. The external solutions contained 140 mM TEA-MeSO₃, 10 mM HEPES (pH 7.4), and 5 mM BaCl₂, were adjusted to 300 mOsm with TEA-MeSO₃ and pH 7.4 with TEA-OH. Cells were held at a potential of −80 mV and family test pulses from −80 to +50 mV, with repeat intervals of 20 s were used. For neurons, current clamp experiments were performed using the same experimental setup. Internal solution contained K gluconate (130 mM), EGTA (0.1 mM), MgCl₂ (1 mM), MgATP (2 mM), HEPES (10 mM), NaCl (5 mM), KCl (11 mM), Na2Phosphocreatine (5 mM); pH7.4, adjusted to 290–295 mOsm. And the external solution contained aCSf (NaCl 124 mM, KCl 2.5 mM, NaH₂PO₄ 1.2 mM, NaHCO₃ 24 mM, HEPES 5 mM, glucose 10 mM, CaCl₂ 2 mM, MgCl₂ 1 mM) is adjusted to 300–305 mOsm, pH 7.4. Measurements were made using family current injections from −0.1 to 4.5 pA current. MATLAB (Mathworks) software was used for analysis of AP firing rate.

Statistics and reproducibility

The data provided includes several biological replicates. No statistical method was used to predetermine sample size. No data were excluded from the analyses. The cultured cells were randomized to experimental conditions. Investigators were not blinded to allocation during experiments and outcome assessment.

All statistical analyses were performed using GraphPad Prism 10. For multiple comparisons, we used a one-way analysis of variance (ANOVA) followed by an appropriate multiple comparisons test. For comparisons of two groups, we used unpaired two-tailed t-tests. For comparisons of probability distributions, we used the Kolmogorov-Smirnov test.

Reporting summary

Further information on research design is available in Nature Portfolio Reporting Summary linked to this article.

Data availability

All data is included in the main text or Supplementary Information file. Source data are provided with this paper.

Code availability

Custom data analysis code for analysis of electrophysiological recordings is available online at: https://github.com/manubenjohny/Whole-Cell-CDI⁶⁸.

References

Bijlmakers, M. J. & Marsh, M. The on-off story of protein palmitoylation. Trends Cell Biol. 13, 32–42 (2003).
Article CAS PubMed Google Scholar
Chamberlain, L. H. & Shipston, M. J. The physiology of protein S-acylation. Physiol. Rev. 95, 341–376 (2015).
Article CAS PubMed PubMed Central Google Scholar
Linder, M. E. & Deschenes, R. J. Palmitoylation: policing protein stability and traffic. Nat. Rev. Mol. Cell Biol. 8, 74–84 (2007).
Article CAS PubMed Google Scholar
Rana, M. S. et al. Fatty acyl recognition and transfer by an integral membrane S-acyltransferase. Science 359, eaao6326 (2018).
Article PubMed PubMed Central Google Scholar
Lemonidis, K. et al. The zDHHC family of S-acyltransferases. Biochem Soc. Trans. 43, 217–221 (2015).
Article CAS PubMed Google Scholar
Roth, A. F. et al. Global analysis of protein palmitoylation in yeast. Cell 125, 1003–1013 (2006).
Article CAS PubMed PubMed Central Google Scholar
Huang, K. et al. Huntingtin-interacting protein HIP14 is a palmitoyl transferase involved in palmitoylation and trafficking of multiple neuronal proteins. Neuron 44, 977–986 (2004).
Article CAS PubMed Google Scholar
Greaves, J. et al. Molecular basis of fatty acid selectivity in the zDHHC family of S-acyltransferases revealed by click chemistry. Proc. Natl. Acad. Sci. USA 114, E1365–E1374 (2017).
Article CAS PubMed PubMed Central Google Scholar
Jiang, H. et al. Protein lipidation: occurrence, mechanisms, biological functions, and enabling technologies. Chem. Rev. 118, 919–988 (2018).
Article CAS PubMed PubMed Central Google Scholar
Duncan, J. A. & Gilman, A. G. A cytoplasmic acyl-protein thioesterase that removes palmitate from G protein alpha subunits and p21(RAS). J. Biol. Chem. 273, 15830–15837 (1998).
Article CAS PubMed Google Scholar
Won, S. J., Cheung See Kit, M. & Martin, B. R. Protein depalmitoylases. Crit. Rev. Biochem. Mol. Biol. 53, 83–98 (2018).
Article CAS PubMed Google Scholar
Camp, L. A. & Hofmann, S. L. Purification and properties of a palmitoyl-protein thioesterase that cleaves palmitate from H-Ras. J. Biol. Chem. 268, 22566–22574 (1993).
Article CAS PubMed Google Scholar
Lin, D. T. & Conibear, E. ABHD17 proteins are novel protein depalmitoylases that regulate N-Ras palmitate turnover and subcellular localization. Elife 4, e11306 (2015).
Article PubMed PubMed Central Google Scholar
Yokoi, N. et al. Identification of PSD-95 Depalmitoylating Enzymes. J. Neurosci. 36, 6431–6444 (2016).
Article CAS PubMed PubMed Central Google Scholar
Tortosa, E. et al. Dynamic palmitoylation targets MAP6 to the axon to promote microtubule stabilization during neuronal polarization. Neuron 94, 809–825. e807 (2017).
Article CAS PubMed Google Scholar
Rocks, O. et al. The palmitoylation machinery is a spatially organizing system for peripheral membrane proteins. Cell 141, 458–471 (2010).
Article CAS PubMed Google Scholar
Martin, B. R., Wang, C., Adibekian, A., Tully, S. E. & Cravatt, B. F. Global profiling of dynamic protein palmitoylation. Nat. Methods 9, 84–89 (2011).
Article PubMed PubMed Central Google Scholar
Blanc, M., David, F. P. A. & van der Goot, F. G. SwissPalm 2: protein S-palmitoylation database. Methods Mol. Biol. 2009, 203–214 (2019).
Article CAS PubMed Google Scholar
Shipston, M. J. Ion channel regulation by protein S-acylation. J. Gen. Physiol. 143, 659–678 (2014).
Article CAS PubMed PubMed Central Google Scholar
Chien, A. J., Carr, K. M., Shirokov, R. E., Rios, E. & Hosey, M. M. Identification of palmitoylation sites within the L-type calcium channel beta2a subunit and effects on channel function. J. Biol. Chem. 271, 26465–26468 (1996).
Article CAS PubMed Google Scholar
Takimoto, K., Yang, E. K. & Conforti, L. Palmitoylation of KChIP splicing variants is required for efficient cell surface expression of Kv4.3 channels. J. Biol. Chem. 277, 26904–26911 (2002).
Article CAS PubMed Google Scholar
Tian, L., McClafferty, H., Knaus, H. G., Ruth, P. & Shipston, M. J. Distinct acyl protein transferases and thioesterases control surface expression of calcium-activated potassium channels. J. Biol. Chem. 287, 14718–14725 (2012).
Article CAS PubMed PubMed Central Google Scholar
Gok, C. et al. Dynamic palmitoylation of the sodium-calcium exchanger modulates its structure, affinity for lipid-ordered domains, and inhibition by XIP. Cell Rep. 31, 107697 (2020).
Article CAS PubMed PubMed Central Google Scholar
Rocks, O. et al. An acylation cycle regulates localization and activity of palmitoylated Ras isoforms. Science 307, 1746–1752 (2005).
Article ADS CAS PubMed Google Scholar
Wedegaertner, P. B., Chu, D. H., Wilson, P. T., Levis, M. J. & Bourne, H. R. Palmitoylation is required for signaling functions and membrane attachment of Gq alpha and Gs alpha. J. Biol. Chem. 268, 25001–25008 (1993).
Article CAS PubMed Google Scholar
El-Husseini Ael, D. et al. Synaptic strength regulated by palmitate cycling on PSD-95. Cell 108, 849–863 (2002).
Article Google Scholar
Sato, I. et al. Differential trafficking of Src, Lyn, Yes and Fyn is specified by the state of palmitoylation in the SH4 domain. J. Cell Sci. 122, 965–975 (2009).
Article CAS PubMed Google Scholar
Ko, P. J. & Dixon, S. J. Protein palmitoylation and cancer. EMBO Rep. 19, e46666 (2018).
Article PubMed PubMed Central Google Scholar
Veit, M. Palmitoylation of virus proteins. Biol. Cell 104, 493–515 (2012).
Article CAS PubMed PubMed Central Google Scholar
Dong, G. et al. Palmitoylation couples insulin hypersecretion with beta cell failure in diabetes. Cell Metab. 35, 332–344 e337 (2023).
Article CAS PubMed PubMed Central Google Scholar
Zhang, M. et al. A STAT3 palmitoylation cycle promotes T(H)17 differentiation and colitis. Nature 586, 434–439 (2020).
Article ADS CAS PubMed PubMed Central Google Scholar
Essandoh, K., Philippe, J. M., Jenkins, P. M. & Brody, M. J. Palmitoylation: a fatty regulator of myocardial electrophysiology. Front. Physiol. 11, 108 (2020).
Article PubMed PubMed Central Google Scholar
Lewis, S. A. et al. Mutation in ZDHHC15 leads to hypotonic cerebral palsy, autism, epilepsy, and intellectual disability. Neurol. Genet. 7, e602 (2021).
Article PubMed PubMed Central Google Scholar
Virlogeux, A. et al. Increasing brain palmitoylation rescues behavior and neuropathology in Huntington disease mice. Sci. Adv. 7, eabb0799 (2021).
Article ADS CAS PubMed PubMed Central Google Scholar
Fraser, N. J., Howie, J., Wypijewski, K. J. & Fuller, W. Therapeutic targeting of protein S-acylation for the treatment of disease. Biochem. Soc. Trans. 48, 281–290 (2020).
Article CAS PubMed Google Scholar
Malgapo, M. I. P. & Linder, M. E. Substrate recruitment by zDHHC protein acyltransferases. Open Biol. 11, 210026 (2021).
Article CAS PubMed PubMed Central Google Scholar
Lan, T., Delalande, C. & Dickinson, B. C. Inhibitors of DHHC family proteins. Curr. Opin. Chem. Biol. 65, 118–125 (2021).
Article CAS PubMed PubMed Central Google Scholar
Azizi, S. A. et al. Development of an acrylamide-based inhibitor of protein S-acylation. ACS Chem. Biol. 16, 1546–1556 (2021).
Article CAS PubMed PubMed Central Google Scholar
Hedberg, C. et al. Development of highly potent inhibitors of the Ras-targeting human acyl protein thioesterases based on substrate similarity design. Angew. Chem. Int. Ed. Engl. 50, 9832–9837 (2011).
Article CAS PubMed Google Scholar
Won, S. J. et al. Molecular Mechanism for Isoform-Selective Inhibition of Acyl Protein Thioesterases 1 and 2 (APT1 and APT2). ACS Chem. Biol. 11, 3374–3382 (2016).
Article CAS PubMed PubMed Central Google Scholar
Remsberg, J. R. et al. ABHD17 regulation of plasma membrane palmitoylation and N-Ras-dependent cancer growth. Nat. Chem. Biol. 17, 856–864 (2021).
Article CAS PubMed PubMed Central Google Scholar
Komatsu, T. et al. Organelle-specific, rapid induction of molecular activities and membrane tethering. Nat. Methods 7, 206–208 (2010).
Article CAS PubMed PubMed Central Google Scholar
Inoue, T., Heo, W. D., Grimley, J. S., Wandless, T. J. & Meyer, T. An inducible translocation strategy to rapidly activate and inhibit small GTPase signaling pathways. Nat. Methods 2, 415–418 (2005).
Article CAS PubMed PubMed Central Google Scholar
Benedetti, L. et al. Optimized vivid-derived magnets photodimerizers for subcellular optogenetics in mammalian cells. Elife 9, e63230 (2020).
Article CAS PubMed PubMed Central Google Scholar
Siller, A. et al. beta2-subunit alternative splicing stabilizes Cav2.3 Ca(2+) channel activity during continuous midbrain dopamine neuron-like activity. Elife 11, e67464 (2022).
Article CAS PubMed PubMed Central Google Scholar
Wang, H. G. et al. The auxiliary subunit KChIP2 is an essential regulator of homeostatic excitability. J. Biol. Chem. 288, 13258–13268 (2013).
Article CAS PubMed PubMed Central Google Scholar
Kuo, H. C. et al. A defect in the Kv channel-interacting protein 2 (KChIP2) gene leads to a complete loss of I(to) and confers susceptibility to ventricular tachycardia. Cell 107, 801–813 (2001).
Article CAS PubMed Google Scholar
Parsons, S. J. & Parsons, J. T. Src family kinases, key regulators of signal transduction. Oncogene 23, 7906–7909 (2004).
Article CAS PubMed Google Scholar
Moore, A. R., Rosenberg, S. C., McCormick, F. & Malek, S. RAS-targeted therapies: is the undruggable drugged? Nat. Rev. Drug Discov. 19, 533–552 (2020).
Article CAS PubMed PubMed Central Google Scholar
Martin, B. R. & Cravatt, B. F. Large-scale profiling of protein palmitoylation in mammalian cells. Nat. Methods 6, 135–138 (2009).
Article CAS PubMed PubMed Central Google Scholar
Fegan, A., White, B., Carlson, J. C. & Wagner, C. R. Chemically controlled protein assembly: techniques and applications. Chem. Rev. 110, 3315–3336 (2010).
Article CAS PubMed Google Scholar
Stanton, B. Z., Chory, E. J. & Crabtree, G. R. Chemically induced proximity in biology and medicine. Science 359, eaao5902 (2018).
Article PubMed PubMed Central Google Scholar
Putyrski, M. & Schultz, C. Protein translocation as a tool: the current rapamycin story. FEBS Lett. 586, 2097–2105 (2012).
Article CAS PubMed Google Scholar
DeRose, R., Miyamoto, T. & Inoue, T. Manipulating signaling at will: chemically-inducible dimerization (CID) techniques resolve problems in cell biology. Pflug. Arch. 465, 409–417 (2013).
Article CAS Google Scholar
Jumper, J. et al. Highly accurate protein structure prediction with AlphaFold. Nature 596, 583–589 (2021).
Article ADS CAS PubMed PubMed Central Google Scholar
Qin, N. et al. Unique regulatory properties of the type 2a Ca2+ channel beta subunit caused by palmitoylation. Proc. Natl. Acad. Sci. USA 95, 4690–4695 (1998).
Article ADS CAS PubMed PubMed Central Google Scholar
Lin, Y. C. et al. Rapidly reversible manipulation of molecular activity with dual chemical dimerizers. Angew. Chem. 52, 6450–6454 (2013).
Article CAS Google Scholar
Dong, X. P. et al. PI(3,5)P(2) controls membrane trafficking by direct activation of mucolipin Ca(2+) release channels in the endolysosome. Nat. Commun. 1, 38 (2010).
Article ADS PubMed Google Scholar
Harvey, C. D. et al. A genetically encoded fluorescent sensor of ERK activity. Proc. Natl. Acad. Sci. USA 105, 19264–19269 (2008).
Article ADS CAS PubMed PubMed Central Google Scholar
Kubala, M. H., Kovtun, O., Alexandrov, K. & Collins, B. M. Structural and thermodynamic analysis of the GFP:GFP-nanobody complex. Protein Sci. 19, 2389–2401 (2010).
Article CAS PubMed PubMed Central Google Scholar
Morgenstern, T. J., Park, J., Fan, Q. R. & Colecraft, H. M. A potent voltage-gated calcium channel inhibitor engineered from a nanobody targeted to auxiliary Ca(V)beta subunits. Elife 8, e49253 (2019).
Article PubMed PubMed Central Google Scholar
Gross, G. G. et al. Recombinant probes for visualizing endogenous synaptic proteins in living neurons. Neuron 78, 971–985 (2013).
Article CAS PubMed PubMed Central Google Scholar
Chavda, B., Arnott, J. A. & Planey, S. L. Targeting protein palmitoylation: selective inhibitors and implications in disease. Expert Opin. Drug Discov. 9, 1005–1019 (2014).
Article CAS PubMed Google Scholar
Papke, B. & Der, C. J. Drugging RAS: know the enemy. Science 355, 1158–1163 (2017).
Article ADS CAS PubMed Google Scholar
Simanshu, D. K., Nissley, D. V. & McCormick, F. RAS proteins and their regulators in human disease. Cell 170, 17–33 (2017).
Article CAS PubMed PubMed Central Google Scholar
McClafferty, H., Runciman, H. & Shipston, M. J. Site-specific deacylation by ABHD17a controls BK channel splice variant activity. J. Biol. Chem. 295, 16487–16496 (2020).
Article CAS PubMed Google Scholar
Niu, J. et al. Allosteric regulators selectively prevent Ca(2+)-feedback of Ca(V) and Na(V) channels. Elife 7, e35222 (2018).
Article PubMed PubMed Central Google Scholar
Ben-Johny, M. Custom code for: engineered depalmitoylases enable selective manipulation of protein localization and function. Zenodo. https://doi.org/10.5281/zenodo.15102975.

Download references

Acknowledgements

We thank Ryan Mahling and Lucile Fossier for their insightful comments. These studies were supported by funding from the National Institutes of Health to MBJ (R01 NS110672, R01 HL163576, P01 HL164319). This content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Author information

Authors and Affiliations

Department of Physiology and Cellular Biophysics, Columbia University, New York, NY, USA
Srinidhi Jayaraman, Audrey Kochiss, Pedro J. del Rivero Morfin & Manu Ben-Johny
Barnard College, Columbia University, New York, NY, USA
Thy-Lan Alcalay

Authors

Srinidhi Jayaraman
View author publications
Search author on:PubMed Google Scholar
Audrey Kochiss
View author publications
Search author on:PubMed Google Scholar
Thy-Lan Alcalay
View author publications
Search author on:PubMed Google Scholar
Pedro J. del Rivero Morfin
View author publications
Search author on:PubMed Google Scholar
Manu Ben-Johny
View author publications
Search author on:PubMed Google Scholar

Contributions

S.J., and M.B.-J. conceptualized and designed the project, S.J., A.K., T.A., P.J.D., and M.B-J. performed experiments, S.J., T.A., P.J.D., and M.B.-J. analyzed data; M.B.-J. funding acquisition and supervision; S.J. and M.B.-J. made figures and wrote the original manuscript draft; and all authors revised the manuscript.

Corresponding author

Correspondence to Manu Ben-Johny.

Ethics declarations

Competing interests

M.B.-J. and S.J. (inventors) filed a provisional patent (44010.198US-PROV//CU24204) for application of engineered depalmitoylases to tune protein palmitoylation. The remaining authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Reporting Summary

Transparent Peer Review file

Source data

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Cite this article

Jayaraman, S., Kochiss, A., Alcalay, TL. et al. Engineered depalmitoylases enable selective manipulation of protein localization and function. Nat Commun 16, 3514 (2025). https://doi.org/10.1038/s41467-025-58908-x

Download citation

Received: 21 December 2023
Accepted: 03 April 2025
Published: 13 April 2025
DOI: https://doi.org/10.1038/s41467-025-58908-x