Regulation of Botulinum neurotoxin half‑life through the ubiquitin‑proteasome system

Introduction

Botulinum neurotoxin (BoNT), produced by the bacterium Clostridium botulinum, is widely recognized as one of the most potent biological toxins. Discovered in the late 19th century, its mechanism of action involves the inhibition of acetylcholine release at the neuromuscular junction, resulting in flaccid paralysis (1). Despite its high toxicity, BoNT has been utilized for therapeutic purposes, particularly in the fields of neurology, ophthalmology, dermatology and urology, owing to its ability to induce temporary muscle relaxation (1).

BoNT consists of a heavy chain (HC; 100 kDa) and a light chain (LC; 50 kDa), which are connected by a disulfide bond. The HC selectively binds to synaptic vesicle protein 2 on neuronal membranes, leading to endocytosis (2,3), whereas the LC cleaves specific snap receptor proteins, inhibiting acetylcholine release (4). There are seven serotypes of BoNT (A-G) (5), each with distinct molecular targets. BoNT/A, C and E cleave synaptosomal-associated protein 25 kDa (SNAP-25), while BoNT/B, D, F and G cleave vesicle-associated membrane proteins, and BoNT/C also targets syntaxin-1 (6). Among these, the BoNT/A serotype has the longest duration of action, lasting ≤6 months, and is commonly used for both therapeutic and cosmetic applications (7,8). Meanwhile, BoNT/E has a rapid onset and short duration of action compared with other Botulinum toxin serotypes, making it suitable for specific clinical scenarios, such as transient muscle relaxation in neurological diagnostics or short-term therapeutic trials requiring rapid reversibility (9).

The therapeutic applications of BoNT have revolutionized the management of various medical conditions, ranging from neuromuscular disorders such as dystonia and spasticity to cosmetic treatments for reducing facial wrinkles (7). Previous research has also unveiled its potential roles in alleviating chronic pain, migraines and hyperhidrosis (10–12). However, several considerations must be assessed when using BoNT, including the development of resistance, immunogenicity, variable patient responses and the need for precise dosing regimens (13). While extending the half-life of BoNT may enhance therapeutic efficacy and reduce the frequency of injections, it also raises potential concerns such as prolonged muscle paralysis, unintended diffusion to non-target tissues and the formation of neutralizing antibodies. These risks must be carefully considered when developing BoNT variants with enhanced stability (14,15).

Various strategies have been employed to extend the half-life of therapeutic proteins in vivo, including PEGylation, Fc fusion, albumin fusion and glycosylation, all of which are widely used (16). However, these methods often involve structural modifications that can affect protein size, immunogenicity or functional activity, potentially limiting their clinical utility and compromising protein activity or structural integrity (17,18). By contrast, residue-specific lysine-to-arginine substitution offers a more precise strategy to enhance protein stability without substantially modifying the native structure (19,20). Additionally, modulation of the ubiquitin-proteasome system (UPS) has emerged as a promising approach for regulating protein half-life. These strategies may ultimately support more durable therapeutic effects with reduced dosing frequency, which are particularly beneficial in chronic or repetitive treatment settings. The UPS is a key cellular pathway in eukaryotic cells and is responsible for the degradation of the majority of intracellular proteins. It regulates essential processes such as cell cycle progression, DNA repair and oxidative stress response, thereby maintaining cellular homeostasis. Dysregulation of this system has been implicated in various diseases, including cancer, neurodegenerative disorders and immune dysfunctions (21,22). Proteins targeted for degradation via the UPS are tagged with ubiquitin molecules in a cascade involving E1 (activating enzyme), E2 (conjugating enzyme) and E3 (ligase enzyme). The ubiquitin chains are recognized and degraded by the 26S proteasome (23). Ubiquitin molecules are primarily attached to the lysine residues on proteins, forming polyubiquitin chains by linking multiple ubiquitin units (24,25). In contrast to ubiquitination, deubiquitination occurs when deubiquitinating enzymes (DUBs) remove these ubiquitin molecules (26). Together, these processes regulate protein stability and various cellular functions, maintaining cellular homeostasis (26).

In the present study, an innovative approach was applied to BoNT, aiming to extend its half-life. The primary objective was to enhance the duration of its therapeutic efficacy while minimizing the required dosage. The present study aimed to enhance the stability and half-life of BoNT by introducing lysine-to-arginine mutations at key ubiquitination sites, while also assessing the potential effects on its enzymatic activity. Furthermore, this approach provides new possibilities for broader applications of BoNT in both therapeutic and cosmetic fields.

Materials and methods

Cell culture and transfection

293T (cat. no. CRL-11268; American Type Culture Collection), cells were cultured in Dulbecco's Modified Eagle Medium (Gibco; Thermo Fisher Scientific, Inc.) with high glucose, supplemented with 10% fetal bovine serum (Gibco; Thermo Fisher Scientific, Inc.) and 1% penicillin-streptomycin (Gibco; Thermo Fisher Scientific, Inc.). The cells were maintained at 37°C in a humidified atmosphere with 5% CO2.

Plasmid DNA and antibodies

Wild type (WT)-BoNT/A1, BoNT/A2 and BoNT/E cDNA were synthesized and subcloned into the pCS4-3XFlag-tagged vector (Invitrogen; Thermo Fisher Scientific, Inc.). The pRK5-hemagglutinin (HA)-tagged ubiquitin (HA-Ub) plasmid was provided by Professor Byung Joon Hwang (Kangwon National University, Chuncheon, South Korea) and used for co-transfection to assess ubiquitination levels. For transfection (100 mm dish), 1–3 µg each plasmid DNA was used (Flag-BoNT/A1-WT and -mutants, Flag-BoNT/A2-WT and K335R, Flag-BoNT/E-WT and -mutants and HA-Ub). Transfection of each construct was carried out using polyethyleimine (cat. no. 23966-1; Polysciences, Inc.) according to the manufacturer's protocol. Cells were incubated at 37°C during transfection. After 24 h, the culture medium was replaced with fresh medium, and cells were harvested 48 h post-transfection for downstream experiments. cDNAs (Flag-BoNT/A1, Flag-BoNT/A2, Flag-BoNT/E and HA-Ub) were transformed into the Escherichia coli DH5α (cat. no. RH617; RBC Bioscience Corp.) bacterial strain and the bacteria were grown for 16 h at 37°C in Luria-Bertani broth (cat. no. MB-L4488; Kisan Bio Co., Ltd.). Anti-β-actin (1:3,000; cat. no. sc-4778; Santa Cruz Biotechnology, Inc.), anti-Flag (1:4,000; cat. no. M185-3L; MBL International Co.) and anti-HA (1:1,000; cat. no. sc-53516; Santa Cruz Biotechnology, Inc.) antibodies were used for western blotting and immunoprecipitation.

Site-directed mutagenesis

Site-directed mutagenesis was performed using WT Flag-BoNT/A1, Flag-BoNT/A2 and Flag-BoNT/E plasmids as a template to generate lysine-to-arginine substitution mutants and amplify the mutant plasmids. The PCR conditions were as follows: 95°C for 30 sec, 54°C for 40 sec and 68°C for 6 min, for a total of 15 cycles. The primers used for each substitution are listed in Table I. The amplified products were subsequently subjected to DpnI digestion to remove the template plasmid, followed by transformation into competent cells for propagation and sequence verification.

Table I. Primers for site-directed mutagenesis of BoNT-LC.

Table I.

Primers for site-directed mutagenesis of BoNT-LC.

Target	Mutations	Sequence, 5′-3′
BoNT/A1-LC	K212R	F: GGTGCAGGCAGATTTGCT
		R: AGCAAATCTGCCTGCACC
	K301R	F: CTTAATAAAGCTAGATCAATAGTAGGT
		R: ACCTACTATTGATCTAGCTTTATTAAG
	K320R	F: TTTTAAAGAGAGATATCTCCTA
		R: TAGGAGATATCTCTCTTTAAAA
	K330R	F: CATCTGGAAGATTTTCGG
		R: CCGAAAATCTTCCAGATG
	K335R	F: TTTCGGTAGATAGATTAAATT
		R: AATTTTAATCTATCTACCGAAA
	K340R	F: TTTTGTATAACCTATCAAATT
		R: ACTAAACTAGAAATTTTACTGG
	K417R	F: ACTAAACTAGAAATTTTACTGG
		R: CCAGTAAAATTTC TTAGTTTAGT
BoNT/A2-LC	K335R	F: TCGGTAGATAGATTAAATTTGATAAGTTATACAAAATGTTAACAGAG
		R: ATCAAATTTTAATCTATCTACCGAAAATTTTCCAGATGTATC
BoNT/E-LC	K3R	F: TGCCAAGAATTAATAGTTTTAATTATAATGATCCTGTTAATGATAGAAC
		R: AAAACTATTAATTCTTGGCATGGATCCGTCC
	K62R	F: ACTTCATTAAGAAATGGAGATAGTAGTTATTATGACCC
		R: ATCTCCATTTCTTAATGAAGTAGGCGGATGAAAATC
	K85R	F: TTTTAAGAATAGTCACAAAAATATTTAATAGAATAAATAATAATCTTTC
		R: TTTTGTGACTATTCTTAAAAATCTATCCTTTTCTTCATCACTTTG
	K230R	F: ATTACTACAAGGTATACTATAACACAAAAACAAAATCC CCTAATAAC
		R: TATAGTATACCTTGTAGTAATCCCTTTAGCCCCATATAG
	K281R	F: GATTATAAAAGAATAGCGTCTAAACTTAGCAAAGTAC
		R: CGCTATTCTTTTATAATCAGCTAGAAGATTAGTATAGATATCATTGGAC
	K288R	F: AAACTTAGCAGAGTACAAGTATCTAATCCACTACTTAATCC
		R: TACTTGT ACTCTGCTAAGTTTAGACGCTATTTTTTTATAATCAG
	K342R	F: AACTTGAAATCTAGTTGCTAAATCAAATTCCGTAAAGCTG
		R: TTAGCAACTAGATTTCAAGTTAAATGTAGGCAAACTTATATTG
	K413R	F: TAGATTTTGTAGAAATATTGTTTCTGTAAAAGGCATAAGG
		R: ACAATATTTCTACAAAATCTAATGATTTTTTTTACTAGTCCTCTAC
	K419R	F: TATTGTTTCTGTAAGAGGCATAAGGTCTAGAGGGC
		R: TGCCTCTTACAGAAACAATATTTTTACAAAATCTAATGATTTTTTTTAC

[i] BoNT, Botulinum neurotoxin; LC, light chain; F, forward; R, reverse.

Bioinformatics analysis of potential ubiquitination sites

To identify potential ubiquitination sites in the LC of BoNT/A1 and BoNT/E, bioinformatics tools were employed. Predicted ubiquitinated lysine residues in BoNT protein were identified using the BDM ubiquitination prediction tool (BDM-PUB) (27), which was analyzed in March 2022 for BoNT/A1, and in December 2022 for BoNT/A2 and BoNT/E. The analysis conducted using PyMOL (https://www.pymol.org/; Ver.3.1; Schrödinger Inc.) and AlphaFold (Ver. 2) via Google Colaboratory (https://colab.research.google.com/), which enabled GPU-based computation due to limited local hardware resources (28), revealed K212, K301, K320, K330, K335, K340 and K417 for BoNT/A1, and K3, K62, K85, K230, K281, K288, K343, K413 and K419 for BoNT/E as major ubiquitination sites. Subsequently, protein structure files for BoNT/A1, BoNT/A2 and BoNT/E, were obtained through the Protein Data Bank (PDB; http://www.rcsb.org/), were analyzed using the PyMOL program (PDB ID: 2ISE for BoNT/A1; PDB ID: 5MOY for BoNT/A2; and PDB ID: 7K7Y for BoNT/E). Among the identified lysine residues, those located on the protein surface were considered potential ubiquitination sites. Site-directed mutagenesis was carried out to substitute the selected lysine residues with arginine, generating lysine-to-arginine mutants. These mutants were experimentally validated to assess their effects on ubiquitination levels and protein stability.

Western blotting and immunoprecipitation

Harvested cells were lysed using RIPA lysis buffer [1% Triton X-100, 150 mM NaCl, 50 mM Tris-HCl (pH 8.0), 1 mM EDTA, 10% glycerol, 100 mM PMSF and protease inhibitor cocktail]. Cell lysates were centrifuged at 16,000 × g for 20 min at 4°C, and the supernatants were mixed with 2X SDS buffer and boiled at 100°C. Protein concentration was determined using the Bio-Rad Protein Assay Dye Reagent (cat. no. 5000006; Bio-Rad Laboratories, Inc.). For western blotting, 20–30 µg protein was loaded per lane and was separated by SDS-PAGE on 6 or 8% gels, depending on the molecular weight of the target protein. Subsequently, the proteins were transferred onto PVDF membranes (cat. no. IPVH00010; MilliporeSigma), which were blocked with 5% skim milk in TBS-0.5% Tween 20 (TBST) for 30 min at room temperature. The membranes were incubated with the primary antibodies overnight at 4°C, and after washing three times (7 min each) with TBST at room temperature, they were incubated with the HRP-conjugated secondary antibodies [1:30,000; cat. no. 5220-0341 (074–1806); KPL, Inc.; and 1:10,000; cat. no. 31470; Thermo Fisher Scientific, Inc.] for 1 h at room temperature. The membranes were then washed a three times with TBST (7 min each) and developed using an enhanced chemiluminescence reagent (cat. no. LF-QC0103; Abfrontier). The intensity of each western blot band was measured and semi-quantified using ImageJ software (Version 1.53t; National Institutes of Health) to assess relative protein expression levels.

For immunoprecipitation, cells were lysed using RIPA lysis buffer [1% Triton X-100, 150 mM NaCl, 50 mM Tris-HCl (pH 8.0), 1 mM EDTA, 10% glycerol, 100 mM PMSF and protease inhibitor cocktail]. 1.5–2 mg protein lysates were incubated with 2 µg anti-Flag antibody at 4°C overnight on a rotator. Subsequently, 35 µl Protein A/G PLUS-agarose beads (cat. no. sc-2003; Santa Cruz Biotechnology Inc.) were added and the samples were incubated for 2 h at 4°C. The immunocomplexes were pelleted by centrifugation at 800 × g for 5 min at 4°C and washed three times with ice-cold RIPA buffer. After the final wash, the beads were resuspended in 2X SDS sample buffer and boiled at 100°C for 7 min. The supernatants were then analyzed by SDS-PAGE followed by western blotting.

Ubiquitination assay

For the ubiquitination assay, 293T cells (2×106) were co-transfected with Flag-tagged BoNT-A1 (WT or K212R, K301R, K320R, K330R, K335R, K340R and K417R mutants), BoNT-A2 (WT or K335R mutant) or BoNT-E (WT or K3R, K62R, K85R, K230R, K281R, K288R, K342R, K413R and K419R mutants) along with HA-Ub. After 4 h of treatment with MG132 at 37°C (10 µM cat. no. F1100; Ubiquitin-Proteasome Biotechnologies, LLC), the cells were harvested and lysed using RIPA lysis buffer supplemented with protease inhibitors. Immunoprecipitation was carried out on the cell lysates (1.5–2 mg total protein) using an anti-Flag antibody as aforementioned, and the ubiquitination levels of Flag-BoNT-LC were analyzed by western blotting with anti-HA antibody to detect ubiquitinated proteins.

Sequence homology analysis

Amino acid sequences of BoNT serotypes (A-G) were obtained from the National Center for Biotechnology Information database, (https://www.ncbi.nlm.nih.gov/) GenBank (https://www.ncbi.nlm.nih.gov/genbank/) using their respective accession numbers. Sequence alignment was performed using Clustal X 2.1 (http://www.clustal.org/clustal2/) with default parameters to calculate the percentage identity of each serotype relative to BoNT/A1. The resulting pairwise percentage homologies were tabulated and used to evaluate sequence similarity across serotypes (Table II).

Table II. Comparison of sequence homology percentages among different types of Botulinum neurotoxin.

Table II.

Comparison of sequence homology percentages among different types of Botulinum neurotoxin.

GenBank protein ID	Description	Percentage homology with type A1	Type
AUN16864.1	Botulinum neurotoxin type A (Clostridium botulinum)	NA	A1
CAA51824.1	Botulinum neurotoxin type A2 (Clostridium botulinum)	89.66%	A2
AAA23211.1	Botulinum neurotoxin type B (Clostridium botulinum)	40.35%	B
BAC22064.1	Botulinum neurotoxin type C (Clostridium botulinum)	39.88%	C
CAA38175.1	Botulinum neurotoxin type D (Clostridium botulinum)	39.29%	D
CAA44558.1	Botulinum neurotoxin type E (Clostridium botulinum)	38.77%	E
CAA57358.1	Botulinum neurotoxin type F (Clostridium botulinum)	40.15%	F
CAA52275.1	Botulinum neurotoxin type G (Clostridium botulinum)	33.28%	G

Protein stability assay

To evaluate the stability of the BoNT/A1-, BoNT/A2- and BoNT/E-LCs, cycloheximide (CHX; cat. no. 01810; Sigma-Aldrich; Merck KgaA) chase experiments were conducted. 293T cells were transfected with plasmids encoding either WT LCs or the lysine-to-arginine mutants (K212R, K301R, K320R, K330R, K335R, K340R and K417R for A1, K335R for A2 and K3R, K62R, K85R, K230R, K281R, K288R, K343R, K413R and K419R for E). Following transfection, CHX was added at a final concentration of 100 µg/ml to inhibit de novo protein synthesis. Cells were harvested at designated time points (0, 12, 18 or 0, 6, 12 h post-CHX treatment at 37°C in a 5% CO2 incubator).

Statistical analysis

Data were obtained from at least three independent experiments. Mean values were calculated, and all quantitative data are presented as the mean ± SD unless otherwise specified in the figure legends. Statistical analysis was performed using two-way ANOVA followed by Tukey's post hoc test for multiple group comparisons. All analyses were conducted using GraphPad Prism (Ver.9; Dotmatics). P<0.05 was considered to indicate a statistically significant difference.

Results

Sequence homology between BoNT types

There are seven known types of BoNT (A-G). Sequence homology among these types was compared and is summarized in Table II. BoNT/A2 exhibited the highest similarity to BoNT/A1, with a sequence homology of 89.66%. By contrast, the other BoNT types displayed 33–40% sequence similarity to BoNT/A1. These sequence variations are not associated with differences in their cleavage targets. For example, BoNT/A, BoNT/C and BoNT/E all target SNAP-25, yet BoNT/C and BoNT/E share only 39.88 and 38.77% sequence homology with BoNT/A1, respectively (Table II).

293T cells were transfected with Flag-tagged BoNT/A1-LC, BoNT/A2-LC and BoNT/E-LC constructs. To confirm the ubiquitination of BoNT/A1, BoNT/A2 and BoNT/E, 293T cells were co-transfected with HA-Ub and Flag-BoNT/A1-LC, BoNT/A2-LC and BoNT/E-LC constructs. Following treatment with the proteasome inhibitor MG132, immunoprecipitation was performed using anti-Flag antibody. Compared with the control condition lacking HA-Ub or MG132 treatment, increased ubiquitin chain signal was detected for all three BoNT subtypes, indicating that BoNT/A1, BoNT/A2 and BoNT/E undergo UPS-mediated degradation (Fig. 1A-C).

Figure 1. Ubiquitination of BoNT/A1, BoNT/A2 and BoNT/E-LCs. (A) BoNT/A1-LC, (B) BoNT/A2-LC and (C) BoNT/E-LC were analyzed in the presence of the proteasome inhibitor MG132. Cells were transfected with the indicated BoNT-LC constructs and HA-tagged ubiquitin, followed by IP using the anti-Flag antibody and WB with anti-HA to detect ubiquitinated proteins. Enhanced ubiquitination was observed for all BoNT types under proteasome inhibition. WB, western blotting; BoNT, Botulinum neurotoxin; WCL, whole cell lysate; IP, immunoprecipitation; HA, hemagglutinin; Ub, ubiquitin; LC, light chain.

Reduced ubiquitination in lysine mutants compared with WT BoNT/A1, A2 and E

To predict ubiquitinated lysine residues in BoNT/A1 and BoNT/E, a ubiquitination prediction tool BDM, analyzed in March 2022 (BoNT/A1) and December 2022 (BoNT/A2, BoNT/E)] was used (27). The identified lysine sites were further analyzed using the PyMOL program and structural data from the PDB for each BoNT subtype. Surface-exposed lysines were selected as potential ubiquitination sites, while those buried within the structure of the protein were excluded. Based on these predictions (Fig. 2A), lysine residues K212, K301, K320, K330, K335, K340 and K417 in BoNT/A1 were substituted with arginine to block ubiquitination at these positions. 293T cells were co-transfected with mutant plasmids and HA-Ub. Immunoprecipitation was carried out using anti-Flag antibodies to isolate the mutant proteins. As shown in Fig. 2B (right panel), HA-Ub immunoblots showed substantially decreased polyubiquitin chains in the lanes corresponding to BoNT/A1-K330R, -K335R and -K417R mutants, compared with that in the WT group, supporting their identification as key ubiquitination sites in BoNT/A1 (Fig. 2B).

Figure 2. Ubiquitination levels of WT and lysine mutant forms of BoNT-LCs. (A) Protein is composed of an LC (gray) and heavy chain (black and dark gray). The heavy chain is further divided into HN (black) and HC (C-terminal of heavy chain) (dark gray) regions. Mutation sites were selected based on ubiquitin prediction analysis and mutated accordingly. BoNT/A1, BoNT/A2 and BoNT/E are represented. (B) Ubiquitination levels of WT and K212R, K301R, K320R, K330R, K335R, K340R and K417R mutants of BoNT/A1-LC. (C) Ubiquitination levels of WT and the K335R mutant of BoNT/A2-LC. (D) Ubiquitination levels of WT and K3R, K62R, K85R, K230R, K281R, K288R, K342R, K413R and K419R mutants of BoNT/E-LC. Short exposure images were used to avoid signal saturation and clearly visualize differences in ubiquitination levels between wild-type and mutant proteins. WT, wild type; BoNT, Botulinum neurotoxin; LC, light chain; HN, N-terminus of heavy chain HC; C-terminus of heavy chain; WCL, whole cell lysate; IP, immunoprecipitation; HA, hemagglutinin; Ub, ubiquitin.

Given the high sequence similarity (89.66%) between BoNT/A2 and BoNT/A1, lysine 335 was also analyzed in BoNT/A2. Prediction site analysis confirmed a similar ubiquitination profile between the two subtypes. Using site-directed mutagenesis, the K335R mutant for BoNT/A2 was generated, and ubiquitination assays revealed reduced ubiquitin binding compared with the WT BoNT/A2, consistent with the findings for BoNT/A1 (Fig. 2C).

By contrast, BoNT/E exhibits only ~40% sequence similarity with BoNT/A1, necessitating new predictions for ubiquitinated lysine residues. Bioinformatics analysis identified lysine residues K3, K62, K85, K230, K281, K288, K342, K413 and K419 as potential ubiquitination sites. These residues were individually substituted with arginine, and ubiquitination assays revealed that the K62R and K288R mutants reduced ubiquitin binding in the HA blot, Fig. 2D (right panel); HA-Ub immunoblots showed markedly reduced polyubiquitin chains signal in lanes corresponding to K62R and K288R mutants, compared with that in the WT group. This highlights K62R and K288R as key ubiquitination sites in BoNT/E (Fig. 2D). Overall, the results of the present study identified key ubiquitination sites in BoNT/A1, BoNT/A2 and BoNT/E. Further experiments aimed to assess whether targeting these residues could extend the half-life these proteins, thereby enhancing their therapeutic potential.

Extending protein half-life through lysine mutations

A CHX chase assay was conducted to determine whether lysine-to-arginine mutations, which reduce ubiquitination compared with WT serotypes, could enhance the stability of BoNT proteins by decreasing their degradation rates (Fig. 3A-C) In BoNT/A1 proteins, the K212 mutation exhibited a trend toward reduced protein stability compared with WT. Conversely, mutations at K335 and K417 demonstrated increased protein stability, with K335 showing significant stabilization at 12 h (1.516-fold) and at 18 h (2.05-fold). Similarly, the K417 mutation exhibited significant stabilization at both 12 h (1.528-fold) and 18 h (1.727-fold; Fig. 3A). For BoNT/A2, the K335R mutant displayed significantly enhanced stability compared with WT at both 12 h (1.37-fold) and 18 h (1.58-fold), consistent with the findings for BoNT/A1 (Fig. 3B). To avoid signal saturation and to better resolve subtle differences in protein degradation rates, short exposure blots were included for CHX chase assays. These allow more accurate comparison of stability between WT and mutant forms. In BoNT/E, lysine residues K62 and K288 were analyzed. Both mutations revealed improved protein stability relative to WT. K62 revealed a 1.34-fold increase in stability at 12 h, while K288 demonstrated a 1.32-fold increase at 12 h and a significant 2.23-fold increase at 12 h (Fig. 3C). These results highlight the ability of these lysine residues in regulating protein stability.

Figure 3. Protein stability of WT and lysine mutant forms of BoNT-LCs. (A) CHX chase assay comparing protein stability of BoNT/A1-LC WT and K212R, K301R, K320R, K330R, K335R, K340R and K417R mutants. Protein levels were analyzed by WB. (B) CHX chase assay of BoNT/A2-LC WT and K335R mutant. (C) CHX chase assay of BoNT/E-LC WT, K62R and K288R mutants. WB was carried out to compare protein expression levels of WT and mutant forms under identical experimental conditions. Short exposure images were used to avoid signal saturation and clearly visualize differences in Flag-BoNT transfected cells. *P<0.05 and **P<0.01. WT, wild type; BoNT, Botulinum neurotoxin; LC, light chain; CHX, cycloheximide; WB, western blotting.

These findings underscored the potential of targeting specific lysine residues through lysine-to-arginine mutations to extend the half-life of BoNT proteins, providing a foundation for developing more durable therapeutic formulations.

Discussion

The therapeutic and cosmetic applications of BoNT have considerably advanced over the past few decades, offering effective solutions for treating various diseases and enhancing cosmetology (29–32). Among the seven serotypes, BoNT/A is the most widely used due to its prolonged half-life (~6 months) and stable action, whereas BoNT/E is preferred in specific cases for its rapid onset and shorter duration of action (33–36). Despite these benefits, the limited half-life and short duration of efficacy of BoNTs remain notable challenges in clinical applications. In the present study, the role of the UPS in BoNT degradation was investigated strategies to enhance the stability of BoNTs by inhibiting ubiquitination were explored (8,37).

The UPS is a primary pathway for intracellular proteolysis, carrying out a key role in regulating protein stability and maintaining cellular homeostasis (21,22). The BoNT-LC undergoes degradation primarily through ubiquitination, a significant factor contributing to its limited half-life (38,39). Previous studies have demonstrated that the BoNT/E-LC interacts with the RING finger protein TNF receptor associated factor 2 to facilitate its degradation (38,40). These findings suggest that targeting the ubiquitination pathway could effectively extend the half-life of BoNTs.

The UPS also regulates the degradation of BoNT/A, and alterations in UPS activity have been shown to influence the persistence of this toxin (39,40). One study screened 32 UPS inhibitors to identify compounds that modulate BoNT/A degradation. This previous study aimed to mitigate BoNT poisoning by exploring compounds that regulate ubiquitination and deubiquitination. Among the inhibitors tested, 10 were found to effectively modulate ubiquitination, with WP1130, PR-619 and celastrol being particularly effective in reducing the persistence of BoNT/A. These compounds also demonstrated the ability to inhibit SNAP-25 cleavage even after toxin exposure (38), further supporting their potential utility in managing BoNT activity. These compounds function as broad-spectrum DUB inhibitors, disrupting the removal of ubiquitin chains and thereby promoting proteasomal degradation of target proteins (41). However, due to their lack of specificity, they may inadvertently affect various cellular processes, including oxidation stress responses and proteostasis, leading to potential off-target effects. Consequently, while their therapeutic potential is promising, further optimization is required to enhance their specificity and reduce toxicity (38,42).

Collectively, these studies highlight the potential of regulating the UPS pathway not only for managing BoNT toxicity but also for optimizing therapeutic strategies. By targeting the processes of ubiquitination and deubiquitination, which antagonize each other in regulating protein expression, it may be possible to reduce toxin dosage requirements, prolong its persistence and enhance its therapeutic utility.

The present study revealed structural differences among BoNT serotypes and demonstrated the importance of specific lysine residues in their degradation through ubiquitination. While BoNT/A1 and BoNT/A2 were revealed to share ~89.99% sequence similarity, BoNT/E shares only ~40% sequence similarity with BoNT/A1. This variability results in differences in predicted and identified ubiquitination sites, suggesting that each BoNT subtype has distinct mechanisms of ubiquitination and degradation.

Previous studies have suggested that BoNT degradation is mediated by the ubiquitination of its LC (29,30,38); however, the specific lysine residues involved and their functional roles have not been clearly defined. In the present study, primary ubiquitination sites were identified in BoNT/A1, BoNT/A2 and BoNT/E, and protein stability was enhanced by inhibiting ubiquitination through site-specific mutations. K335 and K417 in BoNT/A1 were identified as major ubiquitination sites, with the mutations reducing ubiquitination. Similarly, the K335R mutation in BoNT/A2 demonstrated increased protein stability. For BoNT/E, K62 and K288 were identified as key ubiquitination sites, with the K62R and K288R mutations effectively inhibiting ubiquitination. Notably, the lysine-to-arginine mutations at K335 and K417 in BoNT/A1, K335R in BoNT/A2, and K62 and K288 in BoNT/E each contributed to increased protein half-life, indicating that site-specific inhibition of ubiquitination can effectively enhance the stability of BoNT LC.

BoNTs are internalized into nerve cells by binding specific receptors on the neuronal membrane (43). After internalization, the BoNT/A-LC translocates into the cytoplasm, where it directly interacts with SNAP-25 (43,44). This interaction is mediated by the α and β-exosite in the LC, which determine the substrate specificity for SNAP-25. The cleavage of SNAP-25 between Gln197 and Arg198 is the primary mechanism of BoNT/A1 action. The low homology domain (LHD) in the BoNT/A1-LC (residues 268–357) is key for stable interactions with SNAP-25 and anchoring to the plasma membrane (44). Structurally, the LHD is divided into three regions: R1 (275–300), R2 (302–334) and R3. R1 forms an α-helix structure that mediates indirect interaction with SNAP-25, whereas R2 adopts a loop-α-helix-loop structure that establishes strong ionic and hydrogen bonds with SNAP-25 (43,44). Since the activity of BoNT relies on its precise interaction with SNAP-25, it is important to evaluate whether lysine-to-arginine mutations, while extending the protein stability of LCs, might affect its neurotoxin function. Given the key role of the LHD and exosite regions in mediating SNAP-25 recognition, it is essential to consider whether lysine-to-arginine substitutions affect residues within or adjacent to these functional domains. In the present study, the substituted lysine residues K335 and K417 in BoNT/A1 are located outside the core motifs for SNAP-25 interaction, indicating that direct interference with substrate binding is unlikely. Previous structural studies have also revealed that mutations distal to catalytic or binding interfaces can enhance protein stability without compromising enzymatic function (45,46). Since BoNT is recombinantly expressed in Escherichia coli, its sequence can be rationally engineered and screened to optimize the balance between structural stability and functional integrity. Nevertheless, to fully assess the therapeutic relevance of these mutations, future studies should directly compare the enzymatic activity and functional efficacy of WT and mutant BoNTs under physiologically relevant conditions.

By extending the half-life of BoNTs through K335R, K417R, K62R and K288R mutations, the present study introduced the potential for prolonged therapeutic effects with lower doses, reducing the frequency of administration. This approach is particularly beneficial in therapeutic settings, improving patient convenience and compliance. Moreover, extending the half-life of BoNTs could mitigate immunogenicity issues associated with repeated administration (47,48). Repeated administration of BoNT can lead to the formation of neutralizing antibodies, reducing the effectiveness of the drug. Extending the half-life of BoNT could reduce administration frequency, thereby minimizing this risk (47,48).

Existing technologies for extending protein half-life, such as PEGylation, Fc fusion, albumin fusion and glycosylation, often compromise the biological activity or structural stability of proteins (47). In contrast to conventional strategies, the approach taken in the present study utilizes residue-specific lysine-to-arginine substitutions, offering a minimally invasive alternative that preserves the native structure of the protein. Conventional strategies, such as PEGylation or fusion with Fc or albumin domains, extend serum half-life by promoting steric hindrance or receptor recycling. However, these strategies often reduce biological activity, increase immunogenicity and pose manufacturing challenges due to their bulky modifications (49–51). By comparison, lysine substitution directly inhibits ubiquitin-dependent degradation without structural enlargement, offering the potential to preserve functional activity and mitigate the risk of immune responses. Nonetheless, it should be noted that each method has inherent trade-offs, and the optimal choice may vary depending on the therapeutic context and specific protein characteristics. Clinically, prolonging the intracellular half-life of BoNT through reduced ubiquitination may extend its duration of neuromuscular blockade, thereby reducing the frequency of injections required for therapeutic efficacy. This could be particularly advantageous in chronic indications such as spasticity or dystonia, where repeated administration is often necessary. Furthermore, fewer injections may help lower the risk of immunogenicity and patient discomfort, both of which are key considerations in long-term treatment. Thus, by interfering with the degradation pathway rather than altering the delivery or formulation, this strategy may preserve the native function of BoNT while supporting improved therapeutic outcomes.

While the present study demonstrated a promising strategy to stabilize BoNT through lysine-to-arginine substitutions, several unresolved limitations should be taken into account when interpreting the results. To further evaluate the physiological relevance of these findings, additional studies using in vivo models are needed. Comparative experiments between WT and mutant BoNTs could assess functional persistence through measurements of muscle response duration, evaluate neurotoxicity profiles via hemidiaphragm assays and determine enzymatic activity by quantifying SNAP-25 cleavage in neural tissues. These approaches would help to establish whether the introduced mutations extend protein stability without compromising biological efficacy. To address these limitations, more precise evaluations of BoNT stability and functionality should be conducted through structural simulations and comprehensive in vivo experiments. Additionally, this approach must be evaluated for its effectiveness in other subtypes (BoNT/B, BoNT/C, BoNT/D, BoNT/F, BoNT/G and BoNT/H) beyond BoNT/A1, BoNT/A2 and BoNT/E. The potential of lysine mutations to reduce immunogenicity also warrants further exploration. While the present study focused on the role of the UPS, BoNTs may also be degraded through other pathways, such as autophagy. Therefore, a comprehensive investigation of the mechanisms involved in BoNT degradation is essential for fully understanding and optimizing the stability and therapeutic potential of these neurotoxins (52,53).

Acknowledgements

Not applicable.

Funding

This research was supported by a research fund from Alkemier Inc. (grant no. Alkemier-2024-1).

Availability of data and materials

The data generated in the present study may be requested from the corresponding author.

Authors' contributions

HSC, HRS, ML and KHB conceptualized and designed the present study. Material preparation, data collection and analysis were conducted by HSC, HRS, MC, HYK and SKJ. The initial draft of the manuscript was prepared by HSC, HRS and KHB. All authors contributed to the review and revision of the manuscript, and the final version was reviewed and approved by KHB. HSC and KHB confirm the authenticity of all the raw data. All authors read and approved the final manuscript.

Ethics approval and consent to participate

Not applicable.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Glossary

Abbreviations

Abbreviations:

References