Enzyme-mediated site-specific functionalization of mRNA-containing TMV nanorods produced in yeast

Virus-like particles (VLPs) derived from Tobacco mosaic virus (TMV) have emerged as attractive platforms for nanotechnology and molecular delivery applications owing to their robust self-assembly, stoichiometric precision, and highly accessible surface architecture. Here we report the development of a yeast-based platform for the production of TMV VLPs – nanorods – that can be site-specific functionalized and that contain RNA of interest. TMV coat proteins engineered with a C-terminal LPETG Sortase A (SrtA) recognition motif were expressed in Komagataella phaffii along with a self-amplifying RNA construct containing the TMV origin of assembly sequence (OAS) – the “packaging signal” – enabling the intracellular formation of nanorods. Transmission electron microscopy (TEM) confirmed the assembly of particles with wildtype TMV virion morphology, while RT-PCR analysis of these purified VLPs preparations demonstrated their packaging of target RNA. The accessibility of the engineered LPETG motif following assembly enabled SrtA mediated conjugation of N-terminal polyglycine-modified ligands, including fluorescent polyG-Alexa probes and a targeting affibody. Successful conjugation was verified by SDS-PAGE and fluorescence analyses, demonstrating efficient post-assembly functionalization while preserving particle integrity. Further, affibody-functionalized TMV nanorods were coupled to gold nanoparticles, and TEM confirmed the surface presentation and accessibility of the conjugated ligands on intact particles. Together, these results establish K. phaffii as a scalable platform for the production of engineered TMV nanorods and demonstrate that genetically encoded SrtA recognition motifs enable versatile post-assembly surface modification. This approach provides a modular framework for the development of customizable TMV-based nanomaterials for in vivo targeting, imaging, and nanobiotechnology applications.

Plant virus-like particles (VLPs) have emerged as powerful platforms for biotechnology, vaccination, and nucleic acid delivery due to their ability to self-assemble into highly ordered nanostructures while retaining the structural properties of native viruses without the risks associated with infectivity (1–3). Their intrinsic capacity to encapsidate nucleic acids, combined with the possibility of engineering their external surfaces, has enabled the development of programmable nanoparticles for applications ranging from vaccines to targeted drug and gene delivery (4–6). Because VLPs are genetically encoded and self-assemble into highly ordered nanostructures (6), they can be synthesized reproducibly at large scale while maintaining precise control over particle size, architecture, and functionality (6).

Among plant-virus-derived nanoparticles, tobacco mosaic virus (TMV) has become one of the most extensively studied systems. TMV is a positive-sense single-stranded RNA virus composed of 2,130 copies of a 17.5-kDa coat protein (CP) assembled around an embedded helically-ordered 6.4-kb RNA genome (7). The resulting virion forms a rigid rod about 300 nm in length and 18 nm in diameter, containing a central hollow core with a diameter of about 4 nm (8).

The high-resolution structure of TMV and its assembly pathway have been extensively characterized (9), making it one of the best understood biological self-assembling systems. TMV assembly is initiated through the interaction of CP oligomers with a highly conserved RNA secondary structure associated with the Origin of Assembly Sequence (OAS). Binding of this sequence by the CP nucleates particle formation, followed by bidirectional elongation of the helical rod capsid through the sequential addition of CP subunits. Assembly can also occur in the absence of the OAS but it is then initiated at multiple sites along the RNA and “branched” rather than straight nanorod structures arise (10). Importantly, the OAS can be incorporated into heterologous RNAs, allowing the assembly of TMV-based VLPs around non-viral cargo molecules (11). This feature has made TMV an attractive platform for the encapsidation of therapeutic RNAs and other functional nucleic acids (12–16). Furthermore, several studies have demonstrated the ability of TMV-derived particles to enter mammalian cells and deliver encapsidated RNA cargos both in vitro (in cell culture) (6, 12) and in vivo (in mice) (6, 13), highlighting their potential as nucleic acid delivery vehicles. In addition, assembled TMV particles exhibit remarkable stability over a broad range of temperatures, pH values, and ionic strengths, making them particularly attractive for biomedical and nanotechnological applications (14).

Beyond their ability to package nucleic acids, TMV particles possess a highly repetitive and accessible surface that can be engineered as a scaffold for the display of functional molecules (15). Numerous studies have explored the chemical modification of exposed amino acid residues on the interior and exterior surface of TMV particles to attach fluorescent dyes, imaging agents, peptides, targeting ligands, and therapeutic cargos (15). Conventional approaches typically rely on the introduction into the CP of reactive residues such as lysines or cysteines, followed by chemical cross-linking reactions after particle assembly (16). Although effective, these strategies often generate aggregated products and provide limited control of the position and orientation of the conjugated molecules. Such limitations become particularly important when displaying biologically active proteins, antibodies, enzymes, or receptor-binding ligands, where improper conjugation site and/or orientation may significantly reduce biological activity.

An alternative strategy consists of genetically encoding short peptide motifs that serve as substrates for site-specific protein conjugation. Systems such as Sortase A (SrtA) mediated transpeptidation (17–19) and the SpyTag/SpyCatcher platform (20, 21) enable the covalent attachment of cargos in a controlled and directional manner. These approaches have been successfully applied to a variety of protein nanoparticles and VLP platforms, including bacteriophages (22), hepatitis B virus (23), and rod-shaped viruses such as papaya mosaic virus (17), enabling post-assembly functionalization while preserving nanoparticle integrity. In contrast to conventional chemical coupling methods, enzymatic conjugation provides greater control over reaction specificity, product homogeneity, and ligand orientation.

For biomedical applications, the scalable production of functionalized RNA-containing VLPs remains an important challenge. Although in vitro assembly of TMV particles provides precise control over assembly conditions and cargo composition, large-scale manufacturing requires the independent production and purification of both CP and RNA cargo, followed by controlled assembly reactions. Such multistep procedures complicate industrial implementation and increase production costs. Consequently, there is considerable interest in developing systems capable of producing conjugatable RNA-loaded TMV particles in cellulo, i.e., directly within living cells.

Assembly competence of TMV CP depends on post-translational processing that occurs in eukaryotic hosts. Specifically, removal of the initiator methionine followed by N-terminal acetylation of the exposed serine residue is required for efficient particle assembly. Consequently, bacterial expression systems are generally unable to produce assembly-competent TMV CP, whereas eukaryotic hosts such as plants, insect cells, and yeast can support proper maturation (24, 25). The methylotrophic yeast Komagataella phaffii (Pichia pastoris) has emerged as a particularly attractive production platform due to its rapid growth, scalability, ease of genetic manipulation, and ability to perform the post-translational modifications required for TMV assembly (26). Furthermore, because TMV assembly is initiated by recognition of the OAS, the system offers the possibility of directing the selective packaging of engineered RNA cargos.

Nodamura virus (NoV) is an insect-derived alphanodavirus with a compact, two-molecule positive-sense RNA genome comprising RNA1 (3,204 nt) and RNA2 (1,336 nt). RNA1 encodes protein A, the viral RNA-dependent RNA polymerase (RdRp), and can replicate autonomously in the absence of RNA2 while producing the subgenomic RNA3, which can be adapted for heterologous gene expression (27, 28). Because NoV RNA replication and replication-dependent expression have been demonstrated in yeast (29, 30), the compact RNA1 replicon provides an attractive system for intracellular amplification of engineered RNA cargos.

In this work, we combined the NoV RNA1-derived replicon with an engineered TMV assembly system in K. phaffii. The replicon was designed to express luciferase and contain the TMV OAS, while the TMV CP was engineered with a C-terminal LPETG motif enabling site-specific functionalization through SrtA-mediated conjugation of N-terminal polyG ligands. Using K. phaffii as a production host (see Figure 1), we demonstrate the intracellular assembly of TMV nanorods containing heterologous RNA transcripts that include the TMV OAS. We show further that the engineered LPETG motif remains accessible after particle assembly, enabling the conjugation of fluorescent dyes and targeting affibodies to intact TMV nanorods, as quantified by SDS PAGE gels and their fluorescence imaging. Finally, gold labeling of the affibody and transmission electron microscopy (TEM) confirms the successful surface presentation of conjugated ligands on assembled particles. Together, these results open the field toward the design of a scalable platform for the production and post-assembly functionalization of TMV nanorods and provide a versatile framework for the development of customizable viral nanomaterials for biotechnology and nanomedicine applications.

Reagents

Unless otherwise stated, reagents were purchased from Thermo Fisher Scientific (Waltham, MA, USA). Salts and general laboratory chemicals were obtained from Sigma-Aldrich (St. Louis, MO, USA). Restriction enzymes, DNA polymerases, and competent cells were purchased from New England Biolabs (Ipswich, MA, USA).

Plasmid construction

Plasmid pIB4-TMV-LPETG-S was designed to express the TMV CP containing a C-terminal GGGS linker followed by the SrtA recognition motif LPETG under control of the AOX1 promoter. The pIB4-TMV-LPETG-S was generated from a parent plasmid containing a C-terminal cutsite-histag (pIB4-TMV-LPETG-/-HT) using primers Fwd- CTGCAGGCATGCAAGCTTCTTAGAC and Rev- CTATGACCCAGTCTCGGGCAAAGATCC. This parent plasmid was created using a gBlock (Integrated DNA Technologies, IDT) containing TMV-LPETG-/-HT insert using primers Fwd- AAGGAATTCATGTCATACTCCATTACAACGCCATCACAATTTG and Rev- TACCTTGGAGATTCAATTCATATATAATTACGCCTGCAGC which contained restriction sites EcoRI and PstI for cloning into a pIB4 backbone (Addgene: 25453, Watertown, MA USA).

The RNA expression plasmid pPICZa-Nod-Luc-OAS contained NoV-RdRp-Luc-OAS. The expression cassette was amplified from plasmid pNod-Luc-OAS (6) and cloned into pPICZa (Addgene: 78171) using KpnI and XbaI restriction sites.

All plasmids were transformed into NEB 5-alpha competent E. coli cells (Ipswich, MA, US to New England Biolabs) and verified by whole-plasmid sequencing.

Yeast transformation

Plasmids pIB4-TMV-LPETG-S and pPICZa-Nod-Luc-OAS were co-transformed into K. phaffii his4, Mut + autotrophic strain GS115 by electroporation. Transformants were selected on media lacking histidine and supplemented with Zeocin. For CP purification experiments, pIB4-TMV-LPETG-HT was transformed independently into K. phaffii and selected under the auxotrophic conditions.

Expression and purification of TMV CP-LPETG-HT

K. phaffii cells transformed with plasmid pIB4-TMV-LPETG-HT were grown in Buffered Glycerol-complex Medium (BMGY) supplemented with 100μg/ml Ampicillin at 30 °C and 200 rpm until reaching an OD600 of approximately 4 A.U. Cells were harvested by centrifugation at 3000 × g for 5 min and resuspended at 10x the pre-culture volume in induction medium consisting of Buffered Methanol-complex Medium (BMMY) supplemented with 100μg/ml Ampicillin and 1%v/v methanol. Protein expression was induced for 5 days at 30 °C with shaking at 200 rpm. Methanol was added every 24 h to maintain induction. Following induction, cells were harvested by centrifugation at 3000 × g for 10 min at 4 °C. Cell pellets were stored at −80 °C until purification.

Frozen cell pellets were resuspended in lysis buffer containing 50 mM Tris-HCl (pH 7.5), 1M NaCl, 10mM imidazole, and 1 mM PMSF at a 1:3 (w/v ratio). Cells were lysed using an EmulsiFlex homogenizer (Avestin, Ottawa, Canada) operated at 20–25 kpsi for 2–3 passes. Cell debris was removed by centrifugation at 16000 × g for 44 min at 4 °C.

The clarified lysate was loaded onto a nickel-affinity column HisPur™ Ni-NTA Resin, Thermofisher (Waltham, MA, USA) pre-equilibrated with binding buffer (50 mM Tris-HCl pH 7.5, 1M NaCl). The column was washed with 3 column volumes of High Salt Buffer (HSB 50 mM Tris-HCl pH 7.5, 4M NaCl). Then the column was washed with 3 column volumes of wash buffer containing 25 mM imidazole. Finally, the bound proteins were eluted using 250 mM imidazole. Fractions containing TMV-LPETG-HT CP were pooled, exchanged into 50 mM Tris-HCl pH 7.5, 1M NaCl, and 1 mM PMSF buffer using Spectra/por dialysis membrane MWCO 3.5kD (Repligen Corporation, Waltham, MA, USA) for 24 h, and analyzed by SDS-PAGE.

Production and purification of TMV nanorods

K. phaffii cells co-transformed with pIB4-TMV-LPETG-S and pPICZa-Nod-Luc-OAS were cultured under conditions similar to those described above for CP expression. Briefly, cells were grown in BMGY medium to an OD600 of approximately 4 A.U., pelleted and then diluted 10x in BMMY and subsequently. Subsequently induced with 1% (v/v) methanol. Methanol was replenished every 24 h for a total induction period of 5 days.

Cells were harvested by centrifugation at 3,000 × g for 10 min at 4 °C and resuspended in 50 mM MES buffer (pH 6.0). Cell disruption was performed using an EmulsiFlex homogenizer (Avestin, Ottawa, Canada) operated at 20–25 kpsi for 2–3 passes. The lysate was clarified by centrifugation at 14,000 × g for 30 min at 4 °C.

Clarified lysates were layered onto a 20%/70% (w/v) sucrose gradient prepared in 50 mM MES buffer (pH 6.0) and centrifuged at 28,000 × rpm for 2 h at 6 °C using an SW28 rotor (Beckman Coulter, Brea, CA, USA). Following centrifugation, visible bands containing TMV nanorods were collected and dialyzed against 50 mM MES buffer (pH 6.0) using Spectra/Por ® dialysis membrane tubing (6–8 kDa MWCO; Repligen Corporation, Waltham, MA, USA) to remove sucrose.

Samples were subsequently purified by CsCl isopycnic density gradient centrifugation. Gradients consisting of 42%, 49%, 57%, and 65% (w/v) CsCl were prepared as previously described (16). Gradients were centrifuged at 32,000 × rpm for 16 h at 6 °C using an SW41 Ti rotor (Beckman Coulter, Brea, CA, USA). Visible particle bands were collected and dialyzed against 50 mM MES buffer (pH 6.0) using Spectra/Por ® dialysis membrane tubing (6–8 kDa MWCO; Repligen Corporation, Waltham, MA, USA). Purified TMV nanorods were stored at 4 °C until further characterization.

Production and purification of polyG-affibody

A plasmid encoding the ZHER2:342 affibody containing an N-terminal 6×His tag, a TEV protease cleavage site, a pentaglycine sequence, and a glycine/serine linker was synthesized by GenScript (Piscataway, NJ, USA). The encoded protein sequence was: MGSSHHHHHHHSSGENLYFQGGGGGSGGGGSGGGGSVDNKFNKEMRNAYWEIALLPNLNNQQKRAFIRSLYDDPSQSANLLAEAKKLNDAQAPKGC.

The plasmid was transformed into E. coli BL21(DE3) cells. Cultures were grown in 2×YT medium supplemented with 2% glucose and 100 μg/mL ampicillin at 37 °C and shaking at 220 rpm until reaching an OD600 of approximately 0.6 A.U. Protein expression was induced by addition of 1 mM IPTG, and cultures were incubated for an additional 4 h at 37 °C.

Cells were harvested by centrifugation at 9000 × g for 15 min at 4 °C and stored at −80 °C until purification.

For purification, cell pellets were thawed on ice and resuspended in denaturing lysis buffer consisting of 7 M urea, 100 mM NaH 2 PO 4, and 100 mM Tris-HCl (pH 8.0) supplemented with 100 U/mL Benzonase at a ratio of 5 mL of lysis buffer per gram of wet cell pellet. Samples were incubated with gentle stirring at room temperature for 30 min and subsequently centrifuged at 10,000 × g for 30 min.

The His-tagged affibody was purified under denaturing conditions using a Ni-NTA affinity column (Thermo Fisher Scientific, Waltham, MA, USA). The column was washed with three column volumes of denaturing wash buffer (7 M urea, 100 mM NaH2PO4, 100 mM Tris-HCl, pH 6.3), and bound protein was eluted using denaturing elution buffer (8 M urea, 100 mM NaH 2 PO 4, 100 mM Tris-HCl, pH 4.5).

Purified affibody was dialyzed overnight against PBS and subsequently treated with TEV protease at a ratio of 1 mg TEV protease per 100 mg affibody for 12 h at 4 °C. DTT was then added to a final concentration of 20 mM, and the sample was concentrated using a 10 kDa MWCO Amicon Ultra centrifugal filter (MilliporeSigma, Burlington, MA, USA). Cleaved monomeric affibody was recovered from the flow-through and stored at −80 °C until use.

Production and purification of SrtA

The plasmid pET30b-7+-SrtA (Addgene plasmid #105602) was transformed into E. coli BL21(DE3) cells. Cultures were grown in LB medium supplemented with 50 μg/mL kanamycin at 37 °C until reaching an OD600 of approximately 0.6.

Protein expression was induced with 0.5 mM IPTG, and cultures were incubated for an additional 18 h at 20 °C. Cells were harvested by centrifugation at 16,000 × g for 45 min and resuspended in lysis buffer containing 50 mM Tris-HCl (pH 7.6), 1 M NaCl, and 1 mM PMSF. Cells were disrupted using an EmulsiFlex homogenizer (Avestin, Ottawa, Canada), and insoluble material was removed by centrifugation at 16,000 × g for 45 min at 4 °C.

SrtA was purified using HisPur™ Ni-NTA resin (Thermo Fisher Scientific, Waltham, MA, USA). The column was washed with three column volumes of HSB followed by three column volumes of wash buffer containing 25 mM imidazole. Bound protein was eluted using 250 mM imidazole. Fractions containing SrtA were pooled, concentrated using 10 kDa MWCO centrifugal filters, exchanged into SrtA reaction buffer, and stored at −80 °C until use.

SrtA-mediated conjugation

SrtA-mediated conjugation reactions were performed using purified TMV CP-LPETG or purified TMV-LPETG nanorods as substrates. Reactions contained 30 μM TMV CP-LPETG subunits, 150 μM glycine-modified ligand, and 6 μM SrtA in reaction buffer consisting of 50 mM MES (pH 6.0) and 150 mM NaCl. Glycine-modified ligands included either a polyG-affibody or an Alexa Fluor 488-labeled Sortagging peptide, C-terminal Sortagging-[Cys(AF488)] (Biosynth, Naperville, IL,USA).

The molar ratio of TMV CP-LPETG subunits: ligand: SrtA was maintained at 1:5:0.2. Reactions were incubated at room temperature for 2 h protected from light. Control reactions lacking either SrtA or glycine-modified ligand were prepared under identical conditions. Following incubation, samples were analyzed directly by SDS-PAGE or prepared for TEM.

RT-PCR analysis

Before RNA was extracted for RT-PCR analysis, VLP samples were treated with RNaseA. For this, the extinction coefficient of WT TMV virus (6) was used to infer the theoretical RNA content in the samples. Then, VLPs were incubated with RNAseA at a 1 to 0.05 mass ratio of RNA to RNAseA at 37C for 30 min. Finally, VLPs were amicon washed using a 100 MWCO centrifugal filter to remove non-protected/digested RNA and free RNAseA. RNA associated with purified TMV nanorods was extracted using QIAamp Viral RNA Mini Kit purification kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. RNA concentration and purity were determined by absorbance measurements at 260 and 280 nm.

Reverse transcription was performed using M-MuLV reverse transcriptase and gene-specific primer Rev- 5’-ACCACTGGCATAAGCCTAGTTCGGG-3’. Reactions were carried out according to the manufacturer’s recommendations.

Two PCR assays were performed to detect RNA species derived from the NoV4 construct. Primer pair Fwd- 5’-ATGCTGAACTACGAGACAATCATCAACGG-3’/Rev- 5’-ACCACTGGCATAAGCCTAGTTCGGG-3’ generated a 4.5 kb amplicon specific for the full-length NoV-RdRp-Luc-OAS transcript, whereas primer pair Fwd- 5’-ATGAGGTTAAGAGCTAATCAAGTCACTTCC-3’/Rev-5’ACCACTGGCATAAGCCTAGTTCGGG-3’ generated a 1.8 kb amplicon corresponding to the subgenomic RNA. PCR reactions were performed using Phusion™ High-Fidelity DNA Polymerase under the following conditions: initial denaturation at 98 °C for 30 sec, followed by 35 cycles of 98 °C for 30 s, 65 °C for 30 s, and 72 °C for 4 min, with a final extension at 72 °C for 4 min.

Amplification products were resolved on 0.8% agarose gels prepared in TAE buffer and stained with GelRed (Biotium, Fremont, CA, USA) prior to imaging.

SDS-PAGE analysis

Protein samples were mixed with Laemmli Sample Buffer (Bio-Rad, Hercules, CA, USA) and heated at 95 °C for 10 min prior to electrophoresis. Samples were separated on 4–20% Mini-PROTEAN ® TGX™ Precast Protein Gels (Bio-Rad, Hercules, CA, USA) using Tris/Glycine/SDS running buffer (Bio-Rad, Hercules, CA, USA) at 120 V for 60 min.

Following electrophoresis, gels were stained with Bio-Safe™ Coomassie Stain (Bio-Rad, Hercules, CA, USA) and destained in deionized water for 48 h. Gel images were acquired using an Azure 200 imaging system (Azure Biosystems, Dublin, CA, USA).

For polyG-Alexa conjugation experiments, fluorescent images were acquired prior to Coomassie staining using the Epi-Blue illumination module of the Azure 200 imaging system. Fluorescence was detected from the Alexa Fluor 488-labeled glycine-containing Sortagging peptide [Cys(AF488)] used as the SrtA substrate.

TEM

Purified TMV nanorods were adsorbed onto carbon-coated copper grids (300 mesh; Ted Pella, Redding, CA, USA) for 1 min. Excess sample was removed using filter paper, and the grids were negatively stained with 2% (w/v) uranyl acetate for 1 min. Grids were allowed to air dry prior to imaging.

Samples were examined using a Talos TEM (Thermo Fisher Scientific, Waltham, MA, USA) operated at 200 kV. Images were collected using a Thermo Scientific Ceta 16M camera (4k × 4k resolution) at nominal magnifications ranging from 72,000× to 92,000×.

Particle lengths were measured using ImageJ software (National Institutes of Health, Bethesda, MD, USA). A total of 200 particles were analyzed to generate the length distribution histogram.

Mass spectrometry

Intact TMV CP-WT purified from TMV virus (NC_597750.1) and His-tagged TMV CP-LPETG were analyzed under denaturing conditions by reverse-phase liquid chromatography–mass spectrometry (LC-MS) using an Agilent C3 column coupled to an Agilent 6530 quadrupole time-of-flight (Q-TOF) mass spectrometer (Agilent Technologies, Santa Clara, CA, USA). Prior to analysis, samples were buffer-exchanged three times into 0.1% (v/v) formic acid in water, and protein concentrations were determined using a NanoDrop spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). A total of 2 µg of each protein sample was injected for intact mass analysis.

Proteins were ionized by positive electrospray ionization (ESI+), and mass spectra were deconvoluted using Agilent MassHunter software (Agilent Technologies, Santa Clara, CA, USA).

Theoretical molecular weights were calculated from the corresponding protein sequences, taking into account removal of the initiator methionine and N-terminal acetylation of the mature CP. Experimental masses were compared with theoretical values to confirm protein identity and the presence of the engineered LPETG modification.

Gold nanoparticle labeling

Affibody-functionalized TMV nanorods were incubated with 5 nm maleimide-activated gold nanoparticles (Cytodiagnostics, Burlington, ON, Canada) in 50 mM HEPES, 100 mM NaCl buffer (pH 6.8) for 1 h at room temperature followed by overnight incubation at 4 °C. Gold nanoparticles were added at a molar ratio of 100:1 relative to TMV CP subunits.

Labeled particles were deposited onto carbon-coated copper grids and analyzed by TEM. To increase the visibility of bound gold nanoparticles, grids were treated with GoldEnhance™ EM Plus (Nanoprobes, Yaphank, NY, USA) according to the manufacturer’s instructions for 10 min. Following enhancement, grids were negatively stained with 2% (w/v) uranyl acetate for 1 min, allowed to air dry, and subsequently imaged by TEM.

Yeast expression of TMV CP containing a C-terminal LPETG motif

To enable site-specific post-assembly functionalization of TMV nanorods, the TMV CP was engineered to contain a C-terminal LPETG motif recognized by SrtA. The LPETG sequence was selected because it provides an effective enzymatic recognition site while minimizing disruption of the native TMV CP structure. In parallel, a self-amplifying RNA was designed to promote intracellular assembly of TMV nanorods in yeast. This RNA encodes the NoV RdRp and a luciferase (Luc) reporter and contains the TMV OAS to promote preferential encapsidation by the TMV CP (NoV-RdRp-Luc-OAS).

The corresponding expression plasmids were sequence-verified (Supplementary Material) and are depicted in Figure 2. pIB4-TMV-LPETG (Figure 2A) encodes the TMV CP-LPETG. pPICZα-NodRep_Luc_OAS (Figure 2B) produces NoV-RdRp-Luc-OAS. Both constructs are transcribed under the control of the alcohol oxidase 1 (AOX1) promoter. In plasmid B, a hepatitis delta virus (HDV) ribozyme is positioned at the 3′ end of the transcript to generate a defined RNA terminus. The Thosea asigna virus 2A (T2A) ribosome-skipping sequence enables the RdRp and Luc to be produced as two separate functional proteins.

The 5′ and 3′ untranslated regions (UTRs) of the NoV replicon are retained to support recognition and replication by the RdRp. Following translation from the initial positive-sense transcript, the RdRp uses this RNA as a template to synthesize a full-length negative-sense replication intermediate. The negative-sense RNA subsequently serves as a template for the production of additional full-length positive-sense replicon RNAs. In addition, a subgenomic promoter present in the negative-sense template directs the synthesis of an approximately 1,800-nucleotide positive-sense subgenomic RNA corresponding to the 3′ region of the plasmid B transcript.

Expression of the engineered TMV CP-LPETG was first evaluated using a construct containing a C-terminal His-tag preceded by a proTEV cleavage site. A wild-type (TMV-CP WT) carrying the same C-terminal His-tag was also expressed and purified for use as a control. Following expression in K. phaffii, both proteins were purified using a nickel-affinity column.

The purified TMV CP-WT-His and TMV CP-LPETG-His were subsequently analyzed by mass spectrometry. The measured molecular weights were consistent with the expected sequences and confirmed the successful expression of both proteins, including the presence of the engineered C-terminal LPETG motif in the TMV CP-LPETG variant (Supplementary Figure S1).

To determine whether the LPETG motif remained accessible and functional following expression in yeast, purified TMV CP-WT-His and TMV CP-LPETG-His were subjected to SrtA-mediated ligation using a polyG-modified affibody as a model substrate. SDS-PAGE analysis revealed the formation of a higher-molecular-weight product exclusively in reactions containing the TMV CP-LPETG, consistent with successful SrtA-mediated conjugation (Supplementary Figure S2). No corresponding product was observed for the TMV CP-WT control. Together, these results demonstrate that the engineered TMV CP-LPETG can be efficiently produced in yeast while retaining its capacity for site-specific SrtA-mediated functionalization.

Production and purification of TMV nanorods produced in K. phaffii

Following validation of CP expression and conjugation capability, plasmids encoding the TMV CP-LPETG and the OAS-containing RNA construct were co-transformed into K. phaffii. Co-expression of these constructs resulted in the production of TMV-derived nanorods that could be isolated from yeast lysates.

To purify the TMV nanorods, yeast lysates were subjected to sucrose cushion ultracentrifugation followed by purification using a CsCl isopycnic density gradient. Purified samples were analyzed by TEM, revealing rod-shaped particles with morphologies characteristic of TMV nanorods (Figures 3A, B). Length analysis revealed a heterogeneous population of particles with a bimodal distribution, displaying major peaks centered at approximately 100 and 200 nm (Figure 3C).

Because TMV particle length is determined by the length of the encapsidated RNA (specifically, 4.7nm of VLP length per 100nt of RNA), observation of two different nanorod lengths suggests the incorporation of two different RNA species into the purified nanorods. The larger particle population is consistent with encapsidation of the full-length NoV RdRp-Luc-OAS transcript 4563 nt, which is predicted to generate particles approximately 211 nm in length. In contrast, the shorter particle population is consistent with encapsidation of a shorter RNA species, such as a subgenomic RNA derived from the NoV construct 1825 nt, which is predicted to generate particles approximately 84 nm in length. The correspondence between the observed and predicted particle lengths suggests that the intended RNA construct is incorporated into the TMV nanorods. Furthermore, the presence of a shorter particle population is consistent with the production of subgenomic RNA species (which can only arise if replication occurs), although additional experiments would be required to directly confirm their identity and determine whether they arise from active replication of the NoV replicon. While some degree of particle fragmentation during purification cannot be excluded, the presence of two predominant rod-length populations suggests that RNA length contributes substantially to the observed heterogeneity.

To determine whether RNA derived from the NoV construct was associated with the purified particle preparations, RNA was extracted from purified TMV nanorods and analyzed by Reverse Transcription Polymerase Chain Reaction (RT-PCR). Two primer sets were used to amplify distinct regions of the NoV RdRp–Luc–OAS construct. The first primer pair generated a 4.5-kb amplicon corresponding to the full-length transcript, whereas the second generated a 1.8-kb amplicon encompassing the region present in the expected subgenomic transcript. Because this region is also present in the full-length RNA, the 1.8-kb product cannot independently demonstrate replication or confirm the presence of a subgenomic transcript. Nevertheless, these results confirm that RNA derived from the NoV-RdRp–Luc–OAS construct was present in the purified TMV nanorod preparations. Two amplicons were generated by RT-PCR and sent for sequencing, both matching the expected lengths and regions of the NoV-RdRp-Luc-OAS transcript (Supplementary Figure S3).

These results indicate that RNA species consistent with both the full-length NoV RdRp-Luc-OAS transcript and a shorter RNA species derived from it were present within purified TMV nanorod preparations. Together with the observed bimodal particle length distribution, these findings reflect the incorporation of RNA molecules of different lengths into the purified TMV nanorods.

SrtA-mediated functionalization of assembled TMV nanorods

The accessibility of the engineered LPETG motif following nanorod production was evaluated using SrtA-mediated conjugation. Two ligands containing N-terminal glycine residues were investigated: a fluorescent polyG-Alexa488 probe and a polyG-affibody (7.9 kDa).

Functionalization with the polyG-Alexa probe was evaluated by SDS-PAGE followed by Coomassie staining and fluorescence imaging (Figures 4A, B). A fluorescent signal in 4B associated with the conjugated (~20-kDa) TMV CP-LPETG band (white asterisk) was detected only in the complete reaction containing TMV-LPETG nanorods, polyG-Alexa, and SrtA (lane 8). In the corresponding Coomassie-stained gel (4A) this band (red asterisk) is seen to be weak compared to the unconjugated CP, indicating a low level of functionalization. Conjugation levels up to 80% are achieved at higher mole ratios of dye-to-CP, but we’ve kept it low here so that the fluorescent bands in lane 8 of Figure 4B are non-overlapping. This result supports successful SrtA-mediated attachment of the fluorescent probe to the TMV CP-LPETG. A diffuse, slower-migrating fluorescent band at approximately 37 kDa was observed in all lanes containing the polyG-Alexa probe (lanes 2, 4, 6, and 8). This band was present even in samples lacking TMV-LPETG nanorods or SrtA because it corresponds to unconjugated polyG-Alexa molecules.

To further evaluate the versatility of the platform, TMV-LPETG nanorods were conjugated to a polyG-affibody. In this case SDS-PAGE analysis (Figure 4C) revealed a strong higher-molecular-weight band at approximately 27.9 kDa (see red asterisk there) in the complete reaction containing TMV-LPETG nanorods, polyG-affibody, and SrtA (lane 8). The molecular weight of this band is consistent with covalent attachment of the 7.9-kDa affibody to the approximately 20-kDa TMV CP-LPETG and is absent in all other lanes because these samples are lacking in one or another of the three reagents needed for conjugation. Further, in contrast with the case of polyG-Alexa functionalization (4A) the high intensity of the red-asterisk band relative to the unligated CP band - around an 80% conjugation efficiency - reflects a high level of sortase-mediated conjugation of the LPETG nanorods.

Affibody-functionalized TMV nanorods were subsequently examined by TEM. Rod-shaped particles remained readily detectable following conjugation, consistent with preservation of the characteristic nanorod morphology (Figure 4D). Together, these results demonstrate that the C-terminal LPETG motif remains accessible on TMV nanorods produced in K. phaffii and enables post-production functionalization with both fluorescent probes and protein ligands.

Gold nanoparticle labeling confirms surface accessibility of affibody-functionalized TMV nanorods

To verify that affibody molecules remained accessible following conjugation to assembled TMV nanorods, affibody-functionalized particles were incubated with 5-nm diameter maleimide-functionalized gold nanoparticles. Because the affibody contains an exposed cysteine residue, the gold particles could be covalently attached through thiol-maleimide coupling, and the resulting particles analyzed by TEM (Figure 5A). Visualization of the small 5-nm gold nanoparticles by TEM was challenging, so GoldEnhance™ EM Plus was applied to the grids to enlarge the bound gold particles prior to imaging. Following enhancement, multiple gold nanoparticles were observed associated with individual TMV nanorods (Figures 5B, C). The localization of gold nanoparticles along the surface of the rods indicates that the affibody molecules remained exposed and accessible following SrtA-mediated conjugation.

These results provide direct structural evidence that affibodies displayed on TMV nanorods remain accessible for subsequent chemical modification. Furthermore, they demonstrate that the LPETG-based functionalization strategy enables surface presentation of proteins on TMV nanorods produced in K. phaffii while preserving particle morphology and ligand accessibility.

In this study, we developed a yeast-based platform for the production and functionalization of TMV nanorods. By engineering the TMV CP to contain a C-terminal LPETG motif and co-expressing it with an RNA construct containing the TMV OAS, we demonstrated the production of TMV nanorods in K. phaffii that preferably contains the RNA of interest. Furthermore, we showed that the LPETG motif remained accessible following nanorod formation, enabling efficient SrtA-mediated conjugation of small molecules (fluorescent dyes) and proteins (affibodies).

The production of TMV-derived particles in heterologous expression systems has attracted considerable interest because it offers an alternative to plant-based production for providing greater control over genetic manipulation and scalability. Previous studies have shown that assembly-competent TMV CP requires eukaryotic post-translational processing, particularly N-terminal acetylation, limiting the use of bacterial expression systems. Consistent with these observations, expression of TMV CP-LPETG in K. phaffii resulted in the formation of rod-shaped particles with morphologies characteristic of TMV nanorods. These findings further support the use of yeast as a practical production platform for engineered TMV-based nanomaterials.

Analysis of purified particles revealed a heterogeneous population of nanorods with two predominant length distributions centered at approximately 100 and 200 nm. Because TMV particle length is determined by the length of the encapsidated RNA (specifically, 4.7nm of capsid length per 100nts of RNA), these observations are consistent with incorporation of RNA species having approximate lengths of 2130 and 4260nt. The larger-particle population closely correlates within a few percent with the length (211 nm) predicted for encapsidation of the full-length NoV-RdRp-Luc-OAS transcript (4560nt), whereas the broader shorter-particle population is consistent with encapsidation of subgenomic RNA and with unavoidable breakage of the longer (211)nm rods.

A major objective of this work was to develop a platform that enables controlled functionalization of TMV nanorods following production. Traditional TMV modification strategies frequently rely on nonspecific chemical conjugation to naturally occurring surface residues scattered throughout the TMV CP and the protein (e.g., targeting antibody, or antigenic protein), which can generate heterogeneous products and frustrate control over ligand orientation and function. In contrast, SrtA-mediated ligation offers a site-specific and modular approach for nanoparticle functionalization. The successful conjugation of both an N-terminal polyG-Alexa fluorescent probe and an N-terminal polyG-affibody demonstrates that the engineered LPETG motif remains accessible after nanorod formation and can serve as an efficient substrate for enzymatic modification. Importantly, TEM confirmed that the characteristic rod-like morphology of the particles was preserved following conjugation.

The accessibility of the displayed affibody was further validated through labeling with maleimide-functionalized gold nanoparticles. Because the gold nanoparticles react with an exposed cysteine residue present on the affibody, the observed decoration of the TMV nanorods provides direct evidence that the conjugated protein remains surface exposed and chemically accessible following its SrtA-mediated ligation. These results demonstrate that the LPETG platform is compatible not only with small-molecule labeling but also with the display of functional protein ligands.

Although the present study focused on proof-of-concept functionalization using fluorescent probes and affibodies, the approach should be readily adaptable to a wide range of targeting ligands, enzymes, antibodies, and imaging agents. Furthermore, the ability to produce RNA-containing TMV nanorods directly in yeast may simplify future manufacturing workflows by eliminating the need for separate large-scale production and in vitro assembly of CP and RNA cargo. Future studies will focus on improving RNA packaging specificity, further characterizing the RNA species incorporated into the particles, and using the sortase-activity of the VLPs to test the cellular uptake efficiencies of different targeting ligands and to evaluate the in vivo biological activity of functionalized TMV nano rods.

In summary, this work establishes a yeast-based platform for the production of RNA-containing TMV nanorods that can be site-specifically functionalized through SrtA-mediated ligation. By combining intracellular particle production, RNA incorporation, and modular post-production surface functionalization, this system expands the toolkit available for the development of TMV-based nanomaterials and provides a foundation for future applications in targeted delivery, molecular imaging, and synthetic virology.