Proteome-Wide Overexpression of Host Proteins for Identification of Factors Affecting Tombusvirus RNA Replication: an Inhibitory Role of Protein Kinase C

Abstract

To identify host genes affecting replication of Tomato bushy stunt virus (TBSV), a small model positive-stranded RNA virus, we overexpressed 5,500 yeast proteins individually in Saccharomyces cerevisiae, which supports TBSV replication. In total, we identified 141 host proteins, and overexpression of 40 of those increased and the remainder decreased the accumulation of a TBSV replicon RNA. Interestingly, 36 yeast proteins were identified previously by various screens, greatly strengthening the relevance of these host proteins in TBSV replication. To validate the results from the screen, we studied the effect of protein kinase C1 (Pkc1), a conserved host kinase involved in many cellular processes, which inhibited TBSV replication when overexpressed. Using a temperature-sensitive mutant of Pkc1p revealed a high level of TBSV replication at a semipermissive temperature, further supporting the idea that Pkc1p is an inhibitor of TBSV RNA replication. A direct inhibitory effect of Pkc1p was shown in a cell-free yeast extract-based TBSV replication assay, in which Pkc1p likely phosphorylates viral replication proteins, decreasing their abilities to bind to the viral RNA. We also show that cercosporamide, a specific inhibitor of Pkc-like kinases, leads to increased TBSV replication in yeast, in plant single cells, and in whole plants, suggesting that Pkc-related pathways are potent inhibitors of TBSV in several hosts.

INTRODUCTION

Replication of plus-stranded-RNA [(+)RNA] viruses requires many components of the host cells, including host proteins and intracellular membranes, which serve as sites of virus replication in infected cells (9, 26, 33, 38, 39, 42, 45, 64). Major advances in cataloging the host factors affecting (+)RNA virus infections have recently been made with several animal and plant viruses (7, 10, 19, 23, 24, 37, 44, 48, 62, 69, 70, 76, 78, 79), yet our knowledge of how many host factors are involved in viral RNA replication is still far from complete.

Tombusviruses, such as Tomato bushy stunt virus (TBSV) and Cucumber necrosis virus (CNV), are single-component RNA viruses with ∼4,800 nucleotides. Only the replication proteins p33 and p92^pol, among the five virus-coded proteins, are essential for TBSV replication (41, 87). p92^pol is the viral RNA-dependent RNA polymerase (RdRp), whereas replication cofactor p33 (which overlaps the N-terminal prereadthrough segment of p92^pol) is an RNA-binding protein and an RNA chaperone (49, 56, 59, 73). Earlier work determined that p33 is involved in template selection and recruitment of viral RNA into replication (34, 46, 56). These proteins interact with each other, with the viral RNA, and with a group of host proteins in cells (25, 27, 32, 46, 51, 60, 61, 67) that leads to the assembly of viral replicase complexes (VRC) on peroxisomal membranes (31, 43, 46).

Systematic genome-wide screens were conducted in yeast (Saccharomyces cerevisiae), a model host, using nonoverlapping gene libraries covering ∼95% of the yeast genome to identify the roles of host genes in TBSV replication (19, 48, 68, 69). These studies led to the identification of ∼150 host genes that either stimulated or inhibited virus replication and RNA recombination. Additional global proteomics approaches, such as protein arrays, a cDNA library screen and mass spectrometry of the purified tombusvirus replicase, have led to the identification of an additional ∼150 host proteins that interact with either p33 and p92 or the TBSV RNA (25, 27, 32, 67). Functions of several of the identified host proteins in TBSV replication have been dissected in yeast and in vitro as well as validated in a native plant host (3, 13, 16, 17, 28, 38, 40, 65, 84, 88).

In spite of the intensive genome-wide and global proteomics screens for TBSV host factors, it seems that previous screens have not reached saturation level, since the overlap among the identified set of host genes from various screens is somewhat low, albeit significant (37, 40). Therefore, we are continuing systematic screening for TBSV host factors in yeast. Accordingly, in this study, we performed a proteome-wide screen with an overexpression library of yeast genes representing over 90% of the yeast proteome to gain further insights into the complexity of TBSV-host cell interaction. Overexpression of 5,500 yeast proteins led to the identification of 141 yeast proteins that affected the accumulation of tombusvirus replicon RNA (repRNA) in yeast. The identified yeast proteins, which either increased or decreased the accumulation of tombusvirus repRNA, are involved in protein metabolism and transport, RNA transcription and metabolism, or other cellular processes. Among these are 36 host proteins that have also been identified in earlier screens, thus validating the idea that these host genes are likely important for TBSV replication.

To further validate the screen results, we chose protein kinase C (Pkc1p), whose overexpression inhibited TBSV replication. The Pkc superfamily of kinases is found exclusively in eukaryotes. They are serine/threonine kinases frequently involved in signal transduction (66). The human Pkc isoenzymes are known to be involved in diseases, such as cancer, diabetes, and Alzheimer's disease (66). While there are four subgroups of Pkc-related kinases in mammals, there is only a single prototypic PKC gene in S. cerevisiae, which greatly facilitates mechanistic and functional studies. Importantly, the yeast Pkc1p contains all the domains identified in the mammalian isoenzymes (66). Pkc1p has been shown to be an essential and multifunctional enzyme regulating cell wall integrity, osmoregulation, conserved MAPK signal transduction, and lipid homeostasis via its effects on the Opi1p repressor of phospholipid biosynthesis, as well as actin filament and cytoskeleton organization and pexophagy (specialized form of autophagy leading to destruction of peroxisomes) (66). In contrast to animal and fungal PKCs, the plant Pkc-like proteins have not yet been characterized in detail (2, 20, 36, 75, 80–83).

In this work, we show that Pkc1p is a potent inhibitor of TBSV replication. Yeast carrying a temperature-sensitive Pkc1p mutant supported TBSV replication at a higher level than wild-type yeast did. Also, an in vitro approach has demonstrated that Pkc1p directly inhibits TBSV RNA synthesis. We also show that cercosporamide, a specific inhibitor of Pkc-like kinases, leads to increased TBSV replication in yeast, in plant single cells, and in whole plants, suggesting that Pkc-related pathways are potent inhibitors of TBSV in several hosts.

MATERIALS AND METHODS

Yeast strains and expression plasmids.

The parental yeast strain (BY4741) was from Open Biosystems. To study the effect of overexpression of selected yeast proteins on TBSV repRNA replication, we used the yeast open reading frame (ORF) collection from Open Biosystems. In this yeast ORF collection, each ORF is expressed from the 2μ plasmid BG1805 under the control of GAL1 promoter and fused to a tandem affinity tag that includes a hemagglutinin (HA) tag and the “zz” domain of protein A at the C terminus. We also used an N-terminally glutathione S-transferase (GST)-tagged ORF library for a limited overexpression screen (obtained from B. Andrews) (72).

The expression plasmid pGAD-His92 (containing CNV p92^pol gene and LEU2 marker) (50) and the dual expression plasmid pGBK-His33/DI-72 (co-expressing p33 from the ADH1 promoter and DI-72 RNA from the GAL1 promoter) have been previously described (19). Expression of nonphosphorylatable p33 mutants (A₂₁₀A₂₁₁, A₂₀₅A₂₁₀A₂₁₁, and D₂₀₅) in yeast was done as described previously (71).

Yeast transformation and cultivation.

Yeast strains were cotransformed with different combinations of plasmids using the lithium acetate (LiOAc)–single-stranded DNA (ssDNA)–polyethylene glycol (PEG) method (12), and transformants were selected by complementation of auxotrophic markers. For the replication assay, the parental strain (BY4741) was cotransformed with three separate plasmids: (i) pGAD-His92, (ii) pGBK-His33/DI-72 (19), and (iii) one of the individual yeast ORF clones (Open Biosystems) or the 2μ plasmid pYES-NT-C (Invitrogen) as a control.

Host protein overexpression studies.

Individual colonies of strain BY4741 transformed with plasmids carrying the selected ORFs under the control of the GAL1 promoter along with pGAD-His92 and pGBK-His33/DI-72 were pregrown overnight in SC-ULH⁻ medium (54) containing 2% glucose in 96-deep-well plates to suppress host protein expression and TBSV repRNA replication. To overexpress the particular host protein and launch TBSV repRNA replication, yeast transformants were transferred into 1.5 ml of SC-ULH⁻ plus 2% galactose for 24 h at 29°C in 96-deep-well plates. The final optical density (OD) was ∼0.7 to 1.0. For the detection of the expressed host and viral proteins, we performed Western blotting as described previously (50). We analyzed 9 to 18 independent samples for each host protein. Examination of a large number of samples was important for numerous host proteins that resulted in highly variable effects on TBSV RNA accumulation.

RNA analysis.

Total RNA isolation and Northern blot analysis were performed as described previously (47, 50). Briefly, for extraction of total RNA, yeast cells were broken by shaking for 1 to 2 min at room temperature with equal volumes of RNA extraction buffer (50 mM NaOAc [pH 5.2], 10 mM EDTA, and 1% sodium dodecyl sulfate [SDS]) and water-saturated phenol and then incubated for 4 min at 65°C, followed by ethanol precipitation. The obtained RNA samples were separated on a 1.5% agarose gel and transferred to a Hybond-XL membrane (Amersham) before hybridization with a DI-72-specific probe (47). For detection of plus-strand repRNA, we prepared a ³²P-labeled RIII/IV(−) probe with T7 transcription from a PCR product obtained with primers 1165 (AGCGAGTAAGACAGACTCTTCA) and 22 (GTAATACGACTCACTATAGGGCTGCATTTCTGCAATGTTCC) on DI-72 templates.

Protein analysis.

Protein analysis was done as described earlier (47, 50). Briefly, a total of 1 ml yeast culture was harvested, and the pelleted cells were resuspended in 150 μl cold extraction buffer (200 mM sorbitol, 50 mM Tris-HCl [pH 7.5], 15 mM MgCl₂, 10 mM KCl, 10 mM β-mercaptoethanol, yeast protease inhibitor mix; Sigma), and 250 μl of glass beads was added to each sample. The cells were broken with a Genogrinder for 2 min at 1,500 rpm. Each sample was further mixed with 600 μl prechilled extraction buffer, and unbroken cells were removed by centrifugation at 100 × g for 5 min. The supernatant was mixed with 0.5 volume of 3× SDS-polyacrylamide gel electrophoresis (PAGE) sample buffer followed by SDS-PAGE and Western blot analysis as described previously (47, 50). The primary antibody was anti-His₆ (Amersham), and the secondary antibody was alkaline phosphatase-conjugated anti-mouse IgG antibody (Sigma).

In vitro replication assay.

Yeast cell-free extract (CFE) was prepared as described earlier (55). The assay mixture consisted of 2 μl yeast extract, 400 ng of purified p33, 100 ng of p92, and 500 ng of DI-72(+) RNA along with 1× (∼200 ng), 2×, and 4× GST or recombinant PKC1 protein in 20 μl reaction mix. The conditions of the in vitro assay are described elsewhere (55). The reaction mixture was incubated at 25°C for 3 h. RNA was purified by phenol-chloroform extraction, followed by isopropanol-ammonium acetate precipitation. The newly synthesized ³²P-labeled repRNAs were analyzed on 5% denaturing PAGE gels as described previously (50).

Use of cercosporamide for Pkc1 inhibition in yeast.

The yeast strain BY4741 was cotransformed with pHISGBK-CUP1-p33ADH-DI-72 and pESC-Ura-CUP1-His-p92 as described above. Transformed yeast cells were inoculated and grown at 28°C in UH⁻ liquid medium without CuSO₄. After 24 h of growth at 28°C, each culture was divided into two portions, each of which had an OD at 600 nm (OD₆₀₀) of approximately 0.2. In one set of cultures, different concentrations of 100% ethanol (0.125, 0.25, and 0.5 μl/ml) were added as a control, while in the other set of cultures, 0.125, 0.25, or 0.5 μg/ml of cercosporamide (Sigma) was added. Yeast cells were grown for another 36 h at 28°C, and total RNA was analyzed as described earlier (6, 48, 50).

Use of cercosporamide for Pkc1 inhibition in protoplasts.

Preparation of Nicotiana benthamiana protoplasts, electroporation with TBSV and Turnip crinkle virus (TCV) RNA, and viral RNA analysis were performed as described previously (49). Cercosporamide was dissolved in ethanol and was added at concentrations of 2 and 4 μM (2- and 4-μg/ml final concentrations) before electroporation to the N. benthamiana protoplasts). Total RNA samples were obtained 40 h postelectroporation.

Plant inoculation with TBSV and treatment with Pkc1 inhibitor.

Lower leaves of 3-week-old N. benthamiana plants were sap inoculated (sap was purified from TBSV-infected leaves). Two days later, ethanol or cercosporamide (2- and 4-μg/ml final concentrations) were infiltrated into the newly emerging top leaves. Total RNA at 4 days postinoculation (dpi) from the infiltrated leaves was isolated and analyzed for TBSV RNA accumulation by Northern blotting as described earlier (50).

RESULTS

Systematic screening of the yeast overexpression ORF collection for host proteins affecting tombusvirus replication.

Using a high-throughput approach (54), we separately overexpressed 5,500 yeast proteins from the galactose inducible GAL1 promoter (11) in yeast cells also replicating TBSV repRNA. Each yeast protein is C-terminally tagged with zz and a His₆ tag to aid detection (11). Comparable amounts of yeast cells were harvested 24 h later, followed by Northern blotting to measure the level of TBSV (+)repRNA produced. We used rRNA as a loading control for normalization of data on repRNA accumulation in yeast. The accumulation level of repRNA in yeast carrying pYES plasmid, which expresses only a short peptide, was taken as 100%. As an additional control, we overexpressed a pseudogene (APT2) which has no enzymatic activity when expressed (1) and which failed to interact with p33 (25). Overexpression of Apt2p led to 80% ± 11% repRNA accumulation compared with yeast carrying the pYES control (data not shown) (25). This suggests that protein overexpression in general could reduce the ability of yeast cells to support TBSV repRNA accumulation under the protein overexpression conditions. Based on the pYES and APT2 overexpression controls, we considered overexpression of a protein inhibitory if it significantly reduced repRNA accumulation below 70% and stimulatory if it significantly increased repRNA accumulation above 130% of the wild-type (wt) level (i.e., in comparison with the pYES control).

Of the 5,500 yeast ORFs tested, we found that ∼1,300 had detrimental effects on both yeast cell growth (seen as a low rRNA level) and TBSV RNA accumulation (based on repRNA level). Since these host proteins likely affect TBSV repRNA accumulation indirectly via changing yeast metabolism due to their cytotoxicity when expressed at elevated levels, we did not consider these host proteins among the “winners,” which included only those that showed more selectivity in inhibition of TBSV repRNA accumulation than their effects on yeast cell growth. We also performed a limited screen with a GST-tagged yeast overexpression library (72) to extend the list of host proteins examined for their effects on TBSV replication.

Altogether, the proteome-wide screen led to the identification of 141 host proteins, which affected TBSV replication (Table 1; also, see the supplemental material). Among these, overexpression of 40 host proteins increased and 101 decreased TBSV accumulation in yeast (Table 1). Also, about 26% (36 of the “winner” proteins) have been identified previously by various screens (19, 25, 27, 32, 40, 48, 67–69), greatly strengthening the relevance of these host proteins in TBSV replication. Moreover, the screen also led to the identification of 105 new host proteins affecting TBSV replication.

Table 1.

List and known cellular functions of the identified yeast genes affecting TBSV replication when overexpressed

Genea	Product and functionc
ABP1	Actin-binding protein of the cortical actin cytoskeleton, important for activation of the Arp2/3 complex, which plays a key role in actin cytoskeleton organization
ACF4	Protein of unknown function; possible role in actin cytoskeleton organization
AFG2	ATPase of the CDC48/PAS1/SEC18 (AAA) family; may be involved in degradation of aberrant mRNAs
AFI1	Arf3p polarization-specific docking factor, required for the polarized distribution of the ADP-ribosylation factor
AIM14	Protein with similarity to iron/copper reductases (FRE1-8), possibly involved in iron homeostasis
ALG2	Presumed early mannosyltransferase involved in the N-linked glycosylation pathway
ALK2	Protein kinase; similar to mammalian haspins
APM1b	Mu1-like medium subunit of the clathrin-associated protein complex (AP-1); binds clathrin
ARP8b	Nuclear actin-related protein involved in chromatin remodeling; has mRNA binding activity
ARR3	Arsenite transporter of the plasma membrane
ASI1	Putative integral membrane E3 ubiquitin ligase
ATG7	Autophagy-related protein and dual-specificity member of the E1 family of ubiquitin-activating enzymes; mediates the conjugation of Atg12p with Atg5p and Atg8p with phosphatidylethanolamine, required steps in autophagosome formation
ATG9	Transmembrane protein involved in formation of CVT and autophagic vesicles; cycles between the preautophagosomal structure and other cytosolic punctate structures not found in autophagosomes
ATG18	Phosphatidylinositol 3,5-bisphosphate-binding protein of the vacuolar membrane; required for recycling of Atg9p through the preautophagosomal structure
AUR1	Phosphatidylinositol:ceramide phosphoinositol transferase (IPC synthase), required for sphingolipid synthesis
BET2	Beta subunit of type II geranylgeranyltransferase required for vesicular transport between the endoplasmic reticulum and the Golgi
BIR1	Essential chromosomal passenger protein
BNI5	Protein involved in organization of septins at the mother bud neck; may interact directly with the Cdc11p septin; localizes to the bud neck in a septin-dependent manner
BRO1b	Cytoplasmic class E VPS factor that coordinates deubiquitination in the MVB pathway by recruiting Doa4p to endosomes
BUD21b	UTP16, component of SSU processosome
BUL2	Component of the Rsp5p E3-ubiquitin ligase complex
CDC34b	Ubiquitin-conjugating enzyme (E2) and catalytic subunit of SCF ubiquitin-protein ligase complex
CDC55	Nonessential regulatory subunit B of protein phosphatase 2A; actin filament organization
CHA4	DNA-binding transcriptional activator
CCZ1	Protein involved in vacuolar assembly, essential for autophagy and the cytoplasm-to-vacuole pathway
CPR1b	Cytoplasmic peptidyl-prolyl cis-trans isomerase (cyclophilin); catalyzes the cis-trans isomerization of peptide bonds N terminal to proline residues
CRM1	Major karyopherin, involved in export of proteins, RNAs, and ribosomal subunits from the nucleus
DBP2b	Essential ATP-dependent RNA helicase of the DEAD box protein family, involved in nonsense-mediated mRNA decay and rRNA processing
DBP7	Putative ATP-dependent RNA helicase of the DEAD box family involved in ribosomal biogenesis
DDR48b	DNA damage-responsive protein
DEG1b	tRNA:pseudouridine synthase
DID2b	Class E protein of the VPS pathway; associates reversibly with the late endosome
DIE2	Dolichyl-phosphoglucose-dependent glucosyltransferase of the ER; has a role in regulation of ITR1 and INO1
DTD1	cm;1>d-Tyr-tRNATyr deacylase; functions in protein translation, may affect nonsense suppression via alteration of the protein synthesis machinery; ubiquitous among eukaryotes
EPS1	Pdi1p (protein disulfide isomerase)-related protein involved in ER retention of resident ER proteins
ERB1b	Constituent of 66S preribosomal particles, homologous to mammalian Bop1
ERG13	HMG-CoA synthase, ergosterol biosynthesis, mevalonate biosynthesis
ESS1b	PPIase
FEN1	Fatty acid elongase, involved in sphingolipid biosynthesis
FPR4	PPIase (proline isomerase) localized to the nucleus
FSH2	Serine hydrolase that localizes to the cytoplasm; sequence is similar to Fsh1p and Fsh3p
FUN26	Nucleoside transporter with broad nucleoside selectivity
GCD2b	Guanine nucleotide exchange factor for eIF2 translation initiation factor
GCN3	Alpha subunit of the translation initiation factor eIF2B, the guanine-nucleotide exchange factor for eIF2
GEA2	Guanine nucleotide exchange factor for ARFs, involved in vesicular transport between the Golgi and ER, Golgi organization, and actin cytoskeleton organization
GIS4	CAAX box containing protein of unknown function
GGA2	Golgi-localized protein with homology to gamma-adaptin, interacts with and regulates Arf1p and Arf2p in a GTP-dependent manner in order to facilitate traffic through the late Golgi
GLN3	Transcriptional activator
GPI8	ER membrane glycoprotein subunit of the glycosylphosphatidylinositol transamidase complex that adds GPI anchors to newly synthesized proteins
GPM2	Molecular function unknown
GPT2	Glycerol-3-phosphate acyltransferase located in both lipid particles and the ER, involved in lipid biosynthesis
HAA1b	Transcriptional activator involved in the transcription of genes encoding membrane stress proteins
HAS1b	ATP-dependent RNA helicase; localizes to both the nuclear periphery and nucleolus
HBS1b	GTPase with similarity to translation release factors
HOP2	Meiosis-specific protein that localizes to chromosomes
HUL4	Protein with similarity to hect domain E3 ubiquitin-protein ligases
IMP4	Component of the SSU processome, which is required for pre-18S rRNA processing; member of a superfamily of proteins that contain a sigma-70-like motif and associate with RNAs
INO2b	Transcription activator that binds inositol/choline-responsive elements, required for derepression of phospholipid biosynthetic genes
IZH4	Membrane protein involved in zinc and lipid metabolism
JJJ1b	Co-chaperone that stimulates the ATPase activity of Ssa1p
KEG1b	Integral membrane protein of the ER
MAK10	Noncatalytic subunit of N-terminal acetyltransferase of the NatC type, required for replication of dsRNA virus
MAP1	Methionine aminopeptidase; catalyzes the cotranslational removal of N-terminal methionine from nascent polypeptides
MDM38b	Mitochondrial protein; facilitates recruitment of mRNA-specific translational activators to ribosomes
MNR2	Putative magnesium transporter; has similarity to Alr1p and Alr2p, which mediate influx of Mg2+ and other divalent cations
MSP1b	Mitochondrial protein involved in sorting of proteins in the mitochondria; putative membrane-spanning ATPase
MST28	Putative integral membrane protein, involved in vesicle formation; forms complex with Mst27p; member of DUP240 gene family; binds COPI and COPII vesicles
NGR1	RNA-binding protein that negatively regulates growth rate; interacts with the 3′ untranslated region of the mitochondrial porin (POR1) mRNA and enhances its degradation; overexpression impairs mitochondrial function
NOG1b	Putative GTPase that associates with free 60S ribosomal subunits in the nucleolus
NOP53b	Nucleolar protein; involved in biogenesis of the 60S subunit of the ribosome
NPL3b	RNA-binding protein that promotes elongation, regulates termination, and carries poly(A) mRNA from nucleus to cytoplasm
NRM1	Transcriptional corepressor of MCB binding factor-regulated gene expression
NSP1	Essential component of the nuclear pore complex, which mediates nuclear import and export
NSR1b	Nucleolar protein that binds nuclear localization sequences, required for pre-rRNA processing and ribosome biogenesis; nucleolin
NUP1	NPC subunit, involved in protein import/export and in export of RNAs, possible karyopherin release factor that accelerates release of karyopherin-cargo complexes after transport across NPC
OLE1	Fatty acid desaturase, required for monounsaturated fatty acid synthesis
PBP2	RNA binding protein with similarity to mammalian heterogeneous nuclear RNP K protein
OTU2b	member of the ovarian tumor-like superfamily of predicted cysteine proteases
PDR17	Phosphatidylinositol transfer protein; downregulates Plb1p-mediated turnover of phosphatidylcholine, found in the cytosol and microsomes; pdr16 pdr17 double deletion mutants exhibit altered lipid levels
PEP7b	Multivalent adaptor protein that facilitates vesicle-mediated vacuolar protein sorting
PEP12	Target membrane receptor (t-SNARE) for vesicular intermediates traveling between the Golgi apparatus and the vacuole; controls biosynthetic, endocytic, and retrograde traffic into the prevacuolar compartment; syntaxin
PEX27	Peripheral peroxisomal membrane protein involved in controlling peroxisome size and no.
PEX29	Peroxisomal integral membrane peroxin, involved in the regulation of peroxisomal size, no., and distribution
PEX30	Peroxisomal integral membrane protein, involved in negative regulation of peroxisome no.
PIB1	RING-type ubiquitin ligase of the endosomal and vacuolar membranes, binds phosphatidylinositol(3)-phosphate; contains a FYVE finger domain
PGA2	Essential protein required for maturation of Gas1p and Pho8p; involved in protein trafficking
PIN2	Protein that induces appearance of PIN+ prion when overproduced
PKC1	Protein serine/threonine kinase essential for cell wall remodeling during growth
PLB2	Phospholipase B (lysophospholipase) involved in phospholipid metabolism
POL30b	Proliferating cell nuclear antigen, functions as the sliding clamp for DNA polymerase delta
POX1b	Fatty-acyl coenzyme A oxidase, involved in the fatty acid beta-oxidation pathway in the peroxisomes
PRE1	20S proteasome beta-type subunit; localizes to the nucleus throughout the cell cycle
PRO1	Gamma-glutamyl kinase, catalyzes the first step in proline biosynthesis
SPG4	Protein of unknown function
RAD61	Protein of unknown function
RDS1	Zinc cluster protein involved in conferring resistance to cycloheximide
RNY1b	Vacuolar RNase of the T(2) family, relocalizes to the cytosol where it cleaves tRNAs upon oxidative stress
ROT1	Protein that may be involved in cell wall function
RPC31	RNA polymerase III subunit C31; contains HMG-like C-terminal domain
RRP4	Protein involved in rRNA processing; component of the exosome 3→5 exonuclease complex
RRP14	Essential protein, constituent of 66S preribosomal particles
RSP5b	E3 ubiquitin ligase of the NEDD4 family
RVS161	Amphiphysin-like lipid raft protein; subunit of a complex (Rvs161p-Rvs167p) that regulates polarization of the actin cytoskeleton, endocytosis, cell polarity
SAR1	GTPase, GTP-binding protein of the ARF family, component of COPII coat of vesicles; required for transport vesicle formation during ER to Golgi protein transport
SCP1	Component of yeast cortical actin cytoskeleton, binds and cross-links actin filaments
SEC17	Peripheral membrane protein required for vesicular transport between ER and Golgi and for the priming step in homotypic vacuole fusion; part of the cis-SNARE complex; has similarity to alpha-SNAP
SEC22	R-SNARE protein; assembles into SNARE complex; cycles between the ER and Golgi complex; involved in anterograde and retrograde transport between the ER and Golgi; synaptobrevin homolog
SED5	cis-Golgi t-SNARE syntaxin required for vesicular transport between the ER and the Golgi complex; binds at least 9 SNARE proteins
SPC97	Component of the microtubule-nucleating Tub4p (gamma-tubulin) complex
SRP14	SRP subunit; interacts with the RNA component of SRP to form the Alu domain, which is the region of SRP responsible for arrest of nascent chain elongation during membrane targeting
SPT14	UDP-GlcNAc-binding and catalytic subunit of the enzyme that mediates the first step in GPI biosynthesis
STM1b	Protein required for translation under nutrient stress; binds quadruplex and purine motif triplex nucleic acids
SUB1b	Transcriptional coactivator; role in the hyperosmotic stress response
SVP26	Integral membrane protein of the early Golgi apparatus; may function to promote retention of proteins in the early Golgi compartment
TIF4631	Translation initiation factor eIF4G, subunit of the mRNA cap-binding protein complex (eIF4F))
TRM3	2′-O-Ribose methyltransferase, catalyzes the ribose methylation of the guanosine nucleotide at position 18 of tRNAs
TRZ1b	tRNA 3′-end processing endonuclease tRNase Z; homolog of the human cancer susceptibility gene ELAC2
TRS85	Subunit of TRAPPIII, a multimeric guanine nucleotide exchange factor for Ypt1p, required for membrane expansion during autophagy and the CVT pathway; directs Ypt1p to the PAS
TRS120	One of 10 subunits of the TRAPP complex of the cis-Golgi which mediates vesicle docking and fusion; involved in ER-to-Golgi membrane traffic
UBP8	Ubiquitin-specific protease that is a component of the SAGA acetylation complex; required for SAGA-mediated deubiquitination of histone H2B
UTP7b	Nucleolar protein, component of the SSU processome
VCX1	Vacuolar H+/Ca2+ exchanger involved in control of cytosolic Ca2+ concn
VPS3	Cytoplasmic protein required for the sorting and processing of soluble vacuolar proteins
VPS25	Component of the ESCRT-II complex, which is involved in ubiquitin-dependent sorting of proteins into the endosome
VPS27	Endosomal protein that forms a complex with Hse1p; required for recycling Golgi proteins
VPS36	Component of the ESCRT-II complex; contains the GLUE (GRAM-like ubiquitin binding in EAP45) domain
VRP1	proline-rich actin-associated protein involved in cytoskeletal organization and cytokinesis
YBR016W	Plasma membrane protein of unknown function; has similarity to hydrophilins
YCK1	Palmitoylated, plasma membrane-bound casein kinase I isoform; shares redundant functions with Yck2p in morphogenesis, proper septin assembly, endocytic trafficking
YCT1	High-affinity cysteine-specific transporter with similarity to the Dal5p family of transporters
YDJ1	Protein chaperone involved in regulation of the HSP90 and HSP70 functions; involved in protein translocation across membranes; member of the DnaJ family
YGL262W	Putative protein of unknown function; null mutant displays elevated sensitivity to expression of a mutant huntingtin fragment or of alpha-synuclein
YGR026Wb	Putative protein of unknown function; GFP fusion protein localizes to the cell periphery
YGR130C	Putative protein of unknown function
YLR126C	Putative protein of unknown function; may be involved in copper and iron homeostasis
YMR1	PI(3)P phosphatase; regulates the localization and levels of PI(3)P; involved in CVT transport
YMR259C	Putative protein of unknown function; GFP fusion protein localizes to the cytoplasm
YOS1	Integral membrane protein required for ER to Golgi transport; localized to the Golgi, the ER, and COPII vesicles
YPT1	Rab family GTPase, involved in the ER-to-Golgi step of the secretory pathway
YPT52	GTPase, similar to Ypt51p and Ypt53p and to mammalian Rab5; required for vacuolar protein sorting and endocytosis
YSY6	Protein whose expression suppresses a secretory pathway mutation in E. coli; has similarity to the mammalian RAMP4 protein involved in secretion

^aGene names are based on the Saccharomyces genome database. Downregulators are underlined; upregulators are in bold.

^bPreviously identified.

^cVPS, vacuolar protein-sorting; MVB, multivesicular body; SSU, small ribosomal subunit; SCF complex, Skp1, Cullins, F-box proteins; HMG-CoA, 3-hydroxy-3-methylglutaryl coenzyme A; PPIase, peptidylprolyl cis-trans-isomerase; ARF, ADP ribosylation factor; GPI, glycosylphosphatidylinositol; dsRNA, double-stranded RNA; NPC, nuclear pore complex; SNARE, soluble N-ethylmaleimide-sensitive factor (NSF) attachment protein receptors; SRP, signal recognition particle; TRAPP, transport protein particle; CVT, cytoplasm to vacuole; SAGA, Spt-Ada-Gcn5-acetyltransferase; GFP, green fluorescent protein; PI(3)P, phosphatidylinositol 3-phosphate.

A large number of vesicular transport proteins affect tombusvirus replication.

The 141 host proteins identified in the above screen code for proteins with different molecular functions in various cellular processes (Saccharomyces Genome Database [http://www.yeastgenome.org]). Bioinformatic analysis of the identified host factors in the current proteome-wide screen revealed that, surprisingly, host proteins involved in protein targeting and vesicle-mediated transport are by far the most numerous group of factors (39 host proteins) (Fig. 1). It is currently not known how these proteins could affect TBSV replication, which occurs on the cytosolic surface of peroxisomes (31, 43, 46). It is possible that some of the identified host proteins involved in protein targeting and vesicle-mediated transport might directly affect the peroxisome-to-endoplasmic reticulum (ER) pathway, which has been suggested to be involved in sorting TBSV replication proteins (31). Additional large groups of factors include (i) protein modifying enzymes/factors, (ii) lipid metabolism and membrane biogenesis factors, (iii) RNA-modifying and RNA metabolism factors, (iv) translation factors and ribosomal proteins, and (v) stress-related proteins (Fig. 1). Altogether, the identified host factors could have either direct or indirect effects on tombusvirus replication.

Fig 1.

Grouping of the identified yeast proteins affecting TBSV replication based on their known cellular functions. Systematic screening of the yeast overexpression library resulted in the identification of 141 unique yeast genes that either promoted or inhibited TBSV replication (Table 1). The identified host genes were grouped into 11 categories based on their known cellular and biochemical functions. The number of genes in each category is shown in parentheses. Proteins with multiple functions were placed arbitrarily only in one of the categories, mainly based on their predicted function in TBSV replication.

A temperature-sensitive kinase mutant of Pkc1p supports increased TBSV replication in yeast.

To validate the results from the proteome-wide screen, we chose the highly conserved Pkc1p, which is an essential gene for yeast growth. We studied Pkc1p in detail here, since our previous work showed that recombinant Pkc1p could phosphorylate the replication proteins p33 and p92 in vitro (74). The phosphorylation sites were mapped to serines and a threonine located next to the RNA-binding site of these replication proteins (Fig. 2A and B). Interestingly, phosphorylation of p33 and/or p92 interfered with the abilities of these proteins to bind to the viral RNA in vitro (74). Thus, Pkc1p might be involved in TBSV replication as a regulator of the RNA-binding function of p33. Using phosphorylation-mimicking mutants of p33, we previously showed that reversible phosphorylation affects TBSV replication in plants (71). Nevertheless, critical information regarding the in vivo role of Pkc1p in TBSV replication was lacking. Accordingly, identification of Pkc1p in the proteome-wide overexpression screen (Table 1) provides the first piece of evidence that Pkc1p is a host factor affecting TBSV replication.

Fig 2.

Overexpression of Pkc1p inhibits TBSV repRNA accumulation in yeast. (A) Schematic representation of the known domains in the tombusvirus replication protein p33. TMD, transmembrane domain; P, phosphorylation site; RPR, arginine-proline-rich RNA-binding domain. S1 and S2 are subdomains of the p33:p33-p92 interaction domain. The PSVP late domain, in connection with ubiquitinated lysines, is involved in binding to the host ESCRT components. (B) Binding of the unphosphorylated (left) and phosphorylated (right) forms of p33 to the viral RNA (71, 74). Phosphorylation is performed by Pkc in vitro, and the phosphorylated serine and threonine residues in p33 located in the vicinity of the RPR motif are shown in bold and labeled with the letter P. Note that the positively charged arginines within the RPR motif, critical in binding to the viral RNA, are predicted to be neutralized by the phosphorylated serine and threonine, as shown, resulting in a lack of RNA binding by p33. (C) Reduced TBSV repRNA accumulation in yeast overexpressing yeast Pkc1p. Overexpression was done from the GAL1 promoter. repRNA replication took place for 24 h at 29°C before RNA analysis. The accumulation level of DI-72 (+)repRNA (shown as percentages) was normalized based on that of 18S rRNA. Each experiment was repeated three times. (D) To launch TBSV repRNA replication, we expressed His₆-p33 and His₆-p92 from the copper-inducible CUP1 promoter and DI-72 (+)repRNA from the constitutive ADH1 promoter in the parental (wt, BY4741) and pkc1^ts yeast strains. The yeast cells were cultured for 36 h at either 23°C (permissive temperature) or 32°C (semipermissive temperature) on 2% glucose SC minimal medium. Northern blot analysis was used to detect DI-72 (+)repRNA accumulation. The accumulation level of DI-72 (+)repRNA was normalized based on 18S rRNA. Each experiment was repeated three times. (E) Overexpression of yeast Pkc1p in pkc1^ts yeast strains reduced TBSV repRNA accumulation. The yeast cells were cultured for 36 h at 32°C (semipermissive temperature).

To confirm the screen results, first we overexpressed the full-length Pkc1p in the wt yeast strain BY4741. As expected, Pkc1p overexpression reduced TBSV repRNA accumulation ∼4-fold (Fig. 2C). In addition, we used pkc1-4 yeast strain (29), which carries a temperature-sensitive (ts) kinase mutant of Pkc1p (Fig. 2D). We found that the ts mutant of Pkc1p allowed increased TBSV replication in yeast, ∼4-fold at the permissive temperature (23°C) and ∼5-fold at the semipermissive temperature (32°C) (Fig. 2D). Also, complementation with the wt Pkc1p in the pkc1-4 yeast strain led to ∼3-fold less TBSV replication than in the control pkc1-4 yeast strain not expressing wt Pkc1p (Fig. 2E), suggesting that pkc1-4 mutation can be complemented by the plasmid-borne expression of wt Pkc1p. Overall, these results support an inhibitory role for Pkc1p in TBSV replication in yeast.

Treatment with cercosporamide increases TBSV replication in yeast.

To further test if Pkc1p is involved in TBSV replication, we used cercosporamide, which specifically inhibits Pkc-type kinases (21, 77). Treatment of wt yeast with cercosporamide increased TBSV accumulation ∼3-fold (Fig. 3A). This result confirmed that Pkc1p is an inhibitor of TBSV replication in yeast cells.

Fig 3.

Effect of a Pkc1 inhibitor (cercosporamide) on viral RNA accumulation in yeast. (A) (Top) Northern blot analysis was used to detect DI-72 (+)repRNA accumulation in a yeast strain treated with cercosporamide (0.5 μg/ml) to inhibit Pkc1p function. rec, recombinant RNA; deg, partially degraded. (Bottom) Ethidium-bromide stained gel of total RNA extracts of the samples used for Northern blotting above. (B to D) Northern blot analysis of TBSV repRNA replication in yeast expressing the wt p33 or the nonphosphorylatable p33-A₂₀₅A₂₁₀A₂₁₁ (B) or partially phosphorylatable p33-A₂₁₀A₂₁₁ (C) and in p33-D₂₀₅ mutants from the ADH1 promoter (D), while wt p92 and DI-72 repRNA were expressed from the CUP1 promoter and the GAL1 promoter, respectively. Yeast was cultured for 36 h at 23°C in the presence of cercosporamide (1.0 μg/ml), 2% galactose, and 50 μM CuSO₄.

To test if Pkc1p inhibits TBSV replication via phosphorylation of the replication protein p33, we used phosphorylation-deficient mutants of p33, which can be only partially phosphorylatable (A₂₁₀A₂₁₁ and D₂₀₅) or nonphosphorylatable (A₂₀₅A₂₁₀A₂₁₁) by Pkc1p in vitro (71, 74), in untreated yeast or yeast treated with cercosporamide. Interestingly, inhibition of Pkc1p by cercosporamide did not increase the replication of the nonphosphorylatable mutant p33-A₂₀₅A₂₁₀A₂₁₁ (Fig. 3B, lanes 1 and 2 versus lanes 3 and 4, and 3D) or p33-D₂₀₅ (Fig. 3D, lanes 1 and 2 versus lanes 3 and 4), while replication moderately increased (by ∼30%) when a partially phosphorylatable mutant, p33-A₂₁₀A₂₁₁, was used to support TBSV repRNA replication (Fig. 3C, lanes 1 and 2 versus lanes 3 and 4). Altogether, these data support the model that Pkc1p plays a role in TBSV replication via phosphorylation of the viral replication protein p33.

Treatment with cercosporamide also increases TBSV replication in plant protoplasts and whole Nicotiana benthamiana plants.

To examine if there is a similar kinase-based regulation of TBSV replication in plant host cells, we treated Nicotiana benthamiana protoplasts (single cells with the cell wall removed) replicating TBSV genomic RNA with cercosporamide at two different concentrations. The higher-concentration cercosporamide treatment increased TBSV genomic RNA accumulation ∼2.5-fold (Fig. 4A). Similar treatment of plant protoplasts with cercosporamide also increased the accumulation of the genomic RNA of TCV, a closely related plant virus to TBSV. Indeed, TCV accumulation increased 2.4-fold compared with that in the ethanol-treated control (Fig. 4B). Thus, it seems that replication of TBSV and TCV is affected by cercosporamide, an inhibitor of Pkc1-like kinases.

Fig 4.

The effect of cercosporamide treatment on TBSV and TCV RNA accumulation in N. benthamiana protoplasts. (A) Northern blot analysis was used to detect genomic (g) TBSV RNA accumulation in protoplasts treated with cercosporamide to inhibit Pkc1-like functions. Protoplasts from N. benthamiana were electroporated with TBSV gRNA and treated with various concentrations of cercosporamide (2- and 4-μg/ml final concentrations). Total RNA samples were obtained 40 h postelectroporation. The ethidium-bromide stained gel at the bottom shows rRNA levels. Note that treatment with ethanol (shown as “0”), which is used to dissolve cercosporamide, was chosen as the control. The accumulation level of TBSV RNA was normalized based on the rRNA. Each experiment was repeated three times. (B) As panel A, except that TCV RNA was used for protoplast electroporation.

Treatment of N. benthamiana leaves with cercosporamide (at 2 dpi, on newly emerging top leaves) intensified the symptoms caused by TBSV (Fig. 5A) and led to a ∼2.5-fold increase in TBSV RNA accumulation at 4 dpi (Fig. 5B). Thus, it seems that cercosporamide can inhibit the antiviral function of Pkc1-like kinases in whole plants as well.

Fig 5.

The effect of cercosporamide treatment on TBSV RNA accumulation in N. benthamiana leaves. (A) Symptom intensification caused by TBSV infection in plants treated with cercosporamide. Note that leaf curling (the young leaves are circled) is more pronounced in TBSV-infected plants 4 days postinoculation. (B) Northern blot analysis was used to detect gTBSV RNA. Treatment of N. benthamiana leaves with cercosporamide promotes the accumulation of TBSV RNAs. Total RNA samples from the inoculated leaves were obtained 4 days postinoculation and used for Northern blotting (top) and gel analysis (bottom) to show rRNA levels.

Recombinant Pkc1p inhibits TBSV replication in a cell-free replication assay.

Although we previously documented the inhibitory effect of Pkc on the in vitro activity of TBSV RdRp and RNA binding by the replication protein p33 (71, 74), we wanted to obtain further direct evidence that Pkc1p can affect TBSV replication using our recently developed authentic yeast cell-free tombusvirus replication assay (53, 55). In the in vitro TBSV replication assay, we programmed the yeast cell extract (CFE) with the in vitro-transcribed TBSV (+)repRNA and purified recombinant p33 and p92^pol obtained from E. coli (schematically shown in Fig. 6A). This led to the in vitro assembly of the viral replicase on the membranes present in CFE and one single cycle of complete TBSV replication, resulting in both negative-stranded and positive-stranded repRNA progeny (53, 55).

Fig 6.

Inhibition of TBSV replication by recombinant Pkc1p in vitro. (A) Scheme of the CFE-based TBSV replication assay. Purified recombinant p33 and p92^pol replication proteins of TBSV and in vitro-transcribed TBSV DI-72 (+)repRNA were added to the whole-cell extract prepared from the wt yeast strain. The purified recombinant yeast Pkc1p was added before (exp. C) or during (exp. B) the CFE-based TBSV replication assay. (B) Denaturing PAGE analysis of the ³²P-labeled TBSV repRNA products obtained in the in vitro CFE-based TBSV replication assay in the presence of recombinant Pkc1p (1× = 200 ng). Each experiment was repeated three times. (C) CFE-based assay similar to that in panel B, except that Pkc1p was preincubated with p33/p92 in the reaction buffer for 30 min at 25°C. (D) Western blot analysis of purified recombinant GST-Pkc1p with anti-GST antibody.

We expressed and affinity purified recombinant yeast Pkc1p from E. coli (Fig. 6D) and added it to CFE prepared from BY4741 yeast (Fig. 6A). We found that the recombinant Pkc1p inhibited TBSV replication in the CFE-based assay ∼3-fold (Fig. 6B, lane 8 versus 9). Interestingly, preincubation of the recombinant Pkc1p with replication proteins p33 and/or p92 prior to the CFE-based assay led to an almost complete block of TBSV replication in vitro (Fig. 6C, lane 7). This suggests that the recombinant yeast Pkc1p is a potent inhibitor of TBSV replication, likely due to phosphorylation of the replication proteins that inhibits their viral RNA binding abilities, as shown previously for purified Pkc (Fig. 2B) (74).

DISCUSSION

High-throughput overexpression of host proteins is a suitable approach to identify host factors affecting tombusvirus replication.

Most genome-wide screens to identify host factors affecting (+)RNA virus infections are based on RNA interference (RNAi) approaches or the yeast deletion library (7, 19, 23, 24, 39, 44, 48, 62, 69, 70, 76, 78, 79). Additional systematic approaches, such as proteome-wide screens, can also lead to identification of novel host factors, as shown for TBSV by using mass spectrometry, protein arrays, and cDNA library screens (25, 27, 32, 67). In this study, we explored the large yeast ORF expression library (11) that covers over 90% of all known yeast genes for proteome-wide studies of viral host factors. Since we found 36 yeast proteins in the current screen, which were previously identified for TBSV, we suggest that similar proteome-wide approaches could also be useful for other virus-host interactions. Identification of over a hundred novel host proteins affecting TBSV replication in this screen, however, indicates that we still have not reached saturation level in identification of host factors with all the previous screens (38, 40). Bioinformatic analysis (data not shown) suggests that many different host pathways could affect TBSV replication. The largest number of host genes identified affects protein targeting and vesicle-mediated transport (Fig. 1). However, it is unlikely that all the identified host proteins involved in protein targeting and vesicle-mediated transport would directly affect the peroxisome-to-ER pathway involved in sorting TBSV replication proteins (31). Instead, it is possible that overexpression of given proteins would affect vesicle or protein transport in yeast, which might compete with TBSV replication for the transport of viral proteins, host protein factors, or host lipids to the site of tombusviral replication, thus indirectly inhibiting the viral replication process. Indeed, many lipid metabolism and membrane biogenesis genes also affected TBSV replication (Fig. 1), further supporting the idea of competition between the host pathways and TBSV for common host resources, like lipids and host proteins. However, an increasing number of host proteins is known to have direct roles or functions in TBSV replication, as shown for Pkc1p in this work and many others investigated earlier, such as the heat shock protein 70 chaperones (Hsp70) (55, 67, 85, 86), glyceraldehyde-3-phosphate dehydrogenase (13, 84), Cdc34p E2 ubiquitin-conjugating enzyme (25, 67, 84), eukaryotic translation elongation factor 1A (eEF1A) (27, 28), eukaryotic elongation factor 1Bγ (eEF1Bγ) (65), Ded1p RNA helicase (22), Pex19p shuttle protein (52), the Nedd40-like Rsp5p E3 ubiquitin ligase (4, 58), nucleolin (18), cyclophilins (30, 32), and a set of ESCRT proteins (3, 5, 27). Thus, additional in-depth analysis of the identified host factors will be needed to determine whether a given host protein acts directly or indirectly and what this particular host protein's mechanistic role is during viral replication. Altogether, we have already found via multiple screens that more than 400 host genes and proteins affect TBSV replication, indicating the complex nature of RNA virus-host interaction, even in a simple eukaryotic host such as yeast.

Direct inhibition of TBSV replication by Pkc1p.

Previous in vitro work has demonstrated that recombinant Pkc could phosphorylate the serine/threonine residues close to the RPR (arginine-proline-rich) motif in p33 and p92 (Fig. 2A), which inhibits the abilities of p33 and p92 to bind viral RNA and in viral RNA template recruitment to the site of replication (56, 59, 71, 74). This study, which used a CFE-based TBSV replication assay, demonstrates that preincubation of the recombinant yeast Pkc1p with the replication proteins p33 and p92 strongly inhibited TBSV replication in vitro (Fig. 6). Thus, Pkc1p seems to be a negative regulator of TBSV replication, likely due to inhibition of the RNA-binding activity of p33 and p92. Thus, this work in combination with previous in vitro data (56, 59, 71, 74) firmly establishes the inhibitory role of Pkc in the functions of the replication proteins p33 and p92.

Importantly, the current work also provides strong evidence that Pkc1p plays a role in TBSV replication in yeast based on overexpression studies and a ts mutant of Pkc1p. Also, the use of a specific inhibitor of Pkc1p (i.e., cercosporamide) further supports the idea that Pkc1p is a potent inhibitor of TBSV replication. Based on the use of the nonphosphorylatable mutant of p33, it seems that Pkc1p inhibits TBSV replication via phosphorylation of the viral p33 replication protein. Similarly, cercosporamide treatment also increased TBSV replication in plant protoplasts and whole plants, suggesting that Pkc-related host kinases inhibit TBSV in a natural host, too.

Various roles for Pkc in relation to several RNA viruses are emerging. Pkc is known to be a major component of the antiviral response in mammalian cells, as it acts in the signal transduction pathway mediated by human alpha interferon (63). RNA viruses also exploit Pkc during their replication. For example, human parainfluenza virus and Sendai virus use the cellular Pkc to phosphorylate the viral P protein, which is critical for its function as transactivator of the viral RNA polymerase (8, 14, 15). Also, inhibitors of the cellular PKCs are potential agents against human immunodeficiency virus, as they inhibit transcription of viral RNAs (57). In addition, a recent kinome-wide RNAi screen in Drosophila also identified several kinases and phosphatases, including a protein kinase C involved in the poxvirus entry process (35).

Overall, the current work has revealed a role for Pkc1p as an inhibitor of RNA virus replication in vivo. This function for Pkc1p seems to be conserved between yeast and plants, as demonstrated by treatment with the Pkc-specific inhibitor cercosporamide, which resulted in increased TBSV RNA accumulation in yeast, single plant cells, and whole N. benthamiana plants.