The Effect of Calcium Buffering and Calcium Sensor Type on the Sensitivity of an Array-Based Bitter Receptor Screening Assay.

The genetically encoded calcium sensor protein Cameleon YC3.6 has previously been applied for functional G protein-coupled receptor screening using receptor cell arrays. However, different types of sensors are available, with a wide range in [Ca2+] sensitivity, Hill coefficients, calcium binding domains, and fluorophores, which could potentially improve the performance of the assay. Here, we compared the responses of 3 structurally different calcium sensor proteins (Cameleon YC3.6, Nano140, and Twitch2B) simultaneously, on a single chip, at different cytosolic expression levels and in combination with 2 different bitter receptors, TAS2R8 and TAS2R14. Sensor concentrations were modified by varying the amount of calcium sensor DNA that was printed on the DNA arrays prior to reverse transfection. We found that ~2-fold lower concentrations of calcium sensor protein, by transfecting 4 times less sensor-coding DNA, resulted in more sensitive bitter responses. The best results were obtained with Twitch2B, where, relative to YC3.6 at the default DNA concentration, a 4-fold lower DNA concentration increased sensitivity 60-fold and signal strength 5- to 10-fold. Next, we compared the performance of YC3.6 and Twitch2B against an array with 11 different bitter taste receptors. We observed a 2- to 8-fold increase in sensitivity using Twitch2B compared with YC3.6. The bitter receptor arrays contained 300 spots and could be exposed to a series of 18 injections within 1 h resulting in 5400 measurements. These optimized sensor conditions provide a basis for enhancing receptomics calcium assays for receptors with poor Ca2+ signaling and will benefit future high-throughput receptomics experiments.

Identification of a drimenol synthase and drimenol oxidase from Persicaria hydropiper, involved in the biosynthesis of insect deterrent drimanes.

The sesquiterpenoid polygodial, which belongs to the drimane family, has been shown to be an antifeedant for a number of herbivorous insects. It is presumed to be synthesized from farnesyl diphosphate via drimenol, subsequent C-12 hydroxylation and further oxidations at both C-11 and C-12 to form a dialdehyde. Here, we have identified a drimenol synthase (PhDS) and a cytochrome P450 drimenol oxidase (PhDOX1) from Persicaria hydropiper. Expression of PhDS in yeast and plants resulted in production of drimenol alone. Co-expression of PhDS with PhDOX1 in yeast yielded drimendiol, the 12-hydroxylation product of drimenol, as a major product, and cinnamolide. When PhDS and PhDOX1 were transiently expressed by agro-infiltration in Nicotiana benthamiana leaves, drimenol was almost completely converted into cinnamolide and several additional drimenol derivatives were observed. In vitro assays showed that PhDOX1 only catalyses the conversion from drimenol to drimendiol, and not the further oxidation into an aldehyde. In yeast and heterologous plant hosts, the C-12 position of drimendiol is therefore likely to be further oxidized by endogenous enzymes into an aldehyde and subsequently converted to cinnamolide, presumably by spontaneous hemiacetal formation with the C-11 hydroxyl group followed by oxidation. Purified cinnamolide was confirmed by NMR and shown to be deterrent with an effective deterrent dose (ED50 ) of about 200-400 µg g-1 fresh weight against both whiteflies and aphids. The putative additional physiological and biochemical requirements for polygodial biosynthesis and stable storage in plant tissues are discussed.

Calcium Imaging of GPCR Activation Using Arrays of Reverse Transfected HEK293 Cells in a Microfluidic System.

Reverse-transfected cell arrays in microfluidic systems have great potential to perform large-scale parallel screening of G protein-coupled receptor (GPCR) activation. Here, we report the preparation of a novel platform using reverse transfection of HEK293 cells, imaging by stereo-fluorescence microscopy in a flowcell format, real-time monitoring of cytosolic calcium ion fluctuations using the fluorescent protein Cameleon and analysis of GPCR responses to sequential sample exposures. To determine the relationship between DNA concentration and gene expression, we analyzed cell arrays made with variable concentrations of plasmid DNA encoding fluorescent proteins and the Neurokinin 1 (NK1) receptor. We observed pronounced effects on gene expression of both the specific and total DNA concentration. Reverse transfected spots with NK1 plasmid DNA at 1% of total DNA still resulted in detectable NK1 activation when exposed to its ligand. By varying the GPCR DNA concentration in reverse transfection, the sensitivity and robustness of the receptor response for sequential sample exposures was optimized. An injection series is shown for an array containing the NK1 receptor, bitter receptor TAS2R8 and controls. Both receptors were exposed 14 times to alternating samples of two ligands. Specific responses remained reproducible. This platform introduces new opportunities for high throughput screening of GPCR libraries.

High-yield production of a human monoclonal IgG by rhizosecretion in hydroponic tobacco cultures.

Rhizosecretion of recombinant pharmaceuticals from in vitro hydroponic transgenic plant cultures is a simple, low cost, reproducible and controllable production method. Here, we demonstrate the application and adaptation of this manufacturing platform to a human antivitronectin IgG1 monoclonal antibody (mAb) called M12. The rationale for specific growth medium additives was established by phenotypic analysis of root structure and by LC-ESI-MS/MS profiling of the total protein content profile of the hydroponic medium. Through a combination of optimization approaches, mAb yields in hydroponic medium reached 46 µg/mL in 1 week, the highest figure reported for a recombinant mAb in a plant secretion-based system to date. The rhizosecretome was determined to contain 104 proteins, with the mAb heavy and light chains the most abundant. This enabled evaluation of a simple, scalable extraction and purification protocol and demonstration that only minimal processing was necessary prior to protein A affinity chromatography. MALDI-TOF MS revealed that purified mAb contained predominantly complex-type plant N-glycans, in three major glycoforms. The binding of M12 purified from hydroponic medium to vitronectin was comparable to its Chinese hamster ovary (CHO)-derived counterpart. This study demonstrates that in vitro hydroponic cultivation coupled with recombinant protein rhizosecretion can be a practical, low-cost production platform for monoclonal antibodies.

Scaled-up manufacturing of recombinant antibodies produced by plant cells in a 200-L orbitally-shaken disposable bioreactor.

Tobacco BY-2 cells have emerged as a promising platform for the manufacture of biopharmaceutical proteins, offering efficient protein secretion, favourable growth characteristics and cultivation in containment under a controlled environment. The cultivation of BY-2 cells in disposable bioreactors is a useful alternative to conventional stainless steel stirred-tank reactors, and orbitally-shaken bioreactors could provide further advantages such as simple bag geometry, scalability and predictable process settings. We carried out a scale-up study, using a 200-L orbitally-shaken bioreactor holding disposable bags, and BY-2 cells producing the human monoclonal antibody M12. We found that cell growth and recombinant protein accumulation were comparable to standard shake flask cultivation, despite a 200-fold difference in cultivation volume. Final cell fresh weights of 300-387 g/L and M12 yields of ∼20 mg/L were achieved with both cultivation methods. Furthermore, we established an efficient downstream process for the recovery of M12 from the culture broth. The viscous spent medium prevented clarification using filtration devices, but we used expanded bed adsorption (EBA) chromatography with SP Sepharose as an alternative for the efficient capture of the M12 antibody. EBA was introduced as an initial purification step prior to protein A affinity chromatography, resulting in an overall M12 recovery of 75-85% and a purity of >95%. Our results demonstrate the suitability of orbitally-shaken bioreactors for the scaled-up cultivation of plant cell suspension cultures and provide a strategy for the efficient purification of antibodies from the BY-2 culture medium.

Monomeric IgA can be produced in planta as efficient as IgG, yet receives different N-glycans.

The unique features of IgA, such as the ability to recruit neutrophils and suppress the inflammatory responses mediated by IgG and IgE, make it a promising antibody isotype for several therapeutic applications. However, in contrast to IgG, reports on plant production of IgA are scarce. We produced IgA1κ and IgG1κ versions of three therapeutic antibodies directed against pro-inflammatory cytokines in Nicotiana benthamiana: Infliximab and Adalimumab, directed against TNF-α, and Ustekinumab, directed against the interleukin-12p40 subunit. We evaluated antibody yield, quality and N-glycosylation. All six antibodies had comparable levels of expression between 3.5 and 9% of total soluble protein content and were shown to have neutralizing activity in a cell-based assay. However, IgA1κ-based Adalimumab and Ustekinumab were poorly secreted compared to their IgG counterparts. Infliximab was poorly secreted regardless of isotype backbone. This corresponded with the observation that both IgA1κ- and IgG1κ-based Infliximab were enriched in oligomannose-type N-glycan structures. For IgG1κ-based Ustekinumab and Adalimumab, the major N-glycan type was the typical plant complex N-glycan, biantennary with terminal N-acetylglucosamine, ß1,2-xylose and core α1,3-fucose. In contrast, the major N-glycan on the IgA-based antibodies was xylosylated, but lacked core α1,3-fucose and one terminal N-acetylglucosamine. This type of N-glycan occurs usually in marginal percentages in plants and was never shown to be the main fraction of a plant-produced recombinant protein. Our data demonstrate that the antibody isotype may have a profound influence on the type of N-glycan an antibody receives.

Expression of natural human ß1,4-GalT1 variants and of non-mammalian homologues in plants leads to differences in galactosylation of N-glycans.

ß1,4-Galactosylation of plant N-glycans is a prerequisite for commercial production of certain biopharmaceuticals in plants. Two different types of galactosylated N-glycans have initially been reported in plants as the result of expression of human ß1,4-galactosyltransferase 1 (GalT). Here we show that these differences are associated with differences at its N-terminus: the natural short variant of human GalT results in hybrid type N-glycans, whereas the long form generates bi-antennary complex type N-glycans. Furthermore, expression of non-mammalian, chicken and zebrafish GalT homologues with N-termini resembling the short human GalT N-terminus also induce hybrid type N-glycans. Providing both non-mammalian GalTs with a 13 amino acid N-terminal extension that distinguishes the two naturally occurring forms of human GalT, acted to increase the levels of bi-antennary galactosylated N-glycans when expressed in tobacco leaves. Replacement of the cytosolic tail and transmembrane domain of chicken and zebrafish GalTs with the corresponding region of rat α2,6-sialyltransferase yielded a gene whose expression enhanced the level of bi-antennary galactosylation even further.

Molecular farming in tobacco hairy roots by triggering the secretion of a pharmaceutical antibody.

Recombinant pharmaceutical proteins expressed in hairy root cultures can be secreted into the medium to improve product homogeneity and to facilitate purification, although this may result in significant degradation if the protein is inherently unstable or particularly susceptible to proteases. To address these challenges, we used a design of experiments approach to develop an optimized induction protocol for the cultivation of tobacco hairy roots secreting the full-size monoclonal antibody M12. The antibody yield was enhanced 30-fold by the addition of 14 g/L KNO3 , 19 mg/L 1-naphthaleneacetic acid and 1.5 g/L of the stabilizing agent polyvinylpyrrolidone. Analysis of hairy root cross sections revealed that the optimized medium induced lateral root formation and morphological changes in the inner cortex and pericycle cells, indicating that the improved productivity was at least partially based on the enhanced efficiency of antibody secretion. We found that 57% of the antibody was secreted, yielding 5.9 mg of product per liter of induction medium. Both the secreted and intracellular forms of the antibody could be isolated by protein A affinity chromatography and their functionality was confirmed using vitronectin-binding assays. Glycan analysis revealed three major plant complex-type glycans on both forms of the antibody, although the secreted form was more homogeneous due to the predominance of a specific glycoform. Tobacco hairy root cultures therefore offer a practical solution for the production of homogeneous pharmaceutical antibodies in containment.

Identification of alg3 in the mushroom-forming fungus Schizophyllum commune and analysis of the Δalg3 knockout mutant.

Alg3 of Saccharomyces cerevisiae catalyzes the mannosyl transfer from Man-P-Dol to Man(5)GlcNAc(2)-PP-Dol resulting in the formation of Man(6)GlcNAc(2)-PP-Dol, which is then further processed to the final precursor oligosaccharide Glc(3)Man(9)GlcNAc(2) for N-glycosylation of proteins. Here, we identified the alg3 gene of the mushroom-forming fungus Schizophyllum commune by homology search. Its function was confirmed by the complementation of the Δalg3 strain of S. cerevisiae. Inactivation of alg3 in S. commune resulted in the production of predominantly Man(3)GlcNAc(2) protein-linked N-glycans. No impact on growth nor a developmental phenotype due to the deletion was observed. This provides a first step toward engineering of a homogeneous, humanized N-glycosylation pattern for the production of therapeutic glycoproteins in mushrooms.

Characterization of the single-chain Fv-Fc antibody MBP10 produced in Arabidopsis alg3 mutant seeds.

ER resident glycoproteins, including ectopically expressed recombinant glycoproteins, carry so-called high-mannose type N-glycans, which can be at different stages of processing. The presence of heterogeneous high-mannose type glycans on ER-retained therapeutic proteins is undesirable for specific therapeutic applications. Previously, we described an Arabidopsis alg3-2 glycosylation mutant in which aberrant Man(5)GlcNAc(2) mannose type N-glycans are transferred to proteins. Here we show that the alg3-2 mutation reduces the N-glycan heterogeneity on ER resident glycoproteins in seeds. We compared the properties of a scFv-Fc, with a KDEL ER retention tag (MBP10) that was expressed in seeds of wild type and alg3-2 plants. N-glycans on these antibodies from mutant seeds were predominantly of the intermediate Man(5)GlcNAc(2) compared to Man(8)GlcNAc(2) and Man(7)GlcNAc(2) isoforms on MBP10 from wild-type seeds. The presence of aberrant N-glycans on MBP10 did not seem to affect MBP10 dimerisation nor binding of MBP10 to its antigen. In alg3-2 the fraction of underglycosylated MBP10 protein forms was higher than in wild type. Interestingly, the expression of MBP10 resulted also in underglycosylation of other, endogenous glycoproteins.

LEW3, encoding a putative alpha-1,2-mannosyltransferase (ALG11) in N-linked glycoprotein, plays vital roles in cell-wall biosynthesis and the abiotic stress response in Arabidopsis thaliana.

N-linked glycosylation is an essential protein modification that helps protein folding, trafficking and translocation in eukaryotic systems. The initial process for N-linked glycosylation shares a common pathway with assembly of a dolichol-linked core oligosaccharide. Here we characterize a new Arabidopsis thaliana mutant lew3 (leaf wilting 3), which has a defect in an alpha-1,2-mannosyltransferase, a homolog of ALG11 in yeast, that transfers mannose to the dolichol-linked core oligosaccharide in the last two steps on the cytosolic face of the ER in N-glycan precursor synthesis. LEW3 is localized to the ER membrane and expressed throughout the plant. Mutation of LEW3 caused low-level accumulation of Man(3)GlcNAc(2) and Man(4)GlcNAc(2) glycans, structures that are seldom detected in wild-type plants. In addition, the lew3 mutant has low levels of normal high-mannose-type glycans, but increased levels of complex-type glycans. The lew3 mutant showed abnormal developmental phenotypes, reduced fertility, impaired cellulose synthesis, abnormal primary cell walls, and xylem collapse due to disturbance of the secondary cell walls. lew3 mutants were more sensitive to osmotic stress and abscisic acid (ABA) treatment. Protein N-glycosylation was reduced and the unfolded protein response was more activated by osmotic stress and ABA treatment in the lew3 mutant than in the wild-type. These results demonstrate that protein N-glycosylation plays crucial roles in plant development and the response to abiotic stresses.

Differential effects of human and plant N-acetylglucosaminyltransferase I (GnTI) in plants.

In plants and animals, the first step in complex type N-glycan formation on glycoproteins is catalyzed by N-acetylglucosaminyltransferase I (GnTI). We show that the cgl1-1 mutant of Arabidopsis, which lacks GnTI activity, is fully complemented by YFP-labeled plant AtGnTI, but only partially complemented by YFP-labeled human HuGnTI and that this is due to post-transcriptional events. In contrast to AtGnTI-YFP, only low levels of HuGnTI-YFP protein was detected in transgenic plants. In protoplast co-transfection experiments all GnTI-YFP fusion proteins co-localized with a Golgi marker protein, but only limited co-localization of AtGnTI and HuGnTI in the same plant protoplast. The partial alternative targeting of HuGnTI in plant protoplasts was alleviated by exchanging the membrane-anchor domain with that of AtGnTI, but in stably transformed cgl1-1 plants this chimeric GnTI still did not lead to full complementation of the cgl1-1 phenotype. Combined, the results indicate that activity of HuGnTI in plants is limited by a combination of reduced protein stability, alternative protein targeting and possibly to some extend to lower enzymatic performance of the catalytic domain in the plant biochemical environment.

Statistical models discriminating between complex samples measured with microfluidic receptor-cell arrays.

Data analysis for flow-based in-vitro receptomics array, like a tongue-on-a-chip, is complicated by the relatively large variability within and between arrays, transfected DNA types, spots, and cells within spots. Simply averaging responses of spots of the same type would lead to high variances and low statistical power. This paper presents an approach based on linear mixed models, allowing a quantitative and robust comparison of complex samples and indicating which receptors are responsible for any differences. These models are easily extended to take into account additional effects such as the build-up of cell stress and to combine data from replicated experiments. The increased analytical power this brings to receptomics research is discussed.

N-glycan occupancy of Arabidopsis N-glycoproteins.

Most secreted proteins in eukaryotes are modified on the amino acid consensus sequence NxS/T by an N-glycan through the process of N-glycosylation. The N-glycans on glycoproteins are processed in the endoplasmic reticulum (ER) to different mannose-type N-glycans or, when the protein passes through the Golgi apparatus, to different complex glycan forms. Here we describe the capturing of N-glycopeptides from a trypsin digest of total protein extracts of Arabidopsis plants and release of these captured peptides following Peptide N-glycosidase (PNGase) treatment for analysis of N-glycan site-occupancy. The mixture of peptides released as a consequence of the PNGase treatment was analyzed by two dimensional nano-LC-MS. As the PNGase treatment of glycopeptides results in the deamidation of the asparagine (N) in the NxS/T site of the released peptide, this asparagine (N) to aspartic acid (D) conversion is used as a glycosylation 'signature'. The efficiency of PNGase F and PNGase A in peptide release is discussed. The identification of proteins with a single glycopeptide was limited by the used search algorithm but could be improved using a reference database including deamidated peptide sequences. Additional stringency settings were used for filtering results to minimize false discovery. This resulted in identification of 330 glycopeptides on 173 glycoproteins from Arabidopsis, of which 28 putative glycoproteins, that were previously not annotated as secreted protein in The Arabidopsis Information Resource database (TAIR). Furthermore, the identified glycosylation site occupancy helped to determine the correct topology for membrane proteins. A quantitative comparison of peptide signal was made between wild type and complex-glycan-less (cgl) mutant Arabidopsis from three replicate leaf samples using a label-free MS peak comparison. As an example, the identified membrane protein SKU5 (AT4G12420) showed differential glycopeptide intensity ratios between WT and cgl indicating heterogeneous glycan modification on single protein. BIOLOGICAL SIGNIFICANCE: Proteins that enter the secretory pathway are mostly modified by N-glycans. The function of N-glycosylation has been well studied in mammals. However, in plants the function of N-glycosylation is still unclear, because glycosylation mutants in plants often do not have a clear phenotype. Here we analyzed which proteins are modified by N-glycans in plants by developing a glycopeptide enrichment method for plant proteins. Subsequently, label free comparative proteomics was employed using protein fractions from wild type and from a mutant which is blocked in modification of the N-glycan into complex glycans. The results provide new information on N-glycosylation sites on numerous secreted proteins. Results allow for specific mapping of multiple glycosylation site occupancy on proteins, which provides information on which glycosylation sites are protected or non-used from downstream processing and thus presumably are buried into the protein structure. Glycoproteomics can therefore contribute to protein structure analysis. Indeed, mapping the glycosylation sites on membrane proteins gives information on the topology of protein folds over the membrane. We thus were able to correct the topology prediction of three membrane proteins. Besides, these studies also identified limitations in the software that is used to identify single modified peptide per protein. This article is part of a Special Issue entitled: Translational Plant Proteomics.

N-glycoproteomics in plants: perspectives and challenges.

In eukaryotes, proteins that are secreted into the ER are mostly modified by N-glycans on consensus NxS/T sites. The N-linked glycan subsequently undergoes varying degrees of processing by enzymes which are spatially distributed over the ER and the Golgi apparatus. The post-ER N-glycan processing to complex glycans differs between animals and plants, with consequences for N-glycan and glycopeptide isolation and characterization of plant glycoproteins. Here we describe some recent developments in plant glycoproteomics and illustrate how general and plant specific technologies may be used to address different important biological questions.

Identification of the gene encoding the alpha1,3-mannosyltransferase (ALG3) in Arabidopsis and characterization of downstream n-glycan processing.

Glycosyltransferases are involved in the biosynthesis of lipid-linked N-glycans. Here, we identify and characterize a mannosyltransferase gene from Arabidopsis thaliana, which is the functional homolog of the ALG3 (Dol-P-Man:Man5GlcNAc2-PP-Dol alpha1,3-mannosyl transferase) gene in yeast. The At ALG3 protein can complement a Deltaalg3 yeast mutant and is localized to the endoplasmic reticulum in yeast and in plants. A homozygous T-DNA insertion mutant, alg3-2, was identified in Arabidopsis with residual levels of wild-type ALG3, derived from incidental splicing of the 11th intron carrying the T-DNAs. N-glycan analysis of alg3-2 and alg3-2 in the complex-glycan-less mutant background, which lacks N-acetylglucosaminyl-transferase I activity, reveals that when ALG3 activity is strongly reduced, almost all N-glycans transferred to proteins are aberrant, indicating that the Arabidopsis oligosaccharide transferase complex is remarkably substrate tolerant. In alg3-2 plants, the aberrant glycans on glycoproteins are recognized by endogenous mannosidase I and N-acetylglucosaminyltransferase I and efficiently processed into complex-type glycans. Although no high-mannose-type glycoproteins are detected in alg3-2 plants, these plants do not show a growth phenotype under normal growth conditions. However, the glycosylation abnormalities result in activation of marker genes diagnostic of the unfolded protein response.