Field of Invention

The present invention relates to detecting viral infections in plants, especially Cacao swollen-shoot virus (CSSV) infections. Recombinant viral coat protein antigens and binding reagents that specifically bind to thereto are provided together with a sensor incorporating the recombinant antigens. A competitive ELISA for detecting viral infections in plants is also provided.

Background to the Invention

Cacao swollen-shoot virus (CSSV) is a plant pathogen Badnavirus that infects Theobroma cacao trees (cacao trees), decreasing cacao yields and ultimately killing the trees within 3-4 years from infection. Surveys show that in Ghana, for example, around 17% of the cacao growing region is infected. Ghana has had a nationwide cutting out and rehabilitation programme which has resulted in at least 34 million trees being cut down since 2006. Despite many years of these procedures, CSSV infection has spread and is increasing. CSSV therefore presents a major problem for commercial cacao production.

CSSV infection occurs only in plant tissue and is not present in seeds. Current control measures include visual inspection by local government agencies of trees and removal of those infected and nearby trees. For visible symptoms to appear, trees may have been infected for many months prior to identification, enabling the possibility of significant disease transmission. Furthermore, CSSV infection is spread by wind carriage of up to fourteen species of the mealybug {Pseudococcidae) vector, leading to a significant range for disease transmission, and control measures for the vector have so far proved ineffective. In addition, although CSSV symptoms once manifested are usually distinctive, it can be difficult to distinguish the disease from other stresses such as nutrient deficiencies and effects of drought.

The current technology that is used to detect CSSV is based on the detection of viral DNA using a polymerase chain reaction (PCR) (Dzahini-Obiatey, 2010 and Oro et al., 2012). This can only be performed in a laboratory environment and by trained staff. There are PCR systems being developed for other applications that could be deployed in the field but the assays are very expensive and require skilled personnel.

There are a number of simple field testing devices for detection of plant virus infection of other crops, these are based on lateral flow technologies (e.g. Pocket Diagnostic, Abingdon Health Ltd UK). Despite the growing threat to cacao plants from CSSV and the increasing demand for cocoa beans to date, however, none have been developed for CSSV. Despite their ease of use, lateral flow devices typically lack sensitivity and interpretation is often subjective.

The need therefore remains for an assay system that is inexpensive and can be used by unskilled personnel on-site to carry out surveillance procedures and, unlike with standard lateral flow devices, is sufficiently sensitive and specific. Such a system would enable cocoa seedlings to be tested prior to planting and mature cocoa plants to be monitored in the field for CSSV infection. This will help to identify the rate and spread of CSSV disease to be mapped and will enable protection methods to be deployed in areas of low or no infection. This is of critical importance to cocoa growers and multinational users. The present invention aims to provide such a system.

Summary of the Invention

Accordingly, in a first aspect, the present invention provides a method for detecting Cocoa Swollen Shoot Virus (CSSV) using a porous membrane based sensor, the sensor comprising at least one recombinant CSSV coat protein antigen having at least 80% sequence identity to the amino acid sequence of SEQ ID NO: 2 or 4, the method comprising the steps of: (i) contacting cocoa plant material with a labelled binding reagent that specifically binds to the recombinant coat protein antigen to produce a mixture comprising the plant material and binding reagent; and (ii) contacting the mixture with the membrane based sensor.

The recombinant CSSV coat protein antigens have been developed by the present inventors to provide a universal CSSV coat protein antigen that can be used to develop binding reagents and in assays. The recombinant coat protein antigens have been prepared based on the analysis of Open Reading Frame 3 (ORF3) of multiple CSSV genomes and using highly conserved regions of CSSV capsid protein. The recombinant coat protein antigens can be used to generate binding reagents capable of accurately detecting multiple CSSV variants, whilst limiting or avoiding cross reactions with other viruses such as other badnaviruses.

In a second aspect the present invention provides an assay for detecting a viral infection in a plant, the assay comprising the steps of:

contacting plant material with a labelled recombinant binding reagent that specifically binds to a viral coat protein antigen to produce a mixture comprising the plant material and labelled binding reagent; and

contacting the mixture with viral coat protein antigen which is immobilised in or on a surface;

removing labelled recombinant binding reagent that is not bound to the immobilised viral coat protein antigen; and

detecting the presence of the remaining labelled recombinant binding reagent to determine the presence or absence of the viral infection,

wherein the amount of labelled recombinant binding reagent detected is inversely correlated with the number of viral particles present in the plant material.

In a third aspect the present invention provides an isolated binding reagent that specifically binds to a recombinant Cocoa Swollen Shoot Virus (CSSV) coat protein antigen having at least 80% sequence identity to the amino acid sequence of SEQ ID NO: 2 or 4 and/or at least 80% sequence identity to the nucleotide sequence of SEQ ID NO: 1 or 3.

In a fourth aspect the present invention provides a porous membrane based sensor for detecting CSSV comprising a recombinant CSSV coat protein antigen. The sensor can be used in the field to detect CSSV in asymptomatic plants and can reduce or avoid the need for expensive laboratory testing. Additionally, the sensor of the present invention allows the testing to be carried out by unskilled personnel with the consequence that testing can made more widely accessible to cocoa growers.

Description

The isolated binding reagent of the present invention specifically binds to a recombinant CSSV coat protein antigen having at least 80% sequence identity to the amino acid sequence of SEQ ID NO: 2 or 4 and/or at least 80% sequence identity to the nucleotide sequence of SEQ ID NO: 1 or 3.

Whilst the genome sequences of several strains of CSSV have been published, these sequences are not fully annotated. Consequently, whilst the skilled person might know approximately where a protein would be located in the genome they would not know exactly where the protein starts and ends within that genome sequence. The present inventors reviewed five published CSSV genome sequences and aligned them with over 20 other badnavirus sequences to identify conserved regions likely to form part of the coat protein antigen. The C-terminal region of the coat protein was identified based on a conserved zinc knuckle domain in ORF 3 and, from previous SDS page analysis of coat proteins, a sequence length of up to 345 amino acids was identified. The inventors then searched the aligned sequences within the boundary of 345 amino acids and identified a conserved site across the five published CSSV reference strains, which forms the N-terminus and generates a 334 amino acid sequence (SEQ ID NO:2). The corresponding DNA sequence (1002bp; SEQ ID NO: l) was identified on a highly virulent New Juaben strain and the recombinant protein was produced by inserting the sequence into a vector and expressing the vector in E. Coli. This produced the recombinant coat protein antigen of SEQ ID NO:2. Subsequently, 30 additional CSSV sequences were published (Muller et a , 2017; Chingandu et al., 2017) and the above approach was repeated to provide the recombinant coat protein antigen of SEQ ID NO:4 and the corresponding nucleotide sequence of SEQ ID NO:3.

The recombinant coat protein antigens described herein therefore do not include the full coat protein sequence and do not correspond to any one naturally occurring CSSV coat protein antigen. The recombinant coat protein antigens can therefore be used to generate binding reagents able to detect multiple strains of CSSV, which contrasts with known anti-CSSV antibodies, which typically only detect one strain. Additionally, because the binding reagents are generated using recombinant coat protein antigens rather than using plant material, high levels of background activity can be reduced or avoided. The binding reagents generated from the recombinant coat protein antigens can therefore provide more sensitive tests because they can correctly identify the presence of lower concentrations of CSSV in a sample. In embodiments of the invention the isolated binding reagent may bind to a recombinant CSSV coat protein antigen having at least 85%, or at least 90%, or at least 95% sequence identity with one or more of SEQ ID NOs: l-4. The recombinant CSSV coat protein antigen may comprise a sequence having at least 98% or 99% or 100% sequence identity with one or more of SEQ ID NOs: l-4. In embodiments of the invention the recombinant CSSV coat protein antigen may consist of a sequence having at least 85%, or at least 90%, or at least 95% sequence identity with one or more of SEQ ID NOs: l-4. The recombinant CSSV coat protein antigen may consist of a sequence having at least 98% or 99% or 100% sequence identity with one or more of SEQ ID NOs: l-4. In embodiments of the invention the recombinant CSSV coat protein antigen may consist of a sequence according to any one of SEQ ID NOs: 1-4. The recombinant CSSV coat protein antigens can be used to generate binding reagents that can detect multiple strains of CSSV, whilst avoiding false positives.

In the description above, the term "identity" is used to refer to the similarity of two sequences. For the purpose of this invention, it is defined here that in order to determine the percent identity of two sequences, the sequences are aligned for optimal comparison purposes (e.g., gaps can be introduced in the sequence of a first sequence for optimal alignment with a second amino or nucleic acid sequence). The nucleotide/amino acid residues at each position are then compared. When a position in the first sequence is occupied by the same amino acid or nucleotide residue as the corresponding position in the second sequence, then the molecules are identical at that position. The percent identity between the two sequences is a function of the number of identical positions shared by the sequences (i.e., % identity = number of identical positions/total number of positions (i.e. overlapping positions) x 100). Generally, the two sequences are the same length. A sequence comparison is typically carried out over the entire length of the two sequences being compared.

The skilled person will be aware of the fact that several different computer programs are available to determine the identity between two sequences. For instance, a comparison of sequences and determination of percent identity between two sequences can be accomplished using a mathematical algorithm. For example, the percent identity between two nucleic acid sequences can be determined using the sequence alignment software Clone Manager 9 (Sci-Ed software - www.scied.com) using global DNA alignment; parameters: both strands; scoring matrix: linear (mismatch 2, OpenGap 4, ExtGap 1).

Alternatively, the percent identity between two amino acid or nucleic acid sequences can be determined using the Needleman and Wunsch (1970) algorithm which has been incorporated into the GAP program in the Accelrys GCG software package (available at http://www.accelrys.com/products/gcg/), using either a Blosum 62 matrix or a PAM250 matrix, and a gap weight of 16, 14, 12, 10, 8, 6, or 4 and a length weight of 1, 2, 3, 4, 5, or 6. A further method to assess the percent identity between two amino acid or nucleic acid sequences can be to use the BLAST sequence comparison tool available on the National Center for Biotechnology Information (NCBI) website (www.blast.ncbi.nlm.nih.gov), for example using BLASTn for nucleotide sequences or BLASTp for amino acid sequences using the default parameters.

The binding reagent that specifically binds to the recombinant CSSV coat protein antigen is typically recombinant and may be an antibody, an aptamer, an affimer, or a DNA binding protein. Preferably the binding reagent is an antibody such as a monoclonal antibody or a polyclonal antibody. Most preferably the binding reagent is a recombinant monoclonal antibody, such as a recombinant F(ab)2 monoclonal antibody.

Suitable monoclonal antibodies may comprise the CDRs of SEQ ID NOs: 8-13 or SEQ ID NOs: 16-21 or SEQ ID NOs: 24-29. For example, the monoclonal antibody may comprise a light chain variable region sequence selected from SEQ ID NOs: 7, 16 or 25 and a heavy chain variable region sequence selected from SEQ ID NOs: 6, 15 or 24. In particular, the monoclonal antibody may comprise the light chain variable region of SEQ ID NO: 7 and the heavy chain variable region of SEQ ID NO: 6; or the light chain variable region of SEQ ID NO: 16 and the heavy chain variable region of SEQ ID NO: 15; or the light chain variable region of SEQ ID NO: 25 and the heavy chain variable region of SEQ ID NO: 24. The monoclonal antibody may comprise or consist of an amino acid sequence of having at least 80% sequence identity to SEQ ID NOs: 5, 14 or 23.

Monoclonal antibodies as described herein have been shown to detect multiple strains of CSSV, including New Juaben, Kpeve and Nsaba. The binding reagent may be conjugated to a label, such as a fluorescent label, to provide a labelled binding reagent. Fluorescent labels that might be used include eosin, fluorescein, cyanine dyes, nanoparticles with fluorescent characteristics (e.g. EUIII and up converting nanoparticles (UNCPs)) and quantum dots.

Many fluorescent labels can have poor signal to noise ratios and decompose on prolonged re-stimulation because of photo-bleaching. Additional problems can be encountered due to quenching of the signal by the sample and other components in the solution, and optical absorption at the measurement wavelengths. Quantum dots provide photostability, broad adsorption spectra and intense narrow emission spectra. Accordingly, whilst other labels can be used, quantum dots are a preferred label.

Suitable labels may have emission wavelengths in the infrared or near-infrared ranges. For example, the label may have emission wavelengths of about 500nm to about 600nm, and/or about 700nm to about lOOOnm, preferably about 700nm to about 900nm, more preferably about 800nm. The emission wavelength of the label preferably avoids peak chlorophyll emission and adsorption (as shown in Fernandez-Jaramillo et al, 2012).

The present invention additionally provides a porous membrane based sensor for detecting CSSV comprising a recombinant CSSV coat protein antigen (as described above). The sensor can be used in the field to detect CSSV in asymptomatic plants and is suitable for use by unskilled personnel. The sensor may be formed from layers of membrane, such as nitrocellulose, polycarbonate or other high protein binding porous membrane, which act as a solid phase for immobilisation of capture reagents and allow the reagents and sample to move through the membranes by capillary flow.

Preferably the sensor comprises a superficial sieve layer having a pore size of at least lpm. The sieve layer acts to remove plant debris from the sample. The sieve layer may have pore size of about lOpm to about lOOOpm, preferably about lpm to about 800pm, more preferably about 20 pm to about 30pm. In embodiments of the invention the sieve layer may have a pore size of about 25pm. The sieve layer may be formed from a material such as rayon polyester. Preferably the sensor additionally comprises a capture layer in the form of a porous membrane layer in which the at least one recombinant CSSV coat protein antigen is immobilised. The membrane may have pore size of about 0.05pm to about 20pm, preferably about 0.1pm to about 1pm. In embodiments of the invention the membrane layer may have a pore size of about 0.8pm. Target present in the sample competes with the immobilised antigen for binding to a labelled binding reagent (as described above). The capture layer may be formed from materials such as nitrocellulose.

Preferably the sensor additionally comprises a sink layer in the form of an absorbent layer adjacent the capture layer. The sink layer acts to draw liquid through the layers of the sensor. The sink layer may be formed from materials such as cellulose, cotton linter fibres, wood pulp, and sodium polyacrylate super- absorbent polymers or mixtures thereof.

Preferably the sensor additionally comprises a blocking layer in the form of a porous non-reflective layer between the capture layer and the sink layer. This layer acts to prevent light exciting any unbound reporter in the sink layer, thereby reducing nonspecific background signal.

Plant material can be mixed with a labelled binding reagent prior to being contacted with the sensor. The sieve layer removes plant debris and any free/unbound labelled binding reagent binds to the immobilised antigen in the capture layer. Labelled binding reagent bound to antigen (CSSV) in the sample is not immobilised in the capture layer and is drawn into the sink layer. The blocking layer prevents labelled binding reagent in the sink layer from being detected. The signal is detected from labelled binding reagent bound to the immobilised antigen in the capture layer and the amount of the labelled binding reagent present will be inversely correlated with the amount of CSSV present in the plant material.

The present invention additionally provides a method for detecting Cocoa Swollen Shoot Virus (CSSV) using a porous membrane based sensor as described above, the method comprising the steps of:

(i) contacting cocoa plant material with a labelled binding reagent as described herein to produce a mixture comprising the plant material and binding reagent; and

(ii) contacting the mixture with the membrane based sensor. The cocoa plant material may be any plant material and is preferably leaf or stem material. The use of cocoa plant stem material may provide particular advantages as the virus can be extracted by soaking the stems (i.e. no macerating required) and because stems do not contain chlorophyll, so they don't have the problems associated with chlorophyll autofluoresence that can be encountered when using leaf material.

The present invention additionally provides an assay for detecting a viral infection in a plant, the assay comprising the steps of:

contacting plant material with a labelled recombinant binding reagent that specifically binds to a viral coat protein antigen to produce a mixture comprising the plant material and binding reagent;

contacting the mixture with viral coat protein antigen which is immobilised in or on a surface;

removing labelled recombinant binding reagent that is not bound to the immobilised viral coat protein antigen; and

detecting the presence of the remaining labelled recombinant binding reagent to determine the presence or absence of the viral infection,

wherein the amount of labelled recombinant binding reagent detected is inversely correlated with the number of viral particles present in the plant material.

The viral coat protein antigen may be a recombinant antigen. The viral coat protein antigen may be a recombinant CCSV coat protein antigen as described above.

The labelled recombinant binding reagent may be an antibody. For example, the recombinant binding reagent may be a binding reagent according to the first aspect of the invention.

The virus may be a Badnavirus, the Badnavirus may be CSSV.

The label may be a fluorescent label as described above. For example, the label may be a quantum dot having an excitation wavelength of at least 600 nm.

Brief Description of the Drawings The invention will now be described in detail, by way of example only, with reference to the figures.

Figure 1 shows a schematic of an embodiment of the porous membrane based sensor. A shows an arrangement of porous layers including : 1. a sieve layer formed of a large pored membrane used to exclude plant debris; 2. a capture layer formed of a porous membrane containing immobilised antigen (e.g. a recombinant viral coat protein); 3. a dark layer formed of a porous layer; and 4. a sink layer in the form of an absorbent material to pull liquid through the superficial porous membranes. B shows movement of the sample through the membrane by capillary action. Free recombinant antibody/reporter (e.g. quantum dots) binds to the recombinant viral coat protein immobilised in the capture layer (2). Anti body/ reporter bound to plant virus in the sample passes through the membrane (3) and is taken up by the sink (4). C The sensor is exposed to a light source which causes the reporter to exhibit a strong fluorescence at a defined wavelength depending on the reporter used, e.g. depending on the diameter of quantum dots. Dark layer 3 prevents light exciting unbound fluorescent reporter in sink layer 4, thereby reducing nonspecific background signal.

Figure 2 shows fluorescence emission excitation matrices (EEMs) for 605 and 705nm Quantum Dots (QDs).

Figure 3 shows pixel density of streptavidin tagged QDs 605nm captured on nitrocellulose membrane at a range of concentrations of biotinylated goat anti- rabbit antibody. Measurements taken with an Alphalmager and analysed using ImageJ software.

Figure 4 shows bar graphs illustrating optimal horizontal (A) and vertical (B) positioning of QDs on the membrane for detection of fluorescence by ESEIog Fluorimeters.

Figure 5 shows results of the assessment of the efficiency of release of virus from CSSV infected Theobroma cacao leaf tissue by ball bearings.

Figure 6 shows a bar chart illustrating the selectivity of candidate recombinant antibodies. Anti-CSSV-CP-01 antibodies that bind to CSSV-CP-01 but not 1/100 dilution of CSSV negative Theobroma Cacao plant extract, BSA, N1-CD33-HIS6 tag nor unrelated protein with N1-CD33-HIS6 tag. (Data provided by manufacturer.)

Figure 7 shows A: results of a direct ELISA assay to confirm binding of fifteen candidate anti-CSSV recombinant antibodies to recombinant CSSV coat protein antigen (CSSV-CP-01), n=3; and B: Competitive ELISA results showing percentage inhibition produced by CSSV positive and negative plant extract, n=2, of the interaction between fifteen anti-CSSV recombinant antibodies to CSSV-CP- 01 antigen.

Figure 8 shows competitive ELISA results for selected recombinant monoclonal antibodies with CSSV positive and negative plant extracts (n=2). The results shown are the average of two experiments with n=3 for each experiment, different leaves were tested in each experiment for selected recombinant antibodies at a range of dilutions of CSSV positive and negative plant extract.

Figure 9 shows the results of a competitive ELISA results with a qPCR confirmed CSSV infected plant samples and the effect of binding of recombinant antigen CSSV-CP-01 and recombinant monoclonal antibodies Ab31988.1, Ab31999.1, AbD31998 and AbD31024.1.

Figure 10 shows detection of binding of recombinant monoclonal antibodies to antigen CSSV-CP-01, using biotinylated recombinant monoclonal antibody AbD31024.3 and a CSSV biosensor as described herein n=2.

Figure 11 shows binding of a range of concentrations of biotinylated CSSV antibody AbD31024.3 to a range of concentrations of recombinant antigen CSSV- CP-01 using a CSSV biosensor as described herein.

Figure 12 shows the binding of different recombinant monoclonal antibodies to a range of concentrations of CSSV-CP-01 antigen using a CSSV biosensor as described herein.

Figure 13 shows the binding of two concentrations of biotinylated recombinant monoclonal antibody AbD 31998.5 binding to a range of concentrations of CSSV- CP-01 antigen using a CSSV biosensor as described herein. Figure 14 shows results of competition of the binding of biotinylated recombinant monoclonal antibody AbD 31998.5 to CSSV-CP-01 antigen by a range of concentrations of free CSSV-CP-01 antigen (n=3).

Figure 15 shows results from a biosensor demonstrating competition by CSSV positive plant extract (n=3).

Figure 16 shows a conceptual diagram of the CSSV biosensor detection system for use in the field.

Figure 17 shows fluorescence obtained with controls and free CSSV-CP-01 measured in the biosensor during field trials.

Figure 18 shows a comparison of the mean number of CSSV copies/cell present in CSSV infected symptomatic leaves, CSSV infected non-symptomatic leaves and uninfected (negative) leaves.

Figure 19A and 19B represent summary data of ELISA assays performed. A: graph showing dose response of CSSV recombinant protein. B: Graph showing mean and SE of multiple leaves taken from the same plant.

Figure 20 shows competitive ELISA results using AbD31998.1 at three plant sample dilutions: A % inhibition of CSSV ELISA using samples at 1/10 concentration; B % inhibition of CSSV ELISA using samples at 1/100 concentration; and C % inhibition of CSSV ELISA using samples at 1/1000 concentration.

Figure 21 shows CSSV competitive ELISA and Taqman qPCR results for individual infected CSSV plant extracts. All samples with >45% inhibition in the CSSV competition ELISA had > 0.4 CSSV DNA copies/cell. These samples are marked as positive for CSSV.

Figure 22 shows a comparison of three anti-CSSV recombinant antibodies in the competitive CSSV ELISA at 1.25 pg/ml of CSSV antigen 1 (CSSV-CP-01).

Figure 23 shows the interaction of recombinant monoclonal antibodies with CSSV-CP-01 and CSSV-CP-02. Examples

Work conducted at The University of the West of England (UWE), Bristol established a strategy to develop a rapid hand held in field biosensor to detect Cocoa Swollen Shoot Virus (CSSV) infection of Theobroma Cocoa.

Evaluation of florescent reporter

Quantum dots (QDs) were evaluated as a fluorescent reporter and were shown to fluoresce at specific wavelengths dependant on the size of the quantum dots (Figure 2). Pilot work showed the proof of concept to use streptavidin tagged quantum dots binding biotin conjugated antibody captured on a rapid flow through paper based cassette (Figure 3). Other membrane types evaluated include PVDF, Polycarbonate and mixed Cellulose Ester (Data not shown). Three ESELog ESML10-MB-3018 confocal fluorescence detectors were commissioned and produced with two excitation wavelengths:

El - 365nm, E2 - 660nm and two emission filters D1 - 625nm , D2 - 720nm.

On evaluation it was found that the positioning of the quantum dots on the membrane gave maximum readings when positioned off centre (see Figure 4A and B).

Sequence analysis of published CSSV sequences to establish the ORF3 region of CSSV associated with the viral coat protein.

Antibodies raised to virus purified from plant leaves have shown high background values in immunoassays. A strategy was developed to determine the CSSV coat protein sequence by alignment with other similar viral sequences and to produce a recombinant CSSV coat protein antigen (CSSV-CP-01). This would then be used to generate recombinant binding reagents e.g. Aptamers, Monoclonal Antibodies or Affimers, by phage display; this would therefore limit background interference as no plant material was used in the selection process. Research was performed into the costs, requirements and risk of the selection process, which led to a final choice of generation of recombinant Monoclonal Antibodies by phage display.

Tables 1 & 2 show reagents and equipment produced.

Table 1 - Reagents produced.

Table 2 - Equipment produced Optimisation of extraction of virus from plant material.

A supply of CSSV infected and non-infected Theobroma cocoa plants was established from Reading university and were housed in the Envirotron at UWE. A range of buffers and methods of extraction were evaluated using qPCR to determine success.

Method of extraction from leaves

The leaves were harvested from CSSV infected and uninfected cocoa plants and roughly chopped with scissors into 0.5cm pieces. 60 mg leaf tissue was weighed and placed in a 7ml bijoux or Eppendorf tube containing lOx 4mm ball bearings and 2 ml 0.1M Phosphate buffer, pH 7.2. The vessel was shaken for approximately 2 minutes. The solution was passed through a sieve layer with a minimum pore size O. lum and collected. The solid dry plant tissue retained by the sieve layer was discarded and ball bearings recovered and washed. The plant sample was then added to the biosensor. A table showing the effect of different amounts of leaf tissue, size of ball bearings and volume of buffer is shown in Figure 5. Method of TaqMan qPCR

Leaf tissue was extracted in on the day of collection and DNA purified immediately from a 20mI sample using Qiagen Plant Dneasy mini kit. The kit was used following the manufactures instructions but omitted the RNAse treatment step and eluted sample twice in the same IOOmI buffer at the end to enrich. Samples were stored at -20°C until use.

TaqMan qPCR was then performed on the DNA samples using a CSSV primer/probe and a plant genomic probe and the Sensifast No RoxMaster Mix (Bioline Cat. BIO-98005). All samples were run in triplicate.

The CSSV and plant genomic probe/ primer mix comprised of:

2.5mI (100mM) probe

10mI (100mM) forward primer

10mI (100mM) forward primer

77.5mI elution buffer (Qiagen)

CSSV ORF3 primers and probe

F-74: 5'-CTGAAGCGAGTAGGCAACAA-3'

R-151 : 5'-CAGTCCAAGGGATGGACTCT-3'

P-129 : 5'-TCCATCAGGTTGCCATGGCA-3' (5'Fam - 3'Tamra)

Primers & probe for nuclear marker in flanking region of single copy T. cacao microsatellite marker mTcCIR25

F-mTcCIR25 : 5'-CAGATAAGGAAAGGTGGAGTTTGG-3'

R-mTcCIR25 : 5 C A AG A AT GTCTCCT AC ATT C ACTACG - 3 ' P-mTcCIR25 : 5'-TTCCCGTAAGCTTCGTCCCAGATGC-3' (5'Fam -

3’Tamra)

Each PCR reaction comprised of: 0.8mI probe/primer mix

10mI Mastermix (Bioline)

4.2mI nuclease free water

5mI DNA The reactions were run on a Rotor Gene Q instrument. Hold 95° 5', Cycle: 95°C 10s (acquiring to cycling A), 60°C 45s (acquiring to cycling B). Cycle is repeated 60 times.

The number of CSSV copies were determined in each sample by comparison to a synthetic CSSV references DNA oligomer of know copy number and the CSSV copy number per cell was estimated from the ratio of CSSV copy to plant cell copy number.

Assay development

Evaluation of the interaction of recombinant antigen CSSV-CP-01 and recombinant antibodies.

Testing of Theobroma cacoa plant samples/extracts by competitive ELISA

Materials:

Coating buffer: 0.1M sodium carbonate buffer pH 8.6

Sodium carbonate MW 106, so 1.06g/100 mis dH20

Sodium hydrogen carbonate MW 84.01, so 0.86g/100 mis dH20, Mix to pH

8.6

Blocking buffer PBS, 1% Bovine serum albumin (Sigma A7030, #SLBL4277V)

Wash buffer (PBS, 0.05% tween 20)

Antibody dilution buffer (PBS, 0.05% tween 20, 0.1% BSA)

Tetramethylbenzidine Liquid Substrate (Thermo)

Microlon high binding plate (Greiner 655061)

CSSVP-CP-01 recombinant fusion protein - Coating concentration 1 to 10pg/ml

Recombinant reagents

Goat anti Human IgG F(ab')2: HRP (Bio-Rad STAR126P)

AbD31998.1 anti-CSSV antibody (BioRad) final concentration 50ng/ml

Plant samples:

Leaves were harvested from Theobroma cacao plants in the museum collection at CRIG, Ghana and roughly chopped with scissors into 0.5cm pieces. 60mg Leaf tissue was weighed and placed in a 7ml bijoux or Eppendorf tube containing 10 x 4mm ball bearings and 2mls of 0.1M Phosphate buffer B pl-17.2 was added and the vessel shaken for approximately 5 minutes.

Procedure:

1. Allow all reagents to reach room temperature before use. Mix all liquids gently prior to use.

2. Prepare CSSVP-CP-01 Antigen, diluted to 1 - 10pg/ml in coating buffer.

3. Take immunoassay plate. Add 50pl of prepared solution to wells. Cover plate and Incubate at 4°C for 18 hours in a humid box.

4. Decant samples and discard. Wash wells 3 times in Wash buffer using a squirt bottle. Tap plate gently on tissue following every wash to remove all liquid.

5. Fill the wells with 200mI Blocking buffer. Incubate at room temperature for 2 hours.

6. Prepare recombinant monoclonal antibody at 200ng/ml in antibody dilution buffer ie. at 2x final concentration.

7. Prepare competing moieties at 2x final concentration in antibody dilution buffer:

i. Plant sample at 1/5

ii. Plant sample at 1/50

iii. Plant sample at 1/500

Recombinant antigen at 1.25 and 0pg/ml.

Equal volumes of antibody and competing recombinant antigen, plant extract or buffer were mixed in an Eppendorf tube and incubated for 30 mins at room temperature.

8. Decant blocking buffer and discard. Wash wells 4 times in Wash buffer.

Add 50mI of primary antibody/competitor to appropriate well. Cover plate and incubate at room temperature for 60 minutes.

9. Wash wells 5 times in Wash buffer.

10. Add 50mI of secondary antibody Goat anti Human IgG F(ab')2: HRP at 1/2500 in antibody dilution buffer to each well. Cover plate and incubate at room temperature for 60 minutes.

11. Wash wells 5 times in Wash buffer.

12. Add 100mI 3, 3', 5 ,5'-Tetramethylbenzidine Liquid Substrate to the wells.

Incubate for 10 minutes at room temperature in the dark.

13. Add 100mI 1M Sulphuric Acid

14. Read absorbance of wells at 450nm in a microtitre plate reader Evaluation of candidate recombinant monoclonal antibodies

Fifteen candidate monoclonal antibodies were supplied and evaluated for their ability to bind to recombinant antigen by ELISA (see Figure 7A). A competitive ELISA was then performed with CSSV positive and negative leaf samples (see Figure 7B). Candidate antibodies were selected for biotinylation based on results with leaf extracts. Antibodies that bound to CSSV-CP-01 antigen that were competed by CSSV positive leaf extracts, but not by CSSV negative leaf extracts were identified. Recombinant antibodies were selected for biotinylation and used in the Biosensor development including AbD31024.2, AbD31988.1, AbD31999.1 and AbD31998.1.

The results of replicate ELISA experiments for the selected antibodies, across a range of plant extract dilutions are shown in Tables 3 and Figures 8 and 9. A strong reproducible dose response was shown by all antibodies. A summary of the selectivity and specificity of each candidate recombinant monoclonal antibody is shown in Table 4.

Table 3: Competitive ELISA results of qPCR confirmed CSSV infected plant samples and binding of recombinant antigen and recombinant monoclonal antibodies.

antibodies binding to CSSV and plant tissue. *Above background. Antibodies highlighted in italics are assay candidates. Antibodies in bold were selected for progression, the amino acid and DNA sequences of antibodies AbD31998.1, AbD31999.1 and AbD31988.1 are shown SEQ ID NOs: 5-31.

Development of competitive flow through assay - CSSV biosensor.

The first available biotinylated candidate antibody AbD31024.3 was used in the development of the biosensor in terms of materials, timing, buffers, QDs and detection with the ESEIog fluorimeter.

Testing of plant samples/extracts using the Biosensor

Materials

MDI membrane 0.8m CLW-040-SH34 (DL191/3 #LD191917G)

MDI Absorbent pads AP080 of 24.5mmx 36mm

Tesco disposable nappy - absorbent layer

CSSVP-CP-01 recombinant fusion protein AbD31998.5 biotinylated 0.65mg/ml

Qdot ^® 605nm streptavidin conjugate (life technologies cat. Q10101MP lot #1826419)

Phosphate buffered saline pH 7.3 (PBS)

Non-fat dried milk (Marvel)

Method for biosensor competition assay

1. The membrane was placed onto a foil (non - porous) solid support which enabled antibody to be dried to the membrane. The membrane was cut to fit into a cassette and was marked with a pencil so that placement of the dots could be navigated in to the read area. CSSV antigen was diluted in PBS to 100pg/ml. 10pls was spotted onto the paper cast Nitrocellulose Membrane of pore size < 1 pm and left to air dry at room temperature for 15 mins.

2. The membranes were then blocked in 5% skimmed milk in PBS for 2 hours at room temperature and washed 3 times in PBS. Membranes were stored at 4°C for 24 hours.

3. Each membrane was assembled onto a stack of 2 MDI Absorbent pads AP080, and a single layer of absorbent material cut to fill the entire area of the cassette.

Biotinylated recombinant monoclonal antibody was diluted to 8pg/ml in PBS (2x final concentration).

4. Competing test samples were diluted in PBS to 2x final concentration. Plant samples were therefore diluted 1/5, 1/50, 1/500 and lpg recombinant antigen was diluted in lOOpl PBS. Sufficient quantities were prepared such that all tests were repeated in triplicate. lOOpl competitor or control PBS was mixed with lOOpl of recombinant monoclonal antibody and left at room temp for 30 mins to pre-incubate prior to assay. All combinations were prepared in triplicate.

5. 200pl of antibody/ competitor was dripped slowly onto the centre of the spot and allowed to flow through.

6. qDots 605 streptavidin conjugate were diluted to lOnM in PBS and lOpl was applied and allowed to flow through. This was followed by 2x PBS washes of 500pl.

7. The fluorescent signal of bound Qdots were measured at an excitation wavelength of 365nm and emission of 625nm using the Fluorescence Detector (E1D1) at a distance of 1.4cm. Detection of binding of monoclonal antibodies to antigen CSSV-CP-01 using biotinylated antibody AbD31024.3 is shown in Figure 10 and 11.

Evaluation of the other candidate recombinant monoclonal antibodies showed AbD31988.5 to give the highest signal and clearest dose response binding to

CSSV-CP-01 as shown in Figure 12. The reduced binding of antibody in the presence of free competing recombinant antigen CSSV-CP-01 is shown in Table 5 and Figures 13 and 14. Figure 16 also shows competition by CSSV positive plant extract.

Table 5: CSSV Biosensor results showing competition by two concentrations of competing recombinant antigen CSSV-CP-01 (n=3)

A compilation of three separate Biosensor experiments is shown in Tables 6-8. The results shown are the competition observed by recombinant antigen in three separate experiments using two different batches of CSSV-CP-Ola and CSSV-CP- 01b.

three separate experiments using two different batches of CSSV-CP-Ola, b (n=3 in each experiment)

Table 7. UWE, CSSV-CP-Ola, CSSV positive plant extract n=3, QDots - 705nm

Table 8. Competitive Flow through assay to demonstrate competition by CSSV- CP-Ola antigen n=3, QDots - 605nm

Report on the evaluation of reagents and immunoassays to detect Cocoa Swollen Shoot Virus (CSSV) at the Cocoa Research Centre of Ghana (CRIG).

Day 1. Familiarisation with laboratories, unpack reagents and materials, stage 1 of ELISA.

Day 2. Collection of leaf samples 1-15 from museum collection at CRIG includes three strains of CSSV, New Juaben, Kpeve and Nsaba. All samples collected were photographed (not shown).

Uninfected plants samples A-D, were collected from a different part of the CRIG compound. An accurate amount of each sample was weighed and the leaf tissue extracted. Plant extracts were stored at 4°C.

The ELISA assay was completed on 15 plant extracts, 1-15 and 4 CSSV negative plant extracts, A-D. Competing recombinant antigen was included as a control.

Day 3. Biosensor assay completed on xl5 plant extracts, 1-15 and x4 CSSV negative plant extracts, A-D. Competing recombinant antigen at was included as a control.

Results and Discussion

Results of the competitive ELISA for CSSV are shown in Tables 9 and 10 (n=3). No plate reader was available at CRIG to measure absorbance, therefore the results were determined by eye by two people. Photographs are available and results will be analysed using image analysis.

Table 9. Competitive ELISA results with a range of CSSV positive and negative plant samples

Table 10. Competitive ELISA results 500ng of recombinant antigen CSSV-

CP-01

Key: Negative Neg; Strong positives P, PP and PPP; Very weak positive (P)

The results show that 14 out of 15 plant leaf samples were positive for CSSV at a 1 : 10 to 1 : 100 dilution, no samples were positive for CSSV at 1 : 1000 dilution. The assay was able to detect CSSV infection by three strains of CSSV, New Juaben, KPEVE and NSABA. The Ghanaian ELISA assay only detects New Juaben. There were no convincing positives among the CSSV negative plant samples tested .

Subsequent laboratory QPCR results on the above samples have revealed that samples 1-15 contained CSSV. Melt analysis revealed two populations the New Juaben samples formed one group and the other strains formed another.

Table 11. CSSV Biosensor results, n=3 (technical)

The CSSV biosensor controls ( italics ) and a CSSV positive Sample 10. in Table 11 show that the Biosensor worked, and are illustrated in Figure 16.

A conceptual diagram of how the CSSV biosensor detection system will be used in the field is shown in Figure 17.

Example 2

Objectives:

• Validate recombinant antibody Ab 31998.1 in the competitive ELISA with CSSV antigen 1 (CSSV-CP-01) with CSSV infected symptomatic, CSSV non- symptomatic and CSSV non- infected plant extracts.

· Compare competitive CSSV ELISA results with the number of CSSV DNA copies/plant cell present in each sample.

• Evaluate 3 antibodies: AbD31998.1, AbD31988.1 and AbD31999.1 in the CSSV competitive ELISA to CSSV-CP-01 antigen competition.

• Compare the binding of 15 recombinant monoclonal antibodies to CSSV antigen 1 (CSSV-CP-01) and CSSV antigen 2 (CSSV-CP-02).

Results:

CSSV Taqman QPCR

A good dose response was observed for 10 ² - 10 ⁵ copies of synthetic CSSV DNA (data not shown). Summary of qPCR Data

• CSSV infected plants - 10 ⁴ to 10 ⁵ CSSV DNA copies detected.

• CSSV infected non-symptomatic plants - 10 ² to 10 ³ CSSV DNA copies detected.

• Uninfected plants - <50 CSSV DNA copies detected.

These data were normalised with plant DNA detected to determine the number of copies of CSSV DNA present per plant cell. This allows for differences in the success of DNA extraction to be accounted for.

Summary of qPCR Data of plant leaf extracts from 'CSSV infected' Theombroma cacoa plants

Table 12: Symptomatic leaves contain 333.5 - 6.4 CSSV copies/cell; Non- symptomatic leaves contain 1.9 - 0.1 CSSV copies/cell.

Summary of qPCR data of plant leaf extracts from 'uninfected' Theobroma cacoa plants 0.3 - 0.1 CSSV copies/cell Table 13: Uninfected leaves contain 0.6 - 0.1 CSSV copies/ cell (mean = 0.24)

Therefore from this experiment any sample containing >0.4 CSSV copies/cell is positive for CSSV. The mean data for these values are displayed in Figure 18.

Validation of reagents in the CSSV competitive ELISA and ability of the assay to detect CSSV in plant extracts. Figures 19A and 19B represent summary data of the ELISA assays performed. Fig 19A shows a dose response of CSSV recombinant coat protein. Fig. 19B shows the mean and SE of multiple leaves taken from the same plant.

Competitive ELISA using antibody AbD31998.1 shows results at three plant sample dilutions: 1/10 (Fig. 20A); 1/100 (Fig. 20B); and 1/1000 (Fig. 20C).

CSSV competitive ELISA and Taqman qPCR results for individual infected CSSV plant extracts are shown in Fig. 21. All samples with >45% inhibition in the CSSV competition ELISA had > 0.4 CSSV DNA copies/cell. These samples are marked as positive for CSSV.

Table 14: CSSV competitive ELISA and Taqman qPCR results for individual non- infected plant extracts.

2/12 CSSV non-infected plants had >45% inhibition in the ELISA, indicating some interference at the highest concentration tested (bold). However, one of those samples also had a qPCR result of > 0.4 CSSV DNA copies/cell indicating a genuine CSSV positive.

Three anti-CSSV recombinant antibodies (Ab 31998.1, Ab 31988.1 and Ab 31999.1) were compared by competitive ELISA at 1.25 pg/ml of CSSV antigen 1 (CSSV-CP-01) (Fig. 22). The observed antibody sensitivity to competition by CSSV-CP-01 was Ab31999 > Ab31998 > Ab 31988.

The interaction of the recombinant monoclonal antibodies with CSSV-CP-01 and CSSV-CP-02 was compared (Fig. 23). Antibodies AbD31997.1 and AbD31998.1 interacted strongly with both antigens CSSV-CP-01 and CSSV-CP-02.

Summary of results

• Validate recombinant antibody Ab 31998.1 in the competitive ELISA with CSSV antigen 1 (CSSV-CP-01)

Using the criteria of >45% inhibition in the ELISA as being positive for CSSV:

• 11/12 symptomatic leaf samples were positive

• 5/12 non symptomatic leaf samples were positive

• 2/12 uninfected plant samples were positive

• Only 1 of the uninfected samples was confirmed as true CSSV positive i.e.

>0.4copies/cells by qPCR, therefore

• 1 false positive was observed

• Evaluate and compare 3 antibodies AbD31998.1, AbD31988.1 and AbD31999.1

• Antibody sensitivity Ab31999 >Ab31998 >Ab31988

• Evaluate binding of 15 antibodies to CSSV antigen 1 and CSSV antigen 2 (CSSV -CP -02)

• All antibodies tested were originally selected for interaction with CSSV-CP- 01. Antibodies AbD31997.1 and AbD31998.1 were found to have the highest binding with both antigens.

• Antibody AbD31998.1 has previously been shown to be the most sensitive to competition.

Conclusion

The Taqman qPCR has supported the results obtained in the CSSV competitive ELISA in that those plants with detectable levels of CSSV above 0.4 copies/cell were also positive in the competitive ELISA. For two out of three plants tested, CSSV was detected in symptomatic and >50% non-symptomatic leaves. These results were obtained with CSSV-CP-01 and antibody AbD31998. Experiments have determined that antibody Ab31999 has greater sensitivity when tested with CSSV-CP-01. Antigen CSSV-CP-02 was produced using sequences from currently circulating strains of CSSV in Ghana and the Cote D'Ivoire; the antibody AbD31998 is also the antibody that binds strongly to both antigens.

References

H. Dzahini-Obiatey and R. T. V. Fox (2010). African Journal of Biotechnology Vol. 9 (5), pp. 593-603,

Muller, E., Jacquot, E., Yot, P. Journal of Virological Methods (2001) 93: 15-22

Oro, F., Mississo, E., Okassa, M. et al. Arch Virol (2012) 157: 509.

Chingandu,N., Kouakou,K., Aka,R., Ameyaw,G., Gutierrez, O. A., Herrmann H.W., and Brown, J.K. Virology Journal (2017) 14: 199.

Muller, E., Ravel, S., Agret,C., Abrokwah.F., Dzahini-Obiatey, H., Galyuon,I., Kouakou,K., Jeyaseelan,E.C., Allainguillaume,J., Wetten,A. Virus Res. 2018 Jan 15;244:235-251 Epub 2017 Nov 21 Fernandez-Jaramillo, A. A., Duarte-Galvan.C., Contreras-Medina, L.M., Torres-

Pacheco, I., de J. Romero-Troncoso, R., Guevara-Gonzalez, R.G and Millan- Almaraz,R.J. 2012 Instrumentation in Developing Chlorophyll Fluorescence Biosensing : A Review. Sensors, 12, 11853-11869