*Article* **Development of a Real-Time Loop-Mediated Isothermal Amplification Assay for the Rapid Detection of Olea Europaea Geminivirus**

**Sofia Bertacca 1,† , Andrea Giovanni Caruso 1,2,† , Daniela Trippa <sup>1</sup> , Annalisa Marchese <sup>1</sup> , Antonio Giovino <sup>2</sup> , Slavica Matic <sup>3</sup> , Emanuela Noris <sup>3</sup> , Maria Isabel Font San Ambrosio <sup>4</sup> , Ana Alfaro <sup>4</sup> , Stefano Panno 1,\* and Salvatore Davino 1,\***


**Abstract:** A real-time loop-mediated isothermal amplification (LAMP) assay was developed for simple, rapid and efficient detection of the Olea europaea geminivirus (OEGV), a virus recently reported in different olive cultivation areas worldwide. A preliminary screening by end-point PCR for OEGV detection was conducted to ascertain the presence of OEGV in Sicily. A set of six real-time LAMP primers, targeting a 209-nucleotide sequence elapsing the region encoding the coat protein (AV1) gene of OEGV, was designed for specific OEGV detection. The specificity, sensitivity, and accuracy of the diagnostic assay were determined. The LAMP assay showed no cross-reactivity with other geminiviruses and was allowed to detect OEGV with a 10-fold higher sensitivity than conventional end-point PCR. To enhance the potential of the LAMP assay for field diagnosis, a simplified sample preparation procedure was set up and used to monitor OEGV spread in different olive cultivars in Sicily. As a result of this survey, we observed that 30 out of 70 cultivars analyzed were positive to OEGV, demonstrating a relatively high OEGV incidence. The real-time LAMP assay developed in this study is suitable for phytopathological laboratories with limited facilities and resources, as well as for direct OEGV detection in the field, representing a reliable method for rapid screening of olive plant material.

**Keywords:** LAMP; OEGV; *Geminiviridae*; olive viruses

## **1. Introduction**

The olive tree (*Olea europaea* L.) belonging to the *Oleaceae* family is the most widely cultivated species of the *Olea* genus. Olive, providing edible fruits and storable oil, has been cultivated in the Mediterranean area since prehistoric times [1], and is regarded as the most economically important fruit tree in the Mediterranean basin [2]. Olive cultivation has, over the centuries, played an important role in the economic development of rural areas in the Mediterranean region, providing noteworthy sources of income and employment opportunities for the population in rainfed agricultural territories [3]. Even today, after thousands of years, the countries in this area produce about 90% of the olive fruits [4], while the olive cultivated area covers about 10 million hectares worldwide. According to the latest data available on FAOSTAT, among Mediterranean countries, Spain ranked the

**Citation:** Bertacca, S.; Caruso, A.G.; Trippa, D.; Marchese, A.; Giovino, A.; Matic, S.; Noris, E.; Ambrosio, M.I.F.S.; Alfaro, A.; Panno, S.; et al. Development of a Real-Time Loop-Mediated Isothermal Amplification Assay for the Rapid Detection of Olea Europaea Geminivirus. *Plants* **2022**, *11*, 660. https://doi.org/10.3390/ plants11050660

Academic Editor: Alessandro Vitale

Received: 7 February 2022 Accepted: 26 February 2022 Published: 28 February 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

major olive producer in 2020 (2,623,720 ha; 8.137.810 tons), followed by Italy (1,145,520 ha; 2,207,150 tons), Morocco (1,068,895 ha; 1,409,266 tons), Greece (906,020 ha; 2,790,442 tons), and Turkey (887,077 ha; 1,316,626 tons). According to ISTAT data [5], among the Southern Italian regions, Sicily (161,661 ha; 255,798 tons) played a significant role in olive and olive oil production, industry and export in 2021, being the third largest producer after Apulia (379,960 ha; 708,400 tons) and Calabria (184,410 ha; 680,275 tons). In Sicily, the olive crop has been cultivated since ancient times, and it is characterized by many ancient landraces/cultivars of high organoleptic quality [6]; its germplasm is distinguished by a wide genetic diversity, possibly related to its past domestication and spread and to some reproductive biological peculiarities such as self-incompatibility [7]. The production of olive oil in Sicily is based mainly on the autochthonous cultivars (cvs) 'Biancolilla', 'Cerasuola', 'Moresca', 'Nocellara del Belice', 'Nocellara Etnea', 'Ogliarola Messinese', 'Santagatese' and 'Tonda Iblea' [8]. The table olive industry is also appreciable, accounting to 10% of the total production of this region [5], mainly relying on the cv. 'Nocellara del Belice' and, to a minor extent, 'Nocellara Etnea', 'Ogliarola Messinese', and 'Moresca', produce large-sized fruits of high commercial value [8]. The current tendency in olive tree cultivation is moving towards the use of local cvs for high quality oil production (such as DOP—protected designation of origin), which is typical of specific geographic areas. For this reason, the local administration currently supports studies and activities aimed at the characterization and recovery of local and ancient cvs, in order to establish germplasm collections that limit genetic erosion [9]. However, a large number of diseases and disorders affect this crop, mostly caused by fungi, such as *Arthrinium phaeospermum*, *Phoma cladoniicola* and *Ulocladium consortiale*, recently discovered as new olive pathogens in Italy [10], but also by systemic pathogens including bacteria and viruses, which provoke significant yield losses. Indeed, in the last decade, olive production has suffered an enormous decline due to the emergence of biotic agents that have significantly undermined the Mediterranean economy related to olive and the olive oil industry; a dramatic example being the *Xylella fastidiosa* epidemic in 2013, which decimated olive trees in Apulia [11] and created huge losses for the local olive economy and oil production outputs, posing critical challenges for its management, as well as dramatic changes in the landscape [12,13]. Furthermore, the vegetative propagation of olive trees using cuttings of semi-wood has contributed over the years to the spread of systemic-pathogens, particularly viruses [14]. Nevertheless, despite the difficulty of associating specific symptoms to a particular virus, many viruses are easily transmitted through infected propagation material [15], and many olive infecting viruses are symptomless. Therefore, it is essential to better elucidate the evolutionary aspects of latent viruses in olive crops. In the last year, thanks also to the application of new technologies such as high-throughput sequencing (HTS), a new geminivirus called Olea europaea geminivirus (OEGV) has been identified in the olive tree [16], but its spread and pathogenicity remain puzzling. Since its first identification in Apulia [16] in the "Ogliarola" and "Leccino" cvs, OEGV was reported in California and Texas [17], Portugal [15], and Spain [18]. OEGV is classified as a putative member within the *Geminiviridae* family [16], currently including 14 genera [19] and few other still unassigned geminiviruses [20]. The evolutionary relationship of OEGV with other geminiviruses indicated that OEGV has distinctive genome features, possibly representing a new genus [15–17]. OEGV is characterized by a bipartite genome containing DNA-A and DNA-B. DNA-A (2775 nucleotides, nts) includes four ORFs, three in the complementary-sense encoding the replication-associated protein Rep (AC1), the transcriptional activator protein TrAP (AC2), the replication enhancer protein Ren (AC3) and one in the virion-sense, (AV1), encoding the coat protein (CP). DNA-B (2763 nts) includes two ORFs, BC1 in the complementary sense, with an unknown function and lacking known conserved domains typical of geminiviral proteins, and BV1 on the virion sense, possibly encoding the movement protein (MP). In bipartite geminiviruses, AC4/C4 protein is a symptom determinant involved in cell-cycle control, and interacts with CP and/or MP in the replicated genome transport from nucleus to cytoplasm and from cell-to-cell [15]. Curiously, no genes encoding AC4/C4 were found on the OEGV genome. In addition, the two DNA molecules present a common region (CR) of 403 nt that contains the TATA box and four replication-associated iterons with a unique arrangement compared to other geminiviruses [15–17]. In a recent survey, Alabi and co-workers [17] detected OEGV-positive olive trees originating from different locations, advancing the concept of a possible worldwide spread of this virus, likely due to the inadvertent movement of germplasms from clonally propagated infected but asymptomatic olive trees. As a matter of fact, OEGV does not appear to be clearly associated to any symptom in olive; moreover, a high degree of sequence conservation has been identified [18].

In this study, we aimed to investigate the presence of OEGV in Sicily and to develop a rapid detection protocol based on the LAMP methodology. In addition, an on-site olive sample homogenization procedure was developed replacing canonical DNA extraction methods, which is useful in evaluating the suitability of the LAMP assay for on-site OEGV testing.

#### **2. Materials and Methods**

#### *2.1. Plant Material Collection*

To investigate the presence of OEGV in Sicily, different surveys were carried out during spring 2021, focusing in particular on two olive producing sites in the Agrigento province (Sicily, Italy). The olive tree samples were randomly collected according to the hierarchical sampling scheme [21], with minor adaptations to olive plants. All samples were geo-referenced with the Planthology mobile application [22], collected from a total of 80 olive trees of 10 different cvs (40 trees randomly sampled for each site). Each sample consisted of 8 branches per plant (two for each plant cardinal point); samples were stored at 4 ◦C and processed within the next 24 h for subsequent molecular analyses.

#### *2.2. DNA Extraction and Sample Preparation*

Total DNA was extracted using the DNA extraction GenUPTM Plant DNA kit (Biotechrabbit GmbH, Berlin, Germany), following manufacturer's instructions with slight modifications. In brief, 3 g of tissue were homogenized in an extraction bag (BIOREBA, Reinach, Switzerland) using the HOMEX 6 homogenizer (BIOREBA, Reinach, Switzerland), with 3 mL extraction buffer (1.3 g sodium sulphite anhydrous, 20 g polyvinylpyrrolidone MW 24–40,000, 2 g chicken egg chicken albumin Grade II, 20 g Tween-20 in one L of distilled water, pH 7.4). Aliquots of 400 µL of the extract were added to the same volume of lysis buffer. The eluted DNA was resuspended in 100 µL RNase-free water; following two measurements with a UV–Vis NanoDrop 1000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA), samples were adjusted to approximately 50 ng/µL and stored at −20 ◦C.

#### *2.3. Preliminary Screening of OEGV by End-Point PCR*

The end-point PCR was conducted using the primer pair A2for/A4rev [16], amplifying an 831 bp fragment within the AV1 gene. PCR was performed in a final volume of 25 µL, containing 1 µL of total DNA extract, 20 mM Tris-HCl (pH 8.4), 50 mM KCl, 3 mM MgCl2, 0.4 mM dNTPs, 1 µM each primer, and 2 U Taq DNA polymerase (Thermo Fisher Scientific, Waltham, MA, USA) and RNase-free water to reach the final volume. Healthy olive plant DNA and water were used as control samples. The PCR was performed in a MultiGene OptiMax thermal cycler (Labnet International Inc., Edison, NJ, USA) with the following conditions: 95 ◦C for 5 min; 40 cycles of 95 ◦C for 30 s, 64 ◦C for 45 s, and 72 ◦C for 1 min; a final elongation at 72 ◦C for 10 min. PCR products were electrophoresed on 1.5% agarose gel, stained with SYBRTM Safe (Thermo Fisher Scientific, Waltham, MA, USA) and visualized by UV light.

#### *2.4. LAMP Primers Design*

The OEGV DNA-A complete sequence (GenBank Acc. No. MW316657) was used to design LAMP primers by the PrimerExplorer version 5 software (http://primerexplorer. jp/lampv5e/, accessed on 5 July 2021), choosing a 540-bp nucleotide sequence elapsing region within the AV1 gene. A set of six primers were selected, including two outer primers (forward and backward outer primer, F3 and B3, respectively), two inner primers (forward and backward inner primer, FIP and BIP, respectively), and two loop primers (forward and backward loop primer, LF and LB, respectively). The specificity of the primer set was tested in silico using the Nucleotide-BLAST algorithm (https://www.ncbi.nlm.nih.gov, accessed on 5 July 2021) available at the National Centre for Biotechnology Information (NCBI), in order to evaluate possible cross reactions with other viruses. This set of primers was also tested against the full genomic sequences of other geminiviruses reported in Italy using the Vector NTI Advance 11.5 software (Invitrogen, Carlsbad, CA, USA), in order to verify their affinity. The list included Tomato leaf curl New Delhi virus (ToLCNDV) (DNA-A: GenBank Acc. No. MK732932 and DNA-B: MK732933), Tomato yellow leaf curl Sardinia virus (TYLCSV) (GenBank Acc. No. GU951759), Tomato yellow leaf curl virus (TYLCV) (GenBank Acc. No. X15656), TYLCV-IL23 (GenBank Acc. No. MF405078), and TYLCV isolate 8-4/2004 (GenBank Acc. No. DQ144621).

## *2.5. OEGV Real-Time LAMP Assay Optimization*

The real-time LAMP assay was performed in a 12 µL reaction mixture containing 1.6 mM each of FIP-OEGV and BIP-OEGV, 0.2 mM each of F3-OEGV and B3-OEGV, 0.4 mM each of forward loop primer (LF-OEGV) and backward loop primer (LB-OEGV), 6.25 µL WarmStart LAMP 2X Mastermix (New England Biolabs, Beverly, MA, USA), and 0.25 µL of LAMP Fluorescent dye (New England Biolabs, Beverly, MA, USA), 1 µL of total DNA as template and nuclease-free H2O was added to reach the final volume. DNA extracted from ten samples previously analyzed by end-point PCR was used in the real-time LAMP assay, including a positive control (PC) and a healthy olive plant DNA as negative control (NC). Each sample was analyzed twice. The LAMP assay was conducted at 65 ◦C (according to manufacturer's instructions) for 60 min and fluorescence was acquired every 60 s, using a Rotor-Gene Q2plex HRM Platform Thermal Cycler (Qiagen, Hilden, Germany). A melting curve was calculated to record the fluorescence using the following protocol: 95 ◦C for 1 min, 40 ◦C for 1 min, 70 ◦C for 1 min and an increase of temperature at 0.5 ◦C/s up to 95 ◦C. During the amplification, the fluorescence data were obtained in the 6 carboxyfluorescein (FAM) channel (excitation at 450–495 nm and detection at 510–527 nm). The relative fluorescence units (RFU) threshold value was used, and the threshold time (Tt) was calculated as the time at which fluorescence was equal to the threshold value.

#### *2.6. Features of Real-Time LAMP Assay: Sensitivity and Comparison to Conventional PCR, Reaction Time and Specificity*

To set up the conditions of the LAMP assay, an amplicon obtained by subjecting an OEGV-positive sample to end-point PCR (see above) was purified from agarose gel using an UltraClean™ 15 DNA purification kit (MO-BIO Laboratories, Carlsbad, CA, USA), following manufacturer's instructions. The purified DNA (named pcr-DNA) was quantified using a UV–Vis NanoDrop 1000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). The number of copies was determined as follows: [Number of copies = (amount of DNA in nanograms <sup>×</sup> 6.022 <sup>×</sup> <sup>10</sup>23)/(length of DNA template in bp <sup>×</sup> <sup>1</sup> <sup>×</sup> <sup>10</sup><sup>9</sup> <sup>×</sup> 650)]. To determine the OEGV real-time LAMP optimal reaction time and sensitivity, ten-fold serial dilutions of the sample (named pcr-DNA) were used as a template for both real-time LAMP assay and end-point PCR. Moreover, to evaluate the specificity of the LAMP assay and to assess potential non-specific cross reactions with other geminiviruses, a LAMP assay was conducted with two OEGV-positive samples together with DNA extracts from other geminiviruses unrelated to OEGV; specifically, ToLCNDV (Acc. No. MK732932) [23], TYLCSV (Acc. No. GU951759) [24], TYLCV (Acc. No. DQ144621) [25], TYLCV-IS76 [26]. Each sample was analyzed in duplicate in two independent real-time LAMP assays. In each run, total DNA from a healthy olive plant (NC) was included. The assay was conducted as above described, including the melting curve steps.

#### *2.7. Set up of a Rapid Sample Preparation Method Suitable for the Real-Time LAMP Assay*

To set up a simple and inexpensive sample preparation procedure, a method that avoided DNA extraction named "membrane spot crude extract" was used. For this, 1.5 g of vegetable tissue was placed in an extraction bag and homogenized with 3 mL of extraction buffer (see above). Five µL of this extract was spotted on a 1 cm<sup>2</sup> Hybond®-N+ hybridization membrane (GE Healthcare, Chicago, IL, USA), dried at room temperature for 5 min, and placed in a 2 mL tube containing 0.5 mL of glycine buffer (0.1 M Glycine, 0.05 M NaCl, 1 mM EDTA). After 20-s vortexing, samples were heated at 95 ◦C for 10 min and 3 µL of the extract were used for the LAMP assay. Ten samples previously analyzed by end-point PCR were used in the real-time LAMP assay, including a positive control (PC) and a healthy olive plant DNA as negative control (NC).

## *2.8. Spread of OEGV in Different Cultivars*

During autumn 2021, in different Sicilian areas, a second sampling was carried out to evaluate the OEGV spread in Sicily, this time sampling 10–15-year-old olive trees, belonging to 70 different cvs. A total of 560 samples were collected. For each cv, eight different trees were sampled and grouped, obtaining a total of 70 different batches. Sampling and geo-referencing were as described above. In this case, samples were prepared with the "membrane spot crude extract" method and subjected to real-time LAMP assays in a 12 µL final volume as described above. In the case of positive sample batches, they were resampled and analyzed individually to determine the effective number of positive plants for each cultivar.

#### **3. Results**

#### *3.1. OEGV Detection by End-Point PCR*

To ascertain the presence of OEGV in Sicily, a total of 80 samples representing 10 different cvs collected from two olive production sites in the Agrigento province were analyzed by end-point PCR. Overall, 44 of them were found to be positive to OEGV, demonstrating the presence of OEGV in Sicily also. However, OEGV was not equally distributed among the cvs tested, and some cvs tested negative for this virus, at least using the primer set mentioned in this manuscript (Table 1).


**Table 1.** Prevalence and cultivar distribution of OEGV analyzed by end-point PCR.

#### *3.2. OEGV Real-Time LAMP Primer Design*

A real-time LAMP assay for the rapid detection of OEGV was developed using a set of six primers designed on the OEGV-AV1 coding region. The sequences and binding sites of the primers are reported in Table 2 and Figure 1, respectively.


Salicina Vassallo 3/8 37.5 Uovo di piccione 1/8 12.5 Vaddara 0/8 0 Zaituna Florida 0/8 0

**Total 44/80 55**

A real-time LAMP assay for the rapid detection of OEGV was developed using a set of six primers designed on the OEGV-AV1 coding region. The sequences and binding sites

#### **Table 2.** Primers used for OEGV detection by LAMP. **Primer Name Sequence (5′-3′) Amplicon Size (bp)**

of the primers are reported in Table 2 and Figure 1, respectively.

*Plants* **2022**, *11*, x FOR PEER REVIEW 6 of 15

*3.2. OEGV Real-Time LAMP Primer Design*

**Table 2.** Primers used for OEGV detection by LAMP.

MW316657).

**Figure 1.** Location of loop-mediated isothermal amplification (LAMP) primer sets designed on the AV1 coding region of OEGV. F3 and B3 are shown in green, FIP (F1c-F2) in blue, BIP (B1c-B2) in pink, and the two loop primers LF and LB in brown. FIP is a hybrid primer consisting of the F1c and the F2 sequences, while BIP is a hybrid primer consisting of the B1c and B2 sequences. The arrows indicate the extension direction. The numbers at the beginning and end of the sequence represent the genomic position of the first and last nucleotide in the selected sequence (GenBank Acc. No. **Figure 1.** Location of loop-mediated isothermal amplification (LAMP) primer sets designed on the AV1 coding region of OEGV. F3 and B3 are shown in green, FIP (F1c-F2) in blue, BIP (B1c-B2) in pink, and the two loop primers LF and LB in brown. FIP is a hybrid primer consisting of the F1c and the F2 sequences, while BIP is a hybrid primer consisting of the B1c and B2 sequences. The arrows indicate the extension direction. The numbers at the beginning and end of the sequence represent the genomic position of the first and last nucleotide in the selected sequence (GenBank Acc. No. MW316657).

Both the in silico analysis of LAMP primers using Nucleotide-BLAST algorithm and the hybridisation analysis against other geminiviruses performed with the Vector NTI 11.5 program allowed for the exclusion of relevant matches with other organisms and, Both the in silico analysis of LAMP primers using Nucleotide-BLAST algorithm and the hybridisation analysis against other geminiviruses performed with the Vector NTI 11.5 program allowed for the exclusion of relevant matches with other organisms and, more specifically, with geminiviruses known to be present in Sicily.

#### more specifically, with geminiviruses known to be present in Sicily. *3.3. OEGV Real-Time LAMP Assay Optimization*

*3.3. OEGV Real-Time LAMP Assay Optimization* To evaluate the performances of the primer set designed for the real time LAMP assay in the identification of the presence of OEGV in olive DNA extracts, the LAMP assay was conducted using a subset of the samples listed in Table 1, selecting them among those that resulted positive in end point PCR. In the assay, a sample that tested negative was also included (i.e., cv. Giarraffa), together with an appropriate negative control (NC). The To evaluate the performances of the primer set designed for the real time LAMP assay in the identification of the presence of OEGV in olive DNA extracts, the LAMP assay was conducted using a subset of the samples listed in Table 1, selecting them among those that resulted positive in end point PCR. In the assay, a sample that tested negative was also included (i.e., cv. Giarraffa), together with an appropriate negative control (NC). The assay was conducted at 65 ◦C. As reported in Table 3 and Figure 2A, the positive samples showed exponential trends between 3 to 13 min. The melting curves of the positive LAMP reactions all had the same peak temperature of approximately 85 ◦C (Figure 2B). As expected, no signal was obtained with the negative control and, according to the end point PCR results, the samples from cv. Giarraffa could not be amplified by LAMP, even at late reaction times.


**Table 3.** Performance of the real time LAMP assay for the detection of OEGV in olive samples collected in Sicily. **Table 3.** Performance of the real time LAMP assay for the detection of OEGV in olive samples collected in Sicily.

assay was conducted at 65 °C. As reported in Table 3 and Figure 2A, the positive samples showed exponential trends between 3 to 13 min. The melting curves of the positive LAMP reactions all had the same peak temperature of approximately 85 °C (Figure 2B). As expected, no signal was obtained with the negative control and, according to the end point PCR results, the samples from cv. Giarraffa could not be amplified by LAMP, even at late

*Plants* **2022**, *11*, x FOR PEER REVIEW 7 of 15

reaction times.

**Figure 2.** Results of the real time LAMP assay for the detection of OEGV. (**A**): Amplification curves of real-time LAMP assay; (**B**): Melting curves of the amplification curves previously obtained, including positive (PC) and negative control (NC).

#### *3.4. Features of Real-Time LAMP Assay: Sensitivity and Comparison to Conventional PCR, Reaction Time and Specificity*

To determine the sensitivity of the real-time LAMP assay compared to the end-point PCR and to evaluate the LAMP efficacy, a comparative experiment was conducted using as a template ten-fold serial dilutions of an amplicon obtained by end point PCR from an

OEGV-positive sample (pcr-DNA), starting from a concentration of 80.9 ng/µL. As can be observed in Table <sup>4</sup> and Figure 3, DNA concentration up to ~80.9 <sup>×</sup> <sup>10</sup>−<sup>8</sup> ng/µL was detected with the LAMP assay, while the end point PCR positive signals were obtained with DNA concentration up to ~80.9 <sup>×</sup> <sup>10</sup>−<sup>7</sup> ng/µL, indicating that real time LAMP was about ten times more sensitive than conventional PCR. as a template ten-fold serial dilutions of an amplicon obtained by end point PCR from an OEGV-positive sample (pcr-DNA), starting from a concentration of 80.9 ng/μL. As can be observed in Table 4 and Figure 3, DNA concentration up to ~80.9 × 10−<sup>8</sup> ng/μL was detected with the LAMP assay, while the end point PCR positive signals were obtained with DNA concentration up to ~80.9 × 10−<sup>7</sup> ng/μL, indicating that real time LAMP was about ten times

To determine the sensitivity of the real-time LAMP assay compared to the end-point PCR and to evaluate the LAMP efficacy, a comparative experiment was conducted using

**Figure 2.** Results of the real time LAMP assay for the detection of OEGV. (**A**): Amplification curves of real-time LAMP assay; (**B**): Melting curves of the amplification curves previously obtained,

*3.4. Features of Real-Time LAMP Assay: Sensitivity and Comparison to Conventional PCR,* 

**Table 4.** Comparison of the sensitivity of the real time LAMP and end-point PCR. more sensitive than conventional PCR.


*Plants* **2022**, *11*, x FOR PEER REVIEW 8 of 15

including positive (PC) and negative control (NC).

*Reaction Time and Specificity*

**Figure 3.** Sensitivity of the end point PCR (panel **A**) and real-time LAMP (panel **B**) for OEGV detection. The assay was conducted using 10-fold serial dilutions of pcr-DNA. Panel (**A**): Agarose gel electrophoresis of PCR products; M: 1 Kb ladder marker, NC: negative control. Panel (**B**). **Figure 3.** Sensitivity of the end point PCR (panel **A**) and real-time LAMP (panel **B**) for OEGV detection. The assay was conducted using 10-fold serial dilutions of pcr-DNA. Panel (**A**): Agarose gel electrophoresis of PCR products; M: 1 Kb ladder marker, NC: negative control. Panel (**B**). Fluorescence of the 10-fold serial dilutions analyzed. Fluorescence increased in positive sample curves (from 10−<sup>1</sup> to 10−<sup>8</sup> ) after 3 to 10 min.

Moreover, even considering the lowest detectable concentration of the pcr-DNA sample in real time LAMP (~80.9 <sup>×</sup> <sup>10</sup>−<sup>8</sup> ng/µL), the results clearly showed that the time required to carry out the experiment was less than 30 min.

In addition, to evaluate the specificity of the LAMP assay and to assess potential nonspecific cross reactions with other geminiviruses present in the agricultural areas where

olive crop samples were collected, we conducted a LAMP assay using the geminiviruses reported in paragraph 2.6 as a template. Results showed that no signals were obtained with any of the geminiviruses used as the outgroup, while the two OEGV-positive olive DNA samples used as controls reacted in real time LAMP with a time value of 10 min and a single peak at 85 ◦C in the melting curve. This allowed us to confirm the specificity of the assay and to exclude the cross-reactivity with unrelated geminiviruses previously isolated in Sicily.

#### *3.5. Set up of a Rapid Sample Preparation Method Suitable for the Real-Time LAMP Assay*

With the purpose of identify a method that allows a simple and inexpensive sample preparation useful for real time LAMP, samples prepared with the two different procedures were tested. For this, the ten samples previously analyzed by end-point PCR and by real time LAMP assay (Table 3) were considered. As reported in Table 5, all samples tested positive in the LAMP assay when extracted with either procedure. Specifically, samples extracted with the commercial kit showed a fluorescence increase ranging between 3–14 min, while the same samples prepared with the "membrane spot crude extract" method could be detected in 10–24 min. This is worthy of note, as it indicates that the rapid method allows for the detection of the presence of OEGV with a delay of only a few minutes compared to the corresponding extract obtained with the commercial kit. As expected, even with this rapid procedure, no reaction was obtained with the samples from cv. Giaraffa.


**Table 5.** Comparison of two different sample preparation methods for the identification of the presence of OEGV in olive samples.

#### *3.6. Spread of OEGV in Sicily*

To investigate the spread of OEGV in different olive cultivars grown in Sicily, a new survey was conducted testing 70 samples, each consisting of eight different trees of the same cv. These samples were extracted with the rapid extraction protocol and tested in real-time LAMP, thus representing a total of 560 olive trees analyzed overall. This analysis showed that 30 out of the 70 cultivars (~43%) were positive for OEGV, indicating a relatively high incidence and prevalence of OEGV in the sampling locations and across cultivars (Table 6). When each of the eight plant samples present in the 30 positive batches were tested individually, the majority (235 out of 240 plants) resulted as being positive for OEGV, except the batch of cv. 'Calatina', where only three plants out of eight were positive (Table 6).


**Table 6.** Incidence of OEGV evaluated using a real-time LAMP assay on samples prepared with the "membrane spot crude extract" method.


**Table 6.** *Cont.*

**Note**: +: positive sample; −: negative sample; NT: Not Tested.

#### **4. Discussion**

The olive tree is affected by many potential pathogens, including viruses. Some of them are reported to be transmitted by different vectors [9,27–29], but the use of infected propagating material might represent the major, though not the only, means of virus spread [29–33]. The first report on a probable viral disease of olive goes back to 1938 [34] and, since then, several virus-like diseases and viruses have been reported over the years in different areas where olive cultivation plays a prominent role [14,35–40]. Some of these are agents of recognized diseases, others cause latent infections with still undetermined effects on the host [29]. The discovery of OEGV adds to the list of unclassified members of the family *Geminiviridae*, whose genome sequences diverge significantly from those of classified members [16]. The identification of this new virus was facilitated by HTS, a technique that allows for the discovery of new plant viruses, especially when symptoms are not evident, as is the case of OEGV. Besides Apulia, OEGV was recently reported in different areas where the olive cultivation is widespread [15,17,18].

To our knowledge, this study represents the first report of OEGV in Sicily. Since PCR-based methods can be affected by several inhibitors [41,42], such as phenols and polysaccharides [43,44] and require nucleic acid extraction methods [45], we aimed to develop a rapid detection method for OEGV based on LAMP. Indeed, this detection technique showed optimal characteristics, providing rapid, sensitive, specific, and easy detection of several pathogens even in the field, showing a reduced sensitivity to inhibitors [42,46]. The AV1 (CP) gene of OEGV was selected as the target region for primer design and the set of the six LAMP primers showed good specificity and stability for OEGV detection. The specificity is crucial to obtain correct discrimination of OEGV from other viruses belonging to the large *Geminiviridae* family, and a high sensitivity is relevant to minimize false negatives. A LAMP assay optimization performed using DNA extracted from OEGV-infected olive samples revealed that the time required to carry out the experiment was 30 min. The LAMP assay could detect the virus presence from infected samples in as little as 3–14 min. Interestingly, the real-time LAMP developed here has proven to have a 10-fold higher sensitivity compared to the end-point PCR for detection of OEGV. Moreover, in this study, the conventional extraction method using a commercial kit was compared with a "membrane spot crude extract" method; the data obtained from this comparison suggests that the LAMP-based detection method could be suitable for direct use in the field, confirming that ease of sample preparation is a crucial requirement for future application for on-site

detection. Specifically, we demonstrated that the rapid sample preparation method allowed for the avoidance of DNA extraction and could be applied for future epidemiological studies, drastically reducing the cost of the analysis. Furthermore, this real-time LAMP technique, associated with other rapid extraction methods developed in other works [47], could be fine-tuned for an efficient and rapid in-field diagnosis.

The rapid extraction method definitely simplified the surveys of the OEGV spread in different cultivars in Sicily. The effectiveness of the techniques developed is essential to understand its spread and to refine effective methods of crop protections, in order to quickly diagnose the presence of a new pathogen in different areas [48–50].

Our survey revealed a considerable presence of the virus in the olive crops in Sicily, probably due to the inadvertent movement of clonally propagated infected but asymptomatic germplasms. Related to this, the development of diagnostic protocols for plant virus detection [51,52] and the epidemiological studies [53] of viral diseases are among the most important and useful steps towards the containment of new epidemics [54–56].

In conclusion, the real-time LAMP assay described in this work is a rapid, simple, specific and sensitive technique for detecting the presence of the recently described OEGV, allowing for the processing of a great number of samples at the same time, especially if associated with the "membrane spot crude extract" method. For this reason, we propose to adopt this method for routine tests in the laboratory and field conditions for a timely detection of OEGV. In particular, this method represents a potential tool for rapidly screening olive plant material useful for large surveys of the spread and pathogenicity of this virus, which to date remain uncertain. Interestingly, as recently reported by Ruiz-García and co-workers [18], the high level of sequence conservation encountered among all OEGV accession so far isolated requires a prompt investigation of the evolutionary and biological significance of this geminivirus in olive, opening new scenarios about its mechanisms of spread in the major olive-growing areas of the world.

**Author Contributions:** Conceptualization, S.D. and S.P.; methodology, A.G.C., S.B., S.D. and S.P.; software, A.G.C., S.B. and S.D.; validation, A.G.C., S.B., S.M., E.N., S.D. and S.P.; formal analysis, A.G.C., S.B., D.T. and A.M.; investigation, A.G.C., S.B., S.D. and S.P.; resources, A.G., S.P. and S.D.; data curation, A.G.C., S.B., S.D. and S.P.; writing—original draft preparation, A.G.C., S.B., S.D. and S.P.; writing—review and editing, A.G.C., S.B., S.M., E.N., S.D. and S.P.; visualization S.M., E.N., A.M., D.T., A.G., M.I.F.S.A. and A.A.; supervision, S.D. and S.P.; project administration, A.G., S.D. and S.P.; funding acquisition, A.G., S.D. and S.P. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** The authors thank 'PSR SICILIA 2014-2020—Programma di Sviluppo Rurale— Misura 16—Cooperazione—Sottomisura 16.1—"Sostegno per la costituzione e la gestione dei gruppi operativi del PEI in materia di produttività e sostenibilità dell'agricoltura"—Gruppo Operativo: ATS ProOlivo—Titolo del progetto: Applicazione di tecnologie "smart" per il monitoraggio, prevenzione e diagnosi precoce delle malattie di interesse economico dell'olivo' for the technical support.

**Conflicts of Interest:** The authors declare that they have no conflict of interest.

#### **References**


## *Article* **Quantitative Real-Time PCR Assay for the Detection of** *Pectobacterium parmentieri***, a Causal Agent of Potato Soft Rot**

**Anna A. Lukianova 1,2 , Peter V. Evseev <sup>1</sup> , Alexander A. Stakheev <sup>1</sup> , Irina B. Kotova <sup>2</sup> , Sergey K. Zavriev <sup>1</sup> , Alexander N. Ignatov <sup>3</sup> and Konstantin A. Miroshnikov 1,\***


**Abstract:** *Pectobacterium parmentieri* is a plant-pathogenic bacterium, recently attributed as a separate species, which infects potatoes, causing soft rot in tubers. The distribution of *P. parmentieri* seems to be global, although the bacterium tends to be accommodated to moderate climates. Fast and accurate detection systems for this pathogen are needed to study its biology and to identify latent infection in potatoes and other plant hosts. The current paper reports on the development of a specific and sensitive detection protocol based on a real-time PCR with a TaqMan probe for *P. parmentieri*, and its evaluation. In sensitivity assays, the detection threshold of this protocol was 10<sup>2</sup> cfu/mL on pure bacterial cultures and 102–10<sup>3</sup> cfu/mL on plant material. The specificity of the protocol was evaluated against *P. parmentieri* and more than 100 strains of potato-associated species of *Pectobacterium* and *Dickeya*. No cross-reaction with the non-target bacterial species, or loss of sensitivity, was observed. This specific and sensitive diagnostic tool may reveal a wider distribution and host range for *P. parmentieri* and will expand knowledge of the life cycle and environmental preferences of this pathogen.

**Keywords:** *Pectobacterium parmentieri*; qPCR; bacterial taxonomy; bacterial identification; sensitivity; soft rot; pathogen detection

## **1. Introduction**

The potato (*Solanum tuberosum*) is one of the most important crops in the world. The world market for potato production exceeds 388 million tons per year (https://www. potatopro.com/world/potato-statistics (accessed on 7 April 2021)) and per capita consumption in Russia exceeds 110 kg (https://www.potatopro.com/russian-federation/ potato-statistics (accessed on 7 April 2021)). Therefore, research related to optimising potato production, increasing yields and reducing losses associated with plant diseases and other factors is essential and urgent. Among the challenges faced by potato growers is potatoes' spoilage as a result of bacterial infections. In particular, the development of rot on tubers during storage and transportation can lead to severe losses—up to half of the harvest [1]. The leading cause of blackleg and soft rot in potatoes is the bacteria of the Pectobacteriaceae family, namely the group of Soft Rot Pectobacteriaceae (SRP), comprising phytopathogens of the genera *Pectobacterium* and *Dickeya* [2]. One of the representatives of this group is *P. parmentieri*.

*P. parmentieri* (Ppa) was first described by Khayi et al. in 2016. It is a species closely related to the previously known pathogen of Japanese horseradish, *P. wasabiae* (Pwa). Several Pwa strains, isolated from potatoes and which cause soft rot, have been scrutinised

**Citation:** Lukianova, A.A.; Evseev, P.V.; Stakheev, A.A.; Kotova, I.B.; Zavriev, S.K.; Ignatov, A.N.; Miroshnikov, K.A. Quantitative Real-Time PCR Assay for the Detection of *Pectobacterium parmentieri*, a Causal Agent of Potato Soft Rot. *Plants* **2021**, *10*, 1880. https://doi.org/10.3390/plants10091880

Academic Editor: Alessandro Vitale

Received: 30 August 2021 Accepted: 8 September 2021 Published: 10 September 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

and finally reclassified as new species [3]. Later on, Ppa was identified among potato pathogens in circulation in Europe and Russia [4,5], Africa [6], Asia [7,8] and America [9,10]. Many strains isolated from potatoes earlier, and initially attributed as Pwa or *Pectobacterium carotovorum* subsp. *Carotovorum*, were proved to represent Ppa.

Thus, *P. parmentieri* can be considered as a worldwide pathogen (https://www.cabi. org/isc/datasheet/48069201 (accessed on 17 May 2021). Strains of Ppa studied are rather diverse [11,12], and two other species related to Ppa/Pwa, *P. polonicum* [13] and *P. punjabense* [14] have been established recently. A recent study of the distribution of *P. punjabense* in Europe [15] demonstrated the need for an appropriate method for discriminatory quantitative diagnostics for newly established SRP species.

Many PCR-based methods have already been developed for generalised and speciesspecific detection of SRP (reviewed in [16,17]). The accumulation of data on bacterial genomics and taxonomic redistributions has encouraged the design of an updated method for the specific diagnosis of newly established SRP species, particularly Ppa. Earlier, PCR diagnostic methods were proposed for Pwa detection, based on the amplification of the phytase/phosphatase (appA) gene [18] or tyrosine-aspartate (YD) repeat region [19]. Both of these assays enabled scientists to discriminate Ppa/Pwa from *P. carotovorum* and other SRP, but not between the former species. The analysis currently used in phytodiagnostics enables an assumption of the approximate specificity of the pathogen, taking into account the source of the isolation of the strain [4]. However, it still does not allow for speciesspecific detection and is somewhat outdated, due to the changed understanding of the taxonomy of the group. Recently, the authors developed a pipeline for searching unique sequences for genomic groups and tested it in the context of *P. atrosepticum*, a genetically distinct species of SRP [20]. This paper describes how this workflow can be used to design a quantitative real-time PCR assay to discriminate closely related species. The aim of the study was the development of a species-specific detection system for *P. parmentieri.*

#### **2. Results**

#### *2.1. Phylogenetic Analysis*

By early 2021, more than 200 bacterial genomes deposited in NCBI GenBank represented the family *Pectobacteriaceae*. *P. parmentieri* was represented by 30 complete and high coverage draft genomes (strains CFIA102, IFB5408, IFB5427, IFB5432, IFB5441, IFB5485, IFB5486, IFB5597, IFB5604, IFB5605, IFB5619, IFB5623, IFB5626, IPO1955, NY1532B, NY1533B, NY1540A, NY1548A, NY1584A, NY1585A, NY1587A, NY1588A, NY1712A, NY1722A, PB20, QK5, RNS-08-42-1A (type strain), SS90, WC19161 and WPP163).

Whole-genome comparisons made with orthoANI (Figure 1) demonstrate that all the strains assigned to *P. parmentieri* possess a close genome similarity, demonstrating high overall average nucleotide similarity (ANI) of 98.9% and above when compared to the type species, whereas the comparable ANI values of non-*parmentieri Pectobacterium* species lies in the range of 87–94%. The concatenated core genes phylogeny also places Ppa strains in a distinct clade (Figure 1).

**Figure 1.** Phylogenetic tree based on the concatenated nucleotide sequences of 92 conservative genes, including the genes of ribosomal proteins and the proteins essential for the transcription and translation processes. Bootstrap support values are shown above their branch as a percentage of 1000 replicates. The scale bar shows 0.01 estimated substitutions per site, and the tree was rooted to *Samsonia erythrinae DSM 16730*. Average nucleotide identity (ANI) values compared to *P. parmentieri* RNA 08-42-1A type strain are shown to the right of the organism name and coloured according to a heat map scale, where a green colour corresponds to the highest value and a red colour corre-**Figure 1.** Phylogenetic tree based on the concatenated nucleotide sequences of 92 conservative genes, including the genes of ribosomal proteins and the proteins essential for the transcription and translation processes. Bootstrap support values are shown above their branch as a percentage of 1000 replicates. The scale bar shows 0.01 estimated substitutions per site, and the tree was rooted to *Samsonia erythrinae DSM 16730*. Average nucleotide identity (ANI) values compared to *P. parmentieri* RNA 08-42-1A type strain are shown to the right of the organism name and coloured according to a heat map scale, where a green colour corresponds to the highest value and a red colour corresponds to the lowest value.

#### sponds to the lowest value. *2.2. Search for Species-Specific Primers*

*2.2. Search for Species-Specific Primers*  The search for species-specific sequences was carried out using the workflow described in a previous study [20]. Briefly, this workflow splits the genome of the type Ppa strain into short sections, then each section is compared with a negative database of "nontarget" genomes and a positive database of "target genomes" and, as a result, regions are The search for species-specific sequences was carried out using the workflow described in a previous study [20]. Briefly, this workflow splits the genome of the type Ppa strain into short sections, then each section is compared with a negative database of "non-target" genomes and a positive database of "target genomes" and, as a result, regions are identified that occur in all Ppa genomes and are not found in genomes of other species.

identified that occur in all Ppa genomes and are not found in genomes of other species. Using this search, a set of unique Ppa species-specific sites was obtained. Regions belonging to the areas of the genome encoding no genes were manually rejected. Next, several potentially suitable sites within the housekeeping genes were selected for further Using this search, a set of unique Ppa species-specific sites was obtained. Regions belonging to the areas of the genome encoding no genes were manually rejected. Next, several potentially suitable sites within the housekeeping genes were selected for further preliminary testing in the conventional PCR mode (Section 2.3) and a further selection of

the most appropriate sequence for qPCR analysis development was made (Section 2.4). Primers and probes were designed for these sites. Table 1 shows the sequences of primers, probe and amplicon for detection based on the ankyrin repeat domain-containing protein sequence that showed the best results and was therefore selected for further study. the most appropriate sequence for qPCR analysis development was made (Section 2.4). Primers and probes were designed for these sites. Table 1 shows the sequences of primers, probe and amplicon for detection based on the ankyrin repeat domain-containing protein sequence that showed the best results and was therefore selected for further study.

preliminary testing in the conventional PCR mode (Section 2.3) and a further selection of

*Plants* **2021**, *10*, 1880 4 of 13

**Table 1.** Primers for amplification of a species-specific region and *P. parmentieri* and the amplicon of ankyrin repeat domain-containing protein. **Table 1.** Primers for amplification of a species-specific region and *P. parmentieri* and the amplicon of ankyrin repeat domain-containing protein.


The selected species-specific sequence belongs to an ankyrin repeat domain-containing protein that is located adjacent to the components of a type VI secretion system. Interestingly, an avirulence factor was located several genes upstream of the locus shown in Figure 2. A type VI secretion system is important for plant-associated bacteria, including the *Pectobacterium* species. It contributes to virulence and grants fitness and colonisation advantages *in planta* [21]. It might be suggested that the gene containing the species-specific sequence is important for the bacterium. The sequence search conducted with BLAST using an nr/nt database confirmed that the chosen amplicon did not have close homologues in other organisms. The selected species-specific sequence belongs to an ankyrin repeat domain-containing protein that is located adjacent to the components of a type VI secretion system. Interestingly, an avirulence factor was located several genes upstream of the locus shown in Figure 2. A type VI secretion system is important for plant-associated bacteria, including the *Pectobacterium* species. It contributes to virulence and grants fitness and colonisation advantages *in planta* [21]. It might be suggested that the gene containing the species-specific sequence is important for the bacterium. The sequence search conducted with BLAST using an nr/nt database confirmed that the chosen amplicon did not have close homologues in other organisms.


**Figure 2.** Region in the *P. parmentieri* RNS 08-42-1AT genome containing a species-specific sequence. The scheme was visualised using Geneious Prime 2021.2.2 (https://www.geneious.com, accessed on 20 January 2021). **Figure 2.** Region in the *P. parmentieri* RNS 08-42-1AT genome containing a species-specific sequence. The scheme was visualised using Geneious Prime 2021.2.2 (https://www.geneious.com, accessed on 20 January 2021).

#### *2.3. Primary Analysis by Conventional PCR 2.3. Primary Analysis by Conventional PCR*

For the initial assessment of the applicability of the primers obtained for the purpose of species-specific PCR detection, a conventional PCR test was carried out on a limited set of strains. The strains marked F… are a part of the local collection of bacterial pathogens associated with potato soft rot. The collection includes comprehensively described type strains, strains with appropriate genomic characterisation and loosely characterised local isolates. The information on the strains used is provided in Supplementary Table S1. The primary testing strain set included several representatives of different Pectobacteriaceae species belonging to the genus *Pectobacterium* (F002, F004, F012, F016, F028, F041, F043, F048, F061, F109, F126, F131, F135, F152, F157, F160, F162, F164, F171, F182, F258), *Dickeya* (F012, F077, F082, F085, F097, F101, F102, F117, F155, F261) and an unrelated pectolytic isolate (F105). In the experiment described in this paper, amplification was expected only for Ppa (F034, F035, F127, F148, F149, F174), and with none of the others. For the initial assessment of the applicability of the primers obtained for the purpose of species-specific PCR detection, a conventional PCR test was carried out on a limited setof strains. The strains marked F . . . are a part of the local collection of bacterial pathogens associated with potato soft rot. The collection includes comprehensively described type strains, strains with appropriate genomic characterisation and loosely characterised local isolates. The information on the strains used is provided in Supplementary Table S1. The primary testing strain set included several representatives of different Pectobacteriaceae species belonging to the genus *Pectobacterium* (F002, F004, F012, F016, F028, F041, F043, F048, F061, F109, F126, F131, F135, F152, F157, F160, F162, F164, F171, F182, F258), *Dickeya* (F012, F077, F082, F085, F097, F101, F102, F117, F155, F261) and an unrelated pectolytic isolate (F105). In the experiment described in this paper, amplification was expected only for Ppa (F034, F035, F127, F148, F149, F174), and with none of the others.

Figure 3 shows the results of such an analysis for the amplification of ankyrin repeat domain-containing protein, as a result of which significant amplification was demonstrated

only with the target strains (marked in the boxes) and in the absence of false-positive results with all other strains. This enabled the assumption of this site's suitability for amplification in qPCR mode, and made it possible to proceed to the validation using an extended range of strains. strated only with the target strains (marked in the boxes) and in the absence of false-positive results with all other strains. This enabled the assumption of this site's suitability for amplification in qPCR mode, and made it possible to proceed to the validation using an extended range of strains.

Figure 3 shows the results of such an analysis for the amplification of ankyrin repeat domain-containing protein, as a result of which significant amplification was demon-

*Plants* **2021**, *10*, 1880 5 of 13

**Figure 3.** The results of conventional PCR visualised in 1.5% agarose gel. The numbers of the strains belonging to Ppa are marked with a frame. The remaining strains belonging to other species acted as negative controls. The lane designated as "+ control" contained PCR results with test plasmid, which served as a positive control. Evrogen 1 kb Ladder used for the evaluation of amplicons sizes. **Figure 3.** The results of conventional PCR visualised in 1.5% agarose gel. The numbers of the strains belonging to Ppa are marked with a frame. The remaining strains belonging to other species acted as negative controls. The lane designated as "+ control" contained PCR results with test plasmid, which served as a positive control. Evrogen 1 kb Ladder used for the evaluation of amplicons sizes.

#### *2.4. qPCR Analysis on an Extended Set of Strains 2.4. qPCR Analysis on an Extended Set of Strains*

This study involved seven strains previously attributed to being Ppa or Pwa on the basis of genomic sequencing or 16S rRNA gene sequencing. Two more strains were previously identified as Pwa using the diagnostic primer set PhF 5′-GGTTCAGTGCGTCAG-GAGAG and PhR 5′-GCGGAGAGGAAGCGGTGAAG [18], which does not distinguish between Pwa and closely related Ppa (№ 1–9, Supplementary Table S1). A test was also conducted for 67 (№ 10–77) isolates of other *Pectobacteriaceae* species and 32 strains (№ 78– 109) related to other species associated with crop rot. These strains were isolated from potato rots and passed through McConkey's medium to exclude *Salmonella* and Grampositive isolates and SVP medium to ensure the presence of pectolytic activity. This study involved seven strains previously attributed to being Ppa or Pwa on the basis of genomic sequencing or 16S rRNA gene sequencing. Two more strains were previously identified as Pwa using the diagnostic primer set PhF 50 -GGTTCAGTGCGTCAGGAGAG and PhR 50 -GCGGAGAGGAAGCGGTGAAG [18], which does not distinguish between Pwa and closely related Ppa (№ 1–9, Supplementary Table S1). A test was also conducted for 67 (№ 10–77) isolates of other *Pectobacteriaceae* species and 32 strains (№ 78–109) related to other species associated with crop rot. These strains were isolated from potato rots and passed through McConkey's medium to exclude *Salmonella* and Gram-positive isolates and SVP medium to ensure the presence of pectolytic activity.

As shown in Supplementary Table S1, all Ppa strains demonstrated a positive PCR signal. Among the strains with alternative Ppa/Pwa attribution (F035 and F178), F035 showed amplification and therefore can be more accurately classified as Ppa, while F178, revealing no positive signal, may be categorised as Pwa. As shown in Supplementary Table S1, all Ppa strains demonstrated a positive PCR signal. Among the strains with alternative Ppa/Pwa attribution (F035 and F178), F035 showed amplification and therefore can be more accurately classified as Ppa, while F178, revealing no positive signal, may be categorised as Pwa.

The historical strain Pwa F007 used in the study did not show any false positive amplification. No positive results were shown for other isolates with pectolytic activity, both Pectobacteriaceae and unrelated ones. The historical strain Pwa F007 used in the study did not show any false positive amplification. No positive results were shown for other isolates with pectolytic activity, both Pectobacteriaceae and unrelated ones.

Additionally, in silico analysis using an nt-database did not presume any amplification of plant genomic DNA using the designed primers. No amplification was observed in the PCR reaction in vitro using potato DNA as a template. Thus, the authors are confident that the possibility of cross-amplification with potato DNA was excluded. Additionally, in silico analysis using an nt-database did not presume any amplification of plant genomic DNA using the designed primers. No amplification was observed in the PCR reaction in vitro using potato DNA as a template. Thus, the authors are confident that the possibility of cross-amplification with potato DNA was excluded.

#### *2.5. Sensitivity 2.5. Sensitivity*

Serially diluted plasmid and genomic DNA were used in qPCR reactions for a sensitivity test. Based on the threshold cycles (Cq) obtained for each concentration of copies in the sample (Table 2), standard curves were plotted. The resulting curves were linear (Figure 4). The correlation coefficient (R2) was 0.99 for both curves, with a slope of −3.34 and −3.33 for plasmid and genomic DNA, respectively, corresponding to a PCR efficiency of 98.9% and 99.62%. Serially diluted plasmid and genomic DNA were used in qPCR reactions for a sensitivity test. Based on the threshold cycles (Cq) obtained for each concentration of copies in the sample (Table 2), standard curves were plotted. The resulting curves were linear (Figure 4). The correlation coefficient (R<sup>2</sup> ) was 0.99 for both curves, with a slope of −3.34 and −3.33 for plasmid and genomic DNA, respectively, corresponding to a PCR efficiency of 98.9% and 99.62%.


**Table 2.** Mean Cq values for qPCR carried out on serial dilutions of genomic DNA of the *P. parmentieri* F149 and corresponding plasmid. SD is standard deviation. **Table 2.** Mean Cq values for qPCR carried out on serial dilutions of genomic DNA of the *P. parmentieri* F149 and corresponding plasmid. SD is standard deviation.

The limit of detection (LoD) was nearly 16 copies per reaction, corresponding to 4 × 102 copies/mL. Figure 5 shows the amplification curves for the sensitivity test and the good

*Plants* **2021**, *10*, 1880 6 of 13

flare-up of the probe during the reaction, even at high dilutions.

**Figure 4.** Standard curves showing the dependence of Cq on the concentration of pathogen DNA in the reaction. The curves are plotted based on the threshold cycles obtained for a series of ten-fold dilutions of the plasmid (**A**) and genomic DNA of the F149 strain (**B**). The standard deviation is shown as error bars. **Figure 4.** Standard curves showing the dependence of Cq on the concentration of pathogen DNA in the reaction. The curves are plotted based on the threshold cycles obtained for a series of ten-fold dilutions of the plasmid (**A**) and genomic DNA of the F149 strain (**B**). The standard deviation is shown as error bars.

The limit of detection (LoD) was nearly 16 copies per reaction, corresponding to 4 <sup>×</sup> <sup>10</sup><sup>2</sup> copies/mL. Figure 5 shows the amplification curves for the sensitivity test and the good flare-up of the probe during the reaction, even at high dilutions.

**Figure 5.** Amplification curves for a sensitivity test using the example of a series of dilutions of plasmid DNA. The numbers represent the dilution number. **Figure 5.** Amplification curves for a sensitivity test using the example of a series of dilutions of plasmid DNA. The numbers represent the dilution number.

#### *2.6. Assays of Plant Samples 2.6. Assays of Plant Samples*

To conduct an experiment simulating a pathogen's detection in infected plants, the tubers of the "Gala" variety were used, one of the most widespread varieties in Russia, and one which is moderately resistant to bacterial diseases. The potatoes were soaked in a 106 cfu/mL suspension of the pathogen for infection and then incubated at 28 °C until the development of soft rot symptoms. On days 3, 4 and 5, a ~100 mg piece of peel was taken from the tubers and total DNA was isolated. Then, qPCR was performed from the DNA obtained, in the same way as in the previous experiments. Control tubers were To conduct an experiment simulating a pathogen's detection in infected plants, the tubers of the "Gala" variety were used, one of the most widespread varieties in Russia, and one which is moderately resistant to bacterial diseases. The potatoes were soaked in a 10<sup>6</sup> cfu/mL suspension of the pathogen for infection and then incubated at 28 ◦C until the development of soft rot symptoms. On days 3, 4 and 5, a ~100 mg piece of peel was taken from the tubers and total DNA was isolated. Then, qPCR was performed from the DNA obtained, in the same way as in the previous experiments. Control tubers were soaked in a sterile LB medium.

soaked in a sterile LB medium. As shown in Table 3, the pathogen was successfully detected in all cases, confirming the possibility of using the analysis to assess the contaminated material. With an increase in the duration of incubation, the titre of bacteria increased proportionally. Amplification was also recorded for the control tuber, indicating a trace presence of the pathogen, which did not lead to noticeable symptoms of rotting. As shown in Table 3, the pathogen was successfully detected in all cases, confirming the possibility of using the analysis to assess the contaminated material. With an increase in the duration of incubation, the titre of bacteria increased proportionally. Amplification was also recorded for the control tuber, indicating a trace presence of the pathogen, which did not lead to noticeable symptoms of rotting.


**Table 3.** Results of qPCR carried out on material obtained from artificially infected potatoes. APC **Table 3.** Results of qPCR carried out on material obtained from artificially infected potatoes. APC permease gene of Ppa was detected using developed primers. SD is standard deviation.

#### **3. Discussion**

**3. Discussion**  According to the species definition, Ppa differs from Pwa by its ability to produce acid from melibiose, raffinose, lactose and D-galactose [3]. This feature was used to differentiate Ppa strains isolated from potato in Southern Europe [4]. However, the biochemical tests made the precise diagnostics more laborious and, thus, raised questions about the value of such fine analysis. Besides the obvious purpose of monitoring the causal agents of plant diseases, in order to develop adapted prevention actions in particular According to the species definition, Ppa differs from Pwa by its ability to produce acid from melibiose, raffinose, lactose and D-galactose [3]. This feature was used to differentiate Ppa strains isolated from potato in Southern Europe [4]. However, the biochemical tests made the precise diagnostics more laborious and, thus, raised questions about the value of such fine analysis. Besides the obvious purpose of monitoring the causal agents of plant diseases, in order to develop adapted prevention actions in particular countries, regions or climate areas, some fundamental arguments exist.

countries, regions or climate areas, some fundamental arguments exist. Information on the role of Ppa in the bacterial pathogenesis of potatoes worldwide is contradictory [22]. According to national monitoring surveys, Ppa occurrence ranges

from single, moderate cases [6] to severe breakouts [10]. While wet weather throughout the year is preferred for the development of the pathogen (https://www.cabi.org/isc/ datasheet/48069201 (accessed on 17 May 2021)), a broad range of conditions is tolerated. The aggressiveness of Ppa is also debatable. As for other SRP, their pathogenesis relies on the production and secretion of plant cell wall-degrading enzymes, which cause the typical symptoms of soft rot. Enzyme synthesis depends on suitable environmental conditions [23]. Generally, the virulence of Ppa is considered to be moderate. However, a number of studies [24,25] have demonstrated that some strains of *P. parmentieri* can cause fast and severe maceration of tubers and plants comparable with *P. atrosepticum* and *P. brasiliense*, which are considered to be the most aggressive among *Pectobacterium*. It is worth noting that the bacterial community in rotting potato tissues is very complex [26] and may include several different pathogenic species. SRP pathogens may interact antagonistically [27] or synergistically [28] with respect to one another. Therefore, the study of the impact of a particular pathogen on the development of the disease requires quantitative differential identification of the SRP species, particularly with Ppa.

Currently, no effective control agents have been developed to prevent or to treat SRP infections [29,30]. A promising approach is the use of bacteriophages (phages), which are bacterial viruses that infect pathogenic bacteria. A number of successful applications of phage control of plant pathogens, including SRP, have been reported (reviewed in [31,32]). Some phages infecting Ppa have been isolated and investigated [33,34]. An important feature of phage therapy is to have a very selective host range of bacteriophages, usually limited to a bacterial species or even a group of strains within a species. This may be considered to be an advantage, because phage treatment does not affect commensal and endosymbiotic microflora of the plant attacking pathogenic bacteria only. However, scientifically sound use of therapeutic bacteriophages requires fine and precise diagnostics of the causative agent of the disease. Existing assays are often too general for efficient phage application, and more focused methods of discriminating SRP are needed.

Besides pectolytic enzymes, a number of other proteinaceous and carbohydrate factors and signal pathways have been found to participate in bacterial adhesion, the colonising of plant tissue and enhancement of the disease (reviewed in [23]). Essential intracellular effectors have been secreted into the plant cell via secretion systems type III (T3SS), type IV (T4SS) and type VI (T6SS) [35]. An important feature of Ppa/Pwa is the absence of a number of essential genes encoding T3SS in the genome [36,37]. This absence may explain the limited host range of *P. parmentieri*. In such conditions, the role of T6SS and other secretion systems becomes more important [38]. The genomic sequence unique to Ppa that was identified was located adjacent to the T6SS apparatus, and its conservation within a species may indicate a unique role in the functioning of the system. This sequence does not belong to any known mobile elements and, thus, may serve as a hallmark of Ppa genomes.

Another important area where qPCR detection of SRP is needed is the establishment of the threshold bacterial population necessary for the development of disease symptoms. While the occurrence of SRP-related blackleg, wilting and aerial rot of vegetating potato depends on numerous environmental factors (reviewed in [39]), the development of soft rot in stored ware and seed potato is a consequence of a latent infection of the tuber surface. The incidence of soft rot, as a minimum, correlates with the population of SRP as revealed by laboratory testing. Most in vitro experiments described in the literature use an application of 106–10<sup>7</sup> cfu/mL aliquots of SRP suspensions applied to unprotected potato tissue (tuber slices) to establish the stable development of soft rot symptoms. This work reports that, starting from almost negligible values, the population of Ppa grew fast at room temperature and reached ~10<sup>6</sup> cfu/mL, resulting in tissue rotting in a few days. On the other hand, undamaged potato tubers with a latent SRP population 104–10<sup>6</sup> cfu/mL on the skin revealed no signs of soft rot being stored in proper warehouse conditions (4–7 ◦C) [40]. Therefore, the monitoring of the bacterial insemination of the tubers may help to estimate the risk of soft rot development in the stored tubers and to reveal the dangerous threshold for each particular SRP species. The designed assay has been shown to

be sensitive enough to detect Ppa within the range of natural latent infection level (102–10<sup>5</sup> cfu). Thus, this analysis is suitable for assessing the quality of potatoes and diagnosing the likely development of rot.

The reported protocol, based on the genomic analysis of an ample amount of recent GenBank data, was successfully tested and demonstrated high sensitivity and suitability for in vivo testing. The species-specific sequence revealed is not only unique to *Pectobacterium parmentieri,* but is also a part of a functional gene which can be important for pathogenic lifestyle of this economically important plant pathogen. The high specificity of the developed assay is particularly important for efficient phage application in the biocontrol of plant diseases caused by SRP bacteria.

#### **4. Materials and Methods**

#### *4.1. Phylogenetic Analysis*

Bacterial genomes were downloaded from the NCBI GenBank bacterial database ( ftp://ftp.ncbi.nlm.nih.gov/genbank (accessed on 27 March 2021)). A phylogenetic tree was generated using an UBCG pipeline, based on 92 core genes including 43 ribosomal proteins, nine genes of aminoacyl-tRNA synthetases, DNA processing and translation proteins and other conservative genes. Bootstrap analysis phylogeny was conducted by aligned concatenated sequences of 92 core genes made by UBCG with MAFFT (FFT-NSx1000, 200 PAM/k = 2). Then, bootstrap trees were constructed using the RAxML program (maximum likelihood method) (GTR Gamma I DNA substitution model). The robustness of the trees was assessed by fast bootstrapping (1000) [41].

Average nucleotide identity (ANI) was computed using orthoANI, with default settings [42].

#### *4.2. Search for Species-Specific Sequences and Primer Design*

To search for species-specific sequences, custom databases were constructed using BLAST (https://blast.ncbi.nlm.nih.gov/Blast.cgi (accessed on 25 February 2021)). The search for species-specific regions for amplification was carried out using the workflow presented in the previous study [20].

Primers and probes were generated with Primer3Plus (https://primer3.ut.ee/ (accessed on 15 March 2021)) and manually checked for the consistency of melting temperatures and for the absence of hairpins and dimers formation using the functions of Geneious Prime and Primer Biosoft (http://www.premierbiosoft.com/NetPrimer/ AnalyzePrimerServlet (accessed on 20 March 2021)).

#### *4.3. Bacterial Strains, Media and Culture Conditions*

A complete list of bacterial strains engaged in this study, with an indication of their species, year and location of isolation, is shown in Supplementary Table S1. Strains were obtained from the Laboratory of Molecular Bioengineering, IBCh RAS. Pectolytic bacteria were cultivated at 28 ◦C on 1.5% LB agar. CVP medium was used to assess pectinolytic activity. *E. coli* NovaBlue strain was used for transformation during the preparation of a plasmid. *E. coli* was cultivated at 37 ◦C on LB agar medium with the addition of ampicillin.

#### *4.4. Genomic DNA Isolation*

Genomic DNA was isolated using overnight bacterial cultures, using a GeneJET Genomic DNA Purification Kit (ThermoScientific, Waltham, MA, USA), according to the manufacturer's protocol.

Potato DNA was extracted using a CTAB-based protocol. For this purpose, a piece of peel of 100 mg was mechanically homogenised with a 0.1% sodium pyrophosphate solution. The resulting homogenate was transferred into 1.5 mL tubes and centrifuged. 40 µL of lysozyme solution (100 µL/mL) and 60 µL of 10% SDS solution were added to the sediment, resuspended and incubated at 37 ◦C for 30 min. Then, 650 µL of 2% STAB was added to the mixture and incubated for another 30 min at 65 ◦C. Then, the mixture

was cooled and 700 µL of chloroform was added, vortexed and precipitated at 12,000 rpm. The supernatant was mixed in a new tube with 600 µL of isopropanol. After subsequent centrifugation, the precipitate was washed twice with 75% ethanol and dried until the volatile solvents completely evaporated, and the resulting DNA was dissolved in water.

The concentration and quality of the extracted DNA was estimated using a NanoProteometer N60 (NanoProteometer, Munich, Germany). After extraction, DNA concentrations were diluted to a single value of 10 ng/µL.

#### *4.5. PCR Conditions*

The conventional PCR was carried out in a volume of 25 µL containing 5 µL of Evrogen ScreenMix (Evrogen, Moscow, Russia,), 0.35 µM of forward and reverse primers and 60 ng of template DNA. Amplification was performed using a T100 Thermal Cycler (Bio-Rad, Hercules, CA, USA) and in the following conditions: 94 ◦C for 300 s, then 45 cycles of 94 ◦C for 10 s, 62 ◦C for 10 s and 72 ◦C for 10 s. The resulting PCR products were separated by electrophoresis in 1.5% agarose/TA buffer gel and visualised by ethidium bromide staining. The size of the bands was eluted using a 1 kb DNA Ladder marker (Evrogen).

#### *4.6. Plasmid Construction for Sensitivity Assay*

For a precise evaluation of PCR sensitivity, we constructed a plasmid containing an insert of the target sequence amplified from the Ppa F149 strain. For this purpose, the product of PCR amplification was purified using ISOLATE II PCR and Gel Kit (Bioline, St. Petersburg, Russia) and cloned to pAL2-T vector using a QuickTA kit (Evrogen). Plasmid DNA used as standard was purified with a QIAprep Spin Miniprep Kit (Qiagen, Hilden, Germany), according to the manufacturer's instructions. Sanger sequencing of the corresponding region in the resulting plasmid confirmed the correctness of the insert.

#### *4.7. qPCR*

The qPCR was carried out in a LightCycler 96 (Roche, Basel, Switzerland). Each 35 µL reaction contained 200 µM of each dNTP, 0.2 µM of probe, 0.35 µM of forward and reverse primers and 60 ng of template DNA. The optimised amplification conditions were as listed in Section 4.5. Each reaction was carried out in four replicates. Water was used as a negative control. Plasmid-based internal control was used to exclude false-negative results, as described earlier [43].

The processing of the amplification curves obtained and the calculation of the threshold cycles were carried out using software supplied by Roche. A sensitivity analysis was carried out on serial three ten-fold dilutions of the test plasmid and genomic DNA of strain F149. The resulting samples were analysed by qPCR. For each defined threshold cycle, the mean and standard deviation were calculated using Roche software. To construct the standard curve, the threshold cycles' mean values were plotted against the concentration of copies of the target sequence in each reaction.

For all values, the standard deviation was calculated.

#### *4.8. Testing the Detection System on Artificially Infected Tubers*

For the experiment, potato tubers of the most widespread variety, "Gala", were obtained from a market. They were washed and soaked in a bacterial suspension to infect the tubers, following the same protocol as in a previous study [20]. Then, the tubers were incubated at 28 ◦C. On days three, four, five and six, DNA was extracted from 100 mg of the infected tuber's peel, as described in Section 4.4, and analysed by qPCR.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/10 .3390/plants10091880/s1, Table S1 Selectivity of the designed qPCR method.

**Author Contributions:** Conceptualisation, A.N.I. and K.A.M.; methodology, A.A.S. and S.K.Z.; investigation, A.A.L., P.V.E. and A.A.S.; software, P.V.E.; validation and formal analysis, A.A.L., A.A.S. and I.B.K.; data curation, A.A.L. and P.V.E.; writing—original draft preparation, A.A.L. and

P.V.E.; writing—review and editing, A.N.I. and K.A.M.; visualisation, A.A.L. and P.V.E.; project administration, K.A.M. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest.

## **References**

