LI Ya, XU Mei-rong, DAI Ze-han, DENG Xiao-ling

1 Agricultural College, Guangdong Ocean University, Zhanjiang 524088, P.R.China

2 Laboratory of Citrus Huanglongbing Research/Guangdong Key Laboratory of Microbial Signals and Disease Control, South China Agricultural University, Guangzhou 510642, P.R.China

Abstract Candidatus Liberibacter asiaticus (CaLas), an uncultured Gram-negative alphaproteobacterium, is the causal agent of Huanglongbing (HLB) in citrus. CaLas resides in phloem sieve tubes and has been shown to be unequally distributed in different tissues. Although HLB is a disease of citrus plants, it has been demonstrated that periwinkle can serve as an experimental host of CaLas, which can be transmitted from citrus to periwinkle via the parasitic plant dodder (Cuscuta spp.). To investigate the distribution of CaLas in various periwinkle tissues, the bacteria were transmitted from an infected periwinkle plant to healthy periwinkles by top-grafting. The movement of the inoculum and associated titer changes were observed over time in various tissues. CaLas could be detected in the leaves, main stems, and roots of infected periwinkle by conventional PCR, and in all three tissues a clear time-dependent change in CaLas titer was observed, with titer increasing soon after inoculation and then decreasing as disease symptoms became severe. The highest titer was found at 25, 35 and 35 days after inoculation in leaves, main stems and roots, respectively. The titer in leaves was much higher than in the main stems and roots at the same time point, and the spatial distribution of CaLas in the leaves, main stems and roots of infected periwinkle was uneven, similar to what has been shown in citrus. The results provide guidance for selecting the proper periwinkle tissues and sampling times for early detection of CaLas.

Keywords: Candidatus Liberibacter asiaticus, titer, distribution, Catharanthus roseus

1. Introduction

Candidatus Liberibacter asiaticus (CaLas) is one of three Liberibacter species associated with citrus Huanglongbing(HLB) or yellow shoot disease, a destructive disease first described in South China in the early 20th century (Zhao 1981). CaLas is an insect-transmitted and obligate plant pathogen that infects all cultivated citrus and related species and causes systemic disease by residing in the phloem (Da Graca 1991; Jagoueix et al. 1996; Folimonova and Achor 2010). CaLas-infected plants gradually develop symptoms including yellow shoots, leaves with blotchy mottle, small lopsided and aborted fruits, and premature defoliation,which lead to the eventual death of the entire plant (Halbert and Manjunath 2004; Bové 2006). HLB is present in most areas of citrus-growing countries including Asia, Brazil and North America (Bové 2006). Among citrus pathogens,CaLas has become the most widespread, destructive and economically important and hence has attracted attention from researchers around the world (Coletta-Filho et al. 2004;Gottwald 2010; Kumagai et al. 2013; Wang and Trivedi 2013;Xu et al. 2013).

Because CaLas cannot be cultured in artificial media,the presence and titer of CaLas in hosts has often been determined using quantitative real-time PCR (qPCR) (Wang et al. 2006; Lu et al. 2013). The uneven distribution of CaLas in citrus makes accurate qPCR detection difficult.In addition, CaLas-positive can only be detected after at least three months of graft inoculation in top-grafted citrus plants, studying HLB in citrus in vivo is time-consuming(Tatineni et al. 2008; Deng et al. 2012; Ding et al. 2015).For these reasons, periwinkle, an alternate experimental host of the HLB bacteria, is a preferred model plant (Zhang et al. 2010). CaLas can be artificially transmitted from citrus to periwinkle via the parasitic plant, dodder (Cuscuta spp.) in the laboratory, and multiplies to a high level in the phloem of the infected periwinkle (Garnier and Bové 1983;Hartung et al. 2010; Zhang et al. 2010). Furthermore, the titer of CaLas in infected periwinkle is much higher than in infected citrus. Infected periwinkles gradually develop HLB-like symptoms similar to those seen in citrus, with initial vein yellowing that progressively expands into the entire leaf, and often die within six months post-infection(Bové 2014). In this study, healthy periwinkles were infected with CaLas using top-grafting, and HLB-like symptoms in infected periwinkles were observed. The distribution and changes in concentration of CaLas in periwinkle tissues were investigated at various time points post-inoculation by conventional PCR and qPCR.

2. Materials and methods

2.1. Plant materials

Periwinkle seeds were planted in March 2014 in a screen house at the South China Agriculture University, Guangzhou City, China. In April 2014, 100 plantlets were transplanted to single pots and divided into two groups, Group I and Group II (50 plants in each group). Seven to eight weeks later, the plantlets of the experimental group, Group I, were top-grafted with scions from yellow shoots of periwinkles that were confirmed to be CaLas positive by conventional PCR (Zhang et al. 2010), and the plantlets of the negative control group, Group II, were top-grafted with scions from healthy periwinkle that were CaLas-negative. After grafting, all plants were kept for a week in a moist chamber and thereafter in a screen house under natural weather conditions.

2.2. Sampling methods and DNA extraction

Three whole periwinkle plants from the experimental group and three from the control group were taken for sampling at 20, 25, 30, 35, 40, and 45 days after inoculation (DAI).Disease symptoms of the infected periwinkle plants were recorded with photography at different DAI. Whole periwinkle plants were washed with double distilled water,and the leaves, main stems, and roots were collected. For total genomic DNA extraction, the leaf midribs, main stems and roots of each plant (tissues from each plant were processed independently) were cut into small pieces using sterilized scissors, and 0.1 g of tissue and two ceramic beads were placed into a separate 1.5-mL microfuge tube.A total of 600 μL of buffer CPL (from the EZNATMHigh Performance Plant DNA Kit, Omega Bio-Tek, Norcross, GA)and β-mercaptoethanol (final concentration 2%) were added to each tube, and then the sample was ground using the MP Fastprep-24 Homogenizer (MP Biomedicals, Santa Ana,CA). DNA extraction was performed using the EZNATMHigh Performance Plant DNA Kit (Omega Bio-Tek, Norcross, GA)according to the manufacturer’s instructions, and quantified with the Qubit?2.0 Fluorometer (Life Technologies, CA).DNA preparations from each sample were adjusted to a final concentration of 35 ng μL-1and stored at -20°C for further use.

2.3. Detection of CaLas by conventional PCR

CaLas was detected using the primer set OI1/OI2c,which is based on a partial sequence of CaLas 16S rDNA: 5′-GCGCGTATGCAATACGAGCGGCA-3′ and 5′-GCCTCGCGACTTCGCAACCCAT-3′ (Jagoueix et al.1994). The PCR reaction was performed in a total reaction volume of 25 μL with 2.5 μL of 10× DNA polymerase buffer,2.5 μL of dNTPs (2.5 mmol L-1of each dNTP), 0.5 μL each of the forward and reverse primers (10 μmol L-1), 1 μL of sample DNA (35 ng μL-1), 0.4 μL of Taq DNA polymerase(5 U μL-1, TaKaRa, Dalian, China) and 17.6 μL of double distilled H2O. Amplification was conducted in a C1000 Thermal Cycler (Bio-Rad, Hercules, CA) with an initial denaturing step of 96°C for 5 min and 35 cycles of the following conditions: 95°C for 45 s, 55°C for 60 s, and 72°C for 1 min. A final extension of 72°C for 7 min was done after the last cycle.Each run contained two negative controls (DNA extracted from CaLas-negative periwinkle and ddH2O) and one positive control (DNA extracted from CaLas-positive periwinkle). The amplified products were evaluated by electrophoresis in 1.0% agarose gels and visualized after staining with ethidium bromide in the Gel Doc? 2000 Gel Documentation System(Bio-Rad, Hercules, CA). The amplification of CaLas by conventional PCR was performed in triplicate.

2.4. Quantification of CaLas by qPCR

The qPCR was performed using BestarTMTaqMan qPCR Mastermix (DBI Bioscience Inc., Shanghai, China)according to the manufacturer’s instructions. Each qPCR reaction consisted of a total reaction volume of 20 μL with 1 μL of DNA template, 10 μL BestarTMTaqMan qPCR Master Mix, 0.4 μL of both CaLas-specific primers, HLBas(5′-TCGAGCGCGTATGCAATACG-3′, 10 μmol L-1) and HLBr (5′-GCGTTATCCCGAAAAAGGTAG-3′, 10 μmol L-1),0.4 μL HLBp probe (5′-CAGACGGGTGAGTAACGCG-3′)and 7.8 μL of double distilled water (Li et al. 2006). Each run contained two negative controls (DNA extracted from healthy periwinkle and ddH2O) and one positive control (a recombinant plasmid clone containing CaLas 16S rDNA amplified by primer set OI1/OI2c). A standard curve was generated from 10-fold serial dilutions of the OI1/OI2c plasmid with concentrations ranging from 1.4×100to 1.4×106fg μL-1. The average cycle threshold (Ct)value was determined in triplicate for each sample. All qPCR amplifications were performed with a 95°C initial denaturation step for 5 min, followed by 40 cycles at 95°C for 10 s and 60°C for 30 s in a CFX Connect Real-time System (Bio-Rad). Based on a previous report, samples were considered CaLas-negative when the Ctvalue＞36(Zhang et al. 2011).

2.5. Data analysis

The Ctvalues from qPCR were converted to concentrations of 16S rDNA (fg μL-1) according to the standard linear regression generated from the 16S rDNA plasmid clone standard curve. The final DNA concentration data were converted into genome equivalent copy number per gram of periwinkle tissue (GECN g-1) to represent CaLas titer. We define CaLas-negative as a Ctvalue＞36 (Zhang et al. 2011).

Data are expressed as mean±standard deviation (SD).When required, the data were subjected to statistical analysis by one-way analysis of variance (ANOVA)followed by Duncan’s new multiple range test using SPSS 11.0 Software (SPSS Inc., Chicago, IL, USA). Statistical significance was defined as P＜0.05.

3. Results

3.1. Symptoms observed in periwinkle infected by CaLas

In the experimental group, 50 healthy periwinkles were topgrafted with scions from yellow shoots of CaLas-infected periwinkle. Of the 50 top-grafted plantlets, 32 showed HLB-like symptoms and were considered CaLas-infected.The earliest appearance of HLB-like symptoms, yellowing of the shoots, on these infected periwinkles was at 15 DAI(data not shown). After initial yellowing of the shoots and the edges of some leaves, yellow blotchy mottle appeared on leaves (Fig. 1). At 20 DAI, only the edges of some leaves from the infected plants showed localized yellowing(Fig. 1-I1). At 25 and 30 DAI, the yellow and chlorotic area of the leaves increased compared with the observations at 20 DAI (Fig. 1-I2 and I3). At 35, 40 and 45 DAI, the entire plant showed chlorosis, and no blossoms were observed(Fig. 1-I4-I6). In contrast, the periwinkle leaves of the control group remained green, and the plants could blossom(Fig. 1-H1-H6). The roots of CaLas-infected periwinkles were also affected; there was severe stunting and collapse of the root system and less development of fibrous roots in comparison with healthy control plants (Fig. 2).

3.2. Detection of CaLas in leaves, main stems, and roots by conventional PCR

Of the 32 periwinkles with HLB-like symptoms, 18 plants were chosen for further analysis. PCR was used to detect CaLas in the leaves, main stems, and roots of these samples at various DAI. The leaves and stems of all 18 plants were CaLas positive at 20, 25, 30, 35, 40 and 45 DAI (Fig. 3-A and B). The roots of the sampled plants were also confirmed to be CaLas-positive at 25, 30, 35, 40 and 45 DAI, but at 20 DAI, only one of three replicates was CaLas-positive (Fig. 3-C). All healthy samples inoculated by CaLas-negative scions tested negative; no PCR products corresponding to CaLas were observed in any tissues at any time point (Fig. 3-A-C).

3.3. CaLas titer in the leaves, main stems and roots of infected periwinkle changes over time

We next carried out qPCR to determine the concentration of CaLas in phloem-containing tissues of the infected periwinkles: leaves, main stems, and roots. CaLas copy number was calculated by converting Ctvalues to bacterial titer based on the CaLas 16S rRNA standard curve. CaLas titer was measured in the leaves, main stems, and roots of the infected periwinkles and changed depending on sampling time point and tissue. CaLas had a peak titer at 25 and 30 DAI in leaves (1.7×106GECN g-1periwinkle leaf) (Fig. 4-A)and 35 DAI in main stems (8.5×105GECN g-1periwinkle stem) (Fig. 4-B) and roots (5.7×105GECN g-1periwinkle root)(Fig. 4-C). The minimum amounts of CaLas pathogen were observed at 20 DAI in leaves (2.4×105GECN g-1periwinkle leaf) (Fig. 4-A) and roots (9.2×104GECN g-1periwinkle root)(Fig. 4-C) and 45 DAI in main stems (1.7×105GECN g-1periwinkle stem) (Fig. 4-B). The differences between the maximum and minimum amounts of CaLas in leaves,main stems and roots were all significant (P＜0.05). The periwinkles grafted with healthy scions were negative for CaLas based on Ctvalues (data not shown).

Fig. 1 Symptoms on periwinkle leaves following Candidatus Liberibacter asiaticus (CaLas) inoculation via grafting. I1, I2, I3, I4,I5, and I6, indicate 20, 25, 30, 35, 40, and 45 days after inoculation (DAI) infected, respectively. H1, H2, H3, H4, H5, and H6,indicate 20, 25, 30, 35, 40, and 45 DAI healthy, respectively.

Fig. 2 Comparison of the root growth and the height of infected and healthy periwinkles during advanced infection. A, the roots of an infected periwinkle (left) and a healthy periwinkle (right).B, the height of an infected periwinkle (left) and a healthy periwinkle (right).

3.4. The spatial distribution of CaLas titer in leaves,main stems and roots of infected periwinkles

CaLas titer in the leaves, main stems and roots of periwinkles at the same time point were compared to determine the distribution of CaLas in vivo (Fig. 5). At 20 DAI, the titers in leaves and main stems were significantly higher than in roots (P＜0.05). At 25 and 30 DAI, the titers were significantly higher in the leaves than in main stems and roots (P＜0.01). At 35 DAI, the titers were only slightly but significantly higher in leaves and main stems when compared with the titers measured in roots (P＜0.05). At 40 DAI, the highest and lowest titers were observed in leaves and roots, respectively, and there was a significant difference in titer between leaves, main stems and roots (P＜0.05). At 45 DAI, the titers in leaf were significantly higher than in the main stems and roots (P＜0.05).

4. Discussion

Fig. 3 PCR detection of the Huanglongbing pathogen Candidatus Liberibacter asiaticus (CaLas) in different tissues of infected periwinkle at different time points. A, leaves. B, main stems. C, roots. M, DNA 2000 marker; P, positive control; N, negative control;W, water control; DAI, days after inoculation; lanes I1-I3, products from infected plants; lanes H1-H3, products from healthy plants.

Periwinkle, an alternate experimental host of CaLas, not only grows quickly but is also efficiently colonized by CaLas(Zhang et al. 2010). The distribution and concentration of CaLas in periwinkle were observed over a 45-day period after graft inoculation. Bacteria were found in the phloemcontaining tissues of CaLas-infected periwinkle, including leaves, main stems, and roots, and the distribution of CaLas in these tissues was unequal. This distribution pattern is similar to that observed in CaLas-infected citrus (Kawabe et al. 2006; Hilf and Lewis 2016). Therefore, infected periwinkle is a convenient model system for studying HLB and understanding the pathogenic mechanism of CaLas,the host plant response to infection, and the interactions between pathogen and host (Kim et al. 2009; Trivedi et al.2010; Yan et al. 2013).

Fig. 4 Comparison of Candidatus Liberibacter asiaticus(CaLas) titer in leaves (A), main stems (B) and roots (C) of infected periwinkles at different days after inoculation (DAI).GECN, genome equivalent copy number. Different letters above the bars indicate significant differences (P＜0.05). Data are mean±SD.

After CaLas inoculation, the first yellowing symptoms in infected periwinkles appeared in the shoot below the graft insertion and then gradually appeared on most leaves.Disease severity increased as time progressed. The location and spread of the disease phenotype suggest the systemic movement of CaLas from the site of initial infection to the other parts of the plant. This result is consistent with observations in citrus (Ding et al. 2015). In the early stages of infection, from 20 to 35 DAI, CaLas titer in periwinkle increased by approximately 7.0-, 3.0- and 6.5-fold in the leaves, main stems and roots, respectively. In the advanced stage of infection, some leaf symptoms, such as chlorosis,were similar to those caused by nutritional deficiencies (Fan et al. 2010). Stunting and the collapse of the root system in the infected plants was also observed. After 35 DAI, CaLas titer decreased and remained low in all tissues of the host plants (Fig. 4). This trend of increasing disease severity with an initial increase in titer followed by a decrease in CaLas titer may be attributed nutritional availability and bacterial movement within the phloem. It is possible that initially healthy periwinkles provide sufficient nutrition for the colonization and replication of CaLas at early stages postinoculation, and then the formation of bacterial aggregates by increasing concentrations of CaLas later blocks the movement of nutrients inside the phloem and thus enhances the severity of the symptoms (Huber and Haneklaus 2007).It also is possible that CaLas might secrete virulence factors or toxins into the phloem, thus affecting the host response and causing the HLB-like symptoms in periwinkle (Duan et al. 2009; Folimonova and Achor 2010).

Fig. 5 Comparison of Candidatus Liberibacter asiaticus (CaLas)titer in leaves, main stems and roots of infected periwinkles at the same days after inoculation (DAI). GECN, genome equivalent copy number. Different letters above the bars indicate significant differences (P＜0.05). Data are mean±SD.

We used PCR to determine the titer of CaLas in the leaves, main stems and roots of infected periwinkle.Compared with the main stems and roots, relatively higher CaLas titers were detected in the leaves at the same DAI.A previous report indicated that the nutrition provided in citrus leaves made them more suitable for the growth of CaLas than other tissues (Cevallos-Cevallos et al. 2009;Gottwald et al. 2012). It has also been found that phloemlimited bacteria, such as CaLas, move freely through the sieve pores along with nutrients, which flow from the leaves to the sugar consuming plant tissues (Christensen et al.2004; Jiang et al. 2004). Therefore, leaves have become the target tissue for detecting CaLas at the early stages of infection (Jiang et al. 2004). CaLas has been detected in the roots of citrus, and this can explain why many symptomless trees pruned to the stump level develop HLB symptoms in new growth flushes (Lopes et al. 2007). We also detected CaLas in periwinkle roots, but not in all plants during the early stages of infection. Our failure to detect CaLas from two root samples of infected periwinkles at 20 DAI might be due to the samples being free from the HLB pathogen or having a bacterial population density that was too low to detect. These results suggest that time is required for the diffusion of CaLas from the graft insertion to the root and that the direction of diffusion tends to be from the top to the bottom of the plant when grafting is done at the top of the plant. The results of studies in citrus also indicate that CaLas can move from the site of infection to different parts of the plant and that the pathogen is transferred systemically within the continuous sieve tube system (Kawabe et al.2006; Trivedi et al. 2010; Hilf and Lewis 2016).

5. Conclusion

We observed the symptoms of periwinkles infected by CaLas and studied the distribution and changes in titer of CaLas in leaves, main stems, and roots. Our findings are consistent with the movement of CaLas in periwinkle phloem and provide experimental evidence for the uneven distribution of CaLas in different tissues periwinkle plants,which was also observed in citrus. The results described here provide guidance for selecting proper plant tissues and sampling times for the early detection of CaLas when using periwinkle as a model system to study HLB.

Acknowledgements

This work was supported by the earmarked fund for China Agriculture Research System (CARS-27) and the Special Fund for Agro-Scientific Research in the Public Interest,China (2010003067). We thank Mr. Melissa Doud,Agricultural Research Service, United States Department of Agriculture, for her editorial corrections and manuscript suggestions.