Preliminary Report on the Acquisition, Persistence, and Potential Transmission of Citrus tristeza virus by Diaphorina citri

Simple Summary Citrus tristeza virus (CTV) is the causal agent of one of the most serious diseases of citrus and is described to be vectored by several aphid species. There have been no published reports of either acquisition or transmission of CTV by other insects, including phloem-feeding sternorrhynchans. The Asian citrus psyllid Diaphorina citri is an economically important pest since it is the vector of the bacterium associated with Huanglongbing (HLB) in citrus crops. We hereby reported the detection of CTV from field-collected D. citri and estimated the ability of these insects to acquire and transmit the virus. Under controlled conditions, D. citri nymphs were shown to acquire CTV from citrus trees, and the virus persisted in the psyllids for over 15 days. Controlled experiments also suggest that D. citri transmit CTV to healthy citrus plants but not to orange jasmine plants, a favorite host of D. citri. The results indicate D. citri is a potential vector of pathogens for two major citrus diseases: HLB and Citrus tristeza. Abstract Citrus tristeza virus (CTV) is one of the most important citrus tree viruses: a graft-transmissible virus that can be vectored by several aphid species. Diaphorina citri is the insect vector of “Candidatus Liberibacter spp.”, a bacterium associated with citrus Huanglongbing (HLB). However, no detailed description of the relationship between CTV and D. citri has been reported. In this study, D. citri adults collected from CTV-infected “Shatangju” mandarin, “Newhall” sweet orange, and “fingered citron” trees in different orchards yielded CTV-positive rates of 40%, 65%, and 95%, respectively, upon detection by conventional PCR. Illumina HiSeq sequencing followed by de novo assembly recovered the primary full CTV genome from the RNA of 30 D. citri adults sampled from CTV-positive citrus plants. Molting and adult emergence did not affect the presence or titers of CTV within the D. citri; however, the persistence of CTV in psyllids varied among different host plant species. Groups of 10 D. citri (from a population 85% CTV-positive) were shown to potentially transmit CTV to two citrus species, “Shatangju” mandarin and “Eureka” lemon, yielding 58.33% and 83.33% CTV-positive plants, respectively. No transmission of CTV to orange jasmine plants occurred. Thus, this study reports on the ability of D. citri to acquire and transmit CTV, making D. citri as a vector of two important citrus pathogens, warranting further attention and investigation.

The Asian citrus psyllid (ACP), Diaphorina citri Kuwayama (Hemiptera: Sternorrhyncha: Liviidae), is considered to be one of the most important global pests of citrus because it transmits "Candidatus Liberibacter spp.", uncultured alpha-proteobacteria associated with the citrus Huanglongbing (HLB) [25,26]. This psyllid was first recorded as a vector of HLB in Asia in 1967 and was demonstrated to transmit "Ca. Liberibacter asiaticus" (CLas) in China in 1977 [27][28][29][30]. CLas is transmitted by D. citri in a persistent-propagative manner [31][32][33]. Globally, HLB is a devastating disease of citrus crops that is managed chiefly through quarantine, production of pathogen-free nursery stocks, removal of infected trees from orchards, and control of D. citri populations [26,34]. Diaphorina citri has spread from South Asia to West Asia, Southeast Asia, parts of Australasia, and the Americas [35]. This psyllid is a phloem-feeding insect associated with the genus Citrus and some other Rutaceae, and its favorite plant host of D. citri is orange jasmine [36,37]. Nouri et al. performed a viral metagenome analysis of sequenced small RNAs and transcriptomes coupled with bioinformatics and found that virus-like sequences within D. citri were highly diverse but reported no known plant viral pathogens [38]. In a previous study, CTV was used as a vector carrying sequences for D. citri genes in RNAi research [39]. Given that D. citri is considered an important pest of citrus orchards, we wondered whether this psyllid would be able to carry other phytopathogens in addition to "Candidatus Liberibacter spp.".
Reanalysis of raw data in RNAseq samples of D. citri from our previous study [40] had recovered some fragments of the CTV genome, which prompted the present investigation. We wondered whether HLB and citrus tristeza disease could be transmitted by the same vector. Therefore, the main objective of this study was to investigate the possibility that D. citri may carry, persist with, and ultimately transmit CTV under controlled conditions.

CTV Orchard Surveys
The presence of CTV in mature trees and D. citri populations was surveyed in three orchards, two in Qingyuan (23. (Table  S1). All of the investigated orchards had over 2000 citrus plants. Local pest management among these orchards was 8-12 times per year. Twenty D. citri adults from different trees and 20 citrus trees were sampled from each orchard. Total RNA was extracted from the samples and assayed for the presence of CTV by conventional PCR.
After confirming that citrus trees and psyllids from all of the three orchards were CTV-positive, 60 4th-instar psyllids were collected from each of the orchards and kept on the original detached CTV-positive young shoots inside controlled environment chambers (with shoot inserted into moistened centrifuge tubes, inside Falcon ® tubes modified to accommodate the centrifuge tubes). To check whether CTV would persist through molts, 20 late-stage 4th instars (i.e., bound to molt in 24 h) from each of three chambers were collected, ten were used for RNA extraction, and the other ten were transferred to detached CTV-negative orange jasmine young shoots until the 4th instars had molted to the 5th instar. Newly molted 5th instars were collected, and RNA from each nymph was extracted. The RNA samples were reverse transcribed to cDNA and then used for CTV detection by reverse transcription real-time quantitative PCR (RT-qPCR). With a similar goal, the same procedure as above was employed by transferring 20 late-stage 5th instars to CTV-negative orange jasmine and testing ten late-stage 5th instars and ten newly emerged adults.

Plant and Insect Materials under Laboratory Conditions
As shown in Table S1, plants comprising "Shatangju" mandarin (Citrus reticulata Blanco "Shatangju") scions grafted on trifoliate orange, "Eureka" lemon (Citrus × limon var. limon (L.) Burm. f.) grafted on rough lemon (Citrus jambhiri Lush), and orange jasmine (Murraya paniculata (L.) Jack) seedings were maintained in a screenhouse at South China Agricultural University (SCAU; 23.16 • N, 113.36 • E), Guangzhou, China. The two citrus species were reported in previous studies to have different tolerance to infection by CTV, and orange jasmine was reported resistant to CTV [41][42][43]. All plants were 2 to 3 years old with an average height of around 50 cm. All plants were confirmed CTV-negative by PCR and ELISA, as described further below. Concomitantly, buds of CTV-positive "Shatangju" mandarin trees from Zhaoqing (23.75 • N, 112.24 • E), Guangdong, were grafted onto CTV-negative "Shatangju" mandarin plants maintained in a separate screenhouse, and were used to rear CTV-positive psyllids (Table S1).
Adult D. citri collected from orange jasmine plants at SCAU campus were used to establish psyllid colonies on orange jasmine plants in a temperature-regulated greenhouse (25 ± 1 • C, 60% ± 2% relative humidity (RH)). The psyllid population had been reared for over 10 generations. Ten D. citri adults from the orange jasmine plants were tested individually by RT-qPCR every two weeks to ensure that psyllids colonies remained CTV-negative throughout the experiment.

Acquisition and Persistence of CTV in D. citri
The experimental design of this phase was based on the outcome of CTV persistence assessments of nymphs after molting and emergence. About 700 CTV-negative 3rd-instar D. citri were collected from 20 CTV-free orange jasmine plants, maintained as above. The nymphs were divided into 10 groups of 60 individuals. Each group of 60 was placed on one of 10 CTV-positive "Shatangju" mandarin plants ( Figure 1A). One week after the emergence of adult D. citri (after an approximately 15-day-acquisition access period (AAP), when most nymphs had passed through 4th and 5th instars to become adults), ten adults from each tree were collected and tested for CTV titers individually. As shown in Figure 1B, the remaining D. citri adults from the 10 CTV-positive "Shatangju" mandarin plants were collected, randomly subdivided into groups of 10, and then each group was transferred to each of 20 CTV-negative "Shatangju" mandarin plants and to each of 16 CTV-negative orange jasmine plants for 5-or 15-day inoculation access periods (IAP).
After the 5-day IAP, all D. citri were collected from 12 of the 20 "Shatangju" mandarin and 8 of the 16 orange jasmine plants. After the 15-day IAP, all psyllids were collected from the remaining 8 "Shatangju" mandarin and 8 orange jasmine plants ( Figure 1B). Adults of D. citri from three CTV-negative "Shatangju" mandarin or three CTV-negative orange jasmine plants were used as negative controls ( Figure 1A). RNA was extracted from all collected D. citri adults individually and reverse transcribed into cDNA.
The CTV infection status of the initially CTV-negative "Shatangju" mandarin and orange jasmine plants were assessed using qPCR immediately after the 5-d or 15-d IAP.
In addition, the "Shatangju" mandarin trees and orange jasmine plants exposed to CTVpositive D. citri for 5 days were also used for the next experiment about transmission. After the 5-day IAP, all D. citri were collected from 12 of the 20 "Shatangju" mandarin and 8 of the 16 orange jasmine plants. After the 15-day IAP, all psyllids were collected from the remaining 8 "Shatangju" mandarin and 8 orange jasmine plants ( Figure 1B). Adults of D. citri from three CTV-negative "Shatangju" mandarin or three CTV-negative orange jasmine plants were used as negative controls ( Figure 1A). RNA was extracted from all collected D. citri adults individually and reverse transcribed into cDNA.
The CTV infection status of the initially CTV-negative "Shatangju" mandarin and orange jasmine plants were assessed using qPCR immediately after the 5-d or 15-d IAP.
In addition, the "Shatangju" mandarin trees and orange jasmine plants exposed to CTVpositive D. citri for 5 days were also used for the next experiment about transmission.  The preliminary studies about the transmission capacity of CTV by D. citri were described under laboratory conditions using three species of the Rutaceae family with different tolerance to CTV. After a 15-day AAP, ten D. citri adults were transferred to a single shoot in each of 12 "Shatangju" mandarin trees, six "Eureka" lemon trees, and eight orange jasmine plants for a 5-day IAP. All 26 plants were tested CTV-negative by both qPCR and ELISA as described below. The psyllids were enclosed in meshed plastic gauze bags (250 mm(L) × 180 mm(W)) without touching the leaves. After the IAP, all psyllids and eggs together with their enclosing bags were removed. All plants were maintained in plant growth chambers (Jiangnan Instrument Company, RXZ-380A, internal dimensions 50 cm L × 50 cm W × 100 cm H), with three trees in each chamber, at 25 ± 1 • C, 65% ± 2% RH, 14:10 h L:D, 8000 lx ( Figure 1C) during the IAP but were moved to the screenhouse after IAP. Leaves of approximately the same age were selected immediately following the IAP and at 15, 30, 45, 60, 75, 90, 120, and 210 days after IAP. Three leaves (=replicates) from each of three shoots, with one of them fed by the psyllids, from each plant were sampled. RNA was extracted from the three-leaf midribs individually. Any samples that tested positive by conventional PCR were further confirmed by RT-qPCR and ELISA. CTV-negative "Shatangju" mandarin, "Eureka" lemon plants, and orange jasmine plants, after exposure to CTV-negative D. citri adults, were used as negative controls for CTV detection ( Figure 1C). Leaf and stem symptoms of CTV-positive plants were checked monthly (30 days) for 210 days. The barks from three branches of each tree were peeled to check the stem pitting status at 210 days after IAP. At the end of the experiment, crown diameter and height of CTV-positive and CTV-negative mandarin and lemon plants were measured with a measuring tape.

RNA Extraction and cDNA Synthesis
The RNA from each D. citri was extracted individually using TRlzol ® Reagent (Life Technologies, Guangzhou, China). RNA from the midribs of either citrus leaves or orange jasmine leaves was isolated using the TransZol Plant kit (TransGen Biotech, Beijing, China). Genomic DNA contamination was eliminated by digestion with RNase-free DNase I (TaKaRa, Shuzo, Kyoto, Japan). Total RNA concentration was estimated with fluorescencebased Qubit™ (Shanghai, China). The purity of total RNA was determined by absorbance using NanoDrop™ One (Thermo Scientific, Shanghai, China), where RNA samples with OD260/OD230 2.0-2.4 and OD260/OD280 of 1.8−2.2 were further reverse transcripted to cDNA. cDNA was synthesized from 500 ng of total RNA using Verso cDNA Synthesis (TransScript) kit (TransGen Biotech, Beijing, China) with random hexamers primers following the manufacturer's instructions. The obtained cDNA samples were used as templates for PCR or qPCR amplification, as described below.

CTV Detection
Four methods were used for CTV detection to ensure genome integrity and reliability of our results. The methods were: conventional PCR (using conserved regions of the CTV genome for rapid detection while avoiding false negatives), RT-qPCR (enabling absolute quantification of CTV because of greater sensitivity and calculability [19,44]), CTV genome detection (ensuring genome integrity), and ELISA (detection of CTV viral protein coat). The results of these tests were standards for labeling samples as CTV-"positive" or "negative", elsewhere in this manuscript. Each method is detailed below.

Reverse Transcription Real-Time Quantitative PCR (RT-qPCR)
For absolute quantification using Taqman™ qPCR, CTV was detected using the primer set (qP20-f: 5 -TAT AGA GGC GAA ACT GCG AAT-3 / qP20-r: 5 -CCT CAT AAC GAA GAA GCC CA-3 / qP20-p: CGG TTT ACT CGC GAC AAG CTG CTC) (size of PCR product = 109 bp) based on the p20 gene [19]. RT-qPCR was performed using a Bestar ® DBI Taqman PCR Reagent Kit (DBI Bioscience, Shanghai, China) according to the manufacturer's instructions. To calculate CTV copy number, the conventional PCR product of CTV using the primer set cquctv9/cquctv10 (size of PCR product = 424 bp) [19] was identified by 10 g·L −1 agarose gel electrophoresis and purified by DNA Gel extraction kit (TransGen, China). The purified PCR product was inserted into a pEASY-T1 vector (TransGen Biotech, China) to build a recombinant cloning plasmid. After Sanger sequencing, a recombinant plasmid containing cloned CTV cDNA was used as a template to construct standard curves. The copy number of each sample was calculated by the Bio-Rad CFX Manager Software v1.6 (Bio-Rad, Hercules, CA, USA). After qPCR, cDNA samples that yield Ct values ranging from 25.52 to 38.48 were confirmed using Uni-f/r primer through conventional PCR. Visual evaluation of bands obtained by conventional PCR in agarose gel indicated those cDNA samples with Ct ≤ 35, which were considered CTV-positive (i.e., producing a clear band in agarose gel; see Figure S1) and used for subsequent analyses. Samples with Ct values ≥ 37 not generating bands in PCR gels were considered CTV-negative. To avoid errors, samples with intermediate Ct values within 35 to 37 without a clear specific PCR band were not used in further analyses ( Figure S1, Table S2). To decrease the concentration difference of cDNA among different samples, CTV titers were assessed using the following formula referenced to the study of Ruiz-Ruiz et al. [46] in CTV detection: (Copy number of CTV p20 gene) / Titer of D. citri or citrus cDNA per µL.

CTV Genome Detection
Adults of D. citri were collected from the CTV-positive "Shatangju" mandarin plants. RNA from D. citri were extracted individually. Based on the qPCR-derived calculation of CTV titers in these cDNA samples, a pool of 30 D. citri RNA samples yielded RT-qPCRderived Ct values between 20 and 22 (i.e., relatively high CTV titers) was sequenced using Illumina HiSeq (Illumina, San Diego, CA, USA). The CTV sequences from D. citri were reassembled by mapping the obtained reads using Bowtie2 v.2.2.6 [47] and CLC Genomics Workbench v.11 (CLC Bio, Denmark) onto the reference CTV isolate genome DQ151548 (19,252 nt) from Chongqing, China. Gaps in the generated draft genome were filled in using de novo assembly with Velvet v.1.2.10 [48], and by recovering any missing sections using cDNA from five D. citri individuals as source material for conventional PCR amplification followed by Sanger sequencing of amplicons. Gene prediction within the CTV genome was performed with ORF Finder from NCBI (National Center for Biotechnology Information), RAST server (http://rast.nmpdr.org/, accessed on 12 August 2019) [49], and from references to published CTV genomes. The obtained genome was analyzed with BLAST in NCBI.

Enzyme-Linked Immunosorbent Assays (ELISA)
To further confirm the presence of CTV in citrus and orange jasmine plants, prior to and after D. citri transmission, we used ELISA to detect the presence of CTV capsid protein [18,50] via a direct triple antibody sandwich protocol (TAS-ELISA). The capture antibody (Catalog No: CAB 78900; Lot No: 00805) specific to the CTV capsid protein was purchased from Agdia Company (Elkhart, Indiana, USA). ELISA sample extracts were prepared in General Extract Buffer (tissue:buffer = 1:10) from three young leaf midribs (each recorded as a separate repetition) from three different shoots, respectively. The specific capture antibody of CTV was diluted with carbonated coating buffer (1.59 g sodium carbonate and 2.93 g sodium bicarbonate 2.93 g in 800 mL water, adjusted to 1 L with water) at a ratio of 1:100. The CTV detection antibody (A: 00832) and Alk phos enzyme conjugate (B: 00815) were mixed with ECI buffer (add 2 g bovine serum albumin and 20 g polyvinylpyrrolidone MW 24-40,000 to 1000 mL PBST) as suggested by manufacturer's instructions. PNP substrate solution was used after the detection step. CTV-negative and positive samples can usually be visually discerned based on intrinsic differences in the color intensity in ELISA, detectable by absorbance with a microplate reader (OD 405 ). Any samples with an OD value higher than 2x the average OD of healthy samples were considered "positive" [51].

Genetic Characterization of CTV in D. citri
Next-generation sequencing (NGS) of CTV-positive D. citri returned 1.47 × 10 8 100-bp reads. Mapping reads to the reference CTV GenBank isolate DQ151548 (19,252 nt) yielded an average coverage of 16.58× (range from 0.00× to 37.00×) from 1973 mapped reads (0.005% from the total 3.87 × 10 7 reads). The total consensus length was 17,904 nt, covering 93.24% of the reference genome but leaving five gaps (Figure 2), all within the CTV polyprotein gene. Thus, the genome could not be fully reconstructed by de novo assembly and was completed using sequences from five PCR amplicons (Figure 2). Single nucleotide polymorphisms (SNPs) (from 0.58% to 3.61%) were found within the gaps of the different individual D. citri RNA samples across five genome regions. Based on this sequence data, the genome was estimated to be 19,252 nt long and was named "CTV isolate FN08" (GenBank accession number of MK491895). The CTV-FN08 genome encoded the 11 ORFS observed in other published CTV genomes, indicating the genome was complete. Online GenBank BLASTn searches indicate CTV-FN08 genome is most similar to CTV isolates T318A (DQ151548), CT11A (JQ911664.1), and SG29 (KC748392), with measured identities of 98.50%, 97.97%, and 97.75%, respectively. the different individual D. citri RNA samples across five genome regions. Based on this sequence data, the genome was estimated to be 19,252 nt long and was named "CTV isolate FN08" (GenBank accession number of MK491895). The CTV-FN08 genome encoded the 11 ORFS observed in other published CTV genomes, indicating the genome was complete. Online GenBank BLASTn searches indicate CTV-FN08 genome is most similar to CTV isolates T318A (DQ151548), CT11A (JQ911664.1), and SG29 (KC748392), with measured identities of 98.50%, 97.97%, and 97.75%, respectively. , and the arrows and letters represent Sanger-sequenced amplicons from 5 PCR primer sets (P1-P5) used to span and bridge the genome, and one primer set (P6) used to suggest a linear structure for CTV genome. "Reads" area span forward reading frames (in red) and reversed reading frames (in green). Figure 3 illustrates the acquisition efficiency of CTV by D. citri and the persistence of the virus inside adults after they were transferred from CTV-positive "Shatangju" mandarin plants to CTV-negative "Shatangju" mandarin plants or orange jasmine plants. The mean percentage of CTV-positive D. citri adults was 85.00% ± 1.23% (85/100) at the time psyllids were transferred to healthy plants (day 0). The RT-qPCR indicated that 72.50% ± 1.01% and 65.00% ± 2.59% of D. citri remained CTV-positive following 5 and 15 days of post-AAP, respectively, on "Shatangju" mandarin plants, but these values taken between two detection times were not significantly different. In contrast, the incidence of CTVpositive adults and their estimated CTV titers decreased significantly after 5 (32.5% ± 7.01% and 1158.78 ± 165.12) and 15 days (5.00% ± 2.67% and 161.53 ± 43.99) following the post-AAP, once they were placed on CTV-negative orange jasmine plants (Figure 3). , and the arrows and letters represent Sanger-sequenced amplicons from 5 PCR primer sets (P1-P5) used to span and bridge the genome, and one primer set (P6) used to suggest a linear structure for CTV genome. "Reads" area span forward reading frames (in red) and reversed reading frames (in green). Figure 3 illustrates the acquisition efficiency of CTV by D. citri and the persistence of the virus inside adults after they were transferred from CTV-positive "Shatangju" mandarin plants to CTV-negative "Shatangju" mandarin plants or orange jasmine plants. The mean percentage of CTV-positive D. citri adults was 85.00% ± 1.23% (85/100) at the time psyllids were transferred to healthy plants (day 0). The RT-qPCR indicated that 72.50% ± 1.01% and 65.00% ± 2.59% of D. citri remained CTV-positive following 5 and 15 days of post-AAP, respectively, on "Shatangju" mandarin plants, but these values taken between two detection times were not significantly different. In contrast, the incidence of CTV-positive adults and their estimated CTV titers decreased significantly after 5 (32.5% ± 7.01% and 1158.78 ± 165.12) and 15 days (5.00% ± 2.67% and 161.53 ± 43.99) following the post-AAP, once they were placed on CTV-negative orange jasmine plants (Figure 3). Figure 4A shows the percentages of "Shatangju" mandarin, "Eureka" lemon, and orange jasmine plants detected to be CTV-positive by RT-qPCR and ELISA results after 5 days of exposure to batches of 10 CTV-positive adult D. citri. Thirty and 90 days after the 5-day IAP exposure to CTV-positive D. citri, 33.33% (4/12) and 58.33% (7/12), respectively, of the "Shatangju" mandarin plants were CTV-positive. The virus was not detected in "Eureka" lemon plants 30 days after the IAP. However, 33.33% (2/6) and 83.33% (5/6) of the plants were CTV-positive at 45 and 90 days, respectively, after IAP ( Figure 4A). Infection by CTV in "Shatangju" mandarin and "Eureka" lemons was detected by RT-qPCR 210 days after IAP and further confirmed by conventional RT-qPCR and ELISA (Table S3). After 5 days IAP of CTV-positive D. citri, the eight orange jasmine plants remained CTV-negative for at least 210 days. The virus was not detected in control plants ("Shatangju" mandarin, "Eureka" lemon, and orange jasmine plants) after they were exposed to CTV-negative D. citri adults for 5 days.  Figure 4A shows the percentages of "Shatangju" mandarin, "Eureka" lemon, and orange jasmine plants detected to be CTV-positive by RT-qPCR and ELISA results after 5 days of exposure to batches of 10 CTV-positive adult D. citri. Thirty and 90 days after the 5-day IAP exposure to CTV-positive D. citri, 33.33% (4/12) and 58.33% (7/12), respectively, of the "Shatangju" mandarin plants were CTV-positive. The virus was not detected in "Eureka" lemon plants 30 days after the IAP. However, 33.33% (2/6) and 83.33% (5/6) of the plants were CTV-positive at 45 and 90 days, respectively, after IAP ( Figure 4A). Infection by CTV in "Shatangju" mandarin and "Eureka" lemons was detected by RT-qPCR 210 days after IAP and further confirmed by conventional RT-qPCR and ELISA (Table S3). After 5 days IAP of CTV-positive D. citri, the eight orange jasmine plants remained CTVnegative for at least 210 days. The virus was not detected in control plants ("Shatangju" mandarin, "Eureka" lemon, and orange jasmine plants) after they were exposed to CTVnegative D. citri adults for 5 days. The results from the RT-qPCR indicated that the CTV titers increased significantly over time in the "Shatangju" mandarin plants (F = 7.4590; df = 6; p = 0.0001) and "Eureka" lemon plants (F = 4.3090; df = 5; p = 0.0094) after 5 days of IAP exposure to CTV-positive D. citri, and the CTV titers of CTV-positive "Shatangju" mandarin plants were significantly higher than those of the CTV-positive "Eureka" lemons ( Figure 4B).

Transmission of CTV by D. citri to Mandarin, Lemon, and Orange Jasmine Plants
The symptoms observed in crown diameter, plant height, leaf, and stem morphology did not differ significantly between CTV-positive and CTV-negative "Shatangju" mandarin plants. However, CTV-positive "Eureka" lemon plants had significantly smaller crown diameters compared to CTV-negative counterparts (35.17 ± 2.08 cm vs. 50.80 ± 4.18 cm (p = 0.0119), respectively) and were significantly shorter (73.90 ± 1.84 cm vs. 90.13 ± 4.33 cm (p = 0.0119), respectively) ( Figure S3A). Compared to the negative controls of "Eureka" lemon plants, the symptoms of the CTV-positive "Eureka" lemon plants included reduced growth of the young shoots, and yellowing, curling, reduced growth, and luster of the mature leaves. The CTV-positive lemon plants exhibited typical symptoms of rapid decline from 120 to 210 days after the IAP (see symptoms in Figure   The results from the RT-qPCR indicated that the CTV titers increased significantly over time in the "Shatangju" mandarin plants (F = 7.4590; df = 6; p = 0.0001) and "Eureka" lemon plants (F = 4.3090; df = 5; p = 0.0094) after 5 days of IAP exposure to CTV-positive D. citri, and the CTV titers of CTV-positive "Shatangju" mandarin plants were significantly higher than those of the CTV-positive "Eureka" lemons ( Figure 4B).
The symptoms observed in crown diameter, plant height, leaf, and stem morphology did not differ significantly between CTV-positive and CTV-negative "Shatangju" mandarin plants. However, CTV-positive "Eureka" lemon plants had significantly smaller crown diameters compared to CTV-negative counterparts (35.17 ± 2.08 cm vs. 50.80 ± 4.18 cm (p = 0.0119), respectively) and were significantly shorter (73.90 ± 1.84 cm vs. 90.13 ± 4.33 cm (p = 0.0119), respectively) ( Figure S3A). Compared to the negative controls of "Eureka" lemon plants, the symptoms of the CTV-positive "Eureka" lemon plants included reduced growth of the young shoots, and yellowing, curling, reduced growth, and luster of the mature leaves. The CTV-positive lemon plants exhibited typical symptoms of rapid decline from 120 to 210 days after the IAP (see symptoms in Figure S3B, S3C, and S3D).

Discussion
Aphids use piercing mouthparts (stylets) to penetrate the plant vascular systems and access sugary fluids in the phloem. This is how viruses are acquired and transmitted during feeding [52]. CTV is acquired and transmitted in a semi-persistent non-circulative manner by several aphid species, including the most important species Aphis aurantii, A. citricidus, A. gossypii, and A. spiraecola [26,53]. It was reported that CTV could also be vectored by Ferrisia virgata (Cockerell) on lime trees in Africa [54]; however, the reported transmission was based on a single experiment requiring further confirmation. There have been no published reports of the acquisition and transmission of CTV by other phloem-feeding sternorrhynchans, such as psyllids, which feed on Citrus and their hybrids. It should be mentioned that D. citri is often present in citrus plantations where CTV occurs.
Reported CTV acquisition intervals and rates of transmission by aphids are highly variable. For example, A. citricidus is regarded as the most effective vector of CTV to sweet orange and Mexican lime trees (C. × aurantiifolia var. aurantiifolia (Christm.) Swingle) [55]. However, the reported AAP for A. citricidus to acquire CTV from sweet oranges and pomelo crops ranges from 1 to 24 h, while the exposure duration necessary for transmission ranges from 30 min to 100 h [1,14,21,22]. Moreover, CTV transmission rates by single and multiple (3 to 100) A. citricidus ranged from 0.0% to 66.6% and from 0% to 100%, respectively, which is also related to the isolates of CTV [24,[55][56][57]. High acquisition rates of CTV by aphids are a necessary precondition for the effective transmission of CTV, as the virus would not multiply within the insect [1,14,21,22].
Recently, a study revealed that CTV was the most abundant virus in Florida D. citri populations by high-throughput sequencing [58]. The results in this study indicate a high prevalence (40% to 100%) of CTV among D. citri collected from orchards. Moreover, the CTV transmission by groups of 10 CTV-positive D. citri adults led to 58.33% detection ratios of CTV-positive rate in "Shatangju" mandarin and 83.33% in "Eureka" lemon. Moreover, NGS sequencing suggests the CTV genome retrieved from positive D. citri samples was found unbroken.
Although we observed no significant differences between the overall aspect of CTVpositive "Shatangju" mandarin plants and CTV-negative plants, obvious symptoms such as decline (without stem pitting) were observed in positive "Eureka" lemon plants. This could be related to literature claims that resistance in the scions and rootstock relate to different plant virus strains or plant species [41,42,59]. In addition, previous research indicated that the expression, intensity, and severity of the symptoms in citrus depend on the specific infecting CTV strain or a mixture of CTV isolates [2,60]. The CTV strain used in our study was similar to the CTV isolates of the "VT-Asian" subtype [17,61]. The VT genotype viruses are regarded as mild isolates (i.e., symptomless for most citrus) or seedling yellows isolates resulting in symptoms characterized by stunting, small pale or yellow leaves, and a reduced root system in sour orange, grapefruit, or lemon seedlings [2,62,63]. We suppose this is the reason why stem pitting, the most typical symptom of CTV infection in some citrus species, was not observed during the research.
The estimated CTV titers decreased significantly after D. citri was transferred from CTV-positive "Shatangju" mandarin trees to orange jasmine plants, while CTV can persist within D. citri for some time following the potential transmission to healthy "Shatangju" mandarin plants. In addition, once CTV is acquired by 4th instars, it will persist in the psyllids through nymphal molts and eclosion of 5th instars to adults, suggesting that CTV is transmitted by D. citri not in a "semi-persistent" manner as with the aphid vectors of this pathogen. Our observations, albeit preliminary, indicate that the transmission efficiency of the virus by the CTV-positive psyllids to previously healthy "Shatangju" mandarin plants increases with the duration of feeding by the psyllids. Thus, CTV might be persistent but not propagative in D. citri, and the mechanisms of CTV transmission require further clarification.
Furthermore, the persistence of detectable CTV within D. citri was influenced by the host plant species. The CTV titers of CTV-positive D. citri transferred to healthy "Shatangju" host plants did not recede after 15 days (Figure 3), possibly indicating that fresh sap from a suitable host may be necessary for the persistence of CTV in D. citri for 15 days or more. Direct evidence of transmission comes from healthy host plants acquiring the virus from positive D. citri. The efficiency of the potential CTV transmission was also influenced by host plant species, which suggests further factors (e.g., viral strains and particular aspects of D. citri biology) might influence CTV transmission in the field, still warranting further study.

Conclusions
Our study demonstrates that the Asian citrus psyllid, D. citri, can acquire CTV with relatively high efficiency and is capable of transmitting the virus to two citrus species. The results in this preliminary report on CTV acquisition and transmission by D. citri were confirmed by different approaches (whole-genome sequencing, conventional PCR, RT-qPCR, and ELISA) relying on detection of viral nucleic acid and capsule proteins, now awaiting further controlled transmission studies and field observations. These findings may raise concerns about the D. citri psyllid as a vector related to two important citrus pathogens.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/insects12080735/s1. Figure S1: Sensitiveness of detection of Citrus tristeza virus (CTV) cDNA in host samples through conventional PCR with the universal primer set Uni-f/r. Figure S2: Citrus tristeza virus (CTV) detection in cDNA extractions of Diaphorina citri from orchards in Qingyuan, Guangdong (A), Ganzhou, Jiangxi (B), and Zhaoqing, Guangdong (C), P.R. China. Figure S3: Representative pictures of symptoms of Citrus × Limon plants after 120 days exposure to CTV-positive Diaphorina citri population in comparison to healthy plants after exposure to the CTV-negative psyllids, Table S1: Different species of host plants used in each experiment about CTV acquisition, persistence, and transmission by Diaphorina citri. Table S2: Viral load thresholds for the detection of Citrus tristeza virus (CTV) in Diaphorina citri cDNA samples used in this study, from correlating RT-qPCR quantitative estimations with a visual band detection by PCR. Table S3

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The data presented in this study are available in the Supplementary Materials. CTV sequence obtained in this study has been deposited in GenBank.