Persistence of Phytoplasma and Control Efficacy of Oxytetracycline Tree Injection for Elaeocarpus sylvestris (Lour.) Poir. var. ellipticus (Thunb.) H.Hara Decline Disease

Abstract

Elaeocarpus sylvestris (Lour.) Poir. var. ellipticus (Thunb.) H.Hara, an evergreen tree species native to Jeju Island, South Korea, has experienced a progressive decline associated with phytoplasma infection. This study aimed to evaluate the efficacy of oxytetracycline-based tree injection for suppressing phytoplasma and improving tree vitality. Two formulations—oxytetracycline hydrochloride (4.3%) and oxytetracycline calcium alkyltrimethyl ammonium (17%)—were administered to 40 infected individuals across two sites using a gravity-fed injection system. Treatment efficacy was evaluated based on chlorophyll content as an indicator of physiological recovery, while phytoplasma presence was assessed by PCR at 150 days after injection. The oxytetracycline hydrochloride group showed the highest suppression, with a 70% phytoplasma non-detection rate as determined by PCR analysis. Treated trees exhibited significantly higher chlorophyll content compared to untreated infected controls. These findings suggest that minimally invasive tree injection using oxytetracycline can provide temporary suppression of phytoplasma and support physiological recovery in E. sylvestris.

1. Introduction

Foliar yellowing symptoms in various plant species were previously believed to be caused by viral infections; however, it was later discovered that cell wall-less, pleomorphic prokaryotes morphologically resembling mycoplasmas (mycoplasma-like organisms; MLOs) reside in the phloem tissues of plants [1]. Since then, these organisms, now known as phytoplasmas, have been reported in more than 1000 plant species worldwide, posing a significant threat to a broad range of host plants, including those in fruit orchards, forest ecosystems, and urban landscapes, and resulting in substantial economic losses [2,3]. As obligate parasites, phytoplasmas are transmitted by phloem-feeding insects or grafting and disrupt host physiology by altering carbohydrate distribution, phytohormone balance, and gene expression [4,5].

Infected plants exhibit a wide array of symptoms, including witches’ broom, yellowing, stunting, and progressive decline of the affected trees [6,7]. The primary method used for detecting phytoplasma is the polymerase chain reaction (PCR) assay, targeting various phytoplasma gene regions such as 16S ribosomal RNA, sec genes, ribosomal protein (rp) operons, and elongation factor Tu (tuf), which have become standard targets for phytoplasma detection [7,8].

The primary strategy for managing phytoplasma-infected plants involves controlling insect vectors, most commonly through the application of insecticides; however, complete eradication of the vectors remains unachievable, and management remains particularly challenging due to the phytoplasmas’ localization within the phloem and their non-culturable nature [9]. Another approach involves the use of tetracycline-class antibiotics administered via tree injection, which can temporarily alleviate disease symptoms, although it has been reported that this method cannot eliminate phytoplasmas entirely from host plants [3,10].

Tree injection is an alternative method for delivering plant protection agents directly into the vascular system of trees, offering a targeted approach that can replace conventional foliar sprays or soil applications. This technique provides several advantages, such as enhanced uptake efficiency, reduced operator exposure, lower environmental impact, and minimized harm to non-target organisms [11,12].

Tree injection methods can be broadly categorized into three types: high-pressure injection using compressed gas, low-pressure injection using a spring mechanism or manual compression, and passive uptake, driven by the tree’s natural transpiration process [13]. A commonly studied chemical control method for phytoplasma-infected plants involves the tree injection of oxytetracycline-class antibiotics [14,15].

As of now, phytoplasma-specific pesticides registered in Korea include oxytetracycline calcium alkyltrimethyl ammonium (17%) wettable powder and oxytetracycline hydrochloride (4.3%) dispersible concentrate, both of which are used for controlling witches’ broom disease in jujube trees [16]. Various wood anatomical features, such as the size and arrangement of vessels, vessel elements, and axial parenchyma cells, determine the pathways and efficiency of substance distribution throughout the tree [17]. However, the optimal drilling depth for maximizing the systemic movement of injected chemicals remains unclear and has not been established for individual tree species.

Thus, novel approaches are required to achieve the complete elimination of phytoplasmas from infected trees. This includes investigations into whether tetracycline antibiotics are systemically translocated throughout the plant, optimization of tree injection timing based on the host’s growth stage, and development of effective antibiotic delivery protocols that consider the anatomical characteristics (e.g., xylem structure) of the target tree species.

Elaeocarpus sylvestris (Lour.) Poir. var. ellipticus (Thunb.) H.Hara, commonly known as “dampalsu”, is a rare plant species whose northernmost distribution is confined to Jeju Island, Korea [18]. Notably, the natural habitat at Cheonjiyeon Waterfall (Natural Monument No. 163) and the specimen at Gangjeong-dong, Seogwipo (Natural Monument No. 544) have been designated as protected cultural properties [19]. In addition, the presence of antioxidant, antibacterial, and antiviral compounds in extracts of E. sylvestris was confirmed through phytochemical analysis [20,21,22]. However, starting in 2013, widespread dieback symptoms were observed for the first time in E. sylvestris trees planted as street trees, in parks, and in native habitats across Jeju City and Seogwipo City, though the causal agent was initially unknown [23]. Subsequently, polymerase chain reaction (PCR) diagnostics in 2017 confirmed phytoplasma infection as the causal agent [24,25]. The conventional method of managing tree diseases and insect pests involves canopy spraying; however, this method has limited efficacy and poses environmental concerns due to extensive pesticide drift and loss. As a result, tree injection, an approach that minimizes pesticide exposure to non-target organisms, is increasingly being adopted [11,26]. In Korea, research on tree injection for phytoplasma-infected trees dates back to 1976, when oxytetracycline was used for controlling witches’ broom disease in jujube (Ziziphus jujuba) trees [27]. Although subsequent studies have been conducted [28,29,30], complete eradication of the phytoplasma has not been achieved, and tree injection has only been effective in slowing the progression of disease symptoms.

This study aimed to evaluate the efficacy of oxytetracycline-class antibiotics delivered via tree injection for suppressing phytoplasma in E. sylvestris, based on chlorophyll content measurements and PCR diagnostics conducted in the same treatment year. Accordingly, we investigated whether a tree injection protocol refined by wood anatomical characteristics, including injection timing and antibiotic concentration, could improve systemic compound distribution within the stem, thereby enhancing phytoplasma suppression and promoting recovery from decline symptoms.

2. Materials and Methods

2.1. Vessel Distribution and Wood Anatomy of E. sylvestris

Prior to tree injection, a wood dyeing assay was conducted to evaluate the efficiency of chemical translocation and to identify functional xylem vessels in Elaeocarpus sylvestris (Lour.) Poir. var. ellipticus (Thunb.) H.Hara. Branches approximately 12 cm in diameter were cut to a length of 1.5 m and immersed in a 1% fuchsin acid solution (Daejung Inc., Siheung-si, Republic of Korea), commonly used for xylem tracing in anatomical studies, for 72 h. After dye uptake, cross-sections were obtained at 20 cm and 80 cm above the immersion site to assess the extent of xylem staining. Additionally, 1.0 L of 1% fuchsin acid solution was administered to a tree with a diameter of 30 cm using a gravity-fed tree injection system (Figure 1). Two drilling depths were applied: a shallow depth of 1 cm, targeting only the outer portion of the sapwood, and a deeper depth of 3 cm, encompassing a broader area of the sapwood. Seven days post-injection, transverse sections were obtained above and below the injection point to evaluate dye distribution and determine the optimal drilling depth for effective chemical delivery. Gravity-fed injection was employed alongside immersion to better simulate field application conditions and to assess vertical dye movement from the injection site.

2.2. Experimental Design and Selection of Tree Injection Sites

The candidate individuals for tree injection were E. sylvestris trees confirmed to be infected with phytoplasma via preliminary PCR diagnostics. For each treatment group, ten infected individuals were selected, with stem diameters ranging from 16 to 24 cm. Negative control individuals (phytoplasma-free) were designated at each location, with three individuals assigned per site. Each chemical agent was applied at two concentration levels: the standard concentration and a higher concentration equivalent to twice the standard level. The tree injection trials were conducted in two regions of Jeju Island, including street tree plantations in the Hwabuk Industrial Complex area of Jeju City and the landscaped planting zones surrounding the International Convention Center in Seogwipo City.

2.3. Selection and Composition of Antimicrobial Agents

Two oxytetracycline-based antimicrobial formulations were used for tree injection in E. sylvestris affected by decline symptoms. These included a 17% wettable powder formulation of oxytetracycline calcium alkyltrimethyl ammonium and a 4.3% dispersible concentrate of oxytetracycline hydrochloride, both of which are registered for phytoplasma control in jujube (Ziziphus jujuba) in Korea. The specific compositions of the formulations used for tree injection are provided in

Table 1. The trees treated with tree injection and their corresponding pesticide formulations in this study.

№	Concentration	Chemical/ Control Agent	Formulation (Dilution Ratio)/ Number of Holes		DBH * (cm)	№	Concentration	Chemical/ Control Agent	Formulation (Dilution Ratio)/ Number of Holes		DBH * (cm)
	Jeju-si						Seogwipo-si
1	Standard	OTC-ca ** (17%)	2.0 g/50 mL (25-fold)	4	20	24	Standard	OTC-ca ** (17%)	2.0 g/50 mL (25-fold)	3	18
2				3	18	25				4	22
3				4	20	26				4	20
4				4	20	27				3	18
5				3	>18	>28				4	24
6		OTC-hy *** (4.3%)	3.0 mL/50 mL (16-fold)	4	22	29		OTC-hy *** (4.3%)	3.0 mL/50 mL (16-fold)	4	22
7				4	20	30				4	22
8				4	22	31				3	18
9				4	20	32				3	18
10				3	18	33				4	20
11	High	OTC-ca ** (17%)	4.0 g/50 mL (12.5-fold)	3	16	34	High	OTC-ca ** (17%)	4.0 g/50 mL (12.5-fold)	4	22
12				3	18	35				4	20
13				4	20	36				3	18
14				4	20	37				4	20
15				4	22	38				3	18
16		OTC-hy *** (4.3%)	6.0 mL/50 mL (8-fold)	4	20	39		OTC-hy *** (4.3%)	6.0 mL/50 mL (8-fold)	3	18
17				4	20	40				4	20
18				3	18	41				4	22
19				4	20	42				3	18
20				3	18	43				4	22
21	Untreated				20	44	Untreated				18
22					18	45					22
23					18	46					24

Note: DBH *: Diameter at breast height; OTC-ca **: Oxytetracycline calcium alkyltrimethyl ammonium; OTC-hy ***: Oxytetracycline hydrochloride.

.

2.4. Tree Injection Method and Application Timing

Tree injection was performed using a gravity-fed uptake system. Tree injection was conducted on a total of 40 E. sylvestris trees at two experimental sites (Jeju-si and Seogwipo-si), with each oxytetracycline formulation applied at both the standard and double concentrations. The injection bottles used had a volume of 50 mL, and 5 mm drill bits were employed for creating injection holes. One hole was drilled per 5 cm of stem diameter. Tree injection was conducted twice, in March and then again in early April of 2024. All injection holes were made in the buttress root region (exposed root flare). The injection tubing and bottles were wrapped in aluminum foil to shield them from sunlight. All injections were performed between 9:30 a.m. and 11:30 a.m., under full sunlight conditions, coinciding with peak physiological activity of the trees.

2.5. Calibration Curve and Recovery Rate Test

Untreated samples were first purified with a QuEChERS-dSPE kit (150 mg magnesium sulfate, 25 mg C18) and centrifuged. The purified extracts were then diluted 5-fold with 0.1% formic acid in acetonitrile/water (50:50, v/v). A 0.5 mL aliquot of the diluted matrix (final volume 125 mL) was mixed with 0.5 mL of blank standard solution (final volume 250 mL). Calibration standards were prepared at concentrations of 0.0001, 0.00025, 0.0005, 0.001, 0.002, and 0.005 μg/mL. Each standard was injected (2 μL) twice into the LC-MS/MS system (Liquid Chromatography; LC, Tandem Mass Spectrometry; MS/MS), and calibration curves were constructed using peak areas from the resulting chromatograms. For the recovery test, 0.05 mg/kg of oxytetracycline (0.2 mL of 0.25 μg/mL standard solution F24063_3_5) was spiked into 5 g of untreated matrix in triplicate. Samples were analyzed following the same analytical procedure, and recovery rates were calculated accordingly.

2.6. Residue Analysis of Oxytetracycline

To assess the residue of oxytetracycline in the canopy of treated trees, two individuals per antibiotic treatment were selected. Branch samples were collected from the terminal canopy positions in four cardinal directions. Five grams of finely chopped twig samples were placed into 50 mL conical tubes and hydrated with 15 mL of water for 30 min. Subsequently, 25 mL of 5% formic acid in acetonitrile was added, and the mixture was shaken vertically at 1300 rpm for 5 min. QuEChERS original extraction kits (4 g magnesium sulfate, 1 g sodium chloride) were added to the mixture, which was again shaken at 1300 rpm for 5 min and centrifuged at 3800 rpm for 5 min. One milliliter of the supernatant was transferred to a dSPE tube containing 150 mg magnesium sulfate and 25 mg C18, vortexed for 30 s, and centrifuged at 12,000 rpm for 5 min. A 0.1 mL aliquot of the purified extract was mixed with 0.9 mL of 0.1% formic acid in acetonitrile/water (50:50, v/v) and analyzed using LC-MS/MS.

2.7. Chlorophyll Content Measurement

To assess visual recovery from disease symptoms following tree injection, chlorophyll content was measured in 40 treated individuals and 6 untreated controls. Additionally, 8 healthy E. sylvestris individuals with no visible symptoms were also analyzed. Chlorophyll content was measured using a SPAD (Soil Plant Analysis Development) meter (Konica Minolta Inc., Tokyo, Japan) approximately 150 days after tree injection. For each sampled tree, SPAD measurements were conducted uniformly between 1:00 p.m. and 3:00 p.m., using sun-exposed leaves located at the outermost part of the canopy.

For each individual, four readings were taken from mature leaves. Statistical comparisons of SPAD values among treatment groups were performed using one-way analysis of variance (ANOVA) in SPSS version 19.0, followed by Duncan’s multiple range test (MRT) for post hoc analysis. Differences were considered statistically significant at p < 0.05.

2.8. PCR Diagnosis Before and After Tree Injection

The effectiveness of four injected antimicrobial treatments was evaluated using PCR diagnostics targeting phytoplasmas with universal primers P1/P7 [31,32] (

Table 2. Universal primer sets for phytoplasma amplification in this study.

Primer	Remark	Target Genes	Sequence (5′-3′)	Tm Value (°C)	Reference
P1	Forward	16S rRNA	AAGAGTTTGATCCTGGCTCAGGATT	55	Deng and Hiruki (1991) [31]
P7	Reverse	23S rRNA	CGTCCTTCATCGGCTCTT		Schneider et al. (1995) [32]

). PCR using P1/P7 primers was performed with an initial denaturation at 94 °C for 7 min, followed by 38 cycles of denaturation at 94 °C for 1 min, annealing at 55 °C for 2 min, and extension at 72 °C for 3 min. A final extension was carried out at 72 °C for 10 min. PCR was performed before injection and at 60-, 90-, 120-, and 150-days post-injection for all individuals. Newly emerged leaves were collected from each individual and used as the source material for DNA extraction. PCR analysis was performed in triplicate for each sample. Healthy trees were used as negative controls.

3. Results

3.1. Anatomical Characteristics of E. sylvestris Wood

Cross-sectional analysis of test samples immersed in 1% fuchsin acid solution for 72 h revealed that the dye had moved approximately 80 cm upward from the point of immersion along the 1.2 m sample length (Figure 2). At the 20 cm cross-section above the immersion site, dye penetration in vessel tissues was observed throughout the sapwood region, excluding the heartwood and visibly decayed areas. In contrast, the 80 cm cross-section showed no visible staining in most of the sapwood, including the heartwood region, except for slight dye traces in vessel tissues located just beneath the bark. These observations suggest that the most recently developed outer vessels serve as the primary pathway for axial translocation of water and injected solutions. Gravity-fed tree injection of 1.0 L of 1% fuchsin acid solution was performed on a 30 cm diameter Elaeocarpus sylvestris (Lour.) Poir. var. ellipticus (Thunb.) H.Hara individual for seven days. Staining was observed up to approximately 57 cm above and 30 cm below the injection site, indicating bidirectional translocation of the dye within the vascular system (Figure 3). Analysis of transverse sections above the injection site revealed that a drilling depth of 1 cm resulted in dye penetration in vessel tissues up to 57 cm, whereas a 3 cm drilling depth led to dye movement of only about 27 cm. This suggests that shallower drilling depths facilitate more effective upward movement of injected substances toward the canopy. In the sections below the injection point, dye penetration reached approximately 30 cm with a 1 cm drilling depth, while the 3 cm depth resulted in penetration up to 25 cm. These findings indicate that shallower drilling allowed for longer-range dye movement in both upward and downward directions. The results demonstrate that in E. sylvestris, shallow drilling depths allow for greater axial translocation of solutions through active vessel tissues. This implies that, under practical application, tree injections administered with minimal drilling depth could enhance systemic distribution of injected chemicals throughout the entire host. Furthermore, the detection of dye both above and below the injection point confirms that the injected solution can move bidirectionally toward both the canopy and the root system.

3.2. Residual Characteristics of Oxytetracycline in E. sylvestris

The calibration curve for oxytetracycline quantification in E. sylvestris exhibited a linear regression equation of y = 33,561,344.707x + 1479.320, with a correlation coefficient (R²) of 1.000, indicating excellent suitability for quantitative analysis (Figure 4). The LOQ (limit of quantification) was 0.05 mg/kg, and the minimum detectable amount was 0.0002 ng (0.0001 μg/mL × 2 μL). Recovery rates ranged between 70% and 120%, which falls within the recommended range specified in the Korean pesticide registration guidelines [33] (

Table 3. LOD, LOQ, and recovery rates of the oxytetracycline method for detecting oxytetracycline in E. sylvestris trees after tree injection.

Replicates	Final Vol. (mL)	Injection Vol. (μL)	LOD a (ng)	Recovery (%)	Mean Recovery (%)	LOQ b (mg/kg)	RSD c (%)
1	250	2	0.0002	97.6	95.3 ± 2.9	0.005	3.0
2				92.1
3				96.3

Note: LOD ^a: limit of detection. LOQ ^b: limit of quantification. RSD ^c: relative standard deviation.

). At 150 days after tree injection, oxytetracycline residue levels in E. sylvestris individuals ranged from 0.005 to 0.193 mg/kg. The highest average residue was observed in the OTC–wettable powder (high concentration) group, with a mean value of 0.1325 (±0.12) mg/kg across all sampling directions (

Table 4. Residual amounts of oxytetracycline in the E. sylvestris trees under each antibiotic treatment.

	Residue Levels of Oxytetracycline (mg/kg)
	Samples ID		Residue Levels	Average
Non-treated	24C59N2_Ctrl	_1	<0.005	<0.005
		_2	<0.005
		_3	<0.005
OTC-ca *	SP-Oxi-06E		0.132	0.0893 (±0.10)
	SP-Oxi-06W		0.036
	SP-Oxi-06N		0.033
	SP-Oxi-06S		0.156
OTC-ca **	SP-Oxi-10E		0.189	0.1325 (±0.12)
	SP-Oxi-10W		0.193
	SP-Oxi-10N		0.111
	SP-Oxi-10S		0.037
OTC-hy ***	SP-Oxi-15E		0.023	0.0278 (±0.02)
	SP-Oxi-15W		0.005
	SP-Oxi-15N		0.046
	SP-Oxi-15S		0.037
OTC-hy ****	SP-Oxi-20E		0.190	0.0998 (±0.11)
	SP-Oxi-20W		0.109
	SP-Oxi-20N		0.088
	SP-Oxi-20S		0.012

Note: OTC-ca *: oxytetracycline calcium alkyltrimethylammonium (standard concentration); OTC-ca **: oxytetracycline calcium alkyltrimethylammonium (high concentration); OTC-hy ***: oxytetracycline hydrochloride (standard concentration); OTC-hy ****: oxytetracycline hydrochloride (high concentration).

). When these residue levels were compared with phytoplasma detection status by PCR, suppression of phytoplasma was observed in individuals with average oxytetracycline levels of 0.0893 (±0.10) mg/kg in the OTC–wettable powder (standard concentration) group, and 0.0278 (±0.02) mg/kg in the OTC–dispersible concentrate (standard concentration) group. In particular, the average residue concentration in the OTC–dispersible concentrate (standard concentration) group was 0.0278 (±0.02) mg/kg, at which phytoplasma suppression was confirmed in this study.

3.3. Chlorophyll Content in Treated and Control Individuals Following Tree Injection

Chlorophyll content was measured in all treated individuals after tree injection. In the oxytetracycline calcium alkyltrimethyl ammonium (17%) group, individuals treated with the high concentration exhibited slightly higher SPAD values than those treated with the standard concentration. Among the oxytetracycline hydrochloride (4.3%) treatment groups, individuals in the Seogwipo site recorded the highest chlorophyll content. In the untreated control individuals, SPAD values were 46.5083 (±4.5) in Jeju City and 46.8083 (±2.7) in Seogwipo, both of which were lower than the values observed in treated individuals. In contrast, healthy E. sylvestris individuals without phytoplasma infection had significantly higher chlorophyll content, recording 62.0375 (±4.2) in both Jeju City and Seogwipo during the same measurement period. One-way ANOVA revealed statistically significant differences in chlorophyll content across treatment groups at both study sites (p < 0.05). Duncan’s multiple range test (MRT) further divided the individuals into three distinct groups in Jeju City and four in Seogwipo (Figure 5). The highest chlorophyll content among all treatments was observed in the oxytetracycline hydrochloride (4.3%) high concentration group in Jeju City and the standard concentration group in Seogwipo (

Table 5. Chlorophyll content in different treatment groups and pesticide types following tree injection.

Chemical/Control Agent	Previous Year’s Leaves
	Jeju-si Average	Seogwipo-si Average
OTC-ca *	52.5770 (±6.8, b)	52.4550 (±4.0, c)
OTC-hy ***	53.0750 (±7.1, b)	56.1110 (±5.5, b)
OTC-ca **	52.9000 (±3.0, b)	53.2900 (±2.8 c)
OTC-hy ****	54.9200 (±4.4, b)	52.0850 (±3.6, c)
Untreated	46.5083 (±4.5, c)	46.8083 (±2.7, d)
Healthy tree	62.0375 (±4.2, a)

Note: a, b, c, d: Different letters indicate Duncan’s MRT test (p < 0.05); OTC-ca *: oxytetracycline calcium alkyltrimethylammonium (standard concentration); OTC-ca **: oxytetracycline calcium alkyltrimethylammonium (high concentration); OTC-hy ***: oxytetracycline hydrochloride (standard concentration); OTC-hy ****: oxytetracycline hydrochloride (high concentration).

).

3.4. Phytoplasma Detection by PCR Before and After Tree Injection

Based on an application volume of 50 mL per 5 cm of stem diameter, complete uptake of the injected solution was observed in some individuals within 5 h post-injection, while most trees absorbed the entire dose within two days.

Prior to tree injection, phytoplasma was detected in the newly emerged shoots of all test individuals, confirming that the selected trees were suitable for evaluating the efficacy of each antibiotic treatment. In the oxytetracycline calcium alkyltrimethyl ammonium (wettable powder) group, PCR at 60 days post-injection showed phytoplasma non-detection in 7 out of 10 individuals (70%) in the standard concentration group and in 6 out of 10 individuals (60%) in the high concentration group. The same detection rates persisted at 90 days post-injection. However, by 150 days post-injection, two additional reinfections were observed in each group, resulting in 5 phytoplasma-negative individuals (50%) in the standard group and 4 (40%) in the high concentration group (Figure 6,

Table 6. Detection of phytoplasma in target trees and regions before and 150 days after tree injection.

№		Chemical/Control Agent	PCR-Based Detection of Phytoplasma Using P1/P7 Primers (Before and After Tree Injection)										Total Rate
			Before		60 Days		90 Days		150 Days		Undetected Rate (PCR)
J *	S **		J *	S **	J *	S **	J *	S **	J *	S **	J *	S **
1	24	OTC-ca (Standard)	○	○	×	○	×	○	×	○	60%	40%	50%
2	25		○	○	×	○	×	○	×	○
3	26		○	○	×	×	×	×	○	×
4	27		○	○	×	×	×	×	×	○
5	28		○	○	○	×	○	×	○	×
11	34	OTC-ca (High)	○	○	○	×	○	×	○	○	40%	40%	40%
12	35		○	○	×	○	×	○	×	○
13	36		○	○	○	×	○	×	○	×
14	37		○	○	×	○	×	○	○	○
15	38		○	○	×	×	×	×	×	×
6	29	OTC-hy (Standard)	○	○	×	×	×	×	×	×	60%	80%	70%
7	30		○	○	×	×	×	×	×	×
8	31		○	○	×	○	×	○	○	○
9	32		○	○	○	×	○	×	○	×
10	33		○	○	×	×	×	×	×	×
16	39	OTC-hy (High)	○	○	×	×	×	×	×	×	80%	60%	70%
17	40		○	○	×	○	×	○	×	○
18	41		○	○	×	×	×	×	×	○
19	42		○	○	○	×	○	×	○	×
20	43		○	○	×	×	×	×	×	×
21	44	Untreated	○	○	○	○	○	○	○	○	0%
22	45		○	○	○	○	○	○	○	○
23	46		○	○	○	○	○	○	○	○

Note: J *: Jeju-si; S **: Seogwipo-si. “○” indicates phytoplasma detected by PCR, while “×” indicates non-detection.

). In the oxytetracycline hydrochloride (dispersible concentrate) group, PCR at 60 and 90 days post-injection showed 8 phytoplasma-negative individuals (80%) in both the standard and high concentration groups. At 150 days, this number decreased slightly to 7 individuals (70%) in each group. Based on the PCR results, the final phytoplasma non-detection rates at 150 days post-injection were as follows: 50% for the OTC–wettable powder (standard concentration), 40% for the OTC–wettable powder (high concentration), and 70% for both the standard and high concentration groups of the OTC–dispersible concentrate.

These findings indicate that the dispersible concentrate exhibited slightly higher phytoplasma suppression compared to the wettable powder. No signs of phytotoxicity, such as leaf scorching or curling, were observed in any of the treated individuals throughout the experiment. In contrast, all untreated control individuals tested positive for phytoplasma consistently from before treatment through 150 days post-injection (Figure 6 and Table 6).

4. Discussion

Anatomical analysis using fuchsin acid staining revealed that the active conducting vessels in Elaeocarpus sylvestris (Lour.) Poir. var. ellipticus (Thunb.) H.Hara are located immediately beneath the bark. This observation is consistent with prior findings that, in diffuse-porous species such as E. sylvestris [34], water conduction primarily occurs through the outermost 1–3 growth rings [35]. Staining patterns indicated that injected solutions moved both upward toward the canopy and downward into the root zone. This bidirectional translocation aligns with earlier reports demonstrating upward and downward movement of injected compounds within tree xylem [35,36,37]. In particular, when dye was injected using a gravity-fed method at a drilling depth of 1 cm, staining extended 57 cm upward and 30 cm downward from the injection site. This finding corresponds with previous studies indicating that upward movement through apoplastic pathways occurs more rapidly, while downward movement is slower due to the symplastic transport mechanisms [38,39,40]. Based on these anatomical characteristics and experimental results, it can be concluded that shallow drilling enhances the systemic distribution of injected compounds by facilitating efficient translocation via active vessel tissues. Furthermore, shallow drilling minimizes wounding, which has been associated with higher rates of wound closure and compartmentalization [41,42], thereby offering a less invasive and environmentally sustainable method for tree injection. The gravity-fed dye injection method used in this study confirmed bidirectional movement of the solution from the point of application. Given that the phytoplasma associated with E. sylvestris decline has been detected in all anatomical regions, including the upper and lower canopy, root tissues, and all cardinal directions (east, west, south, and north) [19], the injection site at the buttress root zone was considered an appropriate location for maximizing systemic delivery of the treatment.

In a study conducted in Japan, Candidatus Phytoplasma asteris was effectively suppressed in chrysanthemum tissues following tissue culture treatment with tetracycline hydrochloride at a concentration of 0.1 mg/kg, whereas phytoplasmas remained detectable at 0.01 mg/kg [43]. These findings suggested that a minimum threshold near 0.1 mg/kg may be required for effective phytoplasma suppression under in vitro conditions. In contrast, the present study demonstrated successful suppression of phytoplasma in E. sylvestris at a lower oxytetracycline concentration of 0.0278 mg/kg following tree injection. This discrepancy may be attributed to differences in host physiology, tissue context (in vitro vs. in planta), or the application method used (tissue culture vs. field injection). While in vitro assays provide controlled environments for assessing antimicrobial efficacy, in planta systems involve complex interactions between xylem transport dynamics, tissue compartmentalization, and host immune responses, all of which may influence the effective concentration required for pathogen suppression.

These results underscore the importance of system-specific calibration when interpreting effective antibiotic thresholds for phytoplasma control. Further investigation is warranted to determine the minimum in planta oxytetracycline concentration required for sustained phytoplasma suppression in E. sylvestris under field conditions, using larger sample sizes and qPCR-based quantification of pathogen load over time.

The results of this study revealed a substantial difference in chlorophyll content between healthy E. sylvestris individuals and those exhibiting decline symptoms. Additionally, individuals that showed symptomatic recovery following antibiotic tree injection exhibited higher SPAD values than untreated, infected individuals. However, chlorophyll content can vary seasonally; thus, to obtain more reliable comparisons, future research should include time-series measurements of individual trees across different seasons [44]. Previous studies have examined the physiological responses of E. sylvestris, including chlorophyll dynamics, under various environmental conditions. However, those studies did not report specific SPAD values. Chlorophyll content has been shown to decrease under drought stress [45], while changes due to increased planting density appear to be limited from a physiological standpoint [46]. Since no prior studies have quantified chlorophyll content in E. sylvestris in Korea, direct comparison with our data was not possible. SPAD meters provide a simple and cost-effective method for in-field chlorophyll estimation. However, because SPAD values are unitless and species-specific, their interpretation requires additional empirical calibration and validation studies [47]. Therefore, future research should conduct systematic SPAD measurements on injected, untreated, and healthy individuals across different time points. Leaf chlorophyll content is a well-established indicator of photosynthetic capacity and offers critical insights into plant physiological status [48]. Accordingly, the chlorophyll data provided in this study may serve as an indirect diagnostic reference for distinguishing between healthy and phytoplasma-infected E. sylvestris and offer a useful foundation for developing future management strategies.

Tree injection trials conducted in Jeju City and Seogwipo City did not result in complete suppression of decline symptoms in E. sylvestris. However, the oxytetracycline hydrochloride (dispersible concentrate) group showed a phytoplasma non-detection rate of 70%, indicating relatively higher efficacy compared to other treatments. Most studies on phytoplasma control using oxytetracycline formulations have focused on oxytetracycline hydrochloride (OTC-HCl) as the primary treatment [15,49], and in this study, OTC-HCl also demonstrated higher control efficacy.

Tetracycline-class antibiotics have long been recognized as offering only temporary alleviation of phytoplasma symptoms and are generally ineffective at completely eliminating the pathogen from host tissues [10]. A previous study reported that stem injection with oxytetracycline hydrochloride (OTC-HCl) resulted in the recovery of over 86% of trees with mild jujube witches’ broom (JWB) disease [49]; however, this study demonstrated an even lower control efficacy.

Meanwhile, further research is needed to assess the presence of phytoplasma before and after tree injection using nested PCR, a more sensitive detection method. This study had limitations in that regard. Although both antibiotics initially resulted in a high initial rate of phytoplasma non-detection following tree injection, new infections were detected over time. This recurrence may be attributed to re-exposure to insect vectors following initial suppression [3] or to incomplete eradication of the pathogen, allowing systemic recolonization of host tissues. This pattern is consistent with previous studies, in which transient symptom remission was observed following tetracycline injection, but symptoms were reported to reoccur thereafter [19]. Therefore, it is essential to conduct studies aimed at identifying the insect vectors responsible for transmitting the phytoplasma associated with E. sylvestris decline disease. As phytoplasma is likely reintroduced through repeated insect feeding, long-term disease mitigation requires not only antibiotic treatment but also the implementation of integrated pest management (IPM) strategies to effectively control vector populations and reduce reinfection risk.

Since the first study on witches’ broom disease in jujube trees in 1976, oxytetracycline-based tree injection has been recognized as a strategy to slow the progression of phytoplasma-related diseases, rather than to cure them [27]. Even if symptoms are partially suppressed or not visible during the year of treatment [30], post-treatment monitoring remains essential to assess disease progression and reinfection risk.

The optimal timing for tree injection varies depending on the species [50]. As E. sylvestris is an evergreen broadleaf species, the conventional injection period used for managing witches’ broom disease in deciduous species such as jujube trees may not be appropriate. In this study, injections were scheduled in March and early April, corresponding to the early growth phase when leaves are actively expanding [51] and translocation rates are known to be high [40]. To further enhance uptake efficiency, injections were administered exclusively during the morning hours, avoiding midday periods when stomatal closure in broadleaf species could limit xylem transport efficiency [26]. Effective tree injection requires the use of a compound appropriate for the target pathogen, applied at the correct physiological timing, using a method that minimizes damage to the host [26]. No phytotoxic symptoms were observed at the doses used in this study, indicating that higher concentrations may further enhance the suppression of phytoplasma within the phloem tissues. Additionally, residue analysis conducted 150 days after tree injection confirmed the presence of oxytetracycline in all sampled tissues, demonstrating the formulation stability of both antibiotic products. Although two injections were administered during the study period, complete phytoplasma eradication was not achieved based on PCR analysis. Therefore, an additional injection during the period of downward sap flow may be necessary to enhance disease control efficacy.

In this study, the anatomical characteristics of E. sylvestris were assessed using fuchsin acid staining and gravity-fed injection. Injections were applied at the buttress root zone targeting active sapwood tissues, while minimizing damage to surrounding woody structures. Although complete suppression of phytoplasma was not achieved, treated individuals exhibited higher chlorophyll content than untreated controls, and some trees remained phytoplasma-negative up to 150 days post-injection based on PCR diagnostics.

However, phytoplasma is a systemic pathogen capable of infecting all tissues within the host [7], and in the case of E. sylvestris decline, it has been detected across all anatomical regions of the tree [19]. Therefore, even if phytoplasma is undetected in certain tissues, it may still persist and spread within the same individual. Therefore, post-injection monitoring and systematic long-term management are essential for mitigating decline progression and reducing mortality in infected trees. Moreover, this study highlights the lack of species-specific standards for drilling depth in conventional tree injection practices, suggesting that shallow drilling may enable efficient systemic translocation of injected compounds while reducing the risk of excessive wounding caused by unnecessarily deep perforation.

This study provides evidence that tree injection with oxytetracycline-based formulations can serve as a minimally invasive and partially effective method for the short-term suppression of phytoplasma in E. sylvestris. Anatomical tracing confirmed bidirectional compound movement following shallow drilling, while chlorophyll content and PCR results supported physiological and molecular improvements in some individuals. However, the incomplete eradication of phytoplasma, recurrence of infection, and variability in treatment efficacy highlight the limitations of relying solely on antibiotic injection. Given the systemic nature of phytoplasma and the likelihood of reinfection via insect vectors, long-term disease mitigation requires an integrated strategy that combines optimized injection timing and dosage, post-treatment monitoring, and vector surveillance and control. These findings support the development of species-specific tree injection protocols and emphasize the need for further research on sustainable, multi-layered management approaches for phytoplasma-associated tree decline.

5. Conclusions

This study confirmed that shallow drilling enhances the systemic delivery of injected compounds in Elaeocarpus sylvestris (Lour.) Poir. var. ellipticus (Thunb.) H.Hara by facilitating efficient translocation through active outer xylem tissues. Dye tracing experiments showed that a 1 cm drilling depth enabled dye movement up to 57 cm above and 30 cm below the injection point, demonstrating bidirectional mobility through functional vessels. Following tree injection, all treated individuals exhibited higher SPAD values than untreated controls, with the highest chlorophyll content observed in the OTC-HCl (4.3%) group at the Seogwipo site. Untreated infected trees recorded SPAD values of approximately 46.5–46.8, while healthy trees without phytoplasma infection showed significantly higher values of 62.0, indicating partial physiological recovery in treated trees but incomplete restoration to healthy levels. The number of treated trees per formulation (n = 10) may be considered adequate; however, the relatively small number of untreated control trees (n = 3) constitutes a limitation of this study, as it restricted the statistical power for comparison with a larger control group.

Among the treatments tested, oxytetracycline hydrochloride (4.3%, dispersible concentrate) was the most effective, achieving a phytoplasma non-detection rate of 70% at 150 days post-injection. Residue analysis confirmed detectable oxytetracycline concentrations (0.005–0.193 mg/kg) in canopy tissues, with excellent analytical reliability (R² = 1.000, recovery rates of 70%–120%).

Although reinfection occurred in some individuals over time, likely due to insect vectors or residual pathogen presence, the treatment delayed symptom progression and improved host vitality. This study has certain limitations, including a relatively low number of replicates for residue analysis, a lack of environmental standardization, and a short-term observation period. Therefore, long-term monitoring and broader field trials under varying environmental conditions are required to validate these findings. Although residual levels of oxytetracycline remained within the acceptable range specified by Korean pesticide registration guidelines, further studies are needed to assess potential environmental impacts and establish safe usage thresholds for repeated tree injection applications. This study suggests that oxytetracycline tree injection may serve as a supportive tool in the integrated management of phytoplasma-infected E. sylvestris, offering partial suppression of the pathogen and symptomatic relief, though reinfection remains a critical concern.